Archaeologists and historians have long investigated societal responses to climate change (see P. Palmer and M. Smith Nature 512, 365–366; 2014). These records are an underused resource in current climate-adaptation research, but offer scope for highly integrative meta-analyses that would be useful to climate scientists, science advisers and policy-makers, and could provide important information for local outreach efforts.

Risk-reduction researchers have pointed out that responses to climate change are a mix of contemporary industrial (technological) measures and pre-industrial (social and community-based) ones. However, the use of palaeoenvironmental data by the Intergovernmental Panel on Climate Change as a basis for drawing up future climate-change scenarios is not matched by an equally sophisticated use of 'palaeosocietal' data for investigating human impacts and adaptive pathways.

Archaeological and historical data could provide a solid evidence base for effective adaptations to climate change. Expanding the chronological scope of climate-adaptation research into deep time would vastly enlarge the database of available case studies without getting into the tricky issues of data access and legal sensitivity. In effect, this approach draws on natural experiments in history to learn from the past (see R. Van der Noort Climate Change Archaeology Oxford Univ. Press; 2013).