Mapping of Surface-Effervescing Soils of Arable Lands in the South of the Volga Upland
摘要
The purpose of this work is to create a digital map of surface-effervescing soils based on space imagery and field research data; and to compare soil mapping procedure using remote and ground-based methods. The research was carried out at the Oroshaemaya experimental station (Volgograd Oblast) located in the south of the Volga Upland. The soil cover is represented by light-chestnut (Kastanozems) and solonetz microassociations on arable and irrigated lands. The article traces all stages of creating a digital map based on space images and field research in 2020–2023. Field studies included route work to assess the intensity (class) of effervescence with 10% HCl at the soil surface. Based on the relationship between the spectral characteristics of the open soil surface in high-resolution Pleiades satellite imagery (April 25, 2020) and the intensity of soil effervescence, the raster space images were classified into 4 classes separately for 26 fields using the Random Forest algorithm. The following layers were created from the raster: (1) vector layers of field boundaries with the calculation of the proportion of effervescent soils of different classes in the field; (2) vector layers of polygons inside the fields with different proportions of different classes. To vectorize the polygons of surface-effervescing soils, the QGIS program and functions were used. Estimates of the proportion of effervescing soils within the field obtained only by a ground survey and estimates based on a digital map created by vectorization of the raster classified by the Random Forest algorithm are comparable to each other with a determination coefficient of 0.68–0.83. A digital vector map of polygons, created using a model that links remote sensing data and ground-based sampling data, allows us to determine specific parameters of the soil cover pattern based on the presence of surface-effervescing soils and to assess the anthropogenic transformation of the soil cover using the previously proposed indicator.