Pierre Gentine, assistant professor of earth and environmental engineering, has won a three-year $258,011 grant from NASA's New Investigator Program (NIP) in Earth Science for his research project—"A Unified Parameterization of Dry and Moist Convection"—to study turbulence in the earth's atmosphere and develop new models that can predict climate more effectively. Twenty-one proposals were funded from 131 applications.
"This is a great honor," says Gentine, who is also a member of the Institute for Data Sciences and Engineering's Smart Cities Center and of the Earth Institute. "We're trying to model something that's fundamentally very challenging: turbulence, the irregular motion of air, and its random nature. Turbulence is especially complicated because it spans such a broad range of spatial scales, from the smaller—which affect cloud formation and turbulent decay by viscous forces—to larger ones that ultimately affect the radiation over the Earth, climate circulation, and ultimately our planet's overall climate."
Earth's climate is in balance if the solar energy from the sun equals the thermal energy given off by Earth. If the equilibrium is disturbed, then our climate changes. Gentine, whose research focuses on land and atmosphere interactions and the inherent feedback between the two systems, is particularly interested in low-level clouds, which, he says, contribute significantly to Earth's radiative, or climate, balance, and to the vertical transport of heat and moisture. The representation of low-level clouds and convection in climate prediction models is tied to an accurate representation of the mixed layer and moist convection.
Much of the uncertainty in current climate prediction is related to clouds and their feedback, especially these low-level clouds. These are ultimately turbulent, Gentine explains, and are represented in climate models by the boundary layer (lower-atmosphere turbulence) and convection (cumulus) models. But these models have well-known biases, show incorrect timing and transition between different types of clouds, and imprecise diurnal cycles of precipitation, so accurate climate prediction has been difficult.
To make predictions more precise, Gentine and his colleagues have developed a prototype based on a new probabilistic approach to the problem. "Instead of considering the problem as being deterministic," he says, "we are assuming that the problem is inherently random. This randomness is being induced by turbulence, which is ultimately the birth and lifecycle of clouds."
He plans to implement this new probabilistic approach into a current climate model: the NASA GISS model-E, validating results across a broad range of test cases and data sets, to confirm that clouds can be simulated more accurately and that this new method improves climate predictions.
NASA's New (Early Career) Investigator Program (NIP) in Earth Science is designed to support outstanding scientific research and career development of scientists and engineers at the early stage of their professional careers. The program aims to encourage innovative research initiatives and cultivate scientific leadership in Earth system science. The Earth Science Division (ESD) places particular emphasis on the investigators' ability to promote and increase the use of space-based remote sensing through the proposed research.