CAREER: Understanding Low-cloud Feedbacks Using Large-eddy Simulation of Spatially Developing Cloud Transitions

A traveler flying from California to Australia might take off through a flat, unbroken cloud layer and watch as the layer gets lumpy, with stratocumulus clouds protruding from the stratus layer as the flight tracks over warmer water. Eventually the cloud deck breaks up and the ocean appears between puffy cumulus clouds, which start out small but grow into towers over the warm equatorial Pacific.

The breakup of the cloud deck, referred to as the stratocumulus to cumulus transition, matters for climate as a cloud deck cools the earth by reflecting sunlight to space while the ocean absorbs most of the sunlight that falls on it. One concern here is that the stratus to cumulus transition is tied to sea surface temperature (SST), thus the warming of SSTs by greenhouse gas increases could lead to additional warming by favoring well-separated cumulus clouds over a continuous stratocumulus cloud deck.

Concern over this feedback, in which warming begets further warming, has prompted intense interest in the stratocumulus to cumulus transition.

Researchers studying stratocumulus clouds have found a powerful tool in Large Eddy Simulation (LES) models, in which the cloudy atmosphere is simulated on a grid mesh with flow variables like wind speed, temperature, moisture, and cloud amount defined on a lattice of gridpoints spaced perhaps a few tens of meters apart. LES models produce surprisingly realistic simulations of clouds, but they require considerable computing power and the cloud fields they simulate are not more than a few tens of kilometers wide.

The small domain is a severe limitation in studies of the cloud transition, which takes place over hundreds of kilometers. A further limitation is that cloud behavior is strongly influenced by the large-scale atmospheric circulation, particularly the planetary-scale overturning of the Hadley cells (one each for the Northern and Southern Hemispheres).

The Hadley cells feature rising motion in the towering cumulus clouds over the warmest SSTs balanced by subsidence over the subtropics. The overturning motion and accompanying large-scale atmopsheric characteristics have to be externally imposed in LES simulations, which is unfortunate as changes in the cloud transition may influence the overturning circulation in ways that could feed back on the cloud transition.

Reseach under this award addresses the mismatch between the accessible domain sizes for LES models and the much larger areas covered by the stratocumulus to cumulus transition and the even larger Hadley cells. The strategy involves the development and use of a hierarchy of three LES model configurations, the first of which is a small-domain LES domain with periodic boundaries (clouds flowing out of the western side of the domain reenter on the eastern side).

The small domain can be slowly moved over progressively warmer SSTs to induce the cloud transition. The second configuration is a long channel with inflow at the northern end (representing conditions near California) and outflow at the southern end (the "Australian" end). The domain is narrow but periodic in the longitudinal direction to save on computational expense, and not as long as an actual Pacific transect.

Nevertheless preliminary work shows that the domain is capable of producing the stratocumulus to cumulus transition if an SST contrast is imposed between the northern and southern ends, along with appropriate subsidence. The third configuration is similar to the second only much deeper and closed at the ends so that it generates its own overturning circulation, smaller than but analogous to a Hadley cell.

Several experiments are planned, for instance with varying carbon dioxide concentration to study changes in the cloud transition as climate warms. Analysis of model results is guided in part by the theory of horizontal convection, meaning the behavior of convection generated by temperature contrasts along the top or bottom boundary of a domain. Horizontal convection has been extensively studied in oceanography but is not a common framework for understanding clouds and large-scale atmospheric overturning.

The education and outreach component of this CAREER award takes advantage of the realism of clouds simulated by LES models. The project works with the Visualization Services and Research Lab (VisLab) at the National Center for Atmospheric Research (NCAR) to create a collection of cloud images and animations that will be offered to the public as an online Cloud Gallery.

A version of the Cloud Gallery will be exhibited at the William Benton Museum of Art (WBMA), a public museum in Connecticut. The project also involves efforts to incorporate art into science teaching, acknowledging the perceptive portrayal of clouds and turbulence found in artworks such as Da Vinci's sketches of waterfalls. One such effort is conducted through the Joule Fellows program, a six week summer program for K-12 teachers. Joule Fellows work with the project team to develop visual materials for their classroom teaching.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.