Identifying regionalized co-variate driving factors to assess spatial distributions of saturated soil hydraulic conductivity using multivariate and state-space analyses

Abstract

Saturated soil hydraulic conductivity (Ksat) is a key factor in hydrological management projects and its variability along the landscape hinders its correct use in the formulation of such projects. Ksat varies under different climatic and hydrological conditions at spatial scales as reported in several studies. However, co-regionalization of Ksat remains a challenging aspect with regard to identifying supportive co-variates and suitable spatial models. The objectives of this study were to (i) identify factors that relate Ksat with soil and topographic attributes and land-use systems along a 15-km transect using principal component analysis, and (ii) describe the spatial continuum of Ksat across the transect through co-regionalization with autoregressive state-space models. The transect was established in the Fragata River Watershed (FRW), Southern Brazil. One hundred soil sampling points were distributed along the transect at equal distances (150 m). Clay and sand fractions, soil organic carbon content, soil bulk density, soil macroporosity, Ksat, and the soil water retention curve were determined for the 0–20 cm layer at each point. Topographic attributes were derived from the digital elevation model and a land-use map was derived from satellite images. The highest and lowest spatial variabilities were exhibited by Ksat and soil organic carbon content, respectively. Applying the state-space approach, spatial relationships among Ksat and soil and topographic attributes, and land-use systems along the transect, could be found. Principal component analysis used jointly with state-space showed that macroporosity could be used as a proxy to estimate the spatial variation of Ksat in the FRW watershed, assessing surface and subsurface runoff potentials at areas of different land-use. Further studies should be carried out to investigate the use of the type of land-use system as a soil structural predictor of the spatial variations of Ksat at the watershed scale since it is nowadays an “easy-to-measure” variable from satellite images.