甘蔗生长中后期条螟发生高峰期预测模型

Predictive model of occurrence regularity of Proceras venosatum in sugarcane middle and later stage

为研究甘蔗生长中后期条螟发生高峰期的规律及影响因子,用逐步回归分析方法,研究广东雷州蔗区条螟成虫发生高峰期与主要气象因子及上一代高峰期的相关性,建立条螟第2、3代成虫发生高峰期的预测模型。结果表明,第2代条螟成虫发生高峰期与6月下旬积温、6月上旬日照时数及6月下旬的降雨量显著相关;第3代条螟成虫发生高峰期与第2代高峰期、8月中旬积温、8月下旬日照时数显著相关。预测模型的回测历史拟合度平均为93.8%和90.3%,预测符合度平均为92.2%和95.3%。因此,生产上可用拟合的预测模型对甘蔗生长中后期条螟发生高峰期进行预测。

To study the occurrence regularity of Proceras venosatum and the influence factors and to establish the predictive model of Proceras venosatum＇s occurrence peak in sugarcane middle and later stages, we used stepwise regression method to analysis the relationship of the Proceras venosatum occurrence peak and the main climate factors or last occurrence peak. The result showed that all of effective accumulated temperature in late June, sunshine hours in early June and precipitation in late June have significant relationships with the second occurrence peak. The result also indicated that all of effective accumulated temperature in middle August, sunshine hours in later August and the second occurrence peak have significant relationships with the third occurrence peak. The degree fitting of the model with history data are about 93.8% and 90.3%, respectively. The degrees of accuracy are 92.2% and 95.3%, respectively. Therefore, the predictive model could be used to forecast the occurrence peak in middle and later stage of sugarcane production.