Investigation Of Vegetation Cover Change In Ma'anshan City Based On Multi-sensor Satellite Data

Fraction of vegetation cover(FVC)is an important indicator of vegetation change,and remote-sensing vegetation coverage data is widely used for ecological conservation and environment protection.The current study exploited the 30m-resolution Landsat-5 Thematic Mapper,Landsat-7 Enhanced Thematic Mapper Plus and Landsat-8 Operational Land Imager data to investigate spatio-temporal variations in vegetation conditions of the Ma'anShan city,Anhui province.For these purposes,remote sensing,geographical information system and statistical methods were used to process six Landsat images acquired in 1984,1990,1997,2005,2011 and 2017.In order to obtain the change of vegetation cover in the study area,this paper selected six remote sensing images from 1984 to 2017 were selected in this dissertation.Hyperion image has been used as the data source of spectral resampling to simulate multispectral data through spectral conversion technology,and then fitted the bands with quadratic polynomials one by one to obtain the conversion relationship among TM,ETM+,OLI.The efficiency Consistency of data is improved.Based on multi-sensor data intercalibration technology,Normalized Difference Vegetation Index(NDVI)was chosen for estimating vegetation coverage,and calculating vegetation coverage under various land cover conditions based on the dimidiate pixel model.The results of two methods were strongly correlated with the actual reference data,and the first is better(R =0.9731),indicting the effectiveness of the dimidiate pixel model;.The estimated results show that the vegetation coverage of Ma'anShan city is in generally good condition.The area with the poorest(FVC<0.1)vegetation coverage is ~5%.The moderate-vegetated and even better(FVC> 0.3)area is over 60%.According to the calculation results of difference method and transfer matrix method,the vegetation coverage has increased in these 34 years.Time series vegetation coverage data of Ma'anShan municipal districts shows the average vegetation coverage has decreased in Dangtu County,and the average increase of Hanshan County is the largest from 1984 to 2017.Most of the area has a stabilized vegetation coverage,followed by the area with vegetation development and the area with vegetation degradation.The current study exploited Landsat satellite data and derived vegetation coverage data based on the dimidiate pixel model.The method can benefit long-term(1984–2017)estimation of vegetation coverage data in the Ma'anShan city,and therefore fill in the datagaps for tracking decadal changes in vegetation coverage.The results also serve as a reference to verify the achievements of ecological conservation and environment protection.