This page has been archived on the Web

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

2006 - 2008 Literature Review Archives - Climate Projections

Arzel, O., Fichefet, T. and Goosse, H. 2006. Sea ice evolution over the 20th and 21st centuries as simulated by current AOGCMs. Ocean Modelling 12 pp 401-415 doi:10.1016/j.ocemod.2005.08.002.
Sea ice results from a new set of simulations performed by over a dozen global climate models (including the Canadian model CGCM3.1) for the IPCC Fourth Assessment Report are presented in this paper. The multi-model average sea ice extent in both March (sea ice maximum) and September (sea ice minimum) agrees reasonably well with observations in both hemispheres although there are substantial differences among models. The multi-model average trend in sea ice for the NH (1981-2000 period) is also close to the observed trend (-2.15 x 105 km2 per decade vs -2.4 x 105 km2 per decade). In the SH, most models show a decrease in sea ice extent while observations indicate a slight increase over the same period. This difference is not yet understood, and clearly models are still having some difficulty in reproducing SH sea ice behaviour. The climate change projections (based on the IPCC SRES AIB scenario) indicate that sea ice decline will continue over the 21st century and half the models show an ice-free summer Arctic by the end of the century.

Baettig, M.B., M. Wild and D. Imboden. 2007. A climate change index: where climate change may be most prominent in the 21st century. GRL Vol 34, l01701, doi:10.1029/2006GL028159, 2007.
A climate change index (CCI) is derived that provides a single measure of the strength of future climate change relative to today's natural variability. Canada and other high latitude regions show relatively high values for the CCI, but when a country's Human Development Index is considered along with its CCI, it is in the tropics that the countries most vulnerable to climate change are found.
The authors of this study have developed an aggregate measure for the strength of future climate change relative to natural variability. Their index is composed of annual and seasonal temperature and precipitation indicators which each measure some change in extreme events (e.g. change in occurrence of 1 in 20 year hot/wet/dry event). The method is based on the assumption that changes in extremes are likely to have the strongest impacts on environmental and social systems. The index is calculated based on gridded global temperature and precipitation projections from 3 GCMs (ECHAM5, HadCM3 and CGCM2) using both moderate (SRES B2) and high (SRES A2) emission scenarios. The results indicate that climate will change most strongly relative to natural variability in the high latitudes and the tropics. When the CCI is averaged for each country, and then compared to the Human Development Index - a measure of socio-economic development - the vulnerability of many tropical countries to future climate change is made evident.
Reference:

Barnett, D.N. et al., 2006.
Quantifying uncertainty in changes in extreme event frequency in response to doubled CO2 using a large ensemble of GCM simulations, Climate Dynamics, January 2006, 22 pages. For the first time, a large ensemble of simulations was produced to calculate, under a 2XCO2 scenario, the changes in extreme temperature and precipitation events, while exploring the uncertainties arising both from the parameterisation of the atmospheric physical processes and the effects of natural variability. The study confirms that, under a 2XCO2 scenario, changes in extremes will be substantial. For example, extremely warm days (> 99th percentile) occurring on one day per hundred under present conditions would, globally, occur on 20-30 days per hundred under 2XCO2 conditions. Also, the most extreme events (> 99th percentile) are likely to increase more than unusual events (> 90th percentile) and this finding holds true for both temperature and precipitation. The study also confirms that changes in precipitation extremes are less robust (less agreement among simulations) and hence more difficult to evaluate than changes in temperature extremes.

Brasseur, G. P., Schultz, M., Granier, C., Saunois, M., Diehl, T., Botzet, M. and Roeckner, E. 2006. Impact of Climate Change on the Future Chemical Composition of the Global Troposphere. Journal of Climate 19(16): 3932-3951.
In this article, scientists report on use of the Model for Ozone and Related Tracers, volume 2 (MOZART-2) with ECHAM5 (the latest version of the German Max Planck Institute's GCM) to examine the impacts of climate change on concentrations of tropospheric ozone and other pollutants. Following a baseline scenario using year-2000 emissions data, the group examined the effects of CO2-doubling and increasing air pollutant emissions (according to the SRES A2P scenario). They also used the model to attempt to isolate the effects of methane, increased biogenic VOC emissions, increased soil NOxemissions and increased biomass burning on concentrations of tropospheric ozone and other pollutants. All results were for the year 2100. Significant results included a reduction in extratropical ozone levels with CO2doubling and accompanying climate change, due to increased atmospheric water vapour at higher temperatures (which tends to react with ozone to remove it from the atmosphere); slight tropical increases in ozone with doubled CO2due to increased lightning activity (which converts atmospheric N2to NOx); increased ozone concentrations in polluted areas with increased pollutant emissions; reduced ozone levels in non-polluted areas with an arbitrary fixing of methane at year-2000 levels (methane participates in smog formation); increases in ozone when levels of NOxfrom soils and of isoprene (a biogenic VOC) were arbitrarily doubled to simulate increased emissions at higher temperatures; and finally, increased ozone with an arbitrary doubling of emissions from biomass burning, especially in regions like Northern Canada where wildfires are projected to increase the most with climate change. This article is consistent with a number of other studies examining the impacts of climate change on tropospheric ozone concentrations.

Butchart, N., A. Scaife, M. Bourqui, et al., 2006. Simulations of anthropogenic change in the strength of the Brewer-Dobson circulation. Climatic Dynamics (2006) 27:272-741.
In this modeling study, ten GCMs were used to investigate the effect of climate change on the Brewer-Dobson circulation and, in particular, the large-scale seasonal-mean transport between the troposphere and stratosphere. The Brewer-Dobson circulation is a global scale cell in the stratosphere of rising air in the tropics and sinking in the extra-tropical regions. The authors performed 13 climate-change experiments using ten GCMs. All the models were able to reproduce the upwelling and downwelling characteristics of the Brewer-Dobson circulation. The results from the experiments unanimously predict that an increase in greenhouse gases will increase the mean troposphere-to-stratosphere mass exchange, irrespective of whether or not the model included interactive ozone chemistry that may change the dynamics. The mean trend was 11 kt / s year or about 2% per decade but there was considerable variability between models. The largest increase is predicted for the winter. The increase is thought to be primarily due to an increase propagation of waves into the stratosphere. An increase in the Brewer-Dobson circulation would increase the rate ozone is transported from the tropics into the extra-tropical regions, offsetting some of the effects of ozone depletion. This is consistent with the finding of the 2007 Scientific Assessment of ozone depletion that recognizes that variability in ozone transport is equally important as anthropogenic chemical loss in determining year-to-year Arctic ozone variability.

Cox, P.M., Harris, P.P., Huntingford, C. et al., 2008. Increasing risk of Amazonian drought due to decreasing aerosol pollution. Nature. 453:212-216.
A new modelling study indicates that the cooling effects of aerosol emissions in the Northern Hemisphere have likely delayed the decrease in precipitation over the Amazon that would otherwise have occurred due to rising greenhouse gas concentrations. It also projects that measures to reduce air pollution will likely reverse this influence and hence result in a dramatic increase in drought conditions in the Amazon region in future decades.
Past climate studies have indicated that variations in the precipitation over the Amazon are linked to changes in the north-south sea surface temperature (SST) gradient in the equatorial Atlantic Ocean. A British team of climate modellers have now shown that such changes may also be linked to changes in atmospheric aerosol concentrations. To undertake their studies, these researchers used a variant of the HadCM3 coupled climate model that includes a bio-geochemical cycle component, and that appears to replicate observed north-south SST gradients and the Amazon climate better than most models. Simulation runs with this model for the time period from 1860 to 2100 included one with IS92a forcing scenario for changes in well mixed greenhouse gas concentrations only, and another that also included projected changes in tropospheric and stratospheric ozone, natural forcings (solar and volcanic) and anthropogenic aerosol emissions. The GHG-only scenario failed to properly capture the observed variations in the latitudinal SST gradient and the related changes in Amazonian rainfall patterns. When aerosol effects were added, the agreement with observations was much better. Authors indicate that the aerosol forcing appears to delay the strengthening of the north-south SST gradient and hence defers the decrease in precipitation over the Amazon that is predicted for the GHG-only scenario. The simulations further indicate that a reduction in aerosol forcings projected in future decades will reverse this response by strengthening the north-south SST gradient. As a result, the projections indicate that the 2005-like drought conditions over the Amazon, which now occur about once every 20 years, are likely to occur every second year by 2025, and 9 out of every 10 years by 2060. Such events also increase the biospheric CO2 emissions into the atmosphere, thus accelerating the increase in atmospheric CO2 concentrations.

Delworth, Thomas L. and Keith W. Dixon (2006). Have anthropogenic aerosols delayed a greenhouse gas-induced weakening of the North-Atlantic thermohaline circulation? GRL vol.33, L02606,doi:10.1029/2005GL024980,2006.
With the use of an ensemble of simulations using GFDL's newly developed coupled ocean-atmosphere climate model CM2.1, the authors explore the impact of various climate change forcing agents on the evolution of the thermohaline circulation (THC) during the 20th and 21st centuries. They conduct ensemble simulations with several subsets of forcing agents which included experiments that isolated the effects of greenhouse gas and aerosol emissions, to see what is the most important in the weakening of the THC. They found that there was no statistically significant decrease in THC, over the period 1860 to 2000. This is the result of the compensation between GHG forcing, which produces a decrease in THC when acting in isolation, and anthropogenic aerosol (sulfate + black and organic carbon) forcing which tends to enhance the THC. Thus, aerosols in the 20th century have delayed a GHG induced weakening of the THC by several decades (40 years with their model), but this could be a temporary situation. The authors also found a significant decrease of the THC several decades into the 21st century, but no evidence of a complete shut down. At that point, the effects of increasing GHG dominate aerosol effects.

Dumas, J.A., G.M. Flato and R.D. Brown. 2006. Future Projections of Landfast Ice Thickness and Duration in the Canadian Arctic. Journal of Climate, Vol. 19 Issue 20 pp. 5175-5189.
Climate change models suggest that much of the Arctic Ocean may be ice free in summer by the end of this century; however, global scale models are unable to resolve the Canadian Arctic Archipelago. Dumas et al., use a one-dimensional landfast ice model to "downscale" for ten locations within the Canadian Arctic Archipelago. Canadian Ice Service provided data for the 1959-1990 period and the CGCM2 projections were used to drive the landfast ice model. Significantly, scenarios for 2041-2060 and 2081-2100 show a decline of 30 and 50 cm in maximum sea ice thickness and a 1 and 2 month reduction of ice cover duration. The projected changes in freeze-up dates vary more regionally than the break-up dates probably due to more heat being stored in the ocean during the longer open water season. (Ice freeze-up is defined as the day the ice reaches a thickness of 30 cm while break-up is the day the ice thickness decreases to 50 cm.) The paper demonstrates the feasibility of developing physically-based location-specific scenarios for future ice conditions in the Canadian arctic to support adaptation decision making.

Fyke, J.G. and Weaver, A.J. 2006. The effect of potential future climate change on the marine methane hydrate stability zone. J. Climate 19:5903-5917.
In this study, researchers apply results of climate change simulations with the University of Victoria earth system model to a simple model of ocean floor hydrate stability. Results show that, even with the increase in CO2 concentrations already realized to date, ocean floor hydrates may experience a significant decrease in volume in coming centuries. If CO2 concentrations rise to levels similar to those projected under IPCC SRES scenarios, large decreases in hydrate volumes begin to occur after 200 years, although it takes some 5000 years for most of the losses to occur before the system slowly returns to an equilibrium. Greatest losses in the first few centuries occur in shallow water areas where heat penetrates to the ocean floor quickly. Assuming that none of the released methane is absorbed within the ocean and all of it is converted immediately to CO2 upon reaching the ocean surface, the hydrate decay adds a feedback factor on the order of 10%. This simple study suggests that the hydrate feedback may be significant, although not immediate or necessarily catastrophic. However, the study does not investigate the risks related to a sudden pulse of methane (which is a much more potent greenhouse gas) being added to the atmosphere. The average lifetime of methane in the atmosphere before it is converted to CO2 is about a decade.

Hori, M.E., and H. Ueda. 2006. Impact of global warming on the East Asian winter monsoon as realized by nine coupled atmospheric-ocean GCMs. Geophys. Res. Lett., Vol. 33, doi:10 1029/2005GL024961, 2006.
The impact of global warming on the East Asian winter monsoon (EAWM) was studied using nine coupled atmospheric-ocean general circulation models. All the models started their integration from the "20th Century Climate in Coupled Model" (20C3M) run, which is based on historical data of the late 19th century though the 20th century. From the end of the 20C3M run, SRES-A1B conditions were imposed until 2100 then concentrations were fixed until 2200. The authors felt the SRES-A1B condition best fit the rapid economic and population growth in East Asia. They found that most models showed a weakening of the EAWM complemented by a strong anticyclonic anomaly over the North Pacific. That latter is linked to a northward shift in the Aleutian low-pressure area. Possible explanations for the changes are the weakening of the tropical local Hadley circulation in the model that leads to a substantially weakened East Asian Jet and the resultant EAWM. Other interconnectivities considered were changes in modelled SST (which contributed to a weaker Walker circulation) and the negative sea level pressure anomaly over Eurasia as a result of weaker Siberian high-pressure area and Aleutian low. Further research is needed on the influence of boundary conditions such as snow and air-sea coupling in the SST distribution.

Keenlyside, N.S., Latif, M., Jungclaus, J., Kornbleuh, L. and Roeckner. 2008. Advancing decadal-scale climate prediction in the North Atlantic sector. Nature 453:84-88; Wood, R. 2008. Natural ups and downs. Nature 453: 45; Kerr, R.A. 2008. Mother nature cools the greenhouse, but hotter times still lie ahead. Science 320:595.
A new study indicates that models may now be able to predict the combined effect of radiative forcing and natural climate oscillations with some skill for one decade into the future. Results suggest that a cooler phase in the North Atlantic oscillation will offset, and possibly exceed global warming effects for the next decade within the North Atlantic sector, but then reverse in subsequent decades. These results remain uncertain.
Past observational data and model studies have indicated that the long-term rise in global surface temperatures during the past century has been modulated by natural climate variability. Inter-decadal temperature fluctuations are usually associated with ocean circulation oscillations such as El-Nino cycles, Pacific Decadal Oscillations and North Atlantic Oscillations. As a result, average temperatures can rise rapidly during one or more decades and much more slowly -even cool - the next. This variability is particularly evident at the regional scale. However, until recently, model simulations have been unable to predict the timing of such fluctuations, and hence how these will modulate future global warming from one decade to the next has remained uncertain. A recent study by a German team of climate modelers, published in Nature, suggests that it may now be possible to develop such predictions with some skill for at least one decade into the future. In a series of simulations of climates since 1860, they make adjustments to ocean surface temperatures at the first year of the simulation run to make them agree with observations, thus allowing the runs to start with the correct ocean conditions. They demonstrate that this simple adjustment at the start of the run improves the skill of decade to decade predictability of ocean temperature variations in most regions of the world. They then extend the simulations to 2030, using the A1B forcing scenario beyond 2000. While the long term trend to 2030 using this method is similar to that from previous climate projection studies, the shorter term projection differs. During the next decade to 2015, the simulations using adjusted starting conditions suggest that a weaker phase in the variations of North Atlantic circulation will cause a short-term regional cooling effect that more than offsets the expected warming from greenhouse gas forcing during the same time period. This regional cooling may be enough to result in minimal increase in global average temperatures during the decade. However, others note that, while this improvement in decadal climate prediction skills is encouraging, the methods used are still very simplistic, and that another similar study recently completed by a UK team show different results. Therefore, these pioneering results must still be viewed with caution.

Kharin, V.V., F.W. Zwiers, X. Zhang and G.C. Hegerl (2007). Changes in Temperature and Precipitation Extremes in the IPCC Ensemble of Global Coupled Model Simulations, Journal of Climate, Vol. 20, pp. 1419-1444.
Climate Research Branch scientists publish results of an analysis of how extreme temperature and precipitation events may change with global warming. For North America, extreme precipitation events occurring now once every 20 years are projected to occur once every 12-13 years by mid-century and once every 8-9 years near the end of the century. Waiting times for extreme 20-year warm events are reduced by a factor of 2-4 over mid-latitude land areas by mid-century. In the lower latitudes, such events are projected to occur virtually every year by around mid-century.
Three CRB scientists and an American colleague have published the results of an evaluation of global coupled climate models - those participating in the IPCC AR4 - in simulating and projecting temperature and precipitation extremes under current conditions and under different global warming scenarios. Increased frequency of extreme temperature and precipitation events has long been considered a robust consequence of global warming. This paper quantifies how those frequencies can be expected to change. Extremes are evaluated in terms of the 20-year event for three time periods: 1981-2000 ('current climate'), 2046-2065, and 2081-2100. Overall, the climate models simulate present-day warm extremes reasonably well on a global scale, while cold extremes, particularly over sea-ice covered areas, proved more difficult to capture. Current 20-yr precipitation events are captured reasonably well by the coupled models outside the tropics. In the future, what is now considered a 20-yr warm event is projected to become a 10-yr event over high latitudes and a 5-yr event over more moderate latitudes over land by mid-century. Late 20th century warm extremes are exceeded virtually every year mid-century in the lower latitudes. Cold extremes warm faster than the warm extremes by about 30% over Europe and up to 50% over North America, with the greatest response over the Arctic. Waiting times for current 20-year extreme precipitation events are reduced almost everywhere around the globe. Waiting times in the tropics, and mid-high latitudes, are reduced by a factor of about 2 mid-century and by a factor of 3 later in the century with GHG emission scenarios that range from moderate to high.

Knutson, T.R., J.J. Sirutis, S.T. Garner, G.A. Vecchi & I.M. Held. 2008. Simulated reduction in Atlantic hurricane frequency under twenty-first century warming conditions. Nature Geoscience 18 May 2008; doi:10.1038/ngeo202.
A new regional modelling study shows a reduction in Atlantic hurricane frequency in a warmer climate.
A recent study addresses the issue of how Atlantic hurricane activity will change in a warmer world by using a new regional modelling framework to downscale GCM projections. Downscaling techniques, such as this, are necessary since the resolution of GCMs is too coarse to realistically model small-scale features such as tropical cyclones. With this regional model, Knutson et al. are able to reproduce the observed rise in Atlantic hurricanes over the past 25 years, although simulated hurricanes are not as intense as those observed. In a warmer future, the model generates substantially fewer tropical storms (-27%) and hurricanes (-18%). This is apparently due to higher levels of wind shear and other changes in cyclone development processes. These model results suggest that the observed rise in Atlantic hurricanes is not due to absolute increases in sea surface temperature in the Atlantic, but rather due to the warming of the tropical Atlantic relative to other tropical basins. They also showed a very small increase in the intensity of storms, and increased rainfall associated with storms, agreeing with conclusions from the IPCC's Fourth Assessment Report.

Kurz, W.A., Stinson, G. and Rampley, G. 2007. Could increased boreal forest ecosystem productivity offset carbon losses from increased disturbances? Phil. Trans. R. Soc. B. doi:10.1098/rstb.2007.2198.
Research by Canadian Forestry Service scientists projects that the Canadian boreal forest is likely to become a net carbon source in the future.
Canadian Forestry Service researcher Werner Kurz and colleagues use the latest version of the Carbon Budget Model of the Canadian Forest Sector (CBM-CFS3) to simulate rate of change in disturbance, growth and decomposition of a hypothetical boreal forest landscape and thus explore the implications of future environmental change for net carbon fluxes. Model results suggest that significant increases in Net Ecosystem Productivity (NEP), where increases in net primary productivity would significantly exceed concurrent increases in respiration through decomposition, would be required to balance carbon losses from the ecosystems due to natural disturbances. Current projections suggest that the average area of Canadian forests burned annually by wildfire will increase significantly in coming decades as the climate changes. The hypothetical CBM simulations suggest that, if this fire loss rises by 20%, an increase in NEP of 12% would be required in order to maintain a net flux balance. The authors suggest that this is highly unlikely because the rates of both photosynthesis and respiration are expected to increase significantly. Hence, if disturbance rates increase as projected, the Canadian boreal forest is likely to become a net source of CO2.

Lambert, S., J.C. Fyfe, 2006. Changes in winter cyclone frequencies and strengths simulated in enhanced greenhouse warming experiments: results from the models participating in the IPCC diagnostic exercise. Climate Dynamics, DOI 10.1007/s00382-006-0110-3.
Two of Environment Canada's scientists use an array of climate models to determine the frequency and strength of simulated winter storms over the mid-latitudes for enhanced greenhouse gases scenarios. Twelve different models were used to simulate the climate under three different SRES scenarios. As storms are not directly simulated in climate models, daily averaged mean sea level pressure was used as a proxy. The authors found a consensus among the model results which is for fewer storms over all, but a greater number of intense storms.

Li, W., Fu, R. and Dickinson, R. E. 2006. Rainfall and its seasonality over the Amazon in the 21st century as assessed by the coupled models for the IPCCAR4. JGR. Vol. 111, D02111, doi:10/1029/2005JD006355, 2006.
Li et al., use all available climate models for the IPCC AR4 to predict the changes of Amazon rainfall in the 21st century. Under the SRES A1B scenario, five of these eleven models predict an increase of annual rainfall, while three of them predict a decrease of rainfall; and the other three predict no significant changes in the Amazon rainfall. To further understand the causes of these convoluted results, the authors further analyze the projections from two models (UKMO-HadCM3 and GISS-ER), which are among the best in duplicating the past rainfall pattern of the Amazon. UKMO-HadCM3 model predicts decreasing rainfall during the wet season, due to an El-Niño type sea-surface temperature (SST) change and a warmer northern tropical Atlantic, will lead to a drier Amazon in the 21st century. On the other hand, GISS-ER predicts a weaker SST warming in the western Pacific and the southern tropical Atlantic, which in turn will increase moisture transport and rainfall of the Amazon. The findings are significant in that the majority of models did not show a decrease in rainfall in the Amazon bringing into question the implications of a 2004 Nature paper by Cox et al., that projected future decreases in rainfall leading to a positive feedback on GHG emissions due to an abrupt dieback of the Amazon forest.

Liao, Hong et al., 2006. Role of climate change in global predictions of future tropospheric ozone and aerosols. JGR Vol 111, D12304, doi:10.1029/2005JD006852.
In this paper, the authors describe the results of model runs with a fully coupled chemistry-aerosol-climate GCM within the Goddard Institute for Space Studies GCM II. Simulations were done with various combinations of year 2000 air pollutant levels, 2000 CO2 levels, 2100 SRES A2 air pollutant emissions and a 2100 doubling of CO2. Doubling CO2 led to a temperature increase of 4.8 ºC by 2100, specific humidity increases of up to 3-4.5 g H20/kg air, an increase in global average precipitation of +0.32 mm/day and alterations of wind patterns. These physical changes in turn affected the chemistry of tropospheric ozone and aerosols. In the scenario of doubled CO2 but unchanged pollutant emissions, the authors observed increased ozone levels over heavily populated areas and areas of extensive biomass burning (largely due to increases in biogenic hydrocarbons for the latter category); lower primary organic aerosol concentrations at mid and high latitudes because of increased precipitation; decreases in sulphate aerosols, except in the tropics and in highly polluted regions; a significant temperature-related decrease in nitrate levels; and a reduction in the global burden of dust, largely related to increased precipitation. With A2 projected emissions and CO2 doubling, the 2100 atmosphere showed the following characteristics: larger increases in ozone over populated and biomass burning areas; and increases of all anthropogenic aerosols except sulphates (lower emissions and higher wet deposition). Radiative forcings for the air pollutants in question were calculated for both the simulation of today's climate/A2's pollutant emissions and the simulation of doubled CO2 climate/A2's pollutant emissions to estimate the effect of climate change alone. Some pollutants showed essentially no change in radiative forcing (e.g. organic carbon) while others changed by up to +0.31 w/m2m (in the case of nitrate). The results show that while climate change can reduce the global burden of ozone, it can actually exacerbate the situation in highly polluted areas. In the scenario with A2 (high) emissions and doubled CO2, surface-layer O3 levels increased over populated and biomass burning areas by 30-70 ppb (in comparison, the current Canada-wide Standard for ozone is 65 ppb).

Lobell, D.B., G. Bala, C. Bonfils and P.B. Duffy, July 14, 2006. Potential bias of model projected greenhouse warming in irrigated regions. Geophysical Research Letters 33(13) doi:10.1029/2006GL026770.
In this paper the authors attempted to estimate the impact of irrigation schemes on the accuracy of global- and regional-scale climate models. Changes in irrigation have been shown to have significant effects on local climate, and irrigation has not been accounted for in GCM predictions, especially where warming is shown to reduce soil moisture and thereby increase temperature. Since the drying of agricultural soils with increased temperature reduces the amount of heat energy absorbed by the evaporation of soil moisture, it thus contributes a positive feedback to warming. In this study, four 50-year simulations were run on a coupled model consisting of the Community Atmosphere Model and the Community Land Model. Two of these examined the impacts of a doubling of atmospheric CO2 under the models' default soil moisture conditions, while the other two examined the effect of the same doubling with all of the world's agricultural areas maintained with a soil saturation moisture percentage. Full saturation of all agricultural lands worldwide is acknowledged by the authors as an unrealistic extreme case which they say is nevertheless useful to illustrate the effect in question. The reduction of the soil moisture feedback in the saturated soil scenario reduced the magnitude of global temperature increase over 50 years from 2.2 º C to 2.1 º C, with no impact on precipitation. While many agricultural regions showed little difference in climate response with soil saturation, others experienced warming that was 0.5 º C (southern India and Southwest Asia) and 1.0 ºC (much of Europe) less. The possible increase of warming by atmospheric water vapour due to irrigation was found by the investigators to be negligible. These results suggest that current regional-scale climate models may overestimate global warming in irrigated agricultural areas with strong simulated soil moisture feedbacks, and that irrigation may mitigate the climate response to increased GHG concentrations. More realistic scenarios are needed to evaluate these effects with greater accuracy.

Mailhot, A., S. Duchesne, D. Caya and G. Talbot (2007). Assessment of future change in intensity-duration-frequency (IDF) curves for Southern Quebec using the Canadian Regional Climate Model (CRCM), Journal of Hydrology, 347, 197-210.
Southern Quebec could see more frequent and intense extreme rainfall events between May and October in a future climate.
In the first Canadian study of this kind, scientists have looked at the changes in sub-daily extreme precipitation events in Southern Quebec between the current (1961-1990) and a future climate (2041-2070). To do so, the investigators used the Canadian Regional Climate Model (CRCM_3.7.1) driven by a Canadian global coupled climate model (CGCM2) simulation with SRES-A2 emission forcing. The study considers the May to October annual maximum (MOAM) rainfall for 2, 6, 12 and 24 hour durations, for different return periods (2, 5, 10, 25 and 50 years). The authors first compared the statistical characteristics of observed rainfall records at 51 stations with those simulated for current climate by the CRCM. They then compared these with rainfall characteristics for the future climate scenario. Statistical analyses were done at individual grid boxes (CRCM) and stations, but also at a regional level (area averages of many stations and integrating many grid boxes). A first set of results at the grid box scale shows that for extreme rainfalls of short duration (2 and 6 hours), return periods will be reduced approximately 50% in the future climate, while they will decrease by a third for longer duration events (12 and 24 hours). In addition, on a regional basis, future rainfall amounts are systematically higher than their corresponding values for the current climate. This is an indication towards more intense heavy rainfalls in a future climate over Southern Quebec. The results also suggest that systems generating heavy precipitation for given durations will be more localized in a future climate and presumably more convective in nature, even for the 24 hour events. However, the simulation of daily extreme precipitation can be highly dependent on the parameterization (formulation) of a RCM. Thus, the authors note that further simulations using multi-model ensembles will have to be done to better estimate the future changes in rainfall characteristics.

Meehl, G.A. and Teng, H. 2007. Multi-model changes in El Niño teleconnections over North America in a future warmer world. Climate Dynamics 29:779-790.
Future El Niños may cause weaker warming anomalies over the Canadian Arctic, but greater cooling anomalies over low latitudes of North America.
Past climate model studies into the response of El Niño behaviour to enhanced greenhouse gas forcing show varied results, with some models suggesting a more intense El Niño amplitude and others a weaker amplitude. Researchers at the American National Center for Atmospheric Research (NCAR) have recently undertaken a new multi-model study to examine the implications of warmer climates on El Niño patterns as well as amplitudes. In their study, they use three models that project weaker El Niño amplitudes under warmer climates, and three with stronger amplitudes. They find that all of these models replicate current El Niño behaviour quite well, including the geographical pattern of El Niño related changes in temperature and pressure. They also find that projected changes under two future scenarios - a moderately high (SRES A1B scenario at 2100) and a very high (4x CO2) scenario - show a consistency in changes of circulation patterns over North America, regardless of whether the model projects a stronger or weaker El Niño amplitude. Results suggest a moderated warming anomaly over the high latitudes of North America (i.e. reduced variability) during future El Niños relative to that during present events, but an intensified El Niño induced cooling anomaly over southern North America. These changes are linked to a north-east shift and weakening of the low pressure system that develops over the north Pacific during El Niño events.

Meehl, G.A., Teng, H. and Branstator, G. 2006. Future changes of El Niño in two global coupled climate models. Climate Dynamics doi 10.1007/s00382-005-0098-0.
Using a variety of future radiative forcing scenarios, Meehl et al., run a series of experiments with the two NCAR coupled climate models (the Parallel Climate Model and the Community Climate System Model version 3) to examine how El Niño behaviour might change under warmer climates. Both models simulate current El Niño behaviour quite well. Results suggest a decrease in mean El Niño intensity as climates warm, although the decreases only become statistically significant relative to control runs once the warming reaches levels expected under 4xCO2 conditions. Authors note, however, that studies with other climate models have produced different El Niño responses. Hence, there is as yet no definitive answer to the question of how El Niños may change in the future.

Meehl, Gerald A., Washington, Warren M., Santer, Benjamin D. et al., 2006. Climate Change Projections for the Twenty-First Century and Climate Change Commitment in the CCSM3. Journal of Climate, 19(11), 2597-2616.
Climate change simulations with the NCAR Community Climate System Model version 3 (CCSM3) are reported on in this paper. The following scenarios were used in the simulations: emission stabilization at year 2000 emissions, and SRES A2, A1B and B1 scenarios. For the A1B and B1 scenarios, climate change over both the 21st century and beyond to year 2300 were simulated. For the latter timeframe, emissions were stabilized at year 2100 levels. The authors find a committed warming of 0.4º C by 2100 based on stabilization at 2000 concentrations, and a much greater relative increase in sea-level. Temperature and sea-level rise for the three SRES scenarios are all greater than the year 2000 stabilization scenario by 2100, with the highest (A2) scenario projecting a mean warming of 3.3º C and a sea-level rise of 31 cm. In all cases, sea-level rise is based exclusively on thermal expansion, ignoring effects from glacier and icecap melting. The 2100-2300 stabilization scenarios show stabilization of temperature but not sea-level, due to the slower response of the oceans to climate change. There is an observed El-Niño-like response in the equatorial Pacific. The authors evaluate impacts of the various scenarios on soil moisture budgets in midlatitudes, and find very little significant summer drying. There is a continuous decrease in polar sea-ice extent and thickness throughout the 21st century for all scenarios, with the A2 scenario resulting in an ice-free Arctic during summer by 2100.

Mikolajewicz, U., Groger, M., Maier-Reimer, E. et al., 2006. Long-term effects of anthropogenic CO2 emissions simulated with a complex earth system model. Climate Dynamics DOI 10.1007/s00382-006-0204-y.
This team of European and American scientists uses an earth system model to assess the implications of future global warming on the global thermohaline circulation system. While coupled climate model simulations have been used for similar assessments in the past, in this case the model includes dynamic ice sheet, marine biochemistry and dynamic vegetation models as well as circulating oceans and atmosphere. This allows the simulation to include slow feedbacks on multiple century time scales that may not be important in simulating climates of the 21st century, but are important on longer time scales. Multiple simulations are conducted for each of SRES B1, A2 and A1B emission scenarios to 2100, followed by exponential reduction in emissions thereafter. Without vegetation and ocean feedbacks, atmospheric concentrations of CO2 under the highest scenario (A2) reach 969ppm, about 3.5 times pre-industrial levels. With feedbacks, this reaches 1416ppm (5x pre-industrial CO2). Under A2, the global thermohaline circulation shuts down completely by 2300, and does not recover within the next 5000 years. Despite a global warming of 4.9 C, the North Atlantic cools. For B1 (peak CO2 concentration of 416ppmv), it slows down and slowly recovers. Under A1B (concentration of 665ppmv), three of five simulations cause a THC shutdown, suggesting this scenario may be the tipping point for such collapse. The cooling in the north Atlantic also causes a reduced melting of the Greenland ice sheet, and related contributions to sea level rise are largely offset by increased ice accumulation in Antarctica. Hence, while these simulations suggest that a collapse of the thermohaline circulation is unlikely within the next century, emissions to 2100 at the upper range of those proposed by IPCC may make such surprises unavoidable in subsequent centuries.

Murazaki, K., and P. Hess (2006) How does climate change contribute to surface ozone change over the United States?, J. Geophys. Res., 111, D05301, doi:10.1029/2005JD005873.
The purpose of this study was to examine how climate change affects surface ozone levels and therefore, changes in U.S. air quality. Two 10 year periods (1990 to 2000 and 2090 to 2100) were simulated using a global chemical transport model MOZART 2. In each case, MOZART 2 was driven by meteorology from NCAR's coupled Climate System Model (CSM) forced with IPCC Special Report on Emissions Scenarios (SRES) A1 scenario. During both periods the chemical emissions were kept constant at the 1990 levels so that only changes in climate under IPCC A1 scenario were allowed to impact the ozone levels. Results showed that the response of ozone to climate change was not the same in polluted areas as in remote regions. The model indicated a general decrease in background ozone levels across the country of 0 to 2 ppbv by the end of the century while in polluted areas, increases in locally produced surface ozone of up to 6 ppbv were projected. Over the Western U.S., the decrease in background ozone approximately cancels the increase in locally produced ozone. However, in the Northeast U.S., future regional increases of up to 5 ppbv resulted in up to 12 additional days each year exceeding the max 8-hour average ozone limit of 80 ppbv. Various climatic factors were identified that impact upon the net future increase in surface ozone over the U.S. including changes in temperature, water vapour, clouds, transport and lightning NOx. This study only considered stable emissions at the 1990 levels, which the authors feel are unrealistic as future demands for energy will increase along with biogenic emissions in an increasingly industrial World. Therefore, despite some uncertainty about the impact of regional climate change on air quality in the U.S., the authors feel that deterioration in air quality in the future is a robust result.

Perlwitz, J., S. Pawson, R. Fogt, J.E. Nielsen and W. D. Neff. 2008. Impact of stratospheric ozone hole recovery on Antarctic climate. GRL 35, L08714, doi:10,1029/2008GL033317.
A new study into the significant effects of stratospheric ozone recovery on future Antarctic tropospheric circulation underscores the importance of including changes in stratospheric ozone concentrations in model simulations of human induced climate change.
The South Hemisphere polar climate has undergone significant change over the past three decades due to the formation each spring of the ozone hole and the associated cooling in the lower stratosphere. The result has been a one month delay in the breakdown of the winter polar vortex, a strengthening of the austral summer circumpolar westerlies, a surface cooling over the Antarctic interior and warming over the Antarctic Peninsula. This study investigated how a recovery of the ozone layer might modify the effects of increases in GHGs on future Antarctic climate. To do so, the investigators used a version of the Goddard Earth Observing System Chemistry-Climate Model that includes radiative coupling between predicted effects of ozone changes in the middle stratosphere and of increases in GHG concentrations. The model simulation indicates that the tropospheric circulation change that has occurred since the 1970s as a result of ozone depletion - the strengthening of the circumpolar westerlies during austral summer - is reversed in the middle of this century. This recovery of the ozone layer in the model leads to a warming of the polar stratosphere during the spring, weaker westerly zonal winds in the stratosphere during the late spring and early summer and in the troposphere during summer. These results were significantly different from those obtained from the IPCC AR4 models, which generally fail to include future changes in stratospheric ozone. These results underscore the importance of a more complete representation of the middle atmosphere in modeling the stratospheric ozone recovery for Antarctic climate.

Raisanen, J., 2008. Warmer Climate: less or more snow? Clim. Dyn., 30, 307-319.
A study of northern hemisphere snow amounts, over the 21st century, projects a decrease in length of snow season, and an increase in mid-season snow amounts in the coldest northern regions. Conversely, the midlatitudes show a decrease in snow volume throughout the winter. Projections for changes in peak winter snow amounts in the region between the cold north and warmer midlatitudes show little agreement among models.
An interesting question is what will happen to snowfall amounts as the planet warms. The amount of snow accumulation is important for many sectors including farming, power generation, and municipal water supply management, so understanding how this resource will change in time is very important. In this study, simulations from 20 climate models, using the SRES A1B (moderate) emission scenario, are used to examine snow water equivalence amounts for the northern hemisphere. The most obvious result is that the snow season shortens everywhere over the 21st century. Winter precipitation amounts generally increased over the northern hemisphere. In the northern cold regions, north of the winter -20°C isotherm line, this means an increase in snow amounts in the height of winter. The results for the midlatitude region showed a robust finding of less snow throughout the winter season. In the areas that fall in-between the northern-most and midlatitude regions, models showed conflicting results, indicating sensitivity to the projected temperatures. More work is required to better understand and resolve disagreement between the various models in this transitional region.

Ramanathan, V. and Y. Feng, 2008. On avoiding dangerous anthropogenic interference with the climate system: formidable challenges ahead. PNAS, 105 (38): 14245-14250.
Study shows that if the cooling effect of aerosols is eliminated, fixing atmospheric concentrations of GHGs at current levels still commits the world to a global warming of 2.4°C above pre-industrial (with a range between 1.4-4.3°C), a rise in temperature that many would consider 'dangerous'.
In this recent article, the authors argue that, when considering the climate change 'commitment', it is more appropriate to consider only the effect of greenhouse gases, most of which are long-lived. They support this argument by saying that the negative forcing from anthropogenic aerosols is at best a masking effect of GHG warming. An unmasking of the underlying warming will occur rapidly with the implementation of regulations to reduce air pollution. The authors use a scenario where GHGs concentrations remain constant at 2005 levels for the next century. They use the combined GHG forcing for long-lived GHGs and ozone from the IPCC AR4 of 3 W/m2 (2.6 to 3.5) along with the IPCC-AR4 estimate of equilibrium climate sensitivity of 3°C (2°C to 4.5°C) to determine the committed global warming. No other anthropogenic forcings are considered. They conclude that the increase in GHGs from pre-industrial levels has already committed the planet to a global warming of 2.4°C (range of 1.4-4.3°C), of which about 90 percent will most likely be experienced in the 21st century. This is much higher than the IPCC projected committed warming of about 1.3°C from pre-industrial to end of the century since that estimate was based on fixing all forcings, including those from aerosols, at year 2000 levels. In their paper, the authors note that proposed reductions in emissions of CO2 and other GHGs will have no impact on the estimated 2.4°C committed warming or delay the time for realizing it. This means that future efforts to reduce emissions of GHGs can only limit further additions to the committed warming but cannot reduce it. The authors remark that their work makes even clearer the urgency for aggressive GHG mitigation given the magnitude of the committed warming in light of perceived thresholds for dangerous anthropogenic climate change. The authors point towards reductions in emissions of short-lived warming agents like black carbon and ozone as important options for reducing the warming commitment.

Raper, S.B. and Braithwaite, R.J. 2006. Low sea-level rise projections from mountain glaciers and icecaps under global warming. Nature 19 January, 2006, pp311-313. doi:10.1038.
The IPCC Third Assessment Report (TAR) provided an estimated range for global average sea-level rise (SLR) of about 10-90 cm by 2100, based on a range of IPCC SRES scenarios and a range of climate models. For scenario A1B, a mid-range scenario, the mean global average SLR from a number of models was projected to be about 38.7cm by 2100 (see Appendix II IPCC TAR). Thermal expansion is expected to make the largest contribution (28.8cm), while the contribution from melting of land ice (outside Greenland and Antarctica) was estimated at 10.6cm. The paper by Raper and Braithwaite provides new, much lower estimates for the contribution of melting land ice to future SLR. Using the A1B scenario to drive future climate change with two GCMs (GFDL and PCM), and a new method for determining ice volume, the authors estimate the contribution from melting land ice (excluding Greenland and Antarctica) to be about 5 cm (4.6 cm and 5.1 cm for GFDL and PCM models respectively) or about half of the previous IPCC estimate for the same emission scenario. It should be noted that the IPCC estimate of 10.6 cm SLR from melting land ice under the A1B emission scenario is an average result from many models whereas Raper and Braithwaite present results for two models independently.

Robock, Alan and Li, Haibin. 2006. Solar dimming and CO2 effects on soil moisture trends. Geophysical Research Letters 33, L20708, doi: 10.1029/2006GL027585.
Summer soil moisture increased significantly from 1958 to the mid 1990s in Ukraine and Russia, a trend that cannot be explained by changes in precipitation and temperature alone. The authors investigate the potential contribution of solar dimming and increasing carbon dioxide to this trend using a state-of-the-art land surface model (a modified version of the Community Land Model 3.0). When a slow dimming was applied to the model experiment, using a 0.5% per year decrease in solar insolation between 1961 and 1980 (and 1960 level concentrations of atmospheric CO2), results projected a 5% reduction in evapotranspiration for the Ukraine, and 9% reduction for Russia. This increased to 16% and 20%, respectively, when the rate of dimming was increased to -1% per year. Repeating these experiments using rising CO2 concentrations as observed since 1960 produced results that only slightly reduced evapotranspiration relative to that for constant CO2 levels. Comparison with observed changes in soil moisture suggest that the best fit occurs for experiments with enhanced solar dimming. While the CO2 fertilization effects seems to have been very small for this region, authors note that it could be much more significant in regions where evapotranspiration is composed of primarily transpiration (e.g., Amazon rainforest).

Scibek, J. and Allen, D.M. 2006. Comparing modeled responses of two high-permeability unconfined aquifers to predicted climate change. Global and Planetary Change 50: 50-62.
In this study, Scibek and Allen used a 3-D groundwater flow model (MODFLOW V3.1.0) under the IPCC IS92a (Greenhouse Gases + Aerosols) climate change scenario, provided by Canadian Global Climate Model 1 (CGCM1), to simulate the impacts of climate change on groundwater systems. Two small, regional, unconfined aquifers, which are in southern British Columbia, Canada and Washington State, U.S.A., were studied in this experiment. The experiment showed observable, but small, changes in these two groundwater systems under modest climate change, which is within the realm of model uncertainty. Of the two systems, stronger impacts were projected for the aquifer that is connected to the Kettle River, because water levels within that groundwater system respond more directly to shifts in peak river flow due to climate change.

Sheffield, J. and E.F. Wood. 2008. Projected changes in drought occurrence under future global warming from multi-model, multi-scenario, IPCC AR4 simulations. Climate Dynamics 31: 79-105.
Global drying of soils is projected to occur by the end of the 21st century, no matter which IPCC emission scenario is considered, leading to more frequent droughts. However, for most regions of the world, these droughts won't show statistically significant changes for several decades.
The prospect of drought conditions becoming more severe in the future is an impact of key concern related to global warming. Recently, two American scientists carried out a large study on global and regional drought occurrence, analyzing soil moisture data for three future climate scenarios (IPCC SRES B1, A1B and A2) from eight AOGCMs that participated in the IPCC AR4. Results were compared with current climate observations and simulations (1961-1990). All future scenarios produce decreases in soil moisture at global scales, with a corresponding doubling of the spatial extent of severe soil moisture deficits and frequency of short-term (4-6 months duration) droughts from the mid-20th century to the end of the 21st. By the end of the 21st century, long-term droughts (12 months and more) become three times more common, but the changes are less spatially extensive. In general, changes under the B1 scenario (lowest emission scenario) are the least and the A1B and A2 results are similar. Regionally, the Mediterranean, west African, central Asian and central North American regions are the most affected, with notable increases in the frequency of long-term drought. Projections for the mid-latitude region of North America (including southern Canada), also show increases in aridity, but with larger variation between scenarios. However, in general, the projected changes in drought are not statistically different from contemporary climate (1961-1990) or natural variability for several decades, and the effects become apparent only around the mid-21st century. On the other hand, changes in annual and seasonal means of terrestrial hydrological variables, such as evaporation and soil moisture are essentially not detectable within the 21st century. Therefore, changes in the extremes of climate and their hydrological impacts may be more detectable than changes in their means. Their results also show that in most places, increases in drought are primarily driven by reductions in precipitation with increased evaporation from higher temperatures modulating the changes. All these results are consistent with a number of earlier studies.

Shukla, J., T. DelSole, M. Fennessy, J. Kinter and D. Paolino. 2006. Climate model fidelity and projections of climate change. GRL vol. 33, L07702, 4 pages.
This study measures 13 coupled global climate models' fidelity in simulating the present climate (surface air temperature during the past 100 years). Here, the fidelity of a model is measured in terms of 'relative entropy', which is a mathematical measure of the closeness of two probability distributions (in this case, simulated and observed climate). This measure was calculated for seasonal and annual cycles. In a second step, the study also computes the sensitivity to changing greenhouse gases for the 13 models, by taking the difference between two emission scenarios: A1B for 2XCO2 (relative to current concentrations) and an "observed" time series of CO2 for the past 100 years. Looking at the results for the two parts of the study, the authors find that models that have higher fidelity in simulating the present climate are more sensitive and produce higher estimates of global warming for a doubling of CO2 than models with lower fidelity. The authors conclude from these results that actual warming over the next century is likely to be closer to the high end of the projected range from the current generation of GCMs.

Sjoukje, P. and G.J. van Oldenborgh. 2006. Shifts in ENSO coupling processes under global warming. GRL Vol. 33, L11704.
Most models that describe ENSO reasonably well in the current climate show only small changes in its characteristics (period, pattern and amplitude) under a doubled CO2 climate, despite large changes in the mean states of the Pacific Ocean. In this paper, the authors try to explain why the changes in ENSO are so small, by investigating the different couplings and feedback loops between the ENSO characteristics and the mean states of sea surface temperature (SST), zonal wind stress, thermocline depth and mixed layer depth (MLD) of the Pacific ocean. Using six AR4-GCMs shown to have the most realistic description of the mechanisms of ENSO, they simulate the current climate and a warmer climate in the period 2200-2300. They find that the most important mechanisms that affect El-Niño are the SST response to thermocline and wind variabilities, but also to damping (cloud feedback), and the wind response to SST perturbations. The most important result of the study is that feedback loops between SST, wind stress and thermocline do show changes, in the same direction, in a warmer climate, following changes in the mean state. On the other hand, the higher mean SST provides higher damping through cloud feedback. All these changes do have large impacts on the ENSO characteristics, but because they have opposing signs, the residual change is almost zero.

Smith, D.M., S. Cusack, A.W. Colman, C.K. Folland, G.R. Harris and J.M. Murphy. 2007. Improved surface temperature prediction for the coming decade from a global climate model. Science 10 August 2007 Vol 317 pp796-799.
A team of British researchers issues the first GCM-based forecast of global surface temperature for the coming decade using a model that is initialized with observed atmosphere and ocean conditions to better capture internal climate variability.
Global climate models provide transient simulations of climate change over the coming century and beyond and so do provide some information on the magnitude of global and regional warming over the next few decades as well as in the longer term. However, the initial state of the atmosphere and the ocean in such models is not based on observed conditions because to do so is a costly and time-consuming undertaking. A team of researchers led by Doug Smith of the UK Met Office's Hadley Centre has developed a decadal climate prediction system that does just this - initializes the climate system by assimilating data on observed conditions, then projects changes in climate over 10 years in response to projected changes in anthropogenic and natural forcing factors. The first results from this new model -the Decadal Climate Prediction System (DePreSys) were published in Science this week. The model is based on the Hadley Centre Coupled Model, version 3 (HadCM3). The researchers first tested the skill of DePreSys by generating a series of hindcasts of 10-year periods over the period 1982-2001. DePreSys hindcasts were significantly more accurate at reproducing surface temperature anomalies at both global and regional scales than were hindcasts generated by the same model but without the initial data assimilation (NoAssim runs). They then provide a climate change projection for the period 2005-2014. With DePreSys, internal climate variability, which is initially based on observed conditions, including cool initial conditions in some parts of the ocean, is shown to offset the effects of anthropogenic forcing in the first few years. Thereafter, strong global warming returns, with the year 2014 predicted to be about 0.30 ± 0.21°C warmer than the observed value for 2004. Furthermore, half of the years from 2009-2014 are predicted to be hotter than the current hottest year on record, 1998.

Sterl, A., C. Severijns, H. Dijkstra, W. Hazeleger, G. Jan van Oldenborgh, M. van den Broeke, G. Burgers, B. van den Hurk, P. J. van Leeuwen and P. van Velthoven, 2008. When can we expect extremely high surface temperatures? Geophysical Research Letters, vol. 35, L14703, doi:10.1029/2008GL034071, 5pp.
Changes in extremely high temperatures (100-year return period) are projected to occur more quickly than changes in average temperature. Temperatures in excess of 50°C in many regions of the world, equatorward of 30°, are projected.
When assessing the consequences of climate change, very high temperatures are much more important than averages temperatures. Earlier studies have shown that cold extremes will warm faster than warm extremes and that warm extremes will warm faster than average temperatures. This study focused on extremely high air temperatures, represented by the 100-year return temperature (T100). (T100 is a specific statistical expression that means that every year you have a 1% chance of getting that temperature.) A 17-member ensemble simulation of climate change in response to the IPCC SRES A1B scenario was carried out using the ECHAM5\MPI-OM climate model. Their results show a projected global-mean temperature increase of 3.5°K by 2100, which is at the upper end of the range given by the models analyzed in the IPCC-AR4. The authors acknowledge that the present generation of climate models, including the one used in this study, tends to overestimate extremes temperature values. However, even after correcting for this bias, they found that by 2090-2100, projected T100 far exceeds 40°C in Southern Europe and the U.S. Midwest and even reaches 50°C in large parts of the area equatorward of 30°, notably in India and the middle East, and also in most of Australia. The projected T100 values, the authors note, should be taken seriously, since they indicate that potential for dangerously high future temperatures in densely populated areas.

Stott, P.A., J.F.B. Mitchell, M.R. Allen, T.L. Delworth, J.M. Gregory, G.A. Meehl and B.D. Santer (2006), Observational constraints on past attributable warming and predictions of future global warming. J. of Climate, vol.19, pp.3055-3069.
Using three coupled global climate models (CGCM) with different sensitivities run with a range of natural and anthropogenic forcings, the authors investigate the impact of aerosol (sulphate) forcing uncertainty on the robustness of estimates of the 20th century warming attributable to anthropogenic greenhouse gas (GHG) emissions. Applying an optimal detection analysis, they also investigate whether there is sufficient information in observations of past near-surface temperature change to constrain predictions, whichever model is used, and, they investigate which indices of large scale temperature are responsible for discriminating the climate response between models.
Their results show that all three models have good simulations of global mean temperature changes during the 20th century, when they include both anthropogenic and natural forcings. Solar and volcanic forcings make a larger contribution relative to the total forcing in the early 20th century warming, whereas anthropogenic forcing is largely responsible for the warming observed in the last three decades. They note that the models differ much more in the projections of future warming rates than in their simulations of past temperature change. They find that the observationally constrained predictions for the three models are in much better agreement than raw model predictions and much less model dependent. The features that constrain the likely temperature response to anthropogenic and natural forcings are the temporal and spatial structure of the observed global mean temperature changes over the 20th century. Distinctive temporal structures in differential warming rates between the hemispheres, between land and ocean and between mid and low latitudes help to discriminate between models and determine the relative roles of greenhouse warming and sulfate cooling.

Sushama, L., Laprise, R., Caya, D. et al., 2006. Canadian RCM projected climate-change signal and its sensitivity to model errors. Int. J. Climatology 26:2141-2159.
Authors use the standard and updated versions of the third generation Canadian Regional Climate Model (CCRM 3.6 and 3.7) to assess the regional response of a number of major river basins (Fraser, Yukon, MacKenzie, Nelson, Churchill and Mississippi) of western/central North America to future climate change. The models are driven at the boundaries by outputs from the Canadian coupled climate model CGCM2, using IPCC emission scenarios IS92a and A2. Changes in river basin hydrology as projected by the two RCMs for 2041-2070 agree in sign but not magnitude. Rainfall increases by 2-14% in all basins except in the Mississippi, and snow cover decreases significantly in all river basins, resulting in lower spring snow melt runoff peaks. While Mississippi runoff decreases, those for the northwestern basins (Fraser, Yukon and Mackenzie) increase significantly, particularly during fall, winter and spring seasons. Number of low flow days below critical thresholds also decrease in all the Canadian basins, although they increase in late-winter/spring.

Sushama, L., R. Laprise, D. Caya, D. Verseghy, M. Allard (2007), An RCM projection of soil thermal and moisture regimes for North American permafrost zones, Geophys. Res. Lett., 34, L20711, doi:10.1029/2007GL031385.
Canadian regional climate model projections indicate that temperatures in the permafrost regions of North America would rise in the 21st century.
The Canadian Regional Climate Model (CRCM4) was used to project changes in soil temperatures and moisture in the permafrost regions of North America. The CRCM4 is a better tool for impact studies than Global models (GCMs) owing to its greater topographical realism and finer scale atmospheric dynamics. The model compared the 2041-2070 period with the 1961-1990 period using the A2 emissions scenario (a world focused on economic growth). Model results are that near surface soil temperatures increased for all four permafrost zones (as classified by the International Permafrost Association) with the largest soil layer increases projected in the region of continuous permafrost (4-6°C) with respect to the 1961-1990 reference period. Accompanying increasing temperatures, the soil frozen water content is projected to decrease while the liquid water content is projected to increase. This projected decrease in frozen soil moisture would lead to increased drainage for all permafrost zones. While RCMS are being used in impact and adaptation studies, model improvements are still needed (e.g. heat storage, deep layer modelling, soil properties data base).

Teng, H., Buja, L.E. and Meehl, G.A. 2006. Twenty-first century climate change commitment from a multi-model ensemble. GRL 33, L07706, doi:10.1029/2005GL024766.
An NCAR team of experts have assessed outputs from 16 global coupled climate models, forced by past changes in atmospheric composition, in order to determine how much additional future change in climate we are already committed to. Although the simulations are extended out to AD 2100, atmospheric composition during the 21st century is maintained at AD 2000 values. By the end of the simulation period, global surface temperatures have stabilized at levels 0.5 +/-0.2 °C above the warming already achieved today. Models that include volcanic aerosol forcing in the historical part of the simulation show less warming to date (because of the temporary cooling by recent volcanoes), and therefore about 20% more committed warming in future decades than those that don't. On average, sea levels rise an additional 10 cm and showing no sign of levelling off by 2100. While these global results are similar to those from past studies, this study also examines regional responses. Authors find significant polar amplification, particularly in the Arctic, where models project committed winter warming by 2100 of between 1 and 3°C. Nine of the 16 models also show an El-Niño like pattern in the simulations, with enhanced precipitation in the equatorial central-eastern Pacific and the Indian Oceans.

Teng, H., Washington, W., Meehl, G., Buja, L., and Strand, G. 2006. Twenty-first century Arctic climate change in the CCSM3 IPCC scenario simulations. Climate Dynamics 26: 601-616 DOI 10.1007/s00382-005-0099-z.
Arctic climate change in the 21st century was simulated using the NCAR Community Climate System Model version 3. The integration started from 1870 after a 300-year control simulation. External climate forcings included solar, volcanic, ozone, GHGs, halocarbons, direct effects of sulfate aerosols, and black carbon aerosols. The simulations from three emission scenarios (A2, A1B, and B1) were analyzed using eight (A1B and B1) or five (A2) ensembles members. The model projects a decline of sea-ice extent in the range of 1.4-3.9% per decade in winter and 4.8-22.2% per decade in summer depending on the forcing scenario. Under the A2 scenario, the Arctic is ice-free by the end of the century during summer whereas under the lower (B1) forcing, this state occurs later. Surface mean temperatures were predicted to increase by 0.7-5oC in winter while in summer, temperatures increased in a range of 4-14oC depending on the scenario. The model indicates that the warming is mainly due to the radiative forcing of greenhouse gases with direct forcing and feedbacks. The wintertime surface air temperature, in particular, was characterized by a strong warming trend and a polar amplification - type of spatial pattern. The Arctic Oscillation, while dominant in the 20th century, plays a secondary role in both surface temperatures and sea-ice concentrations in the 21 century.

Thomas, J. Bracegirdle, William M. Connolley, and John Turner. 2008. Antarctic climate change over the twenty first century. JGR, VOL. 113, D03103, doi:10.1029/2007JD008933, 2008.
Antarctic climate change over the 21st century is explored using a new weighted approach to coupled climate model intercomparison. The results suggest more continental warming and sea ice retreat than previously predicted.
A comprehensive perspective of Antarctic climate change for the 21st century is presented in this paper using the multimodel data set from the World Climate Research Program's Coupled Model Inter-comparison Project Phase 3 (CMIP3). The study applies a weighting scheme to model outputs based on the ability of the model to reproduce the climate of the late 20th century. Models results reported include: increase in magnitude of circumpolar westerlies , with maximum changes (27%) in autumn; a projected rise in surface warming averaged over the continent of
0.34°C/decade with more rapid warming in winter over regions of sea ice retreat around East Antarctica (0.51°C /decade); and a decrease in total sea ice area of 33%. While precipitation scenarios show seasonal and spatial variation, the net precipitation averaged over the continent is projected to increase, with the interior of East Antarctica showing up to a 25% increase. The application of the weighting scheme had consequences for some aspects of Antarctic climate that are particularly important globally. For example, the stronger winds at high latitudes will contribute to changes in the Weddell Sea gyre, which could lead to changes to global ocean circulation. Increased sea ice loss and larger warming along coastal East Antarctica may contribute to sea level rise, as warming can induce break-up of ice shelves. This, in turn, could accelerate ice flow from glaciers. The authors do caution that the ability of a climate model to represent current conditions does not necessarily indicate whether its response to scenario forcing will be realistic.

Trapp, R.J., N.S. Diffenbaugh, H.E. Brooks, M.E. Baldwin, E.D. Robinson, and J.S. Pal. 2007.Changes in severe thunderstorm environment frequency during the 21st century caused by anthropogenically enhanced global radiative forcing. PNAS Vol 104, no. 50, 19719 -19723.
An increase in frequency of severe thunderstorms in the United States during the 21st century is projected due to large scale changes in meteorology associated with human-induced climate change.
Severe thunderstorms over the United States produce high-impact weather such as destructive surface winds, hail, lightning and tornadoes that lead to annual average losses of billions of dollars in property and crops, 25 fatalities, and 285 injuries. This study looked at the possibilities of changes in severe thunderstorm frequency in response to increases in greenhouse gases and the associated changes in global radiative forcing. The authors used global climate models and a high-resolution regional climate model to calculate areas where the convective available potential energy and the vertical wind change or "shear" characteristics would foster the formation of severe thunderstorms. They found a net increase during the late 21st century in the number of days favorable to the formation of conditions for severe thunderstorms development. The primary reason for the increase in frequency is the increase in water vapor within the planetary boundary layer along the coasts of the Gulf of Mexico and Atlantic coastal regions. The increase in water vapor increases the available potential energy by providing more latent energy for thunderstorm development. Although this study did examine two quantitative measures that characterize severe thunderstorm development, further research is needed into the characteristics of storm initiation that may also change with increases in greenhouse gases.

Vavrus, S., Walsh. J., Chapman, W.L., and Portis, D. 2006. The Behaviour of Extreme Cold Air Outbreaks Under Greenhouse Warming. Intn'l Journal of Climatology. 26.1133-1147.
The incidence of extreme cold outbreaks is related to climate variability, yet the relationship is uncertain. The incursion of cold polar air into the mid and lower latitudes is common during the fall and winter and can cause human suffering and harm vegetation. This modeling study investigates the behavior of extreme cold-air outbreaks over the Northern Hemisphere under recent and future climatic conditions. The findings are based on simulations from seven GCM runs for the 20th century and projections for the 21st century using the SRES A1B emission scenario. The results from this study indicate that the frequency of extreme cold-air outbreaks (at least two standard deviations below the local mean wintertime mean temperature) will decline by 50-100% during the 21st century in most of the Northern Hemisphere as the concentration of greenhouse gases in the atmosphere changes and the climates warms. Not all regions will see a decline nor will the intensity of the decline be the same in all regions of the Northern Hemisphere due to the nature of the general circulation changes and internal variability. Areas that are prone to atmosphere blocking events, such as over western North America, are expected to experience relatively small changes.

Vellinga, M. and R.A. Wood. 2008. Impacts of thermohaline circulation shutdown in the twenty-first century, Climatic Change, 91: 43-63.
A new study shows that in the Northern Hemisphere, the cooling effect of a sudden shutdown of the thermohaline circulation would outweigh the warming effect of increasing greenhouse gases for about 100 years. There could also be important impacts on surface temperature, precipitation and sea-level rise.
Vellinga and Wood investigated the impacts of a hypothetical shutdown of the thermohaline circulation (THC) in conjunction with the impacts of global warming. To do so, they analysed modeling experiments in which a sudden THC shutdown was simulated under both global warming (increasing GHG emissions) and pre-industrial conditions, and an experiment in which increasing GHG emissions were associated with a gradual weakening of the THC. The two THC shutdown experiments allowed the authors to investigate if and how the climate response due to THC shutdown depends on the background climate state. For all three experiments, impacts on global and regional temperature, precipitation and sea level were examined. The study finds that the consequence of THC collapse on surface temperature is a cooling of similar magnitude under either pre-industrial or warming conditions, for the first decade after the shutdown. However, there are important regional differences between the two experiments, mainly because of the differences in the sea ice cover in the Northern and Arctic seas. Under a warming climate, the authors note that the THC collapse leads to an average cooling, over the first 10 years, of -1.7°C for Northern Hemisphere land areas. Regionally, the strongest cooling occurs in Western Europe and Scandinavia (-2° to -5°C) where temperature falls below pre-industrial levels. Elsewhere, temperature remains near (1°or less cooler) or above pre-industrial values. In subsequent years, recovery of the THC accelerates the rate of warming, but the surface temperature only catches up with that of the global warming experiment without a THC shutdown by the year 2150. For 10 years after THC shutdown, precipitation is reduced, and snow cover is increased over much of the Northern Hemisphere), which is opposite to that projected when only considering global warming. Also, large regional variations in sea level change are seen. It is worth noting that the experiment of Vellinga and Wood was not intended to represent a plausible scenario, but a means to quantify the potential impacts of such an event on the climate, compared to that of a gradual weakening of the THC under a climate change scenario.

Wentz, F.J, L. Ricciardulli, K. Hilburn and C. Mears (2007), How Much Rain Will Global Warming Brings? Sciencexpress, published on-line May 31, 2007, 8 pages.
Climate models generally project that globally, precipitation will increase with global warming. But by how much? A new study suggests that models may be underestimating the precipitation response to warming.
It is well recognized that global warming will alter the planet's hydrological cycle. The changes in intensity and spatial distribution of rainfall are expected to pose some of the most serious risks associated with climate change. In this study, researchers looked at an apparent inconsistency between observations and climate model simulations of recent changes (1987-2206) in global precipitation. Global climate models generally project that total atmosphere water vapour should increase at a rate of about 7% per degree C of warming. The large increase is due to the fact that warmer air is able to hold more water vapour and observations confirm this projection. However, models also consistently project a more muted response of precipitation and evaporation to warming, of about 1-3% per degree C of warming. Using satellite observations over the ocean, supplemented by land values from the Global Precipitation Climatology Project (GPCP), the authors of this paper assessed recent changes in global precipitation and evaporation and found their data showed precipitation had increased over the past two decades at the same rate as total atmospheric water vapour (i.e. about 7% per degree of warming). The authors then evaluated the source of the discrepancy between observations and models. Mathematically, using a well known formula, they showed that a muted response of precipitation to global warming requires a decrease in global winds, which is what is also shown by the GCMs. In contrast, they note that observations support an increase in winds. The reason for the discrepancy between observational data and the GCMs is not yet clear but does need to be resolved. If global precipitation over the coming century increases at the higher rates supported by this paper, the consequences in terms of the impacts of climate change are likely to be significant. The findings would support the notion of a generally much wetter world, but how the extra rainfall would be distributed is still uncertain.

Winton, Michael (2006), Surface Albedo Feedback Estimates for the AR4 Climate models, Journal of Climate, vol. 19, pp. 359-365.
In this study, a new technique for calculating the surface albedo feedback (SAF) is applied to transient CO2 warming simulations with 12 AR4 global climate models. This technique estimates the impact of a surface albedo change on the surface shortwave budget, and thus on surface air temperature (Ts). Globally, over 80 year runs, the models produce a mean SAF value of 0.3 W m-2 K-1 with a standard deviation of 0.09 W m-2 K-1, which is comparable to but smaller than earlier estimates from equilibrium 2XCO2 runs. The Northern Hemisphere (NH) oceans, Southern Hemisphere (SH) oceans and land areas account respectively for 37%, 39% and 24% of the global SAF, with the land SAF being attributed mostly to the NH. It is interesting to note that while the surface warming pattern and the SAF pattern show significant similarities in the NH, they are uncorrelated in the SH for more than half of the models. These negative SAF contributions have been documented by Murphy (1995) and are possible where the surface albedo increases in spite of global warming, as in the case of a region where there is ice retention due to reduced sea surface salinity and increased ocean stratification. This lack of consistency between the different models suggests that, to resolve differences in the models' SAFs, regional differences in, for example, snow and sea ice changes, will have to be taken into account.

Zhang, X and Walsh, J.E. 2006. Toward a Seasonally Ice-Covered Arctic Ocean: Scenarios from the IPCC AR4 Model Simulations. J. of Climate, vol 19, pp1730-1747.
In this study, Arctic sea ice simulations were generated using the IPCC AR4 models using anthropogenic and natural forcing under the IPCC SRES A1B, A2 and B1 scenarios. The study examined the sea ice simulations for the climate of the 20th century then made climate projections for the 21st century. The multimodel ensembles simulated the climatology well and showed decreasing trends of sea ice area during recent decades consistent with observations. This trend increases throughout the 21st century, however, with the last twenty years of the century in the multimodel ensembles projecting a 21.6% to 33.4% percentage reduction relative to 1979-99. The study also found in the projections that the multi-year ice shrinks away much more rapidly than the total ice area and as a result makes the largest contribution to the entire sea ice area loss. Seasonal ice coverage, which is ice that forms each year during the fall and winter seasons, was calculated to increase. As a consequence of the opposite signed changes of multiyear and seasonal sea ice, seasonal cycles of sea ice coverage are greatly amplified by the models in this century. By the end of the 21st century, the models predict that only seasonal sea ice will cover the Arctic Ocean. The projected changes in multiyear and seasonal ice coverage may have significant implications for the polar energy and hydrological budgets and pathways.

Zhang, Y., Chen, W. and Riseborough, D.W. 2006. Temporal and spatial changes of permafrost in Canada since the end of the Little Ice Age. Journal of Geophysical Research 111, D22103, doi:10.1029/2006JD007284.
Three Natural Resources Canada researchers have simulated the response of permafrost in Canada to changes in climate during the 19th and 20th century using the Northern Ecosystem Soil Temperature (NEST) model. The simulated southern boundary of the permafrost for recent decades is similar to that of the current published map, and the simulated active layer thickness and the depth to permafrost base are comparable to site measurements. The simulated results from the model experiment showed a 5.4% decrease in the total area underlain by permafrost from the 1850s to the 1990s, with two periods of rapid decrease from the 1920s to early 1940s and after the 1950s. The average thickness of the active layer, in the persistent permafrost region over the whole of Canada, showed a simulated increase of 0.65m to 0.87m, a 34% change, while the depth of the permafrost table also deepened by 0.39m. Permafrost degradation also occurred from the bottom-up as the mean depth of the permafrost base decreased by 3m, with most of the change occurring after the 1940s. The authors acknowledge that the next step would be to couple NEST with vegetation dynamics and biogeochemical models.

Date modified: