地理研究 ›› 2003, Vol. 22 ›› Issue (3): 349-359.doi: 10.11821/yj2003030011

• 论文 • 上一篇    下一篇

中国高分辨率温度和降水模拟数据的验证

朱华忠1,2, 罗天祥1,2, Christopher Daly1,2   

  1. 1. 中国科学院地理科学与资源研究所, 北京 100101;
    2. Spatial, Climate, Analysis, Service, Oregon, State, University, Oregon97331, USA
  • 收稿日期:2002-10-20 修回日期:2003-01-19 出版日期:2003-06-15 发布日期:2003-06-15
  • 作者简介:朱华忠(1969-),男,江苏张家港人,副研究员。1995年获得北京大学地图学与遥感专业理学硕士学 位,现从事遥感、GIS应用研究。
  • 基金资助:

    国家基础研究发展规划(G1998040813);美国林务局全球变化研究项目(USDAForestService00CA11242343017);中国科学院地理科学与资源研究所知识创新项目(CX10GC0002CX10GE010203)

Validation of simulated grid data sets of China’s temperature and precipitation with high spatial resolution

ZHU Hua zhong1,2, LUO Tian xiang1,2, Christopher Daly1,2   

  1. 1. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China;
    2. Oregon State University, 316 Strand Agricultural Hall, Corvallis, Oregon 97331, USA
  • Received:2002-10-20 Revised:2003-01-19 Online:2003-06-15 Published:2003-06-15

摘要:

PRISM模型是一种基于地理特征和回归统计方法生成气候图的模型。基于中国及其周边国家地区2450多个气象台站观测数据,以PRISM模型模拟生成了中国2.5′×2.5′(≈4~5km)逐月温度和降水数据。利用独立于模拟数据的中国生态系统研究网络18个野外观测站的长年气候观测数据,检验了PRISM模型的模拟结果,表明PRISM模型较好地模拟了我国温度和降水的空间分布及季节变化,除了在高山和亚热带地区由于地表覆盖和局部地形的差异影响模拟结果,其模拟值与实测值之间的趋势线同1∶1线基本一致,具有显著相关关系,其中降水效果略差

关键词: 平均最高气温, 平均最低气温, 降水, PRISM模型, CERN生态站

Abstract:

Parameterelevation Regressions on Independent Slopes Model (PRISM) is a statistical geographic approach to mapping climatic factors including monthly maximum and minimum mean temperatures and monthly precipitation PRISM brings a unique combination of physical and statistical concepts to the analysis of orographic precipitation and temperature Based on 30 year (1961~1990) observation data of monthly minimum and maximum mean temperatures and precipitation for 2450 stations from China and neighboring countries, the PRISM model was applied to grid monthly precipitation and monthly minimum and maximum temperatures over China with fine resolution of 2.5 minute latitude by longitude (≈4~5km) The purpose of this study is to introduce and validate the PRISM predicted climate data sets We used the meteorological observation records from 18 CERN stations without joining the PRISM simulations to validate the simulated grid data sets The results indicated that the PRISM model predicted well the geographical distribution patterns and seasonal varations of monthly maximum and minimum mean temperatures and monthly precipitation from 18 CERN stations at an almost one to one relationship except for a few sites in mountainous and subtropical regions where land cover changes and fine to middle orography may influence the distribution patterns of the monthly temperatures and precipitation

Key words: maximum mean temperature, minimum mean temperature, precipitation, PRISM model, CERN stations

PACS: 

  • P468.O+21