Comparing laboratory and airborne hyperspectral data for the estimation and mapping of topsoil organic carbon: Feature selection coupled with random forest

•The capability of airborne hyperspectral image to map SOC was investigated.•Laboratory and airborne hyperspectral data were compared to estimate SOC.•Feature selection resulted in improved SOC predictive accuracy.•The CWT-RF method is a simple and effective model for airborne spectra.•Airborne hyperspectral image offered valuable input for digital soil mapping.