Investigation of Systematic Uncertainties in Brewer–Mast Ozone Soundings Using Observations from a Ground-Based Microwave Radiometer

The series of balloon-borne Brewer–Mast ozone soundings is one of the few datasets that allow the reconstruction of the history of stratospheric ozone decline over the last three decades. The reanalysis of these data goes along with the revision of the soundings operation procedure, which was first defined in the 1970s. The present study focuses on the normalization of the recorded profiles to total ozone. Independent microwave remote sensing data are used to investigate the influence of systematic ozonesonde measurement uncertainties on the computed normalization factors. The analysis shows that a significant underestimation of the ozonesonde readings compared to microwave data exists above 20 hPa (∼27 km). This suggests that the loss of the sonde pump efficiency with decreasing atmospheric pressure is only partially compensated by the standard correction applied for this effect. The shape of the constant volume mixing ratio extrapolation commonly used for residual ozone is, moreover, inconsistent with the microwave ozone profile in the upper stratosphere and the mesosphere. It is shown that these inconsistencies combine in column ozone computations to yield no systematic error on the normalization factors. This balance would, however, be perturbed by the removal of one artifact or the other.