Customer Interview
Developing an accurate weather prediction tool
Name: Dr. James D. Doyle, MMS Team Lead
Q: Name a project that incorporated the ARL MSRC’s high performance
computers?
A: I work on a project that involves the development and application of a numerical
weather prediction model called the Coupled Ocean/Atmosphere Mesoscale Prediction
System (COAMPSTM). The model has been developed by a team of researchers at the
Marine Meteorology Division of the Naval Research Laboratory in Monterey, Calif.,
in collaboration with other government research institutions and universities.
Q: That project's impact on DoD?
A: Accurate mesoscale prediction is considered an indispensable capability for
defense and civilian applications. COAMPS is run four-times daily at the Fleet
Numerical Meteorology and Oceanography Center (FNMOC) and provides high-resolution
analyses and forecasts (for up to 72 hours) for various military-critical areas
of the world. The fields produced from these forecasts are used by the Navy throughout
the world for weather guidance in support of real-time operations such as Operation
Iraqi Freedom and Operation Enduring Freedom.
Q: What was the project’s objective?
A: To develop and evaluate COAMPS applied as a coastal/littoral data assimilation
system that can be used to provide high-resolution (grid increment less than
5 km) analysis and short-term coupled forecast guidance of the atmosphere and
ocean for tactical sized areas. Given this, we apply the system to basic and
applied research projects, which we expect will lead to an improvement in our
understanding of atmospheric and oceanic processes, as well as the forecast system.
Q: What was the methodology?
A: In order to produce a state-of-the-art atmospheric forecast model, we needed
to incorporate a suite of approaches based on fluid dynamics theory, representations
of the atmospheric physical processes, data-assimilation techniques, and computer
science. The atmospheric component of COAMPS includes an analysis and quality
control system that can routinely ingest in-situ and remotely-sensed data and
produce analyses of the current conditions. The forecast model is based on the
nonhydrostatic formulation of the primitive equations and includes nested grids,
terrain-following coordinates, and representations for atmospheric processes
such as precipitation. The ocean component of COAMPS also features an analysis
capability that can assimilate in-situ and remotely-sensed observations, and
the NRL Coastal Ocean Model (NCOM), designed for mesoscale ocean applications.
Our approach makes use of the infrastructure within COAMPS to produce unique,
state-of-the-art, high-resolution atmospheric fields for forcing NCOM and other
ocean models.
Q: What were the overall results?
A: The simulations from COAMPS demonstrate that it is a robust data assimilation
and forecast system capable of predictions and simulations on the horizontal
scale of 0.3-9 km for land-sea effects and topographically driven flows.
Q: What hardware at the MSRC did you use?
A: The SGI Origin O3K.
Q: How many computer hours did you use?
A: Our group used more than 600,000 hours at ARL MSRC in 2003 and more than 400,000
hours in 2002.
Q: What ARL MSRC personnel assisted you?
A: Tom Kendall, Phil Matthews and Steve Thompson have all helped with our codes
on the SGI Origins at ARL MSRC.
Q: What has been the highlight of your career?
A: One of the more interesting studies I have been involved with was an investigation
of mountain wave induced turbulence over the Alps as part of the Mesoscale Alpine
Programme (MAP). During the field program that took place in 1999, we used forecasts
conducted in real time from a number of weather models including COAMPS to design
research aircraft observation strategies. I was a member of a team of meteorologists
that flew onboard an instrumented research aircraft that observed mountain waves
and associated turbulence.
Q: What project are you working on now?
A: The investigation of terrain-induced rotors in the atmosphere.
Q: Do you have any hobbies?
A: Outdoor activities such as hiking and bicycling.