Design And Implementation Of Forest Fire Forecasting And Warning Systems In Taiyuan

A wildfire is any uncontrolled fire which causes damage on forests, properties, orhuman lives.Wildfire is a worldwide severe nature disaster.Its extensive size, highoccurring frequency, severe damage to forest resource, threatening human life activities,causing global environmental pollution draw more and more attention on variouscountries. Wildfire occurs suddenly and randomly, however it have some Regularity, theoccurrence of wildfire requires certain weather condition, combustible vegetation,Topographical and geographical conditions.According to incomplete statistics, hundredsof thousands of wildfires occur world widely each year, the affected area includesseveral million hectares, about0.1%of total area of forecast. Since1980s caused byglobal warming, the trend of wildfire occurrence is going up. Especially in late90century, wildfire destroyed Millions of hectares of tropical forests. In our country,wildfire is a serious destructive natural disaster. According to statistics, from1950to2002a total of nearly700,000wildfires occurred, the average annual occurrence ofwildfires exceeds10,000times, about14percent of the world’s average annual numberof wildfires, total affected forest area passed3million hectares. In recent years, how toreduce the occurrence of wildfires, reduce the loss caused by wildfires is a very urgenttask we have to face. This article based on the study of project “Taiyuan wildfireforecasting and early warning system” in2002, taking remote sensing and geographicinformation systems and other modern high-tech IT methods as the main method,studied wildfire forecasting and early warning application method based on thecombination of Moderate-resolution Imaging Spectroradiometer(MODIS) andGeographic information system technology.In this article, through the collection of wildfire risk relevant forecast information,(Eg: the distribution of forests, vegetation map, fire history data, meteorologicalobservation data and other information) on GIS platform, digitalized into databases, useSpatial Interpolation Methods in statistics on meteorological observation data, obtain acertain space distance (eg.1km) meteorological observation of temperature, humidity,rainfall and wind speed map figure. Using the application of satellite remote sensing data in combustible vegetation classification, normalized various types of data and inputin to wildfire forecast model, thus deduced the next24hours wildfire risk forecastresults.On the basis of the review of domestic and foreign research, this articlerespectively researched Satellite real-time monitoring and the data processing of forestfire behavior, factor analysis on a variety of fire risks, and Moderate-resolution ImagingSpectroradiometer data inversing surface brightness temperature and extraction ofvegetation burning information methods. Through the research, acquired based on GISplatform, mainly resourced by MODIS, Application methods of satellite remote sensingdata in wildfire early warning, thus obtained Taiyuan area of wildfire danger ratingforecast system.