Analysis On The Relationship Between Air Quality And Meteorological Condition In Taiyuan City

In recent years, due to the rapid expansion of economy scale and the speeding up of urbanization, the phenomenon of air pollution becomes increasingly grave in most big and medium-sized cities, such a weather phenomenon--haze has arouse the scientific community, government departments and the public’s attention. For a long time, Shanxi province, the mainly coal mine province of our country made contributes to national development, meanwhile, inevitably destroyed the original ecological environment. Now, the environment problem becomes more and more sensitive, Shanxi province needs to restore blue sky and clear water, especially Taiyuan as its provincial city needs fresh outlook to meet the future development and challenge.After years of comprehensive treatment of environment in Taiyuan city, pollution sources are basically constant, but why air quality still up and down? Meteorological conditions are the key factor of atmosphere pollution degree. In order to reveal the pollution situation of atmospheric quality further in Taiyuan city, first, based on daily haze observation data from 109 stations in Shanxi province during 1984 to 2014 to analysis the characteristics and the cause of the haze in resent 30 years. Focus the analysis on 2013 which the most haze happened year in resent 30 years, used the atmospheric particulate monitoring data which include PM10, PM2.5,PM1, NO2, SO2, CO and O3 from Taiyuan meteorological department since January 1,2013 to analysis the influence of meteorological conditions on the environmental air quality. Through analyzed the change trend of air pollution in Taiyuan city, the author divided I to Ⅶ weather classes into three categories and make statistical comparison analysis on meteorological element respectively, meanwhile, analysis typical day of air environmental quality by selecting multiple cases in heating period (not) from the three categories respectively to find out the most influenced meteorological element. Besides, discussed the process characteristics of the typical haze weather in detail, statisticed the occurrence probability of haze weather and comprehensive analysised the cause of the air pollution in Taiyuan city. Finally, the author using BP neural network established PM2.5 concentration prediction models in winterand summer half year, meanwhile, the author using data(2013 summer half year and 2014 winter half year) to examine the models,and worked well.This paper got some main results:(1) The spatial distribution of haze showed obvious difference in Shanxi, which mainly concentrated in the Taiyuan basin and Linfen basin. Haze days gradually increased along Fen river from north to south. The decadal variability of haze days in Shanxi added 10 days per decade, well above the growth of national average.(2) In 2013, four most of haze history happened in Shanxi Province:the least no haze day in 30 years; the largest scope of haze in 30 years; the most annual average haze days of meteorological station in 30 years; the most happened haze days in a wide range in 30 years. Taiyuan city roughly 30 years of haze days change present a quasi 13 years change cycle.The annual average concentration of PM2.5 in Taiyuan is 81.27μg/m3 which higher than the national average at 23.34μg/m3, only below Shijiazhuang.(3) In Taiyuan, the synoptic situation in heavy haze day and serious polluted day shows that middle and high latitudes at 500hPa is wide low tension, weak cold air intensity and frontal zone towards north. Most current in Shanxi is straight westerlies, uniform pressure field of ground, little wind and low relative humidity in Taiyuan. In Taiyuan, good air quality situation with high pressure system and north wind dominated in the majority; heavy pollution mostly with low pressure system and south wind.(4) In Taiyuan, the days of fine ambient air quality day in winter (October to March) is significantly less than the summer (April to September). The heavy pollutions of atmospheric particulates are mainly concentrated in the first and the fourth quarter in Taiyuan. In the heating period (November to March), the probability of moderate haze and severe haze appears is significantly increased.The most important member of haze is PM2.5. Throughout the year, the proportion of PM25 level at 3 and above accounted for 94.0%.(5) In Taiyuan, the level of haze day and the porbobility of polluted weaher shows positively correlated. With the increation of haze day level, the concention of PM2.5, PM10 and SO2 are also increased. The deterioration of air quality is mainly caused by PM2.5 concentration increased. The concentration of PM2.5 and the speed of ground wind shows negatively correlated, and with the change of relative humidity shows positively correlated. The longer surface layer inversion existed, the longer haze last. The average of thickness and intensity of ground inversion in haze day is higher than no haze day.(6) Establish the PM2.5 concentration prediction models in summer half year and winter half year which based on BP neural network. Using the meteorological data (July 5,2013 to August 2, 2013 and February 2,2014 to March 11,2014) to predict summer half year and winter half year network models and using PM2.5 daily average concentration to inspect during this period, the result shows that the level of forecast accuracy reached 66.4 and 76.3% respectively. The modelhas good prediction ability and the forecast accuracy in winter half year is better than the summer half year. The models provide a simple forecast method for PM2.5 daily average concentration in Taiyuan.