Preliminary Benchmarking Efforts and MCNP Simulation Results for Homeland Security

It is shown that basic measurements made from well-defined source detector configurations can be readily converted into benchmark quality results by which Monte Carlo N-Particle (MCNP) input stacks can be validated. Specifically, a recent measurement made in support of national security at the Nevada Test Site is described with sufficient detail to be submitted to the American Nuclear Society's Joint Benchmark Committee for consideration as a radiation measurement benchmark. From this very basic measurement, MCNP input stacks are generated and validated both in predicted signal amplitude and spectral shape. Not modeled at this time are those perturbations from the more recent pulse-height-light tally feature, although what spectral deviations are seen can be partially attributed to not including this small correction. The value of this work is as a proof-of-concept demonstration that well-documented historical testing can be converted into formal radiation measurement benchmarks. This provides evidentiary support that validated virtual testing could eventually be carried out for various detection system technologies including algorithms, new detector designs, constructions, and arbitrary source and shielding assemblies.