Based On The Ground Monitoring And Remote Sensing Inversion Of Temporal-spatial Distribution Of PM_2.5 At Beijing-Tianjin-Hebei Region

In recent years,with the sustained and rapid development of industrial economy and the process of city, regional air pollution is becoming increasingly serious in china. The primary pollutant affecting air quality of the environment in China is atmospheric particles,especially PM_2.5 on environmental quality and human health risk. Air pollution monitoring in the ground is the basis research of atmospheric pollution, GIS spatial analysis and satellite remote sensing technology are main methods,which distinguish the temporal and spatial distribution of atmospheric particulate on a regional scale.This article choose Beijing-Tianjin-Hebei region as the study area, where is more serious of the PM_2.5 pollution. On the one hand,through using the ground monitoring data for long time series PM_2.5,firstly analyzes the daily, monthly and seasonal variation of PM_2.5 pollution.By using the spatial interpolation method,analyzes the characteristics and trends of PM_2.5 geographical distribution.At the same time,combining with the meteorological elements of PM_2.5 concentration and relative humidity, wind speed, temperature and other meteorological factors correlation analysis,the correlation analysis have been done.On the other hand, with the same period of MODIS high resolution version sixth（C6）MOD/MYD04₃K aerosol product data,analyzed the temporal and spatial distribution of aerosol optical thickness.Then the vertical elevation and humidity correction of aerosol optical thickness data were carried out by using GEOS meteorological data, the multiple linear regression model using regression analysis method of aerosol optical thickness is established to estimate PM_2.5 concentration, we measure the concentration of PM_2.5 and use the reservation data to validate the model,analysis and inverse of the spatial distribution characteristics of PM_2.5 concentration.Finally, the results of remote sensing retrieval,spatial interpolation and ground monitoring data were compared and analyzed, and the accuracy evaluation was carried out by using statistical method.the conclusions are as follows:（1）According to the 《 environmental air quality （AQI） technical regulations（Trial）in 2014,PM_2.5 daily average concentration value up to substandard accounted for 46.1% days in the Beijing-Tianjin-Hebei region,The more severe pollution and the number of days the proportion of 13.7%,The annual average concentration of PM_2.5 is 93.21 μg·m-3. The diurnal changes of PM_2.5 has a bimodal distribution, the concentration fluctuation between 71～102 μg·m-3,PM_2.5 month concentration showed "U" pattern distribution,The average monthly concentration peak appeared in January,144.72 μg·m-3,the lowest values were attained in June,61.19 μg·m-3;The concentration of PM_2.5 showed a clear seasonal distribution, The concentration of PM_2.5 was highest in winter, followed by autumn,summer and spring,The main causes for the seasonal variation of seasonal emission pollutants and climatic conditions.In 2014,alonging the Beijing, Baoding,Shijiazhuang,Xingtai and Handan line PM_2.5 daily and annual average concentration exceed the standard rate is higher; in the Beijing-Tianjin-Hebei region,the spatial distribution of season and month showed regional spatial distribution pattern is the South higher than the North, From May to September in the summer with PM_2.5 concentration gradient change, variation is not obvious.（2）The PM_2.5 concentration in Beijing-Tianjin-Hebei region and relative humidity,wind speed, sunshine hours has greater relevance, due to the different seasonal precipitation differences and increasing rainfall superimposed atmospheric relative humidity and other factors have high specificity. It lead to a low correlation coefficient and precipitation. Relative humidity, average air pressure with the concentration of PM_2.5 showed a positive correlation;between air temperature, wind speed, sunshine duration, precipitation and other meteorological factors and the concentration of PM_2.5 was negatively correlated.（3） The aerosol optical thickness （AOD） and the concentration of PM_2.5 have good coherence in the spatial distribution,but they make a great difference in the numerical distribution, the monthly variation shows an inverted"U" type, regulation of season change showed that summer>spring>autumn>winter, the aerosol optical thickness through the correlation coefficient of vertical correction and ground monitoring of PM_2.5 concentration in spring and summer, autumn and winter seasons were increased by 10.73%,72.56%,7.05%,46.3%, After the correction with the vertical humidity increased by 25.85%,31.53%, revised 21.35%,6.33%.Thus, the vertical elevation and relative humidity in different extent affect the relationship between aerosol optical thickness and surface monitoring of the concentration of PM_2.5.（4） Considering the influence of vertical elevation and relative humidity, the application of regression analysis method of aerosol optical thickness （AOD） was established to estimate the multivariate linear regression model of PM_2.5 concentration,to verify the estimated value and the measured value of the correlation coefficient which reached to 0.840 on the mode, The average relative deviation is 4.66%;The inversion model estimated by the Beijing Tianjin Hebei region in spring, summer, autumn and winter seasons, the average PM_2.5 concentration values were 78.81 (μg·m-3,62.67 μg·m-3,93.51 μg·m-3,122.46 μg·m-3;the spatial distribution pattern and spatial interpolation results are consistent, estimating the PM_2.5 concentration remote sensing space to get uniform continuity, good visual effect is better than that of spatial interpolation results, sensing inversion model estimated PM_2.5 concentration remote in the Beijing Tianjin Hebei region of each city in autumn and winter are large deviation, mainly affected by dark pixel inversion method, remote sensing retrieval results in spring and summer is lower than the precision of spatial interpolation method, spatial interpolation method is higher than that in autumn and winter.This study report pollution characteristics and distribution of PM_2.5 in Beijing-Tianjin-Hebei region, using the ground monitoring and remote sensing data, A scientific research method is provided for the development of PM_2.5 pollution of fine particulate matter in the atmosphere, and to provide scientific basis for the prevention and control of atmospheric environmental pollution.