Modeling Spatio-temporal variation of monthly precipitation (Case Study Golestan province)

Precipitation is a highly variable Meteorological and hydrologicalmeteora. Investigate variation of spatio-temporal of precipitation for water resources planning and management in different parts of the catchment is essential.Autocorrelation of this kind of data however makes their analysis complex, but increases their analysis accuracy. The goal of this study is, modeling Spatio-temporal variationand interpolation of monthly precipitation in Golestan province. So, 30-yr monthly precipitation variations for 30 meteorology and rain gauge stations was studied. To analyse spatio-temporal precipitation,variance hemogenity and space/time stationarity was assessed. By examining different models of the Spatio-temporal variogram, SumMetric model with RMSE of 0.14 mm was selected as the most suitable spatio-temporal variogram model. Then, with this model, by different neighbors, interpolation was done.therefore, 360 monthly precipitation distribution maps was produced. Kriging interpolation with 20 neighborhoods minimum estimation error is revealed.The mean of kriging estimation error is close to zero. Based on the results, pattern of variation in precipitation in this maps with observed maps, haveSimilarity that show the suitability of the modeling approach.Results of this study can be used in determining precipitation in zones with no station in every time (under the influence of spatio-temporal autocorrelation).