Search

Based on the output data of Beijing Climate Center Atmospheric General Circulation Model version 2.0.1 (BCC_AGCM2.0.1), this study evaluates the simulated precipitation and associated circulation during the Meiyu period over the Yangtze-Huaihe River Basin (YHRB). The results show that BCC_AGCM2.0.1 can reasonably reproduce the spatial pattern of Meiyu over the YHRB. However, compared to the observation, the model underestimated (overestimated) the torrential and heavy (moderate and light) rainy days, indicating that the more (less) simulated light (torrential) rainy days result in more (less) modeled total rainy days (Meiyu precipitation amount). The simulated Meiyu onset begins in early June, earlier about 20 days than the climatological observation. For duration, Meiyu persists for 15 days, about 5 days shorter than the observation, which may be due to the simulated land-sea thermal contrast is earlier in early June. During the Meiyu period, the distribution patterns of the upper troposphere atmospheric circulation can be well simulated by BCC_AGCM2.0.1, but the obvious deficiency lies in the lower-troposphere air temperature and wind field, especially over Tibetan Plateau area. [ABSTRACT FROM AUTHOR]

Following the practice of the numerical weather ensemble prediction, a nonlinear statistical ensemble prediction model has been developed based on a neural network technique with a Particle Swarm Optimization (PSO) algorithm. The model is validated by short-range climate forecasts of monthly mean rainfall at 37 stations in Guangxi, China during the first rainy season (April, May, and June). Independent prediction results show that the Particle Swarm Optimization Neural Network ensemble prediction model is clearly better than the traditional linear statistical method, such as the multiple regression method and the stepwise regression method. It is also suggested that by applying multiple ensemble members with each member objectively determined by the PSO algorithm, the generalization capacity of the ensemble prediction model is enhanced, demonstrating a vast range of possibilities for operational short-range climate prediction. [ABSTRACT FROM AUTHOR]

Precipitation, which is the predominant component of the East Asian summer monsoon (EASM), may have large uncertainties among reanalysis datasets. We comprehensively evaluate the performance of five reanalysis datasets in reproducing the EASM precipitation. These datasets are NCEP/NCAR Reanalysis 1 project (NCEP1), NCEP/US Department of Energy Atmospheric Model Intercomparison Project II reanalysis (NCEP2), Japanese 25-year Reanalysis project (JRA-25), Interim ECMWF Reanalysis (ERA-Interim), and Modern Era Retrospective-analysis for Research and Applications (MERRA). Results show that the five reanalysis datasets can generally reproduce the climatology and interannual variability of EASM precipitation. Especially, MERRA and ERA-Interim have the highest skills. Considering different-class precipitation, large uncertainties exist in the category of non-rainfall and heavy rainfall. The five reanalysis datasets overestimate the non-rainfall frequency, and JRA-25 and NCEP2 overestimate the heavy rainfall frequency. The well-known interdecadal variation around the mid-1990s can also be better depicted by ERA-Interim and MERRA. For the linear trend of precipitation, only MERRA can reasonably reproduce the increasing tendency over southern China and the western Pacific and the decreasing tendency over the Indo-China Peninsula. Based on EOF analysis, the spatial-temporal structure of EASM precipitation has been examined. MERRA, NCEP1 and ERA-Interim can better capture both the spatial patterns and principle components of the first two EOF modes. Based on our evaluation, the preferential reanalysis datasets for investigating the EASM precipitation are ERA-Interim and MERRA, which also permit the more precise investigation of interannual to decadal variability. [ABSTRACT FROM AUTHOR]