The USCS Mapping of Continental Croatia using Multinominal Logistic Regression and Inverse Distance Weighting Spatial Interpolation

The Unified Soil Classification System (USCS) is the basis for numerous methods for soil trafficability research. Since USCS data are not always available, various other data and methods are used to correlate and predict the USCS soil group. This paper describes two methods used for the purpose of mapping Continental Croatia (CCro) according to USCS to a depth of 50 cm. In the first method, the possibility of transfer of 308 profile samples according to the International Soil Science Society (ISSS) classification system into USCS was examined. The results show that it is impossible to directly transform ISSS data into USCS. In the second method 414 USCS profiles were used to analyse the weights of factors in the spatial analysis with inverse distance weighting (IDW). The analysis included layers of dominant and associated soil units of the Basic Soil Map of Croatia (BSM), Geological Map of Croatia (GM), drainage and catchment areas. The obtained weights were as follows: BSM 47,12%, catchment area 27,12%, GM 17,67% and drainage 8%. The results showed that CCro is covered with fine-grained soils, with clay covering almost the entire area and silt dominating in the western and north-western parts of the country.