Key Factors Contributing To Population Of Dendrolimus Punctatus Under Drought

Masson pine caterpillar, Dendrolimus punctatus (Walker), the most serious pest of massonpine (Pinus massoniana) in South China, was reported outbreak frequently and caused the mostserious economical loss. As one of the most important factors contributing to the outbreak,drought changed the condition of masson pine and made it vulnerable to the insect infestation.To find out what the host plant had changed has the theoretical and practical significance forpests’predicting and controlling.Eight spots with different defoliation levels in host plants were tested. Carbon, nitrogen andwater contents in soil were also determined. The cocoons and larvae of D. punctatus werecollected to test the relationships between pests and plants or soil. From the life history recordsof D. punctatus in Guangxi,71related meteorological factors with different levels were selectedto analyze the correlations with the outbreak. Dual selective stepwise regression analysis wereused to find out the most related climatic factors contributing to the10years records of D.punctatus in the23tested points in Guangxi on three damage levels. A back-propagation (BP)algorithm was applied to predict the reserved points in Guangxi. And the trained network wasalso used to predict the damage levels in Fujian while adjust the climatic factors by regional lifecycle differences of D. punctatus. The results are as follows:1. Masson pines exhibit a weak position with lower moisture, total nitrogen and organiccarbon in foliage under drought condition in2010. With the results that the soil nitrogen contentwas much higher in the drought year than in the ordinary year, we concluded that one of mostsignificant influence that pines reacted to drought was reducing the ability of concentratingnitrogen from soil.2. Masson pines released much more monoterpinenes under drought when compared withthe normal year. However,3-carene content was lower under drought than the normal year.Considering the negative relation between3-carene and number of larvae and empty cocoons, we conclude that3-carene was the key factor to the first generation after the overwinteringgeneration. The α-pinene and β-pinene content in needles were much higher in the non-outbreakarea. High content of β-pinene in slash pines in same area might explain its resistance for thepest.3. Total number of D. punctatus had a significantly positive correlation with the content oftotal nitrogen in foliage, implying that the significance of the host needle nitrogen content on theoutbreak of D. punctatus. No significant correlation between the number of cocoons (emptycocoons included) and larvae and the tested parameters of plants and soil was observed. Whilenegative relationship between nitrogen content and moisture content of needles means that thestress of lacking of water could raise the content of needle total nitrogen content of masson pinesand reduce the organic carbon content, which may affect the outbreak of pests.4. In a large scale, the most related factors contributing to the slight damage level were theaverage temperature of February, March and July. While the high temperature of January couldaggravate the medium damage level. And the most significant factors affecting the severeoccurrence were the total precipitation of April, May and June, and the temperature of Januaryand March. The temperature of January affected both the medium and severe damage levelmeans that the overwinter generation was the key factors to the next generation and all thedamage levels.5. Artificial neuron network had the advantage on dealing with non-linear issues. By usingthe empirical functions and several tests, the hidden layers of BP network was determined. Thepredicting results of neuron network was meaningful, and that means the factors selected topredicting were meaningful and could reflect the impact of climatic factors to pine caterpillar.But the accuracy were only at68%with the medium level prediction in Fujian while adjust theclimatic factors by regional life cycle differences, the other levels did not show the consistentresults. Anyhow, this research can provide reference for using climatic factors to predict D.punctatus in a large scale under global climate change.