International challenge to predict the impact of radioxenon releases from medical isotope production on a comprehensive nuclear test ban treaty sampling station

https://doi.org/10.1016/j.jenvrad.2016.03.001Get rights and content

Highlights

  • 133Xe releases from a medical isotope production facility impacts IMS noble gas samplers.

  • International study to predict airborne radioxenon concentrations at an IMS noble gas sampler.

  • Model ensembles predicted concentrations better than individual models.

Abstract

The International Monitoring System (IMS) is part of the verification regime for the Comprehensive Nuclear-Test-Ban-Treaty Organization (CTBTO). At entry-into-force, half of the 80 radionuclide stations will be able to measure concentrations of several radioactive xenon isotopes produced in nuclear explosions, and then the full network may be populated with xenon monitoring afterward. An understanding of natural and man-made radionuclide backgrounds can be used in accordance with the provisions of the treaty (such as event screening criteria in Annex 2 to the Protocol of the Treaty) for the effective implementation of the verification regime.

Fission-based production of 99Mo for medical purposes also generates nuisance radioxenon isotopes that are usually vented to the atmosphere. One of the ways to account for the effect emissions from medical isotope production has on radionuclide samples from the IMS is to use stack monitoring data, if they are available, and atmospheric transport modeling. Recently, individuals from seven nations participated in a challenge exercise that used atmospheric transport modeling to predict the time-history of 133Xe concentration measurements at the IMS radionuclide station in Germany using stack monitoring data from a medical isotope production facility in Belgium. Participants received only stack monitoring data and used the atmospheric transport model and meteorological data of their choice.

Some of the models predicted the highest measured concentrations quite well. A model comparison rank and ensemble analysis suggests that combining multiple models may provide more accurate predicted concentrations than any single model. None of the submissions based only on the stack monitoring data predicted the small measured concentrations very well. Modeling of sources by other nuclear facilities with smaller releases than medical isotope production facilities may be important in understanding how to discriminate those releases from releases from a nuclear explosion.

Introduction

The International Monitoring System (IMS) is part of the verification regime for the Comprehensive Nuclear-Test-Ban-Treaty Organization (CTBTO, 2014). The verification regime is designed to detect nuclear explosions no matter where they occur on the earth. When complete, 80 of the IMS stations will have aerosol measurement systems sensitive enough to detect releases from nuclear explosions at great distances. At entry-into-force, half of the 80 stations will also have equipment that measures concentrations of four radioactive xenon isotopes (131mXe, 133Xe, 133mXe, and 135Xe) produced in a nuclear explosion, and following entry-into-force, a plan to add xenon monitoring capabilities to the other 40 stations will be reviewed (Comprehensive Nuclear-Test-Ban Treaty, 1996). An understanding of natural and man-made radionuclide backgrounds can also be used in accordance with the provisions of the treaty (such as event screening criteria in Annex 2 to the Protocol of the Treaty) for the effective implementation of the verification regime.

A number of studies of the release and transport of radioxenon from nuclear explosions, nuclear power plants, and medical isotope production facilities have been published (Becker et al., 2010, Eslinger et al., 2014, Hoffman et al., 2009, Kalinowski et al., 2008, Saey et al., 2010b, Wotawa et al., 2010, Wotawa et al., 2003, Zähringer et al., 2009). These studies confirm that fission-based production of 99Mo for medical purposes is the largest routine contributor of radioxenon to worldwide background levels. The 99Mo (half-life of 66 h) decays into 99mTc (half-life of 6 h) and the resulting 99mTc is used in approximately 30–40 million medical procedures per year (Peykov and Cameron, 2014) and the demand is expected to increase in the future.

A reduction in radioxenon releases to relatively low levels (Bowyer et al., 2013) has the potential to reduce background radioxenon to levels that don't significantly impact treaty verification activities. However, medical isotope production facilities meet regulatory release requirements and their releases don't pose public health risks, thus the operators have no financial incentive to reduce releases. Another way of mitigating the impact on treaty verification activities is to use stack monitoring data, if they are available, and atmospheric transport modeling. In the modeling context, one could attempt to model background sources accurately enough to subtract a background contribution from any sampled value. Given the uncertainties (source terms, modeling), simulated peaks may not accurately represent reality. Thus, alternately, when a xenon peak is observed, one could check whether the simulated background increases during the same period (synchronization in time). If that is the case, the observed peak could be linked to the rise of the radioxenon background.

Unfortunately, the details of the stack monitoring data needed, such as the time resolution, the accuracy, and whether or not local weather data are needed is not well known. There have been questions about whether stack data would be useful in a practical way at all, depending on the type of data made available and when it could be made available from a producer. To date, only one published study (Schöppner et al., 2013) has addressed the impacts the time resolution of stack monitoring data have on predicted concentrations at an IMS station location. The minimum source term resolution considered in that study was one day. Atmospheric modeling studies using inert tracers have been conducted since the early 1980s (Ferber et al., 1986, Gudiksen et al., 1984). This study addresses the difficult nuance of whether atmospheric models currently in wide use can yield information on the accuracy and timing of the source term data needed to faithfully reproduce sampling data.

This paper describes a challenge exercise formulated to start to answer some of these questions. Namely, to ascertain the level of agreement that can be achieved between atmospheric transport models using stack monitoring data and xenon isotopic concentration measurements at IMS stations. An evaluation criterion is used to measure the level of agreement. However, the real value of the exercise is in discussions resulting from the challenge without over-analyzing the evaluation criterion. The challenge is expected to spark discussions on what techniques are best, what gaps exist in our knowledge, and what type of data fidelity is needed from stack monitors. In general, this challenge will help inform the international treaty verification community of the status of the current capability.

The general approach of the exercise was to challenge atmospheric transport modeling groups to reproduce the time-history of 133Xe measurements at an IMS station using stack monitoring data from a medical isotope production facility. Participants received stack monitoring data that included the location, UTC date and time of releases, the measured activity concentrations of 133Xe in Bq m−3, an average stack flow rate (80,000 m3 h−1), and the height (m above ground level) of the release. All other data were gathered by the participants. Each participant used the atmospheric transport model and the associated meteorological data of their choice. The individuals participating in the challenge are identified in Table 1. Participants were asked not to use the IMS sampling data, if they had access to them, until after completing the modeling exercise.

Section snippets

Atmospheric transport models and meteorological data

The participants used several transport codes and several different sources for meteorological data. Several participants submitted results for more than one model. Some of the submissions were averages of other models or low and high resolution runs for the same model. Model metadata are provided in Table 2. Although the analysis considers all twenty six submissions, a subset of the submissions was selected to discuss common model characteristics. The reduced set of submissions is identified

Comparison measures

The purpose of this challenge was to ascertain the level of agreement one can achieve between simulated concentrations and IMS measurements using only the stack data and an atmospheric transport model, as might be expected for situations in which there was a detection of radioxenon at an IMS station and very little other information. Concentration estimates from this modeling exercise are expected to be quite variable (Draxler et al., 2015), thus it is useful to explore the general

Release and detection data

Participants in the modeling challenge received 133Xe stack emission data from the Institut des Radioéléments (IRE) radiopharmaceutical plant in Fleurus, Belgium. Releases from IRE have a measurable influence on 133Xe concentrations collected at DEX33 (Saey et al., 2010a) which is located 376 km from the IRE stack. The emission data covered the period 10 Nov 2013 through 8 Dec 2013. The measured concentration values for the stack data are based only on the 81 keV decay energy level and have an

Model comparison results

Thirteen participants submitted 26 solutions containing modeled concentrations of 133Xe at the sampler (DEX33) in Germany on the time periods used by the sampler. A plot of modeled concentrations for all 26 submissions and the concentrations at the sampler (black dots connected by a dotted line) is provided in Fig. 3. One submission had two predicted concentration values larger than 100 mBq m−3, but the upper limit on this plot partially obscures that fact. Some of the values were zero, thus

Discussion

The ranking and ensemble analysis in this paper suggests that combining multiple models may provide more accurate predicted concentrations than almost any single model. One ensemble selection technique was used in this paper. Further research is needed to identify optimal methods for selecting ensemble members, and those methods may depend on the nature of the transport problem. Although this exercise only addressed release and transport of a nondepositing noble gas, other radionuclides of

Acknowledgments

Participants in the atmospheric transport modeling challenge received 133Xe emission data from the Institut des Radioéléments (IRE) radiopharmaceutical plant in Fleurus, Belgium. IRE granted permission to use the data for the challenge.

The German national authority Bundesamt für Strahlenschutz (BfS) granted permission to use the 133Xe concentration data collected at the IMS noble gas sampler (DEX33) in Schauinsland, Germany for the challenge. Clemens Schlosser and Verena Heidmann of BfS

References (51)

  • P.R.J. Saey et al.

    The influence on the radioxenon background during the temporary suspension of operations of three major medical isotope production facilities in the Northern Hemisphere and during the start-up of another facility in the Southern Hemisphere

    J. Environ. Radioact.

    (2010)
  • M. Schöppner et al.

    Atmospheric transport modelling of time resolved 133Xe emissions from the isotope production facility ANSTO, Australia

    J. Environ. Radioact.

    (2013)
  • E. Solazzo et al.

    The Fukushima-137Cs deposition case study: properties of the multi-model ensemble

    J. Environ. Radioact.

    (2015)
  • A. Stohl et al.

    Validation of the lagrangian particle dispersion model FLEXPART against large-scale tracer experiment data

    Atmos. Environ.

    (1998)
  • G. Wotawa et al.

    Atmospheric transport modelling in support of CTBT verification—overview and basic concepts

    Atmos. Environ.

    (2003)
  • A. Becker et al.

    Backtracking of noble gas measurements taken in the aftermath of the announced october 2006 event in north korea by means of PTS methods in nuclear source estimation and reconstruction

    Pure Appl. Geophys.

    (2010)
  • M. Buehner et al.

    Implementation of deterministic weather forecasting systems based on ensemble–variational data assimilation at environment Canada. Part I: the global system

    Mon. Weather Rev.

    (2015)
  • M. Buehner et al.

    Four-dimensional ensemble-variational data assimilation for global deterministic weather prediction

    Nonlin. Process. Geophys.

    (2013)
  • J.C. Chang et al.

    Air quality model performance evaluation

    Meteorol. Atmos. Phys.

    (2004)
  • M. Charron et al.

    The stratospheric extension of the canadian global deterministic medium-range weather forecasting system and its impact on tropospheric forecasts

    Mon. Weather Rev.

    (2012)
  • Comprehensive Nuclear-Test-Ban Treaty

    Text of the Comprehensive Nuclear-Test-Ban Treaty

    (1996)
  • J. Côté et al.

    The operational CMC–MRB Global Environmental Multiscale (GEM) model. Part II: results

    Mon. Weather Rev.

    (1998)
  • CTBTO

    Verification Regime

    (2014)
  • R. D'Amours et al.

    The canadian meteorological centre's atmospheric transport and dispersion modelling suite

    Atmos. Ocean

    (2015)
  • R. D'Amours et al.

    Application of the atmospheric lagrangian particle dispersion model MLDP0 to the 2008 eruptions of Okmok and Kasatochi volcanoes

    J. Geophys. Res. Atmos.

    (2010)
  • Cited by (36)

    • Impact of industrial nuclear emissions on nuclear explosion monitoring

      2023, Journal of Environmental Radioactivity
      Citation Excerpt :

      The future also holds promise for overcoming the background histogram in Fig. 2. Several approaches can be used to confidently raise the interest level of detections below the threshold: prediction of industrial xenon observations using facility stack monitors and atmospheric transport models (Eslinger et al., 2016), isotopic discrimination when two or more isotopes are observed (Kalinowski et al., 2014), higher time resolution of observations (Haas et al., 2017), and reductions of nuisance signals at the source (Bowyer et al., 2013; Sameh, 2013). Of course, the arrival of detectable xenon below the 95th percentile, if timed to match detection via seismic, hydroacoustic, or infrasound technology, would still be of elevated interest.

    View all citing articles on Scopus
    View full text