资源环境科技发展态势分析平台

In the present study, the performance of the National Centers for Environmental Prediction/Department of Energy (NCEP/DOE) Atmospheric Model Intercomparison Project (AMIP-II) reanalysis (NCEP-2) and NCEP Climate Forecast System Reanalysis (CFSR) seasonal temperature data for China was quantitatively evaluated by using detrended fluctuation analysis. The results indicate that the quality of the NCEP-2 temperature data is the highest in autumn, while the quality of CFSR data set is the highest in summer. Both the quality of NCEP-2 and that of CFSR temperature data is the lowest in winter. The quality is very low in eastern Tibetan Plateau and most of Xinjiang at a significance level of 0.05 for the NCEP-2 and CFSR reanalysis temperature data including three elements: the daily average temperature (T-mean), the daily maximum temperature (T-max) and the daily minimum temperature (T-min). The performance of the NCEP-2T(mean) data is poorer than those of the NCEP-2T(max) and T-min data for all four seasons, while the performance of CFSRTmean is better than that of CFSRTmax and T-min for all four seasons. In addition, the NCEP-2T(max), T-min and CFSRTmean, T-max data perform well in most of central and eastern China for all four seasons. The quality of the CFSRTmean is better than that of the NCEP-2T(mean) for all four seasons. However, the NCEP-2T(max) and T-min data have higher credibility than that of the CFSRTmax and T-min data for all four seasons. Therefore, it is important to consider the reliability of the NCEP-2 and CFSR daily temperature reanalysis data in different districts and seasons when drawing conclusions from the different reanalysis data.