Evaluation And Projection Of The Impact Of Climate Change On Surface PM_2.5 And Ozone Concentration Over China

In recent years, the issue of air quality over China has caused public concern.Among the many species of pollutants, surface fine particulate matter(PM_2.5) and ozone have the most impact upon human health and can also make a crucial contribution to radiative forcing. The assessment of surface PM_2.5 and ozone concentration can bring benefit to the development of next-generation global climate models, and can also provide reference for the other users of these models. And the projection of these pollutants will help policymakers in their efforts to limit emissions.In this paper, by employing the data from Atmospheric Chemistry and Climate Model Intercomparison Project（ACCMIP）, Coupled Model Intercomparison Project phase 5（CMIP5） models and multiple observations, and calculating root-mean-square error, temporal correlation coefficient and trend pattern, the ability of these models in simulating surface PM_2.5 and ozone concentration was evaluated on different spatial and temporal scales. Then, the change of these two pollutants concentration over China was projected based on the assessments of ACCMIP and CMIP5. Moreover, the projections from ACCMIP and CMIP5 were compared. The main conclusions of this paper are summarized as follows:The ACCMIP models can reproduce the spatial distribution of surface PM_2.5concentration over Eastern part of China well, but the simulation shows an overestimate over the Western part of China. The consistency of surface PM_2.5concentration simulated by ACCMIP models over the most parts of China is remarkable. The ability of performance-based weighted averaging（PWA） is higher than the one of equal-weighted averaging（EWA） to simulate surface PM_2.5concentration. The ensemble of ACCMIP can reproduce the trend of surface PM_2.5concentration over Northeast and Central China, Western part of Xinjiang and the coastal regions of East China well. Moreover, the two types of model ensemble cansimulate the second mode of EOF（Empirical Orthogonal Function） well.The ozone concentration from ACCMIP shows an underestimate over China. The ability of PWA is higher than the one of EWA to simulate surface ozone concentration.The poor consistency of surface ozone concentration has been found in Central part of Xinjiang and Western part of Inner Mongolia. The performance of GFDL-AM3,STOC-HadAM3 and PWA is good on different temporal scales.In this paper, by using the data of precipitation, wind speed at 10 m and 500 hPa from CMIP5 and observation, the number of air stagnation days over China was calculated. The majority of CMIP5 models can reproduce the observed spatial distribution well. The variation of modeled time evolution is slighter than the one of observation. The PWA of CMIP5 can simulate the spatial distribution of air stagnation days over China well. And the statistical relation between the observed surface PM_2.5,ozone concentration and the number of air stagnation days was established. By using this relation and the number of air stagnation days under four Representative Concentration Pathways（RCPs） from CMIP5, the surface PM_2.5 and ozone concentration over China were projected.The results from the ACCMIP and CMIP5 simulations projected the surface PM_2.5 and ozone concentration in the 2030 s and 2100 s over China respectively. Some differences can be found among these projections. In all, the projections from the ACCMIP simulations show that the concentration of surface PM_2.5 and ozone over China will decrease at the end of the 21 st century, while the projections from CMIP5 show that the concentration of these pollutants at the end of 21 st century will be close to the one at present.