DRIFT dispersion model predictions for the Jack Rabbit II model inter-comparison exercise

Autor: Harvey Tucker, S E Gant, Graham Tickle, Adrian Kelsey
Rok vydání: 2021
Předmět:
Zdroj: Atmospheric Environment. 244:117717
ISSN: 1352-2310
Popis: In 2015 and 2016, the US Army conducted nine large-scale chlorine releases at the Dugway Proving Ground in Utah, known as the Jack Rabbit II trials. The purpose of these experiments was to improve our understanding of pressure-liquefied chlorine releases and atmospheric dispersion, and to provide useful practical knowledge for emergency responders. Data from three of the Jack Rabbit II trials (Trials 1, 6 and 7) were subsequently selected for an international model inter-comparison exercise, which included modelling contributions from the US, UK, Canada, France, Germany, Sweden, Finland and the European Commission. This paper provides details of one of the UK contributions to that exercise, using the DRIFT integral dispersion model. Participants in the exercise were given a set of prescribed model input conditions. The methodology used in DRIFT to model these conditions is described here, which consisted of three runs: a baseline case and two sensitivity tests that examined the effect of variations in the modelled wind-speed profile and the dry deposition rate. A brief review is also provided of the experimental data used for the model inter-comparison exercise. Gas concentrations were measured in the Jack Rabbit II trials using an extensive array of gas sensors arranged on arcs downwind from the release point. Four types of sensors were used, which each had calibration and/or saturation limits. In some of the trials, sensors that recorded the highest concentrations on an arc were saturated with gas and under-recorded the actual concentrations. In other cases, especially near the source, there were only a few sensors on an arc and therefore it is uncertain whether the plume maximum concentrations were recorded. Results from DRIFT are compared to the Jack Rabbit II data in two ways. Firstly, against the full dataset of maximum arc-wise concentrations, and secondly against the subset of measurements that were unaffected by saturation issues or there being a sparse sensor array. In the first case, DRIFT gives higher concentrations than were measured on average, noting that some measurements probably under-recorded the concentration owing to the previously mentioned sensor saturation issues. In the second case, using the more reliable dataset, the statistical performance of DRIFT falls within commonly-used dense-gas model acceptance criteria. Sensitivity tests on the wind speed profile and dry deposition rate had an effect on the DRIFT predictions in Trial 1, but little impact in Trials 6 and 7. This behaviour is likely to be due to the changing meteorology during Trial 1, and the lower wind speed and smaller chlorine mass released in that trial. Remaining knowledge gaps and possible future directions for research are discussed. These include further work on dry deposition, modelling of the other Jack Rabbit II trials (potentially using source models as well as prescribed common source conditions), and comparisons of predicted and measured toxic load, calculated from the time-varying concentrations.
Databáze: OpenAIRE
Externí odkaz: