Purpose of the Guidance Document

(PDF; 6.56 MB)

In 1996, Environment Canada undertook an assessment of the aquatic effects of mining in Canada (AQUAMIN 1996¹), which provided recommendations regarding the review and amendments of the Metal Mining Liquid Effluent Regulations (currently titled the Metal Mining Effluent Regulations [MMER]) and the design of a national Environmental Effects Monitoring (EEM) program for metal mining. The MMER under the Fisheries Act direct metal mines to conduct EEM as a condition governing the authority to deposit effluent (MMER, Part 2, section 7). EEM is a science-based performance measurement tool used to assess the adequacy of the regulations. Although this guidance document is not a legal document, it is intended to provide guidance for mines in meeting their EEM requirements and conducting EEM studies. For the regulatory EEM requirements, refer to section 7 and Schedule 5 of the MMER. This guidance document replaces the 2002 version.

1 AQUAMIN. 1996. Assessment of the Aquatic Effects of Mining in Canada. Environment Canada – report available upon request by email at EEM-ESEE@ec.gc.ca

AAS: atomic absorption spectrophotometry

AES: Auger electron spectrometry

AETE Program: Aquatic Effects Technology Evaluation Program

ANCOVA: analysis of covariance

ANOVA: analysis of variance

AQUAMIN: Assessment of the Aquatic Effects of Mining in Canada

APHA: American Public Health Association

AVS: acide volatile sulphides

EPS: Environmental Protection Service of Environment Canada

ASPT: average score per taxon

ASTM: American Society for Testing and Materials

ATW: Aquatic Toxicity Workshop

AWWA: American Water Works Association

BACI: before/after control-impact

BAR: B.A.R. Environmental Inc.

B-C Index: Bray-Curtis Index

BEAK: Beak International Inc.

BL: biotic ligand

BLM: biotic ligand model

BMWP: biological monitoring working party

CABIN: Canadian Aquatic Biomonitoring Network

CAEAL: Canadian Association for Environmental Analytical Laboratories

CALA: Canadian Association for Laboratory Accreditation

CALK: combined alkaline stream

CBR: critical body residues

CCME: Canadian Council of Ministers of the Environment

CES: critical effect size

CETTP: Complex Effluent Toxicity Testing Program

C-I: control-impact

COSEWIC: Committee on the Status of Endangered Wildlife in Canada

CPUE: catch per unit effort

CVAAS: cold vapour atomic absorption spectrometry

CVAFS: cold vapour atomic fluorescence spectrometry

D.L.: detection limit

DDW: double distilled water

df: degrees of freedom

DGT: diffusive gradient thin film

DOC: dissolved organic carbon

DQOs: data quality objectives

EC: Environment Canada

EC₂₅: 25% effect concentration

EC₅₀: 50% effect concentration

EDA: effect directed analysis

EDTA: Ethylenediaminetetraacetic acid

EEM: environmental effects monitoring

ELAP: Environmental Laboratory Approval Program

ESG: ESG International Inc.

EU: European Union

exp.: exposure

FDP: final discharge point

FF: far-field

FRAP: Fraser River Action Plan

GC: gas chromatography

GFAAS: graphite furnace atomic absorption spectrometry

GLP: good laboratory practice

GIME: gel integrated minielectrode

GM-IC₂₅: geometric mean of all IC₂₅s

GPS: global positioning system

GSI: gonadosomatic index

HALW: hardness-adjusted laboratory water

HFPLM: hollow fibre permeation liquid membrane

HPLC: high performance liquid chromatography

HSB: hyper-saline brine

IC₂₅: 25% inhibition concentration

IC₅₀: 50% inhibition concentration

ICP-AES: inductively coupled atomic absorption spectrophotometry

ICP-MS: inductively coupled plasma mass spectrometry

ID: internal diameter

INRS: Institut national de la recherche scientifique

INAA: instrumental neutron activation analysis

IOC: investigation of cause

IOS: investigation of solutions

IARC: International Agency for Research on Cancer

ISO/IEC: International Organization for Standardization

LC₅₀: median lethal concentration

LCL: lower control limit

LIMS: laboratory information management system

LOE: lines of evidence

LPL: lowest practical level

LSI: liver somatic index

LT₂₅: time to 25% mortality

LT₅₀: time to 50% mortality

LWL: lower warning limit

MC-I: multiple control-impact

MDDEP: Ministère du Développement durable, de l’Environnement et des Parcs du Québec

MDL: method detection limit

MFO: mixed function oxygenase

MG: multiple gradient

MME: metal mine effluent

MMER: Metal Mining Effluent Regulations

MOE: Ministry of the Environment

MS: mass spectrometry

MS: mean square

MSE: municipal sewage effluent

MSI: mantle somatic index

MT: metallothionein

NABS: North American Benthological Society

NAMC: The North American Metals Council

NF: near-field

NOM: natural organic matter

NRBS: Northern River Basins Study

NSERC: Natural Sciences and Engineering Research Council

OECD: Organisation for Economic Co-operation and Development

PLC: Public Liaison Committee

PME: pulp mill effluent

PPER: Pulp and Paper Effluent Regulations

QA/QC: quality assurance / quality control

R²: coefficient of variation

RCA: reference condition approach

ref.: reference

RG: radial gradient

RISS: Regulatory Information Submission System

SD: standard deviation

SE: standard error

SEC: size exclusion chromatography

sem: simultaneous extracted metals

SEM: scanning electron microscopy

SETAC: Society of Environmental Toxicology and Chemistry

SG: simple gradient

SIMS: secondary ion mass spectrometry

SOP: standard operating procedure

SPE: solid phase extraction

SQT: sediment quality triad

SRM: standard reference material

SS: sum of squares

TER: toxicity emission rate

TIE: toxicity identification evaluation

TKN: Total Kjeldahl Nitrogen

TRE: toxicity reduction evaluation

TOC: total organic carbon

TSRI: Toxic Substances Research Initiative

U.S. EPA: United States Environmental Protection Agency

UCL: upper control limit

UWL: upper warning limit

UV: ultraviolet

VECs: valued ecosystem components

WAWW: whole-animal wet weight

WEF: Water Environment Federation

WHO: World Health Organization

WMI: Whitehorse Mining Initiative

WQG: water quality guideline

XPS: X-ray photoelectron spectroscopy

XRF: X-ray fluorescence

XAFS: X-ray fluorescence spectroscopy

XRM: X-ray microanalysis

YOY: young of the year

List of Acronyms

Chapter 1: Overview of the Metal Mining Environmental Effects Monitoring Program

Chapter 2: Study Design, Site Characterization and General Quality Assurance and Quality Control

Chapter 3: Effects on Fish and Fisheries Resources

Chapter 4: Effects on Fish Habitat: Benthic Invertebrate Community Survey

Chapter 5: Effluent Characterization and Water Quality Monitoring

Chapter 6: Sublethal Toxicity Testing

Chapter 7: Sediment Monitoring

Chapter 8: Data Assessment and Interpretation

Chapter 9: Alternative Monitoring Methods

Chapter 10: Information Management and Interpretative Reports

Chapter 11: Public Involvement in Metal Mining Environmental Effects Monitoring

Chapter 12: Investigation of Cause

Chapter 13: Report on Historical Information

Disclaimer

The objective of this document is to provide guidance to mines on how to meet the environmental effects monitoring regulatory requirements under the Metal Mining Effluent Regulations (MMER). This is not a legal interpretation of the MMER. For the Regulations, refer to the MMER.

Acknowledgements

The National Environmental Effects Monitoring (EEM) Office would like to thank the many people who contributed to the updating of this technical guidance document. The content was greatly improved by contributions from the members of the EEM National Team and Science Committee. The quality of the document was vastly improved through the efforts of Environment Canada’s editing team and individual members of the National EEM Office.

1. Overview of the Metal Mining Environmental Effects Monitoring Program

1.1 Purpose of the Guidance Document

1.2 The Metal Mining Effluent Regulations

1.3 Description of Environmental Effects Monitoring Studies

• 1.3.1 Effluent and Water Quality Monitoring Studies
  • 1.3.1.1 Effluent Characterization
  • 1.3.1.2 Sublethal Toxicity Testing
  • 1.3.1.3 Water Quality Monitoring
• 1.3.2 Biological Monitoring Studies
  • 1.3.2.1 Defining and Confirming Effects
  • 1.3.2.2 Historical Information
  • 1.3.2.3 Biological Monitoring Studies to Assess Effects
    • 1.3.2.3.1 Fish Population Survey
    • 1.3.2.3.2 Benthic Invertebrate Community Survey
    • 1.3.2.3.3 Fish Tissue Survey
  • 1.3.2.4 Biological Monitoring Studies to Investigate Effects
    • 1.3.2.4.1 Magnitude and Geographic Extent
    • 1.3.2.4.2 Investigation of Cause

1.4 Steps in Conducting and Reporting Environmental Effects Monitoring Studies

• 1.4.1 Conduct and Submit Results for Effluent Characterization, Sublethal Toxicity Testing and Water Quality Monitoring
• 1.4.2 Submit Study Design
  • 1.4.2.1 Study Design for Biological Monitoring Studies to Assess Effects
  • 1.4.2.2 Study Design for Biological Monitoring to Investigate Effects
• 1.4.3 Conduct Biological Monitoring Study
• 1.4.4 Conduct Data Assessment
• 1.4.5 Submit Interpretative Report
  • 1.4.5.1 Interpretative Report for Biological Monitoring Studies to Assess Effects
  • 1.4.5.2 Interpretative Report for Biological Monitoring Studies to Investigate Effects

1.5 Recognized Closed Mines

1.6 Identifying a Path through the Metal Mining Environmental Effects Monitoring Program

• 1.6.1 Critical Effect Sizes
• 1.6.2 Magnitude of Confirmed Effects
• 1.6.3 Decision Process for the Metal Mining Environmental Effects Monitoring Program
  • 1.6.3.1 Assessing Effects
  • 1.6.3.2 Investigating Confirmed Effects
    • 1.6.3.2.1 Level of Effort for Investigating Effects
  • 1.6.3.3 Timing of Studies to Assess Effects and Magnitude and Geographic Extent and to Investigate Cause

1.7 References

List of Tables

• Table 1-1: Effect indicators and endpoints for the fish population survey
• Table 1-2: Effect indicators and endpoints for the benthic invertebrate community survey
• Table 1-3: Critical effect sizes for metal mining environmental effects monitoring program
• Table 1-4: Evaluation of magnitude of confirmed effects in two consecutive phases

List of Figures

• Figure 1-1: Decision tree for the metal mining EEM program

1. Overview of the Metal Mining Environmental Effects Monitoring Program

1.1 Purpose of the Guidance Document

In 1996, Environment Canada undertook an assessment of the aquatic effects of mining in Canada (AQUAMIN 1996¹), which provided recommendations regarding the review and amendments of the Metal Mining Liquid Effluent Regulations (currently titled the Metal Mining Effluent Regulations [MMER]) and the design of a national Environmental Effects Monitoring (EEM) program for metal mining. The MMER under the Fisheries Act direct metal mines to conduct EEM as a condition governing the authority to deposit effluent (MMER, Part 2, section 7). EEM is a science-based performance measurement tool used to assess the adequacy of the regulations. Although this guidance document is not a legal document, it is intended to provide guidance for mines in meeting their EEM requirements and conducting EEM studies. For the regulatory EEM requirements, refer to section 7 and Schedule 5 of the MMER, located on the following website. This guidance document replaces the 2002 version.

The MMER prescribes that EEM studies be performed using documented and validated methods, and that their results be interpreted and reported in accordance with generally accepted standards of good scientific practice (MMER, Part 2, subsection 7(3)). The methodologies recommended in this guidance document are based on generally accepted standards of good scientific practice, and incorporate improvements based on program experience, input from multi-stakeholder working groups, and external research initiatives responding to EEM needs. As well, the Metal Mining EEM Review Team, which comprised a group of experts from government, industry, and environmental and Aboriginal groups, was established by Environment Canada to examine the experiences and results of the EEM program from the first phase of metal mining EEM studies and to provide recommendations to Environment Canada for improving the program. The final report, Review Team Report from Metal Mining Environmental Effects Monitoring Program (August 2007), is available on the EEM website. This document also reflects the changes to EEM requirements established by the 2006 and 2012 MMER amendments.

It should be emphasized that the methodologies provided in this guidance document do not constitute an exhaustive list of the possible means of conducting EEM. It is assumed that each study leader has sufficient knowledge to apply these recommendations using generally accepted standards of good scientific practice and is able to determine if unique conditions exist that would warrant modification of the generic study designs, while ensuring that regulatory requirements are met. For a scientific discussion on sound science, refer to Chapter 12 on investigation of Cause. Mines are encouraged to contact the Environment Canada regional EEM coordinators² for EEM-related questions.

This first chapter provides an overview of the metal mining EEM program, including a decision tree to assist mines in identifying an appropriate path, based on their respective situation, as they move through the EEM program. Additional information and documents are available on the EEM website.

Top of Page

1.2 The Metal Mining Effluent Regulations

The MMER permit the deposit of mine effluent if the effluent pH is within a defined range, if the concentrations of the MMER deleterious substances in the effluent do not exceed authorized limits, and if the effluent is demonstrated to be non-acutely lethal to rainbow trout. These discharge limits were established to be minimum national standards based on best available technology economically achievable at the time that the MMER were promulgated. To assess the adequacy of the effluent regulations for protecting the aquatic environment, the MMER include EEM requirements to evaluate the potential effects of effluents on fish, fish habitat and the use of fisheries resources.

Regulations Amending the MMER were published in the Canada Gazette, Part II, in October 2006. The purpose of these amendments was to clarify the regulatory requirements by addressing matters, related to the interpretation and clarity of the regulatory text, which had emerged from the implementation of the Regulations. Additional amendments also occurred at later dates and on several occasions; however, these did not pertain to the EEM requirements.

Additional amendments to the MMER were published in the Canada Gazette, Part II, in February 2012. The following changes were made to improve the EEM provisions of the MMER:

• modifications to the definition of an “effect on fish tissue” in order to be consistent with the Health Canada fish consumption guidelines and to clarify that the concentration of total mercury in tissue of fish from the exposure area must be statistically different from and higher than its concentration in fish tissue from the reference area;
• addition of selenium and electrical conductivity to the list of parameters required for effluent characterization and water quality monitoring;
• exemption for mines, other than uranium mines, from monitoring radium 226 as part of the water quality monitoring, if 10 consecutive test results showed that radium 226 levels are less than 10% of the authorized monthly mean concentration (see subsection 13(2) of the Regulations);
• change to the time frame for the submission of interpretative reports for mines with effects on the fish population, fish tissue and benthic invertebrate community from 24 to 36 months;
• change to the time frame for the submission of interpretative reports for magnitude and geographic extent of effects and for investigation of cause of effects from 24 to 36 months; and
• minor changes to the wording for consistency within Schedule 5.

Top of Page

1.3 Description of Environmental Effects Monitoring Studies

EEM studies are designed to detect and measure changes in aquatic ecosystems (i.e., receiving environments). The metal mining EEM program is an iterative system of monitoring and interpretation phases that is used to assess the effectiveness of environmental management measures, by evaluating the effects of effluents on fish, fish habitat and the use of fisheries resources by humans.

EEM goes beyond the end-of-pipe measurement of chemicals in effluent to examine the effectiveness of environmental protection measures directly in aquatic ecosystems. Long‑term effects are assessed using regular cyclical monitoring and interpretation phases designed to investigate the impacts on the same parameters and locations. In this way, both a spatial and temporal characterization of potential effects to assess changes in receiving environments are obtained.

EEM studies consist of:

• effluent and water quality monitoring studies comprising effluent characterization, sublethal toxicity testing and water quality monitoring (MMER, Schedule 5, Part 1); and
• biological monitoring studies in the aquatic receiving environment to determine if mine effluent is having an effect on fish, fish habitat or the use of fisheries resources (MMER, Schedule 5, Part 2).

Top of Page

1.3.1 Effluent and Water Quality Monitoring Studies

1.3.1.1 Effluent Characterization

Effluent characterization is conducted by analyzing a sample of effluent and recording the hardness, electrical conductivity and alkalinity, as well as the concentrations of aluminum, cadmium, iron, molybdenum, selenium, ammonia and nitrate (MMER, Schedule 5, subsection 4(1)). Mercury in effluent is also analyzed and recorded but may be discontinued if the concentration is less than 0.10 µg/L in 12 consecutive samples (MMER, Schedule 5, subsection 4(3)). Guidance on effluent characterization can be found in Chapter 5. Other parameters relevant to EEM, such as arsenic, copper, cyanide, lead, nickel, zinc, total suspended solids and radium 226, are also analyzed as part of Schedule 4.

Top of Page

1.3.1.2 Sublethal Toxicity Testing

Sublethal toxicity testing is conducted on effluent from the mine’s final discharge point that has potentially the most adverse environmental impact (MMER, Schedule 5, subsection 5(2)). This testing monitors effluent quality by measuring survival, growth and/or reproduction endpoints in marine or freshwater organisms in a controlled laboratory environment. In the case of effluent deposited into marine or estuarine waters, sublethal toxicity testing is conducted on a fish species, an invertebrate species and an algal species. In the case of effluent deposited into freshwater, sublethal toxicity testing is conducted on a fish species, an invertebrate species, a plant species and an algal species (MMER, Schedule 5, subsection 5(1)). Guidance to determine the appropriate final discharge point to sample can be found in Chapter 2. Guidance on sublethal toxicity testing can be found in Chapter 6.

Top of Page

1.3.1.3 Water Quality Monitoring

Samples for water quality monitoring are collected from the exposure area surrounding the point of entry of the effluent into water from each final discharge point and the related reference areas, as well as from the sampling areas selected for the biological monitoring study (MMER, Schedule 5, subsection 7(1)). Water temperature and dissolved oxygen concentrations are recorded for all samples. As for effluent characterization, the concentrations of aluminum, cadmium, iron, molybdenum, selenium, ammonia and nitrate are measured and recorded in water quality monitoring. Mercury in the water quality monitoring samples is also analyzed and recorded if required for effluent characterization (MMER, Schedule 5, subsection 4(3)). In the case of effluent deposited into freshwater, the pH, hardness, electrical conductivity and alkalinity are recorded. In the case of effluent deposited into estuarine waters, the pH, hardness, electrical conductivity, alkalinity and salinity are recorded. In the case of effluent deposited into marine waters, the salinity is recorded. The concentrations of the following deleterious substances set out in Schedule 4 are also recorded: arsenic, copper, cyanide (if used as a process reagent), lead, nickel, zinc, total suspended solids and radium 226 (unless the conditions specified in subsection 13(2) of the Regulations are met) (MMER, Schedule 5, paragraph 7(1)(d)). Guidance on water quality monitoring can be found in Chapter 5.

Top of Page

1.3.2 Biological Monitoring Studies

Biological monitoring studies are conducted in 36- or 72-month phases. The requirements for each study are dependent on the results of studies from the previous phase(s). Biological monitoring studies to assess effects are described in section 1.3.2.3 and biological monitoring studies to investigate effects are described in section 1.3.2.4.

To assess effects, biological monitoring studies are conducted for the following three components (MMER, Schedule 5, section 9):

• fish population to assess effects on fish health;
• benthic invertebrate community to assess fish habitat or fish food; and
• mercury concentration in fish tissue to assess the human usability of the fisheries resources, in terms of fish consumption.

To investigate effects, biological monitoring studies are conducted for the purpose of:

• assessing the magnitude and geographic extent of effects; and
• determining the cause(s) of effects.

Top of Page

1.3.2.1 Defining and Confirming Effects

The studies on the fish population, fish tissue, and benthic invertebrate community are conducted in both exposure and reference areas. The exposure area refers to all fish habitat and waters frequented by fish that are exposed to effluent,and the reference area refers to water frequented by fish that is not exposed to effluent and that has fish habitat that, as far as is practical, is most similar to that of the exposure area (MMER, Schedule 5, section 1).

The MMER defines effects on the fish population, fish tissue, and benthic invertebrate community (MMER, Schedule 5, section 1) and further prescribes the data assessment required for specific indicators (MMER, Schedule 5, section 16). An “effect” on the fish population or benthic invertebrate community is defined as a statistical difference between data collected in an exposure area and in a reference area or sampling areas within an exposure area where there are gradually decreasing effluent concentrations at increasing distances from the effluent discharge. An effect on fish tissue refers to concentrations of total mercury, exceeding 0.5 micrograms per gram (µg/g) wet weight in fish tissue taken in an exposure area and that are statistically different from and higher than the concentrations of total mercury in fish tissue taken in a reference area. Chapter 8 provides information on conducting statistical analyses on EEM data.

Data collected on specific‑effect endpoints (Tables 1-1 and 1-2) are assessed to determine if statistical differences are present in order to establish if there are any effects on the indicators. To confirm that observed effects are not artifacts (or due to confounding factors) and are mine‑related, biological monitoring studies to assess effects are repeated in a subsequent three‑year phase. If a similar type of effect (same endpoint in same direction from zero relative to reference levels) on the fish population, fish tissue or benthic invertebrate community is determined in studies from two consecutive phases, the effect is considered confirmed (MMER, Schedule 5, section 19). Confirmation of an effect for fish endpoints need not be limited to the same sex or same species, unless site-specific conditions warrant a different approach.

If effects are confirmed in one or more components (fish population, fish tissue, benthic invertebrate community), the mine must investigate those effects in subsequent phases (section 1.3.2.3). All confirmed effects must be investigated. If the lack of effects is confirmed in all three components, a mine must proceed to a reduced biological monitoring frequency (MMER, Schedule 5, paragraph 22(2)(b)).

Attributing cause of an effect to a mine’s effluent may be difficult in some circumstances. Environment Canada recommends that where the previous study has determined there is an effect, and there is doubt that the effect is caused by the mine effluent, the second study confirming the effect be designed in a way that maximizes the confidence in establishing that the effect is or is not mine‑effluent‑related. Adjustments to the study design to eliminate confounding factors are described in the other chapters and could include increased sampling effort in both reference and exposure areas; increase or change in sampling areas; or the use of alternative studies, such as mesocosms or caged bivalves.

Top of Page

1.3.2.2 Historical Information

Mines may have conducted biological monitoring studies prior to becoming subject to the MMER. These studies may be used as part of the EEM program if they determine whether the effluent was causing an effect on fish population, fish tissue or the benthic invertebrate community. However, if the mine operation or environmental conditions changed or any event which may have modified biological effects occurred after the historical monitoring was conducted, then any historical data should be used with caution when interpreting currently observed effects. The results of the historical biological monitoring studies must be submitted to the Authorization Officer³ along with a report that contains scientific data to support the results, not later than 12 months after the day on which the mine becomes subject to the Regulations (MMER, Schedule 5, paragraph 14(b)). Refer to sections 1.4.2 and 1.4.5 for requirements on timelines for submission of study design and interpretative reports for mines using historical information. Further details on historical information are provided in Chapter 13.

Top of Page

1.3.2.3 Biological Monitoring Studies to Assess Effects

To assess effects, biological monitoring studies are conducted for the three components: fish population, fish tissue (mercury concentration) and benthic invertebrate community.

Top of Page

1.3.2.3.1 Fish Population Survey

A fish population survey (Chapter 3) measures indicators of fish population health in exposure and reference areas, or along an exposure gradient, to determine if mine effluent has an effect on fish. A fish survey is required if the concentration of effluent in the exposure area is greater than 1% at a distance of 250 metres from the final discharge point (Schedule 5, paragraph 9(b)).

The MMER defines the fish population survey effect indicators as growth, reproduction, condition and survival (MMER, Schedule 5, subparagraph 16(a)(i)). When conducting a standard adult fish survey, the collection of adult males and females of two sentinel species is recommended. Data collected on the specific effect endpoints listed in Table 1-1 are evaluated to determine if statistical differences in the effect indicators are present.

Although the standard fish survey is recommended above, other survey designs, modified methods such as a non-lethal fish survey (Chapter 3) or alternative methods (Chapter 9) may be considered under conditions where the standard survey is not effective or practical.

Top of Page

1.3.2.3.2 Benthic Invertebrate Community Survey

Mines must conduct a benthic invertebrate community survey (Chapter 4) to determine if their effluent has an effect on fish habitat. Benthic invertebrates are collected to determine if there are changes in the effect indicators between exposure and reference areas or along an effluent concentration gradient. Data collected on the specific effect endpoints listed in Table 1-2 are evaluated to determine if statistical differences in the effect indicators are present (Schedule 5, subparagraph 16(a)(iii)). See Chapter 4 for definitions and other details on benthic invertebrate community endpoints.

If the designs in Chapter 4 are not effective or practical, an alternative survey may be appropriate (Chapter 9).

Top of Page

1.3.2.3.3 Fish Tissue Survey

A fish tissue survey (Chapter 3, section 3.11) is conducted to assess if mercury from mining effluent may affect the use of the fisheries resources. A survey respecting the fish tissue is required if, during effluent characterization, the concentration of total mercury in the effluent is equal to or greater than 0.10 µg/L (MMER, Schedule 5, paragraph 9(c)).

Top of Page

1.3.2.4 Biological Monitoring Studies to Investigate Effects

To investigate effects, mines assess the magnitude and geographic extent of all confirmed effects and investigate their causes.

Top of Page

1.3.2.4.1 Magnitude and Geographic Extent

When the results of the two previous biological monitoring studies indicate a similar type of effect (same endpoint, same direction from zero) on the fish population, fish tissue or the benthic invertebrate community, an assessment of the magnitude and geographic extent of the effect is required (MMER, Schedule 5, paragraph 19(1)(d)). Magnitude and geographic extent must be assessed for all confirmed effects. The assessment of the magnitude and geographic extent may require additional monitoring efforts to extend the sampling area further downstream, or the necessary information may already exist as part of previous study results. Guidance on magnitude and geographic extent studies can be found in Chapters 2, 4 and 7.

Top of Page

1.3.2.4.2 Investigation of Cause

If the results of the previous biological monitoring study indicate the magnitude and geographic extent of an effect on the fish population, fish tissue or benthic invertebrate community, an investigation of cause (IOC) study is required (MMER, Schedule 5, subsection 19(2)). The goal of an IOC study is to determine the cause of each confirmed effect. Guidance on IOC studies can be found in Chapter 12.

Top of Page

1.4 Steps in Conducting and Reporting Environmental Effects Monitoring Studies

Conducting and reporting EEM studies, as per the MMER, involves the following key steps:

• Conduct and submit results for effluent characterization, sublethal toxicity testing and water quality monitoring
• Submit study design
• Conduct biological monitoring study
• Conduct data assessment
• Submit interpretative report

Top of Page

1.4.1 Conduct and Submit Results for Effluent Characterization, Sublethal Toxicity Testing and Water Quality Monitoring

Effluent characterization is conducted four times per calendar year and not less than one month apart, with the first characterization carried out not later than six months after the day on which the mine becomes subject to the MMER (Schedule 5, subsection 4(2)). Effluent characterization is conducted on an aliquot of effluent collected for the analysis of deleterious substances under Schedule 4. Refer to Chapter 5 for more information on effluent characterization.

Sublethal toxicity tests are conducted two times each calendar year for three years, and once each year after the third year. Sublethal toxicity tests are conducted on an aliquot of effluent collected for the analysis of deleterious substances under Schedule 4. The first testing is to occur not later than six months after the mine becomes subject to the Regulations (MMER, Schedule 5, subsection 6(1)). Sublethal toxicity testing may be conducted once each calendar year, if the results of six sublethal toxicity tests conducted (after December 31, 1997), on a fish species, an invertebrate species and either an aquatic plant species or an algal species, are submitted to the Authorization Officer not later than six months after the mine becomes subject to the Regulations (MMER, Schedule 5, subsection 6(2)). Refer to Chapter 6 for more information on sublethal toxicity testing.

Water quality monitoring is conducted, starting not later than six months after the day on which the mine becomes subject to the Regulations, four times per calendar year, on samples collected not less than one month apart, while the mine is depositing effluent. Water quality monitoring is also conducted on samples collected at the same time that the biological monitoring studies are conducted (MMER, Schedule 5, subsection 7(2)). Refer to Chapter 5 for more information on water quality monitoring.

An annual report on the effluent and water quality monitoring studies conducted during a calendar year is submitted to an Authorization Officer not later than March 31 of the following year (MMER, Schedule 5, section 8). Most of the annual effluent and water quality monitoring reporting requirements may be met by submitting the data results electronically to Environment Canada using the “Regulatory Information Submission System” (RISS) provided on the following website: https://www.riss-sitdr.ec.gc.ca/riss/Global/Index.aspx. For the reporting requirements that are not supported by the RISS, a hard copy submission is required to be submitted to Environment Canada also not later than March 31 of the following year. These requirements include the methodologies used to conduct effluent characterization, sublethal toxicity testing and water quality monitoring, as well as the quality assurance and quality control measures implemented and data related to the implementation of those measures.

Top of Page

1.4.2 Submit Study Design

The study design describes how the biological monitoring study will be conducted to meet the regulatory requirements (MMER, Schedule 5, sections 10 and 19). This guidance document is intended as a starting point for study designs and allows for flexibility in the design of studies to accommodate site-specific needs. Various examples of potential study designs are presented in Chapter 4 (see also Chapters 2, 3, 9 and 12 for information related to study designs). When multiple mines discharge to the same drainage basin, joint EEM studies are encouraged, where practical.

The first study design is submitted not later than 12 months after the day on which the mine becomes subject to the Regulations (MMER, Schedule 5, paragraph 14(a)) or not later than 24 months after the day on which the mine becomes subject to the MMER for mines submitting historical information (MMER, Schedule 5, paragraph 14(b)). The study design for the first, second or subsequent biological monitoring study is submitted to the Authorization Officer at least six months before the biological monitoring study is conducted (MMER, Schedule 5, subsections 15(1) and 19(1)). For mines that have applied to become recognized closed mines, the final study design is submitted not later than 6 months after providing the notice informing of the intention to become a recognized closed mine (MMER, Schedule 5, subsection 23(1)).

A mine could be conducting different types of studies, such as a standard fish survey and a magnitude and geographic extent study for the benthic invertebrate community, at the same time. The study design would then describe how these two studies would be conducted.

The information to be provided in the study design depends on the type of biological monitoring study to be conducted.

Top of Page

1.4.2.1 Study Design for Biological Monitoring Studies to Assess Effects

In cases where effects have not been assessed or confirmed, where the most recent interpretative report indicates the cause of the effect or where the 2 most recent interpretative reports indicate no effects, the designs for biological monitoring studies shall include (MMER, Schedule 5, section 11; guidance in Chapter 2):

• a site characterization that describes effluent mixing in the exposure area and a measure of the effluent concentration at 250 metres from the final discharge point;
• descriptions of the exposure and reference area habitat;
• the type of production process and the environmental protection practices at the mine;
• a summary of any federal, provincial or other laws applicable at the mine regarding effluent and environmental monitoring; and
• a description of any anthropogenic, natural and other factors not related to the effluent that may reasonably be expected to contribute to any observed effect.

Also included is the scientific rationale for selecting the fish species, sampling areas, sample size, sampling periods, and field and laboratory methodologies, as well as the methodology for determining whether the effluent has an effect on the fish population, fish tissue or benthic invertebrate community. Descriptions of the quality assurance and quality control measures that will be implemented to ensure validity of the data collected must also be included as must the summaries of results from previous biological monitoring studies.

Where available, historical data may provide useful information for site characterization and assist in developing EEM study designs, using lessons learned in previous monitoring. If historical information was submitted, the first study design must include a summary of the results of biological monitoring studies completed before the mine became subject to the Regulations.

Top of Page

1.4.2.2 Study Design for Biological Monitoring to Investigate Effects

If the results of the two previous studies indicate a similar type of effect (same endpoint in same direction from zero relative to reference levels) on the fish population, fish tissue or the benthic invertebrate community, the study design shall include, in addition to the information detailed in section 1.4.2.1, a description of one or more additional sampling areas within the exposure area that shall be used to assess the magnitude and geographic extent of the effect (MMER, Schedule 5, paragraph 19(1)(d)).

If the results of the previous biological monitoring study indicate the magnitude and geographic extent of an effect, the study design shall include a detailed description of the field and laboratory studies that will be used to determine the cause of the effect (MMER, Schedule 5, subsection 19(2)).

Top of Page

1.4.3 Conduct Biological Monitoring Study

The biological monitoring study is conducted according to the submitted study design. If circumstances arise that make it impossible to follow the study design, the owner or operator of the mine must inform the Authorization Officer without delay of the circumstances requiring deviation from the study design and of how the study will be conducted (MMER, Schedule 5, subsections 15(2) and 24(2)). It is recommended that the mine’s environmental personnel or consultants also notify the Environment Canada regional EEM coordinator of any deviation from the study design.

Top of Page

1.4.4 Conduct Data Assessment

After completing the fieldwork, data assessment and interpretation are conducted to determine if mine effluent is causing an effect or effects (MMER, Schedule 5, section 16). Data assessment and interpretation also determine future monitoring requirements. The specific analyses conducted to determine if there are effects on fish population, fish tissue or the benthic invertebrate community are described in Chapter 8. Data assessment for mines that have confirmed effects entails determining the magnitude and geographic extent of the effect(s) and assessing cause(s) of any confirmed effect(s). Guidance on IOC studies can be found in Chapter 12.

Top of Page

1.4.5 Submit Interpretative Report

The first interpretative report is submitted not later than 30 months after the date on which the mine becomes subject to the Regulations or not later than 42 months after the date on which it becomes subject to the Regulations, if the mine has submitted a report utilizing historical biological monitoring information (MMER, Schedule 5, section 18).

Subsequent interpretative reports are submitted 36 or 72 months after the day on which the most recent interpretative report was required to be submitted, depending on the results of the previous interpretative report.

The supporting data from biological monitoring studies are submitted to Environment Canada in the electronic format provided on the EEM website (see Chapter 10 for further information).

The MMER outline the information to be contained in interpretative reports for biological monitoring studies (MMER, Schedule 5, sections 17, 21 and 25). Chapter 10 describes interpretative reports in more detail. Brief descriptions of the different types of interpretative reports are provided below.

Top of Page

1.4.5.1 Interpretative Report for Biological Monitoring Studies to Assess Effects

Interpretative reports for biological monitoring studies to assess effects include, among other items, results of monitoring studies, raw data, results of data assessments, and identification of any effects.

Top of Page

1.4.5.2 Interpretative Report for Biological Monitoring Studies to Investigate Effects

If the magnitude and geographic extent of a confirmed effect on fish population, fish tissue or benthic invertebrate community is not known, then the interpretative report shall include, among other items, the results of a magnitude and geographic extent study. If the magnitude and geographic extent of the confirmed effect is known but the cause of the effect(s) is not known, the interpretative report shall include a description of the cause of the effect. The IOC interpretative report contains the study results and statement identifying the cause of the effect on fish population, fish tissue and/or benthic invertebrate community. If the cause of the effect(s) was not determined, an explanation of why and a description of any steps that must be taken in the next study to determine the cause shall be included in the interpretative report.

Top of Page

1.5 Recognized Closed Mines

An owner or operator of a mine that has ceased operation, and who intends to have that mine become a recognized closed mine, shall provide written notice of that intention to the Authorization Officer and shall maintain the mine’s rate of production at less than 10% of its design-rated capacity for a continuous period of three years starting on the day that the written notice is received by the Authorization Officer. A final biological monitoring study must be conducted during the three-year period (MMER, section 32). The final study design shall be submitted to the Authorization Officer not later than six months after the closure notice is provided (MMER, Schedule 5, section 23). The mine shall base the final monitoring phase on the results of the previous biological monitoring study. The final interpretative report shall be submitted to the Authorization Officer not later than 36 months after the day on which the notice to close the mine was provided (MMER, Schedule 5, section 26). Effluent characterization, sublethal toxicity testing and water quality monitoring requirements continue until the mine becomes a recognized closed mine.

1.6 Identifying a Path through the Metal Mining Environmental Effects Monitoring Program

The metal mining EEM program involves monitoring to assess effects, investigate confirmed effects (magnitude and geographic extent and IOC), and reassess effects. When an effect has been confirmed (i.e., similar type of effect in two consecutive studies), the mine is required to assess the magnitude and geographic extent of the effect (MMER, Schedule 5, paragraph 19(1)(d)), and then to investigate the cause of the effect (MMER, Schedule 5, subsection 19(2)).

Top of Page

1.6.1 Critical Effect Sizes

A critical effect size (CES) is a threshold above which an effect may be indicative of a higher risk to the environment. The Metal Mining EEM Review Team recommended that CESs be established for each of the metal mining EEM effect endpoints following the second national assessment of the EEM data from metal mines (Metal Mining EEM Review Team, 2007).

CESs for the fish population and benthic invertebrate community endpoints were initially developed for the pulp and paper EEM program after EEM data showed that most mills observed an effect in at least one of the effect indicators. Once validated, these CESs were adopted for use in the metal mining EEM program (Table 1-3).

¹ Differences in fish population effect endpoints are expressed as percentage (%) of reference mean, while differences in benthic effect endpoints are expressed as multiples of within-reference-area standard deviations (SDs).
² Problems associated with determining the age of some species of fish should be discussed and reviewed before effects on weight-at-age and age are used to choose a path through the EEM program. Refer to Chapter 3 for recommendations on age determination.

Top of Page

1.6.2 Magnitude of Confirmed Effects

The magnitude of each effect observed on fish population, fish tissue or benthic invertebrate community can be further evaluated to determine if the magnitude of a confirmed effect is above or below the CES. Table 1-4 outlines how effects from two consecutive studies are to be grouped to determine if confirmed effects are below or above the CES.

Top of Page

1.6.3 Decision Process for the Metal Mining Environmental Effects Monitoring Program

Figure 1-1 is a decision tree to assist mines in identifying an appropriate path through the EEM program, based on their respective situation. CESs are applied to EEM results to assist mines in identifying the level of effort for investigations of confirmed effects. The structure of the decision tree is based on the MMER regulatory requirements, recent scientific knowledge and the experience and knowledge gained through implementing the EEM program.

Site‑specific knowledge as well as effluent and water quality data need to be considered before identifying a mine’s path through the EEM program. Confirmed effects in supporting endpoints are used as part of the site-specific evaluations to support decisions regarding a path forward (see chapters 3 and 4 for information on supporting endpoints).

Mines are required to continue conducting effluent and water quality monitoring and reporting the results using the timeline prescribed in the MMER and as outlined in section 1.4.1 of this chapter. This requirement is independent of the timeline for conducting biological monitoring studies and submitting interpretative reports.

Figure 1-1: Decision tree for the metal mining EEM program

Details can be found below

Figure 1-1 is a flow chart which describes the decision-making process through the different phases of monitoring of metal mining effluents and the timing of submission of interpretative reports. The top portion of the flow chart comprises the biological monitoring studies to assess effects and the bottom part of the flow chart comprises the biological monitoring studies to investigate confirmed effects. Based on one’s answer to a question in the diagram, one will be prompted by arrows to answer a related question. Follow-up questions differ based on the answer to the previous questions.

Top of Page

1.6.3.1 Assessing Effects

The interpretative report of the second and any subsequent biological monitoring study is submitted no later than 36 months after the day on which the interpretative report of the previous biological monitoring study was required to be submitted, under the following scenarios:

No effects observed

• The results of a single study indicate no effects on fish population, fish tissue and the benthic invertebrate community (MMER, Schedule 5, subsection 22(1)).

Effects observed

• The results of a single study indicate an effect on fish population, fish tissue or benthic invertebrate community (MMER, Schedule 5, subsection 22(1)).
• The results of a single study indicate an effect on fish population, fish tissue and benthic invertebrate community (MMER, Schedule 5, paragraph 22(2)(a)).

The interpretative report is submitted not later than 72 months after the day on which the interpretative report of the previous study was required to be submitted, under the following scenario:

No effects confirmed

• The results of the previous two consecutive biological monitoring studies indicate no effect on fish population, fish tissue and the benthic invertebrate community (MMER, Schedule 5, paragraph 22(2)(b)).

For the purpose of determining the timing of the submission of interpretative reports, if a study respecting the fish population is not required because of the concentration of effluent in the exposure area, as per Schedule 5, paragraph 9(b), then the effluent is considered to have no effect on the fish population. Similarly, if a study respecting fish tissue is not required because of the concentrations of mercury in the effluent as per Schedule 5, paragraph 9(c), then the effluent is considered to have no effect on fish tissue.

Top of Page

1.6.3.2 Investigating Confirmed Effects

If the results of the previous two consecutive biological monitoring studies indicate a similar type of effect (same endpoint in same direction from zero relative to reference levels) on fish population, fish tissue or the benthic invertebrate community, and if the magnitude or geographic extent of the effect or cause of the effect is not known, then the interpretative report is submitted not later than 36 months after the day on which the interpretative report of the previous study was required to be submitted (MMER, Schedule 5, paragraph 22(2)(c)).

Top of Page

1.6.3.2.1 Level of Effort for Investigating Effects

Mines are required to investigate all confirmed effects. The following paragraphs provide recommendations as to how the confirmed effects can be investigated depending on the magnitude of the effects (below or above CESs).

Confirmed effects of magnitude greater than or equal to CESs
Mines with confirmed effects of a magnitude greater than or equal to CESs (Table 1-4) would conduct a field study to assess magnitude and geographic extent of the effects and submit the next interpretative report in 36 months. Subsequently, the mine would conduct field and/or laboratory studies to determine the cause(s) of the effects and submit the IOC interpretative report in another 36 months. If the magnitude and geographic extent of the effect has already been determined, the mine may move directly to determining the cause(s) of the effects. In this case, the mine could report the magnitude and geographic extent of the effects in the IOC study design.

Confirmed effects of a magnitude below the CESs
If a confirmed effect has a magnitude below the CES, it is not expected that larger effects be observed farther away from the final discharge point. The mine could therefore assess the magnitude and geographic extent of a confirmed effect below the CES by providing a scientifically sound rationale using the results and other existing information from studies, and then move directly to determining the cause(s) of the effects. Under this scenario, if the mine uses existing information to assess the magnitude and geographic extent of the effects, it is recommended that this information be reported in the IOC study design and that the next interpretative report be submitted in 36 months. The cause(s) of the effect could be determined by conducting field and/or laboratory studies or by examination and presentation of solid evidence using existing data, alone or in combination with new data and/or a literature review.

Once the cause(s) of the effects has (have) been determined, the next interpretative report is submitted 36 months after the day on which the most recent interpretative report was required to be submitted. In this case, the study design must describe biological monitoring studies to assess effects (see section 1.4.2.1).

Top of Page

1.6.3.3 Timing of Studies to Assess Effects and Magnitude and Geographic Extent and to Investigate Cause

There are different stages in assessing and investigating effects (Figure 1-1). In many cases, the process of assessing and investigating effects may not move together (concurrently) for the different components (fish population, fish tissue and benthic invertebrate community). Once an effect has been confirmed, mines are required to assess the magnitude and geographic extent of the effect and determine the cause of the effect. The magnitude and geographic extent and IOC studies are required for all confirmed effects.

While conducting a study to assess the magnitude and geographic extent of an effect observed in one component (fish population, fish tissue, and benthic invertebrate community), mines are also required to continue monitoring the component(s) where effects were not previously observed or confirmed. Therefore a mine may conduct a study to confirm an effect in one component, or lack thereof, while conducting a study to determine the magnitude and geographic extent of an effect in another component.

While conducting an IOC study for an effect confirmed in one component (fish, fish tissue, benthic invertebrates), mines do not need to conduct simultaneous studies on component(s) where effects were not confirmed. If effects were confirmed in more than one component or for more than one endpoint within a component, then all cause(s) of all confirmed effects need to be determined during the next phase or, if not possible, promptly over subsequent phases.

Top of Page

1.7 References

Metal Mining Environmental Effects Monitoring Review Team. 2007. Metal Mining Environmental Effects Monitoring Review Team Report. National EEM Office, Environment Canada, Gatineau, QC.

Figures and Tables 

Table 1-1 outlines the effect indicators and effect endpoints in a fish population survey. Effect indicators include growth, reproduction, condition, and survival. Effect endpoints include size-at-age, relative fish gonad size, condition, relative liver size, and age. Data collected on the specific effect endpoints are evaluated to determine if statistical differences in the effect indicators are present.

Return to the table

Table 1-2 outlines the effect indicators and effect endpoints in a benthic invertebrate community survey. Effect indicators include total benthic invertebrate community density, evenness index, taxa richness, and similarity index. Effect endpoints include number of animals per unit area, Simpson’s evenness, number of taxa, and Bray-Curtis index. Data collected on the specific effect endpoints are evaluated to determine if statistical differences in the effect indicators are present.

Return to the table

Table 1-3 outlines the critical effect sizes for the metal mining environmental effects monitoring program. Fish effect endpoints and benthic effect endpoints are aligned with their respective critical effect sizes. Fish effect endpoints include weight-at-age, relative fish gonad size, relative liver size, condition, and age. Benthic effect endpoints include density, Simpson’s evenness, taxa richness, and the Bray-Curtis index.

Return to the table

Table 1-4 provides an evaluation of the magnitude of confirmed effects in two consecutive phases. The table outlines how confirmed effects from two consecutive studies are to be grouped. The confirmed effects are separated into two types: effects above or equal to CES, and effects below CES.

Return to the table

1 AQUAMIN. 1996. Assessment of the Aquatic Effects of Mining in Canada. Environment Canada.

2 Contact information for regional EEM coordinators is available on the EEM website: www.ec.gc.ca/eem.

3 The Authorization Officer for each province is described in Schedule 1 of the MMER. Contact information for Authorization Officers is available on the EEM website: http://www.ec.gc.ca/esee-eem/.

2. Study Design, Site Characterization and General Quality Assurance and Quality Control

2.1 Overview

2.2 Study Design and Site Characterization

• 2.2.1 Site Characterization
  • 2.2.1.1 Plume Delineation
  • 2.2.1.2 Measuring of Constituents in the Effluent
  • 2.2.1.3 Habitat Mapping and Classification
  • 2.2.1.4 Aquatic Resource Inventory
  • 2.2.1.5 Classification Scheme for Reference Area Selection
  • 2.2.1.6 Framework for Rivers
  • 2.2.1.7 Framework for Lakes
  • 2.2.1.8 Use of Sublethal Toxicity for Site Characterization
  • 2.2.1.9 Characteristics of Mining Environments
• 2.2.2 Exposure and Reference Areas
  • 2.2.2.1 Selection of Final Discharge Point for Monitoring
  • 2.2.2.2 Selection of Exposure and Reference Areas
    • 2.2.2.2.1 Exposure Area
    • 2.2.2.2.2 Reference Area
• 2.2.3 Reporting of Field Station Positions
• 2.2.4 Modifying or Confounding Factors
• 2.2.5 Tributaries and Other Point- and Nonpoint-Source Discharges
• 2.2.6 Natural Variation in Environmental or Habitat Conditions
• 2.2.7 Historical Damage

2.3 General Quality Assurance / Quality Control and Standard Operating Procedures

• 2.3.1 Quality Assurance and Quality Control
• 2.3.2 Standard Operating Procedures

2.4 References

Table

• Table 2-1: Site characterization information for preparing an EEM study design

2. Study Design, Site Characterization and General Quality Assurance and Quality Control

2.1 Overview

This chapter includes information on study design, site characterization, and general quality assurance / quality control (QA/QC) information for the metal mining environmental effects monitoring (EEM) program. The requirements for the study design and site characterization are listed in the Metal Mining Effluent Regulations (MMER) (Schedule 5, sections (s.) 10–14) and Chapter 1. This includes information such as timelines for EEM studies (first studies, studies aim to confirm absence or presence of effect, magnitude and geographic extent, investigation of cause and final biological monitoring studies prior to a mine closing), content of study-design reports and submission dates. Each chapter of this document contains additional information on recommended methodologies for the study design for fish, fish tissue, benthic invertebrates and alternative method studies. In addition, each chapter provides more detailed information on QA/QC.

2.2 Study Design and Site Characterization

The objective of a study design is to describe how the biological monitoring studies (a fish survey, fish tissue analysis and benthic invertebrate community survey) are to be conducted.

Study designs should describe the following (MMER, Schedule 5, s. 10–14):

• a summary of previous biological monitoring and effluent and water quality monitoring;
• information related to site characterization, including the results of plume delineation studies;
• the objectives of the field monitoring program, including overall approach and rationale for biological monitoring, which may be based on previous monitoring results;
• statistical design criteria, hypotheses, statistical methods and data needs;
• a description of how the biological monitoring studies are to be conducted to determine if there are effects, taking confounding influences into consideration;
• field sampling plans, including what will be measured, where and when it will be measured, location of exposure and reference sites, and rationale for selection of final discharge point;
• QA/QC measures that will be taken to ensure validity of data; and
• schedules for field monitoring and submission of the interpretative report.

Top of Page

2.2.1 Site Characterization

Site characterization information is submitted as part of the EEM study design (MMER Schedule 5, s. 10 (a)). The requirements for site characterization are described in Schedule 5, s. 11 of the MMER. Table 2-1 summarizes site characterization information that should be included in the study design. For the first EEM study design, site characterization is included in detail. For subsequent EEM studies the site characterization can be submitted in summary format, but new information (e.g., production rates) should be updated in detail. In most cases, mines will have most site characterization information available from previous assessments and historical studies. If information critical to the design of the EEM study is not available, additional field data may be required to provide adequate background for the first EEM study design, particularly with respect to hydrology and aquatic resources.

Site characterization information is used to identify suitable sampling areas that have similar habitats in the exposure and reference areas, and to obtain information on other discharges and confounding factors that may affect the interpretation of data obtained from those areas. Information on some of the unique environmental characteristics of mine sites that should be taken into consideration during the site characterization can be found in Section 2.2.1.9.

For mines with insufficient historical information to locate reference and exposure areas, exploratory sampling may be useful. Exploratory sampling can also be used to identify habitat characteristics for effective selection of sampling stations.

An experienced field crew should be able to approximate the effluent field based on field measurements of water quality tracers (e.g., specific conductance) or preliminary dye study results, and can often identify likely depositional areas based on observed receiving water flow and circulation patterns. Thus, it is usually possible to choose some appropriate water and sediment sampling stations in the field and to complete exploratory sampling of the receiving environment concurrent with plume and depositional zone studies and critical resource/habitat inventories in a single campaign.

Much of the site characterization information can be effectively reported in map form. Maps should be of sufficient scale (e.g., 1:5000) to show the features of the study area in adequate detail. The actual scale should be reported on any map used. The geographic extent of the study area to be mapped should be determined on a site-specific basis, and should include the discharge point as well as the exposure and reference areas.

The requirements of the site characterization section of the EEM study design are outlined in the MMER. Additional information relevant to the site characterization that should be reported may include the following (including Table 2-1):

1. An identification of the major chemical reagents used in the overall production process since January 1, 1996. Mines are encouraged to report current quantities of reagents used. This should list reagents of the following types:
   • activators
   • flocculants
   • pH modifiers
   • depressants
   • frothers
   • collectors
     
2. An inventory of all discharge points where effluent is deposited into the exposure area. This inventory should identify all known sources of effluent to the aquatic environment, including those regulated under the MMER, as well as any others (e.g., nonpoint sources) that may have the potential to cause an effect on the aquatic environment.
   
3. Information on the local climate, particularly seasonal precipitation patterns.

Top of Page

Top of Page

2.2.1.1 Plume Delineation

A description of the manner in which the effluent mixes within the exposure area, including an estimate of the concentration of effluent in water at 250 m from each final discharge point (MMER, Schedule 5, s. 11 (a)), is to be described in the site characterization. For subsequent biological monitoring studies the site characterization information may be summarized along with, where applicable, a detailed description of any changes to that information since the submission of the most recent study design (MMER, Schedule 5, s. 19 (1) (a)). This description should include an indication of relative flow of the effluent and receiver, as well as seasonal variations in flow. This will give an indication of dilution rate. The description should also give an indication of the density of the effluent, and where within the water column the effluent is likely to be, prior to complete mixing. This estimate may be based on direct measurements in the field or modelling, but it is recommended that modelling be validated with field measurements.

A fish population study is conducted if the concentration of effluent is greater than 1% in the area located within 250 m of a final discharge point (MMER, Schedule 5, s. 9 (b)). If such a study does not need to be conducted due to the effluent concentration being less than 1%, it is recommended that more rigorous plume delineation methods be used to document the effluent concentrations in the exposure area.

It is recommended that the description of the manner in which effluent mixes within the exposure area include the following:

• identification of where in the exposure area the effluent is located, prior to mixing with the receiving water;
• estimation of where in the exposure area the effluent and receiving water begin mixing, and where mixing is complete;
• estimation of the effluent dilution ratio at points downstream of effluent discharge; and
• identification of significant sources of dilution, other than the primary receiver (i.e., tributaries or other streams); and
• how the above vary with the tides and seasons.

For extensive guidance on plume delineation, please consult the Revised Technical Guidance on How to Conduct Effluent Plume Delineation Studies, available from Environment Canada (2003) at www.ec.gc.ca/esee-eem/D450E00E-61E4-4219-B27F-88B4117D19DC/PlumeDelineationEn.pdf. This document was prepared for the pulp and paper EEM program but can also be applied to the metal mining EEM program. Additional plume delineation information pertaining to metal mining is discussed below.

Top of Page

2.2.1.2 Measuring of Constituents in the Effluent

Conductivity Surveys

If the conductivity of effluent is consistent during the period of study, a conductivity survey can be used to locate the effluent plume within the exposure area. Survey results can be assessed semi-quantitatively, or conductivity measurements can be converted into relative effluent concentrations lying between 1 (effluent) and zero (background) by applying the following formula:

C_r = (C_a – C_b)/(C_e – C_b)

The relative concentration is an expression of the dilution ratio. Temperature readings need to be taken along with the measurements obtained from the conductivity meter, since conductivity rises approximately 2% for every 1°C rise in temperature. Further information on theoretical considerations for effluent plume delineation using conductivity can be found in relevant reference material (e.g., Fischer et al. 1979; Freeze and Cherry 1979).

Although conductivity surveys can provide valid and cost-effective estimates of the location of effluent within the receiving environment, the natural variability of conductivity in surface waters can interfere with locating the edge of the effluent plume. Such natural variability in conductivity may be observed with depth as well as with surface measurements. The presence of multiple tributaries or receiving water bodies can further exacerbate this difficulty.

Tracing by Effluent Metals

The location of the effluent plume within the exposure area may also be estimated by choosing a reference parameter present in the effluent and tracking its fate in time and space by measuring its concentrations in water samples taken at specific locations. The selection of such a tracer needs to be based on its stability and consistency of concentration, as well as on representativeness and ease of measurement. Since the parameter selected has to be a conservative substance, metals such as copper or nickel may serve as a “natural” tracer. Sulphate is often a good tracer of base metal mine effluents, particularly in massive sulphide deposits.

It should be cautioned that the effluent parameter selected for tracing may sometimes be present at the same order of magnitude in receiving waters. This would rule out its application for determining effluent dilution. Other parameters may be specific to the effluent alone, but present in such low concentrations that it makes their detection difficult. Several parameters may be present at significantly greater levels in effluent, which would make them ideal tracers for dilution measurement. However, as a result of complications such as the costs of analysis, instability of the substance, difficulty in measurement, or the lack of an adequate in situ measuring apparatus, these parameters may not always prove to be appropriate “natural” tracers. The potential for effluent metals to be used as tracers for plume delineation should therefore be evaluated on a case-specific basis.

Top of Page

2.2.1.3 Habitat Mapping and Classification

Some elements of habitat mapping and classification, as well as aquatic resource inventory, are included as part of site characterization. More detailed habitat mapping may be helpful in identifying habitat types present in the exposure and reference areas. This section provides guidance on habitat mapping and classification.

The recommended method to create a habitat map is to perform a habitat classification. The recommended framework for classifying aquatic features is the classification system developed by the U.S. Fish and Wildlife Service, Classification of Wetlands and Deepwater Habitats of the United States (Cowardin et al. 1979; Busch and Sly 1992). This system allows for classification of a wide range of continental, aquatic and semi-aquatic habitats. Cowardin et al. (1979) also provide guidance on habitat description for coastal and estuarine situations.

Classification systems for marine shorelines to deep coastal areas include Frith et al. (1993), Booth et al. (1996), Robinson and Levings (1995), Hay et al. (1996) and Robinson et al. (1996). Specifically, estuarine classification has been reviewed by Matthews (1993), Scott and Jones (1995), Finlayson and van der Valk (1995) and Levings and Thom (1994). In the United States, the most widely used system is that of Cowardin et al. (1979) and Cowardin and Golet (1995), with expansions proposed by other authors.

Listed below are examples of environment-specific conditions for various habitats:

Rivers: It is recommended that river habitat descriptions include information on elevation gradient; the location of dams, falls and other barriers to fish migration; mean annual discharge and ranges; and general substrate characteristics of each river (preferably in the form of a gradient profile chart). Upstream and downstream inputs (e.g., storm water, sewer overflow, effluent from other industrial sites) should be mapped and described.

Lakes: Important habitat features of lakes include bathymetry, the locations of major inlets and outlets, and general oxygen-temperature conditions (e.g., thermal stratification, occurrences of oxygen depletion in deep water).

Open coastlines: Suggested additional mapping parameters for open coastlines (marine, Great Lakes) include depth contours, nearshore substrate characteristics, shoreline configuration, and the locations of inflowing rivers and other discharges and activities.

Estuaries: Estuaries are best described in terms of their general salinity gradients, flows, bathymetries and general substrate features. A description of tidal cycles is recommended for all marine and estuary locations. Most of the above features can be described from navigational maps, topographic maps, government publications on tides and river discharge records, and through interviews with local government officials and knowledgeable individuals.

Natural wetlands: A wetland is defined as land that is saturated with water long enough to promote wetland or aquatic processes as indicated by poorly drained soils, hydrophytic vegetation and various kinds of biological activities that are adapted to a wet environment (Metal Mining EEM Review Team 2007). Wetlands include bogs, fens, marshes, swamps and shallow waters (usually two metres deep or less) (Metal Mining EEM Review Team 2007). During the Metal Mining Program Review (2007) the review team recommended that natural wetlands for EEM studies should be avoided. Where a mine final effluent flows into a natural wetland area, EEM studies should be conducted downstream of the wetland when studies upstream are not possible. This recommendation is consistent with the Federal Policy on Wetland Conservation. This policy is found at the website.

It is recommended that bottom substrates be described. Further guidance on aquatic habitat assessment can also be found in the Department of Fisheries and Oceans and the British Columbia Ministry of the Environment and Parks (1987), Orth (1989), Ontario Ministry of Natural Resources (1989), Plafkin et al. (1989), and the Department of Fisheries and Oceans (1990).

Depositional zones in the exposure area should be identified and illustrated on the habitat map. Any information on sediment characterization (chemistry, toxicity) should be reported. Depositional zones occur where water velocity decreases, resulting in particles settling out; the finest particles settle out in the slowest current speeds. Historical contaminant or benthic invertebrate community data may be helpful in identifying sampling stations within a depositional exposure area. To compare resident benthic invertebrate communities, similar (but uncontaminated) sediment depositional zones should be located in the reference area. In situations where historical contamination was from a source other than the mine, two reference areas could be used; one with and one without the historically contaminated sediment.

Top of Page

2.2.1.4 Aquatic Resource Inventory

An aquatic resources inventory includes the identification of fish and shellfish (resident and transient) that are presently being fished commercially and non-commercially (both sport [including stocked fish] and subsistence fishery). The inventory should make particular note of fish species that may be present in sufficient numbers to be considered as a sentinel species, and of utilization (e.g., spawning, nursery) of the exposure area by fish species. In addition, any species recognized by federal, provincial or territorial authorities as rare, threatened or endangered should also be included. The Committee on the Status of Endangered Wildlife in Canada website (www.cosewic.gc.ca); district fisheries biologists in federal, provincial or territorial regulatory or museum agencies; local conservation officials; and members of the local community (fishermen, Aboriginal people and public interest groups) are all sources for this type of information.

The potential success of field programs increases with familiarity of the study area. It is recommended that fieldwork be undertaken to verify historical information if this information is not detailed or recent.

Stocked fish are not appropriate for EEM-type monitoring, as these fish are predominately sport fish and are not appropriate indicator species since their growth and reproduction may be altered depending on how and when they were stocked and raised. As well, stocked fish generally have no apparent reproductive success; therefore, this effect indicator could not be evaluated.

Top of Page

2.2.1.5 Classification Scheme for Reference Area Selection

Because reference areas will vary among different landscapes, approaches have been developed to classify land through which rivers run or in which lakes reside in order to predict aquatic biotic assemblages (Corkum 1989, 1992; Hughes 1995, Maxwell et al. 1995; Omernik 1995). A classification system is a way of simplifying sampling procedures and management strategies by organizing a variable landscape (Conquest et al. 1994). The assumption is that the classification scheme is hierarchical. The advantage of a hierarchical classification scheme is that it “offers a way to discriminate among features of the landscape at several scales of resolution” (Conquest et al. 1994). The classification scheme is based (with modifications) on one developed by the U.S. Department of Agriculture’s Forest Service (Maxwell et al. 1995). The hierarchical classification scheme is presented as a guide in the a prioriselection of sampling areas.

Habitat-Specific Allocation of Reference and Exposure Areas

The following specific points should be considered during the selection of reference and exposure areas and/or stations:

For Rivers:

• The size of the drainage basin selected is based on stream order. For example, if a mine site is located on a second-order stream, the drainage basin area is delineated at the point the stream becomes third-order (i.e., at the junction of two second-order streams).
• If there are no upstream inputs or confounding factors, reference area(s) can be within the drainage basin and upstream of the mine.
• If confounding factors, such as nonpoint- or point-source inputs, occur upstream of the effluent, the reference area(s) can be selected in nearby drainage basins with comparable habitat features (Figure 4-4).
• If physical disturbance of the river valley is associated with the mine, effluent effects may be confounded by the disturbance. Accordingly, reference areas should be selected to match the physical disturbance, if possible.
• The following features should be similar between reference and exposure areas: ecoregion, drainage basin area, stream order, bankfull width, channel gradient, channel pattern, habitat types, water depth, water velocity, substratum composition, riparian vegetation, shoreline structure and land use, etc.

For Lakes:

• In lakes with a single-mine effluent and without nonpoint sources of pollution, the sphere of influence originating from the effluent should be determined. This is particularly important for lakes in which effluent flow is not unidirectional.
• If effluent plume delineation and former studies indicate that mine effects are likely to be local and restricted, select reference areas within the lake in which the mine discharge occurs. These reference areas should occur in separate but comparable bays or basins of the lake.
• If effluent plume delineation indicates that the identified effluent is dispersed throughout the lake, select reference area(s) in the nearest comparable lake within the same or adjacent drainage basin.
• If nonpoint- or other point-source inputs occur elsewhere on a lake, select reference area(s) in the nearest comparable lakes within the same or adjacent drainage basin.
• If the mine effluent is associated with physical disturbance in the area, effluent effects may be confounded by the disturbance. Accordingly, physically matched reference areas should be selected, if possible.
• The following features should be similar between reference and exposure areas: ecoregion, geological origin, drainage basin area, morphometry, slope from shoreline, habitat types, substratum composition, riparian vegetation, shoreline structure and land use, etc.

For Marine Environments:

• The reference area should be within the same water body and hydrographic current or tidal regime as the exposure area. In other words, the closer the reference area is to the exposure area, the better. Benthic invertebrate communities in marine ecosystems are considerably higher in species richness, have more complex trophic relationships, faunal size ranges and reproductive strategies, than benthic invertebrate communities in freshwater ecosystems. Because of this complexity, and the multitude of interactions between species in marine benthic invertebrates, small shifts in physical or chemical conditions can dramatically alter the overall benthic faunal community. Add to this the effect of increasing variation in chance larval settlement with increasing geographic distance (geographic “drift” in community structure) and physical barriers in complex coastlines, and it is very rare to find similar invertebrate communities from one bay or fjord to the next, and very difficult to predict specific benthic community structure based on sediment factors (for recent review on marine invertebrate sediment interactions, see Snelgrove and Butman 1994)). In order to have some confidence that the “natural” benthic community is similar enough from one coastal area to the next, there should be sufficient water exchange between them. This is more likely in open coastal areas than in isolated bays and fjords.
• Reference areas that are not in the same water body or hydrographic regime may only be suitable for comparisons of summary characters such as shifts in abundance or species richness. If the habitat conditions are similar enough to the exposure area, it may also be possible to compare larger-scale biotic factors such as the presence of characteristic, long-lived depth/substrate specific taxa described by Thorson (1957) as “parallel communities”.
• Reference and exposure areas should have very similar habitat type, shoreline structure (steep, mountainous, delta, marsh, etc.), bottom topography (sills, sandbars, exposure to open oceanic influences, etc.), substrate type (particle size, sorting, natural chemistry), depth properties, current regimes, physical water properties, nutrient regimes, confounding inputs and drainage characteristics.
• Some special considerations are important for determining the suitability of reference areas for marine and estuarine mines. Physical factors in the estuarine/marine environment that tend to be more complex than in freshwater are salinity (including seasonal freshwater influence), tides (and tidal currents) and sediment sulphides. Other important physical factors include ice-scour or buildup, freezing, water column stagnation due to large summer freshwater runoff, re-suspension due to surface freezing in winter, dams or log-booms, extraordinary siltation or clogging from logging, and periodic flooding.
• In addition to the above important characteristics, the following specific points should be similar between reference and exposure areas:
  • intertidal areas: shoreline slope, wave exposure, light and tidal exposures, shoreline vegetation, and encrusting fauna (although the latter may be part of the benthic taxa being monitored for a response to mine effluent).
  • sub-tidal areas: seasonal water column stability and bottom oxygen depletion (stagnation).

Ecoregions

The first step in reference site selection is to use terrestrial attributes (ecoregions) with similar features. Ecoregions are defined as “part of an ecoprovince characterized by distinctive ecological responses to climate as expressed by vegetation, soils, water, and fauna” (Wiken 1986; Wickware and Rubec 1989). Ecoregion maps for Canada are available here.

Drainage Basin and Geographic Scale

Catchments or drainage basins have clear hydrographic boundaries. A drainage basin is defined as the area that has a common outlet for its surface runoff. Although inter-basin transfer occurs among biota, the geoclimatic history of large basins (1:2 000 000) are known to create barriers to dispersal through hydrographic divides and climate (Maxwell et al. 1995). It is essential to establish the geographic scale appropriate to the study design. In large-scale, synoptic surveys in which relationships are sought between landscape features and aquatic biota, the mapping scale for drainage basins is 1:250 000 (Corkum 1989, 1992, 1996; Reynoldson and Rosenberg 1996). These basins are subdivided into progressively smaller sub-basins.

Land/water interactions with respect to sediment and nutrient transport off the land and from upstream sources is integral in developing predictive models that link environmental variables and associated biota. Drainage basins may occur within ecoregions or may cross different ecoregions. Aquatic fauna are more similar to one another in drainages that occupy the same ecoregion than in drainages that occupy different ecoregions (Corkum 1992; Hughes et al. 1994).

Land Use and Vegetative Buffer Strips

Although ecoregions are defined in terms of climate and natural vegetation, natural vegetation is disturbed with human development. Land-use type is a simple measure of disturbance within the drainage basin. If there is a change in land use (e.g., land clearing for agricultural uses or logging, or fire), the biological assemblages in receiving waters will respond to those changes (Corkum 1992, 1996). Accordingly, site selection should be in drainages with comparable land use.

The degree (width and type) of a vegetated buffer strip adjacent to rivers and lakes should be recorded at all sampling areas. In reference areas where human disturbance cannot be avoided, the effect of a vegetated buffer strip moderates temperature fluctuations through shading (Budd et al. 1987), removes or reduces sediment from runoff (Young et al. 1980), and regulates nutrients and metals entering the water body (Peterjohn and Correll 1984).

Top of Page

2.2.1.6 Framework for Rivers

The river sampling design provides a framework for characterizing habitats at multiple scales (Meador et al. 1993). The framework for rivers is based on how they are organized in hierarchical space and how they change through time (Frissell et al. 1986). The riverine system has several hierarchical or nested levels: drainage or catchment basin, valley segment, stream reach and channel unit (Conquest et al. 1994).

Valley Segment and Stream Order

Valley segments are distinctive sections of drainage basins that possess geomorphic properties and hydrological transport characteristics that distinguish them from other segments (Cupp 1989). Montgomery and Buffington (1993) identified three valley segment types: colluvial (channelized and unchannelized), alluvial and bedrock. Valley segments can be filled with colluvium (sediment and organic matter from landslides) or alluvium (sediment transported by flow). The third valley segment has little soil and is dominated by bedrock.

Valley segments are distinguished by six criteria (Conquest et al. 1994):

1. Stream order (position in drainage network)
2. Valley slide-slope gradient
3. Ratio of valley bottom width to active channel width
4. Channel gradient
5. Stream-corridor geomorphic surface deposits
6. Channel pattern.

Channel segments are assigned stream orders (Strahler 1957) for a particular map scale or aerial photograph (e.g., 1:250 000) (Newbury and Gaboury 1993).

Stream Reach

Stream reaches consist of homogeneous associations of topographic features and channel geomorphic units (Bisson and Montgomery 1996). Stream reaches can be used to predict local stream response to perturbations (Montgomery and Buffington 1993). Stream reaches are useful in assessing habitat quality, aquatic productivity, fish distributions and stream health (Maxwell et al. 1995). Stream reach classification is determined using map scales of 1:12 000 to 1:24 000. Criteria used to classify stream reaches include:

• channel pattern
• channel entrenchment
• channel width
• hydraulic radius
• basin area
• channel material
• stream gradient
• bed form
• riparian vegetation

Simpler approaches have been adopted to identify stream reaches. For example, a straight channel has an undulating bed with alternating riffles and pools spaced at repeating intervals of 5-7 channel widths (Leopold et al. 1964; Leopold 1994). Newbury (1984) defined a stream reach to be equivalent to six times the channel width.

Channel Unit

Channel units are subdivisions of stream reaches that describe uniform microhabitats (depth and flow) and are used to identify factors that limit both invertebrate and fish populations within a stream reach. Hawkins et al. (1993) proposed a three-tiered system of channel units in which the first level distinguishes riffles from pools. The second level identifies turbulent and non-turbulent riffles and distinguishes between pools formed by scour or dams. Dammed pools retain more sediment and organic debris and have more cover than scour pools. The third subdivision identifies microhabitats based on hydraulic processes and structure. Channel units are typically 10 m or less and typically cannot be mapped at a scale appropriate for land management.

Criteria for subdivision of riffles include:

• gradient or water surface profile
• percentage of super-critical flow
• bed roughness
• mean velocity
• step development

Criteria for subdivision of pools include:

• location (main channel or off-channel)
• longitudinal and cross-sectional depth profiles
• substrate characteristics
• pool-forming constraints

Top of Page

2.2.1.7 Framework for Lakes

The geological origin, hydrology and morphometry (obtained from maps and aerial photographs) of lakes are important in identifying sediment-water interactions and productivity of lakes (Wetzel 1975). Although thermal stratification can be predicted from morphological features, field verification is necessary. The mapping scale for lakes is typically 1:24 000 or 1:63 000 (Maxwell et al. 1995).

Origins, Location and Hydrological Linkages

Reference and exposure lakes should have the same origin, location and hydrological linkages. Lake geology ultimately affects the physical, chemical and biological characteristics of water bodies. For example, Hutchinson (1957) identifies 11 types of geomorphic processes (tectonic, volcanic, landslides, glacial activity, solution, fluviatile, wind, shorelines, organic accumulation, anthropogenic and natural dams, meteorite impact). Surface geology and location (altitude, latitude and longitude) affect lake chemistry and thermal regimes (Winter 1977). These variables, which can be obtained from maps, are used to predict the biological composition and productivity of lakes (Dolman 1990; Winter and Woo 1990). Hydrological linkage refers to the “connection of a lake to surface or ground water” and can forecast information about lake biota (Maxwell et al. 1995). Maxwell et al. (1995) describe three types of hydrological linkages: riverine linkage (outlets and/or inlets or unconnected), groundwater linkage (gaining, losing, neutral or no recharge), and water storage regime (perennial or intermittent).

Morphometry

Lake morphometry has been used historically to predict fish yields (Ryder 1965; Kerr and Ryder 1988) and to determine species diversity (Eadie and Keast 1984; Marshall and Ryan 1987). With the exception of depth (and volume), other features can be obtained from maps. Hypsographic (cumulative depth-area or cumulative depth-volume) graphs are useful for comparing basin shapes of lakes and predicting surface area or volume for water-level control of reservoirs. Common morphological features of lakes include surface area, volume, mean and maximum depth, shoreline development, and hydraulic residence time.

Trophic State

Many lake classification systems are based on a measure of productivity (oligotrophy, mesotrophy, eutrophy). A fourth lake type (dystrophy) is used to describe lakes that receive large amounts of organic matter from external sources; these lakes are heavily stained and are known as brown-water lakes. Productivity of dystrophic lakes is low and so some limnologists group dystrophic lakes as a subclass of oligotrophic lakes. The following variables have been used to describe the trophic status of lakes:

• dissolved oxygen
• thermal mixing (lake stratification)
• total phosphorus
• soluble reactive phosphorus
• total nitrogen
• nitrite + nitrate
• ammonium
• chlorophyll a
• transparency
• organic matter

Zone

Lakes are subdivided into an open-water pelagic zone, a shoreline or littoral zone inhabited by autotrophic vegetation, and a deeper benthic region free of vegetation (the profundal zone). The reference and exposure areas should always be located in the same zone.

Top of Page

2.2.1.8 Use of Sublethal Toxicity for Site Characterization

Since historical sublethal toxicity data for some or all of the required tests (see Chapter 6) have been generated for effluents discharged from a number of mine sites across Canada, the operator may want to use this information during site characterization in the following ways:

1. To aid in determining sampling areas for fish or the benthic invertebrate community. If the operator has no historical field data on fish, fish food sources or fish habitat from their effluent exposure zone, historical effluent sublethal toxicity data (if available in sufficient quality) could be used to estimate the potential zone of influence to help in establishing sample collection locations for the fish and benthic invertebrate community surveys in the first biological monitoring (i.e., estimation of extent of response in the high effluent exposure area). Details on how to estimate the geographic extent of a sublethal toxicity response are provided in Chapter 6.
   
2. As an aid in comparing effluent discharge sources. If there are numerous effluent discharge locations at a mine site, one of the recommended sublethal toxicity tests could be used to quantify the degree of sublethal toxicity contributed from the different effluent discharge sources (see chapter 6).

Top of Page

2.2.1.9 Characteristics of Mining Environments

Many mines and mining activities share some similarity in environmental features. They are briefly outlined below:

Headwater location of mine sites: Many mine sites are located near the headwaters of rivers or streams. In some instances the effluent can constitute a significant portion of the flow downstream of the discharge point. This will have an impact on how to characterize the mine’s exposure area. Headwater streams, because of their size, gradient or intermittent flow, are often not suitable fish habitat. Therefore, the mine effluent is frequently discharged to receiving waters with small or no fish populations, although the diluted effluent will, in most cases, eventually reach fish habitat. Nonetheless, some fish species often use accessible headwater areas at some stage of their life cycle, such as for spawning, and this information should be considered when designing an EEM study. Mines may have to move progressively downstream until they reach an area with suitable number and species of fish; however mines should assess fish populations in the immediate receiving environment first.

Effluent quality and volume: Effluent quality and quantity are influenced by several factors including the nature of the ore and host rock, processing methods, effluent treatment methods, climate, and site hydrology. Discharge rates will vary both in volume and duration based on site-specific factors. In Canada, some mining operations discharge seasonally. Reduced natural degradation of substances such as cyanide and ammonia under cold conditions sometimes makes it difficult to achieve discharge limits; therefore, wastewater is discharged during spring and summer. Wastewater may also be discharged during early spring to release large quantities of snow meltwater that has accumulated during the winter months.

The presence of fish in the initial receiving environment can also be a factor in the discharge of effluent volume. For example, the need to protect overwintering fish in pools when natural stream flow is minimal may cause a mine to reduce its effluent discharge to the exposure area. Other mines minimize discharges in late summer when stream flow is low and fish may be spawning. Conversely, spring discharge rates may be high, allowing water inventories to be relieved while stream flows are high. However, not all mines have sufficient reservoir capacity to optimize effluent releases.

Mine ore bodies are variable, resulting in variations in effluent discharges. Each ore is different, not only between mines but also within each mine itself. One ore body may contain more or less of certain minerals. In addition, while mines may have a short life span, the mill at the mine site may process ore from several mines. Virtually every mine has its own distinct suite of parameters of concern in effluents that influence site-specific factors related to bioavailability and hardness. However, similarities in operations among mines in a region may allow for efficiencies to be realized by a regional EEM study design.

Mines incorporate a variety of effluent treatment options (e.g., lime addition, settling ponds, water treatment plants) in their processes, and the type and effectiveness of each treatment option will influence the effluent discharged to the environment. The retention time in tailings ponds can affect the composition of the effluent. For example, cyanide breakdown and settlement of particulate matter are time-dependent. A sufficiently long residence time can potentially modify the concentrations in the effluent.

Local geology: Mine location is determined by the regional geology and the exact locations of ore deposits. Local mineralization around ore deposits influences the natural background levels of metals in streams. The net result is often a naturally higher background concentration of metals in streams located near mine sites, which should be considered during reference area selection such that comparability to the exposure area is optimized.

Bioavailability of metals: The bioavailability of metals is an important aspect to consider when assessing the effects of mine effluent in the EEM program, as ongoing research continues to identify modifying factors that can change their bioavailability. For example, the effects of soluble metals on the biota can be mitigated during effluent treatment (e.g., adding lime to precipitate metals). In addition, the effect of mine effluent and associated sediments on the aquatic environment can vary during the mine’s life cycle as the bioavailability of certain parameters changes.

Top of Page

2.2.2 Exposure and Reference Areas

An area is qualitatively defined for sampling purposes and relates to the appropriate geographical scale encompassing one or more fundamental sampling locations (“stations”). A station is a fixed sampling location that can be recognized, re-sampled and defined quantitatively (e.g., latitude/longitude). Within EEM, the overall study area is subdivided into reference and exposure areas (for control-impact designs) or within an exposure area where there are gradually decreasing effluent concentrations (for gradient designs). The MMER definition of exposure area is “all fish habitat and waters frequented by fish that are exposed to effluent,” and the definition of reference area is “water frequented by fish that is not exposed to effluent and that has fish habitat that, as far as practical, is most similar to that of the exposure area.” (MMER, Schedule 5, s. 1).

2.2.2.1 Selection of Final Discharge Point for Monitoring

In cases where the mine has more than one final discharge point, it is recommended that sampling be done in an exposure area where the effluent has the greatest potential to have an adverse effect on the receiving environment. The mass loadings of the deleterious substances, the manner in which the effluent mixes in the exposure area, and the sensitivity of the receiving environment should all be considered when selecting which final discharge point should be used for biological monitoring.

Top of Page

2.2.2.2 Selection of Exposure and Reference Areas

The selection of the sampling areas is one of the most critical components of the study design and should be considered carefully to maximize the quality of the information gained from the study. The design of biological surveys is site-specific, and various examples of potential study designs are presented in Chapter 4. However, this guidance is not intended to limit the mine’s flexibility to propose other potential study designs that may be suitable to the site.

2.2.2.2.1 Exposure Area

Exposure area sampling should be done in an area proximate to the effluent discharge where effects may be found. Sampling areas should ideally support both appropriate habitat for the benthic invertebrate community and populations of the selected fish species. The study design should also consider the use of the exposure area by fish species (e.g., spawning, nursery). Identification of the exposure area and its habitat features should precede the selection of reference areas, because reference areas will, as far as practicable, match the physical and chemical habitat features of the exposure area (other than the features expected to change due to the effluent).

The exposure area may extend through a number of receiving environments (e.g., different stream orders, lakes or marshes, estuarine to marine, or intertidal to sub-tidal) and contain a variety of habitat types. In most cases, the boundary of the exposure area is defined by the zone of effluent mixing. Within an exposure area, there may be a high effluent and a low effluent exposure area often referred to as near-field and far-field areas, respectively. Additional sampling areas within the exposure area may be used during phases to assess magnitude and geographic extent of effects, or during periodic monitoring to provide an enhanced study or address site-specific needs. High effluent exposure (near-field) areas are outside the initial discharge zone (as described below) and have higher exposure to effluent than far-field areas. The initial discharge zone is the area where the effluent exceeds the velocity of the receiving water and the effluent is buoyant. The initial discharge zone is often characterized by visual turbulence and typically does not extend more than 5-50 m from the outfall. At least one of the high effluent exposure (near-field) stations should be as near as possible to the effluent discharge point but located outside the initial discharge zone. For magnitude and geographic extent studies, the exposure area extends along the effluent gradient so that additional lower effluent exposure areas (far-field areas) are included. The exposure area extends geographically until a return to reference area conditions (regulatory definitions of exposure and reference areas are provided above). Lower effluent exposure areas are recommended to be positioned close to the boundary of the zone of effluent mixing. Multiple sampling stations in each defined area should be used to determine spatial variation. In a gradient design, there is no reference area per se, but the response variables are evaluated along the effluent gradient.

In practical use, there will probably always be one or more low effluent exposure (far-field) areas for media other than fish (e.g., water, sediment and benthos). Recommended positioning of low effluent (far-field) exposure areas should be such that each area differs in regard to degree of effluent exposure. If possible, all exposure areas should be located so as to minimize or avoid exposure to non-mine discharges.

Top of Page

2.2.2.2.2 Reference Area

Reference areas do not need to represent pristine (pre-European settlement) conditions, but, rather, can comprise areas in which anthropogenic impacts, unrelated to the mine effluent, are similar to exposure areas (Simon 1991; Omernik 1995).Where feasible, the reference area should be located in the same water body as the effluent discharge, upstream of or beyond any influence from the discharge. The reference area should be suitable physically and biologically, and outside the influence of the mine or other confounding factors. When a mine is located at a headwater, and/or where no suitable reference area on the same water body is available (e.g., dams and reservoirs may be upstream), the reference area should be located in an adjacent water body with similar characteristics or a non-impacted tributary to the receiving water body. Another possibility is sampling a number of exposure areas at increasing distances from the discharge point, representing an exposure gradient (gradient design). More than one reference area may be used, where appropriate. During magnitude and geographic extent studies, it may be necessary to sample more than one reference area if multiple exposure areas with different habitat types are sampled. As well, a more regional approach could be considered, particularly for benthic invertebrate community surveys, such as looking at several non-impacted streams (or lakes) in the area (i.e., reference condition approach).

Where historical monitoring data exist, the mine should consider using the same sampling areas from previous studies, provided they are appropriate for use in the EEM program. This will help ensure that monitoring data collected as part of the EEM program may be compared with historical data.

Baseline data (pre-effluent discharge) and multiple reference areas may assist in data interpretation. It is possible to use historical data as a baseline comparison to determine effects, but it should be treated as data additional to the mine’s own. The mine’s design should therefore include both a reference area and an exposure area (or follow a gradient design). This ensures that reference conditions have not changed and that changes observed are not incorrectly attributed to the mine, because there can be changes in parameters related to changes in environmental conditions (e.g., due to flooding or variability in annual temperatures). A reference area should be used to allow characterization of those changes that are mine-related compared to those that are not. Practitioners can also take advantage of the environmental assessment phase for new projects to provide information complimentary to the EEM program (Kilgour et al. 2007).

Where possible, sampling areas for different components (fish, benthic invertebrate community, water) should be co-located. The characteristics of the selected fish species, (e.g., mobility, habitat usage) and the different sampling gear may not always make this practical. The reference areas for benthic invertebrate sampling in some cases can be directly upstream of the exposure area, which may not be the case with fish (due to mobility). In addition, mines are encouraged to conduct benthic invertebrate community and fish monitoring studies concurrently, if justifiable biologically (e.g., if the ideal time for sampling fish reproductive parameters coincides with a suitable time for benthic community sampling; see Chapter 3 for additional information on timing for fish reproduction). Where there is more than one mine in close proximity and effluents are discharged to the same drainage basin, joint EEM studies are encouraged. Where studies are proposed jointly, sampling areas may be shared.

Data obtained from reference areas when compared to exposure areas can detect impairment of aquatic life (Yoder 1991), diagnose stressors (Hughes et al. 1994), provide data on temporal and spatial trends (Yoder 1989) and provide data for water resource summaries for government agencies (OEPA 1990). The identification of “least impacted” areas will differ across the country. Reference areas in extremely disturbed areas may be impossible to locate. Here, studies should be designed so that reference areas with minimal degradation are located in comparable drainage basins within the same ecoregion (Hughes et al. 1994).

Usually, coastal mines do not have strictly “upstream” areas for reference sampling, because of variable directionality of current due to tidal effects. Estuarine mines may have upstream areas that are too different physically and biologically to be suitable for reference sampling. The reference area is therefore usually at least periodically downstream from the effluent discharge. Accordingly, it is important to understand the current flow patterns in the area to determine whether a potential reference area is “outside” mine effluent exposure.

Selection of a reference area in the initial phase should not necessarily dictate its use in future phases.

In selecting sampling areas, the mine should take the following factors into consideration:

• the location of sampling areas in previous surveys;
• the location of confounding influences;
• the size of area needed to accommodate the number of samples to be collected;
• habitat type;
• site access; and
• other issues that could affect the mobility of fish.

In general, both sampling areas should be as follows:

• As similar as possible except for exposure to effluent. Although the two areas are unlikely to be identical, it is assumed that the differences in natural characteristics (e.g., depth, substrate, flow and water quality) will, other than mine-related factors, be small relative to the potential effect associated with the presence of effluent. If this is not the case, it should become apparent, and study design changes should be made in subsequent phases.
• Situated as closely as possible to each other (but sufficiently distant to be confident that fish from the reference area are not exposed to effluent).
• Accessible, and offer safe sampling during the most appropriate season (i.e., when measurements on fish growth, reproduction, condition and survival can be taken).
• Described in as much detail as possible, including the latitude and longitude as well as a written description of the area (physical, chemical and biological habitat, including measurements of temperature, depth and flow).

At a minimum, for a control/impact study, sampling should be conducted at no less than 1 reference area and 1 exposure area during the first EEM study, and subsequent EEM studies (studies aim to confirm presence or absence of effect and magnitude and geographic extent). The use of multiple reference areas offers the greatest statistical power to detect a meaningful difference between a reference area and an exposure area (Foran and Ferenc 1999). It can also give an indication of variability among reference areas (Munkittrick et al. 2000). Differences found in the exposure area that are outside the range of values seen at a number of reference areas are more likely to be ecologically relevant (Munkittrick et al. 2000). Sampling multiple reference areas is preferred over increasing sample size (e.g., number of fish) at a single area (Environment Canada 1997).

When possible there are advantages to selecting similar sampling areas for benthic and fish sampling so that the data can be used to help interpret responses. However, optimal benthic sampling areas may be inappropriate for the fish survey because of the characteristics of the fish species selected, the mobility of the fish, different habitat selection, and the type of sampling gear required. The sampling areas may be the same in many circumstances, but this should not be a sufficient criterion in itself for selecting the fish sampling area.

Top of Page

2.2.3 Reporting of Field Station Positions

The interpretative report shall contain the latitude and longitude of sampling areas in degrees, minutes and seconds; a description of the sampling areas sufficient to identify them is required (MMER Schedule 5, s. 17(b)). The latitude and longitude coordinates can be obtained using a variety of methods. Global positioning systems are a common tool for locating the position of field stations and are recommended for this purpose. In some instances, the coordinates with stream-wise distances (e.g., river kilometres) may be useful. The recommended positioning accuracy should be determined on a site-specific basis. In some cases, where there are multiple outfalls, industrial sites may decide to collaborate on their studies.

Additional stations may be included to better represent spatial patterns in a large zone of effluent mixing, such as at a location with transects (right, mid, left), and reference/high effluent exposure area (near-field) /low effluent exposure areas (far-field) sampling areas.

Top of Page

2.2.4 Modifying or Confounding Factors

Modifying or confounding factors can alter the interpretation of the results of biological monitoring. If the sampling areas are fairly similar, effects of modifying factors can be considered negligible. However, when there are significant differences among sampling areas, the survey design becomes confounded. In this case, it may be difficult to differentiate the effects of mine effluent from the effects of the modifying factor(s) on the response variable. For example, if the habitat type (e.g., pool) downstream of the mine was different from the habitat found upstream (e.g., riffle), the effects of habitat type on variables would confound any effects related to mine effluent; both mine effluent and habitat type may be good predictors of any differences in variables observed upstream and downstream of the effluent discharge.

Incorporating multiple reference locations into the study design can aid in designing against spatial confounding factors, and practitioners are encouraged to do so. Design considerations for the detection of anthropogenic disturbances have been presented in the literature (see Green 1993 and references therein; Underwood 1994; Underwood 1997), and practitioners are encouraged to incorporate these considerations into their study designs.

Some examples of potential confounding factors include:

• tributaries and other point- and nonpoint-source discharges (e.g., other industrial discharges, agricultural runoff, aquaculture facilities, sewage treatment plants);
• natural environmental/habitat variables; and
• historical damage.

Attention to potential confounding factors identified during site characterization should be considered during the study design. In this way confounding factors can be minimized, or accounted for in the study design so that their influence can be assessed during data interpretation.

Top of Page

2.2.5 Tributaries and Other Point- and Nonpoint-Source Discharges

Tributaries provide dilution water to a main channel, lake, estuary or open ocean. This dilution water may or may not have similar chemical properties to the water body under study. Tributaries also require time and distance to mix with the water body under study. Therefore, if there is a tributary between the reference and exposure area, the additional flow from the tributary can potentially confound the interpretation of data.

Other point- and nonpoint-source discharges may make it difficult to distinguish between effects due to the mine effluent and other discharges, particularly if they are found within close proximity to the mine effluent discharge. When other discharges are present immediately above the mine, multiple reference areas should be used. One reference area should be set between the other discharge and the effluent discharge. In this way it may be possible to account for the influence of the other discharge. As well, the reference area should have similar background levels of metals. If no differences between the two reference areas are found, they can be pooled to compare against the exposure area.

If there are other point-source effluent discharges not related to mines in the study area, the study design should attempt to minimize the potential effects of the confounding factors. When it is not possible to resolve the confounding factors by modifying the study design, alternative sampling designs and methods should be evaluated.

Top of Page

2.2.6 Natural Variation in Environmental or Habitat Conditions

Natural biological communities can differ both temporally and spatially. Particularly if study areas are extensive, it is possible that natural biological communities and characteristics will be different from one location to another. It may be difficult to distinguish the influences of mine effluent, if any, relative to natural variation.

Examples of common, naturally occurring and sometimes confounding factors include:

• habitat type (riffle, run, pool);
• substrate type (organic content, particle size);
• water depth;
• water flow rate/discharge;
• tidal action / currents / wave exposure;
• salinity;
• dissolved oxygen/temperature;
• emergent/submergent vegetation cover;
• water chemistry (conductivity, hardness, pH, etc.); and
• biological properties.

Once present in the study design, confounding effects cannot be eliminated. Only by giving careful attention to potential modifying factors identified during the pre-design phase or the previous phases of the survey can the influence of modifying factors be removed or controlled in subsequent phases. Where it is not possible to eliminate confounding factors, increasing the number of sampling areas or including additional chemical and/or biological parameters may allow the investigator to assess their influence on data interpretation.

When it is not possible to resolve the confounding factors by modifying the study design, alternative sampling designs and methods (Chapter 9) should be considered.

Top of Page

2.2.7 Historical Damage

When the area in which a mine releases its effluent has been subject to damage from previous activities, it may be difficult to determine differences due to current effluent-release practices. The use of an alternative method may have to be used in these situations.

2.3 General Quality Assurance / Quality Control and Standard Operating Procedures

2.3.1 Quality Assurance and Quality Control

Detailed QA/QC is described in each chapter. QA/QC is a documented system incorporating adequate review, audit and internal quality control. The objective of a QA/QC program is to ensure that all field sampling and laboratory analyses produce technically sound and scientifically defensible results.

QA is a planned system of operations and procedures, the purpose of which is to provide assurance to the client that defined standards of quality are being met. Analytical QA defines the way in which tasks are to be performed in order to ensure that data meet predefined data quality goals. These tasks include not only the analysis itself but all aspects of sample handling and data management.

QA encompasses a wide range of internal and external management and technical practices designed to ensure data of known quality commensurate with the intended use of the data. External QA activities include participation in relevant inter-laboratory comparisons and audits by outside agencies. Outside audits may be based on performance in analysis of standard reference materials, or on general review of practices as indicated by documentation of sampling, analytical and QA/QC procedures, test results, and supporting data. QC is an internal aspect of QA. It includes the techniques used to measure and assess data quality and the remedial actions to be taken when data quality objectives (DQOs) are not realized. Within the context of a particular study, assurance of adequate data quality is only possible when DQOs have been defined. Users of the data should play a lead role in defining DQOs for a study and in ascertaining whether laboratory quality-control limits are consistent with these objectives.

Data quality measures should be defined in the same terms as DQOs, so that the two can be compared in project evaluation. DQOs are normally derived from intended data uses (e.g., hypotheses to be tested, summary statistics involved, and total uncertainty that can be tolerated). Total uncertainty includes imprecision (sampling, analytical, environmental) plus any analytical bias that may occur (Taylor 1987). Objectives can be established for each component and for total uncertainty, and should be incorporated into the QA project plans. The various components of imprecision can be estimated using field replicate data and laboratory replicate data.

QA functions, the personnel responsible for each QA function, and corrective actions when performance limits are exceeded should be identified in the quality management plan.

QC activities define the boundaries of acceptable performance for the measurement system, and include the routine checks (data quality measures) that indicate whether the system is performing to specification. Data reporting generally stops and corrective actions are initiated when the system goes out of control. Range and average-control charting methods have been described elsewhere (OMOE 1984; ASTM 1985, 1986; Dux 1986).

An outline of recommended QA/QC requirements for specific components of the fish survey (Chapter 3), benthic invertebrate survey (Chapter 4), effluent and water quality analysis (Chapter 5), and sediment quality analysis (Chapter 7) are presented in each respective chapter. This information focuses on QC in the field, laboratory, data analysis, and reporting.

Top of Page

2.3.2 Standard Operating Procedures

Standard operating procedures (SOPs) are fundamental to any QA/QC program. All field and lab procedures should be conducted according to SOPs to ensure quality control. SOPs should describe the following in detail:

• the field program’s requirement for sampling methods and procedures, sample handling, labelling, equipment, preserving, record keeping, and shipping; and
• the analytical methods and procedures, sample handling, labelling, equipment, test system implementation, record keeping and so forth of all laboratory analyses.

Each SOP should be a written, detailed method accessible to each analyst. SOPs should be based on procedures developed by a standard-setting organization such as Environment Canada, the U.S. Environmental Protection Agency, the American Society for Testing and Materials, or the American Public Health Association. Where methods are not well validated, it is recommended that the SOP be thoroughly referenced to the relevant literature and contain all the elements outlined in CALA (1991). In-house validation of data should be appended to the SOP and should contain the QA/QC procedures, including the types and frequencies of QC samples to be analyzed, expected levels of precision, accuracy and recovery, and the method detection limits.

While chemical analysis procedures tend to be reasonably well documented, sampling procedures in general, and sampling design in particular, are often overlooked. Sampling error is usually a large component and often the dominant component of uncertainty in environmental measurements. SOPs that include field operations will help to reduce this uncertainty or at least ensure that it is quantified. All field staff should be familiar with the SOPs for any field survey work.

Emphasis should be placed on measures to prevent inadvertent contamination of samples and to ensure sample integrity. In addition, SOPs should specify the proper preparation of all sampling gear and supplies, and the proper calibration of all instrumentation (such as meters).

Top of Page

2.4 References

[ASTM] American Society for Testing and Materials. 1985. Standard practice for establishing conditions for laboratory sensory evaluation of foods and beverages. Philadelphia (PA): American Society for Testing and Materials. ASTM E480-84.

[ASTM] American Society for Testing and Materials. 1986. Physical requirement guidelines for sensory evaluation laboratories. Philadelphia (PA): American Society for Testing and Materials. ASTM STP 913.

Bisson PA, Montgomery DR. 1996. Valley segments, stream reaches, and channel units. In Hauer FR, Lamberti GA, editors. Methods in stream ecology. San Diego (CA): Academic Press. p. 23-52.

Booth J, Hay D, Truscott J. 1996. Standard methods for sampling resources and habitats in coastal subtidal regions of British Columbia: Part I - Review of mapping with preliminary recommendations. Canadian Technical Report of Fisheries and Aquatic Science 2118.

Budd WW, Cohen PL, Saunders PR, Steiner FR. 1987. Stream corridor management in the Pacific Northwest: 1. Determination of stream-corridor widths. Environ Manag 11:587-597.

Busch W-DN, Sly PG, editors. 1992. The development of an aquatic habitat classification system for lakes. Boca Raton (FL): CRC Press.

[CALA] Canadian Association for Laboratory Accreditation: formerly CAEAL (Canadian Association of Environmental Analytical Laboratories). 1991. Code of practice and QA manual for laboratory analysis of sewage treatment effluent in support of the MISA Program; Draft report prepared for CAEAL and the Ontario Ministry of the Environment by Zenon Environmental Laboratories.

Conquest LL, Ralph SC, Naiman RJ. 1994. Implementation of large-scale stream monitoring efforts: sampling design and data analysis issues. In Loeb SL, Spacie A, editors. Biological monitoring of aquatic systems. Boca Raton (FL): Lewis Publ. p. 69-90.

Corkum LD. 1989. Patterns of benthic invertebrate assemblages in rivers of northwestern North America. Freshwat Biol 21:191-205.

Corkum LD. 1992. Spatial distributional patterns of macroinvertebrates along rivers within and among biomes. Hydrobiologia 239:101-114.

Corkum LD. 1996. Responses of chlorophyll a, organic matter and macroinvertebrates to nutrient additions in rivers flowing through agricultural and forested land. Arch Hydrobiol 136:391-411.

Cowardin LM, Carter V, Golet FC, LaRoe ET. 1979. Classification of wetlands and deepwater habitats of the United States. U.S. Fish and Wildlife Service. FWS/OBS-79/31.

Cowardin LM, Golet CF. 1995. US Fish and Wildlife Service 1979 wetland classification: a review. Vegetatio 118:139-152.

Cupp CE. 1989. Valley segment type classification for forested lands of Washington. Timber, Fish & Wildlife AM-89-001.

Department of Fisheries and Oceans. 1990. Coastal/estuarine habitat description and assessment manual. Part II. Habitat description procedures. Coquitlam (BC): Prepared by G.L. Williams and Associates Ltd.

Department of Fisheries and Oceans and British Columbia Ministry of the Environment and Parks. 1987. Fish Habitat Inventory and Information Program: Stream Survey Field Guide.

Dolman WB. 1990. Classification of Texas reservoirs in relation to limnology and fish community associations. Trans Am Fish Soc 119:511-520.

Dux JP. 1986. Handbook of quality assurance for analytical chemistry laboratory. New York (NY): Van Nostrand Reinhold Co.

Eadie JM, Keast A. 1984. Resource heterogeneity and fish species diversity in lakes. Can J Zool 62:1689-1695.

Environment Canada. 2003. Revised technical guidance on how to conduct effluent plume delineation studies. Available from: www.ec.gc.ca/esee-eem/D450E00E-61E4-4219-B27F-88B4117D19DC/PlumeDelineationEn.pdf

Environment Canada. 1997. Fish Survey Expert Working Group report. EEM/1997/6.

Finlayson CM, van der Valk AG. 1995. Wetland classification and inventory. A summary. Vegetatio 118(1-2)185-192.

Fischer HB, List EJ, Koh RCY, Imberger J, Brooks NH. 1979. Mixing in inland and coastal waters. San Diego (CA): Academic Press Inc.

Foran J, Ferenc S. 1999. Multiple stressors in ecological risk and impact assessment. Pensacola (FL): SETAC Press.

Freeze RA, Cherry JA. 1979. Groundwater. Englewood Cliffs (NJ): Prentice-Hall.

Frissell CA, Liss WL, Warren CE, Hurley MD. 1986. A hierarchical framework for stream habitat classification: Viewing streams in a watershed context. Environ Manag 10:199-214.

Frith HR, Seraring G, Wainwright P, Harper H, Emmett B. 1993. Review of habitat classification systems and an assessment of their suitability to coastal B.C. Unpub. report to Environment Canada from L.G.L. Ltd., Sidney (BC).

Green RH. 1993. Application of repeated measures designs in environmental impact and monitoring studies. Australian J Ecol 18:81-98.

Hay DE, Waters RD, Boxwell, editors. 1996. Proceedings Marine Ecosystem Monitoring Network Workshop, Nanaimo, B.C. March 28-30. 1995. Canadian Technical Report of Fisheries and Aquatic Science 2108.

Hawkins CP, Kershner JL, Bisson PA, Bryant MD, Decker LM, Gregory SV, McCullough DA, Overton CK, Reeves GH, et al. 1993. A hierarchical approach to classifying stream habitat features. Fisheries 18:3-12.

Hughes RM. 1995. Defining acceptable biological status by comparing with reference conditions.. InDavis WS, Simon TP, editors. Biological assessment and criteria: tools for water resource planning and decision making. Boca Raton (FL): Lewis Publishers. p. 31-47.

Hughes RM, Heiskary SA, Matthews WJ, Yoder CO. 1994. Use of ecoregions in biological monitoring. In Loeb SL, Spacie A, editors. Biological monitoring of aquatic systems. Boca Raton (FL): Lewis Publ. p. 125-149.

Hutchinson GE. 1957. A treatise on limnology. Vol I. Geology, physics, and chemistry. New York (NY): John Wiley & Sons Inc.

Kerr SR, Ryder RA. 1988. The applicability of fish yield indices in freshwater and marine ecosystems. Limnol Oceanogr 33:973-981.

Kilgour BW, Dubé MG, Hedley K, Portt CB, Munkittrick KR. 2007. Aquatic environmental effects monitoring guidance for environmental assessment practitioners. Environ Monit Assess 130:423-436.

Leopold LB. 1994. A view of the river. Cambridge (MA): Harvard University Press.

Leopold LB, Wolman GM, Miller JP. 1964. Fluvial processes in geomorphology. San Francisco (CA): W.H. Freeman & Co.

Levings CD, Thom RM. 1994. Habitat changes in the Georgia Basin: Implications for resource management and restoration. In Wilson RCH, Beamish RJ, Aitkens F, Bell J, editors. Review of the marine environment and biota of Strait of Georgia, Puget Sound and Juan de Fuca Strait: Proceedings of the BC/Washington Symposium on the Marine Environment, Jan 13,14 1994. Canadian Technical Report of Fisheries and Aquatic Science 1948. p. 330-351.

Marshall TR, Ryan PA. 1987. Abundance and community attributes of fishes relative to environmental gradients. Can J Fish Aquat Sci 44:196-215.

Matthews GWT. 1993. The Ramsar Convention: its history and development. Glan (CH): Ramsar Convention Bureau.

Maxwell JR, Edwards CJ, Jensen ME, Paustian SJ, Parrott H, Hill DM. 1995. A hierarchical framework of aquatic ecological units in North America (Nearctic Zone). St. Paul (MN): U.S. Department of Agriculture, Forest Service, North Central Forest Experimental Station. Gen. Tech. Rep. NC-176.

Meador MR, Hupp CR, Cuffney TF, Gurtz ME. 1993. Methods for characterizing stream habitat as part of the national water-quality assessment program. Raleigh (NC): U.S. Geological Survey. Open-File Report 93-408.

Metal Mining EEM Review Team. 2007. Report of the Metal Mining EEM Review Team. Available from: http://www.ec.gc.ca/Publications/default.asp?lang=En&xml=933E74AE-6C80-4A54-9CAB-62E5A7440114.

Montgomery DR, Buffington JM. 1993. Channel classification, prediction of channel response, and assessment of channel condition. Olympia (WA): Department of Natural Resources. Washington State Timber/Fish/Wildlife Agreement. Report TFW-SH10-93-002.

Munkittrick KR, McMaster M, Van Der Kraak G, Portt C, Gibbons W, Farwell A, Gray M. 2000. Development of methods for effects-based cumulative effects assessment using fish populations: Moose River Project. Pensacola (FL): SETAC Press.

Newbury RW. 1984. Hydrological determinants of aquatic insect habitats. InResh VH, Rosenberg DM, editors. The ecology of aquatic insects. New York (NY): Praeger. p. 323-357.

Newbury RW, Gaboury MN. 1993. Stream analysis and fish habitat design. A field manual. Newbury Hydraulics Ltd., The Manitoba Habitat Heritage Corporation, Manitoba Fisheries Branch.

[OEPA] Ohio Environmental Protection Agency. 1990. Ohio water resource inventory. Columbus (OH): Ohio Environmental Protection Agency.

Omernik JM. 1995. Ecoregions: a spatial framework for environmental management. In Davis WS, Simon TP, editors. Biological assessment and criteria. Tools for water resource planning and decision making. Boca Raton (FL): Lewis Publishers. p. 49-62.

Ontario Ministry of Natural Resources. 1989. Manual of Instructions. Aquatic Habitat Inventory Surveys. Toronto (ON): Ontario Ministry of Natural Resources.

[OMOE] Ontario Ministry of the Environment. 1984. Principles of control charting. King DE, Ronan RC, editors. Laboratory Services Branch, Data Quality Report Series. Rexdale (ON): Ontario Ministry of the Environment.

Orth DJ. 1989. Aquatic habitat measurements. In Neilson LA, Johnson DL, editors. Fisheries Techniques. Bethesda (MD): Am. Fish. Soc. p. 61-84.

Peterjohn WT, Correll DL. 1984. Nutrient dynamics in an agricultural watershed: observation on the role of a riparian forest. Ecology 65:1466-1475.

Plafkin JL, Barbour MT, Porter KD, Gross SK, Hughs RM. 1989. Rapid bioassessment protocols for use in streams and rivers: benthic macroinvertebrates and fish. EPA/444/4-89-001.

Reynoldson TD, Rosenberg DM. 1996. Sampling strategies and practical considerations in building reference data bases for the prediction of invertebrate community structure. In Bailey RC, Norris RH, Reynoldson TB, editors. Study design and data analysis in benthic macroinvertebrate assessments of freshwater ecosystems using a reference site approach. Technical Information Workshop North American Benthological Society, 44th Annual Meeting, Kalispell, Montana. p. 1-31.

Robinson CLK, Levings CD. 1995. An overview of habitat classification systems, ecological models and geographic information systems applied to shallow foreshore marine habitats. Canadian Management Report of Fisheries and Aquat Science 2322.

Robinson CLK, Hay DE, Booth J, Truscott J. 1996. Standard methods for sampling resources and habitats in coastal subtidal regions of British Columbia: Part 2 - Review of sampling with preliminary recommendations. Canadian Technical Report of Fisheries and Aquatic Science 2119.

Ryder RA. 1965. A method for estimating the potential fish production of north temperate lakes. Trans Am Fish Soc94:214-218.

Scott DA, Jones TA. 1995. Classification and inventory of wetlands. Vegetatio 118:1-16.

Simon TP. 1991. Development of biotic integrity expectations for the ecoregions of Indiana. I. Central Corn Belt Plain. U.S. Chicago (IL): Environmental Protection Agency, Region V, Environmental Sciences Division, Monitoring and Quality Assurance Branch. Ambient Monitoring Section. EPA 905/9-91/025.

Snelgrove RVR, Butman CA. 1994. Animal sediment relationships revisited: cause versus effect. Oceanogr Mar Biol Ann Rev 32:111-178.

Strahler AN. 1957. Quantitative analysis of watershed geomorphology. Am Geophys Union Trans 38:913-920.

Taylor JK. 1987. Quality assurance of chemical measurements. Chelsea (MI): Lewis Publishers Inc.

Thorson G. 1957. Bottom communities(sublitorral or shallow shelf). In Hedgpeth JW, editor. Treatise on marine ecology and paleoecology. Vol 1. Memoirs of the Geological Society of America 67. p. 461-534.

Underwood AJ. 1994. On beyond BACI: sampling designs that might reliably detect environmental disturbances. Ecol Applications 4(1):3-15.

Underwood AJ. 1997. Experiments in ecology: their logical design and interpretation using analysis of variance. Cambridge University Press. United Kingdom.

Wetzel RG. 1975. Limnology. Philadelphia (PA): W.B. Saunders Co.

Wickware GM, Rubec CDA. 1989. Ecoregions of Ontario. Ecological Land Classification Series 26. Environment Canada.

Wiken E. 1986. Terrestrial ecotones of Canada. Ecological Land Classification Series 19. Ottawa (ON): Environment Canada.

Winter WT. 1977. Classification of hydrological settings of lakes in the north-central United States. Water Resources Research 134:753-767.

Winter WT, Woo MK. 1990. Hydrology of lakes and wetlands.. In Wolman MG, Riggs HC, editors. Surface water hydrology. Boulder (CO): The Geological Society of America. p. 159-187.

Yoder CO. 1989. The development and use of biological criteria for Ohio surface waters.. In Water quality standards for the 21st century. Washington (DC): U.S. Environmental Protection Agency, Office of Water. p. 139-146.

Yoder CO. 1991. Answering some concerns about biological criteria based on experiences in Ohio. In Water quality standards for the 21st century. Washington (DC): U.S. Environmental Protection Agency, Office of Water. p. 95-104.

Young RA, Huntrods T, Anderson W. 1980. Effectiveness of vegetated buffer strips in controlling pollution from feedlot runoff. J Environ Qual 9:483-497.

Table 

Table 2-1 provides additional information relevant to the site characterization that should be reported when preparing an EEM study design. The primary information types include general characteristics, hydrology, anthropogenic influences, aquatic resource characteristics, and environmental protection systems and practices. Each information type is accompanied by a list of recommended information to be reported.

Return to the table

3. Effects on Fish and Fisheries Resources

3.1 Overview

3.2 Study Design Considerations

• 3.2.1 Selection of Reference and Exposure Areas
  • 3.2.1.1 Sampling of Exposure and Reference Areas
  • 3.2.1.2 Adequacy of the Reference Area
• 3.2.2 Confounding Factors
• 3.2.3 Marine Discharges
• 3.2.4 Historical Data

3.3 Selection of Sentinel Fish Species

• 3.3.1 Community Survey
• 3.3.2 Immature Fish
• 3.3.3 Small-bodied Fish Species
• 3.3.4 Live-bearers

3.4 Effect Indicators

• 3.4.1 Lethal Sampling
  • 3.4.1.1 Survival
  • 3.4.1.2 Energy Use (Growth and Reproduction)
  • 3.4.1.3 Energy Storage (Condition)
  • 3.4.1.4 Abnormalities
• 3.4.2 Non-lethal Sampling
  • 3.4.2.1 Survival (Size Distributions)
  • 3.4.2.2 Energy Use
  • 3.4.2.3 Condition
• 3.4.3 Wild Molluscs

3.5 Timing of Sampling

3.6 Verification of Fish Exposure

3.7 Power Analysis

• 3.7.1 Power and Significance Level
• 3.7.2 Effect-Size

3.8 Fish Sampling Methods

• 3.8.1 Bycatch
• 3.8.2 Remote Sensing
• 3.8.3 Alternative Methods

3.9 Fish Survey Quality Assurance and Quality Control

• 3.9.1 Field Practices to Improve Data Analysis and Interpretation
• 3.9.2 Quality Control in the Field
• 3.9.3 Determination of Sampling Effort
  • 3.9.3.1 Examples of Calculations of Sampling Effort
• 3.9.4 Consultation with Regional EEM Coordinators and Implementation in the Field
• 3.9.5 Data Entry
• 3.9.6 Quality Control in the Laboratory

3.10 Data Analysis

• 3.10.1 Statistical Analysis

3.11 Methods for Analysing Fish Usability

• 3.11.1 Selection of the Fish Species
• 3.11.2 Tissue Sample Collection and Preparation
• 3.11.3 Supporting Analyses: Lipid and Moisture Percentages
• 3.11.4 Guidance for Fish Tissue Analysis for Mercury Using Non-Lethal Methods
  • 3.11.4.1 Recommended Methodology

3.12 References

List of Tables

• Table 3-1: Required fish survey measurements, expected precision and summary statistics
• Table 3-2: Suggested aging structures for Canadian fish species
• Table 3-3: Fish survey effect indicators and endpoints for various study designs
• Table 3-4: Generalizations and suggested optimal sampling times for fish species in EEM
• Table 3-5: Fish species commonly used in EEM, aspects to consider during study design, and recommended sampling times
• Table 3-6: Suggested reporting format for the parameters (A) and the resulting regressions (B) required for the fish survey analysis

3. Effects on Fish and Fisheries Resources

3.1 Overview

Fish monitoring for the environmental effects monitoring (EEM) program may consist of a fish population survey and tissue analyses to determine if the mine effluent is having an effect on fish and fisheries resources. Detailed requirements and timelines are found in Chapter 1 and in the Metal Mining Effluent Regulations (MMER) (SOR/2002-222).

For the purposes of EEM, fish includes shellfish, crustaceans and marine animals, as per section 2 of the Fisheries Act, but excludes parts of these organisms (MMER Schedule 5, section (s.) 1).

The fish survey provides an assessment of whether there are differences in the growth, reproduction, condition and survival of the fish population between exposed and reference areas or within an exposure area along a gradient of effluent concentrations. Note that a mine is required to conduct a study of the fish population if the concentration of effluent in the exposure area is greater than 1% in the area located within 250 metres (m) of a final discharge point (MMER, Schedule 5, s. 9(b)).

In addition to the fish survey, biological monitoring studies may also include a study respecting fish tissue if, during effluent characterization conducted under paragraph 4(1)(d), a concentration of total mercury (inorganic and organic mercury) in the effluent is identified that is equal to or greater than 0.10 µg/L (MMER, Schedule 5, s. 9(c)). “Effect on fish tissue” is defined as measurements of concentrations of total mercury that exceed 0.5 µg/g wet weight in fish tissue taken in an exposure area and that are statistically different from and higher than the measurements of concentrations of total mercury in fish tissue taken in a reference area (MMER, Schedule 5, s. 1).

Top of Page

3.2 Study Design Considerations

General information regarding study designs is discussed in Chapter 2. The study design requirements and the definitions of effect for the fish population survey and fish tissue survey are discussed in Chapter 1.

To evaluate the effect of effluent on fish, the following questions should be answered:

• Is there an effect?
• Is the effect mine-related?
• Is the magnitude and extent of the effect known?
• Is the mine-related cause of the effect known?

Each mine’s EEM representatives or consultants should consult with the regional EEM authority to review the results of the previous phase’s site selection, species selection, fishing effort, etc., and to discuss the selection of the most appropriate options for the current phase. The results of previous phases, historical data, and local knowledge should be used to assess:

• the suitability and capture success of selected sentinel species
• the adequacy of the reference area
• the appropriateness of sampling methods and required equipment.

Mines may want to make changes between phases, including increasing sampling effort; changing sampling methods, equipment or fish species; selecting different exposure or reference areas; or using alternative monitoring techniques. Changes in the study design may need to be made for various reasons, including the following:

• The results indicate that power was insufficient in the previous phase due to collection of a low number of fish or high variability.
• The species characteristics were not measurable, not suitable, or there are concerns about the status of fish populations.
• It is uncertain if the fish were exposed to effluent.
• Reference sites were inappropriate.

Concerns raised about EEM studies (and field studies in general) can be separated into concerns about the adequacy of the reference sites, the potential impacts of confounding factors (e.g., potential influences of genetics on the variability of species characteristics), the ecological relevance of effect indicators used, the influences of natural variability, and concerns over statistical design issues. This guidance will attempt to provide input to deal with these issues.

Top of Page

3.2.1 Selection of Reference and Exposure Areas

The two main study designs are control-impact and gradient designs. The choice of reference area is the number one problem with control-impact field studies (Munkittrick et al. 2009). Ideally, a reference site would be located upstream, in similar habitat, and free of confounding influences, with a natural barrier that limits movement between sites. This situation is seldom available. The main issues cited regarding reference areas include whether the reference site is (a) comparable in terms of habitat; (b) free from the issue of concern (i.e., exposure) and from confounding influences (further discussed in Chapter 2); (c) open to movement of fish from the exposure area (fish in an upstream reference area could have been exposed previously or fish in the exposure area could be transient, reducing exposure to potential effects); and (d) whether the exposed fish were exposed to the effluent or stressor of interest.

No reference site is perfect. The ideal situation involves having data from before construction or initiation of the stressor of interest (e.g., before/after control-impact [BACI] design; outlined further in Chapter 4). Study sites that have barriers that prevent fish from moving between sites (e.g., dams, waterfalls, beaver dams) may be a good alternative, providing that the barrier does not alter the habitat. In open-water areas, choosing a species that has limited mobility improves the confidence that fish are not moving, but increases the potential influences of local differences in habitat. One difficult situation to interpret arises when there are no statistical differences in fish measurements between the areas, and there are no barriers restricting movement. In these cases, an indicator of exposure to the effluent is recommended, which can be chemical or physiological (e.g., liver enzyme induction, stable isotope signatures [Galloway et al. 2003; Dubé et al. 2006]).

If there are significant differences in fish characteristics between reference and exposure areas, there can be high confidence that fish are not moving between sites. Although differences are seen, variability in fish parameters (e.g., growth, weight, condition) is a function of a number of factors, not all of which will be related to the discharge of effluent. The selection of appropriate fish species for monitoring, survey timing and sampling gear will also facilitate the interpretation of any differences detected. Nevertheless, other natural and anthropogenic factors may influence effects on the fish and fish tissue and confound interpretation of the data. The requirement to confirm effects was developed to increase the confidence, over two phases, that effects are mine-related.

Top of Page

3.2.1.1 Sampling of Exposure and Reference Areas

The exposure area should be selected to ensure that the fish collected have been exposed to the effluent. It should be sampled first to determine which fish species are present, and their relative abundance within the area. The reference area can then be selected to provide fish of the same species that are available at the exposed site. Timing of sampling should be as close as possible between sites, to minimize temporal variability. The choice of time period for sampling will depend on factors such as time of year, stage of reproductive development, and potential habitat differences between sites (water temperature differences, etc.), but it is recommended that, if possible, all sampling be done within the same week. If a longer time period is required, reference samples should be collected before and after the collection of exposure samples, to allow comparison.

If fish are found in the reference area, but not in the exposure area where they are expected to occur (i.e., fish were historically found in the sampling area), the absence of fish in the exposure area should be reported as an effect. More information on reference site selection can be found in Chapter 2.

Top of Page

3.2.1.2 Adequacy of the Reference Area

It is now common in research programs to use a large number of reference sites. As an example, over the first 3 cycles of monitoring for the pulp and paper EEM program, there has been a trend toward using more reference sites. In Cycle 1, 3% of studies used multiple reference sites; in Cycle 2, 9%; and in Cycle 3, 25%. Including additional reference sites increases the ability to evaluate issues related to natural variability, ecological relevance and confounding factors, and improves the ability to evaluate the adequacy of the chosen reference site. Studies that use a gradient approach and multiple reference sites are statistically stronger than studies that depend only on a single reference site.

Other new approaches include reference condition approaches (Bailey et al. 1998), and using negative reference sites (using the exposed site as your reference). Regardless, the existence of consistent changes over two phases increases the level of confidence that changes are real. Follow-up studies must evaluate the adequacy of the reference site, especially if consistent results are not found.

Top of Page

3.2.2 Confounding Factors

In Cycle 2 of the pulp and paper EEM program, almost 90% of studies that detected effects also concluded that factors other than pulp mill effluent were responsible for such observations. Potential confounding factors exist at most sites and include other outfalls, habitat changes, historical uses and contamination, tributaries and non-point-source inputs. In highly confounded situations, alternative methods should be considered, but it should be emphasized that it is possible to obtain interpretable field results at most sites with adjustments to the study design. Given the complexity of certain situations, it is recommended that as much data as possible be gathered in order to demonstrate that other discharges or contaminant sources are primarily responsible for observed changes or an absence of observed changes. If changes are seen and determined to be influenced by confounding factors, the objective of subsequent study designs should be to eliminate the confounding factors or determine their significance.

Top of Page

3.2.3 Marine Discharges

Metal mines discharging into marine or estuarine receiving waters may face a number of problems and confounding factors that should be considered when developing an EEM fish survey study design. These problems may include the following:

• Some marine and estuarine areas are difficult to sample (e.g., tides, currents, high flushing rates or unsuitable habitat) and alternative approaches should be considered.
• There can be gradients for current, temperature and salinity, which can affect physical processes and the uptake of contaminants as well as have consequences for physiological changes within organisms.
• Selection of reference areas can be more difficult in marine situations.
• Different life stages of fish may utilize different habitats at different times of the year.
• Species availability can be low in marine environments. In many situations, small-bodied resident fish species are available and should be investigated. These species may be multiple spawners or live-bearers, or species for which there is little background information. However, this should not restrict or inhibit attempts to use these species, especially if they are abundant. The assumption inherent in an EEM program is that a fish community should be intact, with the normal abundant species present. The second priority, and underlying assumption, is that a fish population which shows a growth rate, reproductive development, and an age distribution indistinguishable from a reference area, is unaffected.

Potential solutions to these difficulties include using alternative species or caged bivalves, or mesocosms for confounded receiving environments. New facilities that will have collected baseline information prior to initiating effluent discharge will be in a better position to assess the effect of their effluent on the receiving environment compared to confounding factors.

Top of Page

3.2.4 Historical Data

Mines have the opportunity to submit historical data if there is previous biological monitoring information that could determine if there are effects on fish, fish tissue or the benthic invertebrate community. Historical data can be used to assist in the development of the first EEM study. See Chapter 13 for additional information on historical data.

Top of Page

3.3 Selection of Sentinel Fish Species

The recommended method for carrying out the fish survey is to monitor adults of two species of relatively sedentary finfish that have been exposed to effluent over a long period of time. Sexually mature finfish are preferred, but where they are not available, it is possible to design a program using shellfish or juvenile fish, although it will not be possible to analyze all the same effect endpoints. If available, at least one of the species selected should be a benthivore. The most important factors when selecting fish species for the EEM program are exposure, abundance, relevance to the study area (Munkittrick et al. 2000; McMaster et al. 2002), and sensitivity to effluent. In selecting the two species, the species used in previous EEM studies at the site (if applicable) should be considered, and preference should be given to:

• resident (non-migratory) fish species identified in a site characterization
• sexually mature female and male fish species that are abundant in both the exposure and reference areas
• fish species for which fishing or sampling permits can be obtained
• fish species that have the highest exposure to effluent

At any given site, there may be limited choices of potential species for monitoring. It will often be necessary to obtain the advice of an experienced fisheries biologist with knowledge of fish species present in the study area. More than 60 species have been used as sentinels in EEM pulp and paper and metal mining programs to date, and mines and their consultants are encouraged to contact regional, federal and provincial government agencies for fisheries information and additional guidance.

Some receiving environments do not support adequate numbers of fish for sampling. In situations where it has been determined that fisheries resources may be impacted by a destructive fish survey, non-lethal sampling techniques may be used. In environments that do not support adequate numbers of fish to meet the recommended sample sizes or where there are not two suitable fish species for monitoring, the following options, in order of preference, may be considered:

• one sexually mature fish species and one sexually immature fish species
• two sexually immature fish species
• one sexually mature fish species
• one sexually immature fish species.

The mine should consider changing its study design (e.g., species, methods of collection) if the results from the previous phase suggest that the species is long-lived (> 30 years); that it was not possible to measure all survey parameters on the fish (e.g., age, liver and gonad weight); that an insufficient number of individuals were collected; and that the degree of variability was such that the numbers of fish required by power analysis for subsequent designs are unreasonable, and it is not possible to reduce this by selective sampling methods. If the fish species available at a site are present in the high effluent exposure (near-field) area only during certain times of the year or life-history stages, the life stage and sampling time should be selected to maximize exposure to effluent.

Some of the challenges with species selection may relate to attempts to design a single program for multiple purposes. Concerns about contamination of fishery resources for human consumption would direct the study design to collect a species that is long-lived (so that contaminants accumulate longer), is piscivorous (so that biomagnification is greater), matures late (to increase concentration), preferably focuses on male fish or species that do not spawn every year (so that elimination of contaminants through egg deposition is lessened), and are of importance for local consumption. To improve the sensitivity of detecting environmental impacts, it is preferable that species are benthic (because generally they will move less), are not commercially or recreationally important (because it obscures the determining cause), mature early, contribute much energy to reproduction (so that energy demands are high), and are short-lived (so that impacts are recent)--with a focus on female fish (environmental impacts are often more serious on female egg producers).

A number of other factors need to be considered when selecting a sentinel species (see Munkittrick and McMaster 2000; Munkittrick et al. 2000), including ensuring that the species are active participants in the local aquatic food web. Other life-history characteristics, such as spawning time and migration, need to be evaluated site-specifically, because the interaction between discharge site, spawning habitats, seasonal changes in flow and dilution can all influence results and potentially impact the sensitivity of the monitoring program.

A key consideration when selecting a species is the mobility and residence time of that species, as this determines effluent exposure. Species that are resident in the system for most or all of their life cycle and exhibit territorial behaviour or limited mobility relative to the size of the study area are preferred, because the observed responses of these species reflect their localized environment. Species that are migratory or spend only a small proportion of their life cycle in the system under investigation (e.g., anadromous salmonids, some marine fishes) are not suitable, because exposure to effluent is minimal or transient and difficult to determine. This is also true for species that are highly mobile and are likely to be moving in and out of the effluent exposure area. In some cases, it may be possible to select more mobile species (e.g., Mountain Whitefish) (Swanson 1993), due to physical constraints that limit movement (e.g., dams, natural barriers, changes in habitat). In general, the greater the likelihood that a fish species is exposed to effluent, the greater its value as a monitoring species.

Top of Page

3.3.1 Community Survey

If a mine is new or has no historical survey information available, a fish community survey should be done to aid in the selection of appropriate fish species. Fish community surveys evaluate whether there are differences between areas in the diversity and abundance of fish species present.

A change in fish community has occurred when species that are expected to be abundant from the collections conducted at reference areas are not present in the effluent discharge area. If the exposure areas do not support one or more of the abundant species found at the reference area, it will be necessary to document the geographical extent of this absence. When the fish community composition has changed because of the presence of an effluent, there will also likely be measurable changes in the fish populations that remain. Results from the EEM program should document this, and may help in determining whether other fish species are at risk of disappearing from the exposure area.

Fish communities often include a number of species that are not abundant for a variety of reasons that may be unrelated to the presence of mine effluent. Non-lethal techniques (e.g., electro-fishing) are preferred for the community survey where possible, and field sampling should be designed to limit mortality of the existing species.

Top of Page

3.3.2 Immature Fish

The recommended method for carrying out a fish survey is to monitor adults (sexually mature fish) of two species of relatively sedentary finfish that have been exposed to effluent over a long period. However, there have been situations where no adult fish can be collected in a receiving environment. For example, some areas may not be inhabited by adult finfish, but are nursery areas for their juveniles. If sexually mature fish do not reside in an effluent exposure area, the suitability of juvenile fish may be considered. When sexually immature species are used, there is no direct measurement of reproductive development. However, the relative abundance of young of the year (YOY) can be used as a measurement of reproductive success.

Relevant measures for juvenile fish would be similar to those of mature fish, but without gonad measurements: growth (length, weight, or weight-at-age, if possible); condition (length-at-body-weight relationships); liver-weight-to-body-weight ratio; abundance (YOY survival, percent composition of age classes); deformities associated with exposure to effluents, such as vertebral fusions and compressions, spinal curvatures including lordosis and scoliosis, and fin erosion; and growth in juveniles exposed to effluent compared to juveniles in the reference area. Methods for the collection of juvenile fish are well established and many juvenile fishes can be aged (e.g., Secor et al. 1995).

Top of Page

3.3.3 Small-bodied Fish Species

The trend toward the increasing use of small-bodied forage-fish species (Munkittrick et al. 2002) has continued, rising from their use in 10% of surveys in the pulp and paper Cycle 1, to 26% in Cycle 2 and 34% in Cycle 3. A small-bodied fish can be considered a fish species that has a maximum size of 150 mm or less. Their use has several advantages and disadvantages. On a practical level, small-bodied fish species are usually more abundant, easy to capture, and more sedentary than larger-bodied fish species. Small-bodied fish have also been shown to be more sensitive to environmental changes, such as pH (Shuter 1990). Their home-range size has been positively correlated with body size (Minns 1995), and many small-bodied species integrate local conditions very well.

On the other hand, small-bodied fish require more sensitive analytical balances and more careful measurements. They are more sensitive to microhabitat differences because they integrate the local habitat so well. They are also more sensitive to differences in timing of sampling (see section 3.5).

In addition, small-bodied fish often have a shorter life span, so if they are chosen as one or both of the fish species, an additional 20 sexually immature fish (0+ and 1+) should be collected to aid in size-at-age (growth) analysis. Also, because a small-bodied fish species may only have a life expectancy of 3 to 4 years, the 0+ and 1+ will constitute a significant portion of the population (e.g., the 0+ and 1+ Slimy Sculpin [Cottus cognatus] are up to 50–70% of the population). This measurement of the proportion of a sample composed of YOY fish does add another surrogate measurement for reproductive performance (Gray et al. 2002).

There are other considerations as well. The life history, biology, and reproductive characteristics of some small-bodied species are unknown, making it difficult to determine the best sample areas, times and methods. Some are multiple spawners, which means reproductive effort in these species is difficult to estimate from a single sample because the reproductive tissue can be turned over almost completely between clutches (i.e., most of the mass of ova in the ovary will be spawned, and then a new clutch of mature ova will be developed). The ovary will generally contain two or more class sizes of ova and the spawning season may last from several weeks to more than a month. The number of clutches produced during the spawning season becomes the important reproductive variable and is difficult to estimate for an individual female in the field, even with frequent sampling. It will be difficult to evaluate the significance of changes in egg production in multiple spawners if they show normal reproduction in the first clutches.

Species identification of small-bodied fish should be verified, especially for cyprinids, which can appear very similar without careful examination. Useful references for this purpose include Scott (1967), Scott and Crossman (1973), Roberts (1988), Nelson and Paetz (1992), Jenkins and Burkhead (1993) and Coad et al. (1995). The smaller organ size of these fish requires a more sensitive balance. Dissecting microscopes may be necessary for removing the organs properly and avoiding extraneous tissue or moisture, which could affect results. Dissection on recently collected, fresh fish is recommended. Differentiation among tissues and separation of the liver and gonads from intestinal tissue is easiest when the tissue is fresh. Dissection of frozen specimens of small fish can be difficult and lead to errors in organ measurements. Preservation in a formalin solution may give adequate results, but care must be taken to treat exposure and reference fish the same (e.g., duration of storage) in order to minimize preservation distortion.

Measurement of fecundity and egg weight requires special consideration. Many small species have few, large eggs. Gonadal estimates will be easier closer to spawning. The timing of sampling will also be affected by residency, and the two factors have to be optimized. The entire gonad should be preserved and fecundity counts conducted with the aid of a dissecting microscope.

Top of Page

3.3.4 Live-bearers

Live-bearers are not common in Canadian freshwater receiving environments, but if used, require special attention regarding measurement of reproductive variables. Live-bearing species have been used successfully for detecting responses in exposures to Swedish pulp mills (Larsson et al. 2000, 2002; Larsson and Forlin 2002). To estimate fecundity, the gonad must be preserved and the number of live and dead embryos counted. Proper sampling requires some preliminary data on spawning time and gonadal development so that sampling procedures can be optimized.

3.4 Effect Indicators

The effect indicators for the various types of study designs for the fish survey are listed in Table 3-3. For a much more detailed discussion on these topics, consult Munkittrick et al. (2009), where the authors re-emphasize the original purpose of the EEM program and discuss why the current EEM effect indicators are used in place of other levels of monitoring. Additional issues raised and addressed by Munkittrick et al. (2009) include the influence of natural variability (i.e., the tendency for parameter values to change from year to year, or potentially from site to site), genetic adaptation, and four important statistical design issues (site selection, pseudo-replication, power analysis, and concern over the number of comparisons made).

The EEM program focuses on parameters measurable in groups of individuals, for several reasons:

• The approach offers a compromise between the sensitivity and reversibility of biochemical approaches, and the relevance of community-level parameters.
• Monitoring at the community level will miss reversible, important effects at the population level.
• Changes to fish growth, reproduction, condition or survival puts fish at risk, and therefore, focusing on these population level parameters addresses the overall objective of the Fisheries Act, which is to protect fisheries resources.
• Knowing this level of risk is important to the management of ecosystems.

Top of Page

3.4.1 Lethal Sampling

In answer to the question “have fish been modified by the effluent?” effects on growth, reproduction, condition and survival of the fish population are examined. The program recommendation for the fish survey is that key indicators be measured in both sexes of adults of two species of fish. The precision for the measurements is listed in Table 3-1. The intention is to obtain estimates of age or size distribution, how well fish are using available energy for growth and reproduction, and the storage of energy as reserves. The required numbers of samples can be calculated from a statistical equation using the standard deviation (SD) of gonad sizes for the species and site (from previous samples), and a critical effect size (CES) of 25% (see section 3.7.1). The minimum sample size recommended for a lethal fish survey when there are insufficient data to calculate sample size by power analysis is 20 sexually mature males and 20 sexually mature females of 2 fish species, in each sampling area. The rationale for using 20 fish of each sex for lethal sampling is that there is little change in the 95% confidence limits with increasing sample size beyond 20 fish. For example, Munkittrick (1992) found that there was little improvement in White Sucker variance estimates with a sample size above 16.

When there is background information available, it should be used to calculate adequate sample-size requirements prior to conducting the fish survey. It is important that sample size and variability be examined early in the study design phase so that the study can be redesigned if the variability estimates are sufficiently high for the survey not to achieve adequate power. Fish surveys benefit most by decreasing variability. When variability is so high that sample sizes are not justifiable or cost-effective, the first consideration should be to redesign the study to a) reduce variability, b) select alternative species that may be less variable, or c) consider an alternative method.

It is strongly recommended that sampled fish be processed and sexed immediately in the field on sample days to ensure the collection of fish with an equal sex ratio. Subsequent sexing of the fish in the lab using frozen samples may show a skewed sex ratio if it is assumed that fish sampled in the field displayed a 1:1 sex ratio.

It is important to identify immature fish (fish not developing to spawn) so that they can be excluded from the statistical analysis. There are three situations where gonadal development of fish is not uniform: a) situations with multiple spawning species where spawning is not synchronized; b) multiple spawning species where the number of spawns per year is influenced by fish size or age; and c) in northern populations, where fish may not acquire sufficient energy reserves to spawn each year. In all cases, fish should be analyzed within a group: comparisons should be conducted between fish developing to spawn and fish that are not. As well, the proportion of fish in each category can be analyzed. In situations where the existence of two or more groups is known before sampling, it may be possible to separate fish into categories during sampling based on condition or fish size.

The EEM program operates in an iterative fashion, so it is not necessary to develop a full assessment of the fish populations in a single sample, and the measurements are meant to act as surrogates to assist in the development of an assessment over more than one phase. Any effects in the fish survey must be confirmed in a subsequent phase, and be assessed against the CESs before studies progress (CESs are discussed in Chapter 1). While the measurements listed below are the required measurements, it may be necessary to provide alternative measurements due to site-specific or species-specific issues.

* If caudal fin is forked, use fork length (from the anterior-most part to the fork of the tail). Otherwise, use total length, and report type of length measurement conducted for each species. In cases where fin erosion is prevalent, standard length should be used.

** Fecundity can be calculated by dividing total ovary weight by weight of individual eggs. Individual egg weight can be estimated by counting the number of eggs in a sub-sample. The sub-sample should contain at least 100 eggs.

*** For small-size fish weights, use at least a 3-decimal scale.

Top of Page

3.4.1.1 Survival

Mean age is meant to give an assessment of the relative ages of the reference and exposed populations. If size-selective gear such as gillnets are used, and there is a significant difference in mean ages of fish sampled at both sites with identical gear, the difference indicates a need to further investigate the population and the reason for the difference in subsequent phases. More detailed information can be obtained through age distributions (or size distributions if aging is not possible), if adequate sample sizes are available or if aging is not possible. Furthermore, since many fish species have short life spans (< 4 years), it may be necessary to obtain immature fish and juveniles in order to conduct an appropriate assessment of this effect indicator. It is also very difficult to obtain a 25% difference in age when species are short-lived, and it may be possible to substitute a difference in average size (length) of 25% as a surrogate for age when species are short-lived.

A list of appropriate aging structures for a variety of potential sentinel species is provided in Table 3-2. In addition, there are many references that can be referred to for aging methods (e.g., Mackay et al. 1990). Methods of aging should be consistent at each sampling area and among phases, and appropriate quality assurance / quality control (QA/QC) procedures should be followed (e.g., independent confirmation). It is recommended that all aging structures be archived for future reference. If fish cannot be aged reliably or if it is not cost- or time-effective, the age can be determined by using size-frequency distributions. This may be especially useful when sampling small-bodied fish species or when conducting non-lethal sampling. It may also be possible to confirm the size-frequency distributions by aging representative sub-samples from each size class. For more information on size-frequency distributions, consult Nielsen and Johnson (1983).

Top of Page

3.4.1.2 Energy Use (Growth and Reproduction)

Growth and reproduction measures give an assessment of the ability of fish to use the food available to them. Growth is the change in size (weight or length) with time or age. In the case of growth, it may be helpful to collect information on other age classes, such as whether there are changes in growth of early life stages. This will assist in determining the magnitude of the effect. Subsequent phases should focus on confirming responses detected and examining the relevance of the changes to other size classes and species.

Reproduction is expressed as reproductive effort, fecundity, egg weight or gonad weight relative to body size. Reproduction may be the most sensitive measurement in resident fish. Changes in reproductive investment can be evident within a year, because the reproductive tissue is generally turned over annually. Fecundity and gonad weight are easy to measure if an appropriate sampling time is chosen. Confirmed changes in gonad size could lead to additional work related to magnitude, such as determining whether the change occurs at other times of the year (for multiple spawners) or whether the changes are present in other species in the same area.

Top of Page

3.4.1.3 Energy Storage (Condition)

Measures of energy reserves provide valuable information on the availability and quality of food to the fish. The EEM program uses condition (body-length-to-body-weight relationships) and liver size as indicators of energy reserves. As with other indicators, the consistency in response between indicators is important. Liver size can increase for several reasons, including storage of lipids and glycogen and enhanced detoxification activity.

Top of Page

3.4.1.4 Abnormalities

During the fish survey, a visual examination of fish is also conducted in order to identify the presence of any internal or external abnormalities, such as of body form, body surface, fins, eyes, lesions, tumours, neoplasms, scars or other abnormalities such as eroded, frayed or hemorrhagic fins, internal lesions, abnormal growths, parasites, and any other unusual observations. An area on the data sheet should also be included for other significant observations. Photographs can be a useful tool to document any obvious abnormalities.

It is recommended that a rough illustration of the selected fish species be incorporated into the data collection sheet for the recording of abnormalities in the external appearance. This information can then be used by others at a later date if significant differences exist between reference and exposure areas.

More information on fish anatomy can be found in general fish biology textbooks. Instructions on tumour descriptions are available in Gross Signs of Tumors in Great Lakes Fish: A Manual for Field Biologists (www.glfc.org/tumor/tumor1.htm).

* Fish survey effect endpoints used for determining effects as designated by statistically significant differences between exposure and reference streams. Other supporting endpoints can be used to support analyses.

Top of Page

3.4.2 Non-lethal Sampling

Non-lethal sampling should only be used in situations where it is warranted, i.e., where there is a concern about the potential impacts of sampling on small fish populations. Lethal sampling of adults is preferred where possible, although information on non-lethal samples can be valuable when large numbers of fish are collected during the sampling procedures. The indicators used for non-lethal sampling are contained in Table 3-3, and additional information on statistical analysis for the non-lethal sampling is contained in Chapter 8.

If the only option for a facility is to do a non-lethal sampling of fish in order to evaluate the effects of effluents on the fish population at a facility, a minimum of 100 fish older than YOY is recommended from each study site. The YOY acquired during the collection for the 100 non-YOY fish should also be retained and sampled (measured). YOY can usually be separated from older age-classes by size distributions; however, this may not be possible for species with extended spawning periods. The proportion of fish that are YOY should be estimated from the first 100 fish collected. If YOY abundance is extremely high (> 80-90%), sampling should then continue until 100 non-YOY are captured to calculate size-distributions of older fish. The collection of the additional non-YOY fish allows for a higher discrimination of the older fish classes to be achieved. The fish older than YOY that are collected should represent the whole range of fish sizes and be representative of the population (mature and immature). The recommended sample sizes in each area will give a good idea of the population distribution when plotting parameters such as the length or weight frequency. As well, when examining differences between the relative abundance of young versus mature fish, fairly good resolution is achieved (Gray et al. 2002).

When possible, sampling should be conducted when YOY are a catchable size in the gear being used. The same sampling gear should be used in both the exposure and reference areas; if it is not possible to use the same gear, or multiple gears must be used, the size distributions within a site should be compared between gears. If there are differences in the sizes of fish collected with different gear, comparisons between sites should be restricted within gear type. The sampling techniques and relative effort should be the same in all sampling areas. Pooling of data from different fish-sampling techniques should be avoided, and all methods used should be reported. If more than one gear type is used, the records of fish caught by each method should be reported, and any pooling of data clearly described. Fish should be measured for length (±1 mm), weight (± 0.01 g) (Gray and Munkittrick 2005), assessed for the presence of abnormalities, and external sex determination should be made, if possible. All fish should then be released. If possible, a small number of larger fish should be sacrificed to verify ages of older individuals. If only adults are used, the priority should be to sample prior to or at the start of the spawning season (see guidance on preferred sampling times in Table 3-5). However, if YOY are to be collected, the timing should move to the late fall, when it will be easier to measure YOY for most species. Fall sampling of YOY will be much more difficult if the fish are not single, synchronous spring spawners, as the size distributions of YOY fish will be broad.

A large number of areas can typically be sampled when conducting a non-lethal survey and the facility is encouraged to sample multiple exposure and reference areas. Programs that sample adults and YOY will allow for maximal assessment of effect indicators.

Species selection for non-lethal sampling can be difficult and is often based on availability. When choices are available, a synchronous spring spawner will offer the most advantages in terms of differentiating YOY from older year-classes. Discrimination of year-classes can also be affected by the longevity of the species. An annual species such as silverside will have a single year-class, eliminating the need to differentiate year-classes. A short-lived species (2-3 years) with fast growth and easily distinguished year-classes also offers advantages. However, these species are not always available.

When multiple species are available to choose from, it is recommended to collect initial samples and examine the ability to discriminate YOY and age-classes between species.

Top of Page

3.4.2.1 Survival (Size Distributions)

There are challenges to using age information on many short-lived species of fish. If a fish only lives 2-3 years, it will not be possible to measure a 25% difference in mean age. If non-lethal aging structures have not been validated for the sentinel species being used, size-distribution should be examined as a surrogate for differences in age.

Size distributions should be compared between exposure and reference areas with the Kolmogorov-Smirnov test, although this test is not very sensitive. Size comparisons should also examine distributions for YOY alone, for both sizes combined. If a site difference is present, subsequent phases should focus on understanding the difference and possible causes. When possible, verifying the ages of larger fish and YOY can be useful.

Top of Page

3.4.2.2 Energy Use

It should be possible at most sites to get estimates of growth and reproduction using non-lethal methods. Growth can be evaluated by the size of YOY at the end of the growing season and by the size of the 1+ fish. A comparison of the size of YOY fish between sites gives a good indicator of growth, as it is a direct indicator, in comparison to size-at-age, which is indirect. YOY are used because all of their growth is attributable to environmental conditions since the spawning time, and growth is not complicated by diverging energy into reproductive development. Differences between sites in spawning times will be integrated into this analysis. It is also possible to get a growth estimate by a shift in size distributions over time (e.g., repeating measurements 2 months apart at the same sites), or differences in average size (this would require a second sampling trip to determine). If the fish species chosen is externally sexually dimorphic, it is possible to examine whether there are gender-specific differences in growth rate.

Reproduction can be assessed using relative age-class strength or by the relative abundance of YOY individuals (Gray et al. 2002) or by YOY survival, which requires two sampling periods. A length-frequency distribution may be plotted as a surrogate of an age-frequency distribution. Size-frequency analysis can be used to examine size distributions and distributions of condition factors (using length and weight data), and can be used to infer age distributions and size-at-age data (if ages can be inferred) (Gray et al. 2002). It is recommended that, if possible, aging structures be collected from a sub-sample of each size-class, for situations where age may need to be verified (as in section 3.4.1.1, the utility of the age information is reduced in situations where the species is short-lived). In Slimy Sculpin, rapid growth of YOY fish in the spring can cause some overlap with the 1+ age-class, making resolution difficult (Gray et al. 2002). The ability to discriminate the YOY will depend on the duration of the spawning season, and the amount of time elapsed between the spawning time and sampling time. It may be easiest (for spring and early-summer spawning species) to examine length-frequency distributions using late summer and early fall data, when the YOY should be easiest to distinguish. To test for differences in relative abundance of YOY between the exposure and reference areas, a Kolmogorov-Smirnov test can be performed on length-frequency distributions with and without the YOY included. If inclusion of the YOY changes the interpretation of the significance of the difference (i.e., it is different with them included, and not different without them), there is then a difference in the relative abundance of YOY. Alternatively, replicate areas can be sampled to allow for the use of more statistical approaches, or the proportions of YOY can be tested using a Chi-squared test.

It may not be possible to distinguish YOY in species that spawn multiple times, in northern areas where YOY may emerge later in the year, or in situations where there are habitat-preference differences that are age-dependent in a species. In those cases it will not be possible to easily infer potential reproductive impacts. Some professional judgement will be required. If the species lives multiple years and immature fish can be distinguished non-lethally (condition near spawning time can be used in many situations for this), the proportion of immature fish can be used as a substitute. In cases where this is not possible, interpretation will need to be made based on size distributions alone, and care must be exercised to be conscious of the potential impacts of adult mortality on interpretation.

It is important to remember that a difference in water temperature between sites will affect spawning time. End-of-summer differences in size distributions could as easily result from differences in spawning time due to temperature as other potential causes. If there are temperature differences between sites that are suspected to be a major cause in the differences in size distribution observed, then subsequent studies should determine whether these site differences are a consequence of the facility or an inadequacy in choosing reference sites.

If it is possible to make multiple sampling trips, it may also be possible to measure changes in condition before and after spawning as an indicator of reproductive investment. For some small-bodied species, spawning females are very easy to distinguish by condition factor. Differences in condition factor of females between sites before spawning, or an indication of the change before and after spawning in females, could be used to infer reproductive investment, if females can be distinguished after spawning.

Top of Page

3.4.2.3 Condition

Condition factor (k) can be evaluated by the relationship (k = 100 000* (wt/l³)) of the fish examined (where wt = weight [in grams] and l = length [in mm]). The appropriate analysis for final interpretation is an analysis of covariance (ANCOVA) of weight versus length, by site.

Top of Page

3.4.3 Wild Molluscs

Where there are no appropriate finfish present, collection of wild molluscs, such as oysters or mussels, may be considered. Shellfish are included under the definition of fish in the Fisheries Act and they have been used by some pulp and paper mills in the EEM program. However, there are some drawbacks, including difficulties in aging individuals and in estimating reproductive investment in some species. Crabs and lobsters are not suitable species because they cannot be reliably aged at the present time (Environment Canada 1997). Currently, guidance is available on the relative gonad index (mantle somatic index) for bivalves (see mesocosm guidance in Chapter 9).

Molluscs are a diverse taxonomic group that include bivalves and gastropods, and are widely distributed throughout Canada. Molluscs possess many qualities that a species for monitoring should exhibit:

• they are relatively sedentary, although some species (i.e., unionids) may migrate short distances (metres) within their habitat;
• they are widely distributed across Canada and are identified with limited taxonomic expertise;
• most unionid bivalves are large enough to provide sufficient tissue for analyses;
• several bivalve species have been shown to readily accumulate many chemicals from a variety of pathways (water, sediment, food) and show sublethal effects associated with exposure; and
• bivalve growth is relatively easy to measure and has been shown to be as sensitive or more sensitive than mortality in other standard assays on species such as Daphnia, Fathead Minnow and Rainbow Trout (see Salazar and Salazar 2001).

In general, reproductive periods for molluscs and patterns of abundance are related to climate and the abundance of food supply. For most freshwater lotic or lentic habitat types, sampling is best conducted during the fall when the majority of taxa will be present and/or are large enough to be easily collected. In marine environments, sampling should be conducted in late summer or fall, as populations with spring recruits have stabilized by this time.

3.5 Timing of Sampling

A variety of factors need to be considered when deciding on a time to sample, including potential migratory behaviour of the sentinel species, water conditions (e.g., flow, turbidity, wave action), accessibility, and the cycle of gonadal development for the sentinel species. Where historical data exist, it would be useful to examine the data and, if appropriate, conduct the survey during similar periods so that the surveys can be compared.

The timing of sampling should be synchronized with the development of sufficient gonadal tissue so that effects on the reproductive function can be assessed. However, such information is unavailable for many species of fish. Species for which there exists extensive background information on their biology and life history characteristics should be preferred as sentinel species in order to ensure that sampling can be synchronized with sufficient gonadal tissue development.

Recent research has been conducted to evaluate the optimal timing for interpreting gonadal development, using seasonal collections from a variety of species. Five types of fish categorized by spawning characteristics have been identified, and Table 3-4 provides the recommended sampling time based on the following background collection studies: background collections followed Canadian freshwater species that were synchronous spawners (such as Slimy Sculpin; Gray et al. 2005; Brasfield 2007), multiple spawners with few spawnings per year (such as Blacknose Dace [Rhinichthys atratulus]; Galloway and Munkittrick 2006; Hicks and Munkittrick, unpublished data), multiple spawners with many spawns per season (such as Redbelly Dace [Chrosomus eos]; Carroll 2007), and asynchronous spawners (every few days, such as Mummichog [Fundulus heteroclitus; McMullin et al. 2009). There is a fifth type of freshwater species that has asynchronous development, where individuals may take a year off from spawning because of cold temperatures or low food availability. This variability has a major impact on power and sample size requirements.

Examination of these data confirms that there are specific times when power is higher for detecting differences, and when gonadal development is adequate for detecting impacts. The generalizations in Table 3-4 may not apply to all species or all regions; the regional EEM contact should be consulted for any available updates to regional guidance.

Synchronous spawners show a difference in timing of gonadal development between males and females. For synchronous spring spawners, adequate data can usually be obtained as late as possible in the fall, or prior to spawning in the spring. If previous data are available for a site, the reproductive strategy can usually be estimated from the magnitude of the correlation coefficient (R²) between gonad weight and body weight, if the previous collections were done at a time when the gonads were well developed.

Evaluation in multiple-spawning species is complicated by the duration of the spawning period. In the case of such species, the frequency distribution of age- or length-classes may provide valuable complementary information on the reproductive success.

Multiple spawners with few spawns should be sampled at least 6 weeks prior to the initiation of the spawning season (for information on spawning temperatures, consult references such as Scott 1967; Scott and Crossman 1973; Roberts 1988; Nelson and Paetz 1992; Jenkins and Burkhead 1993; Coad 1995) due to an increased variability in the gonad-weight-to-body-weight relationship as the spawning season approaches, because of a lack of synchronization in timing for the second clutch of eggs (Galloway and Munkittrick 2006). Multiple spawners with many spawns, and asynchronous spawners, should be sampled close to the start of the spawning period because of the rapid development of the gonads in both species.

The consequences of sampling at an inappropriate time have been examined using data from the pulp and paper EEM program (cycles 1 to 3). For large-bodied species, fish were sampled outside of the optimal window in more than 33% of previous studies, but interpretation was not strongly affected when optimal and suboptimal studies were compared. However, small-bodied species were sampled at suboptimal times more than 75% of the time, and data collected outside of the optimal windows failed to detect significant effects on gonad or liver size (Barrett and Munkittrick, unpublished data).

Top of Page

Reproductive Strategies:
S, single spawner; M, multiple spawner (few spawning events); MM, multiple spawner (many spawning events); K, exhibit ''skip'' spawning; G, guard nests and (or) provide some form of parental care to their eggs or young; (GSI), strategy was decided based on GSI data over a reproductive cycle; (I), strategy implied or some evidence supporting a particular strategy (e.g., duration of spawning season); ?, data were unavailable to support a reproductive strategy, the strategy was predicted based on observations by the authors of ova sizes in mature ovaries.
Spawning time:
Integers from 1 to 12 to indicate the months in which the species is known or is believed to spawn in Canada. Ranges correspond to all months in that range (e.g., 5-7 corresponds to May, June, and July).
Spawning temperature:
Single temperature in combination with > or < signs, threshold at which a species has been known to initiate spawning activities; Single temperature without > or < sign simply corresponds to a single spawning temperature provided in the literature; Range of temperatures, range at which spawning activities has been observed; ?, spawning temperature data were unavailable or values were predicted based on data for other species of the same genus.
Sampling times:
Late fall, as late as possible before ice cover; 4-6 weeks pre-spawn, four to six weeks before the first spawning event; Spawning, close to the first seasonal spawning event.

a Reproductive strategy as per Barrett and Munkittrick 2010 is S(I). However, there is evidence from data collections in New Brunswick in 2011 and 2012 that common shiners are multiple spawners (Barrett, pers. comm., April 2013).

b Reproductive strategy, spawn time and recommended sampling time were modified from Barrett and Munkittrick 2010 following availability of data from a more recent study conducted in New Brunswick (Barrett, pers. comm., April 2013).

3.6 Verification of Fish Exposure

It is crucial that studies be designed to maximize the possibility of detecting effects if they are present. This can be accomplished by sampling at the proper time of year, with appropriate gear, at appropriate reference areas and during the period of residence in the effluent area. If fish exposure to the mine effluent is uncertain, redesigning the survey (selecting different species, using tracers, changing sampling time or changing exposure or reference areas) or using alternative monitoring methods should be considered for the subsequent phase.

Controversy arises when fish show no differences in characteristics among sites, and there are no indicators of exposure. In this case, it is difficult to determine whether the fish at both sites belong to the same population. In order to verify the exposure of fish to effluent in the exposure areas, and to verify the lack of exposure at reference areas, it may be necessary to select a tracer which accumulates in fish tissue. The selection of a tracer depends on the type of mine involved and the complexity of the receiving environment.

It is possible to infer exposure by examining metal levels in indicator tissues. The indicators and the tissues will vary with the mine type and species being used. In general, gills, liver and kidney have the greatest potential for estimating exposure and bioavailability of metals. Mercury is the only metal element of concern that has been found to accumulate in muscle tissue, so if mercury is a contaminant of concern, dorsal muscle tissue should be analyzed. Blood and bone tissue may reflect exposure to lead, and might be considered if lead is the primary element of concern (Hodson et al. 1984). Bone concentrations are expected to be most indicative of long-term metal exposure, while blood concentrations are indicative of short-term exposure (AETE 1998). For larger species, samples of liver or kidney can be collected. The tissues should be frozen for later preparation and analysis. For small species (< 10 cm), whole body levels can be examined, or levels in the carcass after removing the digestive tract. See section 3.11 for fish usability methods.

Large statistical differences between sites in whole-organism characteristics in a number of parameters give some confidence that the samples are from different populations of fish. If there are no differences between sites, it may be that fish are moving or that there is no impact. Stable isotopes of carbon and nitrogen can be used to document that there are differences in fish residence times, provided that the stressors in question locally alter stable isotopes (i.e., Farwell 1999; Galloway et al. 2004), or there are local geochemical differences that alter stable isotopic signatures and that can be used to demonstrate local residency (i.e., Gray et al. 2004). However, the stable isotopes are not always sufficiently different between sites to be useful, and their suitability has to be evaluated on a site-specific basis (Dubé et al. 2006).

By selecting a sampling time and fish species that have life history habits that may increase the likelihood of exposure, potential exposure can be maximized. For example, for species having spawning movements that take them away from or temporarily into the effluent exposure area, a survey conducted during the spawning season would be ineffective. Thus, for spring-spawning freshwater species, a fall survey would be appropriate. For fall spawners, a spring or summer survey is appropriate. This may not apply to fish in which ova mature rapidly; for example, as some late-spring-spawning cyprinids should be sampled in early spring, rather than in fall when ova may still be immature, it is pertinent to have some background biological information, if possible.

The timing of sampling and the choice of fish species should be made according to normal operation of the facility to ensure that the effluent is present in the environment. Sampling when effluent has not been discharged for long periods (months) should be avoided. However, the selected sampling gear, flow conditions and effluent conditions may limit the preferred season for the survey.

If no fish are captured (or they are captured in reduced density) and there are no fish resident in the exposure area, it could be interpreted that fish are avoiding the exposure area. The suitability of fish species should be evaluated at the end of each monitoring phase, based on the site-specific nature of the results and the site-specific concerns about residency and exposure.

There are some situations where fish may move freely in and out of the exposure area, and no species spend significant periods of time in the effluent. In these cases, the sampling should be designed to maximize exposure time in the effluent area and possibly during periods of optimal gonadal development.

There are two main issues dealing with residency: whether the fish from reference and exposure areas were mixing; and whether the fish captured in the exposure area were indeed exposed. If fish demonstrate exposure, are collected in the exposure area, and demonstrate differences from reference-area fish, there should be no controversy. Follow-up studies can examine other species to see if they demonstrate effects.

If fish demonstrate exposure, are collected in the exposed area, and show no differences, it is outside the scope of the EEM to determine why exposure-area effects are not seen. If subsequent monitoring phases confirm the absence of demonstrated effects and the study design was adequate, it would be concluded that the conditions of the area allow for fish that are exposed to effluent not to be affected, using the current design.

Top of Page

3.7 Power Analysis

The purpose of defining an effect-size and power level is to determine if the sampling program is collecting sufficient information for decisions to be made. The statistical power of a comparison is a function of the sample size, the variability and the target difference set between areas. To determine the sample size for detecting a specific difference, some knowledge is needed about the statistical power level that is acceptable for the decision-making process and the variability of the population.

Top of Page

3.7.1 Power and Significance Level

Earlier cycles of the pulp and paper EEM program set the power (1-beta [β]) at 0.80 and alpha (α) at 0.05. The EEM program now encourages setting α and β equal to one another. If values are set at α = β = 0.10, the sample sizes required to detect the same effect are approximately the same as in earlier cycles. Where possible, provided sample sizes determined by the power analysis are not unreasonably large, mines are encouraged to reduce α = β = 0.05 (the traditional level for alpha). In many statistical programs, the default β is 0.20, and needs to be adjusted. Again, these recommendations are to help ensure that studies are designed to provide a reasonably high probability of statistically detecting a predetermined effect size if it has occurred, (i.e., the power of the test [1-β] should be high). Refer to Chapter 8 for the rationale for setting α and β at equal levels.

It is important to understand that variability and power will vary with the parameter being studied. Fish are not equally variable across all of their characteristics. Reproductive variables are usually as changeable, or even more on a relative scale, than parameters such as length, weight and liver weight (Environment Canada 1997). If effect sizes are also expressed on a relative scale (i.e., as percent differences), any study that can detect a ± 25% difference in relative gonad size can detect similar or smaller differences in other important parameters.

Top of Page

3.7.2 Effect-Size

It is recommended that the EEM program be designed to detect a difference of 20-30% in gonad size, using a recommended power level of 0.90 (1-β). The magnitude of the difference that could be detected for other parameters would be fixed based on the sample size for determining an effect on gonad size. The power for detecting differences in other parameters should be reviewed during study design to ensure that reasonable power is achieved for as many variables as possible. The same approach used to identify a target effect-size for relative gonad weight should be applied to other variables. Sensitivity analyses using population models should be used to explore the consequences of the effect-size chosen for any and all variables (Environment Canada 1997).

An extensive literature review has shown that CESs that have been defined in other programs are often consistent with a CES of around 25% or 2 SDs for many biological or ecological monitoring variables. This value appears to be reasonable for use in a wide variety of monitoring programs and with a wide variety of variables (Munkittrick et al. 2009). Barnthouse et al. (1989) argue that a 10% change in variables would be societally and ecologically significant, although they were concerned primarily with laboratory toxicity tests and not field surveys. Their proposed effect-size was deliberately conservative (small) because of concerns about the uncertainty in extrapolating laboratory results to the field.

When preliminary analyses show that power will be insufficient given reasonable sample sizes, the assessments should be redesigned. Studies are designed site-specifically, and priority should be given to reducing variability rather than increasing sample size. As variability will also vary between sampling campaigns, the target effect-size should not be a fixed number, but rather should be a range of changes that you wish to detect, such as 20-30% difference. Sample sizes can be calculated using methods described in Green (1989); sample size calculators can also be downloaded from the Internet, such as the common one that can be found at http://biostat.mc.vanderbilt.edu/wiki/Main/PowerSampleSize.

A priori power calculations and CES calculations are described in section 8.6.2.1 of Chapter 8.

Top of Page

3.8 Fish Sampling Methods

Sampling methodology should be chosen site-specifically, and capture gear and effort should be focused on methods shown to be successful. The same sampling methods can be used for population and community surveys. The difference lies with the selectivity of the fishing gear. During a community survey, the gear should be as non-selective and non-destructive as possible. For the population survey, which focuses on one or two species, the gear will be more selective. For example, trap netting may be preferred during a community survey, while a one-size mesh gillnet of the appropriate size could be appropriate for a population survey.

Standardized sampling is a priority. Therefore, in situations where sentinel species are the same as for a previous phase, and the sampling techniques used previously were sufficient to capture the target number of each sentinel species, these same sampling techniques should be retained unless good reasons for change, such as unacceptable bycatch, are documented. The sampling techniques and relative effort should be the same in all sampling areas. Pooling of data from different fish-sampling techniques should be avoided, and all methods used should be reported. If more than one gear type is used, the records of fish caught by each method should be reported, and any pooling of data clearly described.

A number of good guidance documents fully describe fish collection methods (Schneider 2000; Portt et al. 2006). Portt et al. (2006) describe the use and efficacy of 1) gillnets; 2) beach seines; 3) hoop, fyke and trap nets; 4) electro-fishing; 5) underwater observation; 6) Gee or minnow traps; and 7) enclosure (drop, pop and throw) traps. However, methods will usually have to be developed and optimized site-specifically.

Top of Page

3.8.1 Bycatch

It may be possible to obtain samples using the bycatch of commercial, research or other fisheries operations in either marine or freshwater situations. The investigator is responsible for ensuring and documenting that sampling procedures and conditions are met (QA/QC), and that fish are exposed. Capture techniques also have to be standardized between sites.

Top of Page

3.8.2 Remote Sensing

Fish abundance near outfalls can be monitored using video or still cameras mounted on remotely operated vehicles. This technique may be particularly effective in rocky and steep areas where use of fishing gear may be difficult. Camera surveys may also be useful in reconnaissance surveys of bottom conditions before trawls or traps are deployed for fishing. Any proposed methods should be clearly outlined in the study design for review.

Top of Page

3.8.3 Alternative Methods

There may be situations where conducting the fish survey is not suitable. The reasons for this are site-specific, but the most common reasons are the presence of hazardous conditions (e.g., strong currents) or the presence of confounding factors such as other effluent discharges in the exposure area, which will make it difficult or impossible to isolate any effects attributable to the effluent being monitored. Under these circumstances, the mine may select an alternative option to the fish survey and/or the fish usability survey. Recommended alternative monitoring methods for the fish survey are mesocosm studies and caged bivalves. Detailed guidance on how to conduct the alternative monitoring methods and interpret the data can be found in Chapter 9.

Top of Page

3.9 Fish Survey Quality Assurance and Quality Control

3.9.1 Field Practices to Improve Data Analysis and Interpretation

The quality of data collected in the field influences the ease of data analysis and interpretation. The preparation of data recording sheets beforehand will save time in the field, and the use of waterproof paper is encouraged. Field conditions, habitat, gear used and information for catch-per-effort calculations should be recorded. The use of the same balance and measuring device for all measurements, and having the same person taking the measurements, will reduce measurement error. If the person taking the measurements is reporting the data to a person recording the measurements, avoid the use of decimal points and report all measurements as digits and not numbers to avoid transcription errors (e.g., report 14.5 cm as 1-4-5 and use units of mm); some numbers can be easily confused when reported orally, such as “fourteen” vs. “forty.”

It is essential that the sampling gear be consistent between the sampling areas, because most sampling methods select for certain age- or size-classes, and thus inconsistent sampling gear between sampling sites could result in detecting false differences (e.g., in age or size).

Top of Page

3.9.2 Quality Control in the Field

This is the first stage of data collection. QA/QC procedures for the fish survey should be outlined during the development of the study plan and should be followed precisely in order to maintain high-quality data. While a QA/QC plan for field sampling can have many components, some of the main procedures are as follows:

• initiate and maintain communication with local government agencies (e.g., fishing licence, dates of fish collection, location of collection, endangered species, etc.);
• all personnel involved in field sampling should have appropriate education and/or training and be familiar with the written standard operating procedures for the survey;
• all safety measures should be identified, understood and adhered to;
• fish collection methods and equipment should be appropriate for the specific water body and fish species;
• habitat descriptions, including possible modifying factors (water depth and current, dissolved oxygen concentration, temperature, substrate classification, evidence of pollution [discolouration, odour, residues], salinity, conductivity, etc.);
• date and time of collection;
• collection methods need to be consistent throughout the study;
• location of sampling areas and fish collection areas documented (geographic coordinates); photograph the collection location;
• record of the number of fish species and incidental species caught per collection stations;
• estimate of catch per unit effort;
• samples from fish (e.g., ovaries, age structures, stomach content) should be placed in appropriate containers;
• suitable preservatives/fixatives (e.g., ovaries--frozen or formalin) should be used;
• all samples should have appropriate labelling;
• all measurements will be taken using appropriate equipment of acceptable accuracy and precision (this should be documented);
• instruments should be calibrated and maintained in good working order (records and methods should be available);
• detailed field notes should be maintained in a bound notebook; and
• chain-of-custody forms and appropriate shipping and storage procedures should be used.

Top of Page

3.9.3 Determination of Sampling Effort

To aid in assessment of expected effort requirements at individual sites, the study plan submitted to Environment Canada should include details on how fish sampling will be performed.

The following are performance-based criteria and guidance to determine a “reasonable level of fishing effort.” Each site is unique. It is uncertain whether fishing success will be achieved at a site just because a certain level of sampling effort has been successful in the past at other sites or even at the same site at other times.

1. The study design should document all details on how the adult fish sampling will be performed, to aid in assessment of effort. Details to include in the study design (where applicable) are:
   
   
   • how and why the sentinel species were selected;
   • who was consulted on the locations and techniques chosen to collect the proposed sentinel species;
   • contingency plans regarding alternative gear and sentinel species;
   • scheduling of dates for work that will be performed so that EEM contacts can be available for consultation;
   • type, location(s) and dimensions of gear (e.g., gillnet, trapnet, hoopnet, fish trap, trawl; in some cases more than one type of gear may be advisable);
   • mesh type (e.g., nylon, cotton fibre or wire, knotted or knotless) and size;
   • proposed level of gear / fishing effort;
   • sampling time (i.e., time of day);
   • sampling duration (i.e., time interval between gear placement and retrieval); and
   • frequency of checks.

   Any preliminary fish survey results or observations made during pre-design activities should be provided where they have guided selection of sentinel species or procedures. The regional EEM contact will review these data and may request further information to clarify sampling procedures.

2. Proper operating procedures should be used. These include use of gear as outlined in the study plan. Gear should be checked at a frequency that ensures the recovery of sentinel species in useful condition and the release of non-target (especially protected and endangered) species. The use of non-lethal and/or selective techniques should be a consideration. A record of the identity and estimated numbers of non-target fish may be a useful addition to contingency plans. The mill and consultant should have a good understanding of the habitat, the characteristics of the species, and the gear being considered.
   
3. Consultation with local experts (e.g., provincial and federal fisheries personnel, Aboriginal groups, individuals and associations involved in local sport and commercial fisheries, the public, and others with knowledge of local fisheries resources) should be conducted to ascertain that the selection of sentinel species, location of nets, timing of collections, etc., are optimal.
   
4. Personnel tasked with the fish collection and sampling procedures should have documented experience.
   
5. Licences for collecting should be obtained from local fisheries agencies.
   
6. Records should be kept that document the operating procedures used (e.g., mesh size, sampling time, location, frequency of checks, etc.). These records may be required in order to properly assess the manner in which the study was conducted.

Although not required, it is recommended that an estimate of catch per unit effort (CPUE) be provided for each sampling area (e.g., number of fish caught per unit of time or area or net). The CPUE information is useful in documenting the effort expended in situations where collection of the minimum number of fish may be difficult.

Top of Page

3.9.3.1 Examples of Calculations of Sampling Effort

Some examples of fishing methods that have been successful in collecting fish in a timely manner are provided below. These are provided as examples to guide consultants in the development and implementation of their study design, and to indicate when it may be advisable to consult with the Environment Canada regional EEM coordinator.

1. Data from two Ontario lakes indicated that 40 individuals of any of 6 warm-water fish species were collected in 1-6 sets for 24 hours duration. The equipment included a 6-x-6-foot trap net. Details of the process are presented in the Ontario Ministry of Natural Resources Fisheries Assessment Unit Newsletter (FAU Update Issue 94-1, OMNR 1994).
   
2. During the Assessment of the Abundance of Cold Waters Ontario Fish Communities Program, the fishing effort recommended to collect 40 Lake Trout in 7 lakes varied from approximately 12 to 120 hours. The equipment included a 46-m gillnet gang with 3 panels of 15.2 m. Details of the process are in the Ontario Ministry of Natural Resources Fisheries Assessment Unit Newsletter (FAU Update Issue 94-2, OMNR 1994). Mesh size should be consistent and selected according to the target species.
   
3. Experience has shown that gillnets made up of 4 panels of 50 m of net, set 24 hours/day for 5 days (equivalent to 24 000 metre-hours of effort) in freshwater systems, should allow for 20 fish of each sex to be collected. This is contingent on correct deployment of the panels and shifting of the panels to cover areas inhabited by the fish. Mesh size should be consistent and selected according to the target species.
   
4. Alternatively, a good strategy would be to initially set a minimal amount of net to decrease bycatch (< 400 m). If fishing is selective enough, and the amount of bycatch acceptable, up to 2 km of a single mesh size has been necessary in small unproductive rivers.
   
5. In marine situations, experience has shown that 48 hours of beam trawling, long-lining using a variety of hook sizes or other methods including traps (alone or in combination), should provide the 20 fish of each sex.
   
6. Consultation with users of electro-fishing technique (large rivers using boat-mounted apparatus) indicates that all fish can be obtained in one day. Procedures enhancing success include operating at dusk or night, passing over the same area at least three times, and using intermittent pulses of current (since a continuous field may actually chase fish away). In small streams, lakes and rivers, more time will often be necessary for sampling because of the difficulties encountered in moving through these environments.
   
7. In 1999, a consultant was collecting fish for an East Coast paper mill located on a tidal estuary. The target species were Mummichogs and the consultants used a 15-x-1.5-m beach seine with 0.5-cm mesh. One end of the beach seine was extended about 10 m out from shore by a technician wearing chest waders, while a second technician held the other end of the seine along the shore. The technicians towed the seine perpendicular to the shore for a distance of 20-30 m and then the outer end was towed into the shore to close the net. Once both ends of the seine were secured on the shore, the upper and lower lines were carefully retrieved to capture any fish enclosed in the net. The consultants found that fishing was most successful at slack high tide. A total of 108 Mummichogs were captured and retained over 4 days of sampling. Total fishing time was 12.5 hours. Many more Mummichogs were captured and released due to the need to balance male and female numbers. In addition, 11 other species of fish were captured and released (Final Report, Repap New Brunswick Inc. Kraft Mill, Second Cycle Aquatic EEM Study, Jacques Whitford Environment Limited, April 2002).

Although the above techniques and gear may apply to a variety of species, these examples are not all-inclusive because each site is unique and the examples are provided as suggested effort only. Local expertise can serve as further advice. The previous examples are adequate guidelines toward catching a minimum of 20 fish per sex, species and area.

Top of Page

3.9.4 Consultation with Regional EEM Coordinators and Implementation in the Field

If all of the above criteria are met and a mine/consultant is having problems meeting the minimum data requirements of the adult fish survey, the owner or operator may deviate from the study design but is required to inform the Authorization Officer without delay of those circumstances and of how the study was or will be conducted. All reasonable efforts should be made to collect target sample sizes of two species of fish and demonstrate due diligence on the part of the mine.

Possible outcomes and options of the consultation with the EEM coordinator are as follows:

1. Continue – Advice on the following situations will depend on site-specific conditions. Set further consultation dates if required.

1. Absence of target species at reference area:
   • continue with current gear and technique;
   • continue at alternative reference area identified in contingency plan;
   • continue at existing reference area with alternative target species, gear and/or technique identified in contingency plan.
2. Absence of target species at exposure area:
   • continue with current gear and technique;
   • continue at alternative exposure area identified in contingency plan;
   • continue at existing exposure area with alternative target species, gear and/or technique identified in contingency plan.
3. Absence of target species at reference and exposure areas:
   • continue with alternative areas, target species, gear and/or technique identified in contingency plan.

2. Postpone (not to continue) – Existing dangerous conditions; sampling conditions (e.g., weather, cold) will not allow collection of fish; alternative gear is not available; no further contingencies are available (e.g., no further alternative species; further investigation is needed):

• design new sampling plan in consultation with regional EEM contact;
• redeploy at a later date with original or alternative areas, target species, gear or technique, but under more favourable conditions;
• set dates for further consultation.

3. Discontinue – If the full complement of fish is not obtained, the absence (or paucity) of fish will be considered a result that will be thoroughly explained in the study findings, taking all possible contributing factors into account. If the minimum number of fish is not caught, this could result in inflated variance estimates. The decision on whether to continue will be influenced first by safety considerations. In all scenarios refer to the contingency plan where appropriate, and set date(s) for further discussion. Field technicians should speak directly with the regional EEM contact. Sentinel species choices will apply to both reference and exposure areas. Pooling of data from different seasons is not valid.

Top of Page

3.9.5 Data Entry

Data entry and preparation of analysis is discussed in Chapter 8, and reporting is discussed in Chapter 10.

Top of Page

3.9.6 Quality Control in the Laboratory

Although much of the survey information is collected while in the field, variables such as fecundity, egg weight, and age are usually determined later in the laboratory. With each measurement, the primary concern of the laboratory QA/QC program is to ensure consistency (precision) and accuracy of the data. The following issues should be considered as part of the measurement procedures:

• all personnel involved in sample processing and analyses should have appropriate education and/or training;
• measurements should be conducted using recognized protocols and methods (these should be documented), and all instruments used should be properly calibrated and maintained (records, methods available);
• keep fish measurements recorded for each fish (target species);
• keep a record of external lesions, tumours, parasites, etc;
• fecundity data, including methods and sub-sampling precision (if applicable);
• aging data, including methods and independent confirmation of estimates;
• maintain records that describe the sample, measurement, and responsible personnel; if possible, a minimum number of individuals should conduct a particular measurement to maintain consistency and reduce measurement error (especially for age determination);
• if sub-sampling is necessary (e.g., fecundity, egg weight), information describing the efficiency and accuracy of the sub-sampling technique should be documented; this information should also be used to calculate appropriate correction or scaling factors (if needed) to minimize possible differences in methods and efficiency;
• all data should be verified; for example, measurements such as fecundity and egg weight should be replicated to ensure precision and accuracy; a recognized expert should verify estimates of age;
• literature and taxonomic keys used for fish identification should be documented;
• archive samples and voucher specimens; and
• maintain detailed sample processing and laboratory notes in a bound notebook.

Top of Page

3.10 Data Analysis

QA/QC concerns regarding data analysis include data verification and validity, repeatability and robustness of statistical analyses, and rigour and defensibility of analyses. For the most part, validation and verification of data depends on the success of QA/QC procedures during field sampling, sample processing, and laboratory analyses (see above). However, there are other considerations regarding the data verification and analyses:

• conduct screening techniques to identify possible transcription errors, outliers and other potentially questionable data points;
• maintain tabular summaries of the general descriptive statistics (sample size, mean, minimum, maximum, standard error, and SD) of fish measurements (e.g., see Table 3-6);
• provide results of assessing assumptions of normality and homogeneity of variance;
• maintain a record of transformation used;
• provide parameter estimates of variability (analysis of variance [ANOVA] mean square error [MSE], ANCOVA MSE, SD for age-to-maturity);
• provide calculations of sample size requirements for each parameter;
• provide a summary of adherence to data quality objectives, standard operating procedures and identification of any QA/QC problems, which should incorporate considerations related to laboratory and field QA/QC;
• to allow reproduction of analyses and results, provide all raw data in an appendix and archive computer data files for an approved period of time after the analyses are published in a report;
• document in detail the methods used for analyses;
• verify that statistical software packages used produce the same output and results as other packages;
• evaluate the robustness of the analyses, (i.e., the results and conclusions should be similar);
• take note of whether outliers are included or excluded, and whether transformations are used, etc.; the objective is to ensure that results are not a function of some manipulation or assumption prior to or during analyses; and
• maintain detailed notes regarding the analyses of the survey data.

Top of Page

3.10.1 Statistical Analysis

The standard statistical assumptions required for many parametric statistical tests are those of independence, normality, and homogeneity of variances. These three assumptions and additional information on data assessment and interpretation are discussed in Chapter 8.

Note: The percentage difference should be reported as exposed relative to reference site. Statistical significance should be given as p-value.
Legend: Diff = difference, stat sign = statistical significance (p-value), ref = reference, exp = exposed, adj = adjusted, sign interax = significant interaction.

3.11 Methods for Analysing Fish Usability

The objective of the question “has there been a change in fish usability due to effluent?” is to determine whether effluent has altered fish in such a way as to limit their use by humans. Fish usability can be affected by altered appearance, altered flavour or odour, or contaminant levels that exceed consumption guidelines for human health and are statistically different from levels measured in the reference area. This section examines fish usability with respect to contaminant levels of mercury.

Mercury is the only metal for which there is a standard Health Canada tissue consumption guideline for humans, and therefore is a pollutant of national concern. Health Canada recently completed a study on mercury and reaffirmed the standard (maximum limit) of 0.5 µg/g with the exception of fresh/frozen tuna, shark, swordfish, escolar, marlin and orange roughy. Provincial and territorial governments are responsible for implementing fish consumption advisories for sport fisheries with the exception of federal parks. Consumption restrictions for sport fish begin at levels above 0.45 mg/g total mercury.

Biological monitoring studies consist of a study respecting fish tissue, if during effluent characterization conducted under paragraph 4(1)(d) a concentration of total mercury in the effluent is identified that is equal to or greater than 0.10µg/L (MMER, Schedule 5, s. 9(c)).

An effect on fish tissue means measurements of concentrations of total mercury that exceed 0.5 µg/L wet weight in fish tissue taken in an exposure area and that are statistically different from and higher than the measurements of concentrations of total mercury in fish tissue taken in a reference area (MMER, Schedule 5 s. 1). At some mine sites there may be reference areas that have levels of total mercury in fish tissue higher than the guideline (e.g., northern Quebec – Schetagne et al. 1997; Schetagne and Verdon 1999); therefore to be considered an effect in fish tissue, there must be a statistical difference between the areas and an exceedance of the guideline using a one-tailed statistical test.

As discussed in Chapter 5, the method detection limit for mercury in effluent has been changed to 0.01 µg/L (0.00001 mg/L) so that the concentration of 0.1 µg/L specified in Schedule 5, s. 9(c) of the Metal Mining Effluent Regulations can be detected with confidence. Analytical methodologies suitable to achieve this level of detection include cold vapour atomic absorption spectrometry (CVAA), cold vapour atomic fluorescence spectrometry (CVAFS), and inductively coupled plasma mass spectrometry (ICP-MS).

In the Guide to Eating Ontario Sportfish, it is suggested that other metals--lead, copper, nickel, zinc, cadmium, magnesium, chromium, arsenic and selenium (Pb, Cu, Ni, Zn, Cd, Mg, Cr, As and Se, respectively)--may be found in fish tissue, but not at levels for consumption restrictions. On a site-specific basis these metals may be identified as a concern if there are human health consumption guidelines from another regulatory agency (e.g., provincial or territorial), that are applicable to the region where the study is being conducted and if the metal for which there is a consumption guideline is present in the effluent. Local consumption and commercial fisheries should be considered to determine which edible tissues (liver, kidney, bones, flesh or even entire fish) should be analyzed. It is recommended that other metals in fish tissue be analyzed where there are site-specific concerns. The Guide to Eating Ontario Sportfish is available at the following website: http://www.ene.gov.on.ca/en/water/fishguide/index.php.

Molluscs can accumulate metals (Cd, Cu, Zn, Pb, Ni, mercury [Hg], As, silver [Ag], and Cr). Field studies suggest that the relationship between mollusc tissue metal concentrations and ambient metal concentrations are influenced by a number of biological, physical and chemical parameters that need to be taken into account. Ultimately, the relationship is metal-specific and depends on the availability of the metal from the dissolved and particulate phase (AETE 1997).

Below is the EEM protocol for fish tissue analysis. Other protocols may be used provided they meet the minimum EEM standards. For example, the Hydro Quebec protocol for the monitoring of mercury levels in fish (Tremblay et al. 1998) has been widely used in Quebec, as this protocol provides an examination of mercury in different age-classes of fish. The protocol can be found on the EEM website.

Top of Page

3.11.1 Selection of the Fish Species

In selecting the location of the sampling areas for the fish tissue sampling, the same factors considered for the fish survey should be taken into consideration. The species selected for tissue analyses should be, if present, sport, subsistence and/or commercial species, including molluscs and crustaceans, where relevant. The fish species used for the tissue analysis may or may not be the same as the species used in the fish survey. On a site-specific basis, the tissue used for the analysis should be chosen based on the portion of the fish constituting the edible portion locally consumed, including the muscle, liver, eggs, hepatopancreas (crustaceans), bone or any other relevant portion. For molluscs, whole soft body parts should be collected, and it may be necessary to produce a composite sample from more than ten specimens to create an adequate sample weight. For lobster or crab, edible tissue (e.g., muscle, eggs, hepatopancreas) should be collected.

Top of Page

3.11.2 Tissue Sample Collection and Preparation

Tissue analyses should be conducted on 8 samples (to achieve 95% power) of a single species from each of the exposure area and reference area, for a minimum of 16 samples. The 8 samples may be tissue from 8 individual fish, or each sample may be a composite of a number of fish; however, tissue from an individual fish should be used in one sample only. If possible, the samples should be of one sex- and age-class. The sex of each fish making up the sample should be reported. If fish are not of the same age-class, the age-classes of the fish should be consistent among sampling areas. Although the largest (oldest) fish of a similar size are preferred, the size specifications set by the responsible authority for fishing regulations in the jurisdiction where the study is undertaken should be respected.

The amount of tissue collected should be appropriate for the analytical method being used. Fish should be used independently in a sample and not mixed between samples. Tissues collected for analysis should be handled in such a way as to avoid contamination from sources such as boat fuel. Each sample should be clearly labelled, sealed in an appropriate contaminant-free container, frozen and forwarded to the analytical laboratory. The individual samples should be homogenized separately and sub-sampled for mercury analysis.

Top of Page

3.11.3 Supporting Analyses: Lipid and Moisture Percentages

Monomethylmercury (MeHg) comprises almost all (95% or greater) of the total mercury found in muscle tissue of fish regardless of the composition of diet sources and exposure water (Bloom 1992). Because of its strong affinity for sulfhydryl groups of proteins, the relative ease with which it passes through the digestive wall and slower depuration rate relative to inorganic mercury, MeHg is accumulated and retained in biological tissues (Clarkson 1994; Saouter et al. 1993).

Lipid concentration has been used to normalize tissue residues among species or within species between seasons, as well as being a key variable in modelling bioaccumulation. Lipid extraction methods by Randall et al. (1991) and the chloroform-methanol extraction method are recommended. Lipid analysis should only be completed when the contaminant being tested is known to be lipophilic.

Percent lipid and percent moisture determinations should be provided for every sample submitted for total metal analysis. Also, percent lipid values should be reported for the replicates analyzed in the same batch with the submitted sample. The percent lipid precision for the replicate samples should be ± 30% for tissues containing more than 2% and ± 60% for tissues with less than 2% lipid. The method for the lipid determinations would be reported and the solvents used clearly specified.

Top of Page

3.11.4 Guidance for Fish Tissue Analysis for Mercury Using Non-Lethal Methods

Tissue analysis for mercury has been traditionally conducted by extracting a fillet from fish. Non-lethal harvesting methods can produce accurate and reliable measures of fish muscle mercury concentrations provided appropriate analytical techniques are used (Tyus et al. 1999; Baker 2002; Baker et al. 2004; Peterson et al. 2005). The use of non-lethal methodologies for mercury analysis are particularly attractive at sites where destructive sampling methods would be detrimental to fish populations, for example, at sites where fish density is low. The purpose of this section is to describe appropriate non-lethal methodologies for tissue sampling and analysis.

Currently, it is recommended that tissue analysis be conducted on 8 samples (to achieve 95% power) from the exposure area and 8 samples from the reference area of a single species from one sex and age class during a lethal sampling study. This guidance should also be followed in a non-lethal survey with the exception of determining sex. It will not be possible to determine for most species if non-lethal sampling is used. However, several studies failed to find differences in mercury concentrations between males and females, although they can differ in energy requirements (Lange et al. 1994; Henderson et al. 2003; Ward and Neumann 1999).

Baker et al. (2004) demonstrated that small tissue quantities collected with two different types of non-lethal biopsy tools (dermal punch and a Tru-Cut™ biopsy needle) provided accurate and precise estimates of mercury concentration in fish muscle relative to benchmark values from the traditional, fillet-style methods and did not reduce survival of recaptured Northern Pike. Tyus et al. (1999) examined survival of Rainbow Trout and Razorback Sucker subjected to tissue collection using dermal punches, fin punches or liver punches and found no significant differences in growth or survival in any of the treated fish.

Top of Page

3.11.4.1 Recommended Methodology

Reliability of the non-lethal technique can depend on the biopsy tool, analytical methodology and tissue sample weight (Baker et al. 2004). The following recommended methodology for extraction of fish muscle tissue using a non-destructive approach is based on the work of Baker (2002) and Baker et al. (2004).

1. Practice – If at all possible, attempt to collect tissue from archived material or incidental mortalities before trying this method on a living fish. Becoming familiar with a technique will minimize possible handling and sampling stress.
   
2. Capture and anaesthetize fish – Prepare two holding containers, one with well-oxygenated water and another containing an anaesthetic (e.g., MS222). Capture fish by non-lethal means such as angling, short-set gill nets or electrofishing and place in the holding container. Transfer fish to the container containing the anaesthetic, one at a time, as necessary.
   
3. Obtain external fish measurements – Once anaesthetized, weigh and measure the fish, and obtain an aging structure (such as a pelvic fin ray) if appropriate.
   
4. Tissue extraction – Two tools currently available for harvesting small tissue samples include dermal punches or the Tru-Cut™ biopsy needle.
   • Tru-Cut™. Remove two or three scales from the dorsal region of the fish just below the dorsal fin using a sterilized needle. The outer barrel is then inserted to a depth of about 1 cm into the fish muscle tissue beneath the scale at an oblique angle (to minimize penetration depth). The 2-cm-long notched needle (inner barrel) is then extended into the flesh. The containment cover (i.e., sharp outer barrel) slides over the extended needle to cut the tissue and capture it within the notch. The needle is then withdrawn, the barrel opened and tissue slug removed with stainless steel (which should be acid washed between samples) or disposable plastic tweezers and placed in a small labelled vial. Samples obtained are approximately 25 mg. At least two tissue samples should be harvested and composited per fish to obtain a sufficient quantity to permit analysis. Baker et al. (2004) indicate that this procedure requires about 10 seconds for an experienced person to harvest a single sample.
   • Dermal punch. The dermal punch harvests a larger quantity of tissue and, for this reason it is the recommended harvesting method if only cold vapor atomic absorption spectrophotometry (CVAAS) is available for tissue analysis. This method can be used on fish greater than 200 mm in size. A few scales are removed and the dermal punch is placed on the skin. Moderate pressure and twisting action is applied to penetrate the epaxial musculature to harvest a small slug of tissue (approximately 60 mg of tissue). As with the biopsy approach, two samples should be harvested per fish and composited.
     
5. Sample preservation – Samples should be frozen using dry ice or liquid nitrogen to prevent decomposition during storage and transport to an analytical laboratory. Samples should be freeze-dried and weighed prior to analysis.
   
6. Infection prevention – Tissue extraction methods, particularly the dermal punch, leaves an open wound that may lead to an increased likelihood of infection. Sterile crazy glue, such as Nexaband™, which acts like a waterproof bandage, should be used to close the wounds to decrease the chance of infection.
   
7. Monitor and reintroduce fish – Once the tissue samples are harvested, return the fish to the holding container until it appears to have recovered and swims normally. The fish is then released back into the receiving water body.
   
8. Chemical Analysis – Selection of an analytical method must consider the accuracy of chemical measurement for small tissue quantities. CVAAS requires a minimum of 100 mg sample weight. Cold vapour atomic fluorescence spectrophotometry has lower detection limits and is better suited to determining mercury concentrations in small tissue quantities. Combustion atomic absorption spectrometry with gold amalgamation is a simplified and rapid procedure for analyzing small tissue quantities for total mercury (Cizdziel et al. 2002).

Top of Page

3.12 References

[AETE] Aquatic Effects Technology Evaluation. 1997. Technical evaluation of molluscs as a biomonitoring tool for the Canadian mining industry. Prepared by Robin Stewart and Diane F. Malley for Aquatic Effects Technology Evaluation (AETE) Program, CANMET, Natural Resources Canada. March 1997.

[AETE] Aquatic Effects Technology Evaluation. 1998. Technical evaluation of fish methods in environmental monitoring for the mining industry in Canada. Prepared by EVS Environment consultants for Aquatic Effects Technology Evaluation (AETE) Program, CANMET, Natural Resources Canada. Draft, July 1998.

Bailey RC, Kennedy MG, Dervish MZ, Taylor RM. 1998. Biological assessment of freshwater ecosystems using a reference condition approach: comparing predicted and actual benthic invertebrate communities in Yukon streams. Freshwat Biol 39:765-774.

Baker R. 2002. Fish mercury database summary – 2001, British Columbia. Prepared by the Aqualibrium Environmental Consulting Group (now the Azimuth Consulting Group, Vancouver BC) for BC Hydro.

Baker RF, Blanchfield PJ, Paterson MJ, Flett RJ, Wesson L. 2004. Evaluation of nonlethal methods for the analysis of mercury in fish tissue. Trans Am Fish Soc 33:568-576.

Barnthouse LW, Suter II GW, Rosen AE. 1989. Inferring population-level significance from individual-level effects: an extrapolation from fisheries science to ecotoxicology. In Suter II GW, Lewis MA, editors. Aquatic toxicology and environmental fate, 11th edition, ASTM STP 1001. Philadelphia (PA): American Society for Testing and Materials. p. 289-300.

Barrett TJ, Munkittrick KR. 2010. Seasonal reproductive patterns and recommended sampling times for sentinel fish species used in environmental effects monitoring programs in Canada. Environ Rev 18:115-135.

Bloom NS. 1992. On the chemical form of mercury in edible fish and marine invertebrate tissue. Can J Fish Aquat Sci 49:1010-1017.

Brasfield SM. 2007. Investigating and interpreting reduced reproductive performance in fish inhabiting streams adjacent to agricultural operations [doctoral dissertation]. Saint John (NB): University of New Brunswick.

Carroll L. 2007. The reproductive cycle of the redbelly dace (Phoxinus eos) [honours thesis]. Saint John (NB): University of New Brunswick, Department of Biology.

Cizdziel JV, Hinners TA, Heithmar EM. 2002. Determination of total mercury in fish tissues using combustion atomic absorption spectrometry with gold amalgamation. Water Air Soil Pollut 135:355-370.

Clarkson TW. 1994. The toxicology of mercury and its compounds. In Watras CJ, Huckabee JW, editors. Mercury pollution: integration and synthesis. Boca Raton (FL): Lewis Publishers.

Coad BW, Waszczuk H, Labignan I. 1995. Encyclopedia of Canadian fishes. Canada Museum of Nature and Canadian Sportfishing Productions Inc. 928 pp.

Dubé MG, Benoy GA, Wassenaar LI. 2006. Contrasting pathways of assimilation: Stable isotope assessment of fish exposure to pulp mill effluents. J Environ Qual 35:1884-1893.

Environment Canada. 1997. Fish Survey Working Group final report. Recommendations from Cycle 1 review. EEM/1997/6.

Farwell A. 1999. Stable isotope study of riverine benthic food webs influenced by anthropogenic developments [doctoral dissertation]. Guelph (ON): University of Guelph, Dept. Environ. Biol.

Galloway BJ, Munkittrick KR. 2006. Influence of seasonal changes in relative liver size, condition, relative gonad size and variability in ovarian development in multiple spawning fish species used in environmental monitoring programmes. J Fish Biol 69:1788-1806.

Galloway BJ, Munkittrick KR, CurrieS, Gray MA, Curry RA, Wood CS. 2003. Examination of the responses of slimy sculpin (Cottus cognatus) and white sucker (Catostomus commersoni) collected on the Saint John River (Canada) downstream of pulp mill, paper mill, and sewage discharges. Environ Toxicol Chem 22:2898-2907.

Galloway BJ, Munkittrick KR, Curry RA, Wood CS, Dunn S. 2004. Identifying a suitable fish species for monitoring multiple effluents in the Upper Saint John River, Canada. In Borton DL, Hall TJ, Fisher RP, Thomas JF, editors. Pulp and paper mill effluent environmental fate and effects. Lancaster (PA): DesTech Publications. p. 169-181.

Gray MA, Munkittrick KR. 2005. An effects-based assessment of slimy sculpin (Cottus cognatus) populations in agricultural regions of northwestern New Brunswick. Water Qual Res J Can 40:16-27.

Gray MA, Curry RA, Munkittrick KR. 2002. Non-lethal sampling methods for assessing environmental impacts using a small-bodied sentinel fish species. Water Qual Res J Can 37:195-211.

Gray MA, Cunjak RA, Munkittrick KR. 2004. Site fidelity of slimy sculpin (Cottus cognatus): insights from stable carbon and nitrogen analysis. Can J Fish Aquat Sci 61:1717-1722.

Gray MA, Curry RA, Munkittrick KR. 2005. Impacts of nonpoint inputs from potato farming on populations of slimy sculpin (Cottus cognatus). Environ Toxicol Chem 24:2291-2298.

Green RH. 1989. Power analysis and practical strategies for environmental monitoring. Environ Res 50:195-205.

Henderson BA, Collins N, Morgan GE and Vaillancourt A. 2003. Sexual size dimorphism of walleye (Stizostedion vitreum vitreum). Can J Fish Aquat Sci 60:1345-1352.

Hodson PV, Blunt BR, Whittle DM.1984. Monitoring lead exposure of fish. In Cairns VW, Hodson PV, Nriagu JO, editors.Contaminant effects of fisheries. New York (NY): John Wiley and Sons. p. 87-98.

Jenkins RE,. Burkhead NM. 1993. Freshwater fishes of Virginia. Bethesda (MD): American Fisheries Society.

Lange TR, Royals HE, Connor LL. 1994. Mercury accumulation in largemouth bass (Micropterus salmoides) in a Florida lake. Arch Environ Contam Toxicol 27:466-471.

Larsson DGJ, Forlin L. 2002. Male-biased sex ratios of fish embryos near a pulp mill: Temporary recovery after a short-term shutdown. Environ Health Perspect 110:739-742.

Larsson DGJ, Hallman H, Forlin L. 2000. More male fish embryos near a pulp mill. Environ Toxicol Chem 19:2911-2917.

Larsson DGJ, Mayer I, Hyllner SJ, Forlin L. 2002. Seasonal variations of vitelline envelope proteins, vitellogenin, and sex steroids in male and female eelpout (Zoarces viviparus). Gen Comp Endocrinol 125:184-196.

Mackay WC, Ash GR, Norris HJ, editors. 1990. Fish ageing methods for Alberta. Edmonton (AB): R.L. & L. Environmental Services Ltd. in assoc. with Alberta Fish and Wildl. Div. and Univ. of Alberta.

McMaster ME, Frank M, Munkittrick KR, Riffon R, Wood C. 2002. Follow-up studies addressing questions identified during cycle one of the adult fish survey of the pulp and paper EEM program. Wat Qual Res J Can 37(1):133-153.

McMullin VA, Munkittrick KR, Methven DA. 2009. Latitudinal variability in lunar spawning rhythms: absence of a lunar patter in the Northern Mummichog (Fundulus heteroclitus macrolepidotum Walbaum). J Fish Biol 75(4):885-900.

Minns CK. 1995. Allometry of home range size in lake and river fishes. Can J Fish Aquat Sci 52:1499-1508.

Munkittrick KR. 1992. A review and evaluation of study design considerations for site-specificity in assessing the health of fish populations. J Aquat Ecosys Health 1:283-292.

Munkittrick KR, McMaster ME. 2000. Assessment of multiple stressors in aquatic ecosystems by directed assessment of cumulative effects using fish populations. In Ferenc SA, Foran JA, editors. Multiple stressors in ecological risk and impact assessment: approach to risk estimation.Pensacola (FL): SETAC Press. p. 27-65.

Munkittrick KR, McMaster M, Van Der Kraak G, Portt C, Gibbons W, Farwell A, Gray M. 2000. Development of methods for effects-based cumulative effects assessment using fish populations: Moose River Project. Pensacola (FL): SETAC Press.

Munkittrick KR, McGeachy SA, McMaster ME, Courtenay SC. 2002. Overview of freshwater fish studies from the pulp and paper environmental effects monitoring program. Water Quality Res J Can. 37:49-77.

Munkittrick KR, Arens CJ, Lowell RB, Kaminski GP. 2009. A review of potential methods for determining critical effect size for designing environmental monitoring programs. Environ Toxicol Chem 28:1361-1371.

Nelson JS, Paetz MJ. 1992. The fishes of Alberta. Calgary (AB): The University of Calgary Press.

Nielsen LA, Johnson DL. 1983. Fisheries techniques. Bethesda (MD): American Fisheries Society.

[OMNR] Ontario Ministry of Natural Resources. 1994a. Nearshore community index netting (NSCIN): indexing the abundance of the warm water fish community. FAU Update 94-1, Fisheries Assessment Unit Newsletter. Sutton West (ON): Lake Simcoe Fisheries Assessment Unit.

[OMNR] Ontario Ministry of Natural Resources. 1994b. Spring littoral index netting (SLIN): indexing the abundance of the coldwater fish community. FAU Update 94-2, Fisheries Assessment Unit Newsletter. Bracebridge (ON): Muskoka Lakes Fisheries Assessment Unit.

Peterson SA, Van Sickle J, Hughes RM, Schacher JA, Echols SF. 2005. A biopsy procedure for determining filet and predicting whole-fish mercury concentration. Arch Environ Contam Toxicol 48:99-107.

Portt CB, Coker GA, Ming DL, Randall RG. 2006. A review of fish sampling methods commonly used in Canadian freshwater habitats. Canadian Technical Report of Fisheries and Aquatic Science 2604. Catalogue number Cat. No.Fs97-6/2604E.

Randall RC, Lee II H, Ozretich RJ, Lake JL, Pruell RJ. 1991. Evaluation of selected lipid methods for normalizing pollutant bioaccumulation. Environ Toxicol Chem 10:1431-1436.

Roberts WE. 1988. The sculpins of Alberta. Alberta Naturalist 18:121-127.

Salazar M, Salazar S. 2001. Standard guide for conducting in-situ field bioassays with caged marine, estuarine and freshwater bivalves. Philadelphia (PA). American Society for Testing and Materials (ASTM). 2001 Annual Book of ASTM Standards.

Saouter E, Hare L, Campbell PGC, Boudou A, Ribeyre F. 1993. Mercury accumulation in the burrowing mayfly Hexagenia rigida (Ephemeroptera) exposed to CH3HgCl or HgCl2 in water and sediment. Water Res. 27:1041-1048.

Schetagne R, Verdon R. 1999. Mercury in fish of natural lakes of northern Quebec. In Lucotte M, Schetagne R, Thérien N, Langlois C, Tremblay A, editors. Mercury in the biogeochemical cycle: natural environments and hydroelectric reservoirs of northern Quebec. Springer-Verlag. p. 115–30.

Schetagne R, Doyon J-F, Verdon R. 1997. Summary report: evolution of fish mercury levels at the La Grande Complex, Québec (1978–1994). Montréal (QC): Joint report of the Direction principale, Communication et Environnement, Hydro-Québec, and Groupe conseil Genivar Inc.

Schneider, JC, editor. 2000. Manual of fisheries survey methods II: with periodic updates. Fisheries Special Report 25. Ann Arbor (MI): Michigan Department of Natural Resources.

Scott WB. 1967. Freshwater fishes of Eastern Canada. Toronto (ON) University of Toronto Press.

Scott WB,. Crossman EJ. 1973. Freshwater fishes of Canada. Fisheries Research Board of Canada Bulletin 184. Ottawa (ON): Fisheries and Oceans Canada.

Secor DH, Henderson AR, Zapalo A, Piccoli PM. 1995. Can otolith microchemistry chart patterns of migration and habitat utilization in anadromous fishes. J Exp Mar Biol Ecol 192:15-33.

Shuter BJ. 1990. Population-level indicators of stress. Amer Fish Soc Symposium 8:145-166.

Swanson SM. 1993. Wapiti/Smoky river ecosystem study. Prepared for Procter and Gamble Ltd./Weyerhaeuser Canada Ltd., Grand Prairie, Alta. by Sentar Consultants Ltd. Calgary (AB).

Tremblay G, Doyon JF, Schetagne R. 1998. Réseau de suivi environnemental du complexe La Grande. Démarche méthodologique relative au suivi des teneurs en mercure des poissons. Rapport conjoint direction principale Communication et Environnement d’Hydro-Québec et Groupe-conseil Génivar inc.

Tyus HM, Starnes WC, Karp CA, Saunders III JF. 1999. Effects of invasive tissue collection on rainbow trout, razorback and bonytail chub. Nor Am J Fish Manage 19:848-855.

Ward SM, Neumann RM. 1999. Seasonal variation in concentrations in mercury in axial muscle tissue of largemouth bass. Nor Am J Fish Manage 19:89-96.

Top of Page

Tables 

Table 3-1 outlines the expected precision and summary statistics of fish survey measurements. Measurement requirements to be assessed include length, total body weight, age, gonad weight, egg size, fecundity, weight of liver or hepatopancreas, abnormalities, and sex. Each measurement requirement is accompanied by its expected precision, and a reporting of summary statistics.

Return to the table

Table 3-2 provides the suggested aging structures for Canadian fish species. Fish species are categorized based on common structure. Comments are offered regarding the relationships between the aging structures and the fish species.

Return to the table

Table 3-3 outlines the fish survey effect indicators and endpoints for various study designs. The primary effect indicators include survival, growth, reproduction, and condition. Each effect indicator is accompanied by the identification of lethal effect and supporting endpoints; non-lethal effect and supporting endpoints; and sentinel mollusc effect and supporting endpoints.

Return to the table

Table 3-4 exhibits generalizations and suggested optimal sampling times for fish species in EEM. Sample times are identified based on reproduction type. Contingent on the reproduction type and its sample time, the relationship between gonad weight and body weight for reference-site females is provided.

Return to the table

Table 3-5 displays the fish species commonly used in EEM, aspects to consider during study design, and recommended sampling times. Fish are identified by family, species, and scientific name.

Return to the table

Table 3-6 outlines the suggested reporting format for the parameters and the resulting regressions required for fish survey analysis in two parts. Table A offers the suggested format for parameter summaries, while Table B shows the suggested format for regression analyses. The percentage difference should be reported as exposed relative to reference site. Statistical significance should be given as p value.

Return to the table

4. Effects on Fish Habitat: Benthic Invertebrate Community Survey

4.1 Overview

4.2 EEM Phases

• 4.2.1 First and Subsequent EEM Phases
• 4.2.2 Magnitude and geographic Extent
• 4.2.3 Investigation of Cause

4.3 Study Design Considerations for the Benthic Invertebrate Community Survey

• 4.3.1 Power analysis and sample sizes
• 4.3.2 Confounding Factors
• 4.3.3 Standard Nomenclature
• 4.3.4 Reporting of Field Station Positions
• 4.3.5 Recommended Sampling Program Designs
• 4.3.6 Reference and Exposure Area Consideration for the EEM program phases
• 4.3.7 Selection of Ecologically Relevant Habitats
  • 4.3.7.1 General Guidance for Habitat Selection
  • 4.3.7.2 Habitat Considerations for the EEM program
  • 4.3.7.3 Habitat Considerations for Marine/Estuarine Habitats
• 4.3.8 Selection of Ecologically Relevant Sampling Seasons
  • 4.3.8.1 General Guidance for Sampling Season Selection
  • 4.3.8.2 Sampling Season Considerations for the EEM Program

4.4 Statistical Considerations for Study Design

• 4.4.1 Determination of Sampling Effort for RCA Designs
• 4.4.2 Determination of Sampling Effort for Field Sub-sampling
• 4.4.3 The Use of Ordination Probability Ellipses for RCA Designs

4.5 Field Methods for Benthic Invertebrate Community Monitoring

• 4.5.1 Sampler Mesh Sizes
• 4.5.2 Sampling Equipment
• 4.5.3 Artificial Substrates
• 4.5.4 Marine/Estuarine Habitat Sampling Equipment
• 4.5.5 Sample Containers
• 4.5.6 Specimen Fixation and Preservation
• 4.5.7 QA/QC for Benthic Invertebrate Field Operations

4.6 Laboratory Methods

• 4.6.1 NABS Certification Program
• 4.6.2 Taxonomic Level of Identification
• 4.6.3 Reference Collections
• 4.6.4 QA/QC for Benthic Invertebrate Laboratory Operations
  • 4.6.4.1 Sorting Efficiency
  • 4.6.4.2 Sub-sampling

4.7 Data Assessment and Interpretation

• 4.7.1 Data Handling Methods
  • 4.7.1.1 QA/QC for Data Input and Verification
  • 4.7.1.2 Dealing with Outliers
  • 4.7.1.3 Unknown, Immature and Non-benthic Organisms
  • 4.7.1.4 Data Reduction and Transformation

4.8 Data Reporting Guidelines

4.9 Effect Endpoints and Supporting Endpoints for the Benthic Invertebrate Community

4.10 Evaluation of Results

• 4.10.1 Effect on the Benthic Invertebrate Community
• 4.10.2 Next Step

4.11 Additional Tools for Focused Monitoring, Weight of Evidence Approaches and/or Investigation of Cause

• 4.11.1 Use of Weight-of-Evidence Approaches to Establish Cause of Effects
• 4.11.2 Lethal and Sublethal Toxicity Tests
• 4.11.3 Analysis of Sediment Cores for Historic Trends
• 4.11.4 Other Benthic Invertebrate Measures and Organisms

4.12 References

List of Tables

• Table 4-1: Recommended sampling program designs
• Table 4-2: Taxonomic keys for benthic invertebrate taxonomic identification in freshwater environments
• Table 4-3: Recommended level of taxonomic precision for benthic invertebrates in marine environment (for lowest practical taxonomic level approach)
• Table 4-4: List of marine and estuarine taxonomic benthic invertebrate keys for Canada

List of Figures

• Figure 4-1: Examples of area, replicate station and field sub-sample spatial scales for a basic control-impact design
• Figure 4-2: Control-impact designs
• Figure 4-3: Gradient designs
• Figure 4-4: Reference condition approach
• Figure 4-5: Impairment stress levels derived for reference sites in hybrid multidimensional scaling ordination space

4. Effects on Fish Habitat: Benthic Invertebrate Community Survey

4.1 Overview

The objectives of a benthic invertebrate community survey for environmental effects monitoring (EEM) are to delineate the magnitude and geographic extent of habitat degradation due to effluent discharge, and to provide an evaluation of the aquatic food resources available for fish selected for the fish survey (see Chapter 3of the present document). However, without a direct comparison between fish diet and the benthic invertebrate fauna, the benthic community survey is mainly aimed at examining habitat degradation. Therefore, the goal of the benthic community survey is to determine if there are structural differences (i.e., total invertebrate density, number of taxa, shifts in the kinds of dominance of taxa) in invertebrate communities in the vicinity of the mine effluent discharge points relative to reference communities. Design considerations will differ depending on whether the mines discharge into freshwater, estuarine or marine receiving waters; this issue is addressed in Section 4.3. It is also recognized that benthic invertebrate surveys will not always use the same study design as the adult fish or water quality surveys because of the different criteria and challenges inherent in the different sampling protocols.

If the benthic invertebrate community survey is conducted in an area where this is possible, sediment samples shall be collected and assessed for particle-size distribution and total organic carbon (Metal Mining Effluent Regulations [MMER], Schedule 5, subparagraph 16(a)(iii)). Water samples shall be taken from the sampling areas when the benthic invertebrate community survey is conducted (MMER, Schedule 5, subparagraph 7(a)(ii)). For more information on water and sediment sampling, see Chapters 5and 7of the present document.

The objective of this Chapter is to provide guidance on the study design and interpretation of results of a benthic invertebrate community survey in relation to EEM requirements. Specifically, this document expands upon 1) study design considerations, 2) standardization of methodologies and 3) data analyses appropriate to the study design. The Metal Mining Effluent Regulations (MMER, Schedule 5) set the requirements and timelines for the benthic invertebrate community surveys. The overall framework of the EEM program is presented in Chapter 1 of this guidance document.

The benthic invertebrate community descriptors used to determine effects (effect endpoints) include total benthic invertebrate density, taxa richness, evenness index (Simpson’s), and similarity index (Bray-Curtis) (MMER, Schedule 5, section 16 (iii)).

Additional community descriptors that could be calculated and reported to assist in data interpretation but that are not used in the determination of effects (supporting endpoints) include Simpson’s diversity index, taxon (i.e., family) density, taxon (i.e., family) proportion, and taxon (i.e., family) presence/absence. For more information on benthic invertebrate community effect endpoints and supporting endpoints refer to Section 4.9 of the present document.

Top of Page

4.2 EEM Phases

4.2.1 First and Subsequent EEM Phases

The first phase of EEM is intended to characterize the benthic communities in major habitats that may be affected by mine effluent and to establish a baseline against which data from future phases can be compared. This phase will also allow for a critical assessment of the need to refine the study design in future phases or the need for the introduction of alternative monitoring techniques. To address the stated objectives of the benthic invertebrate community survey for Phase 1 mines, study design guidelines are presented below.

One specific objective of a Phase 1 survey is to define areas that are relatively homogeneous in terms of habitat class and that have specific ranges in level of exposure to mine effluent.

The study design for the first benthic invertebrate community survey should include:

1. Sampling during an ecologically relevant season
2. Sampling in both reference and high-exposure areas (e.g., area closest to effluent discharge point)
3. Sampling in ecologically relevant habitats
4. One of 7 site-specific sampling designs (Table 4-1)
5. Site-specific supporting variables
6. Standardization of field and laboratory methods

Subsequent EEM phases are intended to confirm the results of the previous phases, help refine monitoring techniques as needed, and determine the factors leading to any detected effect.

Top of Page

4.2.2 Magnitude and geographic Extent

The objective of magnitude and geographic extent studies is to determine the spatial extent of effects on the benthic invertebrate community that are related to mine effluent. Chapter 1provides information on mines conducting magnitude and geographic extent studies and the critical effects sizes that have been developed by Environment Canada to focus additional monitoring.

Magnitude and geographic extent study designs should include:

1. Study and sampling design elements similar to those of previous monitoring, but with more extensive geographic coverage (additional sampling areas)
2. An evaluation of the adequacy of previously sampled areas. The new geographic extent may include additional habitats and substrata such as higher-order streams and lakes or marine/estuarine areas ranging from intertidal to subtidal. If these new habitats were not represented in the reference areas used in previous monitoring, a re-assessment of the adequacy of these references areas is recommended
3. The sampling of additional ecologically relevant habitats, seasons or invertebrate life stages, if this is appropriate for assessing the magnitude of the effect
4. A consideration of other biotic indicators as tools to assess the magnitude of the effect if their use is appropriate and adds value. The list of optional indicators includes biomass and taxonomic composition of periphyton, phytoplankton, macrophyte or zooplankton communities; sampling of other invertebrate life stages, lower-level invertebrate identification, invertebrate biomass, secondary production, additional sensitive habitats or seasons; and toxicity tests on sediment and water

Magnitude and geographic extent surveys may ask the following questions:

Magnitude:

1. How many taxonomic groups are affected?
2. What is the magnitude (e.g., the amount of change in density) of the effect on the taxonomic groups affected?
3. Is there an effect on other benthic community members, such as periphyton or macrophytes, present in the reference area and expected to be present in the exposure area? Note that this is not a requirement of EEM but could be included in a study of investigation of cause.

Geographic extent:

1. What is the geographic area affected?
2. Are the benthic invertebrate communities at the sampling stations furthest from the effluent discharge similar to those living under reference conditions?

Top of Page

4.2.3 Investigation of Cause

For information on investigations of cause (IOC), see Chapter 12of the present guidance document.

Top of Page

4.3 Study Design Considerations for the Benthic Invertebrate Community Survey

Discussed below are various considerations and recommendations which should be examined during the study design process. Benthic invertebrate community survey study designs will be site-specific. The 7 recommended study designs are outlined in section 4.3.5. They attempt to take into consideration factors and possible constraints related to the availability and spatial distribution of suitable reference areas and the spatial extent and heterogeneity of potential impact areas. It should be emphasized that these guidelines, although considered the most applicable generic designs available, are not an exhaustive list of the possible means and ways of conducting a benthic invertebrate community survey. It is assumed that each study leader has sufficient knowledge to apply these recommendations in a sound scientific manner and to determine if unique conditions exist which would warrant modification of the study designs.

Top of Page

4.3.1 Power analysis and sample sizes

For detailed information on power analysis, refer to Chapter 8of this guidance document.

For the first EEM phase it is recommended that the survey consist of the following:

1. At least 2 study areas: reference and high effluent exposure area
2. At least 5 replicate stations in each of the 2 study areas
3. A minimum of 3 field sub-samples to be taken at each station.

Note that, without a priori information on invertebrate density and variability within a station, the number of field sub-samples required to accurately reflect the true density at each station is arbitrarily set at 3. This amounts to a total recommended sampling effort for mines conducting their first monitoring (Phase 1) of 30 benthic samples. Where study designs other than the control/impact design are appropriate, the same minimum sampling effort should be used, although the distribution of areas, stations and samples may differ.

A further recommendation is that the stations be located such that only the dominant habitat class (see section 4.3.7) is sampled. Restricting sampling to the dominant habitat class reduces data variation. Study areas that have extremely heterogeneous habitats, or two habitats that are equally dominant, may require a greater sampling effort than the minimum previously suggested. Further increases in sampling effort, beyond the minimum, are recommended and could include any of the following: addition of one or more reference areas, addition of a low effluent exposure or a very low effluent exposure area, addition of more stations per area, or the addition of more field sub-samples per station. Increases in sampling effort should be determined in consultation with the Regional Coordinator.

Top of Page

4.3.2 Confounding Factors

Note that the Metal Mining EEM Program does not mandate the Metal Mining Industry to investigate effects of other industries or pollution sources on the benthic invertebrate community under multiple discharge situations.

Are there confounding factors that can be resolved by modifying the study design?

The interpretation of benthic invertebrate community effects may be difficult if confounding factors exist within the study area. A careful review of historical or existing data and site characterization information to inform decisions about study and sampling design elements can often resolve problems with confounding factors. For additional information on confounding factors, see Hauer and Lamberti (1996), Culp et al. (2000), and Lowell et al. (2000).

Four categories of such factors include:

Environmental variables: Environmental variables can confound the interpretation of benthic invertebrate community effects if it is not possible to separate the effect of the mine effluent from the effects of differences in natural habitat variables. Augmenting the design to better characterize reference conditions with representation of all habitat types sampled may reduce the problem. This could include locating reference areas in adjacent or further-afield watersheds or by sampling additional reference areas (e.g., moving from a simple control/impact design to a more appropriate design; see figures 4-3 and 4-4 and Table 4-1). Some of the potentially confounding variables that may be dealt with by applying more appropriate study and sampling designs include depth gradients, substrate particle size, rapid effluent dilution, interannual and rare events, and seasonal and long-term variability in physical characteristics such as temperature and flow regimes. It may be possible to judge the influence of environmental or habitat variation by examining correlations between measurements of these factors and measurements of the benthic indicators.

Multiple discharges or historic effects: The potential for confounding effects exists if areas with varying levels of exposure to the mine effluent also have varying levels of exposure to other effluents or stressors, or from historic habitat modifications such as dams or impoundments. If feasible, changing the sampling design by modifying sampling locations may reduce the problem. The collection of sediment cores may also be useful in depositional environments to resolve confounding factors resulting from historic effects (see Chapter 7for more details on sediment monitoring).

Time of sampling: The time of year or the particular year of sampling may confound the interpretation of benthic invertebrate community effects due to effluent. This can be assessed by knowledge of the phenology of benthic invertebrate community species (i.e., relation between climate and life history characteristics) and examination of data collected in previous years from reference areas.

Sampling methods: If standard methods (e.g., sampler types, mesh sizes, taxonomic levels) have not been used consistently within a study or in consecutive studies, any benthic invertebrate community response to the mine effluent may be obscured. It may be possible to examine the data in more detail and convert the data to a comparable level (i.e., convert all taxonomic identification levels to a higher common level). However, in many cases, a redesigned study ensuring that standard methods are consistently applied may be necessary to resolve these problems. Finally, if environmental or logistical conditions exist that preclude the safe and effective collection of samples, the applicability of alternative methods should be examined.

Currently, the only recommended alternative method for the benthic invertebrate component is the application of mesocosms to conduct on-site community bioassays. However, other scientifically defensible monitoring methods that can determine if the mine effluent is having an effect on the benthic invertebrate community may be proposed by the mine. Mesocosms are also useful as an investigation-of-cause tool (see Chapter 12), and their applicability and methodology are described in detail in Chapter 9. Other alternative methods are also described in Chapter 9.

Top of Page

4.3.3 Standard Nomenclature

Standardized definitions for sampling location nomenclature are essential to the EEM program because these will aid national and regional assessments. The following standard terminology for sampling locations should be adopted and applied in a consistent and rigourous manner for all EEM studies with a benthic invertebrate community survey. A schematic representation of these terms is provided in Figure 4-1.

This section defines the terms field sub-sample and replicate station. Reference and exposure areas are defined in Chapter 2. For the basic analysis of variance (ANOVA) study designs (i.e., control/impact or multiple control/impact), where the objectives are to detect differences between or among areas, each reference or exposure area consists of a number of replicate stations (i.e., the replicates for an ANOVA). Each replicate station consists of a number of pooled field sub-samples. Similarly, gradient or reference condition (i.e., reference condition approach [RCA]) study designs use the replicate station as the spatial scale of replication, with field sub-samples being collected as appropriate. See section 4.3.5 for a description of these approaches.

The concept of area is not directly transferable to the gradient or RCA study designs. When using or developing a gradient or RCA study design, a balanced design with similar numbers of replicate stations located within reference and exposure areas is not the basis for comparisons. For example, in RCAstudy designs, exposure stations are individually compared to a distribution of reference stations, which represent appropriate reference conditions. For gradient designs, the lack of suitable reference or exposure areas may be the direct cause for selecting this study design, and thus the ANOVA type terminology is not directly applicable for this approach. Detailed guidance on dealing with these study designs is included in section 4.3.5.

Further guidance regarding the number of replicates and their allocation for different spatial scales and study designs is provided in sections 4.3.5 and 4.4.2.

Field sub-sample

Field sub-samples consist of individual area or time-limited collections of benthic invertebrates (e.g., a grab, core, cylinder, quadrat, kick or U-net sample). To ensure adequate spatial placement of field sub-samples within a station, they should be collected in a random or stratified-random pattern. For many of the statistical analyses used to assess effects in freshwater and marine environments (section 4.9), data from all field sub-samples within a station are pooled, providing a single value of each descriptor from each station.

The first two factors, resulting in an actual physical pooling of the samples, are considered logistical in nature, and their applicability should be determined on a site-specific or method-specific basis. Note that once this physical pooling is done, the potential information from individual sub-samples is lost. In regards to factor 3, if there is a need for additional information to address study design issues (e.g., to examine species area curves or field sub-sampling precision), field sub-samples may be preserved and processed separately. The resulting unpooled data are then available to address the study design issues and can subsequently be pooled electronically for the appropriate statistical analyses. Electronic pooling for the endpoints should be done in such a manner as to be equivalent to results if field sub-samples were physically pooled. This is particularly important for the taxa richness endpoint. Sample calculations for pooled station density and richness are shown below.

For density endpoints, values should be calculated as follows:

Density from pooled field sub-samples = (# in field sub-sample a + # in sub-sample b + # in sub-sample c)/total area of field sub-samples a, b and c.

Note that the resulting number is also the same as calculating the density of each sub-sample and taking an average.

However, the calculation of taxa richness for a station is not equivalent to taking the average taxa richness for the three sub-samples. Station taxa richness should be calculated as follows:

Station taxa richness = all taxa observed at a station in all sub-samples, not the average number of taxa of the three sub-samples.

Replicate station

A replicate station is a specific, fixed sampling location within an area that can be recognized, re-sampled and defined quantitatively (e.g., latitude and longitude and a written description). For each habitat type, a number of replicate stations should be sampled, each resulting in a single composite sample, preferably consisting of ≥ 3 benthic invertebrate field sub-samples. Stations located within the exposure area should be positioned so as to ensure exposure to the effluent plume. Additionally, sufficient physical separation should exist between the replicate stations to allow them to be considered statistical replicates.

The recommended geographic extent of replicate stations for lakes, streams and rivers is as follows:

Area

General information and definitions of reference and exposure areas are presented in Chapter 2.

Sufficient geographic coverage for a single benthic invertebrate study area is recommended for lakes, streams and rivers, as follows:

Top of Page

4.3.4 Reporting of Field Station Positions

Refer to Chapter 2for general information on the reporting of field station positions.

Top of Page

4.3.5 Recommended Sampling Program Designs

The design of the benthic invertebrate community survey is site-specific, and one of 7 benthic sampling program designs listed below is recommended.

1. Control-impact design (C-I)
2. Multiple control-impact design (MC-I)
3. Before/after control-impact (BACI)
4. Simple gradient (SG) design
5. Radial gradient (RG) design
6. Multiple gradient design (MG)
7. Reference condition approach (RCA)

Examples of these designs are illustrated in Figures 4-2, 4-3 and 4-4.

These designs fall into three basic categories with different “philosophical” approaches, as follows:

1. The C-I or MC-I designs (including BACI) are ANOVA-type designs used to detect differences between discrete exposure and reference areas.
2. The gradient (SG, RG or MG) designs are intended to examine changes in community structure along a physical and/or effluent gradient, and are better suited to regression analyses or analysis of covariance (ANCOVA).
3. The multivariate approach of the RCA compares potential “impaired” or test stations to a selection of appropriate reference stations.

It should be noted that there may be some circumstances where ANOVA analyses are applicable to b) and c) above. Alternative monitoring methods (e.g., mesocosms) are also recommended but must be scientifically defensible. A summary of the attributes, applicability and limitations of the sampling designs is presented in Table 4-1 and described in more detail below.

The following descriptions apply primarily to the design of the first and subsequent phases, although special applications for determining the magnitude and geographic extent of an effect are indicated, where applicable.

Control-impact design

The simplest study design for use in EEM is the control-impact (or reference-exposure) design (Green 1979). In rivers and estuaries, this consists of no less than one reference area and a series of downstream exposure areas. For regular monitoring, this should include, at a minimum, one high effluent exposure area. Levels of exposure to mine effluent differ between exposure and reference areas, but should be similar between the stations within each area. Habitat classes sampled should be consistent among areas and, with the exception of exposure level, these areas are to be as similar as possible in terms of substrate, depth, current velocity, water properties, environmental gradients, land use, etc. The first study design employs ANOVA comparisons among areas and is recommended for simple, homogeneous rivers and streams without confounding upstream or near-site discharges from other sources.

The mine may propose modifications to this basic ANOVA approach providing that the modified design is scientifically defensible and addresses the appropriate monitoring questions. For example, if a reference area cannot be located upstream or in adjacent watersheds due to a confounding factor, but a C-I design would otherwise be applicable, a modification of the C-I design may be appropriate. In this case, the design could be modified so that the reference area is “downstream” instead of “upstream” of the point source. The downstream reference area would have to be outside of the exposure area and meet the same reference area criteria as other designs.

This first study design is also recommended for simple, homogeneous estuaries or narrow inlets or bayswithout confounding upstream or near-site discharges from other sources or where the ecologically relevant habitat occurs in spatially discrete but homogeneous patches (i.e., intermittent rocky outcroppings).

Magnitude and geographic extent

The C-I design can be used to ascertain the geographic extent of an effect by first making use of rapid bioassessment protocols (Plafkin et al. 1989) or other available information to approximate how far the effect extends. Following this, a C-I monitoring program can be used that includes the high effluent exposure area and targets additional exposure areas in localities where the effect is suspected to be dissipating (e.g., additional exposure areas located so as to bracket the suspected furthest reach of the effluent effects, together with the previous reference and exposure areas). ANOVA comparisons among areas can then be made to determine the geographic extent of an effect at a given significance level.

Multiple control-impact design

Two of the major problems associated with the use of a single reference area are 1) it can be easily confounded by other factors, and 2) there is a lack of independence among the stations in a single reference area (pseudoreplication) (Hurlbert 1984). In systems where an appropriate reference area is not available due to confounding factors or where it is determined, after a review of historical information, that more reference areas are desirable, the MC-I design should be used. Schematic diagrams of this design for application in mines discharging to large rivers, lakes or coastal waters are presented in Figures 4-2d, e and f. Sampling schemes should be devised so that additional reference areas are located in adjacent watersheds or bays and that comparable habitat classes spanning the range of habitats found within the exposure area are selected.

The design philosophy of both the C-I and MC-I designs is that a specific difference in magnitude of effect between a series of areas is being examined. This lends itself to a classic ANOVA design with associated power analyses. These methods are statistically tractable and can provide indicators as to whether or not there is a biological effect from the mine effluent on the benthic invertebrate community. These designs assume that effluent exposure and habitat conditions are relatively homogeneous among all stations within a sampling area or that effluent exposure is within an acceptable range for a particular defined area.

Before/after control-impact designs

An improvement to the above C-I and MC-I designs is possible when data can be collected both before and after initiation of effluent discharge into the receiving water area. The same considerations discussed above apply for choice of reference (control) and exposure (impact) areas. But the design is further enhanced by collecting data both before and after the facility becomes operational. This kind of monitoring design has been termed a before/after control-impact (BACI) design (Schmitt and Osenberg 1996). Use of a BACI design helps to distinguish effluent effects from natural differences between reference and exposure areas that may have existed before the initiation of effluent discharge.

Detailed descriptions of several kinds of BACI designs and their statistical analyses are available in Green (1979), Schmitt and Osenberg (1996), Underwood (1997), and references therein. In its simplest form, a BACI design entails collecting monitoring data at least once, both before and after initiation of effluent discharge, in both a reference and an exposure area, after which the data are analyzed using an area-by-time factorial ANOVA (Green 1979). In this situation, evidence for an effluent effect is inferred when the area-by-time interaction term in the ANOVA is significant. When the reference and exposure areas have been sampled repeatedly during both the before and after periods, it is possible to use a BACI paired-series analysis; in this case, potential effects are investigated by testing for a change in delta (difference between reference and exposure) from the before to the after period (Schmitt and Osenberg 1996). The design can be further improved by incorporating multiple reference areas (Schmitt and Osenberg 1996; Underwood 1997). Refer to Chapter 2, section 2.2.2.2.2for additional information on baseline data.

Simple gradient and radial gradient designs

Simple and radial gradient designs (Figures 4-3a, b and c) are suitable for situations where rapid effluent dilution precludes the selection of an exposure area that is comparatively homogeneous in terms of effluent concentration. As with the C-I design, gradient designs can be used in cases where no suitable reference areas are available upstream or in adjacent watersheds or bays. Gradient designs are also useful for determining how far along an effluent path effects are observed (i.e., objective of magnitude and geographical extent).

Philosophically, the gradient approach examines departures from expected (non-impacted) “patterns” of correlated biotic and environmental factors over spatial gradients. This is more suited to a regression type of analysis (or equivalent) in which replication (i.e., five stations within an area) is less appropriate than expending a similar effort to obtain accurate measurements of biotic and habitat variables over a sufficiently broad range of the gradient conditions. In the simplest case, a statistically significant effect would be declared if the slope of the regression of a response variable against distance from the effluent source is significantly different than 0. In this approach, a point-source discharge is expected to have a “declining” gradient of effects away from the source, and it is not always feasible to make the judgment that there either “is” or “is not” an effect at a given station. At a certain point along the gradient it is necessary to judge that this effect is no longer measurable or important. Therefore, in gradient designs, reference information is obtained from the stations furthest away from the effluent source.

A gradient does not necessarily imply straight lines or the even spacing of stations within areas. The spacing of stations may be more or less continuous on a gradient away from the discharge, with less emphasis on distinctly different dilution zones and more on adequate geographic coverage, as compared to a C-I design. There are often no “blank” spaces between distinct sample areas, but rather a continuum of sampling stations along the gradient. However, if a change in effluent dilution within the receiving environment is abrupt, more sampling effort may be desirable over these stretches to accurately track rapid changes in mine effects.

SG designs are particularly appropriate for narrow water bodies such as rivers and streams. In wider water bodies such as lakes or open coastal areas, a radial gradient design may be more appropriate. Sampling is conducted away from the effluent source along several gradient transects. As in the MC-I approach, the use of an RG design will provide a larger number of reference sites. Furthermore, a broader geographic area will be sampled, which can be important in non-homogeneous, open lakeshore or marine areas, which often have complex current and circulation patterns or a variety of equally important major habitat classes or gradients.

For RGs, a comparison of regression patterns for each gradient (e.g., regressions of faunal abundance versus distance from the outfall) may help to illuminate the direction and extent of effects. Alternatively, all data from all gradients can be included in one regression, if the comparison is between biotic and physical factors unrelated to geographic or natural habitat factors. If sufficient sampling is done (e.g., RGs), it may be possible to pick and choose unconfounded replicate stations (e.g., homogeneous habitat conditions) to regress a biotic versus a mine-related variable.

Wherever possible, the exposure gradient should be de-coupled or independent from any environmental gradients. A declining exposure gradient may fall along a path with varying depths, but an SG or RG approach may still be feasible if the exposure and depth gradients are not correlated and the differences in depth are not so great as to obscure any effluent effects. In cases where the exposure gradient is correlated with a co-occurring environmental gradient, an MG design may be more appropriate (see next section). Alternatively, a multivariate approach may be necessary to remove the confounding influence of varying depth.

Gradient designs and magnitude and geographic extent

Due to the layout of sampling stations, gradient designs are particularly well suited for determining the geographic extent of an effect. The simplest design for magnitude and geographic extent would be to allocate sampling stations along a gradient from more to less exposed, ensuring that the most distant stations are located well beyond the likely extent of effects. The geographic extent of effects could then be determined graphically by plotting the response variables against distance from the mine and inspecting the data for an inflection point where the response variable asymptotes to the reference condition. Data from sampling stations arrayed in this manner could also be used, together with measured physicochemical data, in a multivariate analysis (e.g., ordination or clustering) used to identify which distant stations tend to group with reference stations and which tend to group with clearly impacted stations. Both of these approaches (graphical plotting and multivariate analysis) look for patterns in the data to determine the approximate extent of an effect; that is, they do not entail hypothesis testing and therefore a power analysis would not be applicable in these cases (in contrast to the C-I approach to magnitude and geographic extent described above).

It is also possible to design a hypothesis-testing gradient program for examining the geographic extent of an effect. This would entail using field sub-samples as replicates (treating stations as areas) and making station-by-station ANOVA comparisons along a gradient to determine where an effect disappears at a given significance level. However, this latter approach might require extensive sampling effort, depending upon the number of stations along the gradient and the required (by power analysis) number of field sub-samples per station.

Multiple-gradient design

In some cases, it may also be useful to compare reference gradients to those exposed to mine effluent. This would be the case when a co-occurring environmental gradient confounds an effluent gradient in the exposure area. By using a MG design (see charts d) and e) of Figure 4-3), it is possible to make statistical comparisons of the exposure area gradient to a similar environmental gradient in a reference area. The reference gradients should be as similar as possible in depth and habitat to the exposure gradient. Potential effluent impacts would be tested for by using ANCOVA to factor out the influence of the co-occurring environmental gradient.

Reference condition approach

The fundamental concept of the RCA is to establish a database of sites that represents unimpaired conditions (reference stations) at which biological and environmental attributes are measured. This database is used to develop predictive models that match a set of environmental variables to biological conditions. These predictive models then allow a set of environmental measurements to be made at a new station and used in the model to predict the station’s expected biological condition (i.e., the biological conditions of the group of reference stations with similar environmental attributes). A comparison of the actual biological condition at the new station with the predicted conditions allows an assessment of the condition of the new station to be made.

The RCA can reduce the need to find nearby comparable reference sites when studying an impacted system, which can be a problem in some traditional approaches. Rather than identifying and sampling upstream reference sites in a river system or next-bay-over reference sites in a lake, the RCA uses a set of biologically equivalent reference sites selected from an existing database to evaluate an exposure site. Provided that it is kept up-to-date, the reference condition database can be used over a number of EEMphases.

The reference condition database is established by an initial standardized sampling program at a wide variety of geographic scales. The same benthic invertebrate community sampling protocol is used in as many ecoregions and stream orders or lakes as are available in a catchment. A number of environmental variables are measured in conjunction with invertebrate sampling. The data are then subjected to a 3-step multivariate analysis in which:

1. a number of invertebrate groups are formed based on similarity of community structure;
2. biological data are correlated with environmental attributes and an optimal set of environmental variables is identified that can be used to predict group membership; and
3. the biological condition of test (exposure) stations is assessed by using the optimal set of environmental variables to predict group membership. How the test station fits, relative to the group to which it is predicted to belong, establishes whether, and to what degree, the station is different from the reference group. Assessment can be made by either the use of the community descriptors, by determining if the site is within the prescribed range of variation observed at reference sites (2 standard deviations [SDs]), or by the use of ordination methods and determining if the exposure site is within the 95% probability ellipse of the matched reference sites.

Depending on the timing and location of the sampling program, it may also be possible to use the resulting database to make ANOVA comparisons between reference and exposure areas.

Once the reference database is established, the RCA can be used as a rapid bioassessment method and to deal with national and local issues using the same database and software. Due to the intensive initial sampling effort required, the RCA would not be considered a practical approach for use by a single mine in a remote location if a reference database is not already available; however, it may be applicable in areas where multiple industries (including different EEMindustrial sectors) are located. In this case it may be practicable and cost-effective for multiple users to collaborate in the development of the reference database. Additional information on the RCA can be found in Bailey et al. (2003).

To assist industry in locating suitable reference sites for the EEM program, the Cooperative Freshwater Ecology Unit of Laurentian University has led the Northern Ontario Benthic Invertebrate Reference Condition Approach (RCA) Biomonitoring Network (Northern Ontario RCA Network). For additional information on this network refer to the following website.

The Canadian Aquatic Biomonitoring Network (CABIN) is a collaborative program developed and maintained by Environment Canada to establish a network of reference sites through the RCA. This information is available to all users interested in assessing the biological health of freshwater in Canada. For additional information on CABIN, please refer to their website.

Top of Page

Note: Multivariate analyses can be performed on data collected using any of the above designs to look for patterns (i.e., not hypothesis tests) that may be useful for highlighting potential areas of concern.

Top of Page

Figure 4-1: Examples of area, replicate station and field sub-sample spatial scales for a basic control-impact design (text description)

Top of Page

Figure 4-2: Control-impact designs (text description)

Top of Page

Figure 4-3: Gradient designs(text description)

Top of Page

Figure 4-4: Reference condition approach (text description)

Top of Page

4.3.6 Reference and Exposure Area Consideration for the EEMprogram phases

The allocation of reference and exposure areas is dependent on the site-specific study design and the phase of the EEM program.

For Phase 1, the objective is to determine whether there is an effect on the benthic invertebrate community in the high effluent exposure area where an effect is more likely to occur. This spatial limitation is designed to concentrate sampling effort in a cost-effective manner. The study design and allocation of reference and exposure areas should be based on this objective.

For subsequent phases, the objectives are to confirm results, detect changes and allow for trend monitoring data. As these objectives are similar in geographic scale to Phase 1, the selection criteria for reference and exposure areas should be the same. However, as with any ongoing monitoring program, the appropriateness of reference and exposure area selection should be re-evaluated as additional information is gained.

For magnitude and geographic extent the objective is to determine the spatial extent of previously identified effects. Thus, sampling should be conducted at exposure areas located farther away from the mine effluent discharge point, until a return to reference conditions is reached. The physical allocation of multiple exposure areas and stations is dependent on the study design. If a confounding factor is encountered before the reference condition is reached in the low effluent exposure area, and this factor cannot be resolved by modifying the study design (see Table 4-1), then the exposure area may be defined to extend only as far as the confounding factor is encountered. Alternative, cost-effective study designs or methods may be applicable (see the Chapter 9and Table 4-1).

In addition, as part of the review of monitoring information, reference areas sampled in previous monitoring should be re-evaluated to assess whether they are adequate for the magnitude and geographic extentprogram. The new geographic extent may naturally include additional habitats such as higher-order streams or lakes. If these new habitats were not represented in the reference areas that were used for previous monitoring, a reassessment of the adequacy of these reference areas will be necessary. The addition of reference areas should also be considered to allow a more balanced design between the number of reference and exposure areas.

If an RCAstudy design was used during previous monitoring, additional reference areas may not be necessary (assuming they adequately represent the habitat types), but it is recommended that a subset be re-sampled to examine the effects of natural temporal variation.

Top of Page

4.3.7 Selection of Ecologically Relevant Habitats

4.3.7.1 General Guidance for Habitat Selection

The most ecologically relevant habitats should be sampled within the exposure areas, and similar habitats should be located and sampled within the reference areas. The selection of the appropriate habitat types requires consideration of the following questions:

• Which habitat type is present in the highest proportion in the exposure area?
• Which habitat, in the absence of human influences, supports the richest assemblages of invertebrates (benthic invertebrate diversity) within the study area?
• In which habitat are the invertebrates most likely to be exposed to sediment or water-borne contaminants for extended periods of time?
• Is historical information available for a particular habitat?

The first consideration is to sample the habitat that accounts for the greatest proportion of the exposure area. However, other factors can override the importance of geographically dominant habitat including the ecological relevance of sampling highly sensitive and diverse habitats, even if they comprise a lower proportion of the study basin. In streams, riffles can support a diverse assemblage of species that display a wide range of sensitivities to water-quality changes. Therefore, the community in this habitat has the potential for greater change than less species-rich communities. In contrast, the fauna of depositional areas, which are generally less rich taxonomically, are of interest during biomonitoring exercises because they may be directly exposed to concentrations of sediment-borne contaminants for longer periods. Consequently, communities in depositional areas may respond to contaminants differently than the more sensitive but less exposed riffle habitats. For additional guidance regarding stream habitat selection, refer to Cuffney et al. (1993), Plafkin et al. (1989), and Meador et al. (1993).

Top of Page

4.3.7.2 Habitat Considerations for the EEMprogram

The decision about which habitat to sample should be based on site-specific considerations. Decisions about the sampling of more than one ecologically relevant habitat during the same survey depend upon the phase of the EEM program.

For Phase 1 and subsequent phases, the objective is to determine if there is an effect on the benthic invertebrate community; therefore, the habitat most likely to exhibit these effects should be sampled. If more than one habitat is determined to be ecologically relevant, effort could be expended during magnitude and geographic extent monitoring to sample all ecologically relevant habitats. This may lessen the potential for missing an effect on a sensitive habitat and/or the necessity of expanding the survey to additional habitats during future monitoring efforts. If questions regarding magnitude and extent can be addressed by additional sampling during the same field trip, it may be cost-effective to do so.

Sampling of additional ecologically relevant habitat types should not be at the expense of reduced sampling effort in the primary habitat of interest. For most biomonitoring studies, sampling a single habitat is intended to reduce the variability inherent in sampling natural substrates. This variability would be even greater if the same level of effort were spread over a range of habitat types.

For magnitude and geographic extent monitoring, the most ecologically relevant habitats should be sampled within the exposure areas, and similar habitats should be located and sampled within the reference areas. The decision regarding number and type of habitats to sample is made based on a review of the previous monitoring results, site-specific considerations and the objectives of magnitude and geographic extent. For example, to determine the geographic extent of an observed effect, additional habitats such as higher-order streams or lakes may become important. On the other hand, if, during previous monitoring, a number of habitat types were sampled but one particular habitat appeared to show responses and the others did not, this habitat type could be targeted during magnitude and geographic extent monitoring.

Top of Page

4.3.7.3 Habitat Considerations for Marine/Estuarine Habitats

A decision needs to be made to sample either depositional or erosional habitats in the estuarine/marine receiving environment. In addition, decisions about sampling intertidal vs. subtidal substrates for estuarine/marine mines will depend on which is the appropriate receiving environment and on the feasibility of obtaining useful samples. In marine/estuarine habitats, the selection of the appropriate habitat types therefore requires consideration of the following questions:

• What habitats are feasible to sample?

The habitat that is most common geographically and most likely to be affected by the effluent should be selected. However, selection of major habitats is partly related to viability of sampling. For example, if the major habitat is a vertical rock cliff to a depth of 300 m immediately outside the effluent discharge outfall, this is rarely feasible for benthic invertebrate sampling without extraordinary equipment. As another extreme example, if the major habitat is intertidal but consists of a steep rock cliff with heavy wave and wind exposure or ice build-up, sampling may not be possible. When multiple habitats are available and appropriate, some choices need to be made. In some cases, more than one habitat may have to be sampled (radial gradient or similar design). Where there is a choice, sampling of soft substrates is preferred because methods are generally more quantitative.

• What is the habitat that is most biologically active or “important?”

When the subtidal environment most exposed to mine effluent consists of both consolidated and unconsolidated sediments, then either both substrates need to be sampled or a decision on which to sample must be made. All other factors being equal, the unconsolidated sediment is more efficiently sampled quantitatively. However, when it is obvious that a coarse sand substrate is almost devoid of macrofauna within the top 10 cm (or the depth of penetration of a sampling device), whereas the nearby rocky reef is extremely rich and an obvious haven for many fish, it is the most “active.” Similarly, if there is an important fishery resource in one major habitat type that is directly exposed to mine effluent, it may be considered the more biologically important.

• Can the ecologically relevant habitat be “classified” according to recognized physical type and characteristic species?

Habitat classification systems have been discussed and reviewed by many researchers and can be useful for comparing “expected” biotic factors with actual biotic factors present in the mine vicinity. Some marine examples include a comprehensive draft document presented to DFO for delineating the Strait of Georgia on the west coast of Canada and northwestern U.S. (Watson 1997). Some relevant references for marine classification for worldwide shorelines to deep coastal areas include Frith et al. (1993), Booth et al. (1996), Robinson and Levings (1995), Hay et al. (1996), and Robinson et al. (1996). Specifically, estuarine classification has been reviewed by Matthews (1993), Scott and Jones (1995), Finlayson and van der Valk (1995), and Levings and Thom (1994). In the U.S., the most widely used system is that of Cowardin et al. (1979) and Cowardin and Golet (1995), with expansions proposed by other authors.

• Is the effluent discharge depth and/or buoyancy most likely to affect the intertidal or subtidal regions?

If the effluent discharge is and remains mainly intertidal, then this should be the targeted habitat. However, if the effluent affects both intertidal and subtidal habitats, then the subtidal is the preferred habitat, because this area is most likely to show impacts in fish. If suitable, both habitats may be sampled. This question should also take into consideration seasonal water column stability changes, which can affect intertidal areas.

• What habitat type is present in the highest proportion?

In many cases, coastal shorelines will be mixed silt, sand, gravel and rock substrates. In bays near freshwater discharge points, there tends to be accumulations of sandy or silty sediment. Estuarine mine sediments will usually have dominantly soft substrates from river-borne material. If there are similar percentages of both depositional and erosional habitats, the preferred habitat to sample is depositional, because this type will accumulate the discharged material from mine effluents and is more likely to present deleterious effects. Erosional substrates tend to be kept “clean” by high current action or wave or ice scouring.

However, if the percentage of solid substrate habitat is much greater than the soft substrate habitat, or if a previously “clean” rocky shoreline has begun to accumulate sediment related to mine discharge, then this may be the preferred habitat to sample.

• Are there confounding factors that may affect benthic communities?

Benthic communities in naturally or anthropogenically confounded sampling areas are problematic to use for interpreting effects from mines. Obviously they should be avoided. For example, in situations where consolidated and unconsolidated substrates are present, only one of these may be outside the influence of the confounding factors. One source of confounding factors that is particularly important in Arctic areas is the seasonal or year-round effects of freezing or ice scour, particularly in intertidal or estuarine areas, which may seriously disrupt surficial communities.

• What is the environment affected by subtidal discharges?

Obviously, the environment most exposed to effluent should be the targeted sampling area and will also determine the type of sampling design used. In an estuary, if the discharge occurs at the surface where there is a strong and permanent surface freshwater layer with little intrusion of saltwater at high tide at depth, then the habitat to be sampled is downstream from the mine. However, where there is a strong tidal intrusion, sampling will have to go upstream and downstream. There are numerous other factors of this type to consider, all of which require detailed site-specific information about the habitats and the pattern of effluent dispersion.

In summary, if there is a choice of habitats to sample, it is recommended first that subtidal habitats be sampled because they tend to have higher diversity and less patchiness in fauna than the intertidal, due to less extreme or harsh habitat conditions. This is particularly true in Arctic regions, where extreme wintertime conditions may eliminate most of the longer-lived fauna that tend to more clearly integrate the effects of contaminants. Second, if there are a variety of suitable habitats, depositional habitats should be chosen, particularly for subtidal areas, because the methodology allows easier and more quantitative sampling procedures. Depositional areas also tend to accumulate contaminants over time, whereas erosional areas may not.

Top of Page

4.3.8 Selection of Ecologically Relevant Sampling Seasons

4.3.8.1 General Guidance for Sampling Season Selection

All benthic invertebrate community surveys should be completed during the most ecologically relevant season. Sampling should occur during a period of effluent discharge and after the receiving environment has been exposed to the effluent for a sufficient period during which effects would reasonably be expected to occur (i.e., generally within 3-6 months).

The preferred seasonal period for sampling is when biological diversity is highest. In general, this corresponds with the seasonal recruitment cycles of benthic organisms (generally related to climate and food abundance). Many insects with freshwater stages reproduce in the spring and fall, although others have multiple cohorts throughout the open-water period. For many lotic habitat types, sampling is conducted during the fall (September/October), when the majority of taxa are present and/or large enough to be collected by the sampling equipment and flow regimes allow access for sampling. In large lakes where the benthic community is often dominated by annelids, crustaceans and molluscs, insect emergence periods and hydrologic regimes are of less importance in determining the sampling period (Rosenberg and Resh 1993).

If historic benthic invertebrate community surveys exist for the system under investigation, it is useful to examine the data and, if appropriate, conduct the survey during similar periods so that the surveys can be compared. Other factors that may influence the sampling period include seasonal flow disruption such as extreme high- or low-flow conditions, freezing and ice scour, mine effluent discharge conditions, type of sampling gear, and feasibility of sampling and field crew safety. Sampling during periods when effluent is not being discharged should be avoided. An understanding of the seasonal patterns and life cycles of the taxa along with changes in the hydrologic regime found within the specific system is also helpful to determine the appropriate timing for the survey. Rosenberg and Resh (1993), Johnson et al. (1993), Rees (1984), Malley and Reynolds (1979), Barber and Kevern (1974), and Jonasson (1955) provide information that may assist in the selection of a sampling period.

Top of Page

4.3.8.2 Sampling Season Considerations for the EEMProgram

It is recommended that efforts be concentrated within a single seasonal sampling period, unless previous data indicate that there is more than one critical time period for the benthic community in the study basin. As this seasonal period should then be used in subsequent studies, it is important to make this decision after compiling all available site-specific data regarding taxa life history characteristics and hydrologic discharge regimes.

Similarly, the sampling season for magnitude and geographic extent monitoring should be the same as previous monitoring unless, upon review of the previous results, there is scientific or logistical justification for a change. Furthermore, additional seasons may be warranted to help determine the magnitude of the response of the benthic community. For example, if the sampling is done at a time when the life stage of a particular invertebrate is not present, then an additional sampling season may be necessary to determine if effects are seen for this specific invertebrate. Bivalves, for example, are not easily sampled in the fall, which is often a critical period for many other invertebrates. In this case an additional season could be added to the monitoring program with the sampling program designed to answer this site-specific concern (i.e., an additional summer sampling trip where methods designed for bivalve sampling are used).

For most marine or estuarine areas the sampling season could be anytime from spring through mid-fall. For temperate marine environments, benthic sampling is usually conducted in late summer or fall as some benthic forms have planktonic larval stages that do not settle to the bottom until later in the season when populations with spring recruits have stabilized. For Arctic areas, the appropriate time period would likely be late summer or early fall, when the long day-length and warmer temperature have allowed some time for growth and development of flora and fauna and there is no sea-surface ice to contend with. In general, reproductive periods and patterns of abundance of benthic species are related to tidal cycles, season and abundance of food supply.

4.4 Statistical Considerations for Study Design

General statistical guidance (e.g., selecting a and b levels and determining sampling effort) is discussed in Chapter 8. This section provides specific guidance on benthic invertebrate statistics including sampling effort for RCA designs and the use of ordination probability ellipses for RCA designs. In addition, a discussion is included which provides guidance on determining the number of field sub-samples which should be taken at a given station and how this field sub-sample data could be used to improve future study designs.

It should be noted here that although a RCA can be used to present results of a benthic invertebrate communauty survey, mines should also submit information related to the effect endpoints required under the MMER (see section 4.9).

Top of Page

4.4.1 Determination of Sampling Effort for RCA Designs

The issue of replication is somewhat different when using the RCA. Replication is at the station scale and, since variation within a station is often much lower than among stations, single samples are taken at stations and variation among stations is used to describe the reference condition. The number of reference replicates is determined by the number of stations in the group to which the exposed station is predicted to belong using the RCA. This is determined when forming the groups of reference stations in the initial classification (see section 4.3), but has been set to a minimum of 10 stations. The variation among the reference stations forming the reference group determines the Type I error, which has been set at 0.1 by using a 90% probability ellipse. Because, in this approach, single-exposure stations are compared to multiple reference stations (minimum of 10), it is not possible to set Type II error, which requires an estimate of the variance associated with a single station. A surrogate can be applied by taking more than one sample at the exposed station, but this is estimating within-station error rather than the appropriate variation at the among-station level. Clearly, Type II station error cannot be determined when there is only one member of the population of exposed stations. Therefore, the power analyses referred to above would not be applicable for the RCA study design. Consequently, RCAstudies should be designed in a way that provides an accurate and precise determination of reference conditions so as to maximize the likelihood of detecting departures from reference conditions at exposure stations, when they exist.

Top of Page

4.4.2 Determination of Sampling Effort for Field Sub-sampling

The objective of multiple field sub-samples at each replicate station is to ensure that the sampling effort will produce an accurate reflection of all the metrics of interest (e.g., taxa richness, density) for each station that is sampled. This is necessary because species may not be homogeneously distributed throughout a station (which is much bigger than the size of the physical sampling apparatus being used). Inadequate sub-sampling 1) gives an imprecise estimate of the true mean for each station and 2) can contribute to an inflated estimate of the true among-station variance, thereby decreasing power.

Therefore, the allocation of field sub-samples within a replicate station depends on the following two inter-related factors that should be considered during any benthic sampling design exercise. However, in the absence of background information, the recommended minimum number of field sub-samples to obtain from each station is 3.

1. The abundance (or density) and degree of aggregation of organisms in relation to the desired level of precision for station estimates

For a given station, the number of field sub-samples needs to be sufficient to give a mean and variance that provide confidence that a representative number of animals has been captured (for a review, see Burd et al. 1990). The more aggregated a community, the higher the variance of mean abundance for each replicate station. Elliott (1977) and Holme and McIntyre (1984) suggested the same simple method of determining the number of field sub-samples to obtain a predefined level of precision. Elliott (1977) suggests that toleration of an index of precision (D) of 20% (i.e., that the standard error is equal to 20% of the mean) is acceptable for most bottom samples. The number of field sub-samples can then be calculated as follows:

where

= the sample mean
n = the number of field sub-samples
s² = the sample variance
D = the index of precision (i.e., 0.20)

Thus, to determine how many field sub-samples (i.e., grabs) per replicate station will provide an estimate with 20% precision, previous data can be used to determine the mean and variance and, thus, the appropriate number of field sub-samples. This determination may vary from location to location along with changes in the mean-to-variance ratio. It is recommended that the number of field sub-samples be calculated for locations that exhibit the highest variability and that the resulting sample size be applied equally to all areas to standardize sampling effort. Although this recommendation will produce better precision in the less variable habitats, it is a conservative approach and maintains equal sampling effort between areas and replicate stations. Also of note is that, with aggregated populations, although the overall mean should remain the same, depending on the scale of the aggregation in relation to sampler size, variance will change with the size of the sampler. Therefore, sample size estimates using preliminary data are only relevant to a sampling program that would employ the same type and size of sampler with which the preliminary data were obtained. In cases where this sampling effort cannot be determined from a previous phase’s data, counting organisms in field sub-samples from the current survey as they are processed and calculating means and variances will allow determination of how many grabs should be processed in the laboratory. However, this a posteriori approach necessitates that a sufficient number of grabs were obtained during the field survey in the first place, so the effort to calculate the sample size within a replicate station a priori should minimize problems due to insufficient sampling effort.

A related approach uses abundance and variance to determine sub-sampling effort and precision and can be used for determining the number of field sub-samples at all replicate stations. It is derived from the relationship between within-station mean abundance and variance across all replicate stations in the area or gradient being evaluated. Downing (1979, 1986) used Taylor’s power law (1961) to estimate aggregation in a freshwater benthic community and thus determine the sampling effort required to reduce variance to an acceptable level. In effect, a given number of organisms are required for each replicate station in order to produce the precision of within-station mean abundance from sampling. Vezina (1988) used the same approach to determine empirically the degree of aggregation inherent in marine benthic communities. This approach entails calculating a power regression equation that describes the log/log relationship between within-station mean abundance and variance across all stations; this provides a formula that is then used to determine the estimated variance expected for a given abundance of organisms in that survey region. From this, the estimated variance for each mean abundance at each replicate station is calculated and then used in the same way as Elliott (1977) to estimate the number of field sub-samples at that replicate station. The difference between the methods of Elliott (1977) and Downing (1979) is that in Elliot’s method the variance used in the equation to determine the number of field sub-samples is based on sample variance, while in Downing’s method the variance used in the equation is based on the variance calculated for the sample mean from the power regression equation for all the samples in the survey region. Furthermore, the index of aggregation (slope) from the power regression equation can then be used to determine the most appropriate data transformation for parametric statistical analyses. Unfortunately, this method assumes that the overall assemblage has a uniform aggregation throughout the study area, which may or may not be true when an external environmental stressor is applied. However, the degree of goodness of fit of the mean and variance data to the log regression equation provides a good indication of how true the homogeneous aggregation assumption is. If there are extreme outliers, they should be taken out of the analysis to avoid skewing the results. Because the aggregation of benthic communities can change as environmental conditions change either naturally or unnaturally, it is wise to review the relationship between mean and variance every time benthic samples are collected. Finally, it should be noted in the above discussion that the “power regression equation” used here to calculate the number of field sub-samples is unrelated to the “power analyses” used to determine the number of replicate stations discussed in the previous sections.

2. The number and distribution of different species in relation to obtaining a representative collection

To determine if sufficient species have been sampled, simple rarefaction methods such as the “species abundance curve” or species/sampling area curve can be used (for a review, see Burd et al. 1990), which compare the number of species obtained vs. number of individuals for different numbers of pooled replicates. This analysis is particularly important in Arctic areas, where diversity may be high, but only on a geographic scale much larger than is feasible to sample (i.e., number of species relative to abundance is high, but abundance is quite low--this can also occur in the deep sea). Because of assumptions inherent in the underlying distribution of fauna related to logarithmic species abundance curves, a more sophisticated approach is the “similarity/sampling area” curve, which uses similarity indices on presence/absence data to determine the sampling effort to obtain an acceptable overall faunal similarity between replicate stations (Weinberg 1978; Kronberg 1987).

If preliminary data are unavailable or unsuitable for determination of the number of field sub-samples to obtain a representative collection of species, a check on sampling effort could be performed very simply. If it is estimated that X number of grab samples per replicate station is sufficient to achieve a data quality objective of retrieving 95% of benthic species present at any replicate station, more grabs can be collected at a few select replicate stations and analyzed. Determination of a taxa richness plateau from these extra samples determines whether the number of grabs were sufficient to achieve the 95% objective (using a species area curve).

Top of Page

4.4.3 The Use of Ordination Probability Ellipses for RCADesigns

A large-scale water quality survey on rivers conducted in the U.K. in 1990 provided the impetus for the development of methods to circumscribe the continuum of responses into a series of bands that represented grades of biological quality (Clarke et al. 1992). The study produced a simplification of the continuum of responses in sites ranging from good to poor biological quality. It was seen as an appropriate mechanism for obtaining a simple statement of biological quality, which allows broad comparisons in either space or time that are useful for management purposes. From a management perspective it is desirable to assign a degree of impairment. This can be done by setting response categories from mild to severe impairment. In the study by Clarke et al. (1992), a number of schemes for categorizing the response were considered and tested. The threshold between unstressed and stressed sites (band A) was set at the 90% probability level (SD = 1.64) for number of taxa and the biological monitoring working party (BMWP) score and 95% for the average score per taxon (ASPT). In Australia, the threshold is set at 2 SD from the reference site mean for the number of taxa. Finally, 95% is frequently set as the limit for determining a biological effect for univariate data and single community descriptors (Lowell 1997). The strategy employed in the U.K. (Wright 1995) to discriminate between degrees of impairment was to quantify the thresholds for stressed and non-stressed sites via the setting of 3 equal-sized bands, as Wright (1995) argued that there was no logical basis for an alternative scheme for dividing up the continuum of sites.

A similar approach can be adopted for defining degrees of impact using multivariate ordination. The reference invertebrate assemblage can be described by its distribution in ordination space, and the assemblage at any given site is characterized by its position in that XY space (Figure 4-5). The greater the similarity between sites, the closer together they are in XY space. Using this approach to set effect size for an invertebrate assemblage, all the reference sites are plotted in XY space together with a test site. The likelihood of the test site being the same as the reference site is quantified by constructing probability ellipses for the reference site only. Reynoldson et al. (1995) selected the 90% probability ellipse as representing the first band, the threshold for a site being considered equivalent to reference. The rationale for using the 90% ellipse rather than the more typical 95% was based on the fact that a multivariate approach will tend to be noisier than univariate measures and therefore a more conservative threshold was deemed appropriate. Sites located in ordination space inside this smallest ellipse (90% probability) would be considered as equivalent to reference and therefore unstressed. Two other probability ellipses are used (Figure 4-5), which are equal in width, to describe further divergence from the reference state, following the argument used by Wright and co-workers (Clarke et al. 1992; Wright 1995). Sites between the smallest (90%) and next ellipse (99% probability) would be considered possibly different; there is a 1 in 10 chance that sites will fall in this band through normal variability. Sites between the 99% and the largest ellipse (99.9% probability) are considered different: there is a 1 in 100 chance that these sites would incorrectly be described as different. And finally, sites located outside the 99.9% ellipse are designated as very different.

Figure 4-5: Impairment stress levels derived for reference sites in hybrid multidimensional scaling ordination space (text description)

Top of Page

4.5 Field Methods for Benthic Invertebrate Community Monitoring

4.5.1 Sampler Mesh Sizes

Benthic samples typically contain varying amounts of fine sediment and debris. To expedite transfer to sample containers, storage, and shipping, these samples should be reduced in the field by sieving. Field sieving should be done, wherever possible, immediately after sample retrieval and before preservation, as many organisms become fragile and brittle after preservation. Various techniques for sieving are available, but most involve washing the sample with a sieve or sieve bucket device.

The recommendation for sieve and/or mesh size for all freshwater mines is 500 µm.

In fresh water, macroinvertebrates are defined as those retained by mesh sizes of 200–500 micrometres (µm) (Slack et al. 1973; Weber 1973; Wiederholm 1980; Suess 1982), although immature life stages of some taxa may be smaller and some adult life stages may be larger.

Note that these mesh sizes are applicable to all equipment used in the field and laboratory (i.e., both the Nitex mesh on the benthic samplers and sieving apparatus).

In some site-specific circumstances it may be desirable for the field samples to be screened for smaller organisms by using a smaller sieve size (less than 500 mm). For example:

1. for comparative purposes, where historic benthic surveys for the system under investigation utilized smaller mesh sizes, or
2. if sampling needs to be conducted, for logistical reasons, at times when organisms are very small. Rees (1984), Barber and Kevern (1974), and Jonasson (1955) provide information on seasonal effects of mesh size.

In these aforementioned cases, it is highly recommended that a stack of screens be used which minimally have the mandatory sieve sizes and then any other smaller sizes, as appropriate. This procedure simultaneously allows site-specific concerns to be addressed and fulfills EEM objectives by allowing for national or regional comparisons to be conducted on the standardized mesh sizes. Sieving with the finest-scale sieve can be done in the field so long as the appropriate fractionation of the sample is performed in the laboratory before processing.

For marine organisms, samples should be sieved with seawater rather than freshwater, since the osmotic shock of freshwater may cause cell bursting and gross distortion of the animals. Where appropriate, field water used to sieve should be screened for ambient organisms with a mesh smaller than the required minimum screen size used for the study. In addition, extreme care should be taken during washing of samples to avoid breakage of specimens, which can greatly reduce taxonomic efficiency and cost-effectiveness. Methods have been described to reduce breakage, particularly in marine samples (Gray et al. 1990).

In marine systems,it is recommended to use astacked set of 1000-mm and 500-mmscreens in the field, with the 500-mm samples being archived and processed only if appropriate.Marine macrobenthos are typically those retained by sieves with 500–1000-µm mesh (Reish 1959; Thiel 1975; Pearson 1975; Holme and McIntyre 1984; Gray et al. 1990). It is estimated that a 1000-mm sieve will retain about 95% of the biomass of marine macrofauna (Reish 1959), while reducing the numbers of juvenile taxa and meiofauna present in samples that respond functionally differently to environmental perturbation than do adult macrofauna (Schwinghamer 1981, 1983; Warwick 1986).

Studying smaller benthic organisms for magnitude and geographic extent in marine systems may include assessment of meiofauna such as nematodes, copepods and smaller oligochaetes or it may include assessment of living and dead foraminifera (Schwinghamer 1981, 1983) or it may include more detailed assessment of juvenile forms of macrofauna. All of these approaches require the use of smaller mesh sizes and/or different samplers (cores may be more appropriate than grabs: see Holme and McIntyre 1984) than are currently recommended here. However, if smaller forms are important, simply adding an additional sieve may not fulfill this function. The appropriateness of the sampling techniques should be assessed for smaller forms. For marine environments, Gray et al. (1990) noted that meiobenthos are most appropriately collected with core samplers, which are not recommended sampling devices for the EEM program. Thus, before simply screening for smaller organisms, appropriate protocols should be implemented.

Top of Page

4.5.2 Sampling Equipment

Two major considerations for benthic surveys are mesh size (see previous section) and quantitative sampling equipment. Quantitative sampling of benthic communities is carried out using devices that sample a known area or volume of habitat, such as grab samplers or stream net samplers. Each sampling device should be non-selective and suitable for a particular substrate. Benthic samples collected from natural substrates provide an indication of past and current stressors. Therefore, samplers that collect benthic communities from the bottom sediments are recommended unless this is not possible due to physical constraints. Samplers are to be consistent within a habitat class among all stations and areas. However, different samplers may be used in the same survey if they are used to sample different habitat classes. For example, if both erosional and depositional habitat classes are sampled, it would be reasonable to use one of the recommended grabs for the exposure and reference depositional habitat, but a Hess-type sampler for the exposure and reference erosional habitat. It is recommended that grab samplers with screens on the top and top-opening gates be used so that the bow-wave ahead of them is reduced and less substrate is lost, and for examination (and possibly sediment chemical analysis) of undisturbed surface layer in sediment samples.

Standardization of benthic samplers facilitates regional and national comparison of benthic invertebrate survey data. Recommendations of samplers appropriate to the various habitat classes encountered during EEM benthic surveys are provided below. Eleftheriou and Holme (1984), Klemm et al. (1990), and Scrimgeour et al. (1993) discuss the options as they pertain to different receiving environments and summarize the advantages and disadvantages of the recommended samplers. The selection of a sampler may also be influenced by the type used in previous surveys of particular systems. To ensure that surveys can be compared with previous historical surveys, it would be useful to use similar sampling equipment. For more detailed information, the reader is referred to the bibliographies on quantitative samplers and appropriate methodologies prepared by Klemm et al. (1990), Eleftheriou and Holme (1984), Elliott and Tullett (1978, 1983), Rosenberg (1978), Downing (1984), and Mason (1991). See also Rabeni and Gibbs (1978) and Alberta Environment (1990).

The standardization of techniques applies not only to the sampling equipment but also to the level of expertise required to correctly deploy the sampler. Crew members should be properly trained in the use of sampling equipment to minimize variation introduced by operator error. For example, when sampling erosional zones in rivers, the depth to which the substrate is disturbed within the net-sampler should be standardized since some individuals may be more energetic in regards to stirring up the substrate than others. The study leader should be well versed in benthic invertebrate sampling and conduct effective training sessions with the crew members performing the field sampling. If training is done effectively, operator error can be eliminated (Reynoldson and Rosenberg 1996).

Depositional habitats: freshwater

Grab samplers are devices with spring-loaded or gravity-activated jaws that “bite” into unconsolidated substrates (sand, silt, mud, etc.) to enclose a defined surface area of the bottom. These devices are generally lowered on a line or cable from a survey vessel to the bottom, sometimes with the aid of a winch. If the sampler type chosen is not suited to the substratum present, it can affect sampling efficiency. Factors that may affect grab sampling include depth of penetration, completeness of closure of jaws, and subsequent loss of material during retrieval. In depositional zones of freshwater rivers or lakes, Ponar or Ekman grabs are suggested as standard samplers for EEM benthic invertebrate surveys. See Eleftheriou and Holme (1984), Klemm et al. (1990), and Scrimgeour et al. (1993) for additional information on samplers.

Erosional habitats: freshwater

Stream-net samplers are devices used for collecting benthic invertebrates in erosional riverine environments. They use mesh of various sizes (but see Section 4.5.1 for discussion on mesh size) to sieve organisms from water flowing through the mesh after disturbance of a known area of the substrate. It is recommended that erosional habitats in freshwater environments be sampled with Neill-Hess cylinder-type samplers that allow unit area (typically 0.1 m²) estimates to be made. One drawback to cylinder samplers in streams is a potential incompatibility with size of substrate. In some systems, mean particle size may be too great for the Neill-Hess cylinder to effectively sample the benthic invertebrates. In such cases, a U-net sampler (Scrimgeour et al. 1993) can provide area-limited samples and be adjusted accordingly to the size of the substrate. This sampler has been used successfully for a range of substrate sizes (Glozier 1989) and can sample either individual stones or a defined area. Kick-net samplers do not provide an area-delimited estimate, but have been used widely in the United Kingdom, the United States, Australia and Canada in large-scale monitoring programs (Reynoldson et al. 1995). Kick sampling is particularly appropriate for the reference condition approach, where many stations are sampled. A timed kick sample is taken at each station to estimate benthic community descriptors. Standardization of kick-sampling techniques is essential for comparative purposes and can be obtained with minimal training (Reynoldson and Rosenberg 1996). The kick-sampling method involves a single composite sample collected at each station by a 3-minute travelling kick method (Reynoldson et al. 1997). Note that separately preserved field sub-samples are not required for the kick-sampling technique recommended for RCA.

For difficult habitats (e.g., very deep, slow-flowing areas or areas with hard substrates) alternatives such as the metal quadrat or airlift system may be available. However, for national or regional comparative purposes, the list of recommended samplers should be sufficient for sampling the majority of ecologically relevant habitats. If habitats are extremely difficult to sample, alternative approaches may be considered.

Top of Page

4.5.3 Artificial Substrates

The use of artificial substrates for benthic invertebrate collections is generally not recommended as a sampling protocol in EEM.

There is no advantage to be gained from using artificial substrates where conventional sampling techniques provide at least as reliable data without the many drawbacks and difficulties of artificial substrates (AETE 1995). Artificial substrates do not collect a representative sample of the indigenous benthic invertebrate community at the site where they are placed, but rather select for mobile, drift-prone species of hard substrata. In addition, artificial substrates do not effectively monitor the effects of sediments or sediment-bound contaminants on aquatic biota because sediment-dwelling taxa tend to be under-represented in artificial substrate samples. The invertebrate community represented by artificial substrates indicates conditions during the period of exposure only and does not integrate long-term effects. Therefore, the use of artificial substrates for benthic invertebrate collections may fail to indicate the effects from effluents, particularly where non-mobile species, sediment-bound contaminants or longer-term integration of effects are important. However, it is recognized that there may be a limited number of cases where there is either a long history of artificial substrate use in a particular ecosystem or extreme habitat conditions (e.g., very deep, fast-water systems) where the use of artificial substrates is the only feasible field method available. In these cases, the use of artificial substrates may be considered along with other alternatives--provided this method can determine if there are effects on the benthic invertebrate community in a scientifically defensible manner.

Top of Page

4.5.4 Marine/Estuarine Habitat Sampling Equipment

Depositional Habitats: Marine/Estuarine

Depositional habitats in marine environments can be sampled with the Smith-McIntyre grab, a modified Van Veen grab, which is suitable and available in Canada. A good review of marine sampling methods is available in Eleftheriou and Holme (1984). However, in shallow subtidal areas where there is not enough water depth to allow the deployment of the larger grabs, a smaller or mini (petite) Ponar grab can be used. This grab is deployable from small inflatable boats and can be retrieved by hand.

Intertidal soft substrates may be sampled using any device that demarcates an area of at least 0.1 m². The soft substrate is then removed to a standard depth of 10 cm using an appropriate device. Note that, in general, the lowest intertidal level available for sampling is preferred because less harsh physical conditions promote higher species richness and abundance.

Erosional Habitats: Marine/Estuarine

In marine/estuarine environments, large unconsolidated sediments such as gravel may be sampled with grab samplers. If not, hard substrates in erosional habitats (intertidal and subtidal) should be sampled using quadrats with a minimum area of 0.1 m². However, some other quantitative techniques may be recommended to collect marine shellfish and other large species. These may include hand collection by divers, remote sensing techniques from defined surface areas (Eleftheriou and Holme 1984; Gray et al. 1990), and collection from defined boundaries along transects. An outline of a marine sampling protocol is described in a series of Puget Sound Estuary Program Reports (Tetra Tech 1986a, 1986b, 1987). When done properly, photographic surveys can be quantitative, at least for larger epibenthic organisms (c.f. Burd et al. 1990). Processing costs tend to be considerably less than for soft-bottom surveys using grabs or cores.

The intertidal zone should be sampled if the effluent plume impinges substantially on it. Determining the tidal level of greatest interest for examining mine impacts will involve logistical considerations. Basically, the lower in the intertidal area the surveys can be conducted, the better, since less harsh conditions create less patchiness and higher diversity in flora and fauna (inter-sample variability). Coastal plants and animals in this habitat typically exhibit vertical distributions that reflect gradients in environmental parameters such as air exposure, temperature (including freezing), salinity, light intensity and daylength, abrasion due to logs or ice, and wave shock. These gradients should be considered in planning and undertaking biological surveys in the intertidal environment. Sampling protocols for this area will be somewhat different from those described in the earlier sections (for review see Gray et al. 1990). Wherever possible, semi-quantitative surveys using quadrat areas of 0.1 m² should be done. Determining the substrate or habitat type to be sampled depends on sampling limitations and the dominant habitat present (see section 4.3.7 for discussion of dominant habitat selection). However, if it is not feasible or ecologically sound to collect samples, then visual surveys are recommended. If approved, a visual survey would include approaches such as recording and mapping (at a gross scale of 1:5000) the major biological features for assessment of gross changes in the biological community.

Top of Page

4.5.5 Sample Containers

Environment Canada’s Guidelines for Monitoring Benthos in Freshwater Environments (EVS Environment Consultants 1993) specify that sample containers should:

• be large enough to ensure the sample takes up no more than 50% of the container volume, with the remainder of the space allocated for preservative;
• be sturdy enough for routine handling and transportation;
• be leak-proof;
• have physical and chemical properties that are not affected by the fixative/preservative; and
• conform to regulations concerning the transportation of dangerous goods.

Top of Page

4.5.6 Specimen Fixation and Preservation

All samples should be fixed in the field in a 10% buffered formalin solution to prevent damage to freshwater and marine worms. Formalin is also important for the proper preservation of most aquatic insects. Preservation directly in ethanol often results in soft, difficult-to-handle specimens. After preservation in the field, samples should be gently mixed several times to ensure that the preservative has thoroughly penetrated any fine material that may be present in the sample. Because formalin is a carcinogen and an irritant to workers, gloves and protective eye gear are needed and should be considered mandatory safety equipment. Furthermore, unbreakable sample jars should be sealed with parafilm, double-bagged for transport back to the laboratory facilities and adequately labelled. The samples should be preserved as soon as is practical after sampling to prevent predatory invertebrates from preying on others in the samples.

Top of Page

4.5.7 QA/QC for Benthic Invertebrate Field Operations

An outline of the quality assurance/quality control (QA/QC) recommendations for the field components of the benthic invertebrate community survey is presented below.

Field sampling is the first stage of data collection. QA/QC procedures for the benthic invertebrate survey are outlined in the study design and should be followed precisely to maintain high data quality. Field standard operating procedures (SOPs) should specify sampling equipment and protocols appropriate to the study. A QA/QC plan for field sampling has many components. Some of the main procedures are listed below:

1. All personnel involved in the field sampling should have appropriate training and experience with field equipment and objectives.
2. All safety measures should be identified, understood and adhered to.
3. Collection equipment should be appropriate for the specific water body and selected invertebrate group, and should be checked frequently and maintained regularly.
4. There should be some a priori criteria for acceptability of samples obtained and clear directions if acceptability guidelines fail (i.e., when to retake a sample; grab sample penetrations of 10-cm depth would be considered an acceptable sample, Gray et al. 1990). Also, sampling methods need to be consistent throughout the study.
5. A visual description of benthic grab samples should be recorded to describe sediment color, odour, texture and debris.
6. Contamination during chemical sampling should be checked by means of trip blanks and equipment rinsates.
7. Field sieving, if necessary, should be done as soon as possible after retrieval of samples.
8. Samples should be stored in appropriate containers with appropriate preservative to prevent breakage and spoilage.
9. All sample containers should be appropriately labelled.
10. Detailed field notes should be maintained in a bound waterproof notebook.
11. Chain-of-custody forms and appropriate shipping and storage procedures should be applied.

For further information regarding all aspects of QA/QC procedures for benthic invertebrate programs, refer to the 1999 AETE report (Beak 1999).

4.6 Laboratory Methods

For information pertaining to sample sorting and sub-sampling, please refer to the Revised Guidance for Sample Sorting and Subsampling Protocols for EEM Benthic Invertebrate Community Surveys, which can be obtained from the EEMwebsite (http://www.ec.gc.ca/esee-eem/default.as.

Top of Page

4.6.1 NABS Certification Program

The accurate identification of aquatic benthic invertebrates is crucial to monitoring programs like the metal mining EEM program. The North American Benthological Society (NABS) implemented a certification program for benthic invertebrate identification. The program tests the candidate’s knowledge and skills in aquatic invertebrate taxonomy and ensures that individuals are providing high-quality identifications. It is recommended that the identification of aquatic benthic invertebrates be conducted by an individual who has completed the NABS certification program. For additional information, please refer to the following website: http://www.nabstcp.com/.

Top of Page

4.6.2 Taxonomic Level of Identification

Identification of the benthic invertebrates sampled should be adequate to meet the objectives of the assessment program. Research indicates that family-level identification provides sufficient taxonomic resolution to detect community responses to human disturbances (Warwick 1988a, 1988b; Bowman and Bailey 1997). As discussed below, the level of taxonomic resolution used may vary across the different monitoring phases, with finer taxonomic resolution needed to detect more subtle environmental impacts.

The recommended level of taxonomic identification is family for the first and subsequent monitoring of freshwater systems. All summary statistics and descriptive metrics should be calculated and reported at the family level for submission to the first and subsequent monitoring interpretative reports. Organisms that cannot be identified to the desired level of taxonomic precision should be reported as a separate category in the fundamental data set. It is recommended that investigators use taxonomic keys appropriate to the geographic region of study. Table 4-2lists taxonomic references typically used for various groups of freshwater organisms.

For some phases, a lower taxonomic level may be recommended, depending on the questions and objectives of the study. The lowest practical level (LPL) has been defined as genus for most insects and the lowest level possible without special procedures (dissection, microscopy) or reliance on specialist for other groups (Taylor 1997). This definition can be used as a guide if lower level identifications are desired for magnitude and extent and investigation of cause monitoring.

There may be site-specific conditions that warrant a lower taxonomic level for some or all familial groups. For example, historic benthic invertebrate information may be identified to lower taxonomic levels, and it may be desirable to identify subsequent surveys to a similar level for comparative purposes. If a lower taxonomic level has been used, either in historic data or during a current survey, the summary statistics and descriptive metrics can be reported at this level, provided a summarized data set at the family level is also included.

Two objectives of the magnitude and geographic extent survey may require different levels of taxonomic resolution. Determination of the geographical extent of the effect may be addressed adequately with family identification. Family-level identification would provide the information necessary for calculating and reporting the required summary and descriptive statistics related to extent of the effect. This first objective is similar in scope to Phase 1 and subsequent monitoring, the major difference being the addition of low effluent exposure areas further from the effluent discharge.

The second objective of magnitude and geographic extent monitoring, when determining the magnitude of the effect, may use family identification or it could warrant investigation at a lower taxonomic level. The question of magnitude of effect with regard to taxonomic level can be addressed using the following question:

• What is the magnitude of the effect on specific taxonomic groups that may be sensitive to the site-specific mine effluent characteristics (e.g., how many groups within a sensitive family are affected)?

Addressing the magnitude of effect during magnitude and geographic extent monitoring can be accomplished by using one of the options outlined below:

• Identify all samples collected to the lowest practical level. Establishing the magnitude of effect in this way provides additional information that may be useful for the study design exercise at the outset of the investigation of cause.
• Re-analyze families that were significantly affected during the first monitoring to identify indicator taxa that can be used to assess the magnitude of effect at stations farther afield. For example, if an effect during the first monitoring was observed for the family Baetidae (order Ephemeroptera), all Baetidae could be identified to a lower level (e.g., genus) for the magnitude and geographic extent monitoring program. This approach would catalog the “sensitive” taxa within the family, and the magnitude of effects would be established by examining this subset of sensitive taxa.
• Other scientifically defensible approaches may be used to identify magnitude of effect as required.

In marine/estuarine environments, it is recommended that all benthic invertebrate organisms be identified to the family level. In interpretative reports, all summary statistics are calculated and reported to the family level. Various authors have examined the utility of using higher taxonomic classifications for environmental monitoring of organically polluted sites in Europe (cf. Warwick and Clarke [1993] and references therein). For marine benthos, juvenile or non-adult fauna should be identified and enumerated separately from adults, as they show different patterns of response to environmental effects.

Though mines may proceed with benthic invertebrate identifications to a lower level, the recommended level of identification for data reporting and determination of effects is the family level. There may also be site-specific conditions that warrant a lower taxonomic level for some or all familial groups. For example, historic benthic invertebrate information may be identified to lower taxonomic levels and it may be desirable to identify subsequent surveys to a similar level for comparative purposes. If a lower taxonomic level has been used, either in historic data or during a current survey, the summary statistics and descriptive metrics can be reported at this level--provided a summarized data set at the family level is also included.

For marine samples it is suggested that, if sufficient numbers of specimens are available in the reference collections, they could be used for a further purpose: to develop a size and biomass database for each mine as another indicator or tool (see section 4.11.4). For these purposes, 5 to 10 representative specimens per taxa are recommended, with mean width, lengths and blotted wet weights recorded for each group.

Top of Page

4.6.3 Reference Collections

Consistency in taxonomic identifications within and between surveys is essential to obtaining useful information on environmental effects monitoring. Therefore, for comparative purposes and quality control of taxonomic identification, the maintenance of a reference collection of organisms is recommended. In addition, it is recommended that an independent professional taxonomist verify the identifications in the collection. Museums are sometimes prepared to perform this service when remote areas are included in the study and new specimens or distribution records are likely. Reference collections have several benefits, including their use in confirming identifications, ensuring consistent taxonomy between surveys, and the training of personnel. Protocols for establishing and maintaining reference collections for benthic invertebrates are detailed in a report prepared for Environment Canada’s Fraser River Action Plan (Green 1994). Following the recommendations of this report, each mine (or group of mines) should compile and archive a complete reference collection with several specimens of representative-sized individuals for each taxon. The collection should encompass representative organisms from each area in the survey, be labelled according to the location and date of collection, and updated as appropriate (i.e., when a taxon is collected). This type of reference collection will not occupy a large space: a small cupboard should be sufficient and should be in the custodianship of the mine. If a mine does not have the facilities or personnel to maintain their own reference collection, universities or museums may be willing to fulfill this function. However, since considerable effort is involved in the long-term maintenance of preserved biological material, the quantity of material submitted should be minimized.

Top of Page

Table 4-3 lists recommended levels of taxonomy desirable for major taxonomic groups of marine benthic organisms. In general, the level of taxonomy should be consistent in each major group for all samples from a survey and also from survey to survey. Organisms that cannot be identified to the desired level of taxonomic precision should be reported as a separate category in the fundamental data set at the finest level of taxonomic resolution possible. Since the accuracy of the taxonomic work depends on the availability of up-to-date taxonomic literature, a basic library of identification keys is essential. Keys appropriate to the geographic region of study are recommended. A detailed list of taxonomic references for marine and estuarine habitat is found in Table 4-4. Microscope slide mounts should be prepared for taxa requiring detailed microscopic examination for identification. This may involve various steps, including dissection, clearing and staining. Slide preparation techniques are listed in Klemm et al. (1990). For marine benthos, juvenile or non-adult fauna should be identified and enumerated separately from adults, as they show different patterns of response to environmental effects. All identifications should be carried out or verified by a qualified and experienced taxonomist. Existing reference collections may be useful as well. An example is the Atlantic Reference Centre at Huntsman Marine Station in St. Andrews, New Brunswick. Photographic iconographs have been used to advantage (Camburn et al. 1984–1986).

Table 4-3: Recommended level of taxonomic precision for benthic invertebrates in marine environment (for lowest practical taxonomic level approach) (text description)

*Nematodes and copepods (e.g. harpacticoida) are meiofauna, and only a fraction of specimens will be captured by a 500 μm or 1000 μm screen. Therefore, numbers are not representative and should be excluded from analyses (Holme and McIntyre 1984).

Top of Page

Table 4-4: List of marine and estuarine taxonomic benthic invertebrate keys for Canada (text description)

Abbot 1974 (seashells)
Abbott et al. 2001 (Mollusca)
Appy et al. 1980 (Bay of Fundy polychaetes)
Austin 1985 (Pacific invertebrates)
Baker 1980 (Tubificid species)
Banse 1972; Banse and Hobson 1974 (polychaetes)
Berkeley and Berkeley 1952 a,b (Pacific Annelida)
Blake 1971 (Polydora, East Coast)
Blake 1991 (Polychaeta, North Atlantic)
Blake 1988 (Phyllodocidae [Polychaeta], Atlantic)
Bousfield 1960 (Atlantic seashells)
Bousfield and Hendryks 1994, 1995 a, 1995b (Pacific Amphipoda)
Bousfield and Hoover 1995 (Pacific amphipods)
Bousfield and Kendall 1994 (Pacific amphipods)
Bousfield 1973 (Amphipoda, Atlantic)
Brinkhurst 1982 (Oligochaetes)
Brinkhurst and Baker 1979 (Marine Tubificidae) (Oligochaeta)
Brunel et al. 1998 (Catalogue of the invertebrates of the Gulf of St. Lawrence)
Butler 1983 (Pacific shrimps)
Clark 1924 (Holothuroidea)
Clark 1915 (Ophiuroidea)
Coates 1980 (B.C. Enchytraeidae)
Coe 1912 (Echinodermata, Atlantic)
Coe 1943 (Nemertea, Atlantic)
Cutler 1973 (Sipuncula)
Fauchald 1977 (Polychaeta)
Fournier and Petersen 1991 (Polychaeta)
Gibson 1994 (Nemertea)
Gosner 1971
Graham 1988 (Gastropoda)
Hart 1982 (B.C. crabs)
Hobson and Banse 1981 (B.C. polychaetes)
Hyman 1940 (Polycladida [Turbellaria, Atlantic])
Hyman 1944 (Turbellaria, Atlantic)
Keen and Coan 1974 (Mollusca)
Knight-Jones 1978 (Spirorbidae [Polychaeta], Pacific and Atlantic)
Knight-Jones 1983 (Sabellidae [Polychaeta])
Kozloff 1987 (Pacific N.W. invertebrates)
Lambert 1981 (B.C. sea stars)
Laubitz 1972 (Caprellidae)
Light 1977 (Spionidae [Polychaeta], Pacific)
Morris 1951 (Mollusca, Atlantic)
Pettibone 1963 (Polychaeta, Atlantic)
Pettibone 1992 (Pholoidea, Polychaeta)
Pettibone 1993 (Polynoidae, Polychaeta)
Pohle 1990 (Decapoda, Atlantic)
Sars 1895 (Amphipoda)
Sars 1899 (Isopoda)
Sars 1900 (Cumacea)
SBMNH 1994a,b,c; 1995a,b,c; 1996a,b,c; 1997a,b
Schultz 1969 (Isopod crustaceans)
Smith 1964 (Marine invertebrate keys, Atlantic)
Squires 1990 (Decapoda, Atlantic)
Steele and Brunel 1968 (Amphipoda)
Tattersall and Tattersall 1951 (Mysidacea)
Thorp and Covich 1991(Freshwater invertebrate keys)
Ushakov 1955 (Polychaeta)
Wallace 1919 (Bay of Fundy Isopoda)
Watling 1979 (Cumacea, Atlantic)
Weiss 1995 (Marine macrofauna)

Top of Page

4.6.4 QA/QC for Benthic Invertebrate Laboratory Operations

In the laboratory, invertebrate samples are processed and counts of the various taxa are made. It is recommended that the sorted, preserved samples from each survey be retained in an appropriate storage facility for at least 6 years, or until it is determined that no further information will be required from the samples. Samples should be processed in a consistent manner to minimize experimental error in counts. To minimize processing error, the following items should be included in the QA/QC program:

1. All personnel involved in the sample processing and analyses should have appropriate training. NABS implemented a certification program for aquatic invertebrate taxonomists. For additional information see this website.
2. The effects of sub-sampling (if done) on abundance estimates should be examined on a minimum of 10% of the samples, and the effects of sub-sampling on the sample estimates should be documented.
3. Re-sorting of randomly selected samples should be done to determine the success of the initial sorting (see detailed discussions below).
4. Appropriate taxonomic references should be used for the type of habitat and geographic location.
5. A complete reference collection for each mine should be compiled and verified by an external taxonomic expert and updated as appropriate (i.e., when new taxa are recorded).
6. A system for archiving samples should be outlined.
7. Detailed sample processing and laboratory notes should be maintained.

Ecological sample processing involves, as a first step, sorting organisms from debris and, possibly, sub-sampling sorted organisms for detailed identification. Inevitably, processing errors are associated with these activities and should be estimated (e.g., Kreis 1986, 1989).

Top of Page

4.6.4.1 Sorting Efficiency

Verification of sorting efficiency is easily performed on a spot-check basis if the leftover debris from a sample is retained. It is recommended that at least 10% of all samples be re-sorted and that the criterion for an acceptable sort be that ≤ 10% of the total number of organisms were missed. This estimate should be reported in the interpretative report. If ≥ 10% of the total number were missed during the re-sort, then all the samples within that group of samples should be re-sorted.

A re-sort would also be required if an entire group of benthic invertebrates was missed by the sorter (i.e., not recognized as an organism), even if the missed organism constituted &lt; 10% of the total. The factors to consider when determining similar groups of samples include: 1) sampling area, 2) habitat class and 3) individual sorters. The QA/QC guidelines apply independently to each group of samples sorted. Sorted and unsorted fractions are to be retained until taxonomy and sorting efficiency are confirmed.

Top of Page

4.6.4.2 Sub-sampling

Sub-sampling of invertebrate samples in the laboratory is acceptable, providing that the quantitative method is used. Large samples or samples with large amounts of sediment debris may require laboratory sub-sampling prior to sorting. Readers are referred to the Revised Guidance for Sample Sorting and Subsampling Protocols for EEM Benthic Invertebrate Community Surveys (Environment Canada 2002), which can be obtained from the EEM website. The detailed reporting of sub-sampling accuracy and precision for all methods is essential to the QA/QC of EEM benthic invertebrate programs. The criterion for an acceptable sub-sampling protocol is that the estimates of each group of samples should be within 20% of the true counts. If the error exceeds 20% for a particular sub-sampling technique or type of samples (i.e., type and amount of organic matter), the technique should be modified to achieve this level of precision, or all samples within that group should be completely sorted to ensure the sub-sampling process is not compromising data integrity. The estimates are then compared to the actual counts from the sample, and the accuracy of the estimates and the precision between sub-samples can be calculated using the following equation:

% error in the estimate = [1 – (estimated # in sample / actual # in sample)] × 100

The accuracy should be reported in the interpretative report.

It is recommended that a minimum number of 300 organisms be removed from a sample in any sub-sampling program to provide additional standardization. If any sampling stations have not reached the recommended minimum number of organisms during sub-sampling (i.e. 300) or have poor accuracy, the sample should be flagged when reported.

For further information regarding all aspects of QA/QC procedures for benthic invertebrate programs, readers are referred to the 1999 AETE report (Beak 1999).

Top of Page

4.7 Data Assessment and Interpretation

4.7.1 Data Handling Methods

4.7.1.1 QA/QC for Data Input and Verification

After data entry, the first step in data analysis is to check for transcription errors. Failure to do this invalidates further analyses. All computer entries should be verified by checking a hard copy of the file against the raw data sheets. Someone other than the person who originally entered the data should do this cross-checking. Double entry systems and transcription checks against the original data records are useful QC techniques. Missing data should be clearly distinguished from taxon absence by use of unique non-zero missing value codes with code definitions built into each file. Read-only files help to ensure data integrity. QA/QC concerns regarding data analysis include data verification and validity, repeatability and robustness of statistical analyses, and rigour and defensibility of analysis. EVS Environment Consultants (1993) suggest that other investigators should be able to arrive at the same conclusions if they were to use the methods and data set found in the report. Other considerations regarding the data verification and analyses are listed below:

1. Use trained and experienced personnel.
2. Conduct screening exercises to identify transcription errors, outliers and other suspicious data points.
3. Provide raw data in an electronic database format and appendices to reports that summarize the data.
4. Document the methods (specific statistical tests) and software (if applicable) used for analysis.
5. Maintain detailed notes regarding the analyses of the survey data.

For further information regarding all aspects of QA/QC procedures for benthic invertebrate programs, readers are referred to the 1999 AETE report (Beak 1999).

Top of Page

4.7.1.2 Dealing with Outliers

Assuming the data are entered correctly, data should be summarized, screened for erroneous values and outliers, assessed for normality, and transformed if necessary (EVS Environment Consultants 1993). Visual screening techniques such as box-and-whisker plots, normal-probability plots and stem-leaf diagrams can be used to identify extreme values (true outliers and/or data entry errors) (see Tukey 1977). Norris and Georges (1993) recommend examining abundance estimates for each taxon to determine if numbers are reasonable. They also recommend calculating means and standard deviations because aberrantly high or low values can indicate errors. Extreme values or outliers that are not errors of some kind should not be removed from the data set because this will result in the loss of an observation and a loss of power to the benthic invertebrate community survey. Instead, extreme values should be identified in the report and the influence of the extreme value on the results should be determined by reanalyzing the data minus the extreme value.

Top of Page

4.7.1.3 Unknown, Immature and Non-benthic Organisms

There have been several instances where non-benthic organisms have been submitted as part of the metal mining EEM program. If it is documented that a given family of organisms can at some point become benthically attached (e.g., Simocephalus), then it is acceptable to include the organism within the benthic invertebrate community. However, species such as planktonic Daphnia should be removed from the data set.

Some samples may contain immature individuals that cannot be identified to the recommended level of taxonomic precision. A similar situation could also occur when samples are improperly preserved and identifying features are destroyed (e.g., mollusc shells dissolve due to unbuffered formalin). For the purposes of correctly reporting the raw data, these unidentified taxa and their abundances should be provided within the electronic raw data and report appendices. However, for data analysis, investigators need to decide whether or not to apportion the unknown individuals according to the ratio of known specimens. This assumes that the ratio of unidentified specimens is similar to the ratio of identified specimens, which may or may not be true. The choices include:

1. not incorporating immature or damaged forms at all
2. pooling all specimens (i.e., mature/immature, identified/unidentified) and lumping them into one category at the next highest taxonomic level
3. keeping unidentified taxa as a separate category in the analysis

Option (1) is not preferred if the “problem” taxa represent a large proportion of the total benthic invertebrate community. Option (2) assumes that all taxa within a higher taxonomic level respond the same way to effluent-related stressors, which may or may not be true. Option (3) will have variable effects on data interpretation depending on the abundance of unidentified taxa. Whatever choice is made will depend on the expertise and experience of the individual investigator; however, it should be fully documented in the Methods section of the interpretative report.

For marine surveys, it is recommended that immature and juveniles be counted and enumerated separately from adults, whether or not they can be identified to the species level, so that the adult assemblage can be analyzed without the confounding influence of transient juveniles. Thus, data analyses should show results both with and without immatures included. This is because newly settled benthic forms have different survival characteristics than adults, which have been present in the sediment much longer and integrate the effects of habitat perturbations over time. Depending on the timing of sampling, newly settled juveniles may be abundant in samples, but may all die within days due to habitat stressors, predation or competition. This is not to say that data on immatures are not important. Dramatic variations in immature settlement between nearby samples within physically homogeneous habitats may be indicative of varying levels of stress. It is just important to avoid confounding the results by mixing groups together for analysis.

Top of Page

4.7.1.4 Data Reduction and Transformation

Data transformation is often performed without consideration of the effects it has on the interpretation of results. For general information on transformations, see Chapter 8. Transformation should only be applied with a complete understanding of its effect on the data and their interpretation, and only if it is necessary to aid in statistical analyses. Transformations should:

1. make heterogeneous variances homogeneous or make the variance independent of the mean for parametric analyses
2. normalize distributions
3. linearize relationships among variables
4. reduce the effects of extremely dominant taxa within a data set on a multivariate analysis (or ordination)
5. reduce the analytical problem of too many zeros in a data matrix (see Clarke and Green 1988).

Data reductions should be done only to aid in statistical or multivariate analyses, and for the same reasons as data transformation. Data reductions can include eliminating or rolling-up rare taxa or reducing field sub-samples by pooling or averaging. Protocols for data reductions for marine communities are varied, but subsequent interpretations of data analyses should take these reductions into account. For example, elimination of rare taxa may result in the elimination of 90% or more of the biomass within a given station if those rare taxa are large. In some cases, rare taxa are rolled up into higher groups, which prevents loss of information but adds assumptions about the uniform behaviour of mixed taxonomic groups. Reviews of standard methods of data reductions are given in Stephenson and Cook (1980), with some ecological consideration in Burd et al. (1990).

Logarithmic transformations have often been used for benthic invertebrate data because organism abundance typically varies exponentially (Green 1979). A log transformation will reduce the importance of the numerically dominant members and improve the likelihood of resolving structure when differences are due to medium-abundance or rare taxa. However, a log transformation is quite extreme. Other researchers have advocated the use of other geometric conversions such as square root, cube root, fourth-root, natural log, etc. (for reviews, see Hoyle 1973; Tukey 1977; Hoaglin et al. 1983; Downing 1981). Downing (1979) showed empirically that the best overall transformation for stabilizing variance in freshwater benthos was the fourth-root (x^0.25), because this greatly improves the performance of parametric multivariate methods such as ordinations. Vezina (1988) repeated the exercise for marine subtidal communities, concluding that they were empirically less aggregated than their freshwater counterparts and require a less extreme transformation (e.g., x^0.4). However, both researchers emphasize that the mean and variance relationships of any given community need to be analyzed to determine the most appropriate transformation. In this way it is possible to check whether or not the transformation used has stabilized the variance.

4.8 Data Reporting Guidelines

Data are submitted in the electronic database format and in hard copies (the interpretative report), as outlined and provided by Environment Canada (see Chapter 10of the present document for additional information on electronic reporting). The complete fundamental data set, including rare and highly variable taxa and ambiguous identifications, should be stored in this manner, even if data filtering has been applied prior to calculation of community descriptors. Other approaches to data filtering, calculation of community descriptors, and analysis can be employed in reanalysis or meta-analysis. A list of the relevant details for the field, laboratory and data analysis components of the EEMbenthic invertebrate survey is provided below; these details should be included and submitted with the interpretative report.

Field reporting

1. field sheets should be retained for six years
2. replicate station location (grid coordinates)
3. date and time of sampling
4. field crew members
5. habitat descriptions, including measures of the supporting environmental variables
6. sampling method used, including type and size of sampler and sieve or mesh size

Laboratory reporting

1. bench sheets should be retained for six years
2. raw data reported for each individual or pooled field sub-sample, listing taxa present and numbers of individuals
3. method and level of sub-sampling applied in the laboratory sorting process
4. sorting efficiency achieved
5. taxonomic authorities used
6. location of reference collection and report on taxonomic verification

Data analysis reporting

1. tabular listing of the number of individuals per taxon in each sample as an appendix
2. tabular summaries of calculated descriptors with variance estimates
3. estimates of power obtained for the survey
4. effects of outliers or extreme values on the results (if any)
5. a summary of adherence to data quality objectives, standard operating procedures and sampling protocols, and identification of any QA/QC problems

Top of Page

4.9 Effect Endpoints and Supporting Endpoints for the Benthic Invertebrate Community

Total invertebrate density: The total number of individuals of all taxonomic categories collected at the station expressed per unit area (e.g., numbers/m²). Values should be reported for each station, as well as the arithmetic mean ± standard error (SE), ± standard deviation (SD), median, minimum and maximum for the area.

Taxa (i.e., family) richness: The total number of different taxonomic categories collected at the station, and the arithmetic mean ±SE, ±SD, median, minimum and maximum for the area.

Evenness index (Simpson’s Evenness Index) (equitability): Evenness (E) can be quantified for each station, and mean E ±SE, ±SD, median, minimum and maximum for the area should be reported. Evenness is calculated as in Smith and Wilson (1996):

where:
E = evenness
p_i = the proportion of the i^th taxon at the station
S = the total number of taxa at the station

Similarity index (Bray-Curtis [B-C] Index):The B-C Index is a distance co-efficient that reaches a maximum value of 1 for two sites that are entirely different and a minimum value of 0 for two sites that possess identical descriptors. Distance coefficients measure the amount of association between sites, and the B-C Index is a member of the class of distance coefficients known as a semimetric that some prefer to call dissimilarity coefficients. The B-C Index measures the percentage of difference between sites (Legendre and Legendre 1983), where the distance statistic is calculated as below:

where:
B-C = Bray-Curtis distance between sites 1 and 2
Y_i1 = count for taxon i at site 1
Y_i2 = count for taxon i at site 2
n = total number of taxa present at the two sites

The Bray-Curtis distance (B-C) from a calculated reference median will be reported for each station, and the arithmetic mean ±SE, ±SD, minimum and maximum B-C distance is reported for the area. As the use of this index for determination of effects may be novel to some, a brief literature summary and a detailed example is provided below.

Most of the invertebrate community statistics discussed above are measures of total density and taxa richness and provide no quantitative information on what kind of organisms are present. A similarity index is also recommended, as it summarizes the overall difference in community structure between reference and exposed sites in a single number, requires no preconceived assumptions about the nature of the community and only varies in one direction (Taylor and Bailey 1997). Of the various indices available, many reviewers have indicated that the Bray-Curtis Index (Bray and Curtis 1957) is the most reliable (Pontasch et al. 1989; Jackson 1993; Bloom 1981). The Bray-Curtis Index is also unaffected by the nature of the communities being compared (Bloom 1981), and differences contribute the same to the Bray-Curtis (B-C) Index regardless of whether the taxon is rare or abundant. Bloom (1981) showed that, of 4 indices examined, only the B-C Index accurately reflected the true resemblance over its range.

Top of Page

Example of Bray-Curtis Index for use in the EEMprogram

The following steps use an example data set to illustrate how the Bray-Curtis Index should be used for the evaluation of effects in the EEMprogram. In this example, 5 stations were sampled from an exposure area and a reference area, with a total of 5 taxa found to be present.

1. Taxa density is entered into a table.
   
2. For the reference stations, the median taxa density is determined (see example below).
   

   Table showing median taxa densities for reference stations

   	Taxa Density
   Reference Stations	Taxon 1	Taxon 2	Taxon 3	Taxon 4	Taxon 5
   Ref 1	2	3	2	3	1
   Ref 2	3	5	2	4	3
   Ref 3	9	1	1	1	1
   Ref 4	4	6	3	4	1
   Ref 5	5	4	2	3	2
   Reference Median	4	4	2	3	1

   
   
3. A similar table is constructed for the exposure stations without the median calculation.
   

   Table showing median taxa densities for exposure stations

   	Taxa Density
   Exposure Stations	Taxon 1	Taxon 2	Taxon 3	Taxon 4	Taxon 5
   Exp 1	23	4	2	10	1
   Exp 2	12	2	2	8	3
   Exp 3	14	6	1	6	2
   Exp 4	13	1	3	12	2
   Exp 5	15	3	2	4	1

   
   
4. The distance of each station (reference and exposure) from the reference median is calculated as illustrated by the following example for reference station 1.
   
   For this approach, the reference median for particular taxa becomes y_i2, the taxon count for site 2 in the above equation.
   

   Table showing the distance of each station (reference and exposure) from the reference median

   	Taxa 1	Taxa 2	Taxa 3	Taxa 4	Taxa 5
   Ref 1 (yi1)	2	3	2	3	1
   Reference median (yi2)	4	4	2	3	1
   | yi1-yi2 | or Ref 1- reference median	2	1	0	0	0
   (yi1+yi2)	6	7	4	6	2

   
   Substituting into the B-C equation gives:
   
   
   
5. The B-C distance from the reference median is calculated for each station in this manner.
   
6. The result of this calculation should be reported for each station, along with the mean (±SE) for the area. The sample data set would result in the following B-C distances:
   

   Table showing B-C distances

   Station	| yi1-yi2|	(yi1+yi2)	B-C distance from median	Mean ± SE
   Ref 1	3	25	0.12	0.18 ± 0.06
   Ref 2	5	31	0.16
   Ref 3	11	27	0.41
   Ref 4	4	32	0.13
   Ref 5	2	30	0.07
   Exp 1	26	54	0.48	0.43 ± 0.03
   Exp 2	17	41	0.41
   Exp 3	17	43	0.40
   Exp 4	23	45	0.51
   Exp 5	13	39	0.33

   
   
7. Finally, for the purposes of determining an effect at the exposure area, the mean B-C distance between the reference stations and the reference median (0.18 ±0.06) can be compared statistically to the mean distance between the exposure stations and the reference median (0.43 ± 0.03).

Simpson’s Diversity Index: Simpson’s Diversity Index (D) takes into account both the abundance patterns and taxonomic richness of the community. This is calculated by determining, for each taxonomic group at a station, the proportion of individuals that it contributes to the total in the station. D for each station and mean (±SE, ±SD), median, minimum and maximum D for the area should be reported. Simpson’s Diversity Index is calculated as (Krebs 1985):

where:
D = Simpson’s index of diversity
S = the total number of taxa at the station
p_i = the proportion of the ith taxon at the station

Taxa (i.e., family) density: The number of individuals of each family expressed per unit area (e.g., numbers/m²). Values should be reported for each taxon at each station and as the mean (±SE) of each taxon for the area.

Taxa (i.e., family) proportion: The percentage abundance for each taxon at each station and the mean (±SE) percentage abundance of each taxon for the area.

Taxa (i.e., family) presence/absence: A matrix indicating the presence and absence of each taxon at the sampling stations should be reported. The matrix will consist of stations (columns) and taxa (rows).

In addition to the benthic invertebrate endpoints, the sediment monitoring variables are also to be reported (see Chapter 7).

Top of Page

4.10 Evaluation of Results

4.10.1 Effect on the Benthic Invertebrate Community

The objective of the benthic invertebrate component of the EEM program is to answer the following question:

“Is there an effect on the benthic invertebrate community?”

The definition of effect is described in Schedule 5, section 1 of the Metal Mining Effluent Regulations.

During the first phases and in magnitude and geographic extent phases of the monitoring program, the following effect endpoints are calculated, reported and used to determine if there is an effect on the benthic invertebrate community:

1. Total benthic invertebrate density
2. Taxa (i.e., family) richness
3. Evenness index (Simpson’s)
4. Similarity index (Bray-Curtis)

For the benthic invertebrate component, it is recommended that the following supporting endpoints also be calculated and reported:

• Simpson’s Diversity Index
• Taxa (i.e., family) density
• Taxa (i.e., family) proportion
• Taxa (i.e., family) presence/absence

All these endpoints, described into details in the previous section are largely summary metrics selected to encompass the range of effects that may be a result of mine effluent.

Many other benthic invertebrate descriptive metrics are available in the literature and serve to address a wide range of questions regarding benthic invertebrate communities. If desired, additional site-specific descriptors may be calculated and used to support the interpretation of effects. For guidance on selecting these optional descriptive metrics and discussion of their applicability, readers are referred to the reviews by Resh et al. (1995).

For the statistical analyses and determination of sufficient power, the recommendations developed for setting of effect size, a and b and presented in section 8.6.1 are also applicable. The recommendation in this previous section was to set a and b equally at 0.10 or less. The appropriate method of analysis for each of the study design options (e.g., ANOVA, ANCOVA, regression, multivariate analysis) is indicated in Table 4-1.

A final caveat regarding effects on the benthic invertebrate community: it is essential for the mine to select a site-specific study design to allow for an appropriate evaluation. Critical to the study design is the selection of an appropriate reference area or areas. The importance of proper reference area selection is underscored by the following, potentially frequent, example. If a mine performs a simple control-impact design with the reference area placed upstream, then differences between upstream and downstream communities will be those determining the presence or absence of effects. However, if the downstream benthic communities are modified due to a factor such as the restoration of an upstream flow disruption (e.g., from a dam), then these communities, although different from upstream communities, may be more similar to (but perhaps not exactly the same as) the communities at a reference area chosen in a drainage basin adjacent to (or even further afield than) the mine drainage basin. In this example, selection of an additional reference area (see Figure 4-2d for an example) may well be worth the extra cost involved so that site-specific interpretation and the appropriate assessment of effects can be accomplished. Note that this example of significant upstream-downstream differences may not necessarily be considered an effect if sufficient additional evidence suggests otherwise.

Top of Page

4.10.2 Next Step

Once the monitoring data have been analyzed, decisions regarding the next step in the EEM program are made. The next step in the monitoring program is dependent on the relationship between several key factors, which are briefly discussed below.

The statistical outcome of the previous benthic invertebrate survey

There are 3 possible statistical outcomes of the benthic invertebrate community survey:

1. no effect is detected but power is not sufficient (i.e., power &lt; 0.90)
2. no effect is detected and power is sufficient (i.e, power ≥ 0.90)
3. an effect is detected

If any of the effect endpoints (total benthic invertebrate density, taxa richness, evenness index (Simpson’s) and similarity index (Bray-Curtis)) demonstrate a statistical difference between exposure and reference areas (or along a gradient), then the conclusion is that there is an effect on the benthic invertebrate community. This result can be obtained by various statistical methods; the choice of methods depends on the study design of the monitoring program.

If the power was insufficient, the mine may reconsider the number of sampling stations or the sampling design that was used, in order to design a study with sufficient power in the next survey.

EEM program options after an effect has been established

If an effect on the benthic invertebrate community is found, the next question to be addressed is:

Is the effect mine-related?

An assessment of whether the effect is mine-related could include asking the following questions:

• Is the cause of the effect known or suspected?
• Can the effect be related to a natural change in the aquatic receiving environment?
• Can the effect be reasonably correlated to an anthropogenic cause other than the mine effluent?
• Is there a weight-of-evidence approach that can indicate a causal link? (See section 4.11)

This series of questions is provided as an example of the type of approach that may allow for the determination of whether or not the observed effect is mine-related. If the presence of confounding factors makes it difficult to determine the effect of mine effluent on the benthic invertebrate community, the mine should reconsider the study design for the next phase. If the effect has been confirmed, and the cause of the effect is unknown, the mine proceeds to the next step of data assessment and interpretation: determining the magnitude and geographic extent of the effect.

Are the magnitude and geographic extent known?

If an effect has been confirmed (see Chapter 1for details on confirmed effects), and the cause of the effect is unknown, then the mine should proceed to the next step and determine the magnitude and geographic extent of the effect. For additional information refer to section 4.2.2.

4.11 Additional Tools for Focused Monitoring, Weight-of-Evidence Approaches and/or Investigation of Cause

There are a number of alternative approaches and tools possible for investigations of cause in the EEM program. Methods provided in this guidance document are not meant to be exhaustive, and mines may propose additional scientifically defensible approaches. Tools should be cost-effective, recognized in the primary literature, readily available from consulting, academic or government laboratories, and applicable to the EEM program.

Additional information can be found in Chapters 9 and 12of this document.

Top of Page

4.11.1 Use of Weight-of-Evidence Approaches to Establish Cause of Effects

Distinguishing among the cumulative impacts of multiple stressors (which sometimes have confounding effects) requires the establishment of a definitive causal link to the mine effluent under evaluation. The environmental assessment of an aquatic ecosystem is particularly prone to impediments because such ecosystems often receive multiple, interactive effluent discharges. Assessments of monitoring results often rely, in large part, on field monitoring data that can only show correlations rather than clear cause and effect between mine effluent and a presumed effect. Establishing a strong causal link, however, can benefit from a weight-of-evidence approach that combines information from a variety of sources. For additional information on the use of weight-of-evidence approaches, readers are referred to chapters 9 and 12of the present document.

Top of Page

4.11.2 Lethal and Sublethal Toxicity Tests

Lethal and sublethal toxicity test methods can be applied during magnitude and geographic extent and investigation-of-cause surveys when an effect has been identified or when previous work failed to provide a satisfactory explanation of cause. These methods provide a direct determination of lethal or sublethal toxicity and can verify that alterations in benthos are due to the toxicity of the mine effluent rather than confounding factors. For example, adverse effects on benthic community structure may be due to factors other than effluent toxicity, including differences in environmental regime. Concurrent impairment of benthic community structure and toxicity implicates the effluent itself as the cause of changes in the benthos. These methods also provide important information for interpreting field effects in situations where benthic community data are inconclusive, or if only pollution-tolerant species are present in both impacted and reference sites.

Top of Page

4.11.3 Analysis of Sediment Cores for Historic Trends

Sedimentary records from depositional areas of water bodies can be used to indicate limnological conditions in recent and ancient history (Frey 1988). Precise dating of sediments, combined with an inventory of the remains of certain organisms and plant material (e.g., diatoms, zooplankton, insects), provides a chronology of changes that often can be linked to the period of anthropogenic influence. In addition to the water body itself, the history of the watershed and airshed may be deduced, and the influences of natural events may be distinguished from anthropogenic impacts. A substantial volume of literature is available on the subject, with a useful synthesis of the science provided by Frey (1988). Due to the level of expertise needed to undertake this type of analysis, the availability of paleolimnological services is limited. In addition, the analyses are restricted to resolving trends over longer time frames (multiple years to decades) as a result of sedimentation processes such as bioturbation. The costs of the technique will be site-specific.

Top of Page

4.11.4 Other Benthic Invertebrate Measures and Organisms

Benthic invertebrates are recommended as the primary indicator organisms for use in an EEM program for monitoring effects on fish habitat. However, the level of identification and measures recommended in the main text of the guidance document are not an exclusive list of measures for which benthic invertebrates can be evaluated. Additional measures include biomass, lower level of identification, secondary production, and population fitness parameters.

Benthic invertebrate biomass in marine environments can provide additionally useful information because it is related to the availability of energy to other trophic levels (e.g., fish). For marine communities, some investigators suggest that an analysis of benthic abundance and biomass together provide a sensitive indicator of changes in the composition of the benthic community (e.g., Warwick 1986; Warwick et al. 1987; Clarke 1990; Burd et al. 1990). For example, in marine samples, it is in the measurement of distributions of biomass that the three main functional groups of benthic organisms--microfauna (grain surface dwellers), meiofauna (interstitial organisms) and macrofauna (burrowers and epifauna)--can be distinctly separated (Schwinghamer 1981, 1983). Because these three groups of organisms have different reproductive modes, metabolic rates, life histories and habitat adaptations, they respond differently to habitat perturbation. This could be particularly important in Arctic subtidal habitats, where abundance may be low, but individuals may be large. However, because precise biomass measurements are time-consuming and problematic (cf. Crisp 1984) unless collected in more detail and more often than is feasible for EEM requirements, it is only possible to determine relative changes in biomass of samples for the EEMsurveys. This is easily done by taking blotted wet-weight measurements of representative-sized adult specimens of each species for each survey. Since the method is non-destructive, the reference collection may be used for this purpose prior to external verification or archiving. The mean weight of a given species can then be used to transform species abundance data to relative species biomass data for further summary or statistical analyses. These data show relative, large-scale changes only, and cannot be used to infer production or trophic flow rates within benthic communities.

In addition to benthic invertebrates, several other types of aquatic biota were considered for use in the EEM program. The most relevant ones were 1) phytoplankton, 2) macrophytes, and 3) periphyton.

Top of Page

4.12 References

Abbott RT. 1974. American seashells. 2nd ed. New York (NY): Van Nostrand Reinhold Co.

Abbott RT, Zim HS, Sandström GF. 2001. Seashells of North America: a guide to field identification. Golden Field Guides from St. Martin’s Press. New York (NY): St. Martin’s Press.

[AETE] Aquatic Effects Technology Evaluation Program. 1995. Field evaluation of aquatic effects monitoring methods - pilot study. Volume 1. AETE Project 4.1.1. Ottawa (ON): Canada Centre for Mineral and Energy Technology.

Alberta Environment. 1990. Selected methods for the monitoring of benthic invertebrates in Alberta rivers. Environmental Quality Monitoring Branch, Alberta Environment.

Appy TD, Linkletter IF, Dadswell MJ. 1980. A guide to the marine flora and fauna of the Bay of Fundy: Annelida: Polychaeta. St. Andrews (NB): Fisheries and Marine Service. Technical Report No. 920.

Austin WC. 1985. An annotated checklist of marine invertebrates in the cold temperate northeast Pacific. Vol. 1, 2 and 3. ,Cowichan Bay (BC): Khoyatan Marine Laboratory. Report for Fisheries and Oceans Canada.

Top of Page

Bailey RC, Norris RH, Reynoldson TB. 2003. Bioassessment of freshwater ecosystems. Boston (MA): Kluwer.

Baker HR. 1980. Key to the common Tubificid species of the northeast Pacific. Manuscript from Oligochaeta workshop.

Banse K. 1972. Redescription of some species of ChoneKröyer and Euchone Malmgren, and three new species (Polychaeta: Sabellidae). Fish Bull 70(2):459-495.

Banse K, Hobson KD. 1974. Benthic errantiate polychaetes of British Columbia and Washington. (Bulletin of Fisheries Research Board of Canada 185). Ottawa (ON): Fisheries and Marine Service.

Barber WE, Kevern NR. 1974. Seasonal variation of sieving efficiency in a lotic habitat. Freshwat Biol 4:293-300.

Beak International Inc. 1999. Quality assurance program for assessing mine-related effects using benthic invertebrate communities. Ottawa (ON): Natural Resources Canada, Canada Centre for Mineral and Energy Technology. Aquatic Effects Technology Evaluation Program project 2.1.4.

Bednarik AF, McCafferty WP. 1979. Biosystematic revision of the genus Stenonema (Ephemeroptera, Heptageniidae). Can Fish Aquat Sci Bull 201. Ottawa: Department of Fisheries and Oceans.

Berkeley C, Berkeley E. 1952a. Canadian Pacific fauna, 9. Annelida 9b (1) Polychaeta Errantia. (Bulletin of Fisheries Research Board of Canada).Toronto (ON): University of Toronto Press.

Berkeley C, Berkeley E. 1952b. Canadian Pacific fauna 9. Annelida 9b (2) Polychaeta Sedentaria. (Bulletin of Fisheries Research Board of Canada).Toronto (ON): University of Toronto Press.

Blake JA. 1971. Revision of the genus Polydora from the east coast of North America (Polychaeta: Spionidae). Smithson Contr Zool 75:1-32.

Blake JA. 1988. New species and records of Phyllodocidae (Polychaeta) from Georges Bank and other areas of the western North Atlantic. Sarsia 73:245-257.

Blake JA. 1991. Revision of some genera and species of Cirratulidae (Polychaeta) from the western North Atlantic. In Petersen ME, Kirkegaard JB, editors. Systematics, biology and morphology of world Polychaeta. Proceedings of the Second International Polychaete Conference, Copenhagen, August 18-23, 1986. Ophelia Supplement 5: 1-723. p. 17-30.

Bloom SA. 1981. Similarity indices in community studies: potential pitfalls. Mar Ecol Prog Ser 5:125-128.

Booth J, Hay D, Truscott, J. 1996. Standard methods for sampling resources and habitats in coastal subtidal regions of British Columbia: Part I - review of mapping with preliminary recommendations. (Canadian Technical Report of Fish and Aquatic Sciences 2118). Nanaimo (BC): Fisheries and Oceans Canada.

Bousfield EL. 1958. Freshwater amphipod crustaceans of glaciated North America. Can Field Nat 72(2):April-June.

Bousfield EL. 1960. Canadian Atlantic sea shells. Ottawa (ON): Department of Northern Affairs and Natural Resources, National Museum of Canada.

Bousfield EL. 1973. Shallow-water gammaridean Amphipoda of New England. Ithaca (NY): Cornell University Press.

Bousfield EL, Hendryks EA, 1994. The amphipod superfamily Leucothoidea on the Pacific coast of North America. Family Pleustidae: subfamily Pleustinae, systematics and biogeography. Amphipacifica I:2.

Bousfield EL, Hendryks EA. 1995a. The amphipod family Pleustidae on the Pacific coast of North America. Part III. Subfamilies Parapleustinae, Dactylopleustinae, and Pleusirinae: systematics and distributional ecology. Amphipacifica II:1.

Bousfield EL, Hendryks EA.1995b. The amphipod superfamily Eusiroidea in the North Pacific region. I. Eusiridae: systematics and distributional ecology. Amphipacifica I:4.

Bousfield EL, Hoover PM. 1995. The amphipod superfamily Pontoporeioidea on the Pacific coast of North America. II. Family Haustoriidae. Genus Eohaustorias J.L. Barnard: systematics and distributional ecology. Amphipacifica II:1.

Bousfield EL, Kendall JA. 1994. The amphipod superfamily Dexaminoidea of the North American Pacific coast, families Atylidae and Dexaminidae: systematics and distributional ecology. Amphipacifica I:3.

Bowman MF, Bailey RC. 1997. Does taxonomic resolution affect the multivariate description of the structure of freshwater benthic macroinvertebrate communities? Can J Fish Aquat Sci 54:1802-1807.

Brandlova J, Brandl Z, Fernando, CH. 1972. The Cladocera of Ontario with remarks on some species and distribution. Can J Zool 50:1373-1403.

Bray JR, Curtis JT. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecol Monogr 27:325-349.

Brinkhurst RO. 1982. British and other marine and estuarine oligochaetes. (Synopses of the British fauna 21). Cambridge (UK): Cambridge University Press.

Brinkhurst RO. 1986. Guide to the freshwater aquatic microdrile oligochaetes of North America. (Canadian Special Publication of Fisheries and Aquatic Sciences 84). Ottawa (ON): Fisheries and Ocean Canada, Scientific Information and Publications Branch.

Brinkhurst RO, Baker HR. 1979. A review of the marine Tubificidae (Oligochaeta) of North America. Can J Zool 57:1553-1569.

Brunel P, Bosse L, Lamarche G. 1998. Catalogue des Invertébrés marins de l’estuaire et du golfe du Saint-Laurent /Catalogue of the marine invertebrates of the estuary and Gulf of Saint Lawrence. Ottawa (ON): National Research Council of Canada Press.

Burch JB. 1975a. Freshwater Sphaeriacean clams (Mollusca: Pelecypoda) of North America. Hamburg (MI): Malacological Publications.

Burch JB. 1975b. Freshwater Unionacean clams (Mollusca: Pelecypoda) of North America. Hamburg (MI): Malacological Publications.

Burch JB. 1989. North American freshwater snails. Hamburg (MI): Malacological Publications.

Burd BJ, Nemec A, Brinkhurst RO. 1990. The development and application of analytical methods in benthic marine infaunal studies. Adv Mar Biol 26:169-247.

Butler TH.1983. Shrimps of the Pacific coast of Canada. (Bulletin 202). Ottawa (ON): Department of Fisheries and Oceans.

Top of Page

Camburn KE, Kingston JC, Charles DF, editors. 1984-1986. PIRLA diatom iconograph. Contains 53 photographic plates and 1059 figures, plus Figure legends. (PIRLA Unpublished Report Series No. 3). Bloomington (IN): Indiana University.

Chu HF, Cutkomp LK. 1992. How to know the immature insects. 2nd edition. Dubuque (IA): Wm. C. Brown Publishers.

Clark HL. 1915. Catalogue of recent ophiurans. Mem Mus Comp Zool Harv 25:165-376.

Clark HL. 1924. Some holothurians from British Columbia. Can Field Nat 38:54-57.

Clarke AH. 1981. The freshwater molluscs of Canada. Ottawa (ON): National Museum of Natural Sciences.

Clarke KR. 1990. Comparisons of dominance curves. J Exp Mar Biol Ecol 138:143-157.

Clarke RT, Furse MT, Wright JF. 1992. A comparison of single, paired and three season combined macro-invertebrate samples for the biological banding of river quality. Bristol (UK): National Rivers Authority.

Clarke KR, Green RH. 1988. Statistical design and analysis for a “biological effects” study. Mar Ecol Prog Ser 46:213-226.

Clifford HF. 1991. Aquatic invertebrates of Alberta. Edmonton (AB): University of Alberta Press.

Coates KA.1980. Keys to intertidal genera and species of Enchytraeidae found in British Columbia. Manuscript from Oligochaeta Workshop. Victoria (BC) University of Victoria.

Coe WR. 1912. Echinoderms of Connecticut. Connecticut State Geological and Natural History Survey19:1-147.

Coe WR. 1943. Biology of the nemerteans of the west coast of North America. Trans Conn Acad Arts Sci 35:129-328, pls. 1-4.

Cowardin LM, Carter V, Golet FC, LaRoe ET. 1979. Classification of wetlands and deepwater habitats of the United States. Washington (DC): U.S. Fish and Wildlife Service. FWS/OBS-79/31.

Cowardin LM, Golet FC. 1995. U.S. Fish and Wildlife Service 1979 wetland classification: a review. Vegetatio 118:139-152.

Crisp DJ. 1984. Energy flow measurements.In Holme NA, McIntyre AD, editors. Methods for the study of marine benthos. IBP Handbook 16. Oxford (UK): Blackwell Scientific Publications. p. 284-370.

Crocker DW, Barr DW. 1968. Handbook of the crayfishes of Ontario. Toronto (ON): University of Toronto Press.

Cuffney TF, Gurtz ME, Meador MR. 1993. Guidelines for the processing and quality assurance of benthic invertebrate samples collected as part of the national water-quality assessment program. U.S. Geological Survey Open-File Report 93-407.

Culp JM, Lowell RB, Cash KJ. 2000. Integrating mesocosm experiments with field and laboratory studies to generate weight-of-evidence risk assessments for large rivers. Environ Toxicol Chem 19:1167-1173.

Cutler EB. 1973. Sipuncula of the western North Atlantic. Bull Am Mus Nat Hist 152(3):1-204.

Top of Page

Downing JA.1979. Aggregation, transformation and the design of benthos sampling programs. J Fish Res Bd Can 36:1454-1463.

Downing JA. 1981. How well does the fourth-root transformation work? Can J Fish Aquat Sci 38:127-129.

Downing JA. 1984. Sampling the benthos of standing waters. In Downing JA, Rigler FH, editors. A manual on methods for the assessment of secondary productivity in fresh waters. 2nd. edition. (IBP Handbook 17). Oxford (UK): Blackwell Scientific Publications. p. 87-130.

Downing JA. 1986. Spatial heterogeneity: evolved behaviour or mathematical artefact? Nature 323: 255-257.

Dussart B. 1969. Les copépodes des eaux continentales. Paris (FR): N. Boubee & Cie.

Top of Page

Edmunds GF Jr, Jensen SL, Berner L. 1976. The mayflies of North and Central America. Minneapolis (MN): University of Minnesota Press.

Eleftheriou A, Holme NA. 1984. Macrofauna techniques. In Holme NA, McIntyre AD, editors. Methods for the study of marine benthos. (IBP Handbook 16). Oxford (UK): Blackwell Scientific Publications. p. 140-216.

Elliott JM. 1977. Some methods for the statistical analysis of samples of benthic invertebrates. (Freshwater Biological Association Scientific Publication No. 25). Ambleside (UK): Freshwater Biological Association.

Elliott JM, Tullett PA. 1978. A bibliography of samplers for benthic invertebrates. (Freshwater Biological Association, Occasional Publication No. 4). Ambleside (UK): Freshwater Biological Association.

Elliott JM, Tullett PA. 1983. A supplement to a bibliography of samplers for benthic macroinvertebrates. (Freshwater Biological Association Occasional Publication No. 20.) Ambleside (UK) Freshwater Biological Association.

Environment Canada. 2002. Revised guidance for sample sorting and sub-sampling protocols for EEM benthic invertebrate community surveys. Ottawa (ON): Environment Canada, National EEMOffice.

Top of Page

Fauchald K. 1977. The Polychaeta worms: definitions and keys to orders, family and genera. (Science Series 28). Los Angeles (CA): Natural History Museum of Los Angeles County.

Finlayson CM, van der Valk AG. 1995. Wetland classification and inventory: a summary. Vegetatio 118(1-2):185-192.

Fitzpatrick JF. Jr. 1983. How to know the freshwater crustacea. Dubuque (IA): Wm.C. Brown Publishers.

Fournier JA, Petersen ME. 1991. Cossura longocirrata:redescription and distribution, with notes on reproductive biology and a comparison of described species of Cossura (Polychaeta: Cossuridae). Ophelia Suppl 5:63-80.

Frey DG. 1988. What is paleolimnology? J Paleolim 1:5-8.

Frith HR, Seraring G, Wainwright P, Harper H, Emmett B. 1993. Review of habitat classification systems and an assessment of their suitability to coastal B.C. Unpublished report. Sidney (BC): L.G.L. Ltd for Environment Canada, Environmental Emergency Services Branch.

Fullington KE, Stewart KW. 1980. Nymphs of the stonefly genus Taeniopteryx (Plecoptera: Taeniopterygidae) of North America. J Kansas Entomol Soc 53:237-259.

EVS Environment Consultants. 1993. Guidelines for Monitoring Benthos in Freshwater Environments. Prepared for: Environment Canada, 224 West Esplanade, North Vancouver, B.C.

Top of Page

Gibson R. 1994. Nemerteans. 2nd edition. (Synopses of the British Fauna No. 24). Shrewsbury (UK): Field Studies Council.

Glozier N. 1989. The effects of biotic and abiotic factors on the foraging success of a lotic minnow, Rhinichthys cataractae [master’s thesis]. Calgary (AB): University of Calgary.

Gosner KI. 1971. Guide to the identification of marine and estuarine invertebrates. New York (NY): J. Wiley and Sons.

Graham A. 1988. Molluscs: prosobranchs and pyramellid gastropods(2nd ed.). Kermack DM, Barnes RSK, editors. Synopses of the British Fauna (New Series) No. 2. London (UK): The Linnean Society of London.

Gray JS, Clarke KR, Warwick RM, Hobbs G. 1990. Detection of initial effects of marine pollution on marine benthos: an example from the Ekofisk and Eldfisk oilfields, North Sea. Mar Ecol Prog Ser 66:285-299.

Green RH. 1979. Sampling design and statistical methods for environmental biologists. Toronto (ON): John Wiley and Sons.

Green G. 1994. Protocols for reference and voucher collections of aquatic invertebrates stored at the Royal British Columbia Museum. DOE-FRAP 1994-15.

Top of Page

Harper PP, Stewart KW. 1984. 13: Plecoptera. In Merritt RW, Cummins, KW, editors. An introduction to the aquatic insects of North America. 2nd ed. Dubuque (IA): Kendall/Hunt Publ.Co.

Hart JFL. 1982. Crabs and their relatives of British Columbia. Handbook No. 40. Victoria (BC): British Columbia Provincial Museum.

Hauer FR, Lamberti GA, editors. 1996. Methods in stream ecology. San Diego (CA): Academic Press Inc.

Hay DE, Waters RD, and Boxwell TA, editors. 1996. Proceedings, Marine Ecosystem Monitoring Network Workshop, Nanaimo, B.C. March 28-30. 1995. (Canadian Technical Report of Fisheries and Aquatic Sciences 2108). Nanaimo (BC): Fisheries and Oceans Canada.

Hilsenhoff WL. 1995. Aquatic insects of Wisconsin: keys to Wisconsin genera and notes on biology, habitat, distribution and species. Publication Number 3. Madison (WI): Natural History Museums Council, University of Wisconsin.

Hilsenhoff WL, Schmude KL. 1992. Riffle beetles of Wisconsin (Coleoptera: Dryopidae, Elmidae, Lutrochidae, Psephenidae) with notes on distribution, habitat, and identification. Great Lakes Entomol 25(3):191-213.

Hitchcock SW. 1974. Guide to the insects of Connecticut. Part VII. The Plecoptera or Stoneflies of Connecticut. Bulletin of the State Geological and Natural History Survey of Connecticut 107.

Hoaglin DC, Mosteller F, Tukey JW. 1983. Understanding robust and exploratory data analysis. New York (NY): John Wiley.

Hobson KD, Banse K. 1981. Sedentariate and Archiannelid Polychaetes of British Columbia and Washington. Bulletin 209. Ottawa (ON): Department of Fisheries and Oceans.

Holme NA, McIntyre AD, editors. 1984. Methods for the study of marine benthos. (IBP Handbook 16). Oxford (UK): Blackwell Scientific Publications.

Hoyle MH. 1973. Transformations: an introduction and a bibliography. Inter Stat Rev 41:203-223.

Hurlbert SH. 1984. Pseudoreplication and the design of ecological field experiments. Ecol Monog 54:187-211.

Hyman LH. 1940. The polyclad flatworms of the Atlantic coast of the United States and Canada. Proc US Nat Mus 89:449-495.

Hyman LH. 1944. Marine Turbellaria from the Atlantic coast of North America. Am Mus Novitates No. 1266.

Top of Page

Jackson DA. 1993. Multivariate analysis of benthic invertebrate communities: the implication of choosing particular data standardizations, measures of association, and ordination measures. Hydrobiol 268:9-26.

Johannsen OA. 1977. Aquatic Diptera. Los Angeles (CA): Entomological Reprint Specialists. Fourth reprinting.

Johnson RK, Wiederholm T, Rosenberg DM. 1993. Freshwater biomonitoring using individual organisms, populations, and species assemblages of benthic macroinvertebrates. In Rosenberg DM, Resh VH, editors. Freshwater biomonitoring and benthic macroinvertebrates. New York (NY): Chapman and Hall. p. 40-125.

Jonasson PM. 1955. The efficiency of sieving techniques for sampling freshwater bottom fauna. Oikos 6:183-207.

Top of Page

Keen AM, Coan E. 1974. Marine molluscan genera of western North America. Palo Alto (CA): Stanford University Press.

Klemm DJ. 1972. Freshwater leeches (Annelida: Hirudinea) of North America. U.S. Environmental Protection Agency Identification Manual No 8. Washington (DC): U.S. Environmental Protection Agency.

Klemm DJ. 1985. A guide to the freshwater Annelida (Polychaeta, Naidid and Tubificid Oligochaeta, and Hirudinea) of North America. Dubuqe (IA): Kendall/Hunt Publ. Co.

Klemm DJ. 1991. Taxonomy and pollution ecology of the Great Lakes region leeches (Annelida: Hirudinea). Michigan Academician 24:37-103.

Klemm DJ, Lewis PA, Fulk F, Lazorchak JM. 1990. Macroinvertebrate field and laboratory methods for evaluating the biological integrity of surface waters. Cincinnatti (OH): U.S. Environmental Protection Agency, Environmental Monitoring Laboratory. EPA 600/4-90/030.

Knight-Jones P. 1978. New Spirorbidae (Polychaeta: Sedentaria) from the East Pacific, Atlantic, Indian and southern oceans. Zool J Linnean Soc. 64:201-240.

Knight-Jones P. 1983. Contributions to the taxonomy of Sabellidae (Polychaeta). Zool J Linnean Soc 79:245-295.

Kozloff EN. 1987. Marine invertebrates of the Pacific Northwest. Seattle (WA): University of Washington Press.

Krebs CJ. 1985. Ecology: the experimental analysis of distribution and abundance. 3rd edition. New York (NY): Harper and Row.

Kreis RG Jr. 1986. Variability study. Section 17. In Charles DF, Whitehead DR, editors. Paleoecological investigation of recent lake acidification (PIRLA): methods and project description. Palo Alto (CA): Electric Power Research Institute.

Kreis RG Jr. 1989. Variability study: interim results. Section 4. In Charles DF, Whitehead DR, editors. Paleoecological investigation of recent lake acidification (PIRLA):1983-1985: interim report. October 1989. Palo Alta (CA): Electric Power Research Institute.

Kronberg I. 1987. Accuracy of species and abundance minimal areas determined by similarity area curves. Mar Biol 96:555-561.

Top of Page

Lambert P. 1981. The sea stars of British Columbia. (B.C. Provincial Museum Handbook No. 39). Victoria (BC): British Columbia Provincial Museum.

Laplante S, Bousquet Y, Bélanger P, Chantal C. 1991. Liste des espèces de coléoptères du Québec. (Fabreries, Supplément 6). Sillery (QC): Association des entomologistes amateurs du Québec.

Laubitz DR. 1972. The Caprellidae (Crustacea, Amphipoda) of Atlantic and Arctic Canada. (Publications in Biological Oceanography No. 4). Ottawa (ON): National Museums of Canada, National Museum of Natural Sciences.

Legendre L, Legendre P. 1983. Numerical ecology. Amsterdam (NL): Elsevier.

Lehmkuhl DM. 1975a. Field guide to aquatic insect families. Blue Jay 33:199-219.

Lehmkuhl DM. 1975b. Saskatchewan damselflies and dragonflies. Blue Jay 33:18-27.

Lehmkuhl DM. 1976. Mayflies. Blue Jay 34:70-81.

Lehmkuhl DM. 1979. How to know the aquatic insects. Dubuque (IA): Wm. C. Brown Company Publishers.

Leopold LB, Wolman GM, Miller JP. 1964. Fluvial processes in geomorphology. San Francisco (CA): W.H. Freeman and Co.

Leopold LB. 1994. A view of the river. Cambridge (MA): Harvard University Press.

Lewis PA. 1974. Taxonomy and ecology of Stenonemamayflies (Heptageniidae: Ephemeroptera). Cincinnati (OH): U.S. Environmental Protection Agency. EPA-670/4-74-006.

Levings CD, Thom RM. 1994. Habitat changes in the Georgia Basin: implications for resource management and restoration. In Wilson RCH, Beamish RJ, Aitkens F, Bell J, editors. Review of the marine environment and biota of Strait of Georgia, Puget Sound and Juan de Fuca Strait: Proceedings of the BC/Washington Symposium on the Marine Environment, Jan. 13-14, 1994. Canadian Technical Report of Fisheries and Aquatic Sciences 1948). p. 330-351.

Light WJ. 1977. Spionidae (Annelida: Polychaeta) from San Francisco Bay, California: a revised list with nomenclatural changes, new records, and comments on related species from the northeastern Pacific Ocean. Proc Biol Soc Wash 90(1):66-88.

Lowell RB. 1997. Discussion paper on critical effect size guidelines for EEM using benthic invertebrate communities. Report to the Environmental Effects Monitoring Program. EEM/1997/9.

Lowell RB, Culp JM, Dubé MG. 2000. A weight-of-evidence approach for northern river risk assessment: integrating the effects of multiple stressors. Environ Toxicol Chem 19:1182–1190.

Top of Page

Mackie GL, White DS, Zdeba TW. 1980. A guide to freshwater mollusks of the Laurentian Great Lakes with special emphasis on the genus Pisidium. Duluth (MN): U. S. Environmental Protection Agency, Office of Research and Development, Environmental Research Laboratory. EPA-600/3-80-068.

Mackie GL, Huggins DG. 1983. Sphaeriacean clams of Kansas. (Technical Publications of the State Biological Survey of Kansas). Lawrence (KS): University of Kansas.

Malley DF, Reynolds JB. 1979. Sampling strategies and life history of non-insectan freshwater invertebrates. J Fish Res Bd Can 36:311-318.

Mason CF. 1991. Biology of freshwater pollution. 2nd edition. New York (NY): Longman Scientific & Technical.

Matthews GWT. 1993. The Ramsar Convention: its history and development. Gland (CH): Ramsar Convention Bureau.

McCafferty WP, Waltz RD. 1990. Revisionary synopsis of the Baetidae (Ephemeroptera) of North and Middle America. Trans Amer Entomol Soc 116:769-800.

Meador MR, Hupp CR, Cuffney TF, Gurtz ME. 1993. Methods for characterizing stream habitat as part of the national water-quality assessment program. Raleigh (NC): U.S. Geological Survey Open-File Report 93-408.

Merritt RW, Cummins KW. 1984. An introduction to the aquatic insects of North America. 2nd edition. Dubuque (IA): Kendall/Hunt.

Merritt RW, Cummins KW. 1996. An introduction to the aquatic insects of North America. 3rd edition. Dubuque (IA): Kendall/Hunt.

Morihara DK, McCafferty WP. 1979. The Baetis larvae of North America (Ephemeroptera: Baetidae). Trans Amer Entomol. Soc 105:139-221.

Morris PA. 1951. A field guide to the shells of our Atlantic and Gulf coasts. (The Peterson Field Guide Series). Boston (MA): Houghton Mifflin Company.

Top of Page

Newbury RW. 1984. Hydrolic determinants of aquatic insect habitats. In Resh VH, Rosenberg DM, editors. The ecology of aquatic invertebrates. New York (NY): Praeger Publishers.

Newbury RW, Gaboury M. 1993. Stream analysis: fish habitat and design: a field manual. Gibson (BC): Newbury Hydraulics.

Norris RH, Georges A. 1993. Analysis and interpretation of benthic macroinvertebrate surveys. In Rosenberg DM, Resh VH, editors. Freshwater biomonitoring and benthic invertebrates. New York (NY): Chapman & Hall. p. 234-286.

Oliver DR, McClymont D, Roussel ME. 1978. A key to some larvae of the Chironomidae (Diptera) from the Mackenzie and Porcupine river watersheds. (Fisheries and Marine Service Technical Report 791). Ottawa (ON): Agriculture Canada.

Top of Page

Pearson TH. 1975. The benthic ecology of Loch Linnhe and Loch Eil, a sea-loch system on the west coast of Scotland. IV: Changes in the benthic fauna attributable to organic enrichment. J Exp Mar Biol Ecol 20:1-41.

Peckarsky BL, Fraissinet PR, Penton MA, Conklin D, editors. 1990. Freshwater macroinvertebrates of northeastern North America. Ithaca (NY): Cornell University Press.

Pennak RW. 1978. Freshwater invertebrates of the United States: Protozoa to Mollusca. 3rd ed. New York (NY): John Wiley & Sons Inc.

Pettibone MH. 1963. Marine polychaete worms of the New England Region. 1. Aphroditidae through Trochochaetidae. (U.S.NationalMuseum Bulletin 227:1). Washington (DC): Smithsonian Institute.

Pettibone MH. 1992. Contribution to the polychaete family Pholoidae Kinberg. Smithsonian Contributions to Zoology 532:1-22.

Pettibone MH. 1993. Scaled polychaetes (Polynoidae) associated with ophiuroids and other invertebrates and review of species referred to Malmgrenia McIntosh and replaced by Malmgreniella Hartman, with descriptions of new taxa. Smithsonian Contributions to Zoology 538:1-92.

Plafkin JL, Barbour MT, Porter KD. 1989. Rapid bioassessment protocols for use in streams and rivers: Benthic macroinvertebrates and fish. Washington (DC): U.S. Environmental Protection Agency, Office of Water Assessment and Watershed Protection Division. EPA/444/4-89-001.

Pohle GW. 1990. A guide to decapod Crustacea from the Canadian Atlantic: Anomura and Brachyura. (Canadian Technical Report of Fisheries and Aquatic Sciences 1771). St. Andrews (NB): Fisheries and Oceans Canada.

Pontasch KW, Smith EP, Cairns J Jr. 1989. Diversity indices, community comparison indices and canonical discriminant analysis: Interpreting the results of multispecies toxicity tests. Wat Res 23:1229-1238.

Top of Page

Rabeni CF, Gibbs KE. 1978. Comparison of two methods used by divers for sampling benthic invertebrates in deep rivers. J Fish Res Bd Can 35:332-336.

Rees HL. 1984. A note on mesh selection and sampling efficiency in benthic studies. Mar Poll Bull 15:225-229.

Reish DJ. 1959. A discussion of the importance of screen size in washing quantitative marine bottom samples. Ecology 40:307-309.

Resh VH, Norris RH, Barbour MT. 1995. Design and implementation of rapid assessment approaches for water resource monitoring using benthic macroinvertebrates. Austral J Ecol 20:108-121.

Reynoldson TB, Bailey RC, Day KE, Norris RH. 1995. Biological guidelines for freshwater sediment based on benthic assessment of sediment (the BEAST) using a multivariate approach for predicting biological state. Austral J Ecol 20:198-219.

Reynoldson TB, Rosenberg DM. 1996. Sampling strategies and practical considerations in building reference data bases for the prediction of invertebrate community structure. In Bailey RC, Norris RH, Reynoldson TB, editors. Study design and data analysis in benthic macroinvertebrate assessments of freshwater ecosystems using a reference site approach. Technical Information Workshop, North American Benthological Society, 44th Annual Meeting, Kalispell, Montana. pp. 1-31.

Reynoldson TB, Norris RH, Resh VH, Rosenberg DM. 1997. The reference condition: a comparison of multimetric and multivariate approaches to assess water-quality impairment using benthic macroinvertebrates. J North Am Benthol Soc 16:833-852.

Robinson CLK, Levings CD. 1995. An overview of habitat classification systems, ecological models and geographic information systems applied to shallow foreshore marine habitats. (Canadian Manuscript Report of Fisheries and Aquatic Sciences 2322). West Vancouver (BC): Fisheries and Oceans Canada.

Robinson CLK, Hay DE, Booth J, Truscott J. 1996. Standard methods for sampling resources and habitats in coastal subtidal regions of British Columbia: Part 2 - Review of sampling with preliminary recommendations. (Canadian Technical Report of Fisheries and Aquatic Sciences 2119). Nanaimo (BC): Fisheries and Oceans Canada.

Rosenberg DM. 1978. Practical sampling of freshwater macrozoobenthos: A bibliography of useful texts, reviews, and recent papers. (Canadian Fisheries and Marine Service Technical Report 790).

Rosenberg DM, Resh VH, editors. 1993. Freshwater biomonitoring and benthic macroinvertebrates. New York (NK): Chapman & Hall.

Top of Page

Saether Ole A. 1975. Nearctic and palaearctic Heterotrissocladius (Diptera: Chironomidae. (Bulletin of the Fisheries Research Board of Canada 193). Ottawa (ON): Fisheries and Marine Service.

Saether Ole A. 1977. Taxonomic studies on Chironomidae: Nanocladius, Pseudochironomus and the Harnischia complex. Bulletin of the Fisheries Research Board of Canada 196). Ottawa (ON): Fisheries and Marine Service.

Sars GO. 1895. An account of the Crustacea of Norway with short descriptions and figures of all the species. Vol. 1 Amphipoda. Christiana and Copenhagen (DK): A.L.B. Cammermeyers Forlag.

Sars GO. 1899. An account of the Crustacea of Norway with short descriptions and figures of all the species. Volume II. Isopoda. Bergen (NO): Bergen Museum.

Sars GO. 1900. An account of the Crustacea of Norway with short descriptions and figures of all the species. Vol. III. Cumacea. Bergen (NO): Bergen Museum.

[SBMNH] Santa Barbara Museum of Natural History. 1994a. Taxonomic atlas. Vol. 1: Platyhelminthes, and Nemertea. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1994b. Taxonomic atlas. Vol. 2: Porifera. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1994c. Taxonomic atlas. Vol. 4: Oligochaeta and Polychaeta: Phyllodocidae to Paralacydoniidae. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1995a. Taxonomic atlas. Vol. 5: The Annelida, Part 2, Polychaeta: Phyllodocida (Syllidae and scale-bearing families), Amphinomida and Eunicida. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1995b. Taxonomic atlas. Vol. 12: the Crustacea, Part 3, The Amphipoda. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1995c. Taxonomic atlas. Vol. 13: The Bryozoa. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1996a. Taxonomic atlas. Vol. 6: The Annelida, Part 3, Polychaeta: Orbiniidae to Cossuridae. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1996b. Taxonomic atlas Vol. 9: The Mollusca, Part 2: The Gastropoda. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1996c. Taxonomic atlas. Vol. 14: Miscellaneous Taxa. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1997a. Taxonomic atlas. Vol 10: The Pycnogonida, The Crustacea Part 1: The Decapoda and Mysidacea. Santa Barbara (CA): Santa Barbara Museum of Natural History.

[SBMNH] Santa Barbara Museum of Natural History. 1997b. Taxonomic atlas. Vol. 11: The Crustacea Part 2, The Isopoda, Cumacea and Tanaidacea. Santa Barbara (CA): Santa Barbara Museum of Natural History.

Schefter P, Wiggins GB. 1986. A systematic study of the nearctic larvae of the Hydropsyche morosa group (Trichoptera: Hydropsychidae). Toronto (ON): Royal Ontario Museum.

Schmitt RJ, Osenberg CW. 1996. Detecting ecological impacts: concepts and applications in coastal habitats. San Diego (CA): Academic Press.

Schultz GA. 1969. The marine isopod crustaceans. Dubuque (IA): Wm. C. Brown Co.

Schuster GA, Etnier DA. 1978. A manual for the identification of the larvae of the caddisfly genera Hydropsyche Pictet and Symphitopsyche Ulmer in eastern and central North America. Cincinnati (OH): U.S. Environmental Protection Agency.

Schwinghamer P. 1981. Characteristic size distributions of integral benthic communities. Can J Fish Aquat Sci 38:1255-1263.

Schwinghamer P. 1983. Generating ecological hypotheses from biomass spectra using causal analysis: a benthic example. Mar Ecol Progr Ser 13:151-166.

Scott DA, Jones TA. 1995. Classification and inventory of wetlands. Vegetatio 118:1-16.

Scrimgeour GJ, Culp JM, Glozier NE. 1993. An improved technique for sampling lotic invertebrates. Hydrobiologia 254:65-71.

Simpson K, Bode R. 1980. Common larvae of Chironomidae (Diptera) from New York state streams and rivers, with particular reference to the fauna of artificial substrates. Bull New York State Mus 439:1-105.

Slack KV, Averett RC, Greeson PE, Lipscomb RG. 1973. Methods for collection and analysis or aquatic biological and microbiological samples. U.S. Geol Surv Tech Water Resour Invest Book 5.

Smith RI. 1964. Keys to marine invertebrates of the Woods Hole region: a manual for the identification of the more common marine invertebrates. Contribution No. 11. Woods Hole (MA): Systematics-Ecology Program, Marine Biological Laboratory.

Smith B, Wilson JB. 1996. A consumer’s guide to evenness indices. Oikos 76:70-82.

Squires HJ. 1990. Decapod crustacea of the Atlantic coast of Canada. Ottawa (ON): Department of Fisheries and Oceans.

Steele DH, Brunel P. 1968. Amphipoda of the Atlantic and Arctic coasts of North America: Anonyx (Lysianassidae). J Fish Res Bd Can 25(5).

Stephenson W, Cook SD. 1980. Skewness of data in the analysis of species-in-sites-in-times. Proc. Royal Soc. Queensland 91:37-52.

Stewart KW, Stark BP. 1993. Nymphs of North American stonefly genera (Plecoptera). Denton (TX): University of North Texas Press

Suess MJ, editor. 1982. Examination of water for pollution control. A reference handbook. Vol. 3. Biological, bacteriological and virological examination. Oxford (UK): Permagon Press.

Top of Page

Tattersall WM, Tattersall OS. 1951. The British Mysidacea. London (UK): Ray Society.

Taylor LR. 1961. Aggregation, variance and the mean. Nature 189:732-735.

Taylor BR. 1997. Optimization of field and laboratory methods for benthic invertebrate monitoring. Final report. Ottawa (ON): Natural Resources Canada, Canada Centre for Mineral and Energy Technology. Aquatic Effects Technology Evaluation Project 2.1.2.

Taylor BR, Bailey RC. 1997. Technical evaluation on methods for benthic invertebrate data analysis and interpretation. Final report. Natural Resources Canada, Canada Centre for Mineral and Energy Technology. Aquatic Effects Technology Evaluation Project 2.1.3.

Tetra Tech, Inc. 1986a. General QA/QC considerations for collecting environmental samples in Puget Sound. Final report. Seattle (WA): Report prepared for U.S. EPA, Region 10. Report TC-3991-04.

Tetra Tech, Inc. 1986b. Recommended protocols for station positioning in Puget Sound. Final report. Seattle (WA): Report prepared for U.S. EPA, Region 10. Report TC-3090-05.

Tetra Tech, Inc. 1987. Recommended protocols for sampling and analyzing subtidal benthic macroinvertebrate assemblages in Puget Sound. Final report. Seattle (WA): Report prepared for U.S. EPA, Region 10. Report TC-3991-04.

Thiel TH. 1975. The size structure of the deep-sea benthos. Int Rev Ges Hydrobiol 60:575-606.

Thorp JH, Covich AP. 1991. Ecology and classification of North American freshwater invertebrates. San Diego (CA): Academic Press.

Tukey J. 1977. Exploratory data analysis. Reading (MA): Addison-Wesley.

Underwood AJ. 1997. Experiments in ecology: their logical design and interpretation using analysis of variance. Cambridge (UK): Cambridge University Press.

Ushakov PV. 1955. Polychaeta of the far eastern seas of the USSR. Acad. Sci. USSR. Translated 1965. Jerusalem (IL): Israel Program for Scientific Translations.

Top of Page

Vezina AF. 1988. Sampling variance and the design of quantitative surveys of the marine benthos. Mar Biol 97:151-155.

Walker EM. 1933. The nymphs of the Canadian species of Ophiogomphus Odonata, Gomphidae. Can Ent 65:217-229.

Walker EM. 1953. The Odonata of Canada and Alaska. Part I, General, Part II. The Zygoptera (damselflies). Vol. I. Toronto (ON): University of Toronto Press.

Walker EM. 1958. The Odonata of Canada and Alaska. Anisoptera. Vol. 2. Toronto (ON): University of Toronto Press.

Walker EM, Corbet PS. 1978. The Odonata of Canada and Alaska. Anisoptera, Macromiidae, Corduliidae, Libellulidae. Vol. 3. Toronto (ON): University of Toronto Press.

Wallace NA. 1919. The Isopoda of the Bay of Fundy. (University of Toronto Studies, Biol. Series No. 18). Toronto (ON): University of Toronto Library.

Waltz RD. 1994. Field recognition of adult Acentrellaand Heterocloeon (Ephemeroptera: Baetidae). Great Lakes Entomologist 26:321-323.

Warwick RM. 1986. A new method for detecting pollution effects on marine benthic communities. Mar Biol 92:557-562.

Warwick RM, Clarke KR. 1993. Increased variability as a symptom of stress in marine communities. J Exp Mar Biol Ecol 172:215-226.

Warwick RM. 1988a. Analysis of community attributes of the macrobenthos of Frierfjord/Langesundfjord at taxonomic levels higher than species. Mar Ecol Prog Ser 46:167-170.

Warwick RM. 1988b. The level of taxonomic discrimination required to detect pollution effects on marine benthic communities. Mar Pollut Bull 19:259-568.

Warwick RM, Pearson TH, and Ruswahyuni. 1987. Detection of pollution effects on marine macrobenthos: Further evaluation of the species abundance/biomass method. Mar Biol 95:193-200.

Watling L. 1979. Marine fauna and flora of the northeastern United States: Cumacea. National Marine Fisheries Service, circular 423:1-22.

Watson J. 1997. A review of ecosystem classification: delineating the Strait of Georgia. Vancouver (BC): Fisheries and Oceans Canada, Pacific Region, Science Branch.

Weber CI, editor. 1973. Biological field and laboratory methods for measuring the quality of surface waters and effluents. Cincinnati (OH): U.S. Environmental Protection Agency. EPA-670/4-73-001.

Weinberg S. 1978. The minimal area problem in invertebrate communities of Mediterranean rocky substrata. Mar Biol 49:33-40.

Weiss HM. 1995. Marine animals of southern New England and New York: identification keys to common nearshore and shallow water macrofauna. Hartford (CT): State Geological and Natural History Survey of Connecticut, Department of Environmental Protection.

Westfall MJ Jr, May ML. 1996. Damselflies of North America. Gainsville (FL): Scientific Publishers.

Wiederholm T. 1980. Use of benthos in lake monitoring. J Wat Poll Cont Fed 52:537-547.

Wiederholm T, editor. 1983. Chironomidae of the holarctic region. Keys and diagnosis. Part 1 - Larvae. Ent. Scand. Suppl. No. 19. Sandby (SE): Scandinavian Society of Entomology.

Wiederholm T, editor. 1986. Chironomidae of the holarctic region. Keys and diagnosis. Part 2 - Pupae. Ent. Scand. Suppl. No. 28. Sandby (SE): Scandinavian Society of Entomology.

Wiggins GB. 1996. Larvae of the North American caddisfly genera (Trichoptera). 2nd Edition. Toronto (ON): University of Toronto Press.

Wood DM, Peterson, BV, Davies, DM, Gyorkos, H. 1963. The black flies (Diptera: Simuliidae) of Ontario. Part II: larval identification with descriptions and illustrations. Proc. Entomol. Soc. Ontario.

Wright JF. 1995. Development and use of a system for predicting the macroinvertebrate fauna in flowing waters. Austral J Ecol 20:181-197.

Figures and Tables

Table 4-1 outlines recommended sampling program designs. Based on the design type, the receiving environment, reference or control area, impact area, and statistics are identified.

Return to the table

Figure 4-1 is a schematic representation of the spatial scales of reference and exposure areas, replicate stations and field sub-samples for a basic control-impact design.

Return to the figure

Figure 4-2 illustrates six examples of control impact designs. Image (a) shows a control-impact design for simpler freshwater rivers and streams or for homogeneous estuarine habitat. Image (b) illustrates a modified C-I design with downstream reference area for streams, rivers, or estuaries. Image (c) shows a magnitude and geographic extent monitoring design. Image (d) illustrates a multiple control-impact design for freshwater rivers and streams with two reference areas. Image (e) shows a multiple control-impact design for freshwater rivers and streams with multiple reference areas in adjacent drainage. Finally, Image (f) illustrates geographically homogeneous lakes, marine bays or inlets, with habitat characteristics similar to the exposure area.

Return to the figure

Figure 4-3 is an illustration of five gradient design examples. Image (a) shows a simple gradient design for freshwater rivers, streams and estuaries. Image (b) illustrates a simple gradient design for lake or costal sites situated in narrow bays or fjords. Image (c) shows a radial gradient design for lake or coastal situations. Image (d) illustrates a multiple gradient design for freshwater rivers. Finally, image (e) shows a multiple gradient design for lake or coastal sites.

Return to the figure

Figure 4-4 is a schematic representation illustrating how the reference and exposure stations are located relative to the effluent input from multiple sources in a reference condition approach.

Return to the figure

Figure 4-5 is a graph illustrating the impairment stress levels derived for reference sites in hybrid multidimensional scaling ordination space. Bands, based on 90, 99, and 99.9% probability ellipses, are identified as A (unstressed), B (possible stressed), C (stressed) and D (severely stressed).

Return to the figure

Table 4-2 outlines the taxonomic keys for benthic invertebrate taxonomic identification in freshwater environments. Each taxon is provided with taxonomic references typically used.

Return to the table

Table 4-3 provides the recommended level of taxanomic precision for benthic invertebrates in a marine environment, for the lowest practical taxonomic approach. Each taxon is aligned with a level. Levels include family, class, sub-class, genus, and species.

Return to the table

Table 4-4 provides a list of marine and estuarine taxonomic benthic invertebrate keys for Canada. References are listed in alphebetical order.

Return to the table

5. Effluent Characterization and Water Quality Monitoring

5.1 Overview

5.2 Sampling Frequency

5.3 Variables Measured

• 5.3.1 Calculation of Loadings

5.4 Sampling Locations

5.5 Reporting

5.6 Effluent Characterization

• 5.6.1 Sampling Methods and Laboratory Analysis
• 5.6.2 Method Detection Limit Change for Mercury
• 5.6.3 Methods for Determination of Thiosalts

5.7 Water Quality Monitoring

• 5.7.1 Preparation for the Field
• 5.7.2 Field Measurement of Water Quality Parameters
• 5.7.3 Collection of Water Samples for Laboratory Analyses
• 5.7.4 Sample Handling, Storage and Analysis for Water Quality Monitoring
  • 5.7.4.1 Sample Handling and Preservation
  • 5.7.4.2 Sample Shipping and Storage
  • 5.7.4.3 Laboratory Analyses of Samples
• 5.7.5 Comparison of Water Quality Data in Exposure and Reference Areas
• 5.7.6 Estimation of Extent of Elevated Concentrations

5.8 Quality Assurance and Quality Control for Water Quality Monitoring

• 5.8.1 General Aspects of Quality Control in the Field
• 5.8.2 Field Aspects of Quality Assurance
• 5.8.3 Quality Assurance during Sample Handling, Shipping and Storage
• 5.8.4 Use of Blanks and Duplicate Samples
• 5.8.5 Quality Control in the Laboratory
  • 5.8.5.1 Details on Quality Control Aspects of Laboratory Analyses
  • 5.8.5.2 Good Laboratory Practices
  • 5.8.5.3 Calibration Control
  • 5.8.5.4 Precision
  • 5.8.5.5 Accuracy and Bias
  • 5.8.5.6 Recovery
  • 5.8.5.7 Detection Limits
  • 5.8.5.8 Data Reporting Conventions
  • 5.8.5.9 Analytical Precision and Accuracy
• 5.8.6 Quality Assurance in the Laboratory
• 5.8.7 Recording and Reporting of QA/QC Information
  • 5.8.7.1 Documentation
  • 5.8.7.2 Reporting of QA/QC Information

5.9 References

Appendix 5-1: Justifications for Parameters for Effluent Characterization and Water Quality Monitoring

List of Tables

• Table 5-1: Analytical parameters measured for effluent characterization and water quality monitoring
• Table 5 2: Summary of recommended use of blanks and duplicate samples in the field and laboratory

5. Effluent Characterization and Water Quality Monitoring

5.1 Overview

The purpose of effluent characterization and water quality monitoring is to answer the following question: “What is the estimated mine-related change in contaminant concentrations in the exposed area?” Data generated from effluent characterization and water quality monitoring are used to:

• monitor changes in the quality of the effluent and environmental conditions in the receiving environment;
• provide an indication of variability in effluent quality and temporal or seasonal trends; and
• provide supporting environmental variables to help interpret results from the biological monitoring (fish and benthic invertebrate community survey) and the sublethal toxicity testing.

Effluent characterization is conducted by analyzing a sample of effluent to provide information on the concentrations of potential contaminants in the mine effluent.

Water quality monitoring is conducted by collecting and analyzing samples of water from the exposure area surrounding the point of entry of effluent into water from each final discharge point and from the related reference areas. In addition, samples of water are collected and analyzed from sampling areas in receiving environments where biological monitoring is completed (Metal Mining Effluent Regulations [MMER], Schedule 5, section [s.] 7).

Top of Page

5.2 Sampling Frequency

Effluent characterization and water quality monitoring shall be conducted 4 times per calendar year and not less than 1 month apart on the samples of effluent and water collected, while the mine is depositing effluent (MMER, Schedule 5, s. 7). It is recommended that, where possible, samples for effluent characterization and water quality monitoring be collected once in each calendar quarter. It is also recommended that samples for effluent and water be collected on the same day.

The following factors should be taken into consideration to decide when the aliquots of effluent samples are collected for effluent characterization:

• seasonal variability of effluent, based on composition and flow;
• the time of year when previous effluent samples have been collected;
• the time of year when sampling for water quality monitoring is being conducted; and
• the time of year when concentrations of the contaminants are expected to be highest in the exposure area.

For water quality monitoring, the following factors should be taken into consideration to decide when water samples are collected in the receiving environment:

• seasonal variability in water quality and flow in the exposure area;
• the time of year when concentrations in the exposure area of contaminants are expected to be highest;
• the time of year when previous water quality monitoring samples have been collected;
• the time of year when samples for effluent characterization are collected; and
• the time of year when the biological monitoring is conducted.

Top of Page

5.3 Variables Measured

Effluent characterization and water quality monitoring are conducted for the parameters listed in Table 5-1. If a mine does not use cyanide as a process reagent within the operations area, cyanide does not need to be recorded (MMER, Schedule 5, paragraph [p.] 7(1)(d). Also, if the concentration of total mercury is less than 0.10 mg/L in 12 consecutive samples, the recording of the concentration of total mercury may be discontinued (MMER, Schedule 5, subsection [ss.] 4(3)). It is recommended that a letter be sent to the appropriate authority in Environment Canada advising that the mine has fulfilled the above requirement. Table 5-1 also includes optional parameters recommended on a site-specific basis that the owner or operator of a mine may record as additional supporting information in order to conduct a more complete chemical characterization. In addition to the required parameters listed in Table 5-1, the measurement of some effluent parameters such as conductivity, sulphate or chloride concentrations may be useful as tracers to determine the extent of effluent mixing in the exposure area. In addition, the concentrations of calcium, magnesium, chloride, potassium, sodium, sulphate and dissolved organic carbon can be used to estimate the potential toxicity of some metals using the biotic ligand model approach (e.g., US EPA 2007; Reiley 2007). Appendix 5-1 includes the justification for the parameters for effluent characterization and water quality monitoring.

¹ All concentrations are total values; dissolved concentrations may also be reported; effluent loading (MMER, s. 20) will also be calculated and reported.
² In situ measured parameters.
³ These other parameters are potential contaminants or supporting parameters; analysis is optional and may be added based on site-specific historical monitoring data or geochemistry data.
⁴ The recording of the concentration of total mercury in effluent may be discontinued if that concentration is less than 0.10 µg/L in 12 consecutive samples (MMER, Schedule 5, ss. 4(3).
⁵ Cyanide does not need to be recorded if that substance is not used as a process reagent within the operations area (MMER, Schedule 5, s. 7(d)).
⁶ In the case of effluent that is deposited into freshwater, record the pH, hardness and alkalinity of the water samples.
⁷ In the case of effluent that is deposited into estuarine waters, record the pH, hardness, alkalinity and salinity of the water samples.
⁸ In the case of effluent that is deposited into marine waters, record the salinity of the water samples.
⁹ Radium 226 does not need to be recorded if the conditions of ss. 13(2) of the MMER are met.
¹⁰ Please refer to Environment Canada document: Guidance Document for the Sampling and Analysis of Metal Mining Effluent (EPS 2/MM/5) for methods. Temperature calibration, and compensation when measuring conductivity, should be done according to the manufacturer’s specifications.

Top of Page

5.3.1 Calculation of Loadings

The MMER requires mines to record the monthly mass loadings of the MMER-prescribed deleterious substances (MMER, s. 20). As part of effluent characterization for environmental effects monitoring (EEM), it is also recommended that mines calculate effluent loadings of the other parameters monitored. Loading can be calculated by multiplying the mean effluent concentration of the parameter by the total volume of effluent discharge over the time period of interest (typically 1 year for effluent characterization).

Top of Page

5.4 Sampling Locations

Samples for effluent characterization shall be collected from each final discharge point, identified by the owner or operator of the mine and in accordance with the MMER (Schedule 5, ss. 4(2)).

Guidance for determining the sampling location(s) for effluent characterization is provided in the Guidance Document for the Sampling and Analysis of Metal Mining Effluents: Final Report (Fowlie et al. 2001). This document focuses primarily on methods for collection of effluent samples from point sources (end of the pipe). If samples are to be collected from nonpoint sources, proposed sample collection locations and methods should be discussed with the Authorization Officer.

Water quality monitoring is conducted by collecting samples of water from the exposure area surrounding the point of entry of effluent into water from each final discharge point and from the related reference areas (MMER, Schedule 5, ss. 7(1)). These sampling stations will not likely be the same sampling stations used for biological monitoring. In selecting sampling stations for the exposure area, the owner or operator of a mine should take into consideration the location where effluent concentrations are the highest.

In addition to the above, the owner or operator of a mine shall collect samples of water from the sampling areas selected for the fish population and fish tissue studies and the benthic invertebrate community studies. Therefore, water quality monitoring is conducted at the same time as the biological monitoring studies, should the mine be required to conduct these studies (MMER, Schedule 5, p. 7(1)a(ii)). The water samples are analyzed for the water quality monitoring variables outlined in Table 5.1.

It is recommended that at least 3 water samples be collected at each sampling station to provide an estimation of the variability and determine if concentrations of the contaminants are homogeneous within the sampling station. However, this may not be sufficiently robust to assess data statistically. More sampling stations within each area may help to better understand contaminant concentrations in the exposure area. At the minimum, a composite sample, consisting of few sub-samples spaced within the station, should be collected.

It is strongly recommended, where the benthic and/or fish sampling areas are not in close proximity to the sampling stations for water quality monitoring, that samples be collected concurrently at the sampling stations for the routine water quality monitoring. This will help to interpret the results of analyses of water samples collected in the benthic and/or fish sampling areas in comparison with the results of water samples collected under water quality monitoring.

Top of Page

5.5 Reporting

The results of effluent characterization and water quality monitoring shall be submitted to the Authorization Officer as part of an effluent and water quality monitoring report (MMER, Schedule 5, s. 8). As per Schedule 5, s. 8 of the MMER, a report on the effluent and water quality monitoring studies conducted during a calendar year shall be submitted to the Authorization Officer not later than March 31 of the following year. See Chapter 10 for information on electronic reporting of effluent and water quality monitoring data. The annual effluent and water quality monitoring report should include the following information.

(Note that in the list below, the regulatory requirements (as per the MMER, Schedule 5, s. 8) are written in italics; these requirements are followed by further recommendations and descriptions.

a) The dates on which each sample was collected for effluent characterization, sublethal toxicity testing and water quality monitoring:

• four dates for effluent characterization (4 times per calendar year and not less than 1 month apart (MMER, Schedule 5, ss. 7(2)), while the mine is depositing effluent;
• four dates for water quality monitoring (4 times per calendar year and not less than 1 month apart (MMER, Schedule 5, ss. 7(2)), while the mine is depositing effluent;
• dates for sublethal toxicity testing (2 times each calendar year for 3 years and once each year after the third year, with the first testing to occur on an effluent sample collected not later than 6 months after the mine becomes subject to section 7 of these regulations (MMER, Schedule 5, s. 6)). The sublethal toxicity testing date(s) should match the date(s) for effluent characterization, as the sublethal toxicity sample must be an aliquot of the effluent characterization sample; and
• if the required number of tests were not conducted, indicate the reason why (i.e., the number of days that the effluent was being discharged or the habitat conditions that prevented the collection of effluent characterization and/or water quality monitoring samples).

b) The locations of the final discharge points from which samples were collected for effluent characterization, noting that effluent characterization is conducted at ALL identified final discharge points (FDPs).

c) The location of the final discharge point from which samples were collected for sublethal toxicity testing and the data on which the selection of the final discharge point was based, in compliance with the MMER, ss. 5(2):

• Indicate from which FDP the effluent was collected for the sublethal toxicity testing.
• Indicate why that FDP was chosen, if there is more than one FDP at the mine site (e.g., effluent that discharges into a sensitive receiving environment, has the greatest mass loading).

d) The latitude and longitude of sampling areas for water quality monitoring, in degrees, minutes and seconds, and a description that is sufficient to identify the location of the sampling areas:

• If units other than latitude and longitude were taken (e.g., UTMs) at the sampling station, there are Web-based tools provided by Natural Resources Canada that can be used to convert them.
• Provide a written description (possibly supplemented with maps) of the sampling station that is sufficient to identify the location within the sampling areas. For example, “water collected under first bridge.” This description should allow for ease of re-sampling at the same stations.

e) The results of effluent characterization, sublethal toxicity testing and water quality monitoring:

• Include the results from all analyses completed on effluent (chemical and physical parameters), sublethal toxicity testing and water quality monitoring.
• Include results from all required parameters, as well as any optional site-specific parameters that were measured (see Table 5.1).
• For sublethal toxicity testing, the laboratory reports should be included as an appendix in the annual report.

f) The methodologies used to conduct effluent characterization and water quality monitoring, and the related method detection limits:

• Some sampling methods are outlined in the Guidance Document for the Sampling and Analysis of Metal Mining Effluent: Final Report (Fowlie et al. 2001); available at http://dsp-psd.pwgsc.gc.ca/Collection/En49-24-1-39E.pdf.
• Indicate the methodology used (e.g., inductively coupled plasma combined with mass spectrometry [ICP-MS], graphite furnace atomic absorption spectrometry [GFAAS]) for effluent characterization and water quality monitoring.
• Indicate the method detection limits for the methodology used--for MMER deleterious substances, the method detection limits identified in Schedule 3 of the MMER should be met. See section 5.6.2 of these guidelines for the method detection limit for mercury. Note that the Canadian Council of Ministers of the Environment’s Canadian Environmental Quality Guidelines (CCME 1999) (Chapter 4: Canadian Water Quality Guidelines for the Protection of Aquatic Life) or additional provincial water quality guidelines should also be considered.
• Indicate whether the Canadian Water Quality Guidelines for the Protection of Aquatic Life (CCME 1999 - Chapter 4) or additional provincial water quality guidelines are met.

g) A description of quality assurance and quality control measures that were implemented and the data related to the implementation of those measures:

• Provide a brief description of the quality assurance / quality control (QA/QC) measures that were taken and the results of such measures with respect to collection of the effluent and water sampling, shipping and storage.
• See sections 7.3–7.5 of Fowlie et al. (2001) and section 5.8 below for further information.

Since effluent samples for effluent characterization are aliquots of samples collected for effluent compliance monitoring, the measurements of pH and the concentrations of the deleterious substances (arsenic, copper, total cyanide, lead, nickel, zinc, radium 226 and total suspended solids) should be available as part of the effluent characterization and water quality monitoring reports from each mine.

Top of Page

5.6 Effluent Characterization

5.6.1 Sampling Methods and Laboratory Analysis

Since effluent samples for effluent characterization arealiquots of samples collected for effluent compliance monitoring as part of the MMER, the sampling and chemical analysis considerations and recommended procedures provided in the Guidance Document for the Sampling and Analysis of Metal Mining Effluent: Final Report (Fowlie et al. 2001) apply also to effluent characterization conducted as part of the EEM program. The volume of sample taken should be sufficient to allow for all required analyses and tests plus associated quality control samples (e.g., field duplicates, laboratory replicates and spiked sample).

Top of Page

5.6.2 Method Detection Limit Change for Mercury

The method detection limit for mercury in effluent has been changed to 0.01 µg/L (0.00001 mg/L) so that the concentration of 0.1 µg/L specified in Schedule 5, s. 9(c) of the MMER can be detected with confidence. Analytical methodologies suitable to achieve this level of detection include cold vapour atomic absorption spectrometry (CVAAS), cold vapour atomic fluorescence spectrometry (CVAFS) and inductively coupled plasma mass spectrometry (ICP-MS).

Top of Page

5.6.3 Methods for Determination of Thiosalts

Total thiosalts is an optional site-specific parameter that may be measured in mine effluent; however, information on sampling or analysis is not available in Fowlie et al. (2001). Thiosalts are soluble sulphur/oxygen ions that form as a result of the incomplete oxidation of sulphide minerals. They have the potential to be generated whenever sulphide minerals are in contact with oxygen, but in practice they tend to be formed during the processing of ores bearing sulphide minerals. If thiosalts occur in effluent, once the effluent is discharged the oxidation of the thiosalts is completed; this results in the generation of sulphuric acid and the lowering of pH in the exposure area. Such pH alterations in receiving waters could also be related to low levels of thiosalts and to thiosalt speciation, which cannot be addressed entirely with commonly used analytical techniques (Vigneault et al. 2002). At a concentration of 10 ppm, thiosalt degradation can still potentially drop the pH to 3.7 in unbuffered receiving water (Vigneault et al. 2002). Information related to thiosalt speciation may be required to predict pH depression, since individual thiosalt species can produce different amounts of acidity and are stable in markedly different conditions.

Despite the ability of thiosalts to alter pH in receiving water bodies, toxicity due to thiosalts in mine effluent has been limited to a few sites. This may be due to the low toxicity of thiosalts to animals. Thiosalts are not expected to be acutely lethal in mine effluents, with lethal concentrations for Rainbow Trout higher than 800 mg/L (Schwartz et al. 2006). Sublethal toxicity testing suggests further that the sensitivity of aquatic species to thiosalts and the toxicity of the different anions composing thiosalts vary by an order of magnitude. Schwartz et al. (2006) reported that Ceriodaphnia dubia was the most sensitive EEM test species, with a 25% inhibition concentration (IC₂₅) of 60 mg/L for thiosulphate. They further noted that tetrathionate was much less toxic than thiosulphate. Few mine sites in Canada have known thiosalt problems, but the potential for thiosalt generation may exist at many mine sites. As such, total thiosalt determination is optional for effluent characterization and water quality monitoring in the EEM program.

The total concentration of thiosalts is most commonly determined with a titration method having a detection limit around 10 ppm (expressed as thiosulphate) (Makhija and Hitchen 1979). Thiosulphate is stable at neutral pH and unstable at low pH, while the opposite is true for polythionates. Ion chromatography can be used to determine the concentration of different thiosalt species in synthetic solutions in the ppb range, but it is difficult to apply to field samples because of the instability of thiosalt. In order to better predict the environmental impacts of thiosalts and thiosalt degradation, more information regarding in situ speciation and measurement methods with lower detection limits are required.

The main concern with this method is that the samples should be analyzed within 24 hours. Given that every available preservation method has limitations, there is in fact no substitute for immediate analysis (O’Reillyet al. 2001). As a result, total thiosalts analyses should ideally be done on-site. These analytical capabilities are likely restricted to sites with known thiosalt problems. Alternatively, samples may be frozen immediately after collection and be analyzed within 7 days. Longer storage time of frozen samples may affect thiosalt stability. Alternatively, an anion exchange resin can also be used to preconcentrate and preserve thiosalts (Drushel et al. 2003; Vigneault et al. 2002).

5.7 Water Quality Monitoring

5.7.1 Preparation for the Field

The reagents for cleaning, operating or calibrating equipment and collecting, preserving and/or processing samples should be handled by appropriately qualified personnel, and the appropriate data for health and safety (e.g., Material Safety Data Sheets) should be available.

Written protocols and standard operating procedures (including QA/QC requirements) should be readily accessible at all times, to ensure proper and safe operation of equipment. Data forms and logbooks should be prepared in advance so that field notes and data can be quickly and efficiently recorded. Extra forms should be available in the event of a mishap or loss. These forms and books should be waterproof and tear resistant. Under certain circumstances, audio or audio/video recordings may prove valuable.

All equipment used to collect and handle samples should be cleaned and all parts examined to ensure proper functioning (e.g., on-site assembly or operation) prior to going into the field. A repair kit should accompany each major piece of equipment in case of equipment failure or loss of removable parts. Backup equipment, batteries and sampling gear should be available. Sampling equipment used for field measurements of water quality parameters should be properly calibrated or standardized according to the manufacturer’s recommendations.

All sample containers and required preservatives should be provided by the laboratory hired to conduct the analyses of the samples. Bottles should preferably be unused and purchased as certified clean. If bottles are reused, they should be cleaned by a documented cleaning procedure with a bottle lot number-control system, and cleanliness should be demonstrated by the use of blanks (Fowlie et al. 2001).

Storage, transportand sample containers, including extra containers in the event of loss or breakage, should be pre-cleaned and labelled appropriately (i.e., with a waterproof adhesive label to which the appropriate data can be added with an indelible ink pen capable of writing on wet surfaces). The containers should have lids that are fastened securely and the appropriate container lids and lid liners should be used to prevent contamination (e.g., lid liners should be lined with an inert material like Teflon^®, not paper or cardboard). A sample-inventory log and a sample-tracking log should be prepared in advance of sampling. The responsibility for these logs should be assigned to one individual who will be required to monitor the samples from the time they are collected until they are analyzed and disposed of or archived.

Top of Page

5.7.2 Field Measurement of Water Quality Parameters

Standard in situ water quality parameters are dissolved oxygen, pH, conductivity, water temperature and salinity (marine and estuarine environments). Total water depth at the sampling area and water depth from where the water sample was collected should be recorded. Optical depth or transparency should also be measured in the field. Current flow should also be measured in riverine environments. Measurements of standard water quality parameters can be taken in the water directly, from a sample container in the boat or on shore immediately after collection of the sample, as long as the water is collected at the appropriate depth. If dissolved oxygen measurements are conducted on the shore, special care should be taken to ensure that air is not introduced into the sample.

In shallow water bodies ≤ 2 m deep, standard water quality parameters need only be measured at mid-depth. If the depth ranges from 2 to 4 m, standard water quality measurements should be taken at 2 depth intervals: approximately 25 cm above the bottom and 25 cm below the surface. In deeper bodies of water, measurement of standard water quality parameters should be taken throughout the water column. Information on bottom depth and water column profiles of conductivity, pH, hardness, alkalinity, salinity, temperature and dissolved oxygen should be obtained along intervals of 1 to 5 m (depending on total depth). For example, at a depth of 5 m, measurements should be taken every metre. At a depth of 25 m, measurements should be recorded at 5-m intervals.

For deep samples, a peristaltic sampler, with appropriate lengths of Teflon^® tubing, should be used in preference to other types of pumps. If other types are used, they should be Teflon^®-coated and non-metallic. Sampling should proceed from the least-contaminated to the most-contaminated station, with a weak nitric acid and distilled water rinse between stations. The solvent rinsate should be collected and returned to the laboratory for proper disposal. Laboratory blanks of the samplers should be run before and after use to demonstrate that no contamination is imparted to samples (Fowlie et al. 2001).

Profiles can be facilitated through the use of a data logger (or equivalent) equipped with a dissolved oxygen probe and associated stirrer, as well as pH, conductivity, depth and temperature probes, which evaluate water column quality simultaneously. Such a unit is particularly useful for deeper evaluations (> 50 m). During profiling, the operator is able to visually review incoming data, noting particular areas of interest during descent and ascent of the unit (e.g., conductivity spikes, thermocline, unusual data records). This information is recorded either manually or directly stored in the data logger. To supplement computer records, parameter readings should be recorded manually onto field data sheets (every 2 or 5 m) depending on total depth profiled.

At shallow depths, hand-held meters are often the most convenient way to measure in situ water quality parameters. They are light, and several models are now available that can measure standard water quality parameters. The probes and the cables connecting them to the hand-held unit can range from 2 to 5 m, limiting the use of such a unit. These meters tend to require more regular maintenance and calibration, meaning extra care should be taken to make sure that the meters are in proper functioning order. Calibration and maintenance logs should be kept on file.

Water depth can be measured indirectly using a sonar-based fish finder, or directly using a calibrated tape, sounding cable or rod. Recommended accuracy is as follows:

• Water depth less than 2 m: recommended accuracy of ± 25 cm
• Water depth of 2-10 m: recommended accuracy of ± 50 cm
• Water depth greater than 10 m: recommended accuracy of ± 1 m

Optical depth is a measure of the transparency of water, and can be measured with a turbidity meter in the field or in the laboratory. Optical depth can also be measured using a Secchi disk. The disk is 20 cm in diameter, and is painted white in two opposite quarters and black in the other two. The disk is attached to a calibrated tape. To measure optical depth, the disk is lowered into the water in the shade until it has disappeared. It is then raised slowly, and the water depth at which it reappears is recorded. At least two measurements should be made at each station, and optical depth should be estimated based on the median value of the measurements. Measurements should be made at midday, and sunglasses should not be worn while measurements are made (Nielsen and Johnson 1983).

Water quality data should be screened on-site during sample collection to prevent the measurement and recording of false readings, as doing so will permit the use of alternative instrumentation or instrument checks in the event of equipment or sampling error. All sampling and monitoring equipment should be checked and calibrated daily, if necessary, to ensure good working condition.

It is recommended that additional field measurements and observations be recorded:

• sample number, replicate number, site identification (e.g., name);
• time and date of the collection of the sample;
• ambient weather conditions, including wind speed and direction, wave action, current, tide, vessel traffic, temperature of both the air and water, thickness of ice if present;
• sampling area location (e.g., positioning information) and location of any replicate samples;
• type of platform/vessel used for sampling (e.g., size, power, type of engine);
• name of personnel collecting the samples;
• details pertaining to unusual events that might have occurred during the operation of the sampler (e.g., possible sample contamination, equipment failure, unusual appearance, control of vertical descent of the sampler); and
• deviations from standard operating procedures.

Top of Page

5.7.3 Collection of Water Samples for Laboratory Analyses

Water samples collected in the field and sent to a laboratory for analysis make up the bulk of the water quality monitoring program and involve the analysis of metals, nutrients, major anions and cations, and several other general water quality variables.

Total metals analysis (total values) is required (MMER, Schedule 5, s. 4) during water quality monitoring, as studies often found no difference between measuring “total” and “dissolved” metals (ESG 1999). However, significant differences between total and dissolved metals can be found in some cases, and analyzing for metals in both the dissolved and total fractions could be relevant on a site-specific basis especially in the context of investigation of causes.

In general, samples should be collected at 2 depth intervals: the subsurface (epilimnion) and near bottom (hypolimnion) in order to obtain samples from both areas of the water column (above and below the thermocline). If the water depth is ≤ 2 m, it is sufficient to collect water samples only at mid-depth or at least 15 cm below the surface. Samples collected below the surface of the water can be collected by hand directly into the sample bottle.

Water collections at discrete depths should be facilitated through the use of appropriate samplers (e.g., Niskin sampler, non-metallic 2-16–L Van Dorn or 0.5-8–L Kemmerer samplers). For streams, depth-integrated samplers that are representative of the suspended sediment and related substances can be used. These samplers can be used from a boat, bridge or ice surface, and usually require two persons for safe operation. For very deep samples, a peristaltic sampler is preferred to other types. If other samplers are used, they should be Teflon^®-coated.

The water sampler should be triple-rinsed with the water from the sampling station between each sample. In addition, it is recommended that sampling in the reference area be completed first to avoid any potential contamination of the sampler with water from the exposure area. The sampler should be double-rinsed with reagent-grade weak nitric acid between sampling areas, particularly if it is not possible to complete sampling in the reference area first. The solvent residues should be collected and returned to the laboratory for proper disposal. Laboratory blanks of the samplers should be run before and after use to demonstrate that no contamination is imparted to the samples.

When collecting water samples, it is important to use as many of the following ultra-trace techniques and proper water sampling protocols as possible:

• Sampling should proceed from the least-contaminated to the most-contaminated station.
• Sample bottles and caps should be rinsed 3 times prior to water collection.
• No preservatives should be placed into the sampling bottles prior to sample collection.
• Samples should be collected with the bottle mouth facing up-current and away from the sample collector’s hand.
• At no time should the inside of the sample container, the bottle mouth, or the inside of the container lid be touched by sample collectors, even while wearing disposable gloves.
• Sample collectors should wear unlined, powder-free latex or nitryl gloves to avoid contamination of the sample.
• Label all samples immediately and clearly, and follow proper preservation techniques. Record all sampling data in the field notebook immediately.
• Caps of water containers should be held lid-down during sample collection.
• The sampling-point locations should be recorded.

Top of Page

5.7.4 Sample Handling, Storage and Analysis for Water Quality Monitoring

5.7.4.1 Sample Handling and Preservation

The Guidance Document for the Sampling and Analysis of Metal Mining Effluents: Final Report (Fowlie et al. 2001) contains information on sample handling recommendations regarding containers, preservatives and holding times for specific parameters. Where appropriate, preservatives should be added to the sample bottle immediately upon completion of the collection. The actual sample volumes required may vary depending on the needs of the laboratory.

Note that to reduce the number of samples collected, several analytes may be analyzed from one sample bottle. Prior to sample collection, the list of variables should be discussed with the laboratory to determine the number and type of sample bottles required.

When collecting samples, it is useful to have a checklist that lists the collection bottles, corresponding analytes, and whether or not a preservative is required. As a sample is collected it should be checked off the list. In certain situations, a maximum holding time of 7–10 days (major cations and anions, nitrate/nitrite, dissolved organic carbon) may be problematic. If the shipping of a mine’s water samples has been unavoidably delayed but the integrity of samples was retained, the Authorization Officer should be notified without delay.

Top of Page

5.7.4.2 Sample Shipping and Storage

It is recommended that samples be cooled to 4°C during collection and stored at the same temperature for shipping, to minimize degradation. Samples should also be refrigerated, and shipping coolers should be equipped with ice packs or bagged ice to ensure that samples are kept cold.

Samples should be transported to a laboratory as soon as possible after collection (within 24-48 hours maximum). Analyses should be completed within the accepted storage times, which will vary depending on the variable. Storage time is defined as the time interval between the end of the sample collection period and the initiation of analyses. All samples should be stored for as short a time interval as possible and under conditions that minimize sample degradation. Samples should be maintained at temperatures above their freezing point and under 10°C, with minimal exposure to light. Samples digested for metals analysis may be maintained in a sealed container and analyzed within 30 days. For additional information refer to Fowlie et al. (2001).

Top of Page

5.7.4.3 Laboratory Analyses of Samples

Laboratory analyses should be carried out in a qualified laboratory by trained personnel operating under quality-controlled conditions and using documented standard operating procedures. Laboratories contracted by the mining industryshould be accredited under the International Organization for Standardization standard ISO/IEC 17025:2005 entitled “General requirements for the competence of testing and calibration laboratories,” as amended from time to time. The analytical methods selected should be generally accepted and in common use in laboratories in Canada. The overall method principle should be peer-reviewed and published in a widely available publication so that it can be located easily for details.

The analytical methods selected should meet the criteria in this document plus any other objectives identified by the mine (or those acting on the mine’s behalf) or Environment Canada. The project manager and the laboratory need to confirm what parameters of interest will be measured and that holding times can be met. The laboratory and analysis methods should be selected and discussed before the sample is collected, to ensure that the laboratory sample requirements are met.

The methods chosen should reliably measure the detection limits indicated for the deleterious substances identified in Schedule 3 of the MMER (i.e., any concentration above about one-tenth of the maximum authorized sample concentration (Fowlie et al. 2001). Normally accepted methods, method detection limits, and precision and accuracy objectives for metal mining effluent analysis are discussed in Fowlie et al. (2001). For the other required or site-specific recommended water quality parameters, for which detection limits are not specified, if there is a CCME Canadian environmental quality guideline (CCME 1999) for the variable measured, the chosen method’s detection limits should be sufficiently low to determine if the parameters measured exceed these guidelines. CCME guidelines can be found at http://ceqg-rcqe.ccme.ca. Several provinces have also developed water quality guidelines, and in cases where both CCME and provincial water quality guidelines exist for a particular parameter, the provincial guidelines take precedence, although both should be reported.

Top of Page

5.7.5 Comparison of Water Quality Data in Exposure and Reference Areas

It is recommended that the Biological Interpretative Report include a comparison of water quality data in the exposure and reference areas. This comparison would examine all parameters included as part of water quality monitoring. In particular, this comparison should identify any parameters for which there are differences in measurements taken in the exposure and reference areas of more than a factor of 2. This comparison is intended to help with the interpretation of biological data in the interpretative report.

The factor of 2 for exceedance of concentrations from the reference area is intended to ensure that differences between exposed and reference area concentrations are real differences, and not just differences that may be attributed to such factors as low concentrations of target contaminants, analytical variability, small minimal sample size (n = 4) and seasonable variability. At sites where the reference area is on a different water body or watershed than the exposed area, the factor of 2 difference may not be applicable.

Determination of whether or not concentrations are different between the exposure and reference areas should be based on the median value of a minimum of 4 samples collected over a 12-month period from the same exposure and reference area locations. The median in a set of n measurements y₁, y₂, y₃, … y_n is defined to be the value of y that falls in the middle when the measurements are arranged in order of magnitude. If there is an even number of measurements, then the median is the value of y halfway between the two middle measurements. If larger data sets (n >> 4) are available, determination of whether concentrations are elevated in the exposure area could be based on a statistical test such as mean or median greater than the 95% confidence interval or greater than 2 standard deviations. If there are adequate pre-mining water quality data in the exposure area, then pre-mining data may be used as a basis for comparison.

In cases where there are differences of more than a factor of 2, it is recommended that the mine estimate and report the geographical extent for which this condition exists, based on expanded water quality monitoring or modelling. However, before completing such an estimate, there are a number of factors that should be considered:

1. Site-specific water quality objectives: If there is a site-specific water quality objective for a particular parameter, and that objective is exceeded in the exposure area, the extent of this exceedence should be determined, regardless of the concentration in the reference area.
   
2. Water quality guidelines: If there are water quality guidelines for a particular parameter (e.g., federal or provincial), and the concentrations of that parameter in the exposure area are greater than concentrations in the reference area by more than a factor of 2, and are greater than the water quality guideline, then the extent of this exceedence should be determined.
   
   The CCME Canadian Environmental Quality Guidelines (CCME 1999 - Chapter 4: Canadian Water Quality Guidelines for the Protection of Aquatic Life) for water quality monitoring parameters are available at http://ceqg-rcqe.ccme.ca/. Several provinces have also developed water quality guidelines, and in cases where both CCME and provincial water quality guidelines exist for a particular parameter, the provincial guidelines take precedence, although both should be reported.
   
3. Detection limits: In cases where the generic water quality guideline for a parameter is close to the analytical method detection limit, and concentrations of the parameter in the study area are close to the guideline, a factor of 2 difference may not be meaningful (as a result of analytical uncertainty close to the detection limit). In such cases the Authorization Officer should be consulted. McQuaker (1999) provides a comparison of achievable detection limits with generic water quality guidelines and, for most parameters, method detection limits (MDL) significantly less than the water quality guideline (WQG) (at least a ratio of 1:10 MDL:WQG) are available. However, McQuaker concluded that there are some parameters (arsenic, cadmium, mercury, selenium, silver and cyanide) for which an MDL at least 10 times lower than the WQG is not currently achievable. As the MDL:WQG ratio decreases, the measurement uncertainty increases; beyond a ratio of 1:2, the results are not considered to be statistically significant.
   
4. pH: For pH, a difference of a factor of 2 may be particularly important, since the pH scale is logarithmic. If there is a site-specific objective for pH, and if the pH in the exposure area is outside the range specified in the site-specific objective, the geographical extent of pH values outside the range of the site-specific objective should be determined. If there is no site-specific objective for pH, and if the pH in the exposure area is more than 0.5 pH units different than the pH in the reference area and is also outside Canadian environmental guidelines (e.g., CCME 6.5 to 9.0), the geographical extent of the exposure area within which pH is more than 0.5 pH units different from the reference area should be determined. According to the Canadian Environmental Quality Guidelines (CCME 1999), human activity should not alter the pH by more than 0.2 pH units in marine or estuarine environments.
   
5. Location of the reference area: At sites where the reference area is on a different water body or watershed than the exposure area, a difference of a factor of 2 may not be applicable. If it is felt that this is the case, the Authorization Officer should be consulted.

Top of Page

5.7.6 Estimation of Extent of Elevated Concentrations

Two methods may be used to estimate the extent of elevated concentrations:

• direct measurement
• modelling

1) Direct Measurement

Direct measurement requires an increase in the number of sampling stations within the exposure area to determine where within the exposure area concentrations of contaminant(s) of concern are no longer elevated. The number of additional stations needed would be determined on a site-specific basis, but generally, a minimum of 3 stations would be needed:

• the exposure area station used for water quality monitoring, and
• two or more further stations that bracket the location within the exposure area where the concentrations of parameters of concern are no longer expected to be elevated.

2) Modelling

If the seasonal variations of concentrations of the key parameter of concern in effluent and the exposure area are well understood, and if seasonal variations in effluent and receiving environment flow are well understood, it may be possible to predict the location within the exposure area where the concentrations of the parameter(s) of concern are no longer expected to be elevated.

5.8 Quality Assurance and Quality Control for Water Quality Monitoring

General aspects of quality assurance and quality control are discussed in Chapter 2.

5.8.1 General Aspects of Quality Control in the Field

General QC aspects of a field sampling program are as follows:

• All personnel involved in field procedures should have appropriate education and training.
• Sampling methods should be consistently applied among sites throughout the study.
• Samples should be collected according to standard operating procedures that should be available to personnel at all times during the field study.
• Sampling equipment should be appropriate for the habitat being studied, properly cleaned, and accompanied by the appropriate documentation (i.e., manual, calibration and maintenance schedule).
• All samples should be properly labelled with date, location, type, number and collector’s name.
• Samples should be in the proper container with the appropriate preservative or fixative if necessary.
• Field technicians should maintain detailed field notes using indelible ink and waterproof notebooks.
• Personnel should use chain-of-custody / sample submission forms and custody seals for contaminant samples.
• Personnel should follow appropriate shipping and storage methods.
• Standardized field collection forms should be used during the field program.

Top of Page

5.8.2 Field Aspects of Quality Assurance

Field QA for water quality monitoring should be achieved through several methodologies, including duplicate readings, comparison of readings with known standards, collection of profile samples for analytical evaluation, and parameter evaluation using alternate equipment (e.g., Hanna CTD meter, thermometer).

Some of the most common quality problems are the result of mislabelling or switching bottles, failure to add proper preservatives, improper storage conditions, sample contamination from sampling equipment, and exceeding the holding time. Each sample should be clearly labelled in a manner that identifies the sample and distinguishes it from all other samples. Labels should be filled out in indelible ink and fixed to the sample container such that they will not fall off when wet or during transport.

The field logbook is an integral part of the sampling program and forms the basis of the sampling report. Items documented in the logbook are often highly relevant to the interpretation of the laboratory data. Any deviations from the sampling plan or any other observation about the sample or the sampling locations should also be noted in the logbook. Some common deficiencies in field logbooks include the failure to make planning notes, make notes at the time events occur, sign and date entries, and write legibly.

Top of Page

5.8.3 Quality Assurance during Sample Handling, Shipping and Storage

The Canadian Association for Environmental Analytical Laboratories (CAEAL) (currently the Canadian Association for Laboratory Accreditation [CALA 1991]) recommends the following with respect to QA during sample handling, shipping and storage:

1. Chain of Custody: Chain-of-custody forms should be used in the transportation of samples, especially in cases where several contracted parties are involved in the sampling, shipping and analysis of the samples.
2. Sample Inspection: The condition of each sample should be noted upon receipt. Discrepancies between required sample conditions and the observed conditions should be recorded in a logbook or on a computer file. It is preferable to preserve samples in the field immediately. However, the samples should be preserved immediately if submitted unpreserved, and a record made of the preservation methodology.
3. Sample Tracking: Samples should be assigned a unique number or code to identify the sample in a tracking system. The sample tracking system should identify the sample, the source, the date of receipt, analyses, due date, and any other pertinent information. A computerized laboratory information management system (LIMS) is recommended for tracking samples in laboratories processing large numbers of samples for a variety of clients.
4. Sample Storage: Samples should be stored in an assigned location in a refrigerator or sample storage area accessible only to authorized personnel. Samples should be refrigerated at 4°C, where applicable, and removed only for inspection, logging and analysis. The temperature of the refrigerator should be measured and recorded daily.

Top of Page

5.8.4 Use of Blanks and Duplicate Samples

The use of blanks and duplicate samples in the field and laboratory is an important component in a QC program.

Field blanks and field duplicates are essential throughout the execution of a field program involving the collection of water. Field QC samples are used to establish whether any errors are being introduced during the sampling process so that corrective action can be taken if necessary. Field QC samples are distinct from laboratory QC samples in that they measure sampling effects rather than laboratory effects.

Field blanks are used to check contamination from all potential sources of contamination of the sample. These include possible contamination of sample bottles, caps, preservatives, equipment, filter paper (if samples are to be filtered), atmospheric contamination, sampling techniques, and analysis. Field blanks are collected by obtaining blank water (i.e., deionized water) from the laboratory conducting the analyses, transporting the water to the field, and taking it through all sample collection, handling and processing steps that the test samples undergo (e.g., transfer to a sample container, preservation, and exposure to the environment). Field blanks are transported, stored and analyzed in the same manner as test samples (McQuaker 1999).

Duplicate samples should be taken to verify analytical results and equipment reliance. Field duplicates are used to evaluate homogeneity of the sample site and the ability of the sampling system to take the sample the same way every time. A field duplicate is a completely separate sample, not a split of a single sample into two bottles. Duplicate samples should be treated as blind samples, and are not identified to the laboratory.

The last type of QC sample is the trip blank, also referred to as travel or transport blanks. Trip blanks are used to check contamination from sample bottles, caps and preservatives during transport, storage and analysis. A sample bottle is filled in the laboratory with blank water (i.e., deionized water) and preserved in the same manner as the test samples (Fowlie et al. 2001). Trip blanks are transported to the field with regular sample bottles and submitted to the laboratory unopened, together with the test samples. They are opened at the time of analysis, and analyzed in the same manner as the samples (McQuaker 1999).

Field and trip blanks as well as duplicate field samples should be collected at a frequency of 5-10% of the total number of samples. Therefore, if a total of 10 water quality areas were being sampled, only one of each of the QC samples would be needed from each station. This proportion can be increased if necessary, to monitor errors due to sampling and matrix homogeneity. If field and trip QC samples are not used, any inaccuracy introduced due to sampling will go undetected or be inappropriately attributed to the analytical laboratory. The use of blanks and duplicate samples in the laboratory is further discussed in section 5.8.5. Table 5-2 summarizes recommended use of blanks and duplicate samples in the field and the laboratory, for larger sampling programs. For routine sampling, with one station from the exposure area and one from the reference area, it is recommended that a single field blank be submitted together with the test samples. In such cases, these samples will be analyzed by the laboratory as a batch, together with samples from other clients. The laboratory will achieve necessary internal QC using the complete batch.

^a Intrinsic to every batch of 20 samples
^b Used interchangeably with SRM if SRM is not available

Top of Page

5.8.5 Quality Control in the Laboratory

The following are general QC aspects of laboratory analyses performed:

• Data should be verified and validated through transcription checks; chemical data will be verified by reference to the analytical laboratory QA reports accompanying the data.
• Data analyses will be repeatable and robust and will be cross-checked with data quality objectives.
• Data analyses will be rigorous and defensible and should include the rationale for all statistical analyses and data transformations.

Top of Page

5.8.5.1 Details on Quality Control Aspects of Laboratory Analyses

Analytical QC procedures are designed to demonstrate statistical control over calibration, precision, accuracy/bias, and recovery (CALA 1991).

Statistical control over these parameters can be demonstrated by running specific QC samples during each analytical run. The results of these QC samples are compared statistically with confidence intervals calculated from historical data. These confidence intervals or control limits are normally calculated at 3 standard deviations (SDs) of the mean of the controlled variable. Warning limits are frequently set at 2 SDs. Indicators of a run considered out of control include the following:

• two successive results for method blanks, laboratory duplicates, standard reference materials, spiked blanks, calibration control samples, or organic surrogate recoveries;
• one of these results outside of the control limits.

QC data can be plotted on appropriate control charts. Control charts are graphic presentations of the QC data as a function of time or consecutive run number. Control charts demonstrate trends in time and provide graphic evidence of long-term statistical control of the analysis. Control limits and control charts are described in detail in ASTM (1986).

Top of Page

5.8.5.2 Good Laboratory Practices

Well-established good laboratory practices (GLPs) should be followed. The following is a brief listing of recommended laboratory practices (a description of GLPs can be found in greater detail in ELAP 1988):

• Records on reagent preparation should be maintained in a logbook. Prepared reagent containers should be labelled with the reagent, its date of preparation, the expiry date, and the person responsible.
• Instruments should be maintained or serviced on a regular basis. Maintenance records should be kept in a logbook.
• Written instructions should be available for all instruments.
• Standard procedures for cleaning glassware and containers should be followed.
• Routine checks of the purity of the distilled water should be conducted and documented. Distilled/deionized water should be checked on a conductivity meter at least daily.
• Chemical reagents should meet the purity requirements of each analytical method.
• Reagents and solvents should be stored according to the manufacturer’s directions.
• Working standards and stock solutions should be checked to determine changes in concentration.
• Reagents should be prepared and standardized against primary reference standards.
• The temperatures of all refrigerators and incubators should be checked daily and temperature excursions should be recorded.
• Each oven should have a dedicated thermometer and the temperature should be checked prior to and following each use.
• Proper volumetric glassware should be used.
• Glassware should be cleaned according to specifications of the method.
• Gas cylinders should be replaced at 700–1400 kilopascals (kPa).
• Laboratory personnel should have appropriate training in analytical laboratory procedures, and in the particular analysis for which they are responsible.

Top of Page

5.8.5.3 Calibration Control

Statistical control over calibration can be confirmed between runs by means of two control standards, A and B, and within-run by means of blanks and mid-range standards.

1. Between-run Calibration Control: Two control standards, A and B, can be used to analyze and control between-run changes in calibration, once at the beginning of each analytical run. These standards are made up and maintained independently of the calibration standards and are normally chosen to be about 80% and 20% of full scale, respectively. Results are accumulated over many runs and the sums (A + B) and differences (A - B) are plotted on control charts. During a specific run, a significant change in the sum (A + B) from the historical mean implies that a significant change in intercept has occurred, other factors remaining constant. A significant change in the difference (A - B) implies a significant change of slope, other factors remaining constant. Control and warning limits for A - B are calculated for the mean and the SD of the population of differences:
   
   
   • Upper and lower warning limits (UWL, LWL) = X_A-B ± 2 SD_A-B
     
   • Upper and lower control limits (UCL, LCL) = X_A-B ± 3 SD_A-B
     
   Control and warning limits for A + B are similarly calculated using the same SD:
   
   • UWL / LWL = X_A+B ± 2 SD_A-B
     
   • UCL / LCL = X_A+B ± 3 SD_A-B
     
   The run should not proceed until it is shown that A + B and A - B are within control limits. Control limits should not exceed ± 5% of the average value for A + B and A - B.
   
2. Within-run Calibration Control (Inorganic Analyses): Within-run changes in calibration attributable to slope and baseline drift should be checked at regular intervals. This can be accomplished by use of a mid-range standard and reagent blank run after every 20 samples. Control limits should be established by each laboratory for each procedure. The drift should not exceed 10%. If a greater drift is detected, the analysis should be stopped, the instrument recalibrated, and samples run after the last acceptable check sample and blank are reanalyzed.
   
3. Within-run Calibration Control (Organic Analyses): In organic analyses by gas chromatography (GC), within-run changes in calibration should be checked by injection of a mid-level check standard at a frequency of 5% or every 12 hours. This injection is compared to the initial calibration by calculating the percent deviation in the response factor of each analyte in the check standard to the average response factor determined during the initial calibration. If the relative percent difference is greater than 25%, the calibration check should be repeated. If the repeated check standard still has a relative percent deviation greater than 25%, corrective action is recommended.

Top of Page

5.8.5.4 Precision

Precision is the degree of variation among individual measurements of the same variable using a specific analytical method, and is usually expressed as the SD of replicates (US EPA 1990). Statistical control of analytical precision is maintained by analyzing within-run duplicates at a frequency of at least 10%. Laboratory duplicates are separate aliquots split in the laboratory from a single sample.

The absolute difference between within-run duplicates is compared to a control limit determined from historical data. To obtain these control limits, the results of duplicate analyses are accumulated over many runs and sorted according to concentration ranges.

Convenient concentration ranges are 0-20%, 20-50%, and 50-100% of full scale (King 1976). Within each concentration range, control limits for the absolute difference between within-run duplicates is determined from the formula:

UCL = D₄ x R

where D₄ (3.267) is a statistical factor and R is the mean difference between duplicates (ASTM 1986; Taylor 1987).

If the difference between laboratory duplicate analyses exceeds the upper control limit, the situation should be evaluated to determine the most appropriate corrective action.

Top of Page

5.8.5.5 Accuracy and Bias

Accuracy is the degree of agreement between an observed value and the true value as determined by analysis of an accepted reference material (US EPA 1990). The converse of accuracy is the degree of systematic error in the analysis, i.e., the bias. Accuracy is controlled by means of method blanks and certified reference materials. Information on recommended quality control for inorganic analyses can be found in CALA (1991).

1. Method Blanks: A method blank is an aliquot of reagent water equivalent in volume to the samples being processed and run in exactly the same manner as the samples. The method blank quantifies the level of contamination introduced to the samples during sample processing and analysis. Method blanks should be analyzed at a frequency of 10% or 1/run, charted, and controlled at ± 2 SD (warning limits) and ± 3 SD (control limits). If a method blank is judged out of control and contaminated, those samples processed with the blanks and greater than the detection limit should be repeated for the variable(s) affected. In general, a method blank is considered free of contamination if the analysis yields results less than the detection limit or less than 0.1 times the level found in all associated samples (CALA 1991).
   
2. Standard Reference Materials: SRMs are samples available in different matrices that have been extensively analyzed by several laboratories and have concentrations certified by standard-setting organizations such as the National Institute of Science and Technology, the U.S. EPA, the National Water Research Institute of Environment Canada and the National Research Council. When available, an SRM should be analyzed at a frequency of 5% or 1/run (CALA 1991; King 1976). The matrix and concentration of the SRM should be as close as possible to the samples being analyzed. The results of SRMs should be accumulated, and control and warning limits determined as ± 3 SD and ± 2 SD, respectively.

Top of Page

5.8.5.6 Recovery

Recovery of the analyte over the entire analytical process is determined from matrix spikes, spiked blanks, and surrogate spikes.

1. Matrix Spike: A matrix spike is a separate aliquot of a randomly chosen sample to which is added all the analytes of interest before processing of the sample. Analysis of a matrix spike gives an indication of the recovery efficiency obtained for the matrix particular to that sample. The sample should be spiked with all the analytes of interest at a concentration as close as possible to that concentration, giving a response equal to the mid-level calibration standard. The spiking solution should be prepared from a stock source separate from that used for calibration. The recommended distribution of matrix spikes is 10% or 1/run. One method to calculate recovery is:
   
   
   
   The results of matrix spikes should be plotted on separate control charts for each matrix. In-house limits should be set on the basis of ± 3 SD on a minimum of 10 data points. In multi-parameter analyses, at least 90% of the analytes should have recoveries within the specified limits. Recoveries for inorganic analytes should fall within 75-125%. Recoveries for organic variables should fall within the limits specified in Table 4 of CALA (1991). If a matrix spike does not meet these criteria, the spike should be repeated. If the recoveries do not meet the criteria in the repeat analysis and there are no indications of other problems with the analysis, a matrix effect should be noted and reported.
   
2. Spiked Method Blank: The spiked method blank is a separate aliquot of the same reagent water used for the method blank that is spiked with the compound of interest at a concentration as close as possible to the concentration of the mid-level calibration standard. The spiked method blank gives an indication of the reliability of a method without the matrix effects of real samples. The spiked method blank should be processed with and in the same manner as the samples. As with the matrix spike, the spiking solution should be prepared from stocks separate from those used for calibration.
   
   In-house recovery limits should be calculated for the spiked method blank based on ± 3 SD and a minimum of 10 data points. Recoveries for inorganic analyses should fall within 75-125%. Recoveries for organic variables should fall within 70-120%. If a spiked blank recovery does not meet the criteria established, the spike should be repeated. If the spike still does not recover, the samples related to the spike should be repeated. If insufficient sample remains for a repeat analysis, the results should be reported and flagged as suspect with an explanation.
   
3. Internal Standards (Organic Analyses): All analyses using GC should be performed using internal standards, or properly validated methods using external standards. An internal standard is a compound that behaves similarly in an analytical system as the compound of interest, but is unlikely to be found in the sample. Internal standards are added at the same level to all samples, standards, and control samples prior to measurement but after sample preparation. All analyte responses should be normalized for the internal standard response to correct for instrument variability in response to such factors as varying injection volumes, temperature fluctuations, and final extract volume. The response of the internal standard in the sample measurement should be within 20% of the internal response of a calibration standard analyzed within the same 12-hour period. If this criterion is not met, the sample should be repeated. If upon reanalysis the criterion is still not met, the sample results should not be corrected for internal standard response and should be flagged with an explanation.
   
4. Surrogate Spikes (Organic Analyses): A surrogate standard is a compound not expected to be found in the sample that behaves similarly to the analytes of interest during sample preparation and analysis. Where applicable, surrogates should be added to all samples (including QC samples) before sample preparation to indicate method performance and sample matrix effects. Analyses run by gas chromatography / mass spectometry (GC/MS) should have at least two surrogates, while those run by GC should have at least one surrogate. The amount of surrogate added to all samples should be the same as that added to the calibration solutions. In-house control limits for surrogate recoveries are based on ±3 SD on a minimum of 10 data points. In-house control limits for surrogate recoveries should be within 60-120%. If any surrogate is outside the expected recovery range, the sample should be reanalyzed. If, upon reanalysis, the surrogate recovery is still outside the permissible range, the results should be reported with a flag and an explanation.

Top of Page

5.8.5.7 Detection Limits

Detection limits should be reported as the method detection limit (MDL) as described by the U.S. EPA (1984). The MDL is defined as the minimum quantity of an analyte that should be observed to justify the claim to have detected the analyte with a specified risk (normally 5% or 1%) of making a false detection.

One method to calculate the MDL is from the SD of the analysis at the lowest concentration range:

MDL = t_{0.05 n-1} x S

where: t_{0.05, n-1} is the one tailed value of Student’s t for a 5% risk of false detection, n-1 degrees of freedom, and S is the SD.

Ideally the SD is calculated from low-level replicate analysis on real samples having the same or similar sample matrix as the samples under consideration. This SD can be calculated from a minimum of seven replicates in the same run using the standard statistical formula (US EPA 1984). However, it is preferable to calculate S from between-within-run replicate pairs accumulated over many runs.

The SD of low-level replicate pairs accumulated over a large number of analytical runs is:

where D is the individual replicate difference and n is the number of replicate pairs. A minimum of 40 replicate pairs is recommended (OMOE 1988). The value of either SD is then entered in the equation to calculate the MDL.

Values below the detection limit should be reported as < MDL, with the applicable MDL for that sample (Fowlie et al. 2001). There are three common approaches to deal with values that are < MDL when analyzing data: set the value at the MDL, half the MDL, or 0. For the purposes of the EEM program, half the MDL is currently used for all data analysis and interpretation. For additional information on how to interpret non-detectable data, refer to Helsel (2005a, 2005b) and Shumway et al. (2002).

Top of Page

5.8.5.8 Data Reporting Conventions

Established protocols for rounding off analytical results should be followed. If too many figures are rounded off before reporting, information is lost and real differences in the concentrations of samples from different locations or occasions may be concealed. QC may be on a coarser basis than is desirable, or necessary, with the result that values of the mean, SD or other statistics of a set of results may be biased. Conversely, when too many significant figures are reported, relatively small, statistically insignificant differences may appear falsely large (Hunt and Wilson 1986).

The SD of the analysis is the preferred criterion for deciding the number of significant figures (King 1989). The process of rounding off should ensure retention of the digit that is in the same decimal position as the most significant digit in the calculated SD. For example, if the analysis provides a value such as 12.345 and the calculated SD based on within-run replicate analysis at this concentration level is 0.32, the result should be truncated to 12.3.

Top of Page

5.8.5.9 Analytical Precision and Accuracy

Precision is the degree of agreement among replicate analysis of a sample, usually expressed as the SD. Reproducibility is the closeness of agreement between the results of measurement of the same parameter carried out under changed conditions of measurement. Reproducibility is the SD obtained measuring the same sample in different analytical runs and is called between-run precision. Between-run precision includes variability due to calibration on different days, instrument drift and many other factors.

Precision is affected by random errors and is a measurable and controllable parameter. Precision should be estimated for all analyses by processing separate sample aliquots through the entire analytical method. A laboratory should monitor their precision and be able to report precision using several days of data. For most parameters, the precision should be within 10%. For total suspended solids, the precision should be within 15% at concentrations greater than 10 times the MDL. For pH, precision should be within ± 0.1 pH unit (MMER, Schedule 3).

Accuracy is the combination of bias and precision of an analytical method, which reflects the closeness of a measured value to the true value of a sample. Bias is a systematic error caused by something in the measuring system resulting in the data being high or low. Bias can be caused by a number of factors including contamination, mechanical losses, blanks, spectral interference, calibration errors or the influence of different operators. Accuracy is measured as percent recovery of known concentrations such as certified reference materials, spiked samples or reference samples prepared by the laboratory and analyzed as samples.

Whether data are considered accurate or inaccurate is relative to the final use of the data. A laboratory should monitor their accuracy and be able to report this using several days of data. For metals and most other parameters, accuracy should be within 10%. For total suspended solids, accuracy should be within 15% at concentrations greater than 10 times the MDL. For pH, accuracy should be within ± 0.1 pH unit (MMER, Schedule 3).

Top of Page

5.8.6 Quality Assurance in the Laboratory

QA encompasses a wide range of internal and external management and technical practices designed to ensure that data of known quality are commensurate with the intended use of the data.

External QA activities include participation in relevant inter-laboratory comparisons and audits by outside agencies. Outside audits may be based on performance in analysis of standard reference materials, or on general review of practices as indicated by documentation of sampling, analytical and QA/QC procedures, test results, and supporting data.

Top of Page

5.8.7 Recording and Reporting of QA/QC Information

5.8.7.1 Documentation

Documentation of all aspects of the analysis is recommended to confirm the quality and reliability of the analytical results. The owner or operator of a mine shall keep all records, books of account or other documents required by the Metal Mining Effluent Regulations at the mine’s location for a period of no less than five years (MMER, s. 27). For each sample or batch of samples, information on the following is recommended:

1. Method Detection Limits: If MDLs are different from the laboratory-determined MDLs (due to interference, dilutions, etc.), this should be recorded.
   
2. Sample Storage Times: Records should be kept on the sampling date, date of receipt, date of sample preparation, and date of analysis. This information is normally handled as part of the sample-tracking process.
   
3. Instrument Performance and Maintenance: A log should be kept of instrument performance, including records of tuning and instrument response. Maintenance or service records should be kept for each instrument.
   
4. Quality Control Samples: Records of duplicate analyses, blanks, spiked blank recoveries, surrogate recoveries, matrix spike recoveries and results from certified reference materials, and records of calibration and calibration checks should be maintained.
   
5. Sample Reception, Preparation and Analysis: All anomalies in delivery, storage, condition, preparation and analysis of samples should be recorded. These include any deviations from standard operating procedures.

Top of Page

5.8.7.2 Reporting of QA/QC Information

Analytical results are reported as a test or analysis report and should include all relevant data needed to assess the validity of the data, including QA/QC components. The report should be accurate, clear, unambiguous and objective. Items that should appear in the report include:

• a title (Test Report, Report of Analysis, Quality Report);
• name, address and location of laboratory, and location where tested;
• unique identification of the report so it can be traced easily (serial number, group number);
• name and address of client;
• identification or description of the sample tested;
• condition of the test item (unpreserved, leaking bottle--where relevant);
• date of sample receipt, date of report;
• identification of the analysis method and description of any non-standard tests;
• reference to sample date and sampling method (grab sample, time-proportioned composite sample, etc.);
• deviations from the usual test method (filtering, pH, adjustment, standard addition, etc.);
• the analytical results with units clearly identified;
• statement indicating whether the results were corrected for blanks;
• QC data;
• identify if result is qualified (did not pass QC tests, sample size too small, etc.);
• signature of accountable person and date authorization;
• name of technician who completed the test;
• subcontractors clearly identified;
• updates or corrections to reports clearly identified; and
• the laboratory should notify clients if new information invalidates reports already issued.

Data below the analytical detection limit should be clearly reported as such along with the applicable MDL for that sample.

5.9 References

AQUAMIN. 1996. Assessment of the Aquatic Effects of Mining in Canada: Final report. Prepared by AQUAMIN Working Groups 7 and 8 for the AQUAMIN Steering Group.

[ASTM] American Society for Testing and Materials. 1986. Manual on presentation of data and control chart analysis. Committee E-11 on Statistical Methods. ASTM Special Technical Publication 15D. Philadelphia (PA): American Society for Testing and Materials.

[CALA] Canadian Association of Laboratory Accreditation. 1991. Code of practice and QA manual for laboratory analysis of sewage treatment effluent in support of the MISA program. Draft report prepared for CAEAL and the Ontario Ministry of the Environment by Zenon Environmental Laboratories.

[CCME] Canadian Council of Ministers of the Environment. 1999. Canadian environmental quality guidelines. Chapter 4: Canadian water quality guidelines for the protection of aquatic life. Hull (QC): Canadian Council of Ministers of the Environment. Available from: http://ceqg-rcqe.ccme.ca/

Druschel GK, Schoonen MAA, Nordstorm DK, Ball JW, Xu Y, Cohn CA. 2003. Sulfur geochemistry of hydrothermal waters in Yellowstone National Park, Wyoming, USA. III. An anion-exchange resin technique for sampling and preservation of sulfoxyanions in natural waters. Geochem Trans4:12-19.

[ELAP] Environmental Laboratory Approval Program. 1988. Environmental Laboratory Approval Program certification manual. New York State Department of Health.

[ESG] Ecological Services Group. 1999. AETE synthesis report of selected technologies for cost-effective environmental monitoring of mine effluent impacts in Canada (AETE Project No. 4.1.4). Report for AETE program. Ottawa (ON): CANMET, Natural Resources Canada.

Fowlie P, Hart DR, Turle R. 2001. Guidance Document for the Sampling and Analysis of Mine Effluents: Final Report. Ottawa (ON): Environment Canada, Environmental Protection Service, Minerals and Metals Division. Report EPS2/MM/5. Available from: http://dsp-psd.pwgsc.gc.ca/Collection/En49-24-1-39E.pdf

Helsel DR. 2005a. Nondetects and data analysis: statistics for censored environmental data. Hoboken (NJ): Wiley-Interscience. 250 p.

Helsel DR. 2005b. More than obvious: better methods for interpreting nondetect data. Environ Sci Technol 39(20):419A-423A.

Hunt DTE, Wilson AL. 1986. The chemical analysis of water; general principles and techniques. 2nd edition. London (UK): The Royal Society of Chemistry.

[ISO/IEC] International Organization for Standardization. 2005. ISO/IEC 17025: 2005. General requirements for the competence of testing and calibration laboratories. Geneva (CH): ISO/IEC.

King DE. 1976. Quality control and data evaluation procedures. Section I. Analytical responsibility. Special report to Laboratory Services Branch, Ontario Ministry of the Environment.

King DE. 1989. Code of practice for environmental laboratories. Special report to the Ontario Ministry of the Environment. ISBN 0-7729-5874-2.

Makhija R, Hitchen A. 1979. The titrimetric determination of sulphate, thiosulphate and polythionates in mining effluents. Anal Chim Acta 105(1):375-382.

McQuaker NR. 1999. Technical evaluation on water quality design and analysis (AETE Project No. 3.1.1). Draft report for AETE program. Ottawa (ON): CANMET, Natural Resources Canada.

Nielson LA, Johnson DL. 1983. Fisheries techniques. Bethesda (MD): American Fisheries Society. 468 p.

[OMOE] Ontario Ministry of the Environment. 1988. Estimation of analytical detection limits (MDL). Report by the Ontario Ministry of the Environment. ISBN-0-7729-4117-3.

Ontario Ministry of the Environment and Energy. 1993. MISA draft development document for the Effluent Limits Regulation for the Metal Mining Sector. Toronto (ON): Queen’s Printer for Ontario.

O’Reilly JW, Dicinoski GW, Shaw MJ, Haddad PR. 2001. Chromatographic and electrophoretic separation of inorganic sulfur and sulfur–oxygen species. Anal Chim Acta 432(2):165-192.

Reiley M. 2007. Science, policy, and trends of metals risk assessment at EPA: How understanding metals bioavailability has changed metals risk assessment at US EPA. Aquat Toxicol 84(2):292-298.

Schwartz M, Vigneault B, McGeer J. 2006. Assessing the potential toxicity of thiosalts in the context of the Metal Mining Effluent Regulation. Presentation made at the 33rd Aquatic Toxicity Workshop, Jasper, AB.

Shumway RH, Azari RS, Kayhanian M. 2002. Statistical approaches to estimating mean water quality concentrations with detection limits. Environ Sci Technol 36(15):3345-3353.

Taylor JK. 1987. Quality assurance of chemical measurements. Chelsea (MI): Lewis Publishers Inc.

[US EPA] United States Environmental Protection Agency. 1984. Definition and procedure for the determination of the method detection limit – Revision 1.11. Appendix B to Part 136. Federal Register. Vol. 49, no. 209. Oct. 26, 1984, Part VI, 40 CFR Part 136.

[US EPA] United States Environmental Protection Agency. 1990. Proposed glossary of quality assurance related terms. QAMS RD-680. Draft report.

[US EPA] United States Environmental Protection Agency. 2007. Aquatic life ambient freshwater quality criteria--Copper 2007 revision. EPA-822-F-07-001.

Vigneault B, Holdner J, Bélanger J. 2002. Validation of an anion exchange method for the preservation and analysis of thiosalt speciation in mining waste waters. Ottawa (ON): CANMET Mining and Mineral Science Laboratories. Division Report MMSL 03-002(TR).

Appendix 5-1: Justifications for Parameters for Effluent Characterization and Water Quality Monitoring

(text description)

¹ Note: the MISA draft development document is referred to throughout this document. The MISA document summarizes data from one year of comprehensive effluent characterization at mines across Ontario in the early 1990s. These data are not representative of the current state of effluent quality from mines across Canada. Reference: Ontario Ministry of the Environment and Energy. 1993. MISA Draft Development Document for the Effluent Limits Regulation for the Metal Mining Sector. Queen’s Printer for Ontario, Toronto, Canada.

² Note: throughout this document, the Canadian Environmental Quality Guidelines (CEQG) are referred to in order to provide a preliminary assessment of concentrations in the effluent for their ecological significance and their potential effects on the receiving environment. Reference: Canadian Council of Ministers of the Environment. 1999. Updated 2001. Canadian Water Quality Guidelines for the Protection of Aquatic Life, available from: http://ceqg-rcqe.ccme.ca/.

Tables 

Table 5-1 outlines the analytical parameters measured for effluent characterization and water quality monitoring. Effluent quality variables are aligned with water quality variables and site-specific variables.

Return to the table

Table 5-2 provides a summary of the recommended use of blanks and duplicate samples in the field and laboratory. Each parameter is aligned with the number of samples, internal or field quality control, control limits, and a description is provided.

Return to the table

Appendix Table 5-1 outlines the justifications for parameters for effluent characterization and water quality monitoring. Topics include deleterious substances and pH as per Schedule 3 of the MMER; parameters required in Schedule 5 of the MMER for effluent characterization and water quality monitoring; required parameters for water quality monitoring only; additional site-specific recommended parameters for effluent characterization and water quality monitoring; parameters recommended for effluent characterization only; and site-specific parameters recommended for water quality monitoring only.

Return to the table

6. Sublethal Toxicity Testing

6.1 Overview

6.2 Collection of Samples

6.3 Sampling Locations

6.4 Frequency and Reporting

6.5 Tabulation of Sublethal Toxicity Endpoints and Validation of Test Results

• 6.5.1 Validation of Test Results
• 6.5.2 Tabulation of Sublethal Toxicity Endpoints

6.6 Data Interpretation in Relation to the Toxicity Objectives

• 6.6.1 Changes in Effluent Quality
• 6.6.2 Understanding Multiple Discharge Situations
• 6.6.3 Contributions to the Weight-of-Evidence Approach
• 6.6.4 Considerations for Integration of Toxicity Test Results

6.7 Description of Freshwater and Marine Sublethal Toxicity Tests

6.8 Dilution Water in Freshwater Sublethal Toxicity Testing

• 6.8.1 Dilution Water Selection

6.9 Collection, Shipment and Storage of Samples for Sublethal Toxicity Testing

• 6.9.1 Test Organism Acclimation
• 6.9.2 Screening Tests
  • 6.9.2.1 Screening Test Impairment
• 6.9.3 Effluent or Effluent-Exposed Surface Water Toxicity Tests

6.10 Use of Sublethal Toxicity Testing in Resolving Confounding Influences

6.11 References

List of Tables

• Table 6-1: Methodologies for effluent sublethal toxicity tests
• Table 6-2: Descriptions of the freshwater and marine sublethal toxicity tests included in the metal mining EEM program
• Table 6-3: Specifications of the Environment Canada test methods and recommendations for collection, storage and use of site collected dilution waters
• Table 6-4: Dilution/control water and corresponding effluent volumes for sublethal toxicity tests
• Table 6-5: Preparation of different hardness/alkalinities

6. Sublethal Toxicity Testing

6.1 Overview

There are two main uses for sublethal toxicity tests in the environmental effects monitoring (EEM) program: to compare processes and measure changes in effluent quality; and to contribute to the understanding of the relative contributions of the mine in multiple-discharge situations.

The purpose of sublethal toxicity testing in the metal mining EEM program is to provide an estimate of the potential effects on biological components (phytoplankton, zooplankton, benthic invertebrates, fish, macrophytes) in the exposure area, whether or not these components are being directly measured in the field.

To estimate the potential effects on biological components, sublethal toxicity testing shall be conducted by following the applicable methods referred to in the Metal Mining Effluent Regulations (MMER), section (s.) 5, subsections (ss.) 3 and 4 (as amended from time to time). Four freshwater sublethal toxicity tests (fish, invertebrate, algal and plant species) or three marine or estuarine sublethal toxicity tests (fish, invertebrate and algal species) shall be conducted, depending on the receiving environment type (MMER, Schedule 5, s. 5), and the results shall be recorded. The test chosen should primarily be based on the relevance of the species to the local receiving environment, and secondarily on the seasonal availability of test organisms.

Acceptable sublethal toxicity methods are outlined in Table 6-1. The following website contains all the biological test method documents published by Environment Canada’s Biological Methods Section: http://www.ec.gc.ca/faunescience-wildlifescience/default.asp?lang=En&n=0BB80E7B-1. Test report checklists have been developed for assessing the validity of test results for each test option, which are available on the EEM website (www.ec.gc.ca/esee-eem/default.asp?lang=En&n=D450E00E-1). Information on the relative sensitivity of the different sublethal toxicity tests can be found in ESG (1999).

For additional information on assessing changes in effluent over time, and other data-interpretation situations, see sections 6.6 and 6.10.

¹ Where Fathead Minnows are not an indigenous species, Rainbow Trout will be used according to Environment Canada (1998).
² In some jurisdictions, both Environment Canada (2007b) and the MDDEP (2007) testing requirements for Pseudokirchneriella subcapitata are acceptable for the EEM program.
Note: For all marine toxicity test procedures, it is recommended that the effluent salinity adjustment procedure by Environment Canada (2001) be followed. For all sublethal tests where the test organisms are purchased for sublethal toxicity testing, it is recommended that the test organism importation of Environment Canada (1999) be followed.

Top of Page

6.2 Collection of Samples

Sublethal toxicity testing shall be conducted on the aliquots of effluent samples collected in accordance with ss. 4(2) of the MMER (Effluent Characterization) (MMER, Schedule 5, ss. 5(2)).

In choosing when effluent for toxicity tests should be collected, two aspects should be considered:

1. when effluent poses the greatest potential for adverse environmental impact on the environment; and
2. when biological monitoring is conducted to look at potential linkages with effects in the exposure area.

Top of Page

6.3 Sampling Locations

To determine which outfall structure has potentially the most adverse environmental impact, the following should be taken into account:

• the mass loading of deleterious substances;
• the manner in which the effluent mixes in the exposure area; and
• historical characterization or sublethal toxicity data.

In cases where it is not clear which discharge source has the greatest potential to affect the environment, mines may wish to use a series of single-concentration sublethal toxicity tests from each final discharge location to determine the source with the greatest sublethal response.

To estimate the potency of the response from each discharge source, the “time to response” can be observed and calculated as the sublethal toxicity test endpoint (e.g., Ceriodaphnia dubia adults are exposed to undiluted effluent samples from each different effluent discharge, and observations are made as to how long it takes to find a 25% or 50% response). The sublethal toxicity test endpoint would be an LT₂₅ (time to 25% mortality) or LT₅₀ (time to 50% mortality) if survival was the key observation. The single-concentration test would be the more cost-effective approach to screening effluent sources in order to determine the discharge point with the greatest potential to affect the receiving environment.

Top of Page

6.4 Frequency and Reporting

Sublethal toxicity tests shall be conducted two times each calendar year for the first three years and once each year after the third year (MMER, Schedule 5, ss. 6(2)). The first effluent sample shall be collected, and sublethal toxicity testingconducted, not later than six months after the mine is subject to section 7 of the MMER (Schedule 5, ss. 6(1)).

The results of sublethal toxicity testing shall be submitted to the Authorization Officer as part of the Effluent and Water Quality Monitoring Report. Additional information on the content of the Effluent and Water Quality Monitoring report is discussed in Chapter 5. The reporting requirements of the sublethal toxicity results are described in the MMER (s. 23; and Schedule 5, s. 8). See Chapter 10 for information on electronic reporting of sublethal toxicity data.

The test methods in Table 6-1 can be referred to for reporting specifications for each test method.

The report should include the following:

• dates when the samples were collected for sublethal toxicity testing;
• the location of the final effluent discharge point from which samples were collected for sublethal toxicity testing, and data on how this point was chosen;
• the results of sublethal toxicity testing, including the median lethal concentration (LC₅₀), 25% inhibition concentration (IC₂₅¹) and 25% effect concentration (EC₂₅) where applicable, 95% confidence limits, and indication of quantitative statistics employed;
• a description of the quality assurance / quality control (QA/QC) measures that were implemented, and the data related to the implementation of those measures; and
• minimum reporting outlined in the test methods and sublethal toxicity checklists.

In determining whether or not to use historical sublethal toxicity data as part of the EEM program, the mine should take the following factors into consideration:

• laboratory QA using the methods listed in Table 6-1;
• no fewer than three species tested (adequate to answer the question);
• age of the data (testing conducted after December 31, 1997);
• the nature of mine operating conditions (e.g., are mine operations similar to the operations in place at the time the sublethal toxicity tests were conducted?); and
• whether any of the sublethal toxicity testing was conducted on an effluent sample taken concurrent with fish and benthic invertebrate field monitoring.

Toxicity data submitted as part of the EEM program for the metal mining industry should be accompanied by a description of the materials and methods, and calculations for each test. Minimum reporting requirements are detailed in section 8 or 9 of the toxicity test method documents of Environment Canada. Test report checklists have been developed for assessing the validity of test results for each test option, which are available on the EEM website (www.ec.gc.ca/esee-eem/default.asp?lang=En&n=D450E00E-1). The minimum reporting requirements for methods of the U.S. Environmental Protection Agency (EPA) (US EPA 2002) have been prescribed for the purpose of subsequent EEM phases. The U.S. EPA requirements generally conform to Environment Canada specifications.

Top of Page

6.5 Tabulation of Sublethal Toxicity Endpoints and Validation of Test Results

Sublethal toxicity endpoints reported vary depending on the test being conducted (refer to test methods in Table 6-1). However, the IC₂₅ will be discussed below for illustrative purposes. The geometric mean² of all IC₂₅s (GM-IC₂₅) for a given species should be calculated for each phase.

Top of Page

6.5.1 Validation of Test Results

Procedures for QA/QC will be followed by both the field crews collecting environmental samples and the laboratory carrying out the toxicity testing, as discussed in the required toxicity test method documents.

Therefore, a first step in the interpretation of toxicity data for EEM should be the resolution of any problems with QA/QC. In addition to the QA outlined in the individual sublethal toxicity test methods, further requirements and recommendations are as follows:

• reference toxicant test conducted in the same manner as the effluent or effluent-exposed surface water test;
• reference toxicant test conducted within ~ 30 days of the effluent or effluent-exposed surface water test;
• test-specific validity criteria met in all effluent sublethal testing conducted;
• sublethal toxicity testing initiated within 3 days of sample collection;
• quantitative sublethal toxicity endpoints provided for all sublethal toxicity tests conducted on effluent or effluent-exposed samples;
• sublethal toxicity test endpoint between 0.1 and 100% bracketed by at least one test concentration;
• sublethal toxicity tests that fail to meet test method validity criteria repeated on a new sample; and
• reporting of “less than” values as a sublethal toxicity test endpoint will no longer be acceptable.

Data could be declared rejected if one or more essential elements of the test method were not followed (e.g., failure to meet organism health criteria, inappropriate manipulations of the sample, failure to conduct the minimum in-test monitoring, incorrect statistic used for sublethal toxicity endpoint calculation).

Laboratories contracted by the metal mining industry to conduct sublethal toxicity testing should be accredited under the International Organization for Standardization standard ISO/IEC 17025:2005 entitled “General requirements for the competence of testing and calibration laboratories,” as amended from time to time.

Top of Page

6.5.2 Tabulation of Sublethal Toxicity Endpoints

If the effluent does not cause a 25% sublethal inhibition or effect for any of the freshwater sublethal tests, an IC₂₅/EC₂₅ cannot be calculated and it is reported as &gt; 100%.

Mortality in some of the concentrations might also prevent calculation of the IC₂₅/EC₂₅. For example, there might be no measured effects (mortality, growth or reproduction) in 32% concentration, but appreciable mortality in 56%. It would then be impossible to obtain a good estimate of the inhibition of growth or reproduction for the 56% concentration, and hence impossible to determine IC₂₅/EC₂₅. In such a situation, the IC₂₅/EC₂₅ should be assumed to be equal to the higher concentration, which in this case is 56%.

There should be IC₂₅s reported for each of the test species. This information is summarized by calculating the geometric mean for each set of IC₂₅s. For example, testing of a mine effluent on six occasions resulting in measured IC₂₅ values of 10, 15, 17, 23, 25 and 30% would lead to a geometric mean of 19%.

Top of Page

6.6 Data Interpretation in Relation to the Toxicity Objectives

6.6.1 Changes in Effluent Quality

The geometric mean of the IC₂₅ (GM-IC₂₅) for each species can be compared between phases to assess changes in the quality of effluent over time at each mine or between mines and mine types. Improvements are expected as a result of changes in process or effluent treatment.

Top of Page

6.6.2 Understanding Multiple Discharge Situations

Comparison of mine effluent toxicity data from other nearby industrial and/or municipal discharges can help in understanding the relative contribution to the potential impact of each effluent source on the environment. Provided that relevant toxicity, flow and dispersion information is available for the other discharges, the inter-relation or overlap of the effluent fields may be better understood. Another method of comparing relative loading (toxic contribution) from each source is to calculate the TER (toxicity emission rate; = (100/GM-IC₂₅) x flow). However, this calculation does not relate to the receiving environment, because the effluent dispersion and dilution are not taken into consideration. See section 6.10 for more information on confounding influences.

Top of Page

6.6.3 Contributions to the Weight-of-Evidence Approach

Where sublethal toxicity data have an IC₂₅ of less than 30%, it is recommended that mines calculate the geographic extent of the response in the exposure area and identify the zone where the concentration of effluent is comparable to the IC₂₅. Data on effluent and receiving flow for the appropriate month are needed to complete this estimate. The estimation of the potential geographic extent can be effectively reported in map form and reported in the Interpretative Report.

A potential-effects zone may be interpreted as a rough indication of the extent of 25% inhibition by effluent in the environment. If, for example, a mine’s GM-IC₂₅ for a particular test was 1% volume/volume (v/v), this would match the extent of the 1% zone. Invertebrates and plant IC₂₅s are not expected to be similar, due to differing species sensitivities and test method sublethal toxicity endpoints, and therefore may have dissimilar potential-effects zones.

Top of Page

6.6.4 Considerations for Integration of Toxicity Test Results

The following points should be considered when toxicity data are used to estimate a potential-effects zone.

(1) Laboratory results: Sublethal laboratory tests provide estimates of toxicity under strictly controlled laboratory conditions for each test species. These conditions do not replicate environmental conditions at the site under study. Chapman (2000) describes various abiotic and biotic modifying factors present in the uncontrolled receiving environment, which may affect an organism’s response to a toxicant.

(2) Species differences: Species differences in sensitivity to metal mining effluents will be taken into account when extrapolating results from laboratory sublethal toxicity tests to effects on indigenous biota.

(3) Background toxicity: The description above assumed that there were no other upstream contributions of toxicity. That assumption would be erroneous if there were overlapping plumes.

(4) Type of receiving water: Receiving-water pH, hardness, dissolved organic carbon (DOC) and other modifying factors could potentially increase or decrease toxicity of the effluent compared to tests with laboratory water.

(5) Plume uncertainties: Calculations of dilution might be difficult or inaccurate, or the position of the mixing zone might be variable. Where this uncertainty exists, estimating a zone of potential effect would have an equal level of uncertainty.

6.7 Description of Freshwater and Marine Sublethal Toxicity Tests

Table 6-2 provides short descriptions for the freshwater and marine sublethal toxicity tests included in the metal mining EEM program. Information on the relative sensitivity of the different sublethal toxicity tests can be found in ESG (1999).

For freshwater tests, laboratory or site water can be used as dilution/control water. For marine or estuarine environments, the mine has a choice of using uncontaminated sea water or artificial sea water produced from hyper-saline brine (HSB). The recommended procedures for adjusting the salinity of the effluent and dilution water and preparing the HSB are described in Environment Canada (2001).

Where applicable, the test organism importation methodology (Environment Canada 1999) should be referred to, where test organisms are purchased for immediate use in sublethal toxicity tests.

For QA/QC, detailed records of all aspects of the samples, test organisms, culture maintenance, test conditions, equipment and test results are validated and kept by the laboratory. A reference toxicant test is used to establish the validity of effluent toxicity data. Successive reference toxicant data are plotted on a control chart. If results are within expected limits, the performance of the batch of test organisms is ensured. The minimum level of reporting is outlined in each test method.

Technical personnel should be skilled in algae, macrophyte, invertebrate and fish culture, and in conducting toxicity tests following aseptic techniques.

For more detailed descriptions, please refer to the specific test method documents.

6.8 Dilution Water in Freshwater Sublethal Toxicity Testing

6.8.1 Dilution Water Selection

The sublethal toxicity test methods required for the metal mining EEM program clearly define the culture conditions and test procedures that need to be followed (Environment Canada 1992a, 1992b, 1998, 2007a, 2007b, 2007c; US EPA 1994a, 1994b, 1995, 2002). Some testing decisions are left to the discretion of the individual laboratories, as long as the standard test acceptability criteria can be achieved. For example, standard methods for testing Ceriodaphnia, Fathead Minnows, Pseudokirchneriellaand Rainbow Trout allow the use of uncontaminated ground or surface water, dechlorinated tap water or reconstituted water as a source for the culture or the test control/dilution water, as long as the water of choice supports a healthy culture and provides a valid test result.

Most laboratories in Canada use “standard laboratory” water for routine culturing and testing requirements. This water is generally supplied to the laboratory through a natural groundwater system (well) or a local municipal water source, which must be dechlorinated and may be buffered to meet acceptable culturing criteria. Deionized water reconstituted to targeted water quality parameters is also used. Advantages of using laboratory water include the following:

• It can be maintained at a consistent quality with minimal risk of contamination by undesirable and/or harmful chemicals or biota.
• Regular monitoring of water chemistry and culture health, as well as reference toxicant testing, ensure that the water is of acceptable quality for toxicity testing.
• Since cultures are maintained in laboratory water, no additional acclimation is needed for testing effluents or chemicals when laboratory water is used as the control/dilution water.
• Laboratory water, normally used in regulatory testing across Canada, provides a measure of the inherent toxicity of the effluents and allows comparison of effluent quality over time.

During the metal mining EEM program, most sublethal toxicity tests will likely be performed using laboratory water as control/dilution water, in order to attain comparable results among different laboratories and over time. It is also likely that in many sublethal toxicity tests where there is measurable effluent toxicity, this toxicity can be attributed to inorganic substances such as metals and ammonia, and the toxicity of the effluent may also be affected by site-specific characteristics such as pH, alkalinity and hardness. These characteristics can be controlled and reproduced in the laboratory for cases in which test results, reflective of the site conditions, are desired. However, a mine may decide to test its effluent using unexposed surface water (as control and dilution water), providing the sample is not exposed to effluent. Alternatively, a reference area of similar physicochemical characteristics to a mine site could be used to supply control/dilution water.

The use of unexposed surface waters can be especially helpful in obtaining the following information.

Estimating the mitigating or stimulatory effect of unexposed surface site water as dilution water on theexpression of toxicity from the effluent discharge or effluent-exposed surface water

Although parallel testing of effluents and effluent-exposed surface waters using site water and hardness-adjusted laboratory water have produced similar results (BEAK 1998, 1999), it is impossible to simulate all the physicochemical characteristics of site water using laboratory water. Therefore, if characteristics of site water, other than hardness, alkalinity and pH, are suspected to influence in the expression of toxicity, it may be useful to perform toxicity testing using site water in order to account for site-specific effects.

Unexposed surface water includes mine-site-collected water that has been collected upstream of a mine effluent discharge or from a nearby reference area. Unexposed surface water from the mine site area may vary in physical, chemical and biological characteristics over time.

Disadvantages of using unexposed surface water as dilution/control water include the following:

• Relatively large volumes of unexposed surface water may be needed for testing, thus additional expense is incurred for the collection, shipment and storage of site-water samples.
• Some laboratory organisms will need acclimation to the unexposed surface water if it is significantly different in physicochemical characteristics from the laboratory water (refer to section 6.9.1).
• Mandatory screening of the water through 60-µm mesh is required to ensure indigenous populations of micro- or macro-organisms present in surface water do not compete with or impair the health of laboratory test organisms.

In spite of its practical technical disadvantages as control/dilution water, unexposed surface water may provide more site-specific toxicity information. The advantages include the following:

• It reflects the physical/chemical characteristics of the receiving environment.
• It could indicate the potential for non-discharge-related effects.
• Tests conducted may better reflect the influence of receiving environment characteristics on toxicant potency than tests conducted with laboratory water.

In Canada, this practice has been logistically constraining due to the large volumes of water needed to be shipped far distances. However, recently it has been shown that individual sites can identify 1 or 2 test organisms most likely to detect site-specific changes in effluent quality, so tests using receiving water could be conducted on just 1 or 2 species (Taylor et al. 2010). For example, the Pseudokirchneriella test requires smaller volumes of water, making this test an ideal candidate for evaluating the effects of receiving-water chemistry on effluent toxicity (Taylor et al. 2010). Sites should determine which type of water best suits their study objectives.

It should be noted that when using site water, problems may arise with reference water exhibiting sublethal responses. Beak International Inc. (BEAK 1998) attributed this problem to indigenous populations of micro-organisms infecting the laboratory organisms. BEAK staff found that boiling the water prior to use before testing was successful in reducing mortality.

Sprague (1997) prepared an extensive review of studies that compared toxicity test results to receiving-water impact, and concluded that effects measured in sublethal toxicity tests correlate with environmental effects most of the time, especially if water collected upstream of the effluent discharge is used as the control/dilution water.

When the purpose of a sublethal toxicity test is to estimate site-specific effects of contaminants, unexposed surface water from the vicinity of the mine site is recommended for use as control/dilution by Environment Canada, the U.S. EPA and the American Society for Testing and Materials (ASTM) in their method and associated guidance documents (Environment Canada 1992a, 1992b, 1998, 2007a, 2007c; US EPA 1994a; ASTM 1998).

The widespread acceptance of unexposed surface dilution water for predicting site-specific effects is based on knowledge regarding the interaction of contaminants with water quality characteristics. For example, metal toxicity is well known to be influenced by the physicochemical characteristics of water, such as pH, alkalinity, hardness (reviewed by Wang 1997 and Sprague 1995). However, studies comparing results of toxicity tests on effluents and effluent-exposed surface waters from 4 different mine sites indicated that a similar estimation of toxicity could be obtained from tests using unexposed surface water or laboratory water for dilution, especially if the laboratory water was adjusted to the hardness, alkalinity and pH of the site water (BEAK 1998, 1999). This finding indicates that the use of site-collected dilution water may not always be necessary, because laboratory waters can be prepared to reflect site-water characteristics such as hardness, pH and alkalinity.

Naturally elevated levels of organic carbon is one important aspect of some site waters; organic carbon is known to mitigate the effects of metal toxicity by reducing metal bioavailability. Ranges of dissolved organic carbon (DOC) have been measured as high as 58 mg/L in some Ontario lakes (Neary et al. 1990, as cited in Welsh et al. 1993), and such waters may influence the expression of metal toxicity in effluent or effluent-exposed surface water matrices. However, in parallel toxicity tests of mine effluent conducted using site water of moderately high organic carbon (total organic carbon of 9.4 mg/L) and hardness-adjusted laboratory water, there was no significant difference in organism response (BEAK 1999). Of note is that high organic carbon content may not always result in reduced metal bioavailability. In waters of high hardness, concentrations of calcium and magnesium may be high enough to bind with humic acid (which makes up the majority of organic carbon). Therefore, humic acid binding sites would be limited and unavailable to tie up free metal ions. In this situation, decreases in effluent sublethal toxicity due to high receiving water DOC would not occur (Winner 1985). In addition, the type of organic matter will also influence metal bioavailability (i.e., more allochthonous-like organic matter decreases Cu toxicity better than autochthonous-like natural organic matter; Schwartz et al. 2004), which complicates the interpretation of DOC’s role in reducing metal toxicities to aquatic biota.

Comparing toxicity of the discharge or effluent-exposed surface water relative to an impaired upstream water

If upstream water is contaminated by nonpoint or upstream-point sources of pollution that are unrelated to the mine operation, a mine may decide to use that water for test dilution purposes in toxicity testing in order to provide an appropriate comparison of test organism responses, as long as the upstream water can support the health of the test organisms. If the upstream water cannot support health of the test organisms, it could be tested separately in a dilution series to quantify its effects with uncontaminated reference site or regular laboratory water used for test control and dilutions.

6.9 Collection, Shipment and Storage of Samples for Sublethal Toxicity Testing

The procedures for the collection, shipment and storage of site-collected dilution water are outlined in each of the Environment Canada and EPA test methods (Environment Canada 1992a, 1992b, 1998, 2007a, 2007b, 2007c; US EPA 1994a, 1994b). Table 6-4 provides estimates of the volumes of site water needed for performing a suite of EEM tests, and includes estimates for effluents or effluent-exposed surface water volume. As recommended by the U.S. EPA (1994a), site-collected dilution water samples should be representative of the water body and be unaffected by recent runoff or erosion events that may cause the water to have a higher total suspended solids concentration.

All volumes are calculated assuming 1 control and 7 test concentrations, except for the Rainbow Trout embryo test where volumes are calculated assuming 1 control and 5 test concentrations. Fathead Minnow assumes 500 ml test volume and 3 replicates, and Lemna minor assumes 150 ml test volume and 4 replicates.
* Estimated effluent volumes for marine/estuarine sublethal toxicity tests are outlined in Table 6-2.

Top of Page

6.9.1 Test Organism Acclimation

Pre-acclimation of culture organisms is recommended prior to exposure to site water. As the purpose of using site water for test control and dilutions is to more accurately predict receiving-water impact, the most accurate prediction should be achieved using organisms adapted to the physicochemical conditions of the receiving environment. For example, Lloyd (1965) found that fish cultured in hard water need to lose calcium before they are as sensitive to metals as fish cultured in soft water. Therefore, if the site water to be used for test dilution is softer than the lab water, pre-acclimating fish to the site water conditions would provide enough time for loss of calcium prior to test initiation. Alternatively, if site-collected water is higher in hardness, then pre-acclimation would allow fish species to accumulate calcium prior to testing.

The Environment Canada methods for Ceriodaphnia dubiaand Fathead Minnows recommend that cultures be maintained in water of similar hardness, alkalinity and pH (i.e., within 20%) to the site water used for test dilution (Environment Canada 1992a, 2007a). Since many Canadian mines are located beside rivers or lakes of low hardness, it is likely that some acclimation of laboratory cultures would be necessary. BEAK developed a pre-acclimation procedure during the 1997 Aquatic Effects Technology Evaluation (AETE) Program study, later refined in the 1999 study, for Fathead Minnows and Ceriodaphnia dubia(BEAK 1998, 1999). Based on the expected hardness, alkalinity and pH of the site water, cultures are gradually introduced to laboratory water of decreasing hardness over several days until the appropriate hardness is reached. This procedure was adapted from that used by B.A.R. Environmental Inc. (BAR) during its 1996 AETE Program study, in which cultures were gradually acclimated to hardness-adjusted laboratory water and site water if screening of un-acclimated organisms showed impairment to laboratory organisms (BAR 1997).

For pre-acclimation of cultures, laboratory water of reduced hardness may be prepared by diluting standard laboratory water with deionized water. Hardness can be increased by adding salts, used for the preparation of reconstituted water in the appropriate amounts (Table 6-5). When hardness-adjusting water, it is important to keep the alkalinity level appropriate to the hardness, because alkalinity affects the speciation of metals (US EPA 2002; Laurén and McDonald 1986). Appropriate hardness and alkalinity relationships are available in Table 6-5, and additional values may be interpolated (US EPA 1994b).

Detailed procedures for pre-acclimation of Ceriodaphnia dubia and Fathead Minnow are described below. References to hardness assume a corresponding change in alkalinity and pH. No pre-acclimation procedures are described for the Rainbow Trout test, since eggs are delivered from the hatchery and used in testing within 24 hours (Environment Canada 1998). Similarly, Pseudokirchneriella and Lemna cultures are maintained in standard culture media that are different from standard testing media, although Lemna does allow for some pre-acclimation of cultures, since plants are transferred to the test medium 18 to 24 hours prior to initiation of testing (Environment Canada 2007c).

Source: US EPA (1994a)
¹ Add reagent-grade chemicals to deionized water.
² Approximate equilibrium pH after 24 hours of aeration.
³ Expressed as milligrams mg) CaCO₃/L.

Ceriodaphnia dubia

Ceriodaphnia cultures are initiated and maintained according to the Environment Canada standard method. To pre-acclimate cultures to the hardness of the site-collected dilution water, a brood of neonates, less than 24 hours old, is initiated in water, reduced in hardness by 20% from that of laboratory water by addition of deionized water. Each day, organisms are transferred to new solutions, decreased by a further 20-30% in hardness. Once the desired hardness is reached (within approximately 1 week) and the culture organisms pass the health criteria of the Environment Canada method (i.e., production of at least three broods, total neonate production of at least 15 per adult, less than 20% adult mortality), a new culture is initiated. The second-generation cultures are maintained in the hardness-adjusted water until organisms pass the health criteria (approximately 1 week). Selenium and vitamin B₁₂ are added to the hardness-adjusted culture water if low in hardness, as recommended by the standard method.

Fathead Minnows

Acclimation and pre-acclimation procedures carried out by BAR and BEAK during the 1996 and 1997 AETE studies used breeding tanks of Fathead Minnows, gradually changed in hardness and alkalinity to that of the applicable site water (hardness-adjusted laboratory water or HALW) (BAR 1997; BEAK 1998). Eggs were collected from the adjusted-water tanks and reared at the HALW until hatching. However, preliminary work completed by BEAK in early 1999 showed that eggs could be hatched out in HALW without prior acclimation of the breeding tanks. By omitting breeding tanks from the pre-acclimation procedure, the volume of HALW needed for pre-acclimation is reduced while maintaining a supply of eggs for hatching in a number of different water types. Eggs are hatched according to the Environment Canada standard method, with fresh water renewal every 24 hours.

Top of Page

6.9.2 Screening Tests

Once organisms are pre-acclimated to the physicochemical conditions of the site water, they may be exposed to the site water in screening tests. If organisms exposed to the site water meet the test method control acceptability criteria, the site water may be considered suitable for use as dilution water provided the site water meets the control validity criteria for the test method. Comparison of the site-water response to the laboratory response in a screening test may reveal a statistically significant reduction in reproduction and/or survival, even though the site-water-exposed group is within the control acceptability criteria for the test. As long as the site water meets the control validity criteria for the test method, it may be considered suitable for testing. For additional information, consult the Environment Canada test methods.

Top of Page

6.9.2.1 Screening Test Impairment

If organisms are given an adequate opportunity to gradually acclimate to the physicochemical characteristics of the site water (by exposure to hardness-adjusted laboratory water), impairment observed during screening tests should not be due to shock exposure to different water quality characteristics such as hardness, alkalinity or pH. Therefore, impairment would likely be due to the presence of harmful biological agents or toxicants.

Special attention should be paid to any 100% site-water exposures showing significant mortality in one or more replicates. Microbiological organisms present in the site water may impair the health of test organisms, and anecdotal evidence from several ecotoxicity laboratories indicates that impairment by indigenous micro-organisms usually occurs after a few days of exposure. For example, this has been manifested in Fathead Minnow tests conducted at BEAK as a sudden onset of significant mortality in the site-water control, often in one or two replicates only. Occasionally, evidence of fungal or bacterial growth may be observed in the test vessels. If such contamination is indicated by a screening test, acclimation of the organisms will most likely not result in a removal of impairment. Therefore, either the site water should be treated by a suitable means to remove the impairment (i.e., boiling or ultraviolet treatment--see below), or hardness-adjusted laboratory water should be used as a surrogate.

Limited laboratory trials of site water showed that impairment could be removed by boiling site water gently for 10 minutes and cooling it prior to use in testing. Other treatments have been reported in the literature, such as ultraviolet light and 0.45-µm filtration (Grothe and Johnson 1996; Kszos et al. 1997). If site water is collected during late spring to early fall, some form of biological contamination should be expected (unless experience with a particular site water indicates otherwise), and precautions such as boiling should be taken.

If the impairment is due to chemical contamination, the suitability of the site water for use as a dilution and control water is questionable, even though cultures may be acclimated to naturally high levels of metals in site water (see section 6.9.1). If the cultures are exposed to higher levels of contaminants, post-acclimation can result in either higher or lower sensitivity of laboratory organisms, depending on the contaminant, organism and the water characteristics, and the utility of the post–screening test acclimation procedure becomes questionable. If impairment is detected in a screening test, the recommended procedure is to attempt treatment by boiling or to use hardness-adjusted laboratory water as a surrogate dilution water. If it is suspected that site water is contaminated, including a boiled-site-water exposure in screening tests would resolve the question of the effectiveness of that treatment for the site water of interest.

Top of Page

6.9.3 Effluent or Effluent-Exposed Surface Water Toxicity Tests

Table 6-3 summarizes the recommendations for collecting, storing and using site-collected dilution waters in sublethal toxicity testing.

Once site water has been deemed suitable for testing, toxicity tests may be initiated as per the appropriate Environment Canada testing method. In addition to the site-water control, an additional control, using water from the laboratory culture, needs to be included in testing to serve as a check of culture health and site-water quality. No additional control is necessary when the test-control dilution water is the same as the culture water. In the case of Lemna minor and Pseudokirchneriella subcapitatatests, the laboratory control would be the standard-test growth medium, specified in the standard method.

If site-collected water is deemed unsuitable for use as test control and dilution water, a treatment such as boiling should be attempted in order to remove the impairment. If successful, the treated site-collected dilution water should be used in testing. However, boiling large volumes of water may not be practical for tests such as the Rainbow Trout test, and hardness-adjusted laboratory water may be a more suitable alternative. If no practical treatment can be found to remove impairment to test organisms caused by the site-collected dilution water, hardness-adjusted water should be used as the dilution water with pre-acclimated organisms. Care should be taken to match the pH of the characteristics of the site-collected dilution water as closely as possible.

6.10 Use of Sublethal Toxicity Testing in Resolving Confounding Influences

Sublethal toxicology data also have the potential utility to aid in the resolution of confounding factors. A multistakeholder group on metal mining toxicology has elaborated on this third use for sublethal toxicity data.

Estimating the relative contribution of mine effluent releases and other natural and/or anthropogenicinfluences on sublethal toxicity in the same receiving water body

During any phase of the EEM program, sublethal toxicity test data can be used to deal with situations where there are confounding influences. Sometimes the site characteristics do not permit full determination of the mine’s effluent effects even with an adapted study design. Information from sublethal toxicity testing may then help in the interpretation of field results. The choice of when to use sublethal toxicity tests in this application is up to the mine operator and the nature of the confounding influences. However, the confounding influence scenario where sublethal toxicity testing would be most relevant is the multiple-point-source discharge and/or nonpoint-source input situation. Sublethal tests or frequency monitoring should be determined based on the site-specific nature of confounding influence situations.

Estimates of sublethal toxicity can help in understanding the relative contribution of diverse industrial or municipal discharges to effects on aquatic organisms in the receiving water, whether the discharges are from upstream point or nonpoint sources (e.g., municipal landfill leachate, agricultural runoff) or the mine’s property. The upstream contribution of an observed environmental effect can be estimated, given surface water sublethal toxicity data, discharge flow, and features of dispersion into the receiving environment. If plumes from different discharges at a mine site overlap, more effort is necessary to distinguish the toxic contributions of the mine’s discharge sources vs. upstream sources. Samples of surface water from key locations in the high effluent exposure areas could be tested, to estimate the combined toxic contribution of the sources.

The following is a 3-step procedure for assessing the relative contribution from different sources of sublethal toxicity to the high effluent exposure receiving environment:

• Conduct a battery of sublethal toxicity tests on samples collected from all significant discharge sources from the mine’s property. Use standard laboratory water for test dilutions and control, or unexposed site water. This estimates the absolute sublethal toxicity of each mine-site discharge. Repeat the sampling and testing on any discharge that is known to be variable in toxicity, in order to obtain an estimate of the degree of variability.
  
• Conduct a parallel battery of sublethal toxicity tests for each discharge to a river, using water collected directly upstream from the discharge point for dilution and control. For lakes or estuaries, carry out the parallel battery of tests by collecting control/dilution water from outside the zone immediately affected by the discharges. Separate and simultaneous controls should be run using standard uncontaminated water as a QA measure. It should be recognized that “upstream” sources of control/dilution water might already be contaminated by other effluent discharges or sources of toxicants. Accordingly, the upstream dilution water might contribute to significant effects on growth or reproduction in concentrations of the effluents being studied or even in control vessels. This would not invalidate the results, because the purpose of the investigation is to evaluate the relative contributions of discharges to the total toxicity of the receiving water.
  
• Confirmation of the relative contribution of discharges is recommended, and can be achieved by conducting sublethal toxicity tests on samples of surface water from the water body receiving the discharges (so-called “ambient” tests). This can aid in:
  1. confirming whether an effluent has a measurable toxicity after mixing into the receiving water;
  2. estimating the persistence in the receiving water of toxicity from all contributing sources; and
  3. determining the combined toxicity resulting from the mixing of all point and nonpoint sources, as an estimate of the overall effect on the receiving environment.

Testing samples of surface water, which receives discharges or toxicants from multiple sources, should be done synoptically and ideally during low-flow or worst-case periods. At a minimum, sampling should be carried out over as short a period of time as possible (e.g., 1 or 2 days). Repeated rounds of sampling and testing would be desirable if the toxicity of the discharges were variable. The above guidance on conducting toxicity assessment studies to estimate the contribution of multiple-discharge sources to instream effects is based on the 8 site investigations conducted under the U.S. EPA Complex Effluent Toxicity Testing Program (CETTP). Detailed reports on these studies were prepared by Mount and Norberg-King (1985, 1986), Mount et al. (1984, 1985, 1986a, 1986b, 1986c) and Norberg-King and Mount (1986). The results of CETTP testing, including independent critiques and re-analyses, were reviewed by Sprague (1997) during a project commissioned by the Aquatic Effects Technology Evaluation (AETE) Program. Sprague concluded that the U.S. EPA CETTP studies provided valid findings that should be considered by Canadian metal mining companies when designing aquatic environmental monitoring programs at their mine sites.

Mines may elect to conduct additional investigations where the most sensitive species in the effluent produced an IC₂₅of less than 30%. Below are additional recommended investigations. At a minimum, the most sensitive test species can be used to estimate the geographic extent of the potential response. Alternatively, the results of the toxicity test(s) may lead to or trigger other recommended laboratory or field monitoring tools.

A tiered approach to resolving the confounding influences is recommended, starting with these additional recommended investigations:

1. re-testing with the sublethal test that provided the most sensitive IC₂₅ result, using upstream or reference-site water for test control and dilutions; or
2. receiving-water toxicity testing with samples collected from the area where a sublethal response is predicted.

Top of Page

6.11 References

[ASTM] American Society for Testing and Materials. 1998. Standard guide for conducting acute toxicity tests on aqueous ambient samples and effluents with fishes, macroinvertebrates, and amphibians, designation E1192-97. Conshohochen (PA): Annual book of ASTM standards. Section 11: Water and environment technology.

[BAR] B.A.R. Environmental Inc. 1997. Toxicity assessment of mining effluents using upstream or reference site waters and test organism acclimation techniques. Aquatic Effects Technology Evaluation Program. AETE Project 4.1.2a.

[BEAK] Beak International Incorporated. 1998. Additional tool evaluations. Aquatic Effects Technology Evaluation Program. Beak Reference 20776.1.

[BEAK] Beak International Incorporated. 1999. Final report: effects of dilution water on the results of sublethal toxicity tests. Report to Natural Resources Canada. Beak reference 33940.

Chapman PM. 2000. Whole effluent toxicity testing – usefulness, level of protection, and risk assessment. Environ Toxicol Chem 19(1):3-14.

Environment Canada. 1992a. Biological test method: test of larval growth and survival using fathead minnows. Ottawa (ON): Environmental Technology Centre. Report EPS 1/RM/22, February 1992, Amended in September 2008.

Environment Canada. 1992b. Biological test method: fertilization assay with echinoids (sea urchins and sand dollars). Ottawa (ON): Environmental Technology Centre. Report EPS 1/RM/27, December 1992, Amended in November 1997.

Environment Canada. 1998. Biological test method: toxicity tests using the early life stages of salmonid fish (rainbow trout). Report EPS 1/RM/28, 2nd ed., July. Ottawa (ON): Environmental Technology Centre.

Environment Canada. 1999. Recommended procedure for the importation of test organisms for sublethal toxicity testing. Ottawa (ON): Environmental Technology Centre.

Environment Canada. 2001. Revised procedures for adjusting salinity of effluent samples for marine sublethal toxicity testing conducted under environmental effects monitoring (EEM) programs. Ottawa (ON): Method Development and Applications Section, Environmental Technology Centre.

Environment Canada. 2007a. Biological test method: test of reproduction and survival using the Cladoceran Ceriodaphnia dubia. Ottawa (ON): Environmental Technology Centre. Report EPS 1/RM/21, 2nd edition, February.

Environment Canada. 2007b. Biological test method: growth inhibition test using a freshwater alga. Ottawa (ON): Environmental Technology Centre. Report EPS 1/RM/25, 2nd edition, March 2007.

Environment Canada. 2007c. Biological test method: test for measuring the inhibition of growth using the freshwater macrophyte Lemna minor. Ottawa (ON): Environmental Technology Centre. Report EPS 1/RM/37, 2nd edition, January 2007.

[ESG] ESG International Inc. 1999. AETE synthesis report of selected technologies for cost-effective environmental monitoring of mine effluent impacts in Canada. AETE Project 4.1.4, March 1999. Ottawa (ON): Canadian Centre for Mineral and Energy Technology, Natural Resources Canada.

Grothe DR, Johnson DE. 1996. Bacterial interference in whole effluent toxicity tests. Environ Toxicol Chem 15(5):761-764.

Kszos LA, Stewart AJ, Sumner JR. 1997. Evidence that variability in ambient fathead minnow short-term chronic tests is due to pathogenic infection. Environ Toxicol Chem 16(2):351-356.

Laurén DJ, McDonald DG. 1986. Influence of water hardness, pH and alkalinity on the mechanisms of copper toxicity in juvenile rainbow trout, Salmo gairdneri. Can J Fish Aquat Sci 43:1488-1496.

Lloyd R. 1965. Factors that affect the tolerance of fish to heavy metal poisoning. In: Biological problems in water pollution, third seminar, 1962. Washington (DC): U.S. Public Health Service. Publ. 999-WP-25. p. 181-187.

[MDDEP] Ministère du Développement durable, de l’Environnement et des Parcs du Québec. 2007. Détermination de la toxicité : inhibition de la croissance chez l’algue Pseudokirchneriella subcapitata. Centre d’expertise en analyse environnementale du Québec. MA. 500-P.sub 1.0, Revised in September 2007).

Mount DI, Norberg-King TJ, editors. 1985. Validity of effluent and ambient toxicity tests for predicting biological impact, Scippo Creek, Circleville, Ohio. Washington (DC): U.S. Environmental Protection Agency. EPA 600/3-85/044.

Mount DI, Norberg-King TJ, editors. 1986. Validity of effluent and ambient toxicity tests for predicting biological impact, Kanawha River, Charleston, West Virginia. Washington (DC):U.S. Environmental Protection Agency. EPA 600/3-86/006.

Mount DI, Thomas NA, Norberg-King TJ, Barbour MT, Roush TH, Brandes RW. 1984. Effluent and ambient toxicity testing and instream community response on the Ottawa River, Lima, Ohio. Washington (DC): U.S. Environmental Protection Agency. EPA 600/3-84/080.

Mount DI, Steen AE, Norberg-King TJ. 1985. Validity of effluent and ambient toxicity testing for predicting biological impact on Five Mile Creek, Birmingham, Alabama. Washington (DC): U.S. Environmental Protection Agency. EPA 600/8-85/015.

Mount DI, Norberg-King TJ, Steen AE. 1986a. Validity of effluent and ambient toxicity tests for predicting biological impact, Naugatuck River, Waterbury, Connecticut. Washington (DC): U.S. Environmental Protection Agency. EPA 600/8-86/005.

Mount DI, Steen AE, Norberg-King TJ. 1986b. Validity of effluent and ambient toxicity tests for predicting biological impact, Back River, Baltimore Harbor, Maryland. Washington (DC): U.S. Environmental Protection Agency. EPA 600/8-86/001.

Mount DI, Steen AE, Norberg-King TJ. 1986c. Validity of ambient toxicity tests for predicting biological impact, Ohio River near Wheeling, West Virginia. Washington (DC): U.S. Environmental Protection Agency. EPA 600/3-85/071.

Neary BP, Dillon PJ, Munro JR, Clark BJ. 1990. The acidification of Ontario lakes: an assessment of their sensitivity and current status with respect to biological damage. Toronto (ON): Ontario Ministry of the Environment. 171 p.

Norberg-King TJ, Mount DI, editors. 1986. Validity of effluent and ambient toxicity tests for predicting biological impact, Skeleton Creek, Enid, Oklahoma. Washington (DC): U.S. Environmental Protection Agency. EPA 600/8-86/002.

Schwartz ML, Curtis PJ, Playle RC. 2004. Influence of natural organic matter source on acute copper, lead, and cadmium toxicity to rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem 23(12):2889-2899.

Sprague J. 1995. Factors that modify toxicity. In Rand G, editor. Fundamentals of aquatic toxicology. 2nd edition. Washington (DC): Taylor and Francis.

Sprague J. 1997. Review of methods for sublethal aquatic toxicity tests relevant to the Canadian metal-mining industry. Ottawa (ON): Aquatic Effects Technology Evaluation Program, Canadian Centre for Mineral and Energy Technology, Natural Resources Canada.

Taylor LN, Van der Vliet LA, Scroggins RP. 2010. Sublethal toxicity testing of Canadian metal mining effluents: national trends and site-specific uses. Hum Ecol Risk Assess 16(2):264-281.

[US EPA] United States Environmental Protection Agency. 1994a. Short-term methods for estimating the toxicity of effluents and receiving water to freshwater organisms. Third edition. Cincinnati (OH): Environmental Monitoring Systems Laboratory, U.S. Environmental Protection Agency. EPA /600/4-90/027.

[US EPA] United States Environmental Protection Agency. 1994b. Interim guidance on determination and use of water-effect ratios for metals Office of Water, U.S. Environmental Protection Agency. EPA 823-B-94-001.

[US EPA] United States Environmental Protection Agency. 1995. Short-term methods for estimating the toxicity of effluents and receiving water to West Coast organisms. First edition. Cincinnati (OH): Environmental Monitoring Systems Laboratory, U.S. Environmental Protection Agency. EPA 600/R-95/136.

[US EPA] United States Environmental Protection Agency. 2002. Short-term methods for estimating chronic toxicity of effluent and receiving waters to marine and estuarine organisms. Third edition. Cincinnati (OH): Environmental Monitoring Systems Laboratory, U.S. Environmental Protection Agency. EPA/821/R-02/014, October 2002.

Wang W. 1997. Factors affecting metal toxicity to (and accumulation by) aquatic organisms - an overview. Environ Internat 13:437-457.

Welsh PG, Skidmore JF, Spry DJ, Dixon DG, Hodson PV, Hickie BE. 1993. Effect of pH and dissolved organic carbon on the toxicity of copper to larval fathead minnows (Pimephales promelas) in natural lake waters of low alkalinity. Can J Fish Aquat Sci 50:1356-1362.

Winner RW. 1985. Bioaccumulation and toxicity of copper as affected by interactions between humic acid and water hardness. Water Res 19:449-455.

Tables

Table 6-1 outlines methodologies for effluent sublethal toxicity tests. Test descriptions--which include fish early life stage development tests, invertebrate reproduction tests, and plant and algae toxicity tests--are linked to receiving environment, test species, and methods.

Return to the table

Table 6-2 offers descriptions of the freshwater and marine sublethal toxicity tests included in the metal mining EEM program. Each test is aligned with its purpose and results, a description, and the biological test method and cost.

Return to the table

Table 6-3 provides the specifications of the Environment Canada test methods and recommendations for collection, storage and use of site-collected dilution waters. Criteria for acceptable dilution water and site water are identified and aligned with two examples: a fathead minnow, and a rainbow trout embryo.

Return to the table

Table 6-4 outlines dilution/control water and corresponding effluent volumes for sublethal toxicity tests. Tests and testing series are identified, and aligned accordingly with their dilution water volumes (in litres), and their effluent volumes (in litres).

Return to the table

Table 6-5 illustrates the preparation of water with different hardness and alkalinities. Water types include very soft, soft, moderately hard, hard, and very hard. Each water type is aligned with reagent added (mL/L) and the final water quality.

Return to the table

1IC₂₅ is defined as the effluent concentration where a 25% inhibition is observed in the exposed test organisms.

2 The geometric mean may be calculated as the n^th root of n numbers multiplied together. Alternatively, the logarithms of the n IC₂₅s (EC₂₅s or LC₅₀s) may be added together, the sum divided by n, and the antilog of the result is the geometric mean.

7. Sediment Monitoring

7.1 Overview

7.2 Collection of Sediment Samples