The Global O₃Simulation In The20th Century From BCC-AGCM-chem0Model

Ozone（O3），as a kind of important greenhouse gas, has significant temporal andspatial variation characteristics. Greenhouse gas produced by human activities andglobal warming contrary impact the change of atmospheric O3. Therefore, the study ofatmospheric ozone is necessary for people to understand climate change and theinfluence of human activities. Use earth system model to make simulation onatmospheric O3is one of hot trends of climate model research. Earth system modelcould reflect the feedback process between climate and atmospheric chemistry moreobjectively, comparing with the traditional method that using the atmosphericchemistry model alone. Beijing Climate Center (BCC) also focus on develop its ownearth system model, as one preliminary work, has established aglobalchemistry-general circulation model BCC-AGCM-Chem0. This model is built on theframework of general circulation model BCC-AGCM2.1and increase chemistryprocess, which is taken from the formulation of atmospheric chemistry modelMOZART-2. BCC-AGCM-Chem0could simulation atmospheric circulation andchemistry process and their interactions. In this paper, BCC-AGCM-Chem0isemployed to simulation tropospheric ozone from1871to1999. We use Globalarchive of ground-based ozone data products derived from the World Ozone andUltraviolet Radiation Data Centre (WOUDC) as well as the ozone data provided byCMIP5to evaluate the model performance in climatological basic mean, seasonalvariations of O3and evolution trend from1871to1999. The main conclusions in thispaper are as follows:(1) BCC-AGCM-Chem0can well capture the spatial distribution and verticalstructure of climate mean state of ozone from1970to1999. Ozone concentration inthe Northern Hemisphere is higher than that in the Southern Hemisphere, and ispoleward increase. In comparison with the CMIP5data, BCC-AGCM-Chem0underestimated ozone column concentration over the South Ocean south of40oS byabout3~6Dobson units and overestimated it over the latitude band from40oS~50oNby about3~6Dobson units. Model biasesare more pronounced over land than overocean, which may be attributable to surface ozone emission, horizontal and vertical ozone advections. BCC-AGCM-Chem0can well capture the vertical structure of O3concentration. In the upper troposphere, the O3concentration shows its maximumvalue at high latitudes and its minimum in the tropical region; in the lowertroposphere, high O3concentration is more pronounced in the Northern Hemisphere.Simulated vertical ozone profiles agreed well with the ground-based ozone data.(2)Spatial structures of the ozone seasonal evolution are also well reproduced.Spatial correlation coefficients of the ozone column concentration seasonal variationbetween the BCC-AGCM-Chem0simulation and the CMIP5data in January, April,July and October are0.89,0.97,0.86and0.91, respectively. The ozone concentrationreaches its maximum in spring and its minimum in autumn for both the Northern andSouthern Hemispheres. The seasonal evolutions of the ozone concentration showdifferent performances below and above500hPa.(3)BCC-AGCM-Chem0can well simulate the evolutiontrend from1871to1999under the greenhouse gases produced by human activities. Under300hPa，O3concentration continue increased from1871to1999and rapidly increased between1950s to mid-1980s; Rising trend in Northern Hemisphere is higher than that in theSouthern Hemisphere; Fast-rising regions are in eastern North America，central andwestern Europe and east Asia. Above simulation results agree well with CMIP5data.BCC-AGCM-Chem0can well capture the rising evolution trend of ozoneconcentration of Sapporo station in Asia and Hohenpeissenberg station in Europewhile have some bias in other stations.