Tempo-Spatial Variations Characteristics Of Atmospheric Visibility In Hefei City And Their Influencing Factors

Atmospheric visibility,Which is an important indicator of atmospheric transparency and Closely related to human life,Its advantages and disadvantages intuitive reflect the atmospheric environmental quality in a certain area.The reduction of atmospheric visibility is mainly due to atmospheric pollutants and meteorological elements,with high concentration of PM is the main reason to poor atmospheric visibility.Hefei is one of the important central cities in eastern China,The analysis of Hefei atmospheric visibility and its influencing factors,forecasting the future development trend of atmospheric visibility will do some help to improve urben environment,perfect energy conservation and ensure traffic safety.In this paper,using atmospheric visibility and meteorological elements observation data from 6 automatic meteorological stations during 1994-2015 and particle concentration(PM2.5 ? PM10)data from 10 environmental quality monitoring stations during October 2013 and June 2015 to investigate the tempo-spatial variations characterics of atmospheric visibility in Hefei and its influencing factors.Using scientific statistical methods to research tempo-spatial variations characterics of atmospheric visibility and its influencing factors,the relationship of visibility,RH and particle concentration,and using ABC-BP neural network model to forecast the atmospheric visibility.The results show:(1)Atmospheric visibility in Hefei was generally declining,the rate of changing was-1.74km/10 a,and the most obvious decline trend in summer.Atmospheric visibility was concentrated in the third grade,the annual average visibility was 13.6km.The average daily visibility which was the lowest in 8 hour and the highest in 14 hour,The seasonal variation of visibility is the highest in summer,followed by spring and autumn,and the lowest in winter.The space distribution of visibility is different,the south,north and southeast part is higher,the central was lower,the highest area appears in changfeng county and uniform area,feidong and feixi county was lower,the lowest in the municipal district,annual average visibility less than 9 km,This distribution has relationship with distribution of land and the concentration of air pollution.Through correlation analysis,visibility was negatively correlated with relative humidity,and positive correlation with wind speed.PM2.5 and PM10 concentration was negatively correlated with visibility,and the influence of PM2.5 on visibility is than PM10.(2)The distribution characteristics of atmospheric visibility in Hefei are significant,and the low visibility is concentrated in the case of high relative humidity or high particleconcentration.When RH>70%,the change amplitude of contour line of atmospheric visibility was laeger,and the impacts of RH on atmospheric visibility were intensified.The effect of particle concentration on atmospheric visibility was different under different relative humidity,When RH<60%,the coefficients of correlation between PM2.5,PM10 concertrations and atmospheric visibility increased gradually with the increasing RH,When RH>60%,the coefficients of correlation between the particles concertrations in atmosphere and atmospheric visibility showed a decreasing trend,When RH was relatively higher,the atmospheric visibility was mainly affected by the relative humidity,on the contrary,the concertration of particles had a greater influence on the visibility.When50%?RH<60%,the coefficients of correlation between PM2.5,PM10 concertrations and atmospheric visibility were high.According to the linear model and the nonlinear model fitted by the data of RH ? particles concertrations and atmospheric visibility,there may be complex nonlinear relationship between the three.(3)In the contribution rate of each factor to the atmospheric visibility,the contribution rate of PM2.5 and relative humidity was the largest,the wind speed and PM10 concertration were the second,the contribution of air temperature and air pressure was the smallest,and the particulate pollutants were the main influencing factors of atmospheric visibility.In this paper,using ABC algorithm to optimize the weights and thresholds of BP neural network and predict the atmospheric visibility.the results showed that the BP neural network optimized by the ABC has a faster convergence rate and the forecast accuracy of82.1%,Compared with the GA-BP model and the standard BP model,the prediction error of the ABC-BP network model was relatively smaller and the prediction accuracy was higher,which can be used to predict the trend of atmospheric visibility in Hefei.