 |
|
 |
||||||
|
||||||
 |
Home
| Mission
|
about SciDAC
|
Contact Us |
|||||
 |
 |
 |
||||
|
Alumni ProjectMulti-Resolution Climate ModelingFerdinand Baer, Aimé Fournier, Houjun Wang, Univ. of MD SummaryThe purpose of this project is to create a global climate model with features that will provide the best available climate predictions, predictions that can be used effectively to assist in making useful decisions on issues related to climate on regional as well as global scales. To understand overall global change it must be understood that regional climate events may have a pronounced impact on the large climate scales, whereas the reverse is certainly true, and thus prediction on the regional scale is essential. It is on this basis that we have developed our multi-resolution climate model. Concurrent with the development of this model that has the capacity to predict climate "seamlessly" across various scales over the Earth, one must also recognize the inherent variability of a model. This variability is only overcome by determining model climate predictions from statistics of a number of realizations of the model climate. The model computations must therefore be as rapid as possible, to provide those statistics in a practically useful time frame, given the time constraints large models put on limited computing facilities. Our model is uniquely structured to optimize the computing time required on state-of-the-art supercomputers.A large community of global climate modelers and model contributors are currently working to create an optimum model that generates reliable climate predictions. Improvements needed in this arena include a more transparent, efficient and accurate method for producing regional climate predictions such as mesh refinement (see the collateral study by M. Fox-Rabinovitz), and the development of computing methodology which uses the latest in computing hardware (massively parallel processing or MPP) most efficiently and economically, to produce the best prediction results with minimal expenditure of resources. To meet this goal we have developed a Spectra Element Atmospheric Model (SEAM), a fairly recent concept using spectral elements as the basis of generating the dynamical core (DC) component of a climate model. The Earth's spherical domain is tiled with spectral elements that can be arbitrarily sized to meet local scaling requirements, allowing the model to create predictions over a range of scales on the entire global domain without user involvement in the computational process. The method also takes optimum advantage of state-of-the-art MPP by minimizing communication among elements and thereby amongst processors. This procedure has yielded dramatic speedup, making the production of multiple realizations more feasible. We anticipate that the method will also provide a more accurate prediction. Additionally the model will serve as a research and training tool. Since many of the forces that drive the climate system are not yet completely understood, the model can and will be used as an experimental environment to test various hypotheses on system forcing, especially in the area of spatially localized scale interactions. Knowledge gained in this process will help to refine the model, allowing it to produce improved predictions. We have thoroughly tested SEAM as a dynamical core with simple global forcing (denotedHeld-Suarez or H-S forcing) and have found it to run stably and as expected. We have added topography to the model, and have tested this using Local Mesh Refinement (LMR) to check its ability to accept the embedding of a local fine-scale grid. We ran with high resolution (LMR) over the continental USA to test the model with the experiment first done by Fox-Rabinovitz using a stretched grid with the NASA climate model, a project currently funded by the SciDAC/CCPP. In addition we ran integrations for three years starting from a resting atmosphere and analyzed the climate statistics of the third year using three different mesh size cases (fine, medium and course resolution) on a uniform grid with and without topography, then repeated these runs with LMR over the USA to capture the fine resolution of the global grid. These experiments demonstrated the advantage of LMR, especially over the mountains. We have enhanced the model by reprogramming in Fortran90. Additionally, we have implemented η-coordinates, and added a semi-Lagrange transport scheme for moisture transport, each with its own features. For consistency with the community effort in climate modeling, we have built a version of the Community Atmosphere Model (CAM, an NCAR, University and Federal Agency collaboration) using SEAM as the dynamical core (CAM/SEAM). CAM contains a ‘coupler', a module that is structured to accept a dynamical core (DC)such as SEAM and link it to the NCAR dynamical physics package (DP) and a land surface processes package (LP). We have modified the coupler to accommodate SEAM's unstructured grid. To run with DP and LP coupling we use a ‘distance weighted” method to interpolate initial and boundary data to the SEAM grid. Some properties of CAM/SEAM have been tested as follows. In addition to the experiments using H-S forcing, both CAM/Eul (NCAR Eulerian spectral DC) and CAM/SEAM were run for three years with aqua-planet conditions with and without topography and real initial conditions. Both models gave similar results, a good test for CAM/SEAM. We have also run CAM/SEAM with full physics and LP coupling using LMR over the continental USA and compared this with a uniform-grid version of the model. The results showed more sensitivity with the higher resolved grid, but the analysis of the output data is not yet complete. For the upcoming years, tests with various forms of LMR will be explored with CAM/SEAM and comparison studies will be performed with Fox-Rabinovitz using his stretched grid model. CAM/SEAM, with its modified version of the coupler will be made available to the CAM program both for experimentation, and improvement. The computations in SEAM are tuned to MPP architecture and make highly efficient use of it. We have access to the computing facilities at NERSC and ORNL, and use both extensively. We welcome the collaboration of other SciDAC researchers in making our calculations even more efficient. For further information on this subject contact:
|
||||
Home | ASCR | Contact Us | DOE disclaimer |
|
|
