Atmosphere & Biosphere Models
We use physically-based models representing the atmosphere and biosphere in our research. These models run on our Linux-based High Performance Computing Cluster (HPCC). Most of these models are standard releases that have been modified by our group to meet a variety of research requirements. All the models are coded primarily in FORTRAN and make use of parallel processing using the OpenMP and MPI libraries.
Atmosphere Models
We use two types of atmosphere models - general circulation models (GCMs) and regional atmosphere models.
Global Models - The global models we use were developed by the National Center for Atmospheric Research (NCAR). These include the Climate Climate System Model (CCSM) which is made up of the Community Atmosphere Model (CAM), the Community Land Model (CLM), the Parallel Ocean Program (POP), and the Community Sea Ice Model (CSIM). As of 2007, we are using version 3 of CCSM, CAM, and CLM with new versions expected to be released in the middle to late 2008. CCSM3 can also be used without dynamic ocean and sea ice representation (referred to as CAM3-CLM). This modeling configuration is ideal for exploring land-atmosphere interactions because it allows isolation of the interactions between just the land and the atmosphere without the competing influence of the ocean and sea ice processes. This configuration does allow for some ocean influence via a slab ocean model and thermodynamic sea ice. We have also used the older Community Climate Model (CCM) version 3 for some of our earlier work. Much of the core dynamics and atmospheric physics in CCM3 still exists in the more recent CAM3 model.
Regional Models - We also use regional atmosphere model, the Weather Research and Forecasting (WRF) Model, developed jointly by NCAR, NOAA, NCEP (FSL), AFWA, Naval Research Laboratory, Oklahoma University, and the FAA. WRF is a mesocale numerical weather prediction system. While WRF is also used in operational weather forecasting, we use it primarily to study coarser resolution, larger domain research problems where there is a need to resolve regional and synoptic-scale processes and their interaction with the land surface. WRF has multiple dynamical cores, a 3-dimensional variational data assimilation system, and can efficiently be run in parallel. WRF is typically run on much shorter time scales than the global models and over a limited area rather than the entire globe. WRF can represent land surface processes, albeit crudely relative to general circulation models. Currently, the NOAH land surface model, a derivative of LSM, and a slab soil model are coupled to WRF. We have plans to develop a version of the IBIS model for use in WRF.
Biosphere Models
We use a variety of biosphere models to examine the interactions between the atmosphere and the biosphere. These models can either be run offline (driven with climate data) or in coupled mode with an atmosphere (e.g., CAM3-IBIS) model or climate system model (e.g., CCSM3-IBIS). A summary of the models is listed below.
The Integrated BIosphere Simulator (IBIS) - IBIS is developed and maintained by the Center for Sustainability and the Global Environment (SAGE) at the University of Wisconsin, Madison. IBIS is a dynamic global vegetation model (DGVM) that models Earth's terrestrial ecosystems. IBIS simulates a variety of ecosystem processes including: energy, water, and carbon dioxide exchanges between soil, plants and the atmosphere; physiological processes of plants (photosynthesis and respiration) and soil organisms; vegetation phenology (seasonal changes in vegetation) including budburst, senescence, and dormancy of vegetation; plant growth and competition; nutrient cycling and soil physics. IBIS contains all these processes within a single, consistant modeling framework. We run the model both offline with prescribed climate data (CRU05 data set and NCEP-NCAR Reanalysis) as well as coupled to the CAM3 and CCSM3 models. Our group is currently making a variety of model improvements with emphasis placed on better representation of Arctic terrestrial ecosystem processes (biophysical and biogeochemical) for current and future climate.
Agro-IBIS - A derivative of the IBIS model also developed at SAGE, UW. Agro-IBIS is built around the IBIS modeling framework and focuses on North American crops. Crop representation includes corn, soybean, and winter and spring wheat. Agro-IBIS allows for variable fertilizer inputs as well as irrigation and farmer management decisions. While Agro-IBIS is currently only run over North America, future work will include representation of global crops. Currently, Agro-IBIS is run offline, but plans for incorporating Agro-IBIS into the CAM/CCSM climate modeling framework are to begin in 2008 at UW.
Community Land Model (CLM) - CLM is the land model component of the Community Climate System Model (CCSM) and the Community Atmosphere Model (CAM). It was developed and is maintained by NCAR. CLM represents biogeophysics, the hydrologic cycle, biogeochemistry, and dynamic vegeation globally (coupled to CAM or CCSM) or regional (in offline mode). The land surface can be one of five sub-grid land cover types (vegetation, lakes, wetlands, urban, and glaciers) with further division into percentages of plant functional types (PFT). Each PFT has its own leaf and stem area index and canopy height. The individual subgrid land cover types separately calculate energy and water fluxes and the aggregate is passed to the atmosphere model (usually run at a coarser resolution).
