Improving the Spatial Prediction of Soil Organic Carbon Content in Two Contrasting Climatic Regions by Stacking Machine Learning Models and Rescanning Covariate Space

• Published in: Remote sensing (Basel, Switzerland) Vol. 12; no. 7; p. 1095
• Main Authors: Taghizadeh-Mehrjardi, Ruhollah, Schmidt, Karsten, Amirian-Chakan, Alireza, Rentschler, Tobias, Zeraatpisheh, Mojtaba, Sarmadian, Fereydoon, Valavi, Roozbeh, Davatgar, Naser, Behrens, Thorsten, Scholten, Thomas
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 29.03.2020
• Subjects: Arid regions; artificial intelligence; Artificial neural networks; Carbon; Carbon content; Climate change; climatic zones; Deep learning; Digital mapping; digital soil mapping; Information processing; Iran; Learning algorithms; Learning theory; Machine learning; machine learning models; Model accuracy; Neural networks; Organic carbon; Organic soils; Precipitation; prediction; Predictions; Remote sensing; Sea level; Soil depth; Soil improvement; Soil mapping; soil organic carbon; Soil profiles; Soil properties; soil surveys; Soils; Spatial analysis; spatial block cross-validation; Spatial distribution; Stacking; stacking of models; Studies; Teaching methods; Topography; Iran; France

Understanding the spatial distribution of soil organic carbon (SOC) content over different climatic regions will enhance our knowledge of carbon gains and losses due to climatic change. However, little is known about the SOC content in the contrasting arid and sub-humid regions of Iran, whose comple...