Effect of sampling classification patterns on SOC variability in the red soil region, China

Abstract

Field sampling remains a vital step for assessing the spatial distribution of various soil properties. In regional soil surveys, unclassified grid-based mode (Grid), soil type-based mode (SoTy), land-use pattern-based mode (Lu), and land-use pattern–soil type-based mode (Lu–SoTy) are all used popularly in allocating sampling points. However, few studies have identified which mode is most effective for sampling point allocation in SOC studies. This study on the effectiveness of sampling modes was conducted in Yujiang County, Jiangxi Province, China, where red soil prevails on hills. A total of 561 soil samples were collected on the basis of 2 km × 2 km grids. Then the collected samples were classified according to the four aforementioned modes respectively where SOC spatial variability data was compared and assessed. The coefficient of variation (CV) of SOC from the classification of Grid was 47%. Mean CVs from the classification of SoTy were 46%, 38%, and 34% at the soil great group, subgroup, and family categories respectively. The mean CV of SOC from the classification of Lu was 42%, and the mean CVs from the classification of Lu–SoTy were 41%, 37%, and 31% at soil great group, subgroup, and family categories, respectively. The mean CVs of SOC from the three latter classification modes were all lower than that from the classification of Grid. The widest decrease in CV was found in the classification of the Lu–SoTy case. A decrease in the mean CV of SOC indicates a decrease in the uncertainty of sampling point allocation for getting SOC spatial distribution. For soil survey in China's hilly red soil areas with complicated soil types and diversified land-use patterns at a county scale, the most effective way to assess the spatial distribution of SOC is to allocate sampling points based on Lu–SoTy.