Quality checks for satellite and ground-based total ozone observations

The accuracy of total column ozone measurements, ground-based as well as satellite ones, is severely influenced by weather, in particular by the state of cloudiness at the measurement site. It is therefore of interest to find an approach to check the reliability of these measurements. Linear multiple regressions, connecting total column ozone with a variety of meteorological parameters and derived from data of a midlatidude Central European station, are shown to be a suitable quality check for total ozone measurements. Nearly all measurements with more than 20 DU deviation from the statistical estimate were identified as influenced by bad weather. The exceptions could be explained by errors in one of the predictors, the total ozone amount of the preceding day.