Assessment of the error budget for stratospheric ozone profiles retrieved from OMPS limb scatter measurements

Abstract

Abstract. This study presents an error budget assessment for the ozone profiles retrieved at the University of Bremen through limb observations of the Ozone Mapper and Profiler Suite – Limb Profiler Suomi National Polar-orbiting Partnership (OMPS-LP SNPP) satellite instrument. The error characteristics are presented in a form that aims at being compliant with the recommendations and the standardizing effort of the Towards Unified Error Reporting (TUNER) project. Besides the retrieval noise, contributions from retrieval parameters are extensively discussed and quantified by using synthetic retrievals performed with the SCIATRAN radiative transfer model. For this investigation, a representative set of OMPS-LP measurements is selected to provide a reliable estimation of the uncertainties as a function of latitude and season. Errors originating from model approximations and spectroscopic data are also taken into account and found to be non-negligible. The choice of the ozone cross section is found to be relevant, as expected. Overall, we classify the estimated errors as random or systematic and investigate correlations between errors from different sources. After summing up the relevant error components, we present an estimate of the total random uncertainty on the retrieved ozone profiles, which is found to be in the 5 %–30 % range in the lower stratosphere, 3 %–5 % in the middle stratosphere, and 5 %–7 % at upper altitudes. The systematic uncertainty is mainly due to cloud contamination and model errors in the lower stratosphere and due to the retrieval bias at higher altitudes. The corresponding total bias exceeds 5 % only above 50 km and below 20 km. After computing the estimate of the overall random and systematic error components, we also provide an ex-post assessment of the uncertainties using self-collocated OMPS-LP observations and collocated Microwave Limb Sounder (MLS) data in a χ2 fashion.