Support of Homeland Security at the 2002 Winter Olympics

The National Center for Atmospheric Research (NCAR) supported the Department of Defense (DoD) during the 2002 Salt Lake City Winter Olympics by developing and deploying an advanced operational computer-based weather forecast system. The forecasted weather variables were used as input to a DoD system that calculated the transport of hazardous material from potential releases. Figure 1 shows predicted low-level winds in the Salt Lake Valley (yellow lines), the concentration (green) of a gas that was released on the north side of the area, and the dosage (accumulated exposure, red). The brown-white shading indicates mountains that, together with the lakes in the area, produced complex winds with much spatial variation that required the use of a sophisticated forecast system. The forecast system was employed operationally for 100 days.

Predicting Plume Transport For Metropolitan Areas

NCAR has developed a sophisticated wind analysis and forecasting capability that uses operational National Weather Service (NWS) weather-radar data, in combination with other standard weather data, to predict detailed wind patterns in the lower atmosphere. Both analyses of current winds and short-term forecasts of up to 30 minutes can be produced within a couple of minutes on a dual-processor PC. Such high-resolution wind products are critical for computing the transport of hazardous material. Figure 2 shows plumes computed for two hypothetical hazardous-material releases, and computed plumes for releases at the same locations about one-half hour later. The large changes in the plume transport were due to rapidly changing wind patterns, which can only be detected using remote sensors like weather radar coupled with sophisticated analysis software. There is no other wind analysis and forecasting system that can utilize the readily available NWS radar data in this way. In this example, the data used were from the NWS radar at Sterling, Virginia.

Predicting Plume Transport Within Cities

The same software that utilized the radar wind data in the previous example can also be used with laser-based radars called lidars (light-detection and ranging) that can observe winds over smaller areas and with greater detail. The wind data must be obtained from a lidar that is specifically deployed for this purpose because there is no operational lidar network, as there is for weather radars. When lidar data are utilized, winds can be measured every 60 m (~200 feet). Figure 3 shows hazardous material tracked using lidar data. The lidar, in this example, was not in Manhattan, but the street map is used as background to provide a scale reference. Lidar systems are commercially produced. They have a range of about 7 km (~4 miles), so they can measure the winds over a significant fraction of a metropolitan area, or over any area where a high-profile event may cause security concerns.

An Operational Weather Forecasting System Developed for Army Test Ranges

During the last seven years, NCAR has been developing and deploying high-resolution, computer-based weather analysis and forecasting systems for the Army Test and Evaluation Command (ATEC). Five Army test ranges are currently served by such systems that allow them to more efficiently and safely conduct their mission. Because NCAR is a Federally Funded Research and Development Center, and not a commercial provider of services or hardware, our long-term relationship with the ATEC sponsor is unusual. One important aspect of this relationship is the process by which we regularly and seamlessly upgrade the forecasting and weather-product-delivery systems as new science and technology permit, and provide continued user training on these evolving systems. These users are like many that work in high-stress, weather-sensitive settings in that they have very little time for an in-depth analysis of complex weather products. Thus, we integrate our staff into the work-environment of the end users of the products in order to learn how they do their job, and then we build customized, timesaving, intuitive decision-support systems. Figure 4 illustrates the computational area for one of the forecast systems that is used at the Aberdeen Test Center in Maryland. Our approach is to use forecast areas that zoom in on an area of interest. The outer boxes represent the large-scale weather conditions, and the inner boxes progressively focus more on the details of the winds and other variables in the center.

Providing Decision-Support Systems and Weather Products to Emergency Responders

It is critical that first responders have information about weather and hazardous-material transport quickly and in a convenient form. Therefore, NCAR is developing methods for making the required weather and hazard information available on portable devices such as cell phones and Personal Digital Assistants (PDA). For example, first responders would input an estimated hazardous-material release point on the map displayed on a PDA. This information is then transmitted to a centralized weather-forecast system that predicts the expected track of the hazardous materials and sends a graphical product back to the emergency responder’s PDA. Figure 5 shows an image of a PDA with a plume pattern displayed on the screen. Based on this information, emergency-response personnel can conduct necessary evacuation and decontamination procedures.

Contact: Scott Swerdlin