STUDIES ON CHAOTIC PROPERTY OF BROWN PLANTHOPPER (Nilaparvata Lugens St(?)l) OCCURRENCE SYSTEM AND BPH'S FORECASTING MODELS

The problem of the influence of insect pest occurrence changes and anomalies on the development of economics and societies has drawn more and more attentions not only in the worldwide scientific circle but also in many governments. Brown Planthopper(BPH), Nilaparvuta Lugens(st鈒), is an important migration insect pest of rice corps both in tropical and temperate area in the East and South Asia. BPH has become an increasingly serious problem since the 1970s in China. Outbreaks have increased in frequency and the area regularly infested has extended into the Jiangsu province (between the Yangtse and l-luaihe Rivers) and north of the Huaihe River. On average, some tO?0 million hectares of the crop are likely to be affected, with an annual loss of some haifa million tones of grain. In 1991, the worst year on record, severe damage extended over the whole rice-growing region and 17.3 million hectares of paddy fields were infested by Brown Planthopper. in spite of great efforts at prevention and control, more than 2 million tones in 1991, were still lost to these infestations. These greatly affect the long-term development of economics and societies in China. Despite the substantial manpower and material resources is invested to study the BPH occurrence laws and forecast, the long-term forecasting power is still lower in these days. So what about solving the dilemma problem of insect prediction, it has been turning into the concerned hotspot for entomologists. In this paper, based on Chaos Theory and Fractal Theory, we study the analysis method of chaotic diagnosis, and analyses the property and laws in Brown Planthopper occurrence system. We also study the varied methods of Brown Planthopper occurrence forecast by combining the statistics theory, the chaos theory and the neural networks theory of nonlinear science. Firstly, (1) Many researches on Brown Planthopper, Nilaparvata Lugens(st醠) in biology, ecology, migration mechanism, migration rule and source of Brown Planthopper fields, are described briefly. (2) We present a brief review of some principal cOncepts in the theory of chaos and mention the chaotic activities of the systems in insects. Also we introduce some methods in the research on problems in insects by the theory of chaos. (3) we generalize and evaluate pest forecasting methods and its limitations. Probability and unsteady factors of long- term pest forecasting is discussed and iv suniznarized. Those factors are U) the difference between theoretic and practical pest occurrence systems; @) the precision and integrality of data in all subsystems; ?the method, condition and error of calculation; ?the inherent randomicity in nonlinear System of pest occurrence. We paiticularly expatiate the modeling principle and steps of forecasting methods of atificial neural networks (ANN), phase space reconstruction and wavelet transforms based on nonlinear theory developed in recent years and their application in pest forecasting, and foresee their development trend. Secondly, Properties of Brown Planthopper occurrence system based on methods of autoeorrelation function, power spectrwn figure, phase trajectory figure, Poincare section, and return map, are preliminary analyses. Even though the iifluence of random noises, it is distinguished that Brown Pl&ithopper occurrence system has chaotic property. Based on above-mentitmed conclusion, time series data from June to November on...