Observational Studies Of Reactive Trace Gases And Aerosols At Mount Tai

Reactive trace gases(O3,CO,SO2,NO,NO2,NOy,etc.)and aerosols(PM2.5,PM10,etc.)are important constituents of the troposphere and have large effects on the air quality,human health,the Earth's ecosystem and climate.A better understanding of their spatiotemporal distributions,physiochemical processes and impacts on regional air quality provides support for national air pollution prevention and control.In addition,the measurement data of reactive trace gases and aerosols can be used to evaluate the performance of regional/global atmospheric chemistry transport models.In this study,an atmospheric composition monitoring station was built on the top of Mount Tai.Long-term(2 years)continuous measurements of trace gases and aerosols were conducted at this site.During the sampling period(from August 2017 to July 2019),O3 had higher levels in summer and lower levels in winter,SO2 and NO.concentrations were higher in winter than in summer.Meanwhile,neither PM2.5 nor CO showed an obvious seasonality.Reactive trace gases and aerosols at Mount Tai both peaked in the daytime and reached a minimum during the night.Compared to other air pollutants,the timing of the maximum and minimum values of PM2.5 and O3 was delayed by photochemistry.Correlation analysis showed that ozone production efficiency was higher in warm seasons and lower in cold seasons or nighttime,PM2.5 nitrate and sulfate were related to secondary formation from gas precursors.The back-trajectory analysis suggested that Mount Tai was mainly influenced by relatively clean air masses originating from the Northwest at high altitudes.At the meantime,clusters from the Jing-Jin-Ji area and East China had high levels of air pollutants.Comparison with measurements at the Tai'an station(ground-level)showed that Mount Tai had lower concentrations of all air pollutants,except for O3.Moreover,the diurnal and seasonal variabilities of air pollutants at these two sites showed relatively large diffrences.Results showed that the Tai'an site was largely affected by local emissions,and Mountt Tai was more influenced by long-range transport and the evolution of PBL.Aerosol optical properties,their temporal variations,and influencing factors at Mount Tai in the spring and winter seasons of 2017 and 2018 were investigated.During these periods,absorption coefficients of aerosol(aap),scattering coefficients of aerosol at 550 nm(?sp-Green)were lower than reported values for other national monitoring sites,single scattering albedo(SSA)were comparable to the measurements at Mount Tai and Jinan in 2014.BC??ap??sp?back scattering coefficients of aerosol at 550 nm(?bsp)?extinction coefficients of aerosol(?ext)?SSA?mass scattering efficiencies(MSE)exhibited diurnal trends,with values at the daytime higher than the nighttime values,which was mainly caused by aerosol emissions,mountain and valley breezes,the PBL evolution,and wet deposition.backscatter fraction(BSF)showed higher nighttime and lower daytime values.Both PM2.5 and BC were significantly and positively correlated with ?sp?absp and ?ext.SSA was positively correlated with PM2.5,but negatively correlated with BC and BSF.Concentrations,absorption and scattering properties of aerosols at Mount Tai were,to a large extent,affected by the Jing-Jin-Ji area and Central China.To test the performance of low-cost sensors(for CO,O3 and PM2.5)at Alpine station,a 7-month long inter-comparison campaign was carried out at Mount Tai using sensors and the reference instruments.Strong positive linear relationships between the sensor data and reference data were found with r of 0.83.0.79 and 0.62 for CO,O3 and PM2.5,respectively.Assessment of the influences of weather conditions on the performance of sensors showed that when the temperature was above 23?,the measured CO and O3 concentrations were positively biased.The sensor tended to overestimate PM2.5 under high relative humidity conditions(significantly positive biases were found when the relative humidity was higher than 65%).Sensor measured CO concentrations were negatively correlated with the wind speed.A multiple linear regression(MLR)model and a random forest(RF)model were adopted to minimize the influence of meteorological factors.The MLR method increased r to 0.91,0.82 and 0.92 for CO,CO,O3 and PM2.5,respectively.The RF method increased r to 0.96,0.94 and 0.94 for CO,CO,O3 and PM2.5,respectively.Results showed that low-cost sensors have the capability and potential for the measurement of air pollutants at high mountain sites with complex weather conditions.The findings will help better understand the features of air pollutants at the top of the PBL and in the free troposphere in the NCP,and provide scientific support for regional air pollution prevention and control.