Estimation of regional soil organic carbon stocks merging classified land-use information with detailed soil data

Sci Total Environ. 2019 Dec 10:695:133755. doi: 10.1016/j.scitotenv.2019.133755. Epub 2019 Aug 3.

Abstract

The Kulunda Steppe in southwestern Siberia is one of the most intensely used agricultural landscapes in the world. Today it is characterized by depletive soil management practices and intensities that are unadapted to the local site conditions. Severe soil degradation and a loss of soil organic carbon (SOC) occurs at agriculturally used areas. So far, only few studies analyzed the spatial extent of these degradation effects and came to inconsistent assessments. We consider that different results arise from different definitions of SOC changes and from different spatial scales. Thus, local soils under different land-uses were sampled to a depth of 30 cm to determine land use dependent changes in topsoil SOC. Site specific soil data was merged with appropriate land-use classifications and soil maps from the mid-1970s and up-to-date data from 2013/2014 to balance land use corrected SOC pools. Here, we use a hierarchical approach to extrapolate local findings to regional and biome scale for typical and dry steppe. At the test area, land-use changed only moderately after the Soviet period. The steppe biomes are also characterized by large areas that are not affected by land-use change and ensuing SOC loss. Agricultural use led to a mean carbon loss of 23.3% for Chernozem soils and 13.9% for Kastanozems. Natural heterogeneities, such as small scale changes in relief or soil type, are missing in small scale maps due to generalization that leads to varying results. The calculated regionwide SOC loss is maximally 15.2% using the small scale map. Following different calculation approaches, according to different map scales, detailed or coarse resolution SOC-stocks differ by up to 59.7%. Consequently, subsequent calculations or modellings may include this uncertainty into consideration.

Keywords: Agriculture; Land-use change; Map scale dependence; Russia; Soil organic carbon; Steppe soils.