Estimation of total organic carbon storage and its driving factors in soils of Bavaria (southeast Germany)

Precise estimations of soil organic carbon (SOC) stocks at large spatial scales are a precondition for national SOC inventories but challenging due to the high spatial variability of SOC. In this study, a comprehensive data set of 1460 soil profiles completely sampled down to the parent material or at least to a depth of 1 m was used to spatially predict SOC stocks for the state of Bavaria in southeast Germany using a geostatistical modeling approach. The model predicted SOC stocks of the main land uses cropland, grassland and forest with an explained variance of 52% of the total SOC variability within Bavaria. The most important factors, which control the spatial variability of SOC storage, were land use, soil type, soil moisture (indicated by the topographic wetness index) and climate (precipitation, temperature). An analysis of the generated SOC map showed that low to medium SOC stocks within the largest part of Bavaria were explained by land use whereas areas of high SOC stocks in floodplains along rivers, bogs and mountainous regions in the Alps and low mountain ranges were related to soil moisture, soil type and climate. A total SOC stock of 760 Mt was calculated for Bavaria with 223 Mt (29%) in cropland soils, 125 Mt (16%) in grassland soils, 257 Mt (34%) in forest soils, 7-29 Mt (1-4%) in bogs and 159 Mt (21%) under other land uses. In view of high SOC stocks in floodplains and mountainous areas, major anthropogenic disturbances of respective soils (e.g. intensification of the land use) should be avoided in these regions.