干旱气象 ›› 2019, Vol. 37 ›› Issue (4): 529-539.

• 论文 • 上一篇    下一篇

黑河流域气温和降水再分析数据的不确定性评估

赵静学, 郭枝虾, 和鑫磊, 徐同仁, 刘绍民, 徐自为   

  1. (地表过程与资源生态国家重点实验室,北京师范大学地理科学学部,北京100875)
  • 出版日期:2019-08-30 发布日期:2019-09-04
  • 作者简介:赵静学(1995— ),女,山东滨州人,硕士研究生,主要从事水文模型及生态系统水碳通量研究. E-mail:zhaojingxue1995@163.co
  • 基金资助:
    国家自然科学基金(41531174;41671335)资助

Uncertainty Assessment of Temperature and Precipitation Reanalysis Data in Heihe River Basin#br#

ZHAO Jingxue, GUO Zhixia, HE Xinlei, XU Tongren, LIU Shaomin, XU Ziwei   

  1. (State Key Laboratory of Earth Surface Processes and Resource Ecology, Faculty of
     Geographical Science,Beijing Normal University, Beijing 100875, China)
  • Online:2019-08-30 Published:2019-09-04

摘要:

评价区域再分析数据的精度和不确定性对于陆面过程模拟和气候变化分析有重要意义。以黑河流域为研究区,基于站点观测数据对中国区域高时空分辨率地面气象要素驱动数据集(CMFD)、黑河流域2000—2015年大气驱动数据集(WRFOUT)及黑河流域3 km 6 h时空分辨率模拟气象强迫数据(SFD)在不同下垫面条件下气温和降水的模拟精度进行评价,并通过广义三角帽法(TCH)评价3套数据集的不确定性。结果表明:(1)对于气温数据,WRFOUT数据集在草地、灌丛地、荒漠裸地及湿地下垫面精度较高,而CMFD数据集在农田下垫面精度相对较高,5种下垫面中,3套再分析数据的气温产品在灌丛地精度最高;对于降水数据,CMFD数据集在5种下垫面精度均较高,SFD数据集在草地、灌丛地和荒漠裸地,WRFOUT数据集在草地、农田和湿地表现出高估。(2)CMFD与WRFOUT数据集中气温数据的不确定性相对较低,而SFD数据集中气温数据的不确定性相对较大;3套数据集中降水数据均表现出相对较高的不确定性,且区域差异明显,地表植被类型复杂和海拔差异较大的地区数据不确定性较高。

关键词: 气温, 降水;精度评价;不确定性;黑河流域

Abstract:

Evaluating the accuracy and uncertainty of regional meteorological data sets were very important for the simulation of land surface process and analysis of climate change. Based on the sites observation data, the simulation accuracy of air temperature and precipitation of the China meteorological forcing dataset (CMFD), the atmospheric forcing data from 2000 to 2015 in the Heihe River Basin (WRFOUT) and the simulated forcing dataset of 3 km/6-hour in the Heihe River Basin (SFD) under different underlying surface types were evaluated, and uncertainty of  three reanalysis data sets were evaulated through three cornered hat (TCH) method. The results were as follows:(1) For temperature data, the WRFOUT data set had the highest accuracy in grassland, shrub land, desert bare land and wet ground surface, while the accuracy of CMFD data set was relatively high under the cropland, and among the five underlying surfaces, the temperature values of the three sets of reanalysis data had the highest accuracy in the shrub. For precipitation data, the accuracy of the CMFD data set under the five underlying surface types was higher, SFD datasets were overestimated in grassland, shrub land and desert bare land, while WRFOUT datasets were overestimated in grassland, farmland and wetland. (2) Through the TCH method, the uncertainty of temperature data in CMFD and WRFOUT data sets was lower, while the uncertainty of SFD data set was relatively large. The uncertainty of precipitation data of the three data sets was larger, and it was higher in the area with complex vegetation types and elevations.

Key words:  temperature, precipitation, precision evaluation, uncertainty, Heihe River Basin

中图分类号: