Comparison of ozone data derived from SBUV and SAGE with emphasis on longitudinal variations

Abstract

Stratospheric ozone observations by the SAGE and SBUV satellite instruments in March and April, 1979 have been analyzed. All SAGE profiles have been smoothed vertically over 8 km to provide some compatibility with the SBUV vertical resolution. Comparing the zonal mean ozone mixing ratios against smoothed LIMS profiles, it is inferred that SAGE is systematically overestimating ozone by approximately 20% at tropical latitudes at pressures lower than 5 mb and that SBUV is underestimating ozone by approximately 15% at 50–70° latitude at 10 mb. A comparison of the longitudinal variations of ozone by SBUV and SAGE is made and the detectability of planetary waves in ozone is emphasized. The uncorrelated portion of the SAGE variances are found to be approximately consistent with the SAGE noise model. Based on the correlated variances, the amplitudes of the smoothed SAGE planetary waves in ozone are found to be the same, on average, as in the SBUV experiment at mid-latitudes between 1 and 10 mb. Planetary wave detectability is illustrated during two several day periods at mid-latitudes and a persistent and theoretically-consistent relationship between ozone and temperature is noted. These examples, however, indicate that differences between ozone planetary wave amplitudes derived from the two sensors may occur when there is a strong vertical gradient in wave amplitude.