Answers are grounded in this paper and cite their sources.
20th-century reanalysis datasets are an invaluable tool for understanding the climate from the beginning of the last century up to the present. They provide a best guess of the atmospheric state, based on a combination of observations and numerical modeling. Contrary to other reanalysis datasets, however, 20th-century reanalysis uses solely surface observations and is thus much less constrained. Consequently, the uncertainty of the analysis is high compared to reanalysis datasets for the satellite era. In the Arctic, where observations are even more sparse than in other parts of the globe, this issue is especially severe. Therefore, a robust estimation of the uncertainty of the reanalysis product is essential. While state of the art 20th-century reanalysis datasets provide some measures of uncertainty, they do not cover the whole uncertainty. We test whether historic independent measurements – that were not assimilated in the reanalysis – can be used to get a more reliable uncertainty estimation of temperature time-series over the last century. For this aim, we use recently digitized in-situ measurements from Alfred Wegener’s last Greenland expedition. Finally, we assess how the outcome of testing typical hypotheses – such as warming trends or comparison of different periods - is affected when considering the new uncertainty estimations.
No takes yet. Share an insight, caveat, or question.
Scher et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: