Back to search

KLIMAFORSK-Stort program klima

Uncertainty communication and climate changes

Alternative title: Kommunikasjon om usikre klimaendringer

Awarded: NOK 6.1 mill.

Project Number:


Application Type:

Project Period:

2014 - 2019

Funding received from:


Partner countries:

When experts make statements about uncertain events, like the consequences of climate changes, they often use interval rather than point estimates, and probabilistic forecasts are given sometimes in words and sometimes in numbers. Such statements are also used in the public debate by journalists, politicians, and lay people, at the risk of creating an "illusion of communication." When are they used, what do they mean, and how are they understood? The present project is based on existing knowledge within the psychology of judgment and decision making, extended to uncertainty communication about issues of climate change. Three main themes have been singled out for closer examination. Interval estimates. Range estimates are typically too narrow, but people often rely more on a narrow than on a wide interval, which seems to indicate lack of expertise. But wide interval estimates are deemed as more correct after the outcome is known. Often, only minimum or maximum values are reported. Such "one-sided" intervals convey opinions and recommendations in addition to their factual content. Verbal probabilities. IPCC has developed guidelines for how to express probabilistic information in words, which say for instance that very likely means probabilities above 90%, and unlikely indicates probabilities lower than 33%. However, such phrases are not always perceived as intended. For instance, unlikely is typically used by lay people to describe outcomes that are not expected to happen at all. Our analyses further suggest that when people speak of what is possible, or what can happen, they have typically extreme outcomes in mind. Yet they are often perceived as quite likely by a recipient. Revised forecasts. Numerical and verbal forecasts are typically revised over time. We have currently been investigating how such changes can be perceived as trends that are believed to extend into the future. In addition, trends are often described with phrases indicating directions ("more than a 50% chance") The project compares uncertainty statements currently used in expert reports and the media, and investigates how they are used and read by different audiences. Results have been presented in conferences and meetings with involved parties. Several student projects are initiated and completed, in addition to ongoing projects with research partners abroad.


IPCC has developed detailed guidelines for how to express uncertainties in the field of climate change in a consistent way. At the same time, recent cross-national studies have demonstrated extensive misinterpretations of recommended terms. The aim of the present project is to go beyond these demonstrations and uncover how people perceive and understand uncertainty in climatic forecasts, and on this basis provide communicators with informed advice on how their message will be read by different audiences. The project has three parts. (1) Subproject 1 focuses on people?s perceptions of wide vs. narrow uncertainty intervals (range estimates) and upper vs. lower limit estimates. The studies are based on prior research done by Teigen, Jørgensen, Halberg, and Fostervold, indicating (a) a preference for overly narrow intervals, and (b) framing effects of single-limit intervals (different pragmatic implications of maximum vs. minimum estimates). (2) Subproject 2 concerns the interpretations of verbal probabili ties, derived from the ?which outcome approach?, recently developed by Teigen, Juanchich, Filkuková and others. Their studies suggest that predictions involving what will happen, or is certain, typically describe minimum outcomes, and what can happen, or is possible, typically describe maximum outcomes. The current project asks how people interpret such statements for positive versus negative outcomes in a context of climate change. (3) These part projects suggest several overlapping issues, including a p reoccupation with extreme outcomes, a preference for numeric vs. verbal estimates, the role of numeracy, and which kind of probability concepts people have in mind. Findings will provide insights into lay interpretations of uncertainty in scientific fore casts, and will suggest guidelines for how uncertain effects of climate change should be communicated in a reliable way, thereby preparing the public and policy makers for societal consequences of climate changes.

Publications from Cristin

No publications found

No publications found

Funding scheme:

KLIMAFORSK-Stort program klima