Intercomparison of tropospheric ozone column datasets from combined nadir and limb satellite observations

<p>This paper presents an intercomparison between existing tropospheric ozone column (TrOC) datasets obtained using combined limb and nadir observations, i.e., exploiting collocated stratospheric profile and total column information retrieved from limb and nadir satellite observations, respectively. In particular, seven datasets have been considered, covering the past 2 decades and consisting of monthly-averaged time series with nearly global coverage. We perform a comparison in terms of climatology and seasonality, investigate the tropopause height used for the construction of each dataset and the related biases, and finally discuss long-term TrOC drifts and trends. The overall goal of the study is to assess the consistency between the datasets and explore possible strategies to reconcile the differences between them. Despite uncertainties associated with the limb–nadir residual methodology and large biases between the mean values of the considered datasets, we identify an overall agreement of TrOC distribution patterns. The different tropopause height definitions used to construct the datasets did not show a relevant role in explaining the biases between them. We demonstrate that a thorough investigation of the drifts with respect to ground-based observations is needed to evaluate TrOC trends from satellite data and that long-term trends in specific regions can be consistently detected, e.g., a positive trend of up to 1.5 DU per decade over China for the 2005–2021 period.</p>