Correlation Analysis Between Fine Particulate Pollutants And Land Cover Pattern In Beijing

Air pollution is a hot environmental issue today.Understanding the interaction between different land cover landscape patterns and air fine particulate pollutants is of great significance to improve urban ecological environment.Aerosol optical depth(AOD)is the premise for the generation of air fine particle pollutants.By introducing the AOD data and ground fine particle pollutant concentration data measured by AERONET Beijing ground detection point,through fitting analysis.After processing Landsat images to obtain the landscape index of Beijing in 2006,2012 and 2018,the following conclusions are obtained after correlation analysis with the AOD data of the corresponding three years:(1)The MODIS AOD data obtained from the interpretation of remote sensing data can estimate the mass concentration of fine particles on the underlying surface in a macro range.Through the analysis of AOD center of gravity transfer during the study period,it is found that with the passage of time,the AOD center of gravity of central urban area,Daxing District,Tongzhou District and Shunyi District in Beijing shows an outward expansion trend,on the contrary,the AOD center of gravity of other administrative divisions shows a contraction trend towards the central urban area.(2)In terms of the correlation between fine particulate matter represented by MODIS AOD and landscape level index,there is a significant negative correlation between COHESION(connectivity index),LPI(maximum patch index),CONTAG(spread index)and fine particulate matter during the study period,while PD and SHDI indexes have a significant positive correlation with fine particulate matter,combined with the spatial correlation bivariate LISA aggregation diagram,the spatial variation characteristics of this correlation are further illustrated.(3)In terms of the correlation between fine particles represented by MODIS AOD and class level landscape index,except for LSI(landscape shape index),other AI(aggregation index),COHESION(connectivity index),IJI(dispersion and juxtaposition index)and ED(boundary density index)of farmland cover type landscape index throughout the year are significantly negatively correlated with fine particles throughout the year;The above five landscape indexes of forest and water cover types have a significant negative correlation with fine particulate matter throughout the year;Except IJI index,the other five landscape indexes of grassland cover types are significantly negatively correlated with fine particles throughout the year,negative correlation shows that it can play a positive role in reducing fine particulate pollutants;The five landscape indexes of construction land are significantly positively correlated with fine particulate matter throughout the year.Through the overall situation spatial correlation analysis,the Moran’s I scatter diagram of AI,COHESION and LSI indexes of each cover type in 2018 is obtained.The Moran index of farmland,forest,grassland and water cover types is negative value,while the Moran index of construction land cover types is positive value,the results are the same as the Pearson correlation results.(4)Through principal component analysis and multiple regression analysis on the landscape index of fine particles and land cover type level represented by MODIS AOD,it is concluded that most of the first and second principal component coefficients of the landscape index of farmland,forest,grassland and water body in the study period are negative,It shows that it has a positive impact on controlling the rise of the mass concentration of fine particles;The values of the first and second principal components of the landscape index of construction land are basically positive,It shows that it can not inhibit the rise of the mass concentration of fine particles.