Gis-based Urban Meteorological Monitoring Elements Interpolation Analysis

The meteorological elements such as precipitation and temperature are closely related to the daily lives of citizens, and also can affect the programming and construction of the urban flood prevention together with Drainage projects, thus to comprehensively understand meteorological elements becomes a focus of all the cities. However, the limited meteorological observations can not meet the requirements of modern societies for the meteorological products, thus, spatial interpolation techniques become the first choice to acquire the meteorological products covering the whole Monitoring Area where has only limited Monitoring Points. Though many researches had been conducted on the interpolation methods and many methods had been developed to interpolate meteorological data, little information about the methods that can more effectively interpolate urban meteorological data.In this paper, we took the Kunming as the study area, and investigated the Suitability of the spatial interpolation techniques in interpolating urban meteorological data based on GIS. Based on the detailed analysis on the commonly utilized meteorological spatial interpolation techniques, we developed an application system for the urban meteorological data interpolation by using ArcEngine, a secondary development components in ArcGIS platform. The system can perform not only the commonly used spatial interpolation, but also the cross validation, and it can calculate the statistical indexes such as the Average Error, mean absolute difference, Mean Square Error and store them in the database of the system. In order to facilitate Interpolation Analysis, we developed the functions of statistic analysis that based on the statistical indexes calculated by the system. With there functions, we can perform the statistical analysis on the results of the all the interpolation methods and cross validation in meteorological data, and furtherly evaluate the Suitability of all the interpolation methods to every meteorological element.The results showed that:Inverse distance weighing (IDW) and Kriging are more suitable to urban meteorological interpolation; IDW is significantly robust than the other methods in interpolate precipitation and successive precipitation, and it is more efficient when there are more monitoring points; Kriging is better in urban precipitation interpolation in case of rainstorm but the results map is not smooth and the problem that the resulting range exceeding its origin range is more serious than Spline; The phenomenon of the result range exceed the its original range is very serious in Spline and results of its cross validation are relatively accurate due to the characteristics of its functions; in addition, the triangle linear interpolation and natural neighbor interpolation are comparatively poor than other methods.In general, the system we developed for evaluating the suitability of the urban meteorological interpolation, basing on the spatial interpolation models in GIS, can facilitate the analysis and visualizing of the urban meteorological interpolation. This paper is helpful to the evaluation on the urban meteorological products by the spatial interpolation in GIS.