R. P. Addis and R. L. Buckley
Westinghouse Savannah River Company
Aiken, SC 29808

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

This report has been reproduced directly from the best available copy.

Available for sale to the public, in paper, from:  U.S. Department of Commerce, National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161,  phone: (800) 553-6847,  fax: (703) 605-6900
email:  orders@ntis.fedworld.gov
online ordering:  http://www.ntis.gov/help/index.asp

Available electronically at  http://www.osti.gov/bridge

Available for a processing fee to U.S. Department of Energy and its contractors, in paper, from: U.S. Department of Energy, Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831-0062,  phone: (865 ) 576-8401,  fax: (865) 576-5728,  email:  reports@adonis.osti.gov

Prognostic atmospheric dispersion models are used to generate consequence assessments, which assist decision-makers in the event of a release from a nuclear facility. Differences in the forecast wind fields generated by various meteorological agencies, differences in the transport and diffusion models, as well as differences in the way these models treat the release source term, result in differences in the resulting plumes. Even dispersion models using the same wind fields may produce substantially different plumes. This talk will address how ensemble techniques may be used to enable atmospheric modelers to provide decision-makers with a more realistic understanding of how both the atmosphere and the models behave.

Meteorological forecasts generated by numerical models from national and multinational meteorological agencies provide individual realizations of three-dimensional, time dependent atmospheric wind fields. These wind fields are then used to drive atmospheric dispersion (transport and diffusion) models, or they are used to initiate other, finer resolution regional meteorological models, which in turn drive dispersion models. Many modeling agencies now utilize ensemble-modeling techniques to determine how sensitive the prognostic fields are to minor perturbations in the model parameters. However, the European Union programs RTMOD¹ and ENSEMBLE², described elsewhere in this session, are the first projects to utilize a WEB based ensemble approach to interpreting the output from atmospheric dispersion models. The ensembles produced are different from those generated by meteorological forecasting centers in that they are ensembles of dispersion model outputs from many different atmospheric transport and diffusion models utilizing prognostic atmospheric fields from several different forecast centers. As such, they enable a decision-maker to consider the uncertainty in the plume transport and growth as a result of the differences in the forecast wind fields as well as the differences in the models. This provides a better understanding of the atmosphere and plume behavior than would a single model output. Atmospheric models often give the impression of greater accuracy and precision than the science is capable of delivering. The ensemble approach is a powerful way to reassert the concept of having a family of equally valid solutions, while enabling outliers to be identified.

The U.S. Department of Energy’s Savannah River Technology Center has participated in RTMOD and ENSEMBLE. SRTC uses the Regional Atmospheric Modeling System (RAMS)³ and Lagrangian Particle Dispersion Model (LPDM)⁴ to provide plume forecasts in real-time for the European grid as described in figure 1. The NOAA northern hemispheric model, Aviation (based on the Medium Range Forecast, MRF), is used to provide the initial and boundary conditions for RAMS. The model plume forecast data are sent to the ENSEMBLE WEB page in real-time where they may be compared with other model outputs.

On February 5, 2002 the ENSEMBLE modeling groups were notified that a hypothetical release of Cs-137 had occurred at Nantes, France as a result of a fire. The source term provided to the modelers involved a release of 0.9E15 Bq/h from the ground surface up to 1000m, and 0.1E15 Bq/h from 1000 m to 1300 m. The duration of the release was given as 6 hours, commencing at February 5, 2002 11:45 UTC. The requested forecast horizon was to February 7, 21:00 UTC. Figure 1 shows an overlay of the SRTC (crosshatched) initial 60-hour forecast for the plume overlaid on an ensemble of 8 other model outputs. The plume shadings (colors on the WEB page) show the level of consensus for a minimum threshold. This kind of output enables modelers to determine consensus between models and identify if a particular model is an outlier.

Figure 2 shows the instantaneous surface concentration as a function of time at a particular location (in this instance, Luxembourg) for the above-mentioned exercise. This figure demonstrates considerable variation in the arrival time and duration of the plume at Luxembourg. Although there are no field data, since no material was actually released during this experiment, nevertheless it provides useful information about how the models vary in their representation of the plume passage. This tool is particularly useful in assessing the uncertainty in model predictions for plume times of arrival and duration. Traditional approaches to model evaluation have focused on assessing which models perform best. However, it is unfortunate that insufficient experiments hinder clear conclusions. The ensemble approach enables us to understand how the models behave under varying atmospheric conditions.

The traditional approach to providing atmospheric consequence assessment tools to aid decision-makers in response to a release from a nuclear facility is to provide a plume output from a particular model. However, the non-unique nature of solutions to the non-linear equations that govern the atmosphere, and the sensitivity of such equations to perturbations in the initial and boundary conditions, results in any single model output being simply one of many viable solutions. As such, the traditional approach does a disservice to decision-makers by inferring greater definitiveness to the plume forecasts than is reasonable to expect from the model or from the atmosphere itself.

An ensemble approach to atmospheric consequence assessment modeling as demonstrated during the European ENSEMBLE and RTMOD experiments provides decision makers with greater insight into uncertainty in the model outputs, as well as atmospheric behavior.

1. Bellasio, R., Bianconi, R., Graziani, G. and Musca, S., "RTMOD: An Internet based system to analyse the predictions of long-range atmospheric dispersion models", Computers and Geosciences, 25 (7), 819-833, 1999.
2. Galmarini S., R. Bianconi, R. Bellasio, G. Graziani, "Forecasting the consequences of accidental releases of radionuclides in the atmosphere from ensemble dispersion modelling, L. Environmental Radioactivity, 57, 203-219, 2001.
3. Pielke, R.A., et. al. "A Comprehensive meteorological modeling system – RAMS", Meteor. Atmos. Phys., 49, 69-91, 1992.
4. Uliasz, M., "The atmospheric mesoscale dispersion modeling system", J. Appl. Meteor., 32, 139-149, 1993.