1. INTRODUCTION
As an important part of water cycle, precipitation spatial information plays a crucial role in many fields, such as water resource management, drought and flood disaster predication, as well as regional sustainable development. With the development of earth sciences and cross-disciplinary, it is more and more meaningful and valuable to analyze and estimate precipitation spatial distribution. It is unrealistic to get the accurate predication of a specific region, because the formation and distribution of precipitation is a complex process. Therefore, only interpolate with precipitation data of the limited sites in the particular region, can the precipitation throughout the region be achieved. For the complex process, it is a big challenge to build a general precipitation interpolation model.