Search

This study evaluated the performance of the climatological precipitation over Southeast Asia using an ensemble of 48 global climate models/earth system models from the sixth phase of Coupled Model Intercomparison Project (CMIP6) suite of experiments. Multi‐source observational data sets were used to evaluate the historical simulation data of the CMIP6 ensembles towards the semi‐year mean of precipitation during November–April (NDJFMA) and May–October (MJJASO). The results indicate that the CMIP6 models were able to reproduce the large‐scale monsoonal features. Precipitation is overestimated over most eastern ocean areas in NDJFMA and over most areas in MJJASO. In terms of the ensemble bias and inter‐model uncertainty, the diagnosis of moisture budget indicates the dominant item is the dynamic convergence and the other important items have dynamic advection and evaporation. The main types of pattern bias were examined through the inter‐model empirical orthogonal function analysis, which is correlated with the global sea surface temperature and moisture flux. The climatological evaluation yielded a higher pattern correlation and a lower root‐mean square error in NDJFMA than MJJASO. Most of the models have greater standard deviations among the spatial grids than observations, especially in NDJFMA. The models with higher resolutions showed a better performance than those with lower resolutions. A model performance index is defined for the convenient comparison, the best three models are CNRM‐CM6‐1, EC‐Earth3‐Veg and NorESM2‐MM, and the worst three are INM‐CM4‐8, AWI‐ESM‐1‐1‐LR and MCM‐UA‐1‐0. This evaluation study provided useful analytics in understanding model behaviour and processes related to rainfall mechanisms which could be further examined in high‐resolution downscaling experiments in future studies. [ABSTRACT FROM AUTHOR]

This study evaluates the National Aeronautics Space Administration (NASA) Earth Exchange Global Daily Downscaled Projections (NEX-GDDP) dataset that provides statistically downscaled CMIP5 historical and future climate projections (precipitation and temperature), at high spatial (25 km) and temporal (daily) resolutions. The study is performed over Southeast Asia for the historical period 1976-2005 and compared against gridded observations on daily scales. Future climate change is assessed over the future time slices, 2020-2050 and 2070-2099, under two Representative Concentration Pathways (RCP) scenarios 4.5 and 8.5 with respect to 1976-2005. The future climate projections indicate that surface temperatures over Southeast Asia are likely to increase by more than 3.5 °C by the end of the century. As to precipitation, both the mean and extreme rainfall are likely to increase but the biases in the historical simulations could contribute to larger uncertainties in the estimates of rainfall projections. Findings of the study indicate that NEX-GDDP are in good agreement with observations over the historical period only on monthly scales and that they do not capture the observed statistics on daily scales which suggests that these data need a deeper scrutiny on daily scales, especially when used for impacts studies. [ABSTRACT FROM AUTHOR]

The uncertainties in two high-resolution satellite precipitation products (TRMM 3B42 v7.0 and GSMaP v5.222) were investigated by comparing them against rain gauge observations over Singapore on sub-daily scales. The satellite-borne precipitation products are assessed in terms of seasonal, monthly and daily variations, the diurnal cycle, and extreme precipitation over a 10-year period (2000-2010). Results indicate that the uncertainties in extreme precipitation is higher in GSMaP than in TRMM, possibly due to the issues such as satellite merging algorithm, the finer spatio-temporal scale of high intensity precipitation, and the swath time of satellite. Such discrepancies between satellite-borne and gauge-based precipitations at sub-daily scale can possibly lead to distorting analysis of precipitation characteristics and/or application model results. Overall, both satellite products are unable to capture the observed extremes and provide a good agreement with observations only at coarse time scales. Also, the satellite products agree well on the late afternoon maximum and heavier rainfall of gauge-based data in winter season when the Intertropical Convergence Zone (ITCZ) is located over Singapore. However, they do not reproduce the gauge-observed diurnal cycle in summer. The disagreement in summer could be attributed to the dominant satellite overpass time (about 14:00 SGT) later than the diurnal peak time (about 09:00 SGT) of gauge precipitation. From the analyses of extreme precipitation indices, it is inferred that both satellite datasets tend to overestimate the light rain and frequency but underestimate high intensity precipitation and the length of dry spells. This study on quantification of their uncertainty is useful in many aspects especially that these satellite products stand scrutiny over places where there are no good ground data to be compared against. This has serious implications on climate studies as in model evaluations and in particular, climate model simulated future projections, when information on precipitation extremes need to be reliable as they are highly crucial for adaptation and mitigation. [ABSTRACT FROM AUTHOR]

We present preliminary analyses of the historical (1986-2005) climate simulations of a ten-member subset of the Coupled Model Inter-comparison Project Phase 5 (CMIP5) global climate models over Southeast Asia. The objective of this study was to evaluate the general circulation models’ performance in simulating the mean state of climate over this less-studied climate vulnerable region, with a focus on precipitation. Results indicate that most of the models are unable to reproduce the observed state of climate over Southeast Asia. Though the multi-model ensemble mean is a better representation of the observations, the uncertainties in the individual models are far high. There is no particular model that performed well in simulating the historical climate of Southeast Asia. There seems to be no significant influence of the spatial resolutions of the models on the quality of simulation, despite the view that higher resolution models fare better. The study results emphasize on careful consideration of models for impact studies and the need to improve the next generation of models in their ability to simulate regional climates better. [ABSTRACT FROM AUTHOR]

• The model bias is decomposed into the ensemble bias and the inter-model modes. • The CMIP6 models simulate well the meridional march and the annual climatology. • The long-term trends are poorly represented in CMIP6 models. • The CMIP6 models can only reproduce about 23% of interannual variability. • The best 3 models are EC-Earth3, EC-Earth3-Veg and E3SM-1-0. This study evaluated the bias in rainfall over Southeast Asia in the historical simulations of the sixth phase of Coupled Model Intercomparison Project (CMIP6). Using 48 CMIP6 models, a suite of evaluation methods was applied to examine in detail – annual meridional march, annual climatology, long-term trend, and interannual variability. For this study, multi-source observational datasets were employed for comparisons against model simulations. The biases in the climatological state were decomposed into ensemble biases and the leading modes of inter-model uncertainty. The CMIP6 models could simulate well the climatological state of meridional march and annual mean with high pattern correlation coefficients of 0.88 and 0.59, respectively. However, the performance is poor in simulating the long-term trend that exhibits a correlation coefficient of 0.05 between simulation and observation in spatial patterns. The CMIP6 models could only reproduce about 23% of interannual variability. This evaluation not only revealed the systematic biases in 48 CMIP6 models but also illustrated the biases in the different Empirical Orthogonal Function (EOF) modes of inter-model uncertainty and the relative performance ranking for a particular model. Based on the total performance index, the best 3 models are EC-Earth3, EC-Earth3-Veg, E3SM-1-0 and relatively poor performance were seen in INM-CM4-8, AWI-ESM-1-1-LR and MCM-UA-1-0. The study provides an assessment of the overall uncertainties in CMIP6 models in the historical simulations and could also be a useful feedback to the global climate modellers in optimizing the models for better performance. [ABSTRACT FROM AUTHOR]