LUR-based Simulation Of PM_2.5Concentrations In Nanchang Central City And Of Land Use On It

In recent years,the urban air pollution problem in China is becoming more and more serious.PM_2.5,which can do serious harm to the human body,has become the primary pollutant in many cities.It is urgent to control PM_2.5 pollution.Studies have already demonstrated that both meteorological factors and land use have important impact on PM_2.5 pollution.Since the relationship between land use and PM_2.5 pollution is still unclear,it is of great practical significance in preventing atmospheric pollution and protecting human health to explore their relationship.Based on data of monitoring stations for ambient air quality,this paper took Nanchang city as an example,analyzed the spatial and temporal distribution characteristics of PM_2.5concentrations,and the influence of meteorological factors,precursors and so forth on PM_2.5concentrations.The land use regression（LUR）models were applied to simulate the spatial variations of PM_2.5concentrations in Nanchang central city.According to the dominant land use type,the study area was divided into six types of land-use function districts including commercial function districts,industrial function districts,residential function districts,educational function districts,special function districts and ecological service districts.15 points in each type of land-use function districts were randomly-selected as samples to investigate the impact of land use on PM_2.5 concentration.The impact was explored from three aspects including land-use function district type,land use level and landscape pattern.The PM_2.5concentrations were used to characterize the"impact degree"of PM_2.5 pollutants on the population,and population density was used to characterize the"sensitivity"of the population to PM_2.5 pollutants,constructing a coupling model of"exposure risk=impact degree+sensitivity"to express the PM_2.5 exposure risk inside city.The study area was spatial zoned according to exposure risk level.This study explored a new way to couple urban land use with air pollution.The results can provide valuable references for urban land-use optimization and atmospheric pollution control in future.The main research conclusions were summarized as follows:（1）There were obvious spatial-temporal characteristics of PM_2.5 concentrations in Nanchang central city.On the variation of time scale,the PM_2.5 pollution condition in Nanchang city had been significantly controlled from 2013 to 2016.High PM_2.5concentrations in winter and low in summer were observed.The highest PM_2.5concentration occurred in January and lowest in August,and a small peak presented in October.On the variation of spatial scale,low PM_2.5 concentrations in the northwest and high PM_2.5 concentrations in the southeast,decreasing from the city center to the suburb.（2）The impact of meteorological factors on PM_2.5 concentrations were complex.When analyzing the correlation between meteorological factors and PM_2.5,any differences of study area,time scale or data scale may lead to different results.According to analysis results of monthly data,PM_2.5 concentrations showed significant negative correlations with precipitation,temperature and relative humidity,significant positive correlation with air pressure,and negative correlation with wind speed in the whole year.According to analysis results of daily data,PM_2.5 concentrations showed significant negative correlations with precipitation,wind speed and relative humidity in all four seasons throughout the year.The correlation properties between PM_2.5 concentrations and air pressure or temperature varied as seasons varied.（3）SO₂ and NO₂ were two gaseous pollutants that had significant influence on PM_2.5concentrations,and both of them had significant positive correlation with PM_2.5concentrations.The correlation coefficients of SO₂ or NO₂ and PM_2.5 concentrations were greatly influenced by the season,the lowest occurred in summer and the highest in autumn or winter.The CO concentrations were correlated with PM_2.5 concentrations to some extent,while O₃ concentrations were not correlated with PM_2.5 concentrations.（4）It was feasible to simulate PM_2.5 concentrations through LUR models by introducing factors including meteorological,land use,population and so on.The results showed the major meteorological factor influencing PM_2.5concentrations in four seasons were different.Traffic and industrial emissions were the main sources of PM_2.5 pollutants in Nanchang central city.（5）Land use was an important factor influencing PM_2.5 concentrations.There were certain relationships between land-use function district type,land use level or landscape pattern and PM_2.5 concentrations.The significant PM_2.5 concentration differences among different land-use function districts indicated that urban land use had an obvious impact on PM_2.5 concentrations.The six types of land-use function districts could be divided into three groups according to the PM_2.5 concentrations,which showed an order of commercial function districts,industrial function districts>residential function districts,educational function districts,special function districts>ecological service districts.The land use intensity was related to PM_2.5 concentration to some extent.The land use indexes that positively correlated with land use intensity showed positive correlation with PM_2.5concentrations,and the land use indexes that negatively correlated with land use intensity showed negative correlation with PM_2.5 concentrations.The landscape pattern reflected the spatial distribution of PM_2.5 concentrations to a certain extent.At class-level,the higher the construction land percentage and the lower the ecological land percentage,the higher the PM_2.5 concentrations.At landscape-level,the higher the patch density index,boundary density index and convergence index,which means the higher landscape fragmentation degree,and the higher the PM_2.5 concentrations.（6）The combination of GIS spatial analysis technology,population spatialization method and air pollutant simulation model could be used to achieve spatial zonation of population exposure risk in urban areas.Compared with the pollutant concentration method,the coupling model of"exposure risk=impact degree+sensitivity"constructed in this paper had overcome the defect of ignorance of population factor.And response of this model to population factor would not be as sensitive as the residence exposure intensity method.Therefore,we concluded the model could be widely used to spatial zonation of air pollutant exposure risk and identification of key exposure risk area in cities.Results showed that there were three PM_2.5 high-exposure-risk areas in Nanchang central city which should be urgently controlled.They were the Jingchuan and Hujia Village which located in the north of Changnan Avenue,the Donghu District which located in west of Hongdu North Avenue and north of Yangming East Road,and Nanchang Railway Station District.