多种再分析地表温度资料在中国区域的适用性分析

利用1979-2010年中国805个站点的逐日地表温度观测资料与ERA-Interim、ERA-Interim/Land、JRA-55、NCEP/NCAR和NCEP/DOE五种再分析地表温度资料, 对五种再分析地表温度资料在中国区域的适用性进行了比较与分析. 结果表明: 从空间分布特征来看, JRA-55资料与观测资料最为接近; 五种再分析地表温度资料在东部地区的适用性好于西部地区, 且均不能很好地模拟出青藏高原地区和西北西部地区的地表温度分布. 从地表温度的长期变化来看, 五种再分析地表温度资料虽然可以描述地表温度变化趋势, 但均存在低估现象, 且低估现象随着地表温度的增加变得更加严重. 在地表温度人工观测和自动观测交接的时间段, 可以使用再分析资料作为地表温度均一化的参考序列.

Applicability analysis of various reanalyzed land surface temperature datasets in China

Using daily land surface temperature observations from 805 stations in China from 1979 to 2010, and land surface temperature data of ERA-Interim, ERA-Interim/Land, JRA-55, NCEP/NCAR and NCEP/DOE, the applicability of the five reanalyzed land surface temperature datasets in China was analyzed and compared. It is found that (1) regarded from spatial distribution characteristics of the land surface temperature, JRA-55 dataset is the closest to the observation; (2) applicability of the five reanalyzed surface temperature datasets in the eastern regions is better than that in the western regions, but all of the five reanalysis land surface temperature datasets cannot well simulate the land surface temperature on the Tibetan Plateau and the west of Northwest China; (3) from the perspective of long-term change in land surface temperature, all of the five reanalyzed land surface temperature datasets can simulate the changing trend of land surface temperature, but all of them underestimate the land surface temperature, and the underestimation becomes more serious with increasing land surface temperature; (4) during the transition of manual observation and automatic observation, reanalyzed land surface temperature datasets can be used as a reference sequence of land surface temperature homogenization.