Improving the spatial prediction of soil salinity in arid regions using wavelet transformation and support vector regression models

• Published in: Geoderma Vol. 383; p. 114793
• Main Authors: Taghizadeh-Mehrjardi, Ruhollah, Schmidt, Karsten, Toomanian, Norair, Heung, Brandon, Behrens, Thorsten, Mosavi, Amirhosein, S. Band, Shahab, Amirian-Chakan, Alireza, Fathabadi, Aboalhasan, Scholten, Thomas
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.02.2021
• Subjects: agricultural productivity; digital elevation models; Digital soil mapping; Iran; Machine learning; prediction; regression analysis; salinity; soil depth; soil restoration; Soil salinity; Support vector regression; topography; topsoil; uncertainty; uncertainty analysis; wavelet; Wavelet transformation; Iran; Support vector regression; Digital soil mapping; Wavelet transformation; Soil salinity; Machine learning

•Soil salinity was predicted using machine learning algorithms in central Iran.•Wavelet-SVR indicated higher performance compared to the standalone SVR.•The estimated prediction interval for SVR is wider than of W-SVR.•Remote sensing features control the spatial distribution of the salinity widely....