我国甜菜夜蛾间歇性暴发的非均衡性循环波动

摘要] 为进一步了解我国甜菜夜蛾Spodoptera exigua ( Hbner)间歇性暴发成灾的规律，作者应用时间序列分析技术研究了我国甜菜夜蛾间歇性暴发的时间序列波动规律。结果表明，在1949—2008年的60a中，我国甜菜夜蛾大尺度暴发总年频次为120次，各年度暴发次数(年频次)存在着非均衡性循环波动特征，并且在其波动过程中呈明显的上升趋势。按“年频次”强度,可将我国1949—2008年60a中甜菜夜蛾发生过程,划分成两个明显不同的阶段。第一阶段（1949—1984年）为平稳低发阶段，36a总年频次量为4次，年平均仅0.11次；第二阶段（1985—2008年）则为波浪式上升性高发阶段，24a总年频次为116次，年平均4.83次（为第一阶段的43.91倍））。按第二阶段24a（1985—2008）数据进行自相关系数和频谱图分析,结果表明，我国甜菜夜蛾大尺度暴发存在2.8a和11.2a2种不同长度的循环周期，其暴发趋势指数对滞后1a和滞后4a的影响为正相关，而对滞后5a和滞后6a的影响则为负相关。本文根据甜菜夜蛾暴发指数的非均衡周期性特征，建立了以时间序列为自变量的甜菜夜蛾暴发指数非均衡周期性预测模型，经回代结果检验，理论值与实测值之间无显著差异。

Unbalanced cyclical fluctuation pattern of intermittent outbreaks of beet armyworm Spodoptera exigua (Hübner) in China

Abstract: Spodoptera (Laphygma) exigua Hbner is a worldwide insect pest, whose populations often explode intermittently causing disaster in China. Especially in the last 20 years, S. exigua outbreaks in China have been constantly expanding in their scope and quantity of occurrence, and the economic damage caused to agriculture has also been intensifying unceasingly. Although our predecessors, through broad and thorough research, have developed a relatively profound understanding of the pest development laws, as well as the reasons for the S. exigua outbreaks, the understanding of the temporal and spatial dimensions of the intermittence of outbreak of S. exigua is still very limited. There are few explorations on the overall rules governing the population dynamics during multi-years, in the long-term or the nationwide-range of S. exigua, which results in a weak ability for early warning of S. exigua outbreaks. In order to further assess any regular patterns of frequently intermittent outbreaks of S. exigua in China, this paper studied the time series of fluctuations of intermittent outbreaks by using the temporal sequence analysis technique. The results showed that the outbreak frequency of S. exigua in China was 120 in the 60 years from 1949 to 2008. The annual frequency had some obvious characteristics of unbalanced cyclical fluctuation, with an increasing trend in the fluctuation. According to the intensity of annual frequency, the occurrence process of S. exigua outbreaks during these 60 years could be divided into two phases. The first phase (1949-1984) was a steady and low incidence phase, and the total annual frequency and the annual average frequency of which were 4 and 0.11, respectively. The second phase (1985-2008) was wavelike with a high incidence phase, and the total annual frequency and the annual average frequency of which were 116 and 4.83 (43.91 times that of the first phase), respectively; the annual minimum frequency was 1; the annual maximum frequency was 19. Based on the data of the second phase data of 24 years, the autocorrelation coefficient and spectrogram analysis were estimated. The results showed that two different length cycles (2.8-years and 11.2-years) of large-scale outbreaks of S. exigua were present in China, and that the trend index of the outbreak was positively correlated with a lag of 1 or 4 years, and was negatively correlated with a lag of 5 or 6 years. The reasons causing the changes in the insect population are complex and diverse. From these we may conclude 3 kinds of main factors: namely artificial factors, natural factors and insect population factors. The artificial factors include the periodic modification of the widespread or locally applied agricultural cultivation technologies, both the plant culture system (for example, crop allocation, variety change, soil amelioration and so on) and the pest control technology. Natural factors include the periodic exceptional changes of wide-range or local area climate (for example drought, rainstorm and so on). The insect population factors include the pest population processes themselves in response to the new environment or new pesticides, and the process of periodicity reducing and recovering the levels of the natural enemy populations and so on. A non-equilibrium cyclical forecasting model of the outbreak index of S. exigua was established in this paper, taking a time series of years as the independent variable according to the non-equilibrium cyclical characteristics of the outbreak index of S. exigua. The results of the inspection by inverse calculation showed that there was no significant difference between the theoretical value and the actual value, which provided the theoretical means for exploring further the time evolution features of medium or long-term of S. exigua populations and establishing the medium or long-term population forecasting models, with the objective to enhance early warning capacity for S. exigua outbreaks, and to guide timely decisions for the integrated management of S. exigua.