Previous research has shown the importance of oxalate-extractable aluminium (Alox) for predicting soil organic carbon (SOC) contents across diverse geographical regions. However, studies using data from humid continental climates are scarce, and the data used in these studies have not been statistically representative for larger scales. Our study aimed to 1) evaluate the influence of soil physical and geochemical properties (specifically Alox), farm management, and climate on the spatial distribution and storage potential of SOC in Swedish agricultural soils and 2) to assess whether estimates of aggregation, assumed to influence the protection of soil organic matter, could improve predictions. We analyzed a statistically representative subset of mineral soils with pH < 7 from the Swedish soil and crop monitoring program, which covers the country's agricultural land. We identified the most important predictors for topsoil SOC contents using a random forest model. We employed partial dependence plots to visualize and interpret the interactions between key variables and SOC contents. Results showed that Alox was the most important predictor for SOC contents, as evidenced by its high relative importance score and the increased out-of-bag error when removed from the model. Notably, SOC content reached a plateau at Alox contents of about 3.5 g kg−1, suggesting the possibility of SOC under-saturation. Climatic variables were of secondary importance, while farm management did not emerge as a significant predictor. Surprisingly, silt-sized aggregation was not identified as an important variable for predicting SOC content. Our findings emphasize the importance of incorporating geochemical properties, particularly Alox, in addition to soil texture, in predictive modelling and monitoring efforts for enhanced soil carbon management in humid climates.
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号