Validation Test for CAP88 Predictions of Tritium Dispersion at Los Alamos National Laboratory

Abstract

Gaussian plume models, such as CAP88, are used regularly for estimating downwind concentrations from stack emissions. At many facilities, the U.S. Environmental Protection Agency (U.S. EPA) requires that CAP88 be used to demonstrate compliance with air quality regulations for public protection from emissions of radionuclides. Gaussian plume models have the advantage of being relatively simple and their use pragmatic; however, these models are based on simplifying assumptions and generally they are not capable of incorporating dynamic meteorological conditions or complex topography. These limitations encourage validation tests to understand the capabilities and limitations of the model for the specific application. Los Alamos National Laboratory (LANL) has complex topography but is required to use CAP88 for compliance with the Clean Air Act Subpart H. The purpose of this study was to test the accuracy of the CAP88 predictions against ambient air measurements using released tritium as a tracer. Stack emissions of tritium from two LANL stacks were measured and the dispersion modeled with CAP88 using local meteorology. Ambient air measurements of tritium were made at various distances and directions from the stacks. Model predictions and ambient air measurements were compared over the course of a full year's data. Comparative results were consistent with other studies and showed the CAP88 predictions of downwind tritium concentrations were on average about three times higher than those measured, and the accuracy of the model predictions were generally more consistent for annual averages than for bi-weekly data.

Uncertainty of current understanding regarding OBT formation in plants

“…Comparing the results of CAP88-PC, NORMTRI, and ISC-PRIME models for atmospheric transfer with the measured yearly average of air HTO concentration, those models give good predictions, but for monthly or biweekly average of air HTO concentration, large misspredictions are observed (Michelotti et al, 2013;Galeriu and Melintescu, 2012). As it was previously discussed in the present study, plant HTO and OBT concentrations depend on air HTO concentration during the previous day (HTO) or month (OBT).…”

“…Comparing the results of CAP88-PC, NORMTRI, and ISC-PRIME models for atmospheric transfer with the measured yearly average of air HTO concentration, those models give good predictions, but for monthly or biweekly average of air HTO concentration, large misspredictions are observed (Michelotti et al, 2013;Galeriu and Melintescu, 2012 (Melintescu and Galeriu, 2005) was used in that case and a significant seasonal pattern of the public dose was observed with a maximum value in September (when maize is harvested) and a minimum value in winter (Fig. 6).…”

Evaluation of Deuterium Oxide Deposition Velocity Over a Forest Environment

“…In analyses for environmental impacts and facility safety basis, downwind dispersion is modeled using Gaussian plume models such as AERMOD (US EPA 2022) and CAP88. Some of these models, such as CAP88 (Michelotti et al 2013) or UFOTRI (Raskob 1990), have been developed or used for predictions of tritium (T) dispersion in the environment. While models do account for some surface characteristics through the inclusion of turbulence parameters or adjustment of the logarithmic profile, only some, such as UFOTRI, account for vegetation interactions with the airborne plume.…”