Methods for determining the distribution of rainfall in a given area within the GIS were created form desire that information about the natural characteristics of the area are digitalized. This work aims to show and explain the procedure to make a map of the spatial interpolation of rainfall in the catchment area of the river Mirna and the entire Istrian peninsula. The method used is multiple linear regression with input data of annual rainfall at meteorological stations of the Istrian peninsula. The information used on the precipitation are from the period from 1961 to 1990 collected from DHMZ. As a dependent variable it was used annual rainfall, while the independent variable were longitude, latitude, altitude and distance from the sea. Multiple linear regression was performed first for all data from meteorological stations of the Istrian peninsula, then for the division of Istria in the climate zones of mean annual precipitation. From the results we conclude that it is better to use the first option to obtain mean amount of precipitation over large areas, while another at specific locations such as the Mirna river basin, which is also treated in the paper. This was pointed out by statistical analysis of the data processed in the work, and regression analysis to climatic zone IV. The area with the lowest predicted precipitation is the western part of Istria, the largest predicted rainfall we found on the slopes of mount Učka and Ćićarija. In the basin of the river Mirna, area with the most rainfall the area around the town of Hum and least near the town of Novigrad at the mouth of the river.