Prediction of water table in areas of ecotone in Doñana National Park with piezometric data, topographical covariates and geostatistical methods

Abstract

Between marshes and stabilised sands, there are the most biodiverse areas of Doñana National Park, an ecotone where the water table is relatively close to the surface and causes the existence of important plant communities that are concentrated in the wetter areas. In the study area, the phreatic groundwater piezometers are irregularly distributed and it is inadequate to characterise different microenvironments spatially, mainly in areas of ecotone and faraway, where there are not piezometers installed near. In this paper, we compare interpolators and geostatistical methods for estimating the water table with piezometric data and topographic covariates. The results show that the covariates, when they are well correlated and better sampled than primary variables, can improve precision and reduce errors of prediction. In this cases, we recommend Co-Kriging over Kriging and IDW.