Emergency Response Model Evaluation Using Diablo Canyon Nuclear Power Plant Tracer Experiments

The Lawrence Livermore National Laboratory’s (LLNL) Atmospheric Release Advisory Capability (ARAC) provides real-time emergency response support for accidental radiological releases to the atmosphere at Department of Defense (DOD) and Department of Energy (DOE) facilities throughout the U.S. ARAC uses diagnostic three-dimensional (3-D) dispersion modeling as its primary emergency response tool (Dickerson and Orphan, 1976). The regional (20 to 200 km) modeling system is built around the MATHEW (Mass-Adjusted Three-dimensional Wind field) and ADPIC (Atmospheric Dispersion Particle-In-Cell) models (Sherman, 1978; Lange, 1989). MATHEW adjusts the wind field by variational methods to be mass-conservative and to account for terrain effects. ADPIC calculates the time- and space-varying transport and diffusion (using K-theory) of source material using thousands of Lagrangian “mass” particles on a Eulerian grid. To determine the accuracy and transferability of the MATHEW/ADPIC models to a wide variety of settings and meteorological conditions, the models have been evaluated against numerous tracer experiments over the last decade (Dickerson and Ermak, 1988; Lange, 1989). This paper presents a model evaluation using a tracer experiment in the complex, coastal terrain of the Diablo Canyon Nuclear Power Plant.