The first multi-model ensemble of regional climate simulations at kilometer-scale resolution, part I: evaluation of precipitation (original) (raw)
Related papers
2021
In this study, we evaluate a set of high-resolution (25-50 km horizontal grid spacing) global climate models (GCMs) from the High-Resolution Model Intercomparison Project (HighResMIP), developed as part of the EU-funded PRIMAVERA (Process-based climate simulation: Advances in high resolution modelling and European climate risk assessment) project, and from the EURO-CORDEX (Coordinated Regional Climate Downscaling Experiment) regional climate models (RCMs) (12-50 km horizontal grid spacing) over a European domain. It is the first time that an assessment of regional climate information using ensembles of both GCMs and RCMs at similar horizontal resolutions has been possible. The focus of the evaluation is on the distribution of daily precipitation at a 50 km scale under current climate conditions. Both the GCM and RCM ensembles are evaluated against high-quality gridded observations in terms of spatial resolution and station density. We show that both ensembles outperform GCMs from the 5th Coupled Model Intercomparison Project (CMIP5), which cannot capture the regional-scale precipitation distribution properly because of their coarse resolutions. PRIMAVERA GCMs generally simulate precipitation distributions within the range of EURO-CORDEX RCMs. Both ensembles perform better in summer and autumn in most European regions but Published by Copernicus Publications on behalf of the European Geosciences Union. 5486 M.-E. Demory et al.: European daily precipitation in EURO-CORDEX RCMs and HighResMIP GCMs tend to overestimate precipitation in winter and spring. PRI-MAVERA shows improvements in the latter by reducing moderate-precipitation rate biases over central and western Europe. The spatial distribution of mean precipitation is also improved in PRIMAVERA. Finally, heavy precipitation simulated by PRIMAVERA agrees better with observations in most regions and seasons, while CORDEX overestimates precipitation extremes. However, uncertainty exists in the observations due to a potential undercatch error, especially during heavy-precipitation events. The analyses also confirm previous findings that, although the spatial representation of precipitation is improved, the effect of increasing resolution from 50 to 12 km horizontal grid spacing in EURO-CORDEX daily precipitation distributions is, in comparison, small in most regions and seasons outside mountainous regions and coastal regions. Our results show that both high-resolution GCMs and CORDEX RCMs provide adequate information to end users at a 50 km scale.
Resolution effects on regional climate model simulations of seasonal precipitation over Europe
Climate Dynamics, 2010
We analyze a set of nine regional climate model simulations for the period 1961-2000 performed at 25 and 50 km horizontal grid spacing over a European domain in order to determine the effects of horizontal resolution on the simulation of precipitation. All of the models represent the seasonal mean spatial patterns and amount of precipitation fairly well. Most models exhibit a tendency to over-predict precipitation, resulting in a domain-average total bias for the ensemble mean of about 20% in winter (DJF) and less than 10% in summer (JJA) at both resolutions, although this bias could be artificially enhanced by the lack of a gauge correction in the observations. A majority of the models show increased precipitation at 25 km relative to 50 km over the oceans and inland seas in DJF, JJA, and ANN (annual average), although the response is strongest during JJA. The ratio of convective precipitation to total precipitation decreases over land for most models at 25 km. In addition, there is an increase in interannual variability in many of the models at 25 km grid spacing. Comparison with gridded observations indicates that a majority of models show improved skill in simulating both the spatial pattern and temporal evolution of precipitation at 25 km compared to 50 km during the summer months, but not in winter or on an annual mean basis. Model skill at higher resolution in simulating the spatial and temporal character of seasonal precipitation is found especially for Great Britain. This geographic dependence of the increased skill suggests that observed data of sufficient density are necessary to capture fine-scale climate signals. As climate models increase their horizontal resolution, it is thus a key priority to produce high quality fine scale observations for model evaluation.
2022
Since a decade, convection-permitting regional climate models (CPRCM) have emerged showing promising results, especially in improving the simulation of precipitation extremes. In this article, the CPRCM CNRM-AROME developed at the Centre National de Recherches Météorologiques (CNRM) since a few years is described and evaluated using a 2.5-km long 19-year hindcast simulation over a large northwestern European domain using different observations through an added-value analysis in which a comparison with its driving 12-km RCM CNRM-ALADIN is performed. The evaluation is challenging due to the lack of high-quality observations at both high temporal and spatial resolutions. Thus, a high spatio-temporal observed gridded precipitation dataset was built from the collection of seven national datasets that helped the identification of added value from CNRM-AROME. The evaluation is based on a series of standard climatic features that include long-term means and mean annual cycles of precipitati...
2022
Recent studies using convection-permitting (CP) climate simulations have demonstrated a step-change in the representation of heavy rainfall and rainfall characteristics (frequency-intensity) compared to coarser resolution Global and Regional Climate models. The goal of this study is to better understand what explains the weaker frequency of precipitation in the CP ensemble by assessing the triggering process of precipitation in the different ensembles of regional climate simulations available over Europe. We focus on the statistical relationship between tropospheric temperature, humidity and precipitation to understand how the frequency of precipitation over Europe and the Mediterranean is impacted by model resolution and the representation of convection (parameterized vs. explicit). We employ a multi-model data-set with three different resolutions (0.44°, 0.11° and 0.0275°) produced in the context of the MED-CORDEX, EURO-CORDEX and the CORDEX Flagship Pilot Study "Convective P...
Climate Dynamics, 2011
Regional Climate Models (RCMs) constitute the most often used method to perform affordable highresolution regional climate simulations. The key issue in the evaluation of nested regional models is to determine whether RCM simulations improve the representation of climatic statistics compared to the driving data, that is, whether RCMs add value. In this study we examine a necessary condition that some climate statistics derived from the precipitation field must satisfy in order that the RCM technique can generate some added value: we focus on whether the climate statistics of interest contain some fine spatial-scale variability that would be absent on a coarser grid. The presence and magnitude of fine-scale precipitation variance required to adequately describe a given climate statistics will then be used to quantify the potential added value (PAV) of RCMs. Our results show that the PAV of RCMs is much higher for short temporal scales (e.g., 3-hourly data) than for long temporal scales (16-day average data) due to the filtering resulting from the time-averaging process. PAV is higher in warm season compared to cold season due to the higher proportion of precipitation falling from small-scale weather systems in the warm season. In regions of complex topography, the orographic forcing induces an extra component of PAV, no matter the season or the temporal scale considered. The PAV is also estimated using high-resolution datasets based on observations allowing the evaluation of the sensitivity of changing resolution in the real climate system. The results show that RCMs tend to reproduce relatively well the PAV compared to observations although showing an overestimation of the PAV in warm season and mountainous regions.
Climate Dynamics, 2009
One of the main concerns in regional climate modeling is to which extent limited-area regional climate models (RCM) reproduce the large-scale atmospheric conditions of their driving general circulation model (GCM). In this work we investigate the ability of a multi-model ensemble of regional climate simulations to reproduce the large-scale weather regimes of the driving conditions. The ensemble consists of a set of 13 RCMs on a European domain, driven at their lateral boundaries by the ERA40 reanalysis for the time period 1961–2000. Two sets of experiments have been completed with horizontal resolutions of 50 and 25 km, respectively. The spectral nudging technique has been applied to one of the models within the ensemble. The RCMs reproduce the weather regimes behavior in terms of composite pattern, mean frequency of occurrence and persistence reasonably well. The models also simulate well the long-term trends and the inter-annual variability of the frequency of occurrence. However, there is a non-negligible spread among the models which is stronger in summer than in winter. This spread is due to two reasons: (1) we are dealing with different models and (2) each RCM produces an internal variability. As far as the day-to-day weather regime history is concerned, the ensemble shows large discrepancies. At daily time scale, the model spread has also a seasonal dependence, being stronger in summer than in winter. Results also show that the spectral nudging technique improves the model performance in reproducing the large-scale of the driving field. In addition, the impact of increasing the number of grid points has been addressed by comparing the 25 and 50 km experiments. We show that the horizontal resolution does not affect significantly the model performance for large-scale circulation.
International Journal of Climatology, 2013
An ensemble of high-resolution regional climate simulations is used to assess the effect of near future climate change on mean and extreme precipitation in a part of Central Europe with complex topography. The ensemble consists of high-resolution simulations with the COSMO-CLM (CLM) regional climate model (RCM) using several realizations of the driving general circulation models (GCMs) ECHAM5 and HadCM3. The study is focussed on the changes in the near future (2011-2040 compared to 1971-2000) which are relevant for planning purposes. The mean winter precipitation shows a spatially uniform increase, summertime mean precipitation is likely to decrease slightly. For extreme precipitation the simulations exhibit an increase on the average for both seasons, but for different reasons. The changes in winter are proportional to the increase in total precipitation, whereas in summer a broadening of the precipitation distribution is found. The spatial change patterns in summer are much more heterogeneous than in winter, with regions of significant increase and decrease sometimes close to each other. The plausibility of the findings is assessed in terms of ensemble consistency. The area mean changes found for the ensemble of CLM simulations were consistent with the change signals derived from a larger but coarser resolved ensembles using several RCMs and driving GCMs. In addition, it was found that the simulated near future precipitation changes in the study region during summer generally agree with trends observed during the last decades.
2003
Two climate model simulations made with the Rossby Centre regional Atmospheric model version 1 (RCA1) are evaluated for the precipitation climate in Scania, southernmost Sweden. These simulations are driven by the HadCM2 and the ECHAM4=OPYC3 global circulation models (GCMs) for 10 years. Output from the global and the regional simulations are compared with an observational data set, constructed from a dense precipitation gauge network in Scania. Area-averaged time series corresponding to the size and location of the RCA1 grid points in Scania have been created (the Scanian Data Set). This data set was compared to a commonly used gridded surface climatology provided by the Climatic Research Unit (CRU). Relatively large differences were found, mainly due to the fact that the CRU-climatology uses fewer stations and lacks a correction for rain-gauge under-catch. This underlines the importance of the data set chosen for model evaluations. The validation is carried out at a large scale including the whole area of Scania and at the finest resolution of RCA1 (the grid point level). When integrated over the whole area of Scania, RCA1 improves the shape of the annual precipitation cycle and the inter-annual variability compared to output from the GCMs. The RCA1 control climate is generally too wet compared to the observations. At the grid point level, RCA1 improves the simulation of the variability compared to the GCMs. There is a strong positive correlation between precipitation and altitude in all seasons in the observations. This relationship is, however, much weaker and even reversed in the RCA1 simulations. Analysis of the dense rain gauge network reveals features of spatial variability at around 20–35 km in the area and indicates that a finer resolution is needed if the spatial variability in the area is to be better captured by RCA1.
Climate Models and Their Simulation of Precipitation
Energy & Environment, 2014
Current state-of-the-art General Circulation Models (GCMs) do not simulate precipitation well because they do not include the full range of precipitationforming mechanisms that occur in the real world. It is demonstrated here that the impact of these errors are not trivial-an error of only 1 mm in simulating liquid rainfall is equivalent to the energy required to heat the entire troposphere by 0.3°C. Given that models exhibit differences between the observed and modeled precipitation that often exceed 1 mm day-1 , this lost energy is not trivial. Thus, models and their prognostications are largely unreliable.