HR: 16:30h
AN: A54C-03    [Abstracts]
TI: MOPITT Retrieval Performance in Extreme Pollution Conditions
AU: * Deeter, M
EM: mnd@ucar.edu
AF: National Center for Atmospheric Research, Atmospheric Chemistry Division P. O. Box 3000, Boulder, CO 80307, United States
AU: Edwards, D
EM: edwards@ucar.edu
AF: National Center for Atmospheric Research, Atmospheric Chemistry Division P. O. Box 3000, Boulder, CO 80307, United States
AU: Gille, J
EM: gille@ucar.edu
AF: National Center for Atmospheric Research, Atmospheric Chemistry Division P. O. Box 3000, Boulder, CO 80307, United States
AU: Francis, G
EM: gfrancis@ucar.edu
AF: National Center for Atmospheric Research, Atmospheric Chemistry Division P. O. Box 3000, Boulder, CO 80307, United States
AU: Drummond, J R
EM: james.drummond@dal.ca
AF: Dalhousie University, Department of Physics and Atmospheric Science Sir James Dunn Building (Room 130) 6310 Coburg Road, Halifax, NS B3H1Z9, Canada
AB: Satellite-based methods for determining trace gas concentrations must be applicable in widely varying contexts. We analyze the performance of the MOPITT (Measurements of Pollution In The Troposphere) retrieval algorithm for carbon monoxide (CO) in strongly polluted atmospheres. In such conditions, retrievals of CO vertical profiles based on MOPITT observations might be degraded by at least two distinct sources of retrieval error. First, the accuracy of the operational radiative transfer model for MOPITT might be lower in conditions marked by extremely high CO loading than in more typical situations. Second, the radiative effects of aerosols, which are neglected in the operational retrieval algorithm, can potentially mask the spectral signature of CO in the upwelling radiation. In both cases, systematic differences between the calibrated radiances and model-calculated radiances can lead to biases in the CO product. We exploit a variety of approaches to investigate these potential sources of retrieval error. The effect of anomalously high CO mixing ratios can be evaluated directly by retrieval simulations; results based on the regression-based operational forward model are compared with results based on a much more accurate (line- by-line grade) forward model. The effects of aerosols on retrieval quality are more easily analyzed by empirical methods. For example, using near-simultaneous observations of MOPITT and MODIS, errors in MOPITT retrieved sea surface temperature (SST) can be related to MODIS retrievals of aerosol optical depth (AOD). Because MOPITT retrievals of SST exploit a shorter-wavelength infrared spectral band compared to MODIS, they should be more sensitive to the presence of aerosols. Correlation of the SST error and MODIS AOD provides indirect evidence of the effect of aerosols on the MOPITT radiances.
DE: 0345 Pollution: urban and regional (0305, 0478, 4251)
DE: 0365 Troposphere: composition and chemistry
DE: 0394 Instruments and techniques
DE: 1855 Remote sensing (1640)
DE: 3360 Remote sensing
SC: Atmospheric Sciences [A]
MN: 2007 Fall Meeting