The Geographic Information System (GIS) is an effective and reliable tool in estimation of spatial distribution of environmental variables. It is well-known that spatial interpolation methods are commonly used for estimating temperature or precipitation when climate stations are few and widely separated. This research aimed to implement and compare the accuracy of different interpolation methods including Inverse Distance Weighted Averaging (IDWA), regularized and tension spline, spherical, circular, exponential and Gaussian kriging for interpolating yearly precipitation of the Fars province in the south of Iran. Long term average of yearly precipitation of synoptic meteorological and rain gauge stations of the study area have been used. As no single method yields an optimum estimation for all regions, Cross Validation (CV) was used to check which method provides the best estimations. Various statistics including Mean Absolute Error (MAE), Mean Bias Error (MBE), Root Mean Square Error (RMSE) and Correlation Coefficient (r) were considered to evaluate the performance of the interpolation methods used. The results revealed that the exponential kriging was associated with less errors compared to other methods and that it could reasonably predict long term average precipitation (RMSE = 83.5 mm and MAE = 60.0 mm). The tension spline, circular and spherical kriging were also performed well and could be used as the second order of priority. The results showed that the application of the Gaussian kriging, IDWA and regularized spline approaches generated more errors compared to other methods.
S. Bazgeer, E.A. Oskuee, M. Hagigat and A.R. Darban Astane, 2012. Assessing the Performance of Spatial Interpolation Methods for Mapping Precipitation Data: A Case Study in Fars Province, Iran. Trends in Applied Sciences Research, 7: 467-475.