地理科学 ›› 2005, Vol. 25 ›› Issue (2): 233-238.doi: 10.13249/j.cnki.sgs.2005.02.233

• 研究报道 • 上一篇    下一篇

基于PRISM和泰森多边形的地形要素日降水量空间插值研究

朱求安, 张万昌, 赵登忠   

  1. 南京大学国际地球系统科学研究所, 江苏, 南京, 210093
  • 收稿日期:2004-01-04 修回日期:2004-07-01 出版日期:2005-03-20 发布日期:2005-03-20
  • 基金资助:
    国家重点基础研究发展规划项目(2001CB309404);海外青年学者合作研究基金(40128001);教育部科学技术重点项目(2001)。

Topography-based Spatial Daily Precipitation Interpolation by Means of PRISM and Thiessen Polygon Analysis

ZHU Qiu-An, ZHANG Wang-Chang, ZHAO Deng-Zhong   

  1. International Institute for Earth System Science, Nanjing University, Nanjiang, Jiangsu 210093
  • Received:2004-01-04 Revised:2004-07-01 Online:2005-03-20 Published:2005-03-20

摘要: 以黑河流域河西走廊中段地区为例,利用该研究区年、月降水与地形间较强的相关性特点,在PRISM方法的基础上对该地区日降水量进行了空间插值计算。文章提出了以月降水量的PRISM空间插值结果为该月逐日降水空间分布的参考本底,利用泰森多边形方法确定空间日降水的概率,从而实现黑河流域河西走廊中段地区日降水的空间制图方法,并对该方法得到的日降水时空数据集进行了误差分析和评估。分析结果表明,这一方法简单可靠,满足分布式水文模型或相关陆表过程分布式模拟对分布式日降水数据时空精度的要求。

Abstract: The spatial interpolation of monthly and daily precipitation over the Heihe region in the middle range of the Hexi corridor was systematically studied, and a new methodology based on PRISM approach in association with the utilization of Thiessen polygon analyses to counter the problems existed in daily precipitation spatial interpolations was proposed and validated with the meteorological data obtained in 1988 from 28 stations in the study region.In order to upscale the precisely interpolated yearly or monthly precipitation results derived from PRISM applications to the daily ones in the study region, a transferring algorithm was developed based on the fact that spatial regime of the specific day’s precipitation in the specific month is generally similar to that of the monthly precipitation spatial patterns of that month under study except some difference for some specific locations without precipitation in daily precipitation spatial distributions.For accurately locating the area where no precipitation take place in specific days over the study region, Thiessen polygon approach was utilized to determine the probability of daily precipitation in each pixels.The performance of the proposed methodology was validated for 6 stations, and the derived daily spatial precipitation over the study region was statistically analyzed in both temporal and spatial scales.The detailed analysis indicates that the method is simple and reliable and is able to meet the precision need of the distributed daily precipitation in the distributed hydrological process studies or relevant distributed land surface process simulations.

中图分类号: 

  • TP311.132.4