Search

Analyses of the variation characteristics for aridity index (AI) can further enhance the understanding of climate change and have effect on hydrology and agriculture. In this paper, based on the data of 283 standard meteorological stations, the temporal-spatial variations and the influences of climate factors on AI were investigated and the relationship between AI and two climate indices (the Arctic Oscillation (AO); El Nino-Southern Oscillation (ENSO)) were also assessed in northern China (NC) during the period from 1961 to 2012. The results revealed that the annual mean AI decreased at the rate of −0.031 per decade in the past 52 years and the trend was statistically significant at the 0.01 level. The Mann-Kendall (M-K) test presented that the percentages of stations with positive trends and negative trends for AI were 10 and 81.9 % (22.6 % statistically significant), respectively. Spatially, in the western part of 100° E, the extremely dry area declined and the climate tended to become wet obviously. In the eastern part of 100° E, dry area moved toward the east and the south, which resulted in the enhancement of semiarid area and the shrinkage of subhumid area. The contributions of sunshine duration and precipitation to the decline of AI are more than those of other meteorological variables in NC. Moreover, the average temperature has risen significantly and AI decreased in NC, which indicated the existence of 'paradox.' Relationship between climate indices (AO and ENSO) and AI demonstrated that the influence of ENSO on AI overweight the AO on AI in NC. [ABSTRACT FROM AUTHOR]

The parameterization of physical processes is one of the critical elements to properly simulate the regional climate over eastern China. It is essential to conduct detailed analyses on the effect of physical parameterization schemes on regional climate simulation, to provide more reliable regional climate change information. In this paper, we evaluate the 25-year (1983-2007) summer monsoon climate characteristics of precipitation and surface air temperature by using the regional spectral model (RSM) with different physical schemes. The ensemble results using the reliability ensemble averaging (REA) method are also assessed. The result shows that the RSM model has the capacity to reproduce the spatial patterns, the variations, and the temporal tendency of surface air temperature and precipitation over eastern China. And it tends to predict better climatology characteristics over the Yangtze River basin and the South China. The impact of different physical schemes on RSM simulations is also investigated. Generally, the CLD3 cloud water prediction scheme tends to produce larger precipitation because of its overestimation of the low-level moisture. The systematic biases derived from the KF2 cumulus scheme are larger than those from the RAS scheme. The scale-selective bias correction (SSBC) method improves the simulation of the temporal and spatial characteristics of surface air temperature and precipitation and advances the circulation simulation capacity. The REA ensemble results show significant improvement in simulating temperature and precipitation distribution, which have much higher correlation coefficient and lower root mean square error. The REA result of selected experiments is better than that of nonselected experiments, indicating the necessity of choosing better ensemble samples for ensemble. [ABSTRACT FROM AUTHOR]