Carbon emissions must be halved over the next decade to hold the global average temperature increase to the range in the Paris Agreement, “recognizing that this would significantly reduce the risks and impacts of climate change.” But even this bold mitigation will not eliminate the risks humans face from climate change. In the coming years we expect to see increased climate change disruptions related to rising sea levels, extreme droughts, intensified tropical cyclones, and terrestrial and marine heat waves. These climate change hazards interact with increasing vulnerability and exposure of human societies, compounding risks in complex ways over time. Because of this, the Paris Agreement also established “the global goal on adaptation of enhancing adaptive capacity, strengthening resilience and reducing vulnerability to climate change, with a view to contributing to sustainable development and ensuring an adequate adaptation response in the context of the temperature goal.” But what exactly does this goal consist of? It isn’t specified and is an active topic of debate and negotiation. While there have been substantial advances in attributing extreme weather to climate change and in understanding risk dynamics, it remains difficult to specify concrete goals for reducing these climate risks. In contrast, the Intergovernmental Panel on Climate Change (IPCC) Special Report on Global Warming of 1.5°C and working group III’s contribution to the Sixth Assessment Report provided four global mitigation pathways detailing how many gigatons of CO2 emissions needed to be eliminated and the timelines for doing so to avert dangerous anthropogenic climate change. But nothing like this exists in relation to climate change adaptation, neither in terms of precise adaptation goals nor in terms of prospective scenarios to minimize and address adverse climate impacts at the global scale, hence many unresolved questions. This paper is an exploration of the state of adaptation science, understand from the global perspective, and develops a new formation of the global goal on adaptation which is tied to the burning embers and sustainable development frameworks. This formulation of the goal is one that, if followed, will assure sustainable development remains obtainable and is measurable, providing guidance to countries trying to make policies to reduce climate risk.

Two ideas that run through much of Western environmentalist thought are (1) nature is that which is untouched by humans, and (2) intervening in nature is generally bad, morally and epistemically. These ideas continue to be quite influential in environmental conservation. They define what successful outcomes look like, and what strategies are allowable for promoting these outcomes. An important contribution of environmental philosophy and philosophy of science has been to question these ideas about nature and intervening in nature. There is, for example, a rich tradition of challenging the human/nature conceptual dichotomy (e.g. Callicott 2000) and a growing literature that critiques excessive caution about conservation interventions (e.g. Brister et al. 2021). Our paper accepts these critiques and takes them as a starting point. But where the existing literature focuses almost exclusively on environmental conservation, we examine policy and discourse around climate change adaptation. We show that suspicion about “hi-tech” interventions that are perceived as unnatural, such as marine cloud brightening and genetic engineering, is common in adaptation discussions and decision-making (e.g. Van Haperen et al. 2012). At the same time, there is often uncritical acceptance of nature-based solutions (the use of natural features and processes to address environmental challenges) (Holl and Brancalion 2020). This bias against the technological and in favor of what is perceived as natural is familiar, and it has unfortunate consequences. It prejudices planners, managers, and the public against promising adaptation policies and unnecessarily pits so-called “green” versus “gray” solutions against one another, when in fact combining the two is likely crucial for effective adaptation (Seddon et al. 2020). What alternative heuristics might make better guides for decision-making around climate change adaptation? We explore this question in the rest of the paper and introduce a framework for communicating about and evaluating adaptation strategies that emphasizes three key insights from the adaptation literature (e.g. Browder et al. 2019). First, particular adaptation solutions only make sense as elements of larger adaptation packages, which need to deliver benefits over multiple time-scales. Second, the novelty or technical complexity of any given adaptation solution is much less important than its testability. Third, ongoing monitoring of adaptation solutions is critical, both for winning public support and addressing knowledge gaps.

Anthropogenic climate change (CC) poses a serious global threat, and human responses to this problem are usually framed in terms of mitigation (the reduction of human actions that contribute to climate change) and adaptation (the response to actual or expected impacts of changes in the climate with the aim of reducing vulnerabilities and enhancing opportunities). So far, philosophers interested in climate change science have mostly focused on the epistemic and value theoretic issues in climate modelling (Katzav and Parker 2018), independently on how information flows from experts (climate scientists) to lay people (the public, policy makers, etc.), and the decisions-types that take climate change information into account. One promising way in which philosophers have analysed the dependency between the epistemology of climate science and the purpose it serves in the so-called “adequacy-for-purpose” view (Parker 2009, 2020; Baumberger et al. 2017), a pragmatic approach to model evaluation. However, this approach only focuses on physical climate modelling, possibly leaving out some important issues that arise in generating decision-relevant information in the context of climate change adaptation. Here, we take a different approach. We argue that while mitigation and adaptation are both important responses to CC, supporting mitigation and adaptation actions requires a different type of information, that incurs in different epistemic and value theoretical problems. Information derived from climate change science for mitigation action requires information at the global scale, but information for adaptation requires climatic information at local spatial scales and long temporal scales. If this information need is taken to involve only considerations about climate models (as is mostly done for deriving information to support mitigation), the high degrees of uncertainty tied to information for adaptation can hinder effective adaptation decision making (Dessai et al. 2009). These considerations have prompted physical climate scientists and environmental social scientists to develop new approaches to developing information for adaptation. We first highlight some epistemic issues that have been raised in approaching developing information for CC adaptation in the same way as CC mitigation (e.g. Dessai and Hulme 2004, Wilby and Dessai, 2010). We then show that producing CC information requires taking elements that go beyond the physics of climate change into account, such as local vulnerabilities and attitudes towards risk of decision makers (Adger et al 2009, Shepherd and Lloyd 2021). Finally, we illustrate a new methodology that addresses this need (Goulart et al. 2021, Ciullo et al. 2021) and identify what novel epistemic and value theoretic issues this methodology entails.

Climate change policy—including matters involving both mitigation and adaptation—is informed by assessments of the risks and damages caused by climate change. Such assessments, in turn, depend on our ability to attribute specific impacts to climate change. Of particular interest is the exacerbating effect climate change is having on damaging, costly and potentially deadly extreme weather events, including heat waves, droughts, floods, storms and tornado outbreaks. A vigorous debate has arisen among researchers regarding best practices for attributing such events to climate change. This debate hinges on issues that are both scientific and philosophical in nature. Among the complicating considerations are (a) the differing levels of confidence in “thermodynamic” (direct effects of warming) and “dynamical” (indirect effects related to changes in atmospheric circulation and stability) influences, (b) the relative merit of “storyline” vs probability-based approaches, and in the latter case (c) alternative preferences for frequentist vs. Bayesian approaches to statistical inference. Last, but not least, is (d) the limitations of climate model-based attribution approaches in capturing subtle, real-world linkages between climate change and extreme weather events that are not well resolved in current generation climate models. I will review the current state of play in this debate, discussing some of my own research contributions in these areas.