农田土壤有机碳动态模拟模型的检验与应用

 利用大田填埋试验资料和连续７年的土壤改良定位试验资料对所建模型进行了检验。结果表明,模拟值与测定值之间具有良好的一致性,大田填埋试验的测定值与模拟值间的线性相关系数（ｒ２）为０．９２９１（ｎ＝４８）,土壤改良定位试验的测定值与模拟值间的线性相关系数为０．６４３１（ｎ＝６５）。模型预测结果指出,若仅靠作物自然还田,宜兴市万石镇农田土壤（白土）有机碳的变化将从 １９８４年的 ７．８５ｇ／ｋｇ下降到 ２０１４年的 ６．３０ｇ／ｋｇ;若每年收获的稻麦秸秆有５０％（约６．７５ ｔ／ｈａ）还田,则同期内农田有机碳含量将从７．８５ｇ／ｋｇ提高到１１．４２ｇ／ｋｇ;若要在２０１４年达到１３ｇ／ｋｇ,则每年需要有 ９ ｔ／ｈａ的秸秆还田。每年每增加 １ｔ／ｈａ稻麦秸秆还田,则可增加土壤有机碳 ０．０２５ｇ／ｋｇ。

Validation and Application of a Soil Organic Carbon Model

A model developed by the authors was validated against independent data sets. The data sets were obtained from field experiments of crop residue decomposition and a 7-year soil improvement in Yix;ng City, Jiangsu Province. Model validation indicated that soil organic carbon dynamics could be simulated from soil clay content and weather variables. Model simulation in general agreed with the measurements. The comparison between computed and measured resulted in correlation coefficients r2 values of 0. 9291(n=48) and 0. 6431(n=65) for the two experiments, respectively. A long-term model prediction suggested that the soil organic carbon in Wanshi Township of Yixing City would be dropped from the initial value of 7. 85g/kg in 1988 to 6. 30g/kg in 2014 if there would be no additional organic matter amendment. Under a rice-wheat rotation cropping system, the soil carbon in this place would increase up to 11. 42g/kg if there would be additional rice shoots and wheat shoots application with a rate of 6. 7 5 t/ ha per year. According to the model simulation, soil carbon content of 13g/kg in 2014 would be achieved if additional rice shoots and wheat shoots were incorporated with a rate of 9 t/ha per year.