Using soil properties map in the production of detailed soil maps

Abstract

Background and objectives: Detailed soil maps are an essential tool for achieving sustainable management. Despite the advances in digital soil mapping and efforts to produce accurate maps, insufficient reliability on the soil maps remains on different scales. Recent studies have been focused primarily on remote sensing, which pays less attention to the subsurface diagnostic properties of the soil. This study used a new interpolation method between the subsurface soil diagnostic properties to increase the accuracy of the prepared maps. In this study, the production of soil maps with two approaches including the use of geomorphic surfaces, as well as the interpolation of subsurface characteristics of the soil in the Chalous region, was examined. Materials and methods: The study area with an extent of 100 hectares in Bandar village, was located in the suburb of Kelardasht, in Chalous County, Mazandaran province. The thickness, or depth of the upper/lower boundary of the diagnostic horizons, or soil characteristic properties used as criteria for distinguishing soil map units. According to the Comprehensive American Soil Classification System, there are six influential characteristics (Upper Cambic horizon boundary, argillic horizon upper boundary, calcic horizon upper boundary, Mollic horizon thickness, soil profile thickness, and calcareous or non-calcareous parent materials) in soils at the family level, that were used for numerical interpolation and generate thematic maps. The final soil map of the region was obtained by a combination of these six thematic maps. After mapping each of the soil properties separately, all the prepared maps intersected, and homogeneous map units were obtained. The highlands of this region, are of special importance due to their unique ecosystem and the effect of slope direction on microclimate formation, vegetation changes, and high soil diversity. Grid sampling was carried out from 56 profiles and 44 auger points (to assess variability among profiles) as a grid network with 100 m intervals. To study and describe the spatial structure of the variables, a semivariogram was used. The existence of data trends and heterogeneity were also examined. After preparing the variograms, the selection of the best approach was done using the cross-validation method and RMSS index. Results: To prepare the soil map using the geomorphic surface method, out of 20 soil units were identified, 12 soil units were of the consociation type, and 8 unit were of the association type. However, all 20 map units obtained by interpolation of subsurface diagnostic properties were of consociation type. This method had high accuracy in mapping the boundaries among soil units. The main differences between this method and traditional methods are in the production of detailed soil maps, the use of supervised automatic interpolation instead of manual interpolation, and the use of a set of quantitative indicators. Conclusion: One of the major advantages of this method is the use of internal characteristics of the soil as an auxiliary variable along with other climatic factors, topography, living organisms, parent material and time. This study underscored the importance of the role of factors in the SCORPAN model and showed that the internal properties of the soil, which are directly involved in soil classification, can be effectively used to separate soil map units