Engineering for a Sustainable Humanity: Predicting Extreme Weather

Apr 20 2018 | By Allison Elliott | Photo Credit: Timothy Lee Photographers | Image Courtesy of D. Kucharski and K. Kucharska/Shutterstock

Imagine a world where the hidden work of evaporation could help predict weather extremes.

From droughts to floods, climate affects extreme weather in ways that scientists need to understand and predict. But the climate system represents a complex and interconnected web of factors, which makes creating accurate models and long-term projections difficult.

Pierre Gentine, professor of earth and environmental engineering, provides insights into this system by focusing on water—in particular, evaporation—and its role in land/atmosphere interaction, with the ultimate goal of linking this knowledge back to climate response.

Through a combination of multiscale simulations, machine learning, and remote sensing, he builds models to help characterize climate and has steered a number of groundbreaking studies detailing its impact on water resources and agriculture.

One recent study from his group found that almost a third of the factors contributing to subseasonal climate variability can be attributed to vegetation (through the release of water vapor during photosynthesis). Another study revealed that plants, which can regulate evaporation through stomata (small pores at the leaf surface), will have a greater impact on water resources than precipitation or temperature in an enriched CO2 future. Plants that can access carbon more easily will show higher rates of photosynthesis and less water loss, which affects the water cycle.

With colleague Adam Sobel, professor of applied physics and applied mathematics, Gentine has worked on a modeling approach to more accurately simulate the continental tropical climate, especially in the Amazon. Simultaneously, his lab uses machine learning to better simulate clouds and precipitation in climate models, a long-standing challenge for scientists.

Gentine is also engaged in promising work on evaporation as a potential energy source. He contributed to a recent study that showed evaporation from lakes and reservoirs in the U.S. could potentially provide up to 325 gigawatts of power—equal to much of the country’s power needs.

Stay up-to-date with the Columbia Engineering newsletter

* indicates required