Choice Of Regional Climate Models Research Articles

On the representation of atmospheric circulation modes in regional climate models over Western Europe

AbstractAtmospheric circulation is a key driver of climate variability, and the representation of atmospheric circulation modes in regional climate models (RCMs) can enhance the credibility of regional climate projections. This study examines the representation of large‐scale atmospheric circulation modes in Coupled Model Inter‐comparison Project phase 5 RCMs once driven by ERA‐Interim, and by two general circulation models (GCMs). The study region is Western Europe and the circulation modes are classified using the Promax rotated T‐mode principal component analysis. The results indicate that the RCMs can replicate the classified atmospheric modes as obtained from ERA5 reanalysis, though with biases dependent on the data providing the lateral boundary condition and the choice of RCM. When the boundary condition is provided by ERA‐Interim that is more consistent with observations, the simulated map types and the associating time series match well with their counterparts from ERA5. Further, on average, the multi‐model ensemble mean of the analysed RCMs, driven by ERA‐Interim, indicated a slight improvement in the representation of the modes obtained from ERA5. Conversely, when the RCMs are driven by the GCMs that are models without assimilation of observational data, the representation of the atmospheric modes, as obtained from ERA5, is relatively less accurate compared to when the RCMs are driven by ERA‐Interim. This suggests that the biases stem from the GCMs. On average, the representation of the modes was not improved in the multi‐model ensemble mean of the five analysed RCMs driven by either of the GCMs. However, when the best‐performed RCMs were selected on average the ensemble mean indicated a slight improvement. Moreover, the presence of the North Atlantic Oscillation (NAO) in the simulated modes depends also on the lateral boundary conditions. The relationship between the modes and the NAO was replicated only when the RCMs were driven by reanalysis. The results indicate that the forcing model is the main factor in reproducing the atmospheric circulation.

Partitioning uncertainty components of mean climate and climate change in a large ensemble of European regional climate model projections

A study of seasonal mean temperature, precipitation, and wind speed has been performed for a set of 19 global climate model (GCM) driven high-resolution regional climate model (RCM) simulations forming a complete 5 × 4 GCM × RCM matrix with only one missing simulation. Differences between single simulations and between groups of simulations forced by a specific GCM or a specific RCM are identified. With the help of an analysis of variance (ANOVA) we split the ensemble variance into linear GCM and RCM contributions and cross terms for both mean climate and climate change for the end of the current century according to the RCP8.5 emission scenario. The results document that the choice of GCM generally has a larger influence on the climate change signal than the choice of RCM, having a significant influence for roughly twice as many points in the area for the fields investigated (temperature, precipitation and wind speed). It is also clear that the RCM influence is generally concentrated close to the eastern and northern boundaries and in mountainous areas, i.e., in areas where the added surface detail of e.g. orography, snow and ice seen by the RCM is expected to have considerable influence on the climate, and in areas where the air in general has spent the most time within the regional domain. The analysis results in estimates of areas where the specific identity of either GCM or RCM is formally significant, hence obtaining an indication about regions, seasons, and fields where linear superpositions of GCM and RCM effects are good approximations to an actual simulation for both the mean fields analysed and their changes. In cases where linear superposition works well, the frequently encountered sparse GCM–RCM matrices may be filled with emulated results, leading to the possibility of giving more fair relative weight between model simulations than simple averaging of existing simulations. An important result of the present study is that properties of the specific GCM–RCM combination are generally important for the mean climate, but negligible for climate change for the seasonal-mean surface fields investigated here.

Precipitation intensity–duration–frequency curves for central Belgium with an ensemble of EURO-CORDEX simulations, and associated uncertainties

An ensemble of 88 regional climate model (RCM) simulations at 0.11° and 0.44° spatial resolutions from the EURO-CORDEX project is analyzed for central Belgium to investigate the projected impact of climate change on precipitation intensity–duration–frequency (IDF) relationships and extreme precipitation quantiles typically used in water engineering designs. The rate of uncertainty arising from the choice of RCM, driving GCM, and radiative concentration pathway (RCP4.5 & RCP8.5) is quantified using a variance decomposition technique after reconstruction of missing data in GCM×RCM combinations. A comparative analysis between the historical simulations of the EURO-CORDEX 0.11° and 0.44° RCMs shows higher precipitation intensities by the finer resolution runs, leading to a larger overestimation of the observations-based IDFs by the 0.11° runs. The results reveal that making a temporal stationarity assumption for the climate system may lead to underestimation of precipitation quantiles up to 70% by the end of this century. This projected increase is generally larger for the 0.11° RCMs compared with the 0.44° RCMs. The relative changes in extreme precipitation do depend on return period and duration, indicating an amplification for larger return periods and for smaller durations. The variance decomposition approach generally identifies RCM as the most dominant component of uncertainty in changes of more extreme precipitation (return period of 10years) for both 0.11° and 0.44° resolutions, followed by GCM and RCP scenario. The uncertainties associated with cross-contributions of RCMs, GCMs, and RCPs play a non-negligible role in the associated uncertainties of the changes.

Future changes in the climatology of the Great Plains low-level jet derived from fine resolution multi-model simulations

The southerly Great Plains low-level jet (GPLLJ) is one of the most significant circulation features of the central U.S. linking large-scale atmospheric circulation with the regional climate. GPLLJs transport heat and moisture, contribute to thunderstorm and severe weather formation, provide a corridor for the springtime migration of birds and insects, enhance wind energy availability, and disperse air pollution. We assess future changes in GPLLJ frequency using an eight member ensemble of dynamically-downscaled climate simulations for the mid-21st century. Nocturnal GPLLJ frequency is projected to increase in the southern plains in spring and in the central plains in summer, whereas current climatological patterns persist into the future for daytime and cool season GPLLJs. The relationship between future GPLLJ frequency and the extent and strength of anticyclonic airflow over eastern North America varies with season. Most simulations project a westward shift of anticyclonic airflow in summer, but uncertainty is larger for spring with only half of the simulations suggesting a westward expansion. The choice of regional climate model and the driving lateral boundary conditions have a large influence on the projected future changes in GPLLJ frequency and highlight the importance of multi-model ensembles to estimate the uncertainty surrounding the future GPLLJ climatology.

Regional climate model projections of the South Pacific Convergence Zone

This study presents results from regional climate model (RCM) projections for the south-west Pacific Ocean. The regional models used bias corrected sea surface temperatures. Six global climate models (GCMs) were used to drive a global variable resolution model on a quasi-uniform 60 km grid. One of these simulations was used to drive three limited area regional models. Thus a four member ensemble was produced by different RCMs downscaling the same GCM (GFDL2.1), and a six member ensemble was produced by the same RCM (Conformal Cubic Atmospheric Model—CCAM) downscaling six different GCMs. Comparison of the model results with precipitation observations shows the differences to be dominated by the choice of RCM, with all the CCAM simulations performing similarly and generally having lower error than the other RCMs. However, evaluating aspects of the model representation of the South Pacific Convergence Zone (SPCZ) does not show CCAM to perform better in this regard. In terms of the future projections of the SPCZ for the December–January–February season, the ensemble showed no consensus change in most characteristics though a majority of the ensemble members project a decrease in the SPCZ strength. Thus, similar to GCM based studies, there is large uncertainty concerning future changes in the SPCZ and there is no evidence to suggest that future changes will be outside the natural variability. These RCM simulations do not support an increase in the frequency of zonal SPCZ events.

Uncertainty in Regional Climate Model Mean Runoff Projections under Climate Change: Case Study of Labrador's Churchill River Basin

ABSTRACTAn ensemble of seven climate models from the North American Regional Climate Change Assessment Program (NARCCAP) was used to examine uncertainty in simulated runoff changes from a base period (1971–2000) to a future period (2041–2070) for the Churchill River basin, Labrador, Canada. Three approximations for mean annual runoff from each ensemble member were included in the analysis: (i) atmospheric moisture convergence, (ii) the balance between precipitation and evaporation, and (iii) instantaneous runoff output from respective land-surface schemes. Using data imputation (i.e., reconstruction) and variance decomposition it was found that choice of regional climate model (RCM) made the greatest contribution to uncertainty in the climate change signal, whereas the boundary forcing of a general circulation model (GCM) played a smaller, though non-negligible, role. It was also found that choice of runoff approximation made a substantial contribution to uncertainty, falling between the contribution from RCM and GCM choice. The NARCCAP output and imputed data were used to calculate mean and median annual changes and results were presented via probability distribution functions to facilitate decision making. Mean and median increases in runoff for the basin were found to be 11.2% and 8.9%, respectively.

Extreme value indicators in highly resolved climate change simulations for the Jordan River area

[1] Understanding changing trends and frequency of extreme rainfall and temperature events is extremely important for optimal planning in many sectors, including agriculture, water resource management, health, and even economics. For people living in the Jordan River region of the Middle East such changes can have immediate devastating impacts as water resources are already scarce and overexploited and summer temperatures in the desert regions can reach 45°C or higher. Understanding shifts in frequency and intensity of extreme events can provide crucial information for planning and adaptation. In this paper we present results from regional climate model simulations with RegCM3 and MM5 centered on the eastern Mediterranean region. Our analysis focuses on changes in extreme temperature and rainfall events. We show that maximum daily summer temperature is expected to increase by between 2.5°C and 3°C, with an increase in warm spell length. Precipitation extremes are expected to increase with longer dry spells, shorter wet spells, and increases in heavy rainfall. Model agreement for the control period 1961–1990 is higher in the southern region than in the north, perhaps because of the complex topography, suggesting that even small differences in spatial scale play an important role. In addition, we notice that the chosen global model plays an important role in determining future temperature trends, while the choice of regional climate model is critical for understanding how precipitation is expected to evolve.

Uncertainty assessment of climate change impacts on the hydrology of small prairie wetlands

With increasing evidences of climate change in the prairie region, there is an urgent need to understand the future climate and the responses of small prairie wetlands. This study integrated two regional climate models (RCMs), two weather generators and a distributed hydrological model to examine uncertainties in hydrological responses to climate change in the Assiniboia watershed, Canada. Comparing to baseline conditions (1971–2000), annual water yield and evapotranspiration in the period of 2041–2070 were generally unchanged, while annual reservoir storage was generally reduced. However, projected hydrological regimes were less consistent at monthly level, particularly for March and July. Such uncertainties in simulated hydrological responses were derived from the implementations of different integrated downscaling methods, reflecting our imperfect knowledge of the future climate. We identified a warming temperature trend from climatic projections, but had less confidence in the future pattern of precipitation. Uncertainties in integrated downscaling were primarily derived from the choice of RCM, and were amplified through the incorporation of different weather generators. Results of any climate change study based on only one RCM and/or one weather generator should be interpreted with caution, and the ensemble framework should be advised to generate a comprehensive vision of the future climate. This study demonstrated that the incorporation of precipitation occurrence change contributed to a full translation of RCM outputs, but also introduced additional uncertainty. A balance is thus desired between the information loss and the additional uncertainty in order to effectively utilize RCM outputs.

An intercomparison of regional climate model data for hydrological impact studies in Denmark

Summary The use of high-resolution regional climate models (RCM) to examine the hydrological impacts of climate change has grown significantly in recent years due to the improved representation of the local climate. However, the application is not straightforward because most RCMs are subject to considerable systematic errors. In this study, projected climate change data from the RCM HIRHAM4 are used to generate climate scenario time series of precipitation, temperature, and reference evapotranspiration for the period 2071–2100 for hydrological impact assessments in Denmark. RCM output for the present-day period (1961–1990) are compared to an observational data set, with precipitation corrected for undercatch and wetting losses, to quantify systematic model errors. A delta change method is applied to cope with these biases. A question arises as to how variable the climate change signals are, as they are influenced by the forcing GCM, the emissions scenario, and the choice of RCM. Here, we focus on the choice of RCM and the effect of increasing horizontal resolution. A comparison of HIRHAM4 output to seven similar RCMs showed that for precipitation the HIRHAM4 model is as representative as any of the RCMs in the ensemble. For temperature HIRHAM4 at 50 km resolution simulates higher temperatures than the other RCMs, while HIRHAM4 at 12 km resolution is within the range of the other models. This study highlights some of the uncertainties related to the use of RCM data in hydrological studies, which contribute to a range in hydrological effects of climate change, mainly originating from simulated precipitation amounts.

Changes in European ecosystem productivity and carbon balance driven by regional climate model output

AbstractClimate change resulting from the enhanced greenhouse effect together with the direct effect of increased atmospheric CO2 concentrations on vegetation growth are expected to produce changes in the cycling of carbon in terrestrial ecosystems. Impacts will vary across Europe, and regional‐scale studies are needed to resolve this variability. In this study, we used the LPJ‐GUESS ecosystem model driven by a suite of regional climate model (RCM) scenarios from the European Union (EU) project PRUDENCE to estimate climate impacts on carbon cycling across Europe. We identified similarities and discrepancies in simulated climate impacts across scenarios, particularly analyzing the uncertainties arising from the range of climate models and emissions scenarios considered. Our results suggest that net primary production (NPP) and heterotrophic respiration (Rh) will generally increase throughout Europe, but with considerable variation between European subregions. The smallest NPP increases, and in some cases decreases, occurred in the Mediterranean, where many ecosystems switched from sinks to sources of carbon by 2100, mainly as a result of deteriorating water balance. Over the period 1991–2100, modeled climate change impacts on the European carbon balance ranged from a sink of 11.6 Gt C to a source of 3.3 Gt C, the average annual sink corresponding with 1.85% of the current EU anthropogenic emissions. Projected changes in carbon balance were more dependent on the choice of the general circulation model (GCM) providing boundary conditions to the RCM than the choice of RCM or the level of anthropogenic greenhouse gases emissions.