A new method for the derivation of the horizontal distribution of total ozone amounts from the brightness temperature data obtained by the HIRS/2 sensor on board the NOAA satellites has been developed. This method is based on the regression method developed by Lienesch (1988), who added a transmittance term of the ozone layer as a predictor of regression, but the present method also includes the second-order terms of the brightness temperatures of Channels 1, 2, 3, 8 and 9 of the HIRS/2 sensor and the transmittance of ozone layer into the regression calculation. The total ozone data obtained by TOMS were used as the true values in generating the regression coefficients. The transmittance for the slantwise-looking condition was converted into that for the nadir-looking condition using the angle correction method introduced by Muller and Cayla (1983). The angle correction was also made for the brightness temperature at Channel 9 using the corrected transmittance. Horizontal distributions of total ozone amounts were retrieved by this method with a standard deviation of residual errors of ∼4% for the wide latitudinal region from 15° to 60°, including Japan where total ozone varies largely with latitude. Inclusion of the second-order terms into the regression improves the accuracy of retrieval, especially in the low-latitude regions. This regression. method will offer quite valuable data on total ozone because of the high spatial and time resolution, the availability of night-time data and independence of retrieved values from the existence of stratospheric aerosol.
ASJC Scopus subject areas
- Atmospheric Science