| With the rapid development of urbanization and industrialization,the environmental quality in most areas changes rapidly.Emissions from the industrial sector,diesel,or gasoline vehicles,coal-fired and biomass burning in power plants,and residential stoves,among other sources,contribute to serious air pollution in China.With the construction of the Beautiful China in recent years,measures have been taken to combat the serious air pollution problem and effectively reduce the pollution level.China’s ecological environment has improved and air pollution has been controlled,but there are significant differences in PM2.5concentrations across different regions.PM2.5 pollution is affected by many factors,and the deterioration of natural ecological environment caused by human activities is an important reason for the worsening of PM2.5pollution.However,studies on the relationship between PM2.5 and the evolution of land use pattern are few,and the relationship between environmental greenness and PM2.5 concentration at the regional and urban levels,as well as the influence of the evolution of different land cover types on it are still unclear.Shanxi Province as main energy base,coal industry has developed rapidly.Over the past few decades,industrial development,population growth and urban expansion in Shanxi Province have led to significant changes in land use patterns and increasing air pollution problems,which directly threaten the physical and mental health of local people and seriously impede regional sustainability.Based on the above questions,this study selected Shanxi Province as the research area,combined with remote sensing monitoring,spatial analysis method and spatial correlation analysis method.Global Land Cover Dataset(Globe Land30),Normalized Vegetation Index(NDVI)from Moderate-Resolution Imaging Spectroradiometer(MODIS)product data,and high spatio-temporal resolution fine particulate matter concentration data(China High PM2.5)were used to analyze the land use pattern and its evolution characteristics in Shanxi Province in the past 20 years,as well as the regional environmental greenness and PM2.5 pollution characteristics and their spatiotemporal changes.The effects of land use pattern evolution and vegetation greenness change on PM2.5pollution in Shanxi Province in recent 20 years were discussed from regional scale,pixel scale,different cities and different vegetation types.The main conclusions are as follows:(1)The distribution of different land use types in Shanxi has obvious spatio-temporal differences,among which farmland is the most widely distributed,accounting for more than 40% of the total area of Shanxi Province,mainly distributed in the central,northeast,southeast and southwest basins of Shanxi Province.The area of woodland and grassland accounted for 28.28% and 23.78% of the total area,respectively.The construction land was mainly distributed in the central and southeastern basins,accounting for 5.44% of the land use types.The proportion of wetland,water body and wasteland in the study area is small.The area change of different land cover types and its driving factors were different.The area of farmland and grassland decreased by 7.17% and 4.60%,respectively.The expansion of forest land and construction land was the main direct driving factor of grassland reduction.The main direct driving factor of farmland reduction is construction land expansion,followed by conversion of farmland to woodland and grassland.The area of construction land increased the most in Shanxi Province,and the increase rate was the highest.The second is woodland.The increase rate of wetland,water body and bare land is relatively high,but the original distribution area is relatively low,so the increase area is very small.The expansion of construction land is mainly the occupation of farmland and grassland,and the main direct driving factor for the increase of forest land is the degradation of grassland and cultivated land.(2)Environmental greenness in Shanxi Province has been improved in the past 20 years.At the regional scale,the average annual NDVI in the study area showed a significant increasing trend(P<0.01),and the proportion of the regions with a significant increase was about 93.67%,while only 0.97% of the regions had a significant decreasing trend of NDVI,mainly concentrated in the central construction land area.The average annual NDVI of different cities in Shanxi Province increased significantly,but the change trend was different.Jincheng City had the highest average NDVI,while Shuozhou City had the lowest average NDVI.The region with the slowest growth rate of annual average NDVI is Jincheng City,and Lvliang City has the fastest growth rate.The analysis of the annual average NDVI changes of the four main land cover types showed that the woodland was the highest in the annual average NDVI of the four main land cover types.The average NDVI of construction land was the lowest.The average annual NDVI of forest land,grassland,farmland and construction land increased significantly,with the highest rate of increase in grassland and the lowest rate in construction land.(3)The atmospheric environment of Shanxi Province showed a good trend in the past 20 years.The average annual PM2.5 concentration showed a significant downward trend,especially in central and southern regions,which was significantly greater than that in other regions.From 2001 to 2020,the average annual PM2.5 concentration of different cities in Shanxi Province showed an obvious downward trend,with Datong City being the slowest and Yuncheng city being the fastest.The annual average PM2.5 concentration of the four main land use types was analyzed,and it was found that the annual average PM2.5 concentration of the forest land was the lowest among the four main land use types,and the annual average PM2.5 concentration of the construction land was the highest.The average annual PM2.5 concentration of the four main land use types all showed a decreasing trend,but the decreasing degree was different.From 2001 to 2020,the average PM2.5 concentration of construction land in Shanxi Province decreased the most significantly,followed by farmland and grassland,and the forest land decreased the least.(4)The regional environmental greening rate has a significant impact on the decrease of PM2.5concentration in the region.The annual PM2.5 concentration in Shanxi Province was negatively correlated with the annual mean of NDVI,accounting for 83.77% of the total area of the study area,and mainly distributed in the vegetated area.The regional distribution area of insignificant correlation and significant positive correlation is small,and mainly occurs in the area of construction land expansion.There was a significant negative correlation between the average annual PM2.5 concentration and the average annual NDVI in different cities,among which Datong had the strongest negative correlation and Jinzhong had the weakest negative correlation.There was a significant negative correlation between the annual PM2.5concentration and the annual NDVI of the four land use types.Among them,the land use type with the strongest negative correlation between PM2.5 and the annual mean of NDVI is construction land,followed by farmland,and grassland has the weakest correlation with the annual mean of NDVI.In addition,this study also found that the correlation coefficient between the annual PM2.5 concentration and the annual NDVI value increased with the decrease of NDVI.In the area with low vegetation coverage,the increase of environmental greening rate had a stronger effect on the decrease of PM2.5 concentration.(5)The PH tool of TDA is introduced into the PM2.5 pollution prediction analysis,and an automatic correlation analysis model based on PH for spatial and temporal patterns of land use and changes is constructed.From the persistence diagram analysis at multiple scales,it was found that the PDs of 2001 and2020 were significantly different,and ω at lower filtering scales the overall 2D topology in 2020 was more clustered in the higher threshold part than the distribution in 2001,the 2D topology distribution in 2020 was mainly clustered in the high threshold position,while the 2D topology distribution in 2001 was mainly clustered in the middle threshold position,and the overall The overall topology of 2020 has a more persistent homology.The classifier with high classification accuracy for each feature was selected from multiple classifiers of different features as the final classifier of the model,and then four classification metrics were used to evaluate the environmental greenness metrics,vectorize the persistent topological features and fused features,and it was found that the fused features had better performance.Further analysis of the distribution of topological features revealed that HK and PL(Gaussian kernel standard deviation 3.2,1 parametric)in the LGBM boosted tree model were significantly effective in identifying the spatial distribution and changes of land use in Shanxi Province,and at higher scales,the distribution of HK values in 2001 was more dispersed and inconsistent than that in 2020.At the lowest scale,the PL distribution in 2001 and 2020 were significantly different,indicating that the spatial and temporal distribution of NDVI values were more variable throughout the persistent homogeneous filtering,while 2020 was more stable.The above conclusions can provide scientific basis for regional environmental governance,planning decision-making,air quality monitoring and early warning,governance effect evaluation and real-time monitoring of ecological environment quality in Shanxi Province. |
PDF Full Text Request