The Computational Geosciences program at UC Davis focuses on the role of atmospheric models and modeling tools in building an understanding of atmospheric phenomena and improving weather forecasting systems. Computing power has increased exponentially over the past decade, stimulating a significant burst of new research focused on developing software that can take advantage of the rapidly advancing hardware. In particular, it has become increasingly important to design atmospheric models which are capable of scaling on systems with tens to hundreds of thousands of processors. Hence, significant effort has been directed to the study and application of modern numerical techniques to simulating the atmosphere, developing tools for assimilating observations into predictive models and building algorithms for processing and analyzing large climate datasets.
Select a faculty member's name below to visit their web page.
Profesor Biello's research focuses on the structure of the partial differential equations which govern the dynamics of fluids. Particularly fascinating is the wave phenomena in fluids - in particular in rotating fluid systems, of which Earth's atmosphere is our most important example. Nonlinearity in the rotating fluid equations create two important effects on the waves they describe. (1) Wave steepening and breaking is due to the interaction of the nonlinear and the linear properties of the wave. (2) Non-linearity causes interactions across a wide range of length scales, yielding upscale or downscale transports. The tropical atmosphere is a fascinating laboratory to study these waves and the effects of nonlinearity on natural phenomena. On set of phenomena that I have been studying is called tropical/midlatitude interactions. For example, nonlinear coupling causes tropical waves to generate midlatitude Rossby waves; weather patterns over Indonesia can suppress or enhance weather over the continental USA. Furthermore, nonlinear interactions can cause midlatitude Rossby waves to excite precipitating Kelvin waves over the deep tropics. A third phenomenon is the Madden-Julian Oscillation (MJO). The MJO is a classic example up a nonlinear, upscale atmospheric phenomenon wherein small scale structures (storm systems) organize to produce a large scale circulation (the MJO covers more than a quarter of the tropical Earth, and moves across three quarters of it).
Professor Chen's research involves data assimilation and the study of idealized heavy orographic rainfall. A major forecasting problem is that there is not enough in-situ data over the oceans. One of the possibilities for improving model forecasting is to properly utilize remote sensing data to improve model initial conditions. In collaboration with Dr. Francois Vandenberghe at NCAR, Dr. Grant W. Petty at the University of Wisconsin, and Dr. James F. Bresch at NCAR, her group has applied SSM/I and QuikSCAT data to improve hurricane simulations. For heavy orographic rainfall, in collaboration with Dr. Y. L. Lin of North Carolina State University, she has studied the effects of the moist Froude number and the convective available potential energy on flow regimes associated with a conditionally unstable flow over a mesoscale mountain. In addition to these two directions, her research will also be extended to cumulus parameterization and the on-line tracer study in the near future.
Professor Kleeman's research uses supercomputing to perform air quality simulations for large regions (continental U.S.) with fine spatial scales (4km) and time scales (hourly) over long time periods (~decade). These simulations quantify the effect of climate on air quality and establish realistic exposure fields for future health impacts. Professor Kleeman has also developed a source-oriented WRF/Chem model that keeps track of aerosol mixing states more realistically than standard model configurations. The source-oriented WRF/Chem model tracks the mixintg state of black carbon and brown carbon for simulations that consider feedbacks between anthropogenic air pollution and meteorology. Professor Kleeman maintains the largest computational cluster on campus operated by a single PI - currently +400 nodes, ~1200 compute cores, ~400T redundant storage.
Professor Nathan's research is fundamental in nature and focuses on identifying and understanding the physical and dynamical mechanisms that govern the spatial and temporal evolution of large-scale atmospheric circulation systems. Professor Nathan's research involves combining observations with advanced mathematical techniques to study the following: tropical-midlatitude interactions during El Niño and La Niña flow regimes; stability of geophysical fluid flows; nonlinear dynamics of atmospheric circulations; and interactions among radiation, ozone and dynamics in the stratosphere. Additional research includes using proxy data (e.g. dendroclimatic reconstruction and historical records) to examine the impacts of meteorological events on exploration, including the Lewis and Clark expedition.
Professor Paw U studies the physical and biometeorological processes responsible for exchanges of momentum, heat, and gases such as water vapor between the lower atmosphere and vegetated surfaces. These processes are fundamental to understanding how forests, for example, absorb pollutant gases, how agricultural crops utilize water, and how plant communities exchange carbon dioxide with the atmosphere. The plant biometeorology research encompasses experimental observation in the field, numerical modeling, and theoretical analysis of turbulent mechanisms in and above plant communities. Experiments involve using fast response instruments to measure turbulence, such as sonic anemometers and infrared gas analyzers (IRGAs). Current projects include estimating turbulent parameters and dispersion coefficients for a California regional air quality study, and determining, by eddy-covariance and mean advection methods, and the carbon exchange between the atmosphere and a 500-year old, 65 m high forest at the Wind River Canopy Crane Research facility (WRCCRF). Our research group is measuring carbon dioxide fluxes, and is in collaboration with a group measuring biogenic hydrocarbon emissions from the forest canopy. Recent data indicated this old-growth forest is surprising active and is annually sequestering approximately 2 tons of carbon per hectare, similar to younger forests. Other areas of research focus on the observation and analysis of repeatable patterns in the turbulent wind fields. These characteristic motions, or coherent structures, appear to play an important role in the overall exchange process. Numerical modeling work involves two main topics, the first being state-of-the-art Large Eddy Simulation (LES) of turbulence within and above plan canopies, using the NCAR supercomputer system. The second topic is the numerical modeling of plant canopies using higher-order closure turbulence equations linked with radiation, energy budget, and plant physiology models. This set of models has been named the "Advanced Canopy-Atmosphere Simulation Algorithm" (ACASA). It has been connected to the regional scale model MM5 and can provide a regional scale understanding of ecosystem-atmosphere interactions of radiation, the energy balance, carbon, water, other gaseous and particulate emissions, transport, and deposition. In addition, Professor Paw U has studied the thermal budget of animals and humans in response to atmospheric variables.
Professor Ullrich's work focuses on the design and development of next-generation numerical methods for modeling atmospheric dynamics, and the tools needed for processing and managing large climate datasets. His work further focuses on the use of variable-resolution atmospheric models for capturing regional-scale phenomena, features which currently are unresolved in global atmospheric modeling systems. Variable-resolution climate modeling is a cutting-edge technology, with efforts only recently directed towards support for multiple mesh scales within a single framework. It is also a timely endeavor, since demands for fine-scale resolution of atmospheric features have taxed the limits of our most powerful computing systems. However, even as variable-resolution software systems have moved forward, there has been a discernible lag in our understanding of exactly how these systems improve the representation of regional-scale behavior. Intuitive regional metrics which include both expectation and variability of seasonal precipitation and surface temperature, as well as the count and variability of extreme weather events, can be readily estimated by multi-resolution Earth modeling systems. These metrics are of considerable importance in informing regional-scale policy and implementation for agricultural planning, forest fire prevention, urban development, and many other relevant fields. A better understanding of the uncertainty present in the estimates of these metrics is significant for both improving our scientific understanding of the Earth system and lending credibility to the conclusions drawn from these models.