Study of precipitation interpolation at Xiangjiang River Basin based on Geostatistical Analyst

摘要

Precipitation interpolation is a difficult and hot spot in water science. The way to ensure the precision of interpolation outcome is the key point. This paper chose monthly precipitation data of 52 precipitation stations in Xiangjiang River Basin from year 1980 to 2008. Geostatistical Analyst Tools in ArcGIS 10.2 were used to interpolate the precipitation data. Cross validation was used to interpolate, and mean absolute error (MAE), root mean square error (RMSE) and t-test were used to test the outcome. Thereafter linear regressions were used to modify the result. The results show that the integration of Geostatistical Analyst and linear regression is able make the result credible and meaningful.