Thomas Dewitt

Starting at SFI November 2026

Thomas studies the statistical structure of Earth's atmosphere. His Ph.D. work at the University of Utah challenges the logic behind long-standing attempts to carve up the atmosphere into discrete phenomena, from hurricanes and thunderstorms to individual clouds and wind gusts. Does a thunderstorm's size imply it is created by a different mechanism than a cumulus cloud? Should we treat clouds as discrete entities just because they are visible to the human eye?

Rather than some separable ontology, Thomas argues that the atmosphere should be viewed as a single, continuous, dynamical object across the dimensions of space, time, size, and lifetime. His work has shown that observations of clouds and wind gusts are "scale invariant", supporting this continuous view—clouds have no meaningful "typical" size and turbulence is not restricted to the smallest scales, contrary to default assumptions.

Although evidence for atmospheric scale invariance is decades old, the implications remain esoteric and underexplored. At SFI, Thomas aims to develop scale invariance into a framework that is not only descriptive but can also be used for simulation and prediction of Earth's atmosphere. Thomas believes that an emergent theory of scale invariant cascades could one day replace fluids as the core of our understanding of Earth's atmosphere.

When Thomas is not working on his research he also writes a blog and enjoys making ceramics.