Digital Mapping of the Humus Horizon Thickness in Soils of the Cis-Salair Plain Using the Random Forest Machine Learning Algorithm

• Published in: Eurasian soil science Vol. 58; no. 11
• Main Author: Gopp, N. V.
• Format: Journal Article
• Language: English
• Published: Moscow Pleiades Publishing 01.11.2025; Springer Nature B.V
• Subjects: Algorithms; Datasets; Decomposing organic matter; Digital mapping; Digital maps; Earth and Environmental Science; Earth Sciences; Errors; Genesis and Geography of Soils; Geotechnical Engineering & Applied Earth Sciences; Horizon; Humus; Land use; Land use management; Land use planning; Learning algorithms; Machine learning; Mapping; Meadows; Modelling; Soil formation; Soil properties; Soils; Thickness; Training; predictors; gray forest soils; WorldClim; climate; relief; SAGA GIS; chernozems
• ISSN: 1064-2293; 1556-195X
• DOI: 10.1134/S1064229325601568

Results of digital mapping of the humus horizon thickness (HHT) in the soils of the Cis-Salair Plain using the Random Forest (RF) machine learning algorithm implemented on the Google Earth Engine cloud online platform are reported. A total of 92 predictors are employed to characterize the soil formation factors, including climate, relief, vegetation, spatial position, and soil properties. Training ( n = 718) and validation ( n = 130) datasets are constructed based on the archive materials (1974–1984) of ZapSibNIIgiprozem (Western Siberian Research, Design, and Survey Institute for Land Use Planning). The following indicators of the HHT modeling efficacy using the RF algorithm are obtained: coefficient of determination for training dataset = 0.88; coefficient of determination for validation dataset = 0.12; root mean square error RMSE VD = 9.7 cm; mean absolute percentage error MAPE VD = 24.3%; and mean absolute error MAE VD = 6.5 cm. The modeling accuracy estimated with MAPE VD is satisfactory. Actual data show that HHT varies from 3 to 110 cm with the trend of a decrease from northwest to southeast. The lowest (3 cm) average HHT values are typical of meadow–chernozemic solonetz (Solonetz (Salic)) and the highest (61 cm), of ordinary meadow soils (Mollic Gleysols).