HR: 0830h
AN: A21D-0992    [PDF]
TI: TOMS Ozone Anomalies and Ozone Retrieval Errors Over Cloudy Areas
AU: * Liu, X
EM: xliu@cfa.harvard.edu
AF: Atomic and Molecular Physics Division, Harvard-Smithsonian Center for Astrophysics, 60 Gardent Street, Cambridge, MA 02138 United States
AU: Newchurch, M
EM: mike@nsstc.uah.edu
AF: Department of Atmospheric Science, University of Alabama in Huntsville, 320 Sparkman Drive, Huntsville, AL 35085 United States
AU: Kim, J
EM: jaekim@pusan.ac.kr
AF: Department of Atmospheric Science, Pusan National University, 30 Changjeon-Dong, Pusan, 609-735 Korea, Republic of
AU: Bhartia, P K
EM: Pawan.Bhartia@nasa.gov
AF: Atmospheric Chemistry and Dynamics Branch, NASA Goddard Space Flight Center, NASA/GSFC, Code 916, Greenbelt, MD 20771 United States
AU: Loughman, R
EM: loughman@wrabbit.gsfc.nasa.gov
AF: Cooperative Center for Atmospheric Science and Technology, University of Arizona, 1118 E 4th St, Tucson, AZ 85721 United States
AB: This study characterizes TOMS Ozone Retrieval Errors (OREs) associated with incorrect Cloud-Top Pressures (CTPs) and with assuming opaque Lambertian clouds, investigates these errors' effects on tropospheric ozone derivation, and analyzes ozone anomalies over TOMS data. Large errors occurring in TOMS assumed CTPs and inaccurate CTP-caused OREs are most significantly from inappropriately added ozone below clouds. Because OREs are usually within the TOMS retrieval precision when Cloud Optical Depth (COD)$>$20, assuming Lambertian surface is good. Because of In-Cloud Ozone Absorption ENhancement (ICOAEN), assuming opaque clouds can introduce large positive OREs even for optically thick clouds. For a 2-12 km water cloud of COD 40 with 20.8 DU ozone inside the cloud, the ORE is 17.8 DU at nadir. The ICOAEN effect depends strongly on viewing geometry and inter-cloud ozone amount and distribution; it is typically 5-13 DU over the tropical Atlantic and Africa and 1-7 DU over the tropical Pacific for deep convective clouds. The negative errors from using the TOMS Partial Cloud Model (PCM) partly cancel other positive errors. At COD $<$ 5, the TOMS algorithm retrieves approximately the correct total ozone because of compensating errors. With increasing COD up to 20-40, negative PCM effect decreases to almost zero, and the overall positive ORE increases and is dominated by ICOAEN effect. The ICOAEN effect can largely underestimate tropospheric ozone derived from cloudy/clear difference techniques. The convective cloud differential and cloud-clear pair methods use minimum ozone above clouds to cancel positive errors. A Positive or Negative Ozone Anomaly (POA/NOA) is defined to occur if the ozone/reflectivity correlation coefficient in a region is $>$0.5 or $<$-0.5. Average fractions of OA occurrence are 31.8% and 35.8% in Nimbus-7 and Earth-Probe TOMS data, respectively. Most tropical NOAs result from large cloud-height errors; corrections lead to 50-70% POAs in the tropics because of mainly the ICOAEN effect. POAs with fractions of 30-60% occur in marine stratocumulus regions west of South Africa and South America. OREs over clear and cloudy areas cause about half the ozone/reflectivity slope; greater ozone production from frequent low-altitude clouds and rich ozone precursors may cause the remainder. The knowledge of TOMS OREs has important implications for ozone/trace gas retrieval from other satellites.
DE: 0368 Troposphere--constituent transport and chemistry
DE: 1610 Atmosphere (0315, 0325)
DE: 1640 Remote sensing
DE: 9399 General or miscellaneous
SC: Atmospheric Sciences [A]
MN: 2003 Fall Meeting