Rain does not only come from the ocean, it can also come from upwind land evaporation. Forests, lakes, soils, and cities all return moisture to the atmosphere, and that moisture falls again, often hundreds or thousands of kilometers away. The Amazon waters southern Brazil's farms. West African forests rain into the Sahel. Land use change in one place can modify rainfall somewhere else.
Our group studies these atmospheric water connections We analyze how moisture moves at continental scale, how forest loss and agricultural expansion are shifting those flows, and what this means for the cities, farms, and ecosystems on the receiving end. A central thread of the work is that water a teleconnected resource, and that understanding these flows, and governing this water well requires recognizing the dependencies that cross borders and biomes.
Selected papers
Silveira, Keys, Ruhoff, et al. (2025). Observed shifts in regional climate due to Amazon deforestation. Communications Earth and Environment. [link]
Keys, Wang-Erlandsson, Galaz, Ebbesson, & Gordon (2017). Approaching moisture recycling governance. Global Environmental Change. [link]
Keys, Wang-Erlandsson, & Gordon (2016). Revealing invisible water: Moisture recycling as an ecosystem service. PLoS ONE. [link]
Keys, van der Ent, Gordon, Hoff, Nikoli, & Savenije (2012). Analyzing precipitationsheds to understand the vulnerability of rainfall dependent regions. Biogeosciences. [link]
Full publication list on the publications page.
Climate change doesn't arrive as a single number on a thermometer. It propagates through the systems people depend on, including agricultural trade networks, migration corridors, water supplies, food systems. Some of the most consequential risks are often the ones that emerge from how these systems interact. A drought in one country can become a price shock somewhere else.
Our group studies these compound and cascading dimensions of climate change. We work on climate storylines — physically plausible accounts of how specific extreme events might unfold in a warmer world — and on how risks transmit through globally connected systems. We've also helped develop the broader framing of Anthropocene risk: the recognition that the most dangerous futures are not the ones where any single planetary boundary is breached, but the ones where social, ecological, and climatic shifts interact in ways our existing tools were not built to anticipate.
Selected papers
Im & Keys (2025). Climate storylines of annual heatwave patterns in a 3°C warmer continental USA. Environmental Research Letters. [link]
Keys, Barnes, Diffenbaugh, Hertel, Baldos, & Hedlund (2025). Exposure to compound climate hazards transmitted via global agricultural trade networks. Environmental Research Letters. [link]
Keys, Galaz, Nystrom, et al. (2019). Anthropocene risk. Nature Sustainability. [link]
Full publication list on the publications page.
Climate models can simulate the next century in temperature and precipitation — but they cannot tell us what living through it will feel like, what new institutions might emerge, or what surprises lie outside the assumptions baked into our scenarios. Some futures are easier to model than to imagine. Our group uses sci-fi prototyping — stories, games, scenario art, and structured worldbuilding — to stretch how science thinks about the Anthropocene. Past projects range from radical ocean futures to a sea-level-rise interactive fiction game set in 22nd-century Lagos, developed in collaboration with researchers across earth science, social science, and the arts.