| With the rapid development of China's economy and the acceleration of urbanization,air pollution in China is becoming more and more serious.The central-eastern China(CEC),a typical East Asian monsoon(EAM)region,has been experiencing the frequent heavy pollution incidents caused by high concentrations of atmospheric particulate matter.The formation of air pollution is not only controlled by the emissions of anthropogenic aerosol pollutants,but also affected by weather and climatic conditions.Weather and climate conditions can affect atmospheric physical and chemical processes such as air pollutant emission,chemical conversion,dry and wet removals and regional transport.The sea-land dynamic and thermal forcing anomalies control the climate change of EAMs,and its impact on air quality change in the CEC is still a scientific issue to be studies in depth.Therefore,based on the meteorological and environmental observation data for multi-years and through the combined way of climatic statistical analysis and numerical simulation,we studied the effects of EAMs climate change,tropical Pacific thermal forcing and the Tibetan Plateau(TP)topography on the variation of atmospheric particulate matter concentration in the CEC and its mechanism.Meanwhile,using a machine learning method called the random forest,the associated climate factors were applied to the cross-seasonal climate prediction of winter haze days in China.The main research results are summarized as follows:(1)Effect of East Asian monsoons on interannual variations of aerosol concentrations in central and eastern China and its mechanismTo identify the contribution of meteorological factor to air quality change,an aerosol simulation from 1995 to 2004 with the global air quality model GEM-AQ/EC was designed without interannual variations in the anthropogenic aerosol(including sulfate and organic and black carbon)emissions over the 10-years.To assess the impact of interannual variations of EAMs on air quality change in China,this modeling study focused on the CEC,a typical EAM region with high anthropogenic aerosol emissions.The simulation analysis showed that the interannual variability in surface aerosols over CEC was driven by fluctuation in meteorological factors associated with EAM changes.Large amplitudes of interannual variability in surface aerosol concentrations reaching20–30%relative to the 10-year averages were found over southern CEC in summer and over northern CEC in winter.The weakened near-surface winds of EAMs in both summer and winter were significantly correlated with aerosol increases over most areas of CEC.The summer and winter monsoon changes enhance the surface aerosol concentrations with increasing trend rates exceeding 30%and 40%over the southern and northern CEC region,respectively,during the 10 years.The composite analyses of aerosol concentrations in weak and strong monsoon years revealed that positive anomalies in surface aerosol concentrations during weak summer monsoon years were centered over the vast CEC region from the North China Plain to the Sichuan Basin,and the anomaly pattern with“northern higher”and“southern lower”surface aerosol levels was distributed over CEC in weak winter monsoon years.Aerosol washout by summer monsoon rainfall exerted an impact on CEC aerosol distribution in summer;aerosol dry depositions in connection with atmospheric boundary layer conditions resulted in wintertime aerosol variations over CEC.Climate change with regard to EAMs could modulate the interannual variations in aerosols and air quality over CEC by changing near-surface winds,precipitation and atmospheric boundary layer.(2)Climatic modulation of thermal forcing in the Pacific Ni(?)o3.4 region on winter haze pollution change in southern ChinaBased on observational data from 1981 to 2010,we show that the thermal forcing of sea surface temperature(SST)in central and eastern Pacific could significantly exert an influence on the winter haze frequency in Southern China(SC).We identify the influences of anthropogenic emission and meteorology to haze variations over SC through the discrete wavelet transform(DWT)method.The winter haze frequency over SC is negatively correlated with simultaneous Ni(?)o3.4 index with R=-0.51.The number of haze events over SC is lower in the above-normal Ni(?)o3.4 SST winters with 3-5 days and higher in below-normal Ni(?)o3.4 SST winters with 3-5 days,which are attributed to the anomalies of near surface wind speeds,atmospheric stability in the lower-troposphere,anomalous vertical motion and precipitation,accordingly.In the winter of El Ni(?)o,the regional surface wind speed increased,and the abnormal"cold shield"pattern of vertical thermal structure made the atmospheric stratification unstable,which was conducive to the transport and diffusion of pollutants.Meanwhile,more precipitation strengthened the wet removal of pollutants,resulting in less haze pollution.The winter of La Ni(?)a is the opposite.(3)Spatial and temporal variations of atmospheric aerosols over the CEC influenced by the Tibetan Plateau's dynamic and thermal forcingUsing CESM 1.2 model,the altitude of the Tibetan Plateau(TP)was set to 1000m to investigate the effect of topography on the distribution of aerosols over the CEC.The downstream area east of the TP is a weak wind speed zone and the vertical gradient of wind speed is not obvious below 700 h Pa.Aerosol concentration is generally high in the CEC,with high aerosol concentration centers in Sichuan Basin(SB)and North China Plain(NCP).After the topography of TP was removed,the aerosol concentration decreased in the vast area from NCP to SB with a deciline of 5-8?g m-3,but increased in the rest of the CEC,especially over the southeast coastal ar with a growth of 2-6?g m-3.Further analysis showed that besides the effect of barrier and circumfluence was disappear due to the removal of the TP topography,the thermal effect(TP is commonly condidered as a weak cold source in winter)is also removed,which results in the differentiation of climate in CEC.The winter monsoon system retreated northward and weakened,and the NCP was still affected by the monsoon.Additionally,the westerly anomaly in the middle and lower layers in troposphere after the removal of the plateau facilitates the outward transport of aerosols in NCP.The unique positive vertical circulation anomaly in SB,combined with the"cold sheild"vertical thermal structure,is conducive to the mixing of pollutants and transport to the lower reaches out of the basin through the westerly anomaly.The weakening of winter monsoon resulted in decrease in surface wind speed and a lack of precipitation in the rest areas of the CEC,which is unfavorable to aerosol diffusion and wet removal.Adding import of aerosols from the upstream SB,the concentration of aerosols in the region has a positive anomaly.The results of 10-year GEM-AQ/EC simulation experiments show that the atmospheric aerosol concentration in winter over the CEC,compared to winters with a weak heat source on plateau,increases by 30-45%in winters with a strong the heat source.The abnormal thermal conditions on the TP may lead to the abnormal vertical thermal structures of the"warm shield"and"cold shield"over the CEC.The vertical thermal structure of the"warm shield"intensifies the subsidence of the lower troposphere,which is conducive to the accumulation of heavy pollution and the occurrence of haze events.The influence of the"cold shield"is contrary.The thermal forcing anomaly over the TP has an important influence on the atmospheric aerosol concentration in the CEC region.(4)Cross-seasonal climate prediction of haze pollution in winter in southernChina based on key factors of land-sea thermal forcingBased on meteorological and environmental observations from 1980 to 2013,the relationship between the Ni(?)o3.4 SST and the interannual variation of haze pollution in the southern region in winter was determined.The precursory signal of WHD variability can be detected in the tropical central-eastern Pacific SST,which is amplified since August to winter season.Based on the August-October mean SSTAs in Ni(?)o3.4 and three other identified predictors,we developed a seasonal prediction model of the WHD using RF regression method.The correlation coefficient between the predicted values of haze days by the random forest model and the observed values is as high as 0.95,which is obviously better than that predicted by the multiple linear regression model.The results show that the random forest model has potential application in cross-seasonal climate prediction of haze pollution in winter in China.This implies that the SSTAs over tropical Pacific play a significant role in the variability of WHD in the SC. |
PDF Full Text Request