Font Size: a A A

Remote Sensing Monitoring Of Forest Fires And Vegetation Restoration

Posted on:2019-06-21Degree:MasterType:Thesis
Country:ChinaCandidate:H SunFull Text:PDF
GTID:2370330548976647Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
The important content of improving the quality of forest is to study the interference intensity of forest fire,to describe the intensity of the forest fire and the restoration of vegetation,and to improve the monitoring ability of the dynamic change of the forest.For forest stand scale,using multi-modal remote sensing,coordinate the horizontal and vertical structure information of the forest,quantitatively analysis.the effects of fire on forest area,Fractional Vegetation Cover(FVC),above-ground biomass(AGB)and canopy height,improve the efficiency of regional scale forest structure parameter inversion.Exploring the method of forest fire accuracy estimation based on multi-modal remote sensing can improve the efficiency of forest structure parameter inversion at regional scale,provide high-age support for the accurate monitoring of the intensity of forest resources change before and after an interference event,Improve the current status of the forestry sector in assessing fire disturbances and provide scientific evidence for the development of the forestry industry.This study is based on the principle of binary model,support vector machine,zero-mean normalization,linear regression,AGB Conversion,and stereoscopic photogrammetry,with ZY-3,GF-1 and Landsat-TM data as the main data sources,combined with Li DAR point cloud data and CCD images,MODIS17_NPP products,GLASS_LAI products,ASTER GDEM30 m data,etc.to realize the remote sensing dynamic monitoring of forest fire disturbance and restoration of vegetation Carry out relevant work and obtain the following results:(1)The FVC of the city is higher,but the "Jinhe 5·5" fire is seriously damaged.After 15 years of recovery and growth,the FVC of the fired area was improved,but mostly shrubs and clusters of white birch were the main ones,and the group was not restored.It is not a single factor to influence the FVC changes,and the fire interference is the most prominent factor.The effects of topographic and climatic factors are irresistible.In low-altitude and flat slope areas,the FVC in areas prone to human activities,such as roads(301 national road),river(Genhe and Aoluguya River),were not significantly improved or even decreased.Strengthening the implementation of “Natural Forest Protection Project “Scientific afforestation,managing forests,and protecting forest to improve stand structure and reducing the interference of human activities is an effective means of sustainable management of city forest resources.(2)A comprehensive understanding of the extent of forest destruction "Jinhe 5·5" Burned and restore growth conditions.Among them,the area of fire in 2002,2003,2006 and 2016 was 164.34 ha,84295.89 ha,32453.28 ha and 13779.87 ha.According to the "Z-value method" of FVC,the area ratio of light,medium and heavy fire is about 1:3:6.As of 2016,the area of light,medium and high degree of recovery is about 4:5:1.The average of AGB in 2002,2003 and 2006 was 75.11,44.66 and 58.16t/ha.In 2003,the total amount of AGB in the "Jinhe 5·5" fire loss was 261.99 million tons,and the loss of AGB in the fire starting point was the most serious.(3)Considering the natural update status after fire area is more than 10 years,and in the area of natural recovery(natural secondary forest and shrub mixed)inversion,the vegetation height of Li DAR point cloud data acquisition of shrub layer the average height of 1.09 m,The height error of shrub layer obtained by zy-3 is large.Reasons: 1)lack of high quality summer ZY-3 stereo image data;2)zy-3 has low effective point cloud density;3)it is difficult to find the position connection point of the same name due to the high mountain topography shadow interference,which needs to be further studied and solved.Numerous studies in the past have focused more on fire point monitoring and recovery monitoring.This study not only monitors forest parameters before and after the fire,but also use multi-source remote sensing data,to obtain the surface coverage type,FVC,AGB,shrub height and other monitoring indexes,to reflect the degree of burning of fire zone,and the restoration of vegetation situation,effectively solve the voucher only stand factor to describe the problem of the burns.The method for the quantitative evaluation and dynamic monitoring of the burns provides reference basis and technical support.
Keywords/Search Tags:burned area, dynamic monitoring, land cover types, Fractional Vegetation Cover, above-ground biomass, stereo photogrammetry
PDF Full Text Request
Related items