论文部分内容阅读
12年粘虫幼虫量与诱蛾量间的统计结果，单纯依据诱蛾量预测幼虫量的方法出入很大，历史拟合率仅66．7％，难以指导药剂防治，分析了幼虫量与谷草把卵块数与卵粒数之间的关系，相关密切，幼虫量与卵块数r＝0.745，幼虫量与卵粒数相关性也很显著．又对历史资料进行分析，幼虫量与6月份气象要素之间的关系也相当密切r=0.746，与6月上中旬相对湿度相关系数r=0.6682，与该两句降雨天数r＝－0.6019，与6月份平均相对湿度r＝－0．6519．用电子计算机筛选出的5个参效因子，制定出幼虫量预测多元回归模型．Y＝191.56－0.0035X1＋0.0004X2－10.943X3＋9.534X4－2．371X5，公式历史回朔拟合率92％，1993～1995年运用准确.
Based on the statistical results of 12 larvae of armyworm larvae and the amount of trap lures, the method of simply predicting larvae based on the amount of lures was very different, with a history fitting rate of only 66.7%, which made it difficult to guide the control of larvae. The relationship between the number of eggs and the number of eggs is closely related, the number of larvae and eggs, r = 0.745, larvae and the number of eggs is also significant correlation. Furthermore, the historical data were also analyzed. The relationship between the larvae amount and the meteorological elements in June was also quite close (r = 0.746), and the correlation coefficient of the relative humidity in the middle and upper reaches of June was 0.6682 (r = -0.6019, r = -0.6019, June average relative humidity r = -0.6519. The five factors screened by computer were used to develop the multiple regression model of larval forecast. Y = 191.56-0.0035X1 + 0.0004X2-10.943X3 + 9.534X4-2.371X5, formula history backhoe fitting rate of 92%, 1993 ~ 1995 accurate use.