1997 ARSC Cray Grant Recipients
University of Alaska Fairbanks Fairbanks, Alaska Shusun Li, awarded $25,000 Geophysical Institute, 907-474-7676
Automated Topography Production System Based on Synthetic Aperture Radar (SAR) Interferometry Techniques Existing workstation tools developed at the Alaska SAR Facility are the basis for this two year project to develop a fully automated, massively parallel processing capability to produce high quality topographic products from computer compatible signal data (CCSD) products from the Alaska SAR Facility. The main results of the project include an optimized computer code for automated generation of digital terrain elevation data, and accurate digital terrain elevation models in test sites.
Peter Olsson, awarded $26,500 Geophysical Institute, 907-474-6477
Trial High-Resolution Regional Weather Forecasting Project for Alaska Relatively little high resolution atmospheric modeling has been done in high latitudes. The unique ìparameter spaceî found in Alaska -- very stable boundary-layer stratification, multiple thin cloud layers, dramatic topographic variations, and extremely low surface temperatures -- pose several problems not encountered in the midlatitude environment. The thrust of this project is to make modifications to existing computer models using the CRAY T3E system and verify model simulations for Alaskan weather against observations, a task that researchers in the Geophysical Institute are uniquely positioned to do.
Mark W. Govett, awarded $15,500 Forecast Systems Laboratory Boulder, Colorado 303-497-6278
Developing an Automated Meteorological Forecast System on the CRAY T3E Using the RAMS Model Tasked with developing and testing high resolution Numerical Weather Prediction (NWP) forecast models on massively parallel processing systems for technology transfer to the National Weather Service (NWS), NOAAís Forecast Systems Laboratory (FSL) has developed the Scaleable Modeling System (SMS) to improve portability and performance of parallel codes and to reduce development costs. This project will focus on the Regional Atmospheric Modeling System (RAMS) developed at Colorado State University for simulating and forecasting meteorological phenomena. The team will port the SMS version of the RAMS model to the ARSC CRAY T3E supercomputer in support of research endeavors at the University of Alaska Fairbanks to set up an automated forecast system for the Alaska region.
Andrey Proshutinsky, awarded $18,000 Institute of Marine Science, 907-474-7834
High Performance Computer Code for a Coupled Ice-Ocean Models The Arctic Ocean and its adjacent seas, and the Bering and Okhotsk Seas are the major regions of scientific research to benefit from the development of this coupled ice-ocean model with advanced physics. As a major scientific tool to investigate seasonal variability of ice and water dynamics and thermodynamics, it also models the features of the hydrographic regime at different spatial and temporal scales.
University of Montana Missoula, Montana Donald Morton, Jr., awarded $18,333 Department of Mathematical Science, 406-243-2830
CRAY T3E Coupling and Optimization of Parallel Hydrologic and Thermal Models for Arctic Regions Hydrologic and thermal processes play fundamental roles in the global climate, but are not well represented in current general circulation modes (GCMís). This project will integrate spatially distributed hydrologic and thermal models for arctic regions and provide evaporation and condensation parameters to GCMís that are necessary for determining factors such as cloud cover. The integrated model will realistically simulate the positive and negative feedback interactions between the interdependent hydrologic and thermal systems.
Naval Postgraduate School Monterey, California Wieslaw Maslowski, awarded $30,000 Department of Oceanography, 408-656-3267
High Performance Modeling of the Arctic Ocean and Sea Ice in Turbulent Equilibrium Using the capabilities of the CRAY T3E system at ARSC, this project will complete the design and optimization of an eddy resolving, coupled Arctic Ocean and sea ice model in order to conduct century-long integrations. Results obtained to date from a 120 year coupled model simulations have already provided an unprecedented wealth of new information regarding both large- and regional-scale sea ice and ocean circulations. Early tests on the ARSC CRAY T3E system indicate that these codes can already attain three-fold faster speeds than on the older CRAY T3D supercomputer. The major improvements proposed in this project will enable the code to be portable and scaleable for future generations of SGI/Cray machines.