Localness in Climate Change

In The Great Derangement, the novelist and anthropologist Amitav Ghosh asks why we find it so difficult to think about climate change. The question starts with literature but then extends to history and politics. His answer is that in all three arenas, the modern mindset has promoted the primacy of human, indeed individual, agency. In this mindset, there is no place for the uncanny—and climate change, especially at the local scale, is nothing if not uncanny.¹ Not only does climate change require a rather different conception of agency, it also asks us to consider its geography as both uneven and unequal. For instance, the cities established through colonialism have been largely sited on coasts, in defiance of threats from tropical cyclones and coastal flooding, indeed often on landfill. The localness of climate change is perhaps most salient in these vulnerable and exposed places. Yet climate science takes the scale of the global as normative and the local as accidental. It has thereby “detached knowledge from meaning”² and enacts forms of what the philosopher Miranda Fricker has called “hermeneutic injustice.”³ We describe here some ideas emerging from within climate science, informed by other disciplines that express this critique and aim to address it.

The Scale of the Global

Global warming is, as the phrase says, global. Though the rate of human-induced temperature increase is not uniform in magnitude—most dramatically, the Arctic is warming faster than elsewhere—nearly everywhere on earth is warming. And the responsible increases in greenhouse gas concentrations, even more than temperature, are truly global. Carbon dioxide is long-lived and well-mixed in the atmosphere, with a nearly uniform concentration, so a ton of CO₂ emitted anywhere is the same as a ton emitted anywhere else. Its global scale is one of the aspects that makes climate change such a daunting challenge for the human species.

Our confidence in the basic fact of a substantial human influence on climate rests in large part on our solid understanding of the greenhouse effect and its thermodynamic consequences—that is, it directly increases temperature and almost as directly increases atmospheric moisture and raises sea level—which act, fundamentally, at the global scale. Yet the impacts of the warming will be felt at the local scale and will differ according to all the particularities of place, both physical and human: geographic, social, political, economic, and cultural. The models used to make climate predictions have spatial resolutions on the order of 100 kilometers. Resolution in these models is analogous to the resolution, or pixel size, in a digital camera image; nothing of this size or smaller can be represented. Yet a typical city, for example, is much smaller than this. Methods have been developed to “downscale” climate predictions, but these tend to magnify uncertainties, which were already substantial at larger scales.⁴ Thus the need to understand global warming’s impacts on human society and to adapt in order to reduce the harm it does puts pressure on climate science to produce predictions at spatial scales much smaller than those at which it has historically been able to do so with any confidence.

In addition to being global in physical space, the dominant mode of doing and communicating climate science is also global in another, more conceptual sense: namely, it is global in the space of possible trajectories, or histories, of the climate system.

Chaos theory teaches us that immeasurably tiny, unknowable differences in the present state of the weather rapidly become large differences in the future state. This is the reason that useful forecasts of the daily weather more than a couple of weeks in advance are not possible and that while the atmosphere in principle obeys deterministic laws, in practice its behavior has a component that is indistinguishable from that of a stochastic, or random, system. It also means that the actual weather state, and in fact the entire temporal sequence of weather, in a sense, is—within bounds set by larger physical constraints on the climate—an accident, just one possible history among many equally possible...