Statistical language backs conservatism in climate-change assessments

Abstract

The scientific evidence for anthropogenic climate change is empirically settled, but communicating it to nonscientific audiences remains challenging. To be explicit about the state of knowledge on climate science, the Intergovernmental Panel on Climate Change (IPCC) has adopted a vocabulary that ranks climate findings through certainty-calibrated qualifiers of confidence and likelihood. In this article, we quantified the occurrence of knowns and unknowns about “The Physical Science Basis” of the IPCC's Fifth Assessment Report by counting the frequency of calibrated qualifiers. We found that the tone of the IPCC's probabilistic language is remarkably conservative (mean confidence is medium, and mean likelihood is 66%–100% or 0–33%), and emanates from the IPCC recommendations themselves, complexity of climate research, and exposure to politically motivated debates. Leveraging communication of uncertainty with overwhelming scientific consensus about anthropogenic climate change should be one element of a wider reform, whereby the creation of an IPCC outreach working group could enhance the transmission of climate science to the panel's audiences.