GHCN-CAMS和CMFD两种尺度再分析资料对宁夏气温反映能力评估

再分析资料能有效弥补实际观测数据时空分布不均的缺陷,开展再分析资料区域适应性评估对地气过程研究、气候分析等具有重要意义。论文利用宁夏24个气象观测站的平均气温,从2种空间尺度(0.5°×0.5°和0.1°×0.1°)和2种时间尺度(年、月),采用偏差、绝对偏差、均方根误差和相关系数等多个统计指标,评估了再分析资料对宁夏地区地面气温的反映能力。结果表明：① GHCN-CAMS(Global Historical Climatology Network and the Climate Anomaly Monitoring System)和CMFD(China Meteorological Forcing Dataset)2套再分析资料对宁夏气温的反映能力整体上均较强,前者对宁夏气温略高估,后者略低估;② 年和月2种时间尺度上,2种尺度的再分析资料存在阶段性正偏差和负偏差,且年尺度上的相关性好于月尺度的;③ 2套再分析资料对下垫面主要为农田(压砂种植)的气温均存在冷季高估、暖季低估的情况,对城镇两者总体均低估,对草地整体表现为CMFD在冷季低估、暖季略高估,而GHCN-CAMS在冷季高估、暖季低估。总体看,空间分辨率较高的CMFD再分析资料对宁夏气温的反映能力更好一些。

Comparative evaluation of the ability of GHCN-CAMS and CMFD reanalysis data to reflect regional temperature in Ningxia

Due to the lack of enough in-situ observation data, there are uncertainties in geoscience research and climate research. Since the 1990s, several sets of reanalysis data have been developed, such as the NCEP/NCAR reanalysis data, ECMWF reanalysis data, and so on. These reanalysis data can effectively make up for the shortcomings of the uneven spatial and temporal distribution of observation data. However, different reanalysis data have different quality and inconsistent performance in different regions. Therefore it is of great significance to carry out the regional suitability assessment of reanalysis data for the study of geo-atmospheric process and climate analysis. Based on the monthly average temperature of 24 meteorological stations in Ningxia Hui Autonomous Region (Ningxia), the reanalysis data of GHCN-CAMS (Global Historical Climatology Network and the Climate Anomaly Monitoring System) and CMFD (China Meteorological Forcing Dataset) were evaluated for the ability to reflect the surface temperature at two spatial scales of 0.5° × 0.5° and 0.1° × 0.1° and two temporal scales of annual and monthly, using statistical indices of bias, absolute bias, root-mean-square error, and correlation coefficient. The results show that: 1) The reanalysis data of GHCN-CAMS and CMFD all have strong ability to reflect the temperature in Ningxia as a whole, the former slightly overestimates the temperature in Ningxia, and the latter slightly underestimates the temperature. 2) The reanalysis data of the two spatial scales have periodic positive and negative deviations at the annual and monthly scales, and the correlation at the annual scale is better than that at the monthly scale. 3) GHCN-CAMS and CMFD show different ability to reflect the temperature under different underlying surfaces. The temperature of farmland (gravel-mulched field) is overestimated in the cold season and underestimated in the warm season. For urban land, its temperature is overall underestimated. The temperature of grassland is underestimated by CMFD in cold season and slightly overestimated in warm season; while overestimated by GHCN-CAMS in cold season and underestimated in warm season. On the whole, CMFD have a better ability to reflect the surface temperature in Ningxia.