Study On Fire Monitoring Method Combined With Active And Passive Remote Sensing In Qinling Area

Forests play an important role in the cycle of the entire ecosystem,and the protection of forest resources is a major event related to human well-being.Fires can cause great damage to forest resources,leading to the death of vegetation,and the emission of harmful substances such as smoke will have a bad impact on the environment.At present,satellite remote sensing monitoring of forest fires is one of the most widely used and effective means.It is of great significance to make full use of satellite remote sensing technology to monitor forest fires.This paper carries out research on fire monitoring methods in Qinling area combined with active and passive remote sensing:passive remote sensing MODIS satellite can identify fire points,and active remote sensing CALIPSO satellite can identify smoke aerosols.Firstly,according to active and passive remote sensing data,the temporal and spatial distribution characteristics of fire points and smoke aerosols in Qinling area from 2012 to 2021 are reconstructed,and the influencing factors related to fire points are analyzed.Then,the data matching algorithm of fire point and smoke aerosol is used to study the optical properties of smoke aerosol under fire point.Based on the maximum inter-class variance fire point recognition algorithm of MODIS satellite,combined with the active remote sensing CALIPSO satellite,the fire point recognition model of active and passive remote sensing is constructed,and the problem of missed detection of ground fire points caused by cloud occlusion of passive remote sensing MODIS satellite is tried to be improved.The feasibility of the model is verified by real fire cases,which provides new ideas and methods for remote sensing satellites and ground lidar to monitor fire points.Based on the MODIS satellite fire data,the temporal and spatial distribution of fire points in the Qinling Mountains was statistically analyzed.It was found that the number of fire points in the Qinling Mountains was the highest in 2015,accounting for 15.11%of the total number of fire points.In terms of seasonal distribution characteristics,the fire spots in the Qinling Mountains mainly occur in spring and autumn,accounting for 65.41%of the total fire spots.In terms of spatial distribution,Shiyan City has the highest number of fire spots.The fire points are mainly concentrated at the altitude of 992m-1367m,accounting for 49.12%of the total fire points.In the distribution of vegetation cover types,the fire points are mainly distributed in the coniferous and broad-leaved mixed forest and coniferous forest vegetation types,accounting for 25.43%and 20.20%of the total number of fire points,respectively.Smoke aerosols and fire spots have similar characteristics in spatial distribution.Through the correlation analysis of the number of fire points and its influencing factors,the precipitation and NDVI values are negatively correlated with the number of fire points,and the temperature is positively correlated with the number of fire point.Through the matching algorithm of fire point and smoke aerosol data,the smoke aerosol under the fire point in Qinling area from 2012 to 2021 is obtained.The statistical analysis shows that the non-spherical degree of smoke aerosol particles under the fire point decreases from near the ground to high altitude.The color ratio of smoke aerosol particles is between 0.6 and 0.8,and the particles are larger.By comparing the depolarization ratio of smoke aerosol layer integral between fire point and non-fire point,it is found that the depolarization ratio of smoke aerosol layer integral under fire point is significantly larger than that under non-fire point,and the threshold of layer integral depolarization ratio to distinguish them is 0.1.On the basis of the existing MODIS satellite fire point identification algorithm,combined with the method of active remote sensing CALIPSO satellite to identify smoke,a fire point identification model under the cloud is constructed.At the same time,a case is used to verify that the model can improve the MODIS satellite.The problem of missing fire points caused by clouds.Through the fire case in Ankang City on October 12,2013 and the mountain fire case in Qingchuan County,Guangyuan City on February 6,2016,the active and passive remote sensing fire point identification model was verified and compared with the MODIS official fire point identification algorithm.It was found that the model can identify the fire points missed by MODIS satellites due to cloud cover.