Análisis de la distribución espacial de la precipitación anual (1979-2014) aplicando métodos de interpolación en la Región Occidental del Paraguay

Abstract

Less frequent or high intensity rain events followed by long periods of drought put pressure on an ecosystem and can increase the possibility of soil loss due to erosion or increase the vulnerability of communities. The Paraguayan Chaco is a region dominated by livestock use and with extremely vulnerable vegetation that receives little rainfall throughout the year. However, it has very few meteorological stations for monitoring throughout the territory. Thus, geostatistical methods present an opportunity to predict the spatial distribution of precipitation. In this study, daily precipitation data from 1979-2014 were obtained from seven in-situ meteorological stations and combined with reanalysis data from the Climate Forecast System (CFSR). 400 points were selected of which 78 were randomly chosen, with a total of 84 sampling points to interpolate. The ordinary Kriging interpolation was chosen and the theoretical exponential, matern, spherical, linear and Gaussian variograms were compared for the characterization of precipitation. From the comparison of the values ​​obtained in the present study, the model with automatic, exponential, matern, linear and spherical adjustment are those that present a lower error, for which they are considered adequate for estimating precipitation in the Paraguayan Chaco at starting from the sampling carried out.