Days after the horrific floods in Western Europe in early July that destroyed entire towns in Western Germany, articles started appearing in my newsfeed trying to evaluate whether and to what extent climate change is to blame for the extreme precipitation that caused the floods. For decades, climate scientists could only state that generally, climate change causes more and more extreme rainfall. In the past few years, however, a new field of climate science has emerged that can quantify the effect of anthropogenic emissions on a particular extreme weather event more precisely. Interestingly, this field of attribution science is altering the peer-review process to make their findings more relevant to the public discourse.
For the past decades, the peer-review process has been a pillar of scientific discovery and verification. In academia, it is frowned upon to speak publicly about results before they have been vetted and approved by this mechanism. A problem emerges, however, due to the slow nature of peer-review. A year passes typically from submission to publication of an article (Otto & Brackel, 2020). When researching extreme weather events, this clashes with a peak in public interest that wanes before a peer-review can occur. Therefore, scientists have to balance the need for unassailable results with the role of science as a service to society.
This is what attribution science does. Attribution science or probabilistic event attribution (PEA) aims to establish a causal link between certain extreme weather events and anthropogenic climate change. More specifically, the goal is to quantify whether and how much the probability or magnitude of an extreme event was changed by anthropogenic greenhouse gas emissions (Brüggemann et al., 2020). Multiple ensembles of climate models are run to compare the present world (with anthropogenic climate change) with a hypothetical world where all human influence on the climate is removed (Otto et al. 2014) “in order to isolate and quantify the effect of climate change” (Brüggemann et al., 2020). While all PEA studies contain uncertainties, confidence is higher for heatwaves and large-scale rainfall than for localised extreme precipitation or hurricanes (Otto et al., 2014).
Among others, World Weather Attribution, an international team of climate scientists, is working on rapid attribution studies (Otto & Brackel, 2020). They aim to provide reliable results about the causal link between climate change and an extreme weather event within days or weeks of the event, while there is still public interest. But, unfortunately, it is impossible to get results peer-reviewed on such short timescales. So the team decided on a new approach to solve this dilemma: During an event, they start an attribution study based on previously peer-reviewed methods. Then, as soon as the event is over and they are confident in their results, they approach the media. Later on, they get their results verified through a traditional peer-review process.
With this approach, scientists have created a new understanding of the peer-review process that strikes an improved balance between public relevance and sound scientific work. While there was some criticism of the method, it has been thoroughly vetted in an independent report by the National Academies of Sciences and is now almost universally accepted (National Academies of Sciences, Engineering, and Medicine, 2016).
However, there is, of course, a risk associated with temporarily bypassing the peer-review process: any mistakes later uncovered can potentially threaten the public trust in the scientific method. Therefore, climate scientists are compelled to make sure their results are unassailable before going public with their results. In my opinion, however, this is worth the risk as it appears to be the only valid option to influence the public discourse on the causal link between climate change and extreme weather events before interest vanishes.
So could and should this method be adopted in other fields of (climate) science? It depends, in my opinion. For the larger trends in climate, scientists should prioritise public trust in the results over a rapid distribution of results. While they undergo a lengthy process until publication, the assessment reports by the IPCC enjoy great confidence due to their rigorous peer-review. But for more time-sensitive topics – for example, the Covid-19 pandemic – using previously peer-reviewed methods on new data is preferable to the current practice of covering pre-prints that might be of far inferior quality.
References
National Academies of Sciences, Engineering, and Medicine. (2016). Attribution of Extreme Weather Events in the Context of Climate Change. Washington, DC: The National Academies Press.
Otto, F. (2020). Extreme weather events and local impacts of climate change: The scientific perspective. In M. Brüggemann & S. Rödder (Eds.), Global warming in local discourses: How communities around the world make sense of climate change (pp. 245–262). essay, Open Book Publishers.
Otto, F., & Brackel, B. von. (2020). Angry weather: Heat waves, floods, storms, and the new science of climate change. Greystone Books.
Otto, F., James, R., & Myles, A. (2014). The science of attributing extreme weather events and its potential contribution to assessing loss and damage associated with climate change impacts. Environmental Change Institute.