Climate and climate change in western canada as simulated by the Canadian regional climate model

Abstract

A þrst climate simulation performed with the novel Canadian Regional Climate Model (CRCM) is presented. The CRCM is based on fully elastic non‐hydrostatic þeld equations, which are solved with an efþcient semi‐implicit semi‐Lagrangian (SISL) marching algorithm, and on the parametrization package of subgrid‐scale physical effects of the second‐generation Canadian Global Climate Model (GCMII). Two 5‐year integrations of the CRCM nested with GCMII simulated data as lateral boundary conditions are made for conditions corresponding to current and doubled CO₂ scenarios. For these simulations the CRCM used a grid size of 45 km on a polar‐stereographic projection, 20 scaled‐height levels and a time step of 15 min; the nesting GCMII has a spectral truncation of T32, 10 hybrid‐pressure levels and a time step of 20 min. These simulations serve to document: (1) the suitability of the SISL numerical scheme for regional climate modelling, (2) the use of GCMII physics at much higher resolution than in the nesting model, (3) the ability of the CRCM to add realistic regional‐scale climate information to global model simulations, and (4) the climate of the CRCM compared to that of GCMII under two greenhouse gases (GHG) scenarios.     

Dynamic downscaling of 22-year CFS winter seasonal hindcasts with the UCLA-ETA regional climate model over the United States

This study evaluates the UCLA-ETA regional model’s dynamic downscaling ability to improve the National Center for Environmental Prediction Climate Forecast System (NCEP CFS), winter season predictions over the contiguous United States (US). Spatial distributions and temporal variations of seasonal and monthly precipitation are the main focus. A multi-member ensemble means of 22 winters from 1982 through 2004 are included in the study. CFS over-predicts the precipitation in eastern and western US by as much as 45 and 90 % on average compared to observations, respectively. Dynamic downscaling improves the precipitation hindcasts across the domain, except in the southern States, by substantially reducing the excessive precipitation produced by the CFS. Average precipitation root-mean-square error for CFS and UCLA-ETA are 1.5 and 0.9 mm day^?1, respectively. In addition, downscaling improves the simulation of spatial distribution of snow water equivalent and land surface heat fluxes. Despite these large improvements, the UCLA-ETA’s ability to improve the inter-annual and intra-seasonal precipitation variability is not clear, probably because of the imposed CFS’ lateral boundary conditions. Preliminary analysis of the cause for the large precipitation differences between the models reveals that the CFS appears to underestimate the moisture flux convergence despite producing excessive precipitation amounts. Additionally, the comparison of modeled monthly surface sensible and latent heat fluxes with Global Land Data Assimilation System land data set shows that the CFS incorrectly partitioned most of surface energy into evaporation, unlike the UCLA-ETA. These findings suggest that the downscaling improvements are mostly due to a better representation of land-surface processes by the UCLA-ETA. Sensitivity tests also reveal that higher-resolution topography only played a secondary role in the dynamic downscaling improvement.     

Mid-twenty-first century warm season climate change in the Central United States. Part I: regional and global model predictions

A regional climate model (RCM) constrained by future anomalies averaged from atmosphere–ocean general circulation model (AOGCM) simulations is used to generate mid-twenty-first century climate change predictions at 30-km resolution over the central U.S. The predictions are compared with those from 15 AOGCM and 7 RCM dynamic downscaling simulations to identify common climate change signals. There is strong agreement among the multi-model ensemble in predicting wetter conditions in April and May over the northern Great Plains and drier conditions over the southern Great Plains in June through August for the mid-twenty-first century. Projected changes in extreme daily precipitation are statistically significant over only a limited portion of the central U.S. in the RCM constrained with future anomalies. Projected changes in monthly mean 2-m air temperature are generally consistent across the AOGCM ensemble average, North American Regional Climate Change Assessment Program RCM ensemble average, and RCM constrained with future anomalies, which produce a maximum increase in August of 2.4–2.9 K over the northern and southern Great Plains and Midwest. Changes in extremes in daily 2-m air temperature from the RCM downscaled with anomalies are statistically significant over nearly the entire Great Plains and Midwest and indicate a positive shift in the warm tail of the daily 2-m temperature distribution that is larger than the positive shift in the cold tail.     

Standardized estimates of climate change damages for the United States

Nordhaus (1991), Cline (1992), Fankhauser (1992), and Titus (1992) have published comprehensive estimates of annual climate change damages to the United States in about 2060 that vary from $55 billion to $111 billion ($1990). The estimates are comprehensive because they address market and nonmarket impacts. They based their estimates on different assumptions about the rates of climate change and sea level rise, rates of return on investment, and changes in population and income. In addition, many of the damage estimates, although reported for a 2.5–3.0 °C warming, were based on studies that assumed higher rates of warming. Thus, these studies may have overestimated damages associated with a 2.5–3.0 °C warming. In this paper, the results of these studies were standardized for a 2.5 °C warming, a 50-cm sea level rise, 1990 income and population, and a 4% real rate of return on investments. After standardization, the total damage estimates range from $42.3 billion to $52.8 billion, slightly less than 1% of United States GNP in 1990. Yet, within individual sectors, such as agriculture and electricity, standardized damages differ by more than an order of magnitude. In addition, a significant amount of speculation underlies the damage estimates. Thus, the small range of total standardized damages and apparent agreement about the magnitude of such damages should be interpreted with caution.     

Climate variability and projected change in the western United States: regional downscaling and drought statistics

Climate change in the twenty-first century, projected by a large ensemble average of global coupled models forced by a mid-range (A1B) radiative forcing scenario, is downscaled to Climate Divisions across the western United States. A simple empirical downscaling technique is employed, involving model-projected linear trends in temperature or precipitation superimposed onto a repetition of observed twentieth century interannual variability. This procedure allows the projected trends to be assessed in terms of historical climate variability. The linear trend assumption provides a very close approximation to the time evolution of the ensemble-average climate change, while the imposition of repeated interannual variability is probably conservative. These assumptions are very transparent, so the scenario is simple to understand and can provide a useful baseline assumption for other scenarios that may incorporate more sophisticated empirical or dynamical downscaling techniques. Projected temperature trends in some areas of the western US extend beyond the twentieth century historical range of variability (HRV) of seasonal averages, especially in summer, whereas precipitation trends are relatively much smaller, remaining within the HRV. Temperature and precipitation scenarios are used to generate Division-scale projections of the monthly palmer drought severity index (PDSI) across the western US through the twenty-first century, using the twentieth century as a baseline. The PDSI is a commonly used metric designed to describe drought in terms of the local surface water balance. Consistent with previous studies, the PDSI trends imply that the higher evaporation rates associated with positive temperature trends exacerbate the severity and extent of drought in the semi-arid West. Comparison of twentieth century historical droughts with projected twenty-first century droughts (based on the prescribed repetition of twentieth century interannual variability) shows that the projected trend toward warmer temperatures inhibits recovery from droughts caused by decade-scale precipitation deficits.     

ENSO anomalies over the Western United States: present and future patterns in regional climate simulations

Surface temperature, precipitation, specific humidity and wind anomalies associated with the warm and cold phases of ENSO simulated by WRF and HadRM are examined for the present and future decades. WRF is driven by ECHAM5 and CCSM3, respectively, and HadRM is driven by HadCM3. For the current decades, all simulations show some capability in resolving the observed warm-dry and cool-wet teleconnection patterns over the PNW and the Southwest U.S. for warm and cold ENSO. Differences in the regional simulations originate primarily from the respective driving fields. For the future decades, the warm-dry and cool-wet teleconnection patterns in association with ENSO are still represented in ECHAM5-WRF and HadRM. However, there are indications of changes in the ENSO teleconnection patterns for CCSM3-WRF in the future, with wet anomalies dominating in the PNW and the Southwest U.S. for both warm and cold ENSO, in contrast to the canonical patterns of precipitation anomalies. Interaction of anomalous wind flow with local terrain plays a critical role in the generation of anomalous precipitation over the western U.S. Anomalous dry conditions are always associated with anomalous airflow that runs parallel to local mountains and wet conditions with airflow that runs perpendicular to local mountains. Future changes in temperature and precipitation associated with the ENSO events in the regional simulations indicate varying responses depending on the variables examined as well as depending on the phase of ENSO.     

A regional model for studies of atmosphere-ice-ocean interaction in the western Arctic

Summary Asa step in the development of a fully coupled regional model of the atmosphere-ice-ocean system, atmospheric and sea ice models have been adapted to a western Arctic domain centered on the Bering Strait. Lateral boundary conditions derived from operational analyses drive the models through simulations on grids having horizontal resolutions of 21 km and 7 km. Sensitivities to the presence of sea ice are large after only 48 hours, by which time the surface temperatures in the Bering and Chukchi Seas are 10–15°C higher without sea ice than with sea ice. The temperatures, in turn, modify the fields of sea level pressure, surface wind and precipitation. By influencing the surface wind stress through the static static stability, the surface state feeds back to the surface momentum exchange, ice/ocean transport, and the rate of formation of new ice. The results also show a resolution-dependence of the surface winds, precipitation rates and new ice formation rates, particularly in areas in which the coastal configuration and topography are spatially complex. The experiments will be augmented by the implementation of an ocean model on the same grids.With 12 Figures     

A regional climate model downscaling projection of China future climate change

Climate changes over China from the present (1990–1999) to future (2046–2055) under the A1FI (fossil fuel intensive) and A1B (balanced) emission scenarios are projected using the Regional Climate Model version 3 (RegCM3) nests with the National Center for Atmospheric Research (NCAR) Community Climate System Model (CCSM). For the present climate, RegCM3 downscaling corrects several major deficiencies in the driving CCSM, especially the wet and cold biases over the Sichuan Basin. As compared with CCSM, RegCM3 produces systematic higher spatial pattern correlation coefficients with observations for precipitation and surface air temperature except during winter. The projected future precipitation changes differ largely between CCSM and RegCM3, with strong regional and seasonal dependence. The RegCM3 downscaling produces larger regional precipitation trends (both decreases and increases) than the driving CCSM. Contrast to substantial trend differences projected by CCSM, RegCM3 produces similar precipitation spatial patterns under different scenarios except autumn. Surface air temperature is projected to consistently increase by both CCSM and RegCM3, with greater warming under A1FI than A1B. The result demonstrates that different scenarios can induce large uncertainties even with the same RCM-GCM nesting system. Largest temperature increases are projected in the Tibetan Plateau during winter and high-latitude areas in the northern China during summer under both scenarios. This indicates that high elevation and northern regions are more vulnerable to climate change. Notable discrepancies for precipitation and surface air temperature simulated by RegCM3 with the driving conditions of CCSM versus the model for interdisciplinary research on climate under the same A1B scenario further complicated the uncertainty issue. The geographic distributions for precipitation difference among various simulations are very similar between the present and future climate with very high spatial pattern correlation coefficients. The result suggests that the model present climate biases are systematically propagate into the future climate projections. The impacts of the model present biases on projected future trends are, however, highly nonlinear and regional specific, and thus cannot be simply removed by a linear method. A model with more realistic present climate simulations is anticipated to yield future climate projections with higher credibility.     

Using climate impacts indicators to evaluate climate model ensembles: temperature suitability of premium winegrape cultivation in the United States

We explore the potential to improve understanding of the climate system by directly targeting climate model analyses at specific indicators of climate change impact. Using the temperature suitability of premium winegrape cultivation as a climate impacts indicator, we quantify the inter- and intra-ensemble spread in three climate model ensembles: a physically uniform multi-member ensemble consisting of the RegCM3 high-resolution climate model nested within the NCAR CCSM3 global climate model; the multi-model NARCCAP ensemble consisting of single realizations of multiple high-resolution climate models nested within multiple global climate models; and the multi-model CMIP3 ensemble consisting of realizations of multiple global climate models. We find that the temperature suitability for premium winegrape cultivation is substantially reduced throughout the high-value growing areas of California and the Columbia Valley region (eastern Oregon and Washington) in all three ensembles in response to changes in temperature projected for the mid-twenty first century period. The reductions in temperature suitability are driven primarily by projected increases in mean growing season temperature and occurrence of growing season severe hot days. The intra-ensemble spread in the simulated climate change impact is smaller in the single-model ensemble than in the multi-model ensembles, suggesting that the uncertainty arising from internal climate system variability is smaller than the uncertainty arising from climate model formulation. In addition, the intra-ensemble spread is similar in the NARCCAP nested climate model ensemble and the CMIP3 global climate model ensemble, suggesting that the uncertainty arising from the model formulation of fine-scale climate processes is not smaller than the uncertainty arising from the formulation of large-scale climate processes. Correction of climate model biases substantially reduces both the inter- and intra-ensemble spread in projected climate change impact, particularly for the multi-model ensembles, suggesting that—at least for some systems—the projected impacts of climate change could be more robust than the projected climate change. Extension of this impacts-based analysis to a larger suite of impacts indicators will deepen our understanding of future climate change uncertainty by focusing on the climate phenomena that most directly influence natural and human systems.     

Sensitivity studies with the regional climate model REMO

Summary A new regional atmospheric model was set up in a joint effort by DKRZ (Deutsches Klimarechenzentrum), DWD (Deutscher Wetterdienst), GKSS (Forschungszentrum Geesthacht) at the MPI (Max-Planck-Institut fuer Meteorologie). This model, called REMO (REgional MOdel) can be used in the weather forecast mode as well as in the climate mode. It is based on the Europa-Model (EM), the main weather forecast model of the new numerical weather prediction system of the Deutscher Wetterdienst. In addition to the physical parameterizations implemented in the EM, REMO has the possibility of using the same physics as the global climate model (MPI) into which it is nested to assess the scale dependence of physical parameterizations within the same dynamical framework.This paper gives an overview over different case studies investigating the dependence of model results on simulation domain size, horizontal resolution, initial conditions and lateral boundaries especially for long term calculations. A sample of one month long integrations for an arbitrary July month, a four year long run for the Baltic Sea and its drainage basin and two summer seasons of the Indian Monsoon will be used to demonstrate the sensitivity of regional climate model results to different environments.The sensitivity studies show that it is very important to use realistic large scale driving fields at the lateral boundaries. The regional model results are strongly influenced by the driving fields. The domain size and the simulation length are also influencing the results.With 12 Figures     

Potential use of a regional climate model in seasonal tropical cyclone activity predictions in the western North Pacific

This study investigates the potential use of a regional climate model in forecasting seasonal tropical cyclone (TC) activity. A modified version of Regional Climate Model Version 3 (RegCM3) is used to examine the ability of the model to simulate TC genesis and landfalling TC tracks for the active TC season in the western North Pacific. In the model, a TC is identified as a vortex satisfying several conditions, including local maximum relative vorticity at 850?hPa with a value?≥450?×?10^?6?s^?1, and the temperature at 300?hPa being 1°C higher than the average temperature within 15° latitude radius from the TC center. Tracks are traced by following these found vortices. Six-month ensemble (8 members each) simulations are performed for each year from 1982 to 2001 so that the climatology of the model can be compared to the Joint Typhoon Warning Center (JTWC) observed best-track dataset. The 20-year ensemble experiments show that the RegCM3 can be used to simulate vortices with a wind structure and temperature profile similar to those of real TCs. The model also reproduces tracks very similar to those observed with features like genesis in the tropics, recurvature at higher latitudes and landfall/decay. The similarity of the 500-hPa geopotential height patterns between RegCM3 and the European Centre for Medium-Range Weather Forecasts 40 Year Re-analysis (ERA-40) shows that the model can simulate the subtropical high to a large extent. The simulated climatological monthly spatial distributions as well as the interannual variability of TC occurrence are also similar to the JTWC data. These results imply the possibility of producing seasonal forecasts of tropical cyclones using real-time global climate model predictions as boundary conditions for the RegCM3.     

Consequences of climate change for the soil climate in Central Europe and the central plains of the United States

This study aims to evaluate soil climate quantitatively under present and projected climatic conditions across Central Europe (12.1°–18.9° E and 46.8°–51.1° N) and the U.S. Central Plains (90°–104° W and 37°–49° N), with a special focus on soil temperature, hydric regime, drought risk and potential productivity (assessed as a period suitable for crop growth). The analysis was completed for the baselines (1961–1990 for Europe and 1985–2005 for the U.S.) and time horizons of 2025, 2050 and 2100 based on the outputs of three global circulation models using two levels of climate sensitivity. The results indicate that the soil climate (soil temperature and hydric soil regimes) will change dramatically in both regions, with significant consequences for soil genesis. However, the predicted changes of the pathways are very uncertain because of the range of future climate systems predicted by climate models. Nevertheless, our findings suggest that the risk of unfavourable dry years will increase, resulting in greater risk of soil erosion and lower productivity. The projected increase in the variability of dry and wet events combined with the uncertainty (particularly in the U.S.) poses a challenge for selecting the most appropriate adaptation strategies and for setting adequate policies. The results also suggest that the soil resources are likely be under increased pressure from changes in climate.     

Elevation gradients of European climate change in the regional climate model COSMO-CLM

A transient climate scenario experiment of the regional climate model COSMO-CLM is analyzed to assess the elevation dependency of 21st century European climate change. A focus is put on near-surface conditions. Model evaluation reveals that COSMO-CLM is able to approximately reproduce the observed altitudinal variation of 2 m temperature and precipitation in most regions and most seasons. The analysis of climate change signals suggests that 21st century climate change might considerably depend on elevation. Over most parts of Europe and in most seasons, near-surface warming significantly increases with elevation. This is consistent with the simulated changes of the free-tropospheric air temperature, but can only be fully explained by taking into account regional-scale processes involving the land surface. In winter and spring, the anomalous high-elevation warming is typically connected to a decrease in the number of snow days and the snow-albedo feedback. Further factors are changes in cloud cover and soil moisture and the proximity of low-elevation regions to the sea. The amplified warming at high elevations becomes apparent during the first half of the 21st century and results in a general decrease of near-surface lapse rates. It does not imply an early detection potential of large-scale temperature changes. For precipitation, only few consistent signals arise. In many regions precipitation changes show a pronounced elevation dependency but the details strongly depend on the season and the region under consideration. There is a tendency towards a larger relative decrease of summer precipitation at low elevations, but there are exceptions to this as well.     

Representing glaciers in a regional climate model

A glacier parameterization scheme has been developed and implemented into the regional climate model REMO. The new scheme interactively simulates the mass balance as well as changes of the areal extent of glaciers on a subgrid scale. The temporal evolution and the general magnitude of the simulated glacier mass balance in the European Alps are in good accordance with observations for the period 1958–1980, but the strong mass loss towards the end of the twentieth century is systematically underestimated. The simulated decrease of glacier area in the Alps between 1958 and 2003 ranges from −17.1 to −23.6%. The results indicate that observed glacier mass balances can be approximately reproduced within a regional climate model based on simplified concepts of glacier-climate interaction. However, realistic results can only be achieved by explicitly accounting for the subgrid variability of atmospheric parameters within a climate model grid box.     

A study on combining global and regional climate model results for generating climate scenarios of temperature and precipitation for the Netherlands

Climate scenarios for the Netherlands are constructed by combining information from global and regional climate models employing a simplified, conceptual framework of three sources (levels) of uncertainty impacting on predictions of the local climate. In this framework, the first level of uncertainty is determined by the global radiation balance, resulting in a range of the projected changes in the global mean temperature. On the regional (1,000–5,000 km) scale, the response of the atmospheric circulation determines the second important level of uncertainty. The third level of uncertainty, acting mainly on a local scale of 10 (and less) to 1,000 km, is related to the small-scale processes, like for example those acting in atmospheric convection, clouds and atmospheric meso-scale circulations—processes that play an important role in extreme events which are highly relevant for society. Global climate models (GCMs) are the main tools to quantify the first two levels of uncertainty, while high resolution regional climate models (RCMs) are more suitable to quantify the third level. Along these lines, results of an ensemble of RCMs, driven by only two GCM boundaries and therefore spanning only a rather narrow range in future climate predictions, are rescaled to obtain a broader uncertainty range. The rescaling is done by first disentangling the climate change response in the RCM simulations into a part related to the circulation, and a residual part which is related to the global temperature rise. Second, these responses are rescaled using the range of the predictions of global temperature change and circulation change from five GCMs. These GCMs have been selected on their ability to simulate the present-day circulation, in particular over Europe. For the seasonal means, the rescaled RCM results obey the range in the GCM ensemble using a high and low emission scenario. Thus, the rescaled RCM results are consistent with the GCM results for the means, while adding information on the small scales and the extremes. The method can be interpreted as a combined statistical–dynamical downscaling approach, with the statistical relations based on regional model output.     

Urbanization, climate change and flood policy in the United States

The average annual cost of floods in the United States has been estimated at about $2 billion (current US dollars). The federal government, through the creation of the National Flood Insurance Program (NFIP), has assumed responsibility for mitigating the societal and economic impacts of flooding by establishing a national policy that provides subsidized flood insurance. Increased flood costs during the past two decades have made the NFIP operate at a deficit. This paper argues that our current understanding of climate change and of the sensitivity of the urban environment to floods call for changes to the flood policy scheme. Conclusions are drawn on specific examples from cities along the heavily urbanized corridor of northeastern United States. Mesoscale and global models along with urbanization and economic growth statistics are used to provide insights and recommendations for future flood costs under different emissions scenarios. Mesoscale modeling and future projections from global models suggest, for example, that under a high emissions scenario, New York City could experience almost twice as many days of extreme precipitation that cause flood damage and are disruptive to business as today. The results of the paper suggest that annual flood costs in the United States will increase sharply by the end of the 21st Century, ranging from about $7 to $19 billion current US dollars, depending on the economic growth rate and the emissions scenarios. Hydrologic, hydraulic and other related uncertainties are addressed and a revised version of the NFIP is suggested.     

An inter-comparison of regional climate models for Europe: model performance in present-day climate

The analysis of possible regional climate changes over Europe as simulated by 10 regional climate models within the context of PRUDENCE requires a careful investigation of possible systematic biases in the models. The purpose of this paper is to identify how the main model systematic biases vary across the different models. Two fundamental aspects of model validation are addressed here: the ability to simulate (1) the long-term (30 or 40 years) mean climate and (2) the inter-annual variability. The analysis concentrates on near-surface air temperature and precipitation over land and focuses mainly on winter and summer. In general, there is a warm bias with respect to the CRU data set in these extreme seasons and a tendency to cold biases in the transition seasons. In winter the typical spread (standard deviation) between the models is 1 K. During summer there is generally a better agreement between observed and simulated values of inter-annual variability although there is a relatively clear signal that the modeled temperature variability is larger than suggested by observations, while precipitation variability is closer to observations. The areas with warm (cold) bias in winter generally exhibit wet (dry) biases, whereas the relationship is the reverse during summer (though much less clear, coupling warm (cold) biases with dry (wet) ones). When comparing the RCMs with their driving GCM, they generally reproduce the large-scale circulation of the GCM though in some cases there are substantial differences between regional biases in surface temperature and precipitation.     

Influence of modern land cover on the climate of the United States

I have used a high-resolution nested climate modeling system to test the sensitivity of regional and local climate to the modern non-urban land cover distribution of the continental United States. The dominant climate response is cooling of surface air temperatures, particularly during the warm-season. Areas of statistically significant cooling include areas of the Great Plains where crop/mixed farming has replaced short grass, areas of the Midwest and southern Texas where crop/mixed farming has replaced interrupted forest, and areas of the western United States containing irrigated crops. This statistically significant warm-season cooling is driven by changes in both surface moisture balance and surface albedo, with changes in surface moisture balance dominating in the Great Plains and western United States, changes in surface albedo dominating in the Midwest, and both effects contributing to warm-season cooling over southern Texas. The simulated changes in surface moisture and energy fluxes also influence the warm-season atmospheric dynamics, creating greater moisture availability in the lower atmosphere and enhanced uplift aloft, consistent with the enhanced warm-season precipitation seen in the simulation with modern land cover. The local and regional climate response is of a similar magnitude to that projected for future greenhouse gas concentrations, suggesting that the climatic effects of land cover change should be carefully considered when crafting policies for regulating land use and for managing anthropogenic forcing of the climate system.     

Role of dynamic vegetation in regional climate predictions over western Africa

This study examines the role of vegetation dynamics in regional predictions of future climate change in western Africa using a dynamic vegetation model asynchronously coupled to a regional climate model. Two experiments, one for present day and one for future, are conducted with the linked regional climate-vegetation model, and the third with the regional climate model standing alone that predicts future climate based on present-day vegetation. These simulations are so designed in order to tease out the impact of structural vegetation feedback on simulated climate and hydrological processes. According to future predictions by the regional climate-vegetation model, increase in LAI is widespread, with significant shift in vegetation type. Over the Guinean Coast in 2084–2093, evergreen tree coverage decreases by 49% compared to 1984–1993, while drought deciduous tree coverage increases by 56%. Over the Sahel region in the same period, grass cover increases by 31%. Such vegetation changes are accompanied by a decrease of JJA rainfall by 2% over the Guinean Coast and an increase by 23% over the Sahel. This rather small decrease or large increase of precipitation is largely attributable to the role of vegetation feedback. Without the feedback effect from vegetation, the regional climate model would have predicted a 5% decrease of JJA rainfall in both the Guinean Coast and the Sahel as a result of the radiative and physiological effects of higher atmospheric CO₂ concentration. These results demonstrate that climate- and CO₂-induced changes in vegetation structure modify hydrological processes and climate at magnitudes comparable to or even higher than the radiative and physiological effects, thus evincing the importance of including vegetation feedback in future climate predictions.