Video, summary, and presentations.
The Webinar took place on Friday 28 October 2016 at 10:30 CET Time.
It consisted of two talks.
“Communicating uncertainty in climate information: insights from the behavioural sciences”
Summary of the First Talk
In the first one, Dr Andrea Taylor, from the University of Leeds (UK), talked about “Communicating uncertainty in climate information: insights from the behavioural sciences”. She discussed the importance of communicating climate information adequately to decision-makers in order to avoid a false understanding of this concept, and therefore to avoid erroneous assumptions that could lead to bad decisions. The main challenges to communicate with the climate services community were highlighted, such as the preference that users have for deterministic information, which can lead them to not correctly understanding and using probabilistic information. She pointed out at the importance of validating communication in order to ensure that what it is tried to communicate is what the user gets. Finally, some recommendations for correctly communicating uncertainty in climate information were given, such as considering user expertise in order to not present information in a way that could be falsely interpreted as deterministic, or the need to develop a highly tailored communication using an iterative discussion process with users.
Questions to the first talk were:
- Which are the differences between the approaches that could be used to communicate uncertainty in climate projections and in climate predictions?
- How do you tailor communication when you have a large group? How do you validate your communication?
“Event attribution: from research to climate service”
Summary of the Second Talk
In the second talk, Dr Geert Jan van Oldenborgh, from the Royal National Meteorological Institute (KNMI, The Netherlands), gave the talk with the title “Event attribution: from research to climate service”. He explained the importance of providing attribution statements for society. When an extreme climate event takes place, a range of users would like to know as soon as possible not only how rare this particular event is but also whether it could be attributed to climate change or if, instead, it is part of the climate variability one would expect in the absence of climate change. This type of questions is not easy to answer. First, what is and is not considered an extreme event needs an appropriate definition and is confronted with the problem of observational uncertainty. Then, one needs to explore if the probability of occurrence of certain episodes has changed due to climate change, for which the methodologies developed for event attribution are applied. These methodologies not only require data and software, but also a minimum level of experience in the interpretation of climate variability in both models and observations.
Questions to the second talk:
- What is your experience about the challenges to communicate event attribution to the general public?
- Do you see any difference in the needs for communicating extreme event uncertainties (compared to non-extreme events)?
The webinar finished with a brief description of the speakers’ experience in identifying and addressing the challenges for climate modelling and observations raised by climate services.Download the presentation (pdf – 260 KB)
Watch the full video of the webinar