IBM Research

Deep Thunder IBM supports regional weather forecasting at the 1996 Olympic Games Consider, for example, a highly public event like the 1996 Centennial Olympic Games in Atlanta, GA, which was composed of activities that are affected by the weather. The need to improve the quality of forecasts for such an occasion is quite clear, including: The comfort and safety of spectators -- information and warnings for specific venue sites about humidity heat index warnings lightning rain Warnings for event rescheduling for the Olympic Committee or even cancellations Winds must be < 20 mph winds for high platform diving Humidity and water condensation at velodrome cancels cycling Strong wind changes, unfair advantage to some athletes Rainout of events Use of weather conditions by athletes to their advantage Marathon (conserving energy) -- consider heat index Sailing (using information about sea breezes) Improved security: Routing and evacuation planning. As a result, the National Weather Service (NWS) of the National Oceanic and Atmospheric Administration (NOAA) agreed to provide meteorological data, and warning and forecast services for the Atlanta Committee for the Olympic Games (ACOG) via a number of new technologies. To include conditions unique to the locale to increase the accuracy of forecasts, a numerical prediction model was needed that: operates at a higher spatial resolution (e.g., two km) provides effective and timely throughput, and incorporates visualization methods for analyzing and disseminating the results more quickly An overview of the entire forecasting effort for the Olympics is also available for your reference (Lans P. Rothfusz, Melvin R. McLaughlin, Stephen K. Rinard. An Overview of NWS Weather Support for the XXVI Olympiad.  Bulletin of the American Meteorological Society, 79, No. 5, pp. 845–860). Therefore, the aforementioned Regional Atmospheric Modeling System (RAMS) was selected, which provided flexibility for the forecaster in the selection of domains and level of detail. The RAMS model was coupled with a synoptic weather model, simulating the weather across North America on a shared-memory vector processor (Cray C-90). The results of this broad range (29 km resolution) model were periodically transmitted from the National Center for Environmental Prediction of NOAA to the SP in Peachtree City, to establish boundary conditions for RAMS. Various observational data were also used as the initial conditions for the weather forecasts. These included surface data (SAO - 70 stations, mesonet - 90 stations and three buoys at the yachting venue) and analyses, quality control, dynamic adjustment and derived fields, and upper air data (Doppler radar, GOES imagery, ACARS aircraft and models), and analyses. These data were assimilated via a pre-processing step to RAMS provided by the Local Analysis and Prediction System (LAPS) as discussed earlier and interpolated to the RAMS computational grid. Since RAMS focuses on the physics of cloud formation, for example, a much more detail prediction of precipitation was possible. To provide timely production, a parallelized version of RAMS was installed on a 30-node distributed memory supercomputer (IBM RS/6000 SP) at the NWS in Peachtree City, GA. It was used during the 1996 Olympics to provide efficient and accurate weather forecasts for four main Olympic venue areas. RAMS fully exploited the parallel processing power of the 30-node SP system. The SP configuration could provide weather simulations for a 24-hour simulation over the full domain in only a few hours. As an aid in the presentation and analysis of weather simulations at different resolutions (e.g., eight km over the full domain and two km over specific Olympic venues) from this system, interactive three-dimensional visualization methods were introduced. IBM Visualization Data Explorer* (DX) provided meteorologists with new visualization methods to help support forecasting at the Olympics. These methods were developed within a "natural" coordinate system to provide a context for three-dimensional analysis, viewing and interaction. They provide representations of the state of the atmosphere, derived from RAMS output, registered with relevant terrain and political boundary maps. Examples of these techniques are shown below. Visualizations such as these facilitated the dissemination of the computed weather forecasts to the public via the World Wide Web on the Olympics Weather home page and the media, as well as the Olympic organizing committee, participants and spectators. The effort supporting the Olympics representing the first operational use of an integrated mesoscale weather model operating on a distributed memory supercomputer and three-dimensional visualization techniques. The forecasters working at the NWS office were very pleased with the improved accuracy over the synoptic scale model, the performance of the SP, and the products generated with Data Explorer. In addition, the experience in assembling and operating this system established the following characteristics: Inexpensive, with low maintenance costs as compared to high-end supercomputers typically used in operational weather forecasting todayx Easy to install and integrate with other hardware and software Reliable, with sufficient amount of hardware and software redundancy Portable, facilitating its replication in environments with similar infrastructures Easy to operate, manage, and automate the various processes involved, thus making it virtually transparent to the weather forecasters While this approach to operational weather forecasting was of immediate value at this year's Olympics, enabling athletes and attendees to plan around adverse weather conditions, these technologies could be applied in other areas where precision forecasting shows promise. Such potential applications include travel, agriculture, broadcast, pollution monitoring, and fire control and management. A paper that summarizes the visualization work for this project and the results is available for you to read.  It was presented at the American Meteorological Society (AMS) Conference, February 2-7, 1997 in Long Beach, CA. Alternatively, you can download a Postscript copy or a PDF copy of this paper, L. Treinish and L. Rothfusz, Three-Dimensional Visualization for Support of Operational Forecasting at the 1996 Centennial Olympic Games. A paper that summarizes the IBM SP server for this project and the results is available for you to read.  It was presented at the American Meteorological Society (AMS) Conference, February 2-7, 1997 in Long Beach, CA. >Alternatively, you can download a Postscript copy or a PDF copy of this paper, Z. Christidis, J. Edwards and J. Snook, Regional Weather Forecasting in the 1996 Summer Olympic Games Using an IBM SP. You may also download additional papers related to this project that were presented at the AMS conference in either Postscript or Acrobat Portable Data Format (PDF): J. Edwards, J. Snook and Z. Christidis. Forecasting for the 1996 Summer Olympic Games with the SMS-RAMS Parallel Model.  (in PDF). J. Snook, Z. Christidis, J. Edwards and J, McGinley. Forecasting Results from the Local-Domain Mesoscale Model Supporting the 1996 Summer Olympic Games.  (in PDF). P. Stamus and J. McGinley. The Local Analysis and Prediction System (LAPS): Providing Weather Support to the 1996 Summer Olympic Games.  (in PDF). P. Stamus, L. Safford, J. Johnson and L. Rothfusz. The Creation and Use of 'Interobservations' for Olympic Venues.  (in PDF). The Olympic Games weather forecasting system was a joint project between IBM, the NWS and the Forecast Systems Laboratory of the National Oceanic and Atmospheric Administration. Weather Visualizations for the Olympics You can view a number of visualizations of RAMS output and learn about the data and the techniques by selecting the category of interest below. You can look at several examples, which illustrate RAMS output at both 8 km and 2 km resolution. Each example is accompanied with an in-line image. Click on the image to see a full-size image of the same data. There is an MPEG animation for each example, and a VRML geometry corresponding to several of the examples as well as a PanoramIX scene. For your reference, the following images shows the full domain that was supported with RAMS as topography and bathymetry surface, redundantly pseudo-colored by height overlaid with coastline, state boundary and river maps. A zoomed-in representation is shown below, looking from the Atlanta area in the lower left toward the northeast. Major Olympic venue sites are marked with three-dimensional icons that illustrate the events that took place in those locations. You can also see an animation that shows the domain. Clouds and surface heat index and winds Clouds and surface heat index -- early and late forecasts Thunderstorms over Atlanta Weather over Savannah Hurricane Bertha off the Georgia coast Weather over the northern Olympic venue sites More thunderstorms... Closing Ceremonies at the Olympics Example LAPS Visualization You can see additional images and learn more about some of the earlier visualization work on this project. You can learn more about some of earlier work on the regional weather model and the parallelized implementation on an IBM SP. lloydt@watson.ibm.com [ DX Home Page | Contact DX ]   *Trademarks of the IBM Corporation in the United States or other countries or both.