Modeling of the Relationship between Annual and Seasonal Precipitation and Geo-Climate Factors in Kurdistan Province

To model the relationship between three Geo-Climate factors of altitude, latitude and longitude and annual and seasonal precipitation in and Kurdistan Province, monthly data from 83 synoptic, climatology and pluviometer stations for a period of 30 years (1985-2008) were used. Statistical defects of stations in 30 years were completed with the most consistent station based on regression model. Model of choice for establishing this relationship was multiple-regression model using stepwise method of variables. Validity of models was also evaluated by testing four hypotheses of linearity of the relationship, lack of correlation of model errors; normality of remnants and constancy of variance of the remaining were evaluated. The results of implementation of this model on the annual and seasonal precipitation indicated the different combinations of each of these three variables on the spatial distribution of seasonal and annual precipitation. So that, the combination of the two variables of longitude and latitude, respectively justifies 69, 66 and 46% of spatial variability of autumn precipitation, annual precipitation and spring precipitation. The combination of the two variables of longitude and latitude explains about 63% of spatial variability of summer precipitation and combination of the three variables of longitude, latitude and altitude explains 47% of precipitation spatial variability in the winter.