[Spatial Variability of C-to-N Ratio of Farmland Soil in Jiangxi Province]

Spatial variability of soil carbon-to-nitrogen ratio (C/N) at the provincial scale was analyzed using ordinary kriging methods. The effects of the factors influencing C/N were quantified by regression analysis based on 16,582 points of surface soil samples (0-20 cm) collected during the project of soil-test-based formulated fertilization in Jiangxi Province in 2012. The results showed that soil C/N ranged from 2.98 to 52.67, with an average of 11.72. The coefficient of variation was 25.17%, suggesting moderate variability. The nugget-to-sill ratio was 88.44%, meaning that the stochastic factors played a more important role in the spatial variability of soil C/N between the structural and stochastic factors. The spatial distribution of soil C/N was relatively smooth and the high-value areas were mainly distributed in Pengze County-Jiujiang City, Shangli County-Pingxiang City, and Lean County-Fuzhou City. The terrain factors, farmland-use type, parent material, soil type, and the level of nitrogen fertilizer had significant impacts on the spatial variability of soil C/N (P<0.05), but the degree of influence was different for each factor. Soil C/N indicated a significant positive Pearson's correlation with elevation and the slope of slope (P<0.05). The terrain factors explained 0.3% of the spatial variability of soil C/N and the farmland-use could explain 1.4%. The explanatory power of soil groups, subgroups, and soil family were 2.7%, 3.6%, and 5.5% respectively. The level of nitrogen fertilizer could explain 33.4% of the spatial variability of soil C/N, which showed that the amount of nitrogen fertilizer was the main factor that controls the spatial distribution of soil C/N.