I work on a range of problems related to clouds, radiation, and climate. Much of this is the nuts-and-bolts of building models and interpreting remote-sensing observations. Still inspired by one of my first mentors, I'm especially interested in what we can learn from different views of the same aspect of the world.
Radiation is the one of the best-understood aspects of atmospheric sciences but applying this knowledge to atmospheric models, i.e. building parameterizations of radiative transfer, is an exercise in finding acceptable accuracy for acceptable computational cost. Alongside work to develop next-generation parameterizations of radiative transfer and couple them to emerging, highly-detailed treatments of clouds, I'm exploring noisy-but-unbiased parameterizations that exploit model dynamics, especially the insensitivity of the flow to even large amounts of small-scale noise, to reduce computational cost.
Every observation has a model attached to it; for remote sensing observations this almost always includes assumptions the target within the field of view is constant and that the signal isn't influenced by properties outside the field of view (i.e. that "one-dimensional homogenous theory" can be applied.) These assumptions fail egregiously in clouds, and I'm interested in how those failures affect the large-scale view of the world obtained from satellite observations.
The difference between model representations of clouds and what's remotely observable is particularly severe, and I'm interested in narrowing this gap for applications ranging from data assimilation (the blending of observations and forecasts to determine the state of the atmosphere) to climate model evaluation.
I'm the editor in chief of Journal of Advances in Modeling Earth Systems (JAMES), an open-access journal devoted to modeling published by the American Geophysical Union. Send me your interesting manuscripts on modeling of all kinds.
With Masa Kagayama I lead the initiative on Leveraging the Past Record of the WCRP Grand Challenge on Clouds, Circulation, and Climate Sensitivity. The grand challenge seeks to motivate the observational, theoretical, and modeling communities to make progress on a few key questions with special relevance to our understanding of future climate change.
RFMIP, the Radiative Forcing Model Intercomparison Project associated with CMIP6, is an effort to understand radiative forcing in climate models by identifying errors in clear-sky forcing, carefully diagnosing model-specific effective radiative forcing, and bounding historical forcing by aerosols.