A Study Of Grassland Fire Monitoring And Early Warning And Assessment Inner Mongolia

This paper takes the Inner Mongolia Autonomous Region as the researcharea, in accordance with the weak link of our country grassland fireemergency management and assessment, remote sensing estimation of thewithered season fuel weight, fire risk warning, sub-pixel fire areaestimation,risk assessment, loss evaluation and grassland fire ecologyenvironmental impact assessment and etc has been investigated based on3Stechnology via a multitude of field investigations and experiments, resultsmainly including the following several aspects:(1) All types of grassland fuel and first and second channel spectralreflectivity of MODIS data have significant negative correlation. The firstchannel reflectivity of grassland fuel in meadow steppe, typical steppe, desertsteppe and grassland desertification area has higher correlation than secondchannel’s, and in the desert the second channel reflectivity has highercorrelation than the first one. Using the channel which have high correlations has established estimation model, the accuracy test proves that the correlationcoefficient between measured data and the inversion result of typical steppe,steppe-desert and desert are respectively0.78、0.84、0.71, all highly correlated;Meadow steppe and desert steppe are0.71and0.78significant correlated;Various types of grassland fuel estimation model have reached themacroscopic monitoring standard.(2) Selecting7indicators such as withered period grassland fuel and etchas established grassland fire warning model of Inner Mongolia, accuracycould be up to96.42%through the precision test. The result shows that indexselection and hierarchy classification is reasonable for the prediction methodof grassland fire danger rating, therefore, it can be used for practicalapplication of grassland fire danger forecast.(3) The Landsat TM high resolution remote sensing data as an externaldata source for the EOS/MODIS and put forward the basic flow and the keytechnology of the grassland fire sub-pixel fire area estimation, the method canaccurately estimate the sub-pixel fire area. Through application of the methodto estimate the area of fire point, find out that the area of on fire is79.3 km~2.But if ignore the existence of mixed pixels and directly use pixel units tocalculate the area of the fire point,the result is192km~2. So, the method canimprove the accuracy of extraction of fire point’s area.(4)Based on the Graham Kinney method (LEC) and analytical hierarchyprocess method (AHP model) have constructed grassland fire risk evaluationindex and model. Application of the model to calculate the Inner MongoliaAutonomous Region each leagues’ and cities’ fire risk in the next few years,find that Hulunbeir City, Xilingol League, Hinggan League, Tongliao City,Chifeng City have a high risk.(5) This study has established a more regulated grassland fire disasterloss evaluation index system and grassland fire loss assessment model, takingXilingol League April7,2012grassland fire as an example, the result showthat the fire area is778km~2, has caused death2people and minor injury8people, burned forage grass37.38million kg, financial loss calculation ofbured forage grass according to the market price of0.3yuan/kg is11.21million yuan. According to standard of the grassland fire loss level division,the fire belongs to particularly significant(I level) one. (6) In this study taking stipa grassland as the research area, using thedifferent period of winter and spring prescribed fire tests were conducted andthen do vegetation community of field investigation and soilphysico-chemical properties experimental analysis. Selecting10indicatorssuch as biomass established grassland fire ecology environmental impactassessment model, and the fired ecological environment are divided into fourgrades-the obvious better, better, worse and the worst. From the evaluationresults can be informed that compared with not burned fire area the Stipagrassland ecological environment has deteriorated for after burned area.This research provides timely and accurate information services andtechnical support for management departments to do well disaster earlywarning, real time monitoring, and quick reaction after disaster and makescientific countermeasures for disaster prevention and mitigation. Andrealizes changing from the current passive disaster management mode to threephases─the pre-disaster early warning and disaster period emergency andafter disaster relief phases integrated grassland fire disaster comprehensivemanagement and control mode, comprehensively improves the level of emergency management science and technology of China’s grassland firedisaster.