1. Jinlong Li 1, Jun Li 1, Christopher C. Schmidt 1, Timothy J. Schmit 2, and W. Paul Menzel 2 1 Cooperative Institute for Meteorological Satellite Studies (CIMSS), University of Wisconsin-Madison, 2 NOAA/NESDIS/STAR Study of Total Column Ozone Retrieval From the Current GOES Sounder ACKNOLEDGEMENT: This program is supported by NOAA GOES Improved Measurement and Product Assurance Plan (GIMPAP). REFERENCES Li, J., C. C. Schmidt, J. P. Nelson, T. J. Schmit, and W. P. Menzel, 2001: Estimantion of total ozone from GOES sounder radiances with high temporal resolution. J. Atmos. Oceanic Technol., 18, 157-168. 5. Bias estimates and their corrections 1. Introduction 2. Algorithm for total column ozone retrieval Figure 10. (a) OMI level 2 ozone products minus collocated GOES-12 ozone retrievals in May 10 of 2005; (b) GOES-12 ozone retrieval bias estimated from regression model; (c) GOES-12 ozone retrievals after bias correction. 3. Comparison of GOES-8 retrievals and TOMS products 4. Comparison of GOES-12 retrievals and OMI products Bias: TOMS minus GOES-8 retrieval t: current days in the year latitude: pixel latitude O 3 : GOES retrieved total ozone angle: satellite viewing angle The radiance measured by the current Geostationary Operational Environmental Satellite (GOES) Sounder provide hourly information for atmospheric temperature, water vapor as well as total column ozone. CIMSS has been experimentally generated hourly total column ozone product from these measurements. In this study an updated algorithm with new training data sets and a bias correction for GOES total ozone retrieval is addressed. Ozone is one of the most important chemical constituents in the atmosphere. It plays a critical role in the UV radiation, stratospheric temperature, and air quality studies. High spatial and high temporal measurement of ozone with good accuracy from space is needed. Figure 1. The atmospheric temperature, water vapor and ozone mixing ratio (water vapor and ozone are expressed as the logarithm of the mixing ratio) component weighting functions for GOES-8 ozone band. (Li et al., 2001) In generalized form, GOES ozone is retrieved by the following regression expression: O 3 : either ozone profile or total column ozone Tb: radiance measured by first 15 GOES bands j: GOES channel number P s : Surface pressure latitude: pixel latitude month: current month of measurement : satellite viewing angle The regression coefficients were generated from a nearly global training data sets containing 6408 atmospheric temperature, moisture and ozone profiles along with physically assigned surface emissivities and surface skin temperatures. The associated radiances were calculated with a fast forward radiative transfer model. Two retrieval schemes were investigated: one integrating a retrieved ozone profile and another retrieving total column ozone directly. The direct total column ozone retrieval shows a better result and is applied in this study. Figure 2. Comparisons of old algorithm and new algorithm by percentage root-mean-square differences between GOES-8 total ozone retrievals and collocated TOMS products. Figure 3. Scatter plot of collocated GOES-8 total ozone estimates and TOMS level 2 ozone measurements: (a) for June 1998 and (b) for January 1999. (a)(b) Figure 4. Scatter plot of collocated GOES-8 total ozone estimates and TOMS level 2 ozone measurements: (a) for 45N-50N from January to September 1999 and (b) from May 1998 to September 1999. (a)(b) Figure 5. Scatter plot of collocated GOES-12 total ozone estimates and OMI level 2 ozone products in May 10 of 2005. Figure 6. (a) OMI level 2 total ozone products at UTC 17:43 in May 10 of 2005; (b) Total ozone retrievals from GOES-12 at UTC 18:00 in May 10 of 2005. (a) (b) Figure 7. Monthly differences between TOMS ozone products and GOES-8 ozone retrievals from May 1998 to September 1999. Figure 8. Differences between TOMS ozone products and GOES-8 ozone retrievals before and after bias correction from May 1998 to September 1999 (a) for month (b) for latitude. (a)(b) Figure 9. (a) Scatter plot of collocated GOES-8 total ozone estimates and TOMS level 2 ozone measurements after bias correction from May 1998 to September 1999; (b) same as (a) but for GOES-12 retrievals and OMI products in May 10 of 2005. (a)(b) 6. Conclusions Based on statistical regression procedure, a direct total column ozone retrieval method was developed and applied to GOES Sounder radiance measurements. GOES-8 ozone estimates in 1998 and 1999 showed root-mean-square difference (RMSD) of 3-6% with collocated TOMS level 2 ozone measurements. A case study of May 10 2005 showed a similar difference between GOES-12 ozone retrievals and OMI products. Furthermore, GOES ozone retrieval biases were estimated by using GOES-8 retrievals and TOMS data through a regression model. After bias adjustment, GOES Sounder ozone retrievals have been improved for both GOES-8 and GOES-12. More GOES-12 cases are needed to validate these results. (a) (b) (c)