中国西部地区积雪深度的空间插值比较  

Comparison of Spatial Interpolation Methods of Snow Depth in the West of China

作　　者：唐国栋[1] 柯长青[1] 

机构地区：[1]南京大学城市与资源学系,江苏南京210093

出　　处：《遥感技术与应用》2007年第1期39-44,共6页Remote Sensing Technology and Application

基　　金：国家自然科学基金(40301013)项目资助

摘　　要：应用反距离加权法、样条函数法、Kriging法对中国西部地区(79.05-°103.57°E,27.17-°48.05°N)113个气象台站观测的年平均积雪深度进行空间插值比较研究。结果表明反距离加权法和样条函数法的插值结果与积雪深度的实际分布情况有一定的差异;普通Kriging法能够反映出研究区积雪深度分布的空间结构特征,与实际情况比较吻合。影响插值结果精度的主要原因是研究区内气象台站稀少且空间分布很不均匀。可以通过合理的采样设计,将确定性方法和随机方法相结合,并考虑地形、气候等影响积雪分布的因素来提高空间插值的精度。The spatial interpolation methods of Inverse Distance Weighted （IDW）, Spline and Kriging are utilized for comparison study on spatial interpolation of annual average snow depth from 113 observatories in the west of China （79.05°-103.57°E,27.17°-48.05°N）. The principles of these three methods are different from each other. IDW determines cell values using a linear-weighted combination set of sample points. Spline estimates values using a mathematical function that minimizes overall surface curvature. And ordinary Kriging is a powerful statistical interpolation method which assumes that the distance or direction between sample points reflects a spatial correlation that can be used to explain variation in the surface. Compared with the unsatisfactory interpolation results of IDW and Spline, the result of ordinary Kriging is more close to the real snow depth distribution and can represents the spatial structure of snow depth distribution better. The main reasons which affect the precision are the small number of observatories and their asymmetric spatial distribution. However, the accuracy of spatial interpolation can be improved through reasonable design of sampling, combining deterministic and stochastic methods, and considering the influencing factors of snow distribution such as the terrain and climate.

关 键 词：Kriging法 反距离加权法 样条函数法 积雪深度 中国西部地区 

分 类 号：P208[天文地球—地图制图学与地理信息工程]