摘要
研究4种Bt玉米(34B24、NK58-D1、R×601RR/YG和农大61)秸秆分解释放Bt蛋白的土壤降解动态,分别应用一级动力学反应方程、双指数模型和移动对数模型进行了拟合。结果表明,4种Bt玉米Bt蛋白的土壤降解均呈现前期负指数大量快速降解和中后期极少量稳定两个阶段,双指数模型与移动对数模型的拟合结果比一级动力学反应方程更符合实际。判断模型拟合精度时不能只用拟合结果的统计指标值来确定,图示对比实测值与模拟值的差异是避免应用统计学拟合精度指标很高但并不符合实际的模型错误估算DT50的有效办法。在研究转基因作物秸秆中Bt蛋白土壤降解规律时,应该根据其土壤降解前期较快的特点,增加前期取样的次数,以便更好地对模拟模型求得的安全性评价关键参数DT50的准确性进行科学判断。
The soil degradation of Bt protein from stalks of 4 Bt corn hybrids（34B24, NK58-D1, R × 601 RR/YG and Nongda61） were compared and the first-order kinetic model, biexponential model and shift-log model were used to fit the degradation data set in this study. The results showed that the degradation of Bt protein of all 4 hybrids in soil demonstrated a two-stages pattern, a rapid decline in 14 days and then stable at low concentration for a long time. Both the biexponential model and shift-log model fitted the decay pattern of Bt protein much better than the first-order kinetic model. However, model selection should not only be based on the statistic significant index of the model fitting, but also on the deviation between each simulated value and observed value. It was necessary to increase sampling frequently in the rapid degradation stage so as to make a scientific judgment about the accuracy of DT50 and DT90 calculated from models.
出处
《生态学杂志》
CAS
CSCD
北大核心
2005年第9期1063-1067,共5页
Chinese Journal of Ecology
基金
国家自然科学基金项目(3027027)
广东省自然科学基金资助项目(000569
021043)
