Modelling and mapping organic carbon content of topsoils in an Atlantic area of southwestern Europe (Galicia, NW-Spain)

•We modelled the distribution of soil organic carbon (SOC) in topsoils from NW Spain.•SOC prediction using PLS was better than that obtained by PCR and MLR.•At regional scale, SOC accumulation is mainly influenced by climate.•At local scale, both geology and land-use play a significant role in SOC accumulation.•We estimated that epipedons (< 30 cm) store about 0.55 ± 0.12 Pg C.

Spatial differences in the content of organic carbon in topsoils are closely related to differences in environmental parameters such as climate, land use and geology. The spatial distribution of organic carbon in topsoils in Galicia was predicted using a PLS approach and compared with other more frequently applied regression methods (multiple linear regression, and principal component regression). The accuracy of the estimates has been evaluated using an independent dataset. PLS presented the highest model performance (R2 = 0.78, RMSE = 3.48; ME = 0.05) and shows that the amount of SOC can be explained by a combination of climatic factors, land-use and geology, which operate at different intensities and spatial scales. Precipitation rate is the most influential parameter in the accumulation of SOC. We estimated that the total SOC stored in the upper 30 cm in soils from NW Spain is about 0.55 ± 0.12 Pg C.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slide