A good understanding the effects of environmental factors on the spatial variety of soil organic carbon density (SOCD) helps achieve a relatively accurate estimation of the soil organic carbon stock of terrestrial ecosystems. The present study analysed the SOCD of 1033 top soil samples (0–20cm) from the Jinjing catchment located in subtropical China. Spatial variability of SOCD was estimated using a geostatistics method and a geographically weighted regression (GWR) model, and the major environmental factors affecting SOCD were also explored. In the present study, SOCD had a moderate spatial dependence and the best-fitting model was exponential with a nugget-to-sill ratio of 60.72% and a range of 182m. Land use types (woodlands, paddy fields and tea fields) and topography (elevation, slope, topographic wetness index (TWI)) affected the spatial variation of SOCD. Mean SOCD in the paddy fields was higher than in woodland and tea fields (3.50 vs 3.24 and 2.81kgCm–2 respectively; P<0.05). In addition, SOCD was generally higher in the valleys of paddy fields (with low slope and high TWI) and the hills of woodland (with high elevation and increased slope). GWR generated the spatial distribution of SOCD more accurately than ordinary kriging, inverse distance weighted, multiple linear regression model, and linear mixed-effects model. The results of the present study could enhance our understanding of the effects of land use and topography on SOCD, and improve the accuracy in predicting SOCD by GWR in small catchments of complex land use and topography.
Effects of soil types and land use - land cover on soil organic carbon density at Madendere watershed
