Using a coupled dispersion model to estimate depletion of a tritium oxide plume by a forest

Highlights

• A model has been developed to incorporate the effects of forested topography on the atmospheric transport of tritium oxide.

• The forested surface produced conservative deposition velocity values for tritium oxide between 0.7 and 1.2 cm s⁻¹.

• After 10 km of transport, plume concentration was found to be reduced between 40 and 60%.

• This work provides a baseline for assessing safety basis considerations for tritium-related facilities.

Abstract

Tritium processing facilities may release tritium oxide (HTO) to the atmosphere which poses potential health risks to exposed co-located workers and to offsite individuals. Most radiological consequence analyses determine HTO dose by applying Gaussian plume models to simulate the transport of HTO. Within these models, deposition velocity is used to assess the sum of all deposition processes acting on the plume. While this may account for vegetative and soil uptake or respiration processes, it may currently lack inclusion of the complex interactions within heterogeneous forested environments. In this complex morphology, dispersion patterns are significantly altered by changing flow regimes above and below the forest canopy and by the transfer of plume material across the canopy boundary. To determine the effects of a heterogeneous forest canopy on an airborne HTO plume, a Gaussian plume model coupled with an advection-diffusion plume model was applied to estimate transport in the free atmosphere above the forest and within the forest canopy and understory. During 2012, wind speed and wind direction measurements taken at 5 heights, ranging from 2-m to 28-m, on an instrumented meteorological tower located in a loblolly pine forest at the Department of Energy (DOE) Savannah River Site (SRS), near Aiken, SC. From these measurements, model predictions were made over a full spectrum of meteorological conditions. Deposition and resuspension velocities were calculated based on the model-predicted flux of plume material across the top of the forest canopy. Additionally, net deposition velocity of the plume material was calculated as the difference between the deposition and resuspension velocities. The 1st and 5th percentile net deposition velocities were estimated to be 0.7 cm s⁻¹ and 1.2 cm s⁻¹, respectively.

Introduction

Releases of tritium to the atmosphere from the operation of nuclear reactors, reprocessing plants, and tritium processing facilities may result in potential health risks to facility workers, co-located workers, and the public. Most atmospheric releases of tritium consist primarily of its elemental (HT) or oxide (HTO) forms (Kessler, 1983). HT is a low-energy beta emitter with inhalation as the primary dose pathway. However, HTO exposure is of much greater concern due to its molecular similarity to water which is readily exchanged in plants and organic tissue (Ojovan and Lee, 2005), and then rapidly distributed throughout the body, leading to cell damage as it undergoes radioactive decay. Accordingly, HTO has a significantly larger Dose Conversion Factor (DCF) than HT, posing a much greater risk to human health (EPA, 1988).

To reduce potential radiological consequences from HT and HTO releases, Material at Risk (MAR) limits may be placed on tritium processing facility inventories and/or production levels. These administrative controls are typically determined by radiological consequence assessments using atmospheric transport and diffusion models for unmitigated releases which predict potential downwind concentration and dose to the workers and the public. Several similar models are available to perform the assessment, all based on Gaussian dispersion methodology. MELCOR Accident Consequence Code System Version 2 (MACCS2) (US DOE, 2004) and Generation II Environmental Radiation Dosimetry Software System (GENII) (Napier, 2011) are two of the radiological consequence codes that are available through the DOE Central Registry of safety software (Defense Nuclear Facility Safety Board, 2011). The Gaussian dispersion methodology enables an analyst to take a large meteorological data set, spanning a large range of possible atmospheric dispersion conditions, and calculate cumulative frequency statistics of predicted time-averaged plume and potential consequences (Hanna et al., 1982). However, this methodology uses several simplified environmental characteristics (Miller and Hively, 1987) such as temporally- and spatially-uniform meteorological conditions, discrete plume diffusion modes based on typing atmospheric stability; applying roughness length and zero-plane displacement parameters to describe frictional drag imposed by the ground surface, and deposition velocity algorithms to describe an average rate of plume depletion by surface contact.

HTO has been shown to have a complex behavior with the environment. It behaves like water vapor in terms of its interactions with soil and vegetation, but is also subject to uptake and respiration processes. Lee et al. (2012) identified that for facility safety basis modeling, a 2 h residence time of HTO within vegetation or soil should be used, which represents a fairly rapid cycling of HTO in and out of the environment relative to the 24 h period typically used for dose assessment calculations. Its interactions with soil are further dependent on the soil moisture conditions (Garland, 1979). Galeriu and Melintescu (2015) identify that more detailed understanding of the transfer of HTO between the atmosphere, soil and vegetation is still needed and is an important component for accurately performing dose modeling for accident analyses.

The UFOTRI model (Raskob, 1999) has been the most widely used model specifically designed to assess the movement of tritium in the environment. A sensitivity analysis identified that interactions with the surface and vegetation contain the most (Galeriu et al., 1995). A potential shortcoming with UFOTRI and similar models is that, once deposition has occurred, the deposited material remains in that location unless the model accounts for potential resuspension of deposited material. In complex environments, it is possible that movement of the deposited material could occur on rapid timescales. For instance, within a forest canopy, there can be transport within the airspace of the canopy that is distinctly different from the surrounding environment above the canopy. A closer examination of the complex environment may yield interesting facets of transport that would not be captured by the traditional Gaussian models.

Gaussian models typically assess deposition using a single deposition velocity which is designed to account for the complex interactions at the surface using a single number to describe the net effects. This parameter is based on a range of measurements and is highly site- and environment-specific. For this reason, we have undertaken a detailed study of the forested environment at the Savannah River Site (SRS) to assess the range of deposition velocities in the forest and identify what would be considered a conservative deposition velocity for use in safety basis modeling.

A forested environment has a spatially-variant wind flow and turbulence regime; separate and distinct from the free atmosphere above it. Several earlier analyses of wind structure in forested environments have shown that wind direction changes within the forest canopy (Smith et al., 1972), and that the standard logarithmic wind profile is not maintained due to frictional effects of the forest (Garratt, 1980; Parlange and Brutsaert, 1989). While the zero-plane displacement parameter partially accounts for this change by shifting the logarithmic wind profile upward to a more representative height, it does not account for deposition, which is permanently removing material onto an idealized surface. Accordingly, the Gaussian model is much too constrained to accurately depict ongoing HTO transport and fate as it moves from the free atmosphere to the forest canopy atmosphere and then to the understory atmosphere, before recycling back to the free atmosphere.

The presence of a separate flow regime below the forest canopy transports HTO horizontally at a different rate within the forest compared to its transport rate above the forest. Ejection and sweep events (i.e., strong bursts of upward and downward vertical motion across the forest boundary) result in intense turbulent motions (Zhu et al., 2007; Guo et al., 2010), propelling HTO into the forest, or flushing it out of the forest and back into the free atmosphere at a rapid rate (Rannik et al., 2016). While the effects may be minimal through the depth of the plume, the effects on near-surface concentration has a direct impact on the determination of radiological consequences.

In addition, forest vegetation absorbs HTO from the atmosphere as part of natural photosynthetic and evapotranspiration processes (Canadian Nuclear Safety Commission, 2009). The removal of HTO from the atmosphere through the mixing and absorption processes can be inferred in the standard Gaussian models from a determination of a deposition velocity. However, most of the HTO that is taken up by vegetation in the canopy region is returned to the atmosphere through resuspension, with a half-life of less than 1 h (Brudenell et al., 1997; Boyer et al., 2009), leading to the conservative assumption of no deposition. Moreover, these processes will lead to some HTO redistribution well into the forest understory where it will be affected by additional mechanical shear and by changes in both wind direction and wind speed. Unfortunately, the extent to which the forest environment acts to further disperse an airborne plume and how this enhanced dispersion affects net deposition velocity and plume centerline has not yet been fully researched. While these processes are far too complex to capture in a Gaussian modeling architecture, it may be possible, with a more comprehensive model, to determine a representative deposition velocity for HTO that captures a physically realistic net effect on the plume.

With this objective in mind, the Savannah River National Laboratory (SRNL) performed a review of dispersion modeling methodology for HTO, mainly in response to findings issued by the Defense Nuclear Facility Safety Board (DNFSB, 2011). The DNFSB was concerned that the application of a specified deposition velocity (i.e., 0.5 cm s⁻¹) in design safety analysis modeling of HTO plumes at SRS was not supported by current scientific understanding and may not be sufficiently conservative to ensure that no adverse worker or public health risk existed in the event of a release of HTO. A subsequent study by Murphy et al. (2012) examined experimental data collected at SRS on the uptake processes and subsequent resuspension of HTO from surface vegetation. This study concluded that the time scale of uptake and resuspension was of sufficiently short duration that no net deposition would occur during the integration period (i.e., approximately 24 h) considered by safety-related radiological consequence assessments. As a result, the report recommended that modeling for design safety analysis should not credit removal of HTO by deposition in order to maintain a sufficiently conservative upper-bound for dose estimates at key downwind receptors. It should be emphasized that the DNFSB studies used the aforementioned, limited Gaussian modeling techniques, which made it impossible for it to examine the potential influence of the forest on fate and transport (Viner, 2012). Since the presence of extensive forests at SRS creates a micrometeorological environment where wind speed and wind direction will vary from above to below the forest canopy, the application of a Gaussian plume model is not sufficiently robust to capture these complex dispersion patterns. The current study seeks to quantify a deposition velocity that serves as a surrogate for the effective removal of HTO, with respect to a downwind receptor of concern, due to enhanced dispersion resulting from the complex interactions of the plume within the forest canopy and understory.

To demonstrate a more realistic value of deposition velocity, measurements from the Aiken AmeriFlux tower at the SRS were used as input to an atmospheric transport model, which can address two separate flow regimes above and within the forest. This coupled model was developed to quantify the movement of an airborne effluent from the free atmosphere as it moves in and out of the confined forest canopy and understory atmospheres. The SRNL model enables an estimation of the potential decrease in near-surface concentrations that result in comparison to using a simple Gaussian model. Accordingly, the magnitude of the predicted flux can be used to determine suitable deposition velocity magnitudes for use in simpler Gaussian models. In addition to improving the understanding of how forests influence dispersion, this model can also inform decisions regarding the determination of appropriate values of deposition velocity in highly complex environments.