气候因子对靖远松叶蜂暴发的影响

运用SAS8.2中的主成份分析法分析了1988~1997年的气象数据,研究了温度、湿度、降雨量与靖远松叶蜂发生的关系,用实验数据对统计分析结果进行了解释,并运用典型判别分析建立了非时滞和时滞预报模型,进一步研究各种气候因子的综合作用.结果表明,10月份较高的气温与较高的湿度有利于靖远松叶蜂大量结茧,为来年的暴发提供虫源;靖远松叶蜂小幼虫的死亡率与温度呈显著线性正相关(r=0.4,P=0.036),与相对湿度呈弱的线性负相关(r=-0.25,P=0.147),即7月较高的温度与较低的湿度有利于靖远松叶蜂的暴发.1991年靖远松叶蜂开始小规模暴发与1990年初霜期异常推迟有关,8月份过多或过少的降雨可能会妨碍靖远松叶蜂病毒的流行,从而有利于靖远松叶蜂大规模暴发.非时滞预报模型回代正判率为81.8%,时滞预报模型回代正判率为72.7%.典型判别函数表明,温度因子是决定靖远松叶蜂暴发的最主要气象因子.图3表4参9

Effect of Climate on Outbreak of Diprion jingyuanensis(Diprondae, Hemenoptera)

The meteorological data from 1988 to 1997 were analyzed using principal component analysis (PCA) by SAS8.2 to find out the relationship between the outbreak of Diprion jingyuanensis and climate factors. The statistic results were explained by experimental data. One-year lag and non-lag predication models were constructed by canonical discriminant functions for the analysis of integrated effects of climate factors. The results showed that high temperature and relative humidity in October benefited the spinning of cocoons, and led to the outbreak of the pine sawflies in the next year. Mortality of 1~ st ~3~ rd instar larva was positively linear correlated with the mean temperature in continuous 3 days (r= 0.4, P=0.072) and negatively linear correlated with the mean relative humidity in continuous 3 days (r=-0.25, P=0.274). As a result, the higher temperature and the lower relative humidity in July contributed to the outbreak of the pine sawfly. The initiation of the small scale outbreak of D. jingyuanensis in 1991 might have some relations with the remarkable delay of first frost in 1990. In addition, it was possible that too much or too little precipitation in August hindered the epidemics of NPV disease of the sawfly that could cause the large scale outbreak of the sawfly. Corrected discrimination ratio of one-year lag prediction model was 72.7 and non-lag prediction model was 81.8 when the meteorological data from 1988 to 1998 were input to these models. The coefficients of canonical discriminant functions showed that the temperature was the most important factor among meteorological factors to forecast the population of the pine sawfly. Fig 3, Tab 4, Ref 9