The production of policy-relevant research has become a central mandate in the climate change scientific domain, as highlighted by the sheer number of published articles on policy issues.

Climate scientists understand that they play a critical role in informing the policy decision-making process. Many of them work remarkably hard to produce more robust greenhouse gas emissions estimations in relation to climate change targets. The hope, at present, is to illuminate governments gathering together at several upcoming international United Nations meetings, starting this September in New York, to initiate the process that will lead to a new global climate change agreement at the end of next year in Paris.

But scientists do more than just that. They also investigate what mitigation options those estimations entail, including solutions to deliver negative emissions (see Commentary by Fuss et al. on page 850 of this issue). They even discuss how countries could effectively and fairly share the carbon emissions mitigation burden to avoid dangerous climate change, as highlighted in a Perspective by Raupach et al. on page 873. Supposedly all this work, represented at its highest synthesis by the Intergovernmental Panel on Climate Change reports, feeds policy discussions and therefore is policy-relevant. Yet, global climate change policy is not delivering solutions in line with scientific knowledge. We all know this, and scientists now more than ever agree that it's the case. One would expect that, as a result, some changes in both scientific and policy practices would have already begun. Indeed, awareness about the need for change in the science–policy domain is gradually emerging within the broad climate change community, but the way forward remains blurry.

Social scientists are concerned about the nature of the gap between science and policy. Many think that without revisiting the lens through which natural scientists look at policy on one side, and the way negotiators frame policy goals on the other, the gap will remain. In a Commentary on page 853, David Victor lucidly underlines how most of the policy-relevant climate science practice develops around the idealistic view that governments “separate the decisions about how much warming is tolerable from the choices about who pays for policy action”. He goes on to suggest that somehow even diplomats contribute to the same vision by setting “abstract global goals”. This would explain at least in part why the science–policy gap remains. What's next? There are many calls for the proper integration of the social sciences into the climate change research agenda as a crucial step to fill the gap, and to reframe the policy-relevant research questions. However, making this happen hinges on a number of challenging steps, including broadening the notion of scientific relevance and excellence beyond conventional criteria and revisiting our understanding of the societal importance of science. Such a process cannot be accomplished by scientists alone, no matter what research field they belong to. Research funders, scientific outlets, governments and ultimately the public all have a role to play.