Compare CAP‐88 and MCNP calculations of cloudshine. Accuracy of Cloudshine Gamma Dose Calculations in the CAP-88 Dispersion Model Michael W. McNaughton, Jessica M. Gillis, Elizabeth Ruedig, Jeffrey J. Whicker, and David P. Fuehne* Abstract: The U.S. Environmental Protection Agency dispersion model, CAP‐88, calculates ground-level dose using the ground-level concen- tration and the semi-infinite cloud approxima- tion. Doses can be underestimated for elevated plumes during stable atmospheric conditions at receptor locations within a kilometer down- wind of a stack. The purpose of this paper is to identify when CAP‐88 calculations of gamma dose from cloudshine are inaccurate and provide estimates of the inaccuracy. The method used compares CAP‐88 estimates with Monte Carlo N-Particle (MCNP) estimates. Comparisons were made at distances of 800 m and 3,000 m downwind of the stack and for plume heights from 0 to 50 m. For these conditions, the an- nual dose calculated by CAP‐88 is greater than or equal to that calculated by MCNP. Health Phys. 112(4):414–419; 2017. Key words: operational topics; emissions, at- mospheric; modeling, dose assessment; Monte Carlo INTRODUCTION The U.S. Environmental Protection Agency (U.S. EPA) computer model, CAP‐88, (U.S. EPA 2012–2015) is used to calculate the air-pathway contribution to the annual public dose at sites such as Argonne Na- tional Laboratory, Brookhaven Na- tional Laboratory, Fermi National Accelerator Laboratory, Los Alamos National Laboratory, and Oak Ridge National Laboratory. At these sites, emissions of air-activation prod- ucts such as 11 C have the potential to cause external-radiation dose to individuals near the site boundary. This dose is calculated by CAP‐88 and used to demonstrate compli- ance with U.S. EPA regulations. In some cases, this dose can be underestimated, especially for ele- vated plumes during stable atmo- spheric conditions and for receptors within a kilometer downwind of a stack (McNaughton et al. 2017). For example, if the entire plume is overhead and the ground-level air concentration is essentially zero, CAP‐88 will report zero dose be- cause it does not calculate the ex- ternal “cloudshine” dose. The purpose of this paper is to compare the CAP‐88 calculations with those of the Monte Carlo N-Particle (MCNP) computer pro- gram (LANL 2004) to determine when CAP‐ 88 calculations of gamma dose from cloudshine are inaccu- rate, and to estimate the accuracy of the CAP‐88 results. In this paper, the calculations are for distances of 800 m and 3,000 m downwind of the stack and for plume heights from 0 to 50 m because these choices correspond to the major sources of 11 C at U.S. Department of En- ergy facilities, and also because there are direct measurements of cloudshine for these conditions (Bowen 1994). However, the methods described here can also be used for other distances and stack heights (McNaughton et al. 2017). In the following sections, the CAP‐88 and MCNP calculations will be described, followed by a comparison of the two methods. METHODS External doses were calculated using CAP‐88 and MCNP. Calcula- tions were performed for specific cases of geometry, wind direction, wind speed, and atmospheric sta- bility class (A through F). These cases were combined to calculate the annual dose, using a weighted sum with the statistical weighting determined according to the prob- ability of each case. Technical details of CAP‐88 and MCNP The following sections describe the technical details of CAP‐88 and MCNP that are needed for the calculations. The CAP‐88 calculations of concentration CAP‐88 uses a Gaussian-plume model (Slade 1968; NCRP 1984; Eisenbud and Gesell 1997; Cember and Johnson 2009; Till and Grogan *MS J978, Los Alamos National Laboratory, Los Alamos, NM 87545. The authors declare no conflicts of interest. Michael McNaughton is a Certified Health Physicist at Los Alamos National Laboratory. He studied mathematics at the University of Oxford and nuclear physics at the University of London and the University of California. Since then, he has worked for forty years in nuclear physics and health physics at Los Alamos National Laboratory. His email is mcnaught@lanl.gov . Operational Topic 414 www.health-physics.com April 2017 Copyright © 2017 Health Physics Society. Unauthorized reproduction of this article is prohibited.