Influence of different data densities on the mapping of agroclimatological parameters - case of precipitation

Abstract

Different spatial densities of precipitation data were used to illustrate the influence of data density on spatial information about this parameter. Four maps of long term precipitation data (1961-1990) for Slovenia were produced: on a base of all, half, quarter and eighth of data. Comparing the four maps we can see how we lose the spatial information by decreasing the number of stations. The basic shape of precipitation amount pattern remains the same, but with every reduction of data we lose some details. The ordinary kriging technique was used for spatial interpolation of precipitation data. By using auxiliary variable the precipitation maps could be improved. In our case we used for additional information the elevation data in 1 km grid, due to high correlation between the average annual amount of precipitation and elevation. For spatial interpolation in this case the interpolation technique kriging with external drift was used.