Development of a 250-m resolution available soil water map using soilgrids and pedotransfer functions

Abstract

The assesment of available water content in soils is of an broad interest in agricultural practice as well as for crop models used for developing early warnings and decision support systems. SoilGrids is a system for global digital soil mapping that uses state-of-the-art machine learning methods to map the spatial distribution of soil properties across the globe at a spatial resolution of 250 meters. The aim of the study was to develop 250-m resolution available water content (AWC) map of Republic of Serbia that can be used for wide range purposes. Several soil properties (sand, silt and clay fractions, bulk density and organic matter) data were directly extracted from SoilGrids maps and used for the pedotransfer functions to obtain AWC in soil for the depths of 5, 15, 30 and 60 cm. Studied area cover territory of the Republic of Serbia (88361 km2 ). Results showed that soils with AWC <16% vol. cover 0.4% of the study area, as well as soils with AWC >22% vol. The largest areas (58.9%) cover soils with AWC 16-18% vol., followed by soils with AWC 18-20% vol. (35.2%) and those with AWC 20-22% vol. (5.2%). Because analytical methods of obtaining AWC and other soil hydraulic properties are expensive and time consuming, this rapid approach is considered as valuable for agricultural and environmental practice and modeling, both on the regional and national level.