Evaluating emergency response models of radiological dispersion in complex terrain

Abstract

Operational airborne releases of trace quantities of the radioactive noble gas Ar-41 from the HIFAR Nuclear Research Reactor located in Sydney, Australia are valuable for evaluating emergency response models incorporating radiological dispersion. The Australian Nuclear Science and Technology Organisation (ANSTO), where the reactor is located, has a network of meteorological stations and GR-150 environmental gamma dose detectors placed in complex terrain within a 5km radius of the site. The current version of ANSTO's Emergency Response System feeds real-time meteorological data from this network into the wind field model NUATMOS, which is then subsequently used, together with known source emissions of Ar-41 from HIFAR, to drive the Lagrangian mesoscale atmospheric dispersion puff model RIMPUFF. Output from RIMPUFF is compared with data from the gamma dose monitoring network in order to assess the overall performance of the system. An updated model combination LSMC/RIMPUFF has recently been evaluated using Ar-41 peak observations from the detector network during 2002-03 under a range of atmospheric stability conditions. Sensitivity tests of the new model version were performed by perturbing the gridded meteorological data within the model domain, which incorporates complex terrain. Results from statistical analyses of the model output are reported, assessing the suitability of LSMC/RIMPUFF to replace the current models in the Emergency Response System. The LSMC/RIMPUFF evaluation results are compared with two previous evaluations carried out in 2004, of LSMC's predecessor LINCOM/RIMPUFF and the NUATMOS/RIMPUFF combination currently used at ANSTO. Results from these tests have been favourable for LSMC/RIMPUFF, which is consequently now likely to replace the older models within ANSTO's Emergency Response System. The decision support system ARGOS, which also incorporates the LSMC/RIMPUFF model, is currently being evaluated for potential use by a number of Organisations within Australia including ANSTO. Some results of this initial evaluation will also be included here.

Model evaluation of RIMPUFF within complex terrain using an <SUP align="right">41</SUP>Ar radiological dataset

“…Observations of 41 Ar by the gamma detector network from HIFAR's routine releases during 2002-03 have previously been used to evaluate the dispersion model RIMPUFF (RIsø Mesoscale PUFF) (Mikkelsen et al, 1984) with 2 different diagnostic wind models (Williams et al, 2005). More recently, the observations were used to evaluate RIMPUFF with the Local Scale Model Chain (Dyer and Pascoe, 2008), which incorporates modern micrometeorological scaling approaches. As a result of tests against two cases of the 41 Ar dataset comparing observed and predicted dose rates, RIMPUFF has recently had one of its puff growth models modified.…”