A study on temperature interpolation methods based on GIS

This paper compares five spatial interpolation methods of inverse distance weight, Spline, Kriging-exponential, Kriging-spherical, Kriging-Gaussian in temperature interpolation, utilizing GIS software spatial analysis functions. The research data are the minimum temperature, mean air temperatures and maximum temperature of the first ten days of January, April, July and October from 1990 to 2005 in 327 weather stations of China. The result shows that Kriging-exponential and Kriging-spherical interpolation methods are the highest-accuracy methods, inverse distance weight method is less accurate, and Kriging-Gaussian and Spline interpolation methods have the lowest accuracy. Meanwhile, during interpolating four seasons´ temperature, the results of the interpolation error of these five methods are: Error_summer < Error_autumn < Error_spring < Error_winter. When the number of interpolation points is certain, there is a negative correlation between the precision of interpolation methods and standard deviation of the interpolation points.