基于地统计学的山区气象要素空间化方法探索——以广西年均降雨量为例  

Method Exploration of Mountainous Area Meteorological Element Spatialization Based on Geostatistics——Take the average annual rainfall of Guangxi as the example

作　　者：苏文静[1] 卢远[1] 

机构地区：[1]广西师范学院资源与环境科学学院,广西南宁530001

出　　处：《广西师范学院学报（自然科学版）》2012年第1期77-81,共5页Journal of Guangxi Teachers Education University(Natural Science Edition)

摘　　要：采用地统计学的克里格插值方法对广西多年平均降雨量进行空间化处理,具体探索了普通克里格、泛克里格与协同克里格插值方法。与普通克里格法、泛克里格法相比,协同克里格法增加了相关因素(纬度、离海岸线的距离、DEM)对降雨量的影响,在应用中更具潜力。地统计插值可以通过交叉验证和验证对话框中的4个指标,以及实测值与预测值的比较评定精度。三种方法综合比较,协同克里格方法最适合广西多年平均降雨量的空间化。This article uses the Kriging interpolation method of Geostatistics to carry on spatialization processing to the mean annual precipitation for years in Guangxi,explores specifically Ordinary Kriging,Universal Kriging with the CoKriging interpolation method.Compared with Ordinary Kriging,Universal Kriging,the CoKriging increases the influence of correlative factors（for example,latitude,coastline distance,DEM） on rainfall amount,which is more potential in application.By statistical cross validation and the four targets of validation dialog,geostatistics interpolation is employed to assess the accuracy.Bringing the three methods into comparison,CoKriging method has the best accuracy.

关 键 词：气象要素 空间插值 克里格 年均降雨量 广西 

分 类 号：P426.6[天文地球—大气科学及气象学]