A literature review has been conducted to evaluate the range and distribution of atmospheric dispersion model-specific input parameter uncertainties. This review does not consider uncertainties associated with source term and meteorological inputs to an atmospheric dispersion model. Uncertainty ranges relating to the standard deviation of the cross wind (σy) and vertical (σz) Gaussian plume profile, the horizontal diffusivity term, Ku, the vertical diffusivity term, Kw, dry deposition velocity, surface resistance approach to dry deposition modelling, wet deposition scavenging coefficient parameter, a, wet deposition scavenging exponent parameter, b, and roughness length have all been recommended. Where possible accompanying probability distributions were also recommended. Where no alternative recommended approach can be implemented when describing the uncertainty range on the turbulent diffusion scheme for a Gaussian plume model, an alternate “second-choice” stability category, determined by expert judgement, is recommended. And it was identified that there is value in the application of model ensemble approaches for describing the uncertainty associated with the standard deviation of the cross wind and vertical Gaussian plume profile, and the below-cloud and in-cloud scavenging coefficient schemes.
These recommendations aim to facilitate the propagation of uncertainties through the chain of atmospheric dispersion and radiological assessment models both for historical scenarios (such as the accident at the Fukushima Daiichi Nuclear Power Plant) and for hypothetical scenarios. This will help improve understanding of the effect of input uncertainties on model output, and thereby appraise the impact on decision making in the context of an emergency response.