Chapter 5: Time series and trend results

The ultimate goal of LOTUSis to improve confidence in calculated ozone trend values via an improved understanding of the uncertainties. Chapter 3 highlighted many of the challenges facing analyses of long-term ozone time series, and despite the fact that many of those challenges still need to be addressed, it is worthwhile to assess the trend results from this work in such a way as to be able to place those in the context of previous work. This chapter highlights the results of taking the " LOTUSregression" model from Chapter 4 and applying it to the different data sets (i.e., satellite, ground, and model) at different resolutions comparable to those in previous ozone assessments and comprehensive studies (e.g., WMO, 2014; Harris et al., 2015; Steinbrecht et al., 2017). The individual satellite-based trend results are then combined to obtain a single mean ozone trend profile with respective uncertainty estimates. This important yet challenging final step in the assessment has been the cause of debate in the community in recent years. Different methods for combining the individual trend results are discussed and explained, and the final trend profile estimates are analysed for significance.