论文部分内容阅读
于1983～1986年在建平县青松岭20年生左右的油松人工林中,对松毛虫各时期的发生量进行了调查,经定性分析,将影响其数量变动的32个变量因子分为内部因子、特征因子和普通因子。用回归法建立了松毛虫发生量的短中期预测模型,用马尔可夫链法建立了长期预测模型,并对试验区的松毛虫数量发生平衡态进行了探讨。结果表明,下树期、翌年上树期、老熟期虫口数量及蛹量以综合预测和自回归预测效果较好,相关系数可达0.85～0.99,而下代卵量以特征因子、普通因子预测较好,相关系数为0.63。文中建立的模型宜用于生产。
In 1983 ~ 1986 in Ching Chung Ridge Jianping 20-year-old Pinus tabulaeformis plantations, the occurrence of the era of the caterpillar were investigated, the qualitative analysis, will affect the number of changes in the 32 variables are divided into internal Factor, characteristic factor and common factor. The regression model was used to establish the short-and-medium-term prediction model of the occurrence of pine caterpillars. The long-term prediction model was established by Markov chain method, and the equilibrium state of the number of pine caterpillars in the experimental area was discussed. The results showed that under the tree stage, the number of the population and the amount of the pupae in the upper tree stage, the next mature stage were better than those predicted by the comprehensive prediction and the autoregression, and the correlation coefficient was 0.85-0.99. The lower generation ovarian was characterized by the characteristic factor, common factor Good forecast, the correlation coefficient of 0.63. The model established in this paper is suitable for production.