| Fractional vegetation cover can indicate the quality of the ecological environment and is one of the evaluation indicators of ecosystem function and stability.Efficient monitoring of vegetation cover change is the primary prerequisite for assessing the health of regional vegetation ecological environment,which is very important for ecological construction and restoration and improving the quality of people’s living environment.The ecological construction and stable development of the mining area are closely related to the prosperity of the social economy.It is urgent to carry out ecological protection and management of the mining area,repair the ecological function of the mining area,and enhance the value of ecosystem services.How to quickly and accurately monitor the dynamic changes of vegetation in mining areas and realize the common development of economy and ecological construction is a focus problem worthy of continuous attention and urgent solution.In this study,the Google Earth Engine(GEE)platform was used to obtain long-term series of Landsat remote sensing images during the period of vigorous vegetation growth(July-September).The normalized difference vegetation index-dry fuel index(NDVI-DFI)pixel three-point model was used to estimate the annual photosynthetic vegetation coverage(fPV),non-photosynthetic vegetation coverage(fNPV)and bare soil coverage(fBS)in Shenmu City,Shaanxi Province from 2000 to 2020.Based on the fine resolution mapping of mountain environment(FRMM)vegetation coverage data,the reliability of the pixel trisection model to estimate vegetation coverage was verified.The dynamic change characteristics and spatial distribution pattern of vegetation coverage in Shenmu City were discussed by using linear regression analysis,coefficient of variation,spatial autocorrelation calculation and correlation analysis.The effects of precipitation,temperature,population density and land use type transfer on vegetation change were explored.The main conclusions are as follows:(1)The photosynthetic vegetation coverage estimated by the NDVI-DFI pixel trichotomy model is highly consistent with the FRMM vegetation coverage data.The overall distribution trend of photosynthetic vegetation coverage in 2010 and 2018 was highly consistent with the FRMM vegetation coverage data in the corresponding years,and the correlation coefficients in the corresponding years were 0.63(P<0.01)and 0.66(P<0.01),respectively.This indicates that the NDVI-DFI pixel trichotomy model is feasible for estimating vegetation coverage and the results are reliable.(2)From 2000 to 2020,the vegetation coverage in the study area fluctuated greatly,and the vegetation coverage showed an increasing trend in time.The average variation coefficient of photosynthetic vegetation coverage in the past 21 years was 0.8.The vegetation change in the southwestern desert area and the urban agglomeration area in the central and northern parts is the most violent.The vegetation coverage change in the hilly and gully areas suitable for herbaceous plants in the east and south fluctuates greatly.The average photosynthetic vegetation coverage and average non-photosynthetic vegetation increased at an average annual rate of 3.86%and 0.36%,respectively,and the vegetation coverage improved significantly.The area with significant increase of photosynthetic vegetation coverage accounted for about 59.6%,the area with slight increase accounted for 29.4%,the area with significant degradation accounted for about 1.9%,the area with slight degradation accounted for about 8.3%,and the area without significant change accounted for about 0.8%.The photosynthetic vegetation coverage increased rapidly in the eastern,central and southern regions,and increased slowly in the northwestern region.Vegetation degradation occurred in some urban agglomerations.(3)The vegetation distribution in the study area is concentrated and has significant spatial positive autocorrelation.The global Moran’s I index of the average photosynthetic vegetation coverage in Shenmu City in the past 21 years was 0.54(P<0.01),indicating that the spatial agglomeration of vegetation coverage was strong.The local spatial autocorrelation results of the average photosynthetic vegetation coverage in Shenmu City for many years are mainly high-high aggregation,low-low aggregation and insignificant aggregation.Low-low aggregation is mainly distributed in the west and north,and high-high aggregation is mainly distributed in the east,south and around Hongjiannao Lake.(4)The increase of precipitation and temperature can promote the improvement of vegetation to a certain extent.The increase of population density is not conducive to vegetation restoration.The land use transfer activities,which mainly transform cultivated land and unused land into grassland,improve vegetation coverage.On the pixel scale,the proportion of pixels with positive correlation between photosynthetic vegetation coverage and annual cumulative precipitation and annual average temperature in Shenmu City exceeded 80%,and the number of pixels with significant positive correlation exceeded 10%.The number of pixels with a negative correlation with population density accounted for 68.9%,of which the number of pixels with a significant negative correlation accounted for 17.9%.On the regional scale,the average photosynthetic vegetation coverage in Shenmu City is consistent with the interannual variation trend of annual cumulative precipitation and annual average temperature,which is opposite to the interannual variation trend of population density.In terms of land use type change,the proportion of grassland area in Shenmu City increased to 84.4%,and the transfer of land use types was mainly from cultivated land to grassland and unused land to grassland. |
