Digital soil mapping of key secondary soil properties using pedotransfer functions and Indian legacy soil data

•A national-scale database for soil function parameters created.•Digital maps for water and nutrient holding capacities created for Indian soils.•Water and nutrient holding capacities linked to environmental covariates. Water and nutrient holding capacities are primarily controlled by the secondary soil parameters of water retention characteristics and cation exchange capacities (CEC). Complete suite of these secondary soil parameters is rarely included in legacy soil databases generated in soil survey and mapping efforts. Therefore, pedotransfer functions (PTF) are generally used to derive these parameters. We used PTFs and laboratory-measured water contents at field capacity (θFC) and wilting points (θPWP) and CEC values to augment the existing Indian soil legacy database that contained CEC data for 1681 locations and θFC and θPWP data for about a 3rd of these locations across the country. A quantile regression forest (QRF) algorithm was used to map these three soil parameters at six GlobalSoilMap-specified soil depths. Estimated coefficient of determination (R2) values in the 10-fold cross-validated datasets ranged from 0.60 to 0.65 for CEC while a substantially lower accuracies were found for the water holding capacities (R2 for θFC: 0.27–0.34 and R2 for θPWP: 0.33–0.37). Resulting soil function maps show distinct spatial patterns: high values for the water and nutrient holding parameters in soils developed from basaltic parent materials of central and part of peninsular India and low values in soils developed in the deserts of Western Plains and in soils developed from granite-gneiss parent material commonly seen in many eastern Indian region. Terrain attribute showed stronger influence on the distribution of these secondary soil parameters than the climatic parameters. Specifically, valley depth showed strongest negative correlation coefficients (r: −0.68 to −0.73) with layer-wise CEC values; January temperature showed strongest influence on water holding capacities (r: 0.08 to 0.62 with θFC and 0.14 to 0.62 for θPWP). Thus, CEC in Indian soils is primarily driven by the pedogenesis of clays in which monsoon rain favors clay translocation to deeper soil layers followed by subsequent deposition in subsurface horizons during the post-monsoon dry weathers. Similarly, strong linkages between January temperature and water holding capacities reaffirm the conclusion that temperature of driest quarters is a key driver of water relationships in soil.