Estimating the Tropospheric Ozone Distribution by the Assimilation of Satellite Data

Tropospheric ozone is important to the environment, because it acts as a strong oxidant to control the concentrations of many reduced gases (methane, carbon monoxide, ... ), its radiative forcing plays a significant role in the greenhouse effect, and direct contact with ozone is harmful to human health. Tropospheric ozone, whose main sources are intrusion from the stratosphere and chemical production from source gases associated with urban pollution or biomass burning, varies on a wide range of spatial and temporal scales. Its transport and chemistry can be influenced by weather, seasonal, or multiannual variability. Despite the importance of tropospheric ozone, it contributes only about 10% of the total ozone loading in the atmosphere. Consequently, satellite instruments lose sensitivity below the stratospheric ozone peak, and provide little information about middle and lower tropospheric ozone. This talk will discuss recent modifications made to the satellite ozone data assimilation system at NASA's Data Assimilation Office (DAO) in order to provide better tropospheric ozone columns and profiles. We use a version of the system that assimilates only the data from the Solar Backscatter UltraViolet/2 (SBUV/2) instrument. The quality of the assimilated ozone in the tropical troposphere is evaluated by comparison with independent observations obtained from the Southern Hemispheric Additional Ozonesondes (SHADOZ) network. It is shown that the quality of ozone fields is sensitive to the winds used in the transport model. Increasing the vertical resolution of the model also has a beneficial impact. The assimilated ozone in the lower troposphere was substantially improved by inclusion of tropospheric ozone production, loss, and dry deposition rates from the Harvard GEOS-CHEM model. The mechanisms behind these results will be examined and the implications for our understanding of tropospheric ozone will be discussed.