蝗虫生境监测方法研究进展

蝗虫是一种能对农业生产带来毁灭性打击的害虫。预测、监测蝗虫的发生、发展对于防治蝗虫、减轻蝗灾具有重要意义。在分析蝗虫的发生和消长与其生存环境的关系基础上，概述了以往蝗虫发生、发展预测、监测的主要方法。然后，从监测蝗虫生境采用的指标、卫星数据、算法等方面进一步阐述了运用现代遥感、地理信息系统技术监测蝗虫生境、预测蝗虫发生、发展的理论基础和最新进展；最后，结合现代对地观测技术、网络、快速计算和模拟等技术的发展探讨了蝗虫生境监测的发展方向。

Progress on methodology in monitoring locust habitats

Locust over abundance could extremely harm the agricultural yields. It is very important for lessening and controlling the locust plague to predict locust's expanding. The analysis of the relationship between the locust and its environment evidences that the major factors influencing locust's habitats are those including landforms, temperature, humidity, vegetation, soil and the against species. The records that drought immediately following the water flood was very suitable for locust expanding in Eastern China were found in the book of Controlling locust published in 1874 (Chinese Tongzhi Year 13). Hereafter the methodology predicting locust by climate, phenology and ground true investigation were widely applied in predicting the locust situation in China. With the development of remote sensing and geographical information systems in 1970's, the new technology has been widely practiced in monitoring the locust habitats. The great achievements in this studies include monitoring the rainfall with METEOSAT, NOAA AVHRR and GMS IR WEFAX, the vegetation types and vegetation greenness with NOAA APT, NOAA AVHRR, LANDSAT MSS, TM and SPOT VGT; the soil types and soil humidity with LANDSAT MSS and SPOT VGT S1. More recently, MODIS integrated with ETM is synthetically applied in monitoring indicators of the habitats. Supported by the information technology, GIS for monitoring and predicting locusts were developed and implemented. The comparison study among the traditional methods, as well as the new ways supported by the update technology was taken in this paper, which analyzed the suitability and disadvantages for each of the methods. It is obvious that the integration of all methods will drive the research into a bright direction in synthetically monitoring the locust habitats. High resolution and more frequency remote sensing data and information will play a critical role in the monitoring in the coming years, the integration among the information from satellites, in situ survey and historic statistic data could be one of the key issues. The rapid development of computational methodology, internet/ intranet and virtual reality technology will great help the locust habitat monitoring from the individual activities into the systematic and intelligent consortium.