International Network of Engineers and Scientists Against Proliferation

Atmospheric Krypton-85 Transport Modeling for Verification Purposes

Martin B. Kalinowski, Johann Feichter, Ole Roß

The Additional Protocol provides the IAEA with the power to apply environmental sampling. So far, swipe samples are routinely applied inside inspected facilities. But sampling outside of nuclear facilities has not been implemented yet. Several approaches have been proposed and a few have been tested but there is no sufficient scientific evidence about the effectiveness and cost-efficiency of detecting proliferation indicators off-site.

Atmospheric radionuclide monitoring is being implemented to verify the Comprehensive Test Ban Treaty (CTBT) – the first time that it is officially used in the framework of an arms control treaty. Previously, it had been extensively applied clandestinely by governmental agencies and laboratories as well as openly by independent scientists.

Noble gas monitors, sensitive to particular isotopes, such as krypton-85 (⁸⁵Kr), for example, in combination with chemistry transport model simulations, could detect on-going reprocessing activities. Model simulation of the spatial and temporal distribution of ⁸⁵Kr is the only tool that can separate the impact of different emitters at a receptor site and assess biases in emission strength and/or source location from reported krypton releases. The goal of a newly established task for the IAEA is to study this in more detail.

Radionuclides and Atmospheric Transport Modeling

Growing interest in the chemical composition of the atmosphere and the climatic impact of atmospheric tracer species has led to the development of an increasing number of three-dimensional and global atmospheric transport models. The use of such models for verification of international regulations requires careful evaluation. Assessing the accuracy of chemistry transport simulations is a complex task, because the concentrations of tracer species depend on many different processes, which can be tightly bound up with one another. The importance of evaluation is emphasized by a number of model intercomparisons (e.g. IPCC 2007, COSAM, AeroCom, TransCom) to evaluate a large number of CTMs (chemistry transport models driven by observed meteorology) and GCMs (general circulation models which generate their own meteorology) that have been developed to serve different purposes.

Cosmogenic and terrigenic natural radiotracers and radionuclides from nuclear bomb tests have been widely used to evaluate a large variety of relevant processes. Species used as test tracers should ideally meet the following conditions: they should be chemically inert, sources and sinks should be well known, and sufficient observational data should be available for comparison of model results.

Using radioactive tracers in order to understand the atmospheric dynamics and evaluate transport models has a long history. Christian Junge’s pioneering book Air Chemistry and Radioactivity[1] already paid attention to emissions and distributions of radionuclides, and the most voluminous chapter in Elmar Reiter’s review Atmospheric Transport Processes, published in 1978, is Part IV: Radioactive Tracers.[2]

Transport Processes

Transport simulated in numerical models takes place both through resolved large-scale advection and by parameterized sub-grid scale convective and boundary layer turbulent mixing. Small-scale deep convection is primarily induced by strong surface heating and has horizontal scales ranging from a few to tens of kilometers. Convection is strongest in summertime. The importance of these phenomena is that they transport compounds very rapidly from the vicinity of the surface into the free atmosphere, thus facilitating transport to distant locations.

Parameterization of boundary layer meteorology and quantification of pollutant exchange with the free troposphere (generally better in the well-mixed diurnal boundary layer and more problematic in stable situations) is often associated with large uncertainties or biases, particularly in orographically complex terrain. In addition, uncertainties may arise from the meteorology constraining the transport model. Biases in tracer distribution calculations may thus exhibit regional and seasonal variability depending on which processes dominate the transport.

Use of Radiotracers to Test Specific Processes

The usefulness of anthropogenic and natural radionuclides for model evaluation purposes was recently highlighted by an expert meeting on radionuclides organized by the World Meteorological Organization.[3] In order to test the boundary layer transport and the exchange between the boundary layer and the troposphere of a model, the natural terrigen radiotracer ²²²Rn was applied. ⁸⁵Kr was used to evaluate the inter- and intrahemispheric transport times. This long-lived radionuclide is produced by reprocessing. Sources are mostly located in the Northern hemisphere’s mid-latitudes, making them a good agent for inter-hemispheric transport of pollutants. The very pronounced latitudinal profiles that have been measured for ⁸⁵Kr allow an estimation of the interhemispheric exchange time, which is approximately one year.

The cosmogenic ¹⁴CO is a good tracer of stratosphere-troposphere exchange. It can also be used to assess the tropospheric OH abundances. The main source for ¹⁴CO is cosmic radiation and the only sink is by oxidation with the hydroxyl radical OH.

To address the downward transport from the stratosphere to the troposphere, ¹⁴CO, ¹⁴CO₂, ⁷Be, and ¹⁰Be are well suited tracers. The upward transport from the troposphere to the stratosphere occurs mostly in tropical regions where convective systems inject lower tropospheric air into the high troposphere/low stratosphere region.

Radionuclides that condense on ambient particle surfaces allow verification of aerosol physics, e.g. removal by precipitation and dry deposition on the ground. Radiotracers like ²¹⁰Pb, ⁷Be, ¹⁰Be, and ⁹⁰Sr have been used for this purpose. However, research about radionuclides was not only driven by curiosity about atmospheric processes but also by the need to understand the effects of nuclear bomb fallout and to estimate plutonium production.

Global Chemistry Transport Models for Verification Purposes

The observation of episodic largescale transport events and of a continuing rise in the background levels of air pollution has led to the recognition that air pollution cannot only be controlled by local measures but constitutes a problem of global dimension. Long range transport of air pollution is of crucial importance in governing air pollution levels which are underpinned by the Convention on Long- Range Transboundary Air Pollution (CLRTAP). A complicating factor in keeping to emissions’ ceilings or concentration limits imposed by different environmental directives and national and international regulations is the influence of meteorological conditions on air pollutant levels. Numerical models of the atmospheric chemical composition are used to assess sourcereceptor relationships.

Such models provide the only way to integrate measurements from different platforms with varying spatial and temporal dimensions into a consistent analysis of the state of the atmosphere and to assess the source distribution and strength based on atmospheric concentration measurements. Already some decades ago, chemistry transport model simulations have been used to observe atmospheric concentrations of the noble gas ⁸⁵Kr in order to detect ongoing reprocessing activities.

⁸⁵Kr is a radioactive noble gas with a half-life of 10.76 years. The major sources for atmospheric ⁸⁵Kr are reprocessing facilities, which are mainly located in the northern hemisphere. The relatively well-known distribution of the sources and the fact that the only significant sink in the atmosphere is natural decay makes this tracer ideal for assessing the characteristics of large-scale horizontal and interhemispheric transport as depicted in numerical atmospheric circulation models. Several research groups a three-dimensional tropospheric model to simulate the global ⁸⁵Kr distribution.[4]

Although the solubility of ⁸⁵Kr (as of all other inert gases) in water is very low (the ratio of ⁸⁵Kr to water is 1.85x10^-10 g/g at equilibrium) and the gas exchange rates of atmosphere-ocean and surface and deep water indicate that the ocean dissolves no more than 0.1 percent of the annual input of ⁸⁵Kr,[5] ⁸⁵Kr is also a very useful tracer for relatively short-term (decadal) atmosphereocean exchanges and mixing, for the dating of ground water, and to determine the age distribution of water.[6]

The ⁸⁵Kr distribution in the atmosphere can also be used as an indicator for clandestine separation of plutonium, which can then be used to build nuclear weapons. These studies are based on the atmospheric ⁸⁵Kr input function. The first comprehensive emission inventory of ⁸⁵Kr was published in 1986 and expanded on in 1988.[7] A more recent inventory also includes the period from 1986 to 2000 as well as additional sources in the former Soviet Union.[8]

However, all emission inventories lack an appropriate temporal resolution and show only average emissions per year. In reality, the release of ⁸⁵Kr is not continuous at all. The release rates of the few plants that report emissions with better temporal resolution such as La Hague (see Figure 1, monthly emissions), Karlsruhe (weekly and daily emissions),[9] and Tokai Mura (daily emissions)[10] show substantial short-term and seasonal variations due to sporadic reprocessing campaigns. It has also been shown that the variability of the concentrations calculated with emissions that vary from month to month is higher than the variability calculated with constant emissions over one year, and correlates better with actual observations.[11] In addition, the background concentrations calculated on the basis of monthly emissions have nearly always been lower than background concentrations calculated with the yearly emissions, whereas the variability from the monthly emissions is higher.

Another calculation of the global ⁸⁵Kr distribution has been done recently with the MOZART model.[12]

Model Strategies to Assess Detectability of Undeclared Plutonium Production

With the aid of transport models, at each observation site the temporal evolution of the ⁸⁵Kr concentration background, with its contributions from different emitters (mainly the known reprocessing facilities) and the variability of the concentrations, can be calculated. Distinguishing between contributions from different sources facilitates assessment of potential sources whose emissions do not agree with the officially reported source strength. High variability can be considered as an indicator for the influence of sources close to the receptor site. Transport models may also be useful tools to determine the minimum amount of ⁸⁵Kr released which can be detected, given the accuracy of the respective measurement technique. Transport models can and have been used to optimize the allocation of measurement sites.

The Model Additional Safeguards Protocol was introduced by the International Atomic Energy Agency (IAEA) in 1997 and expanded safeguards authorities and activities as a result of the shock caused by Iraq’s clandestine nuclear program, which was uncovered in the early 1990s. However, hardly anything has so far been achieved in providing the IAEA with the technical means to detect clandestine activities from a distance. Satellite imagery is used mainly for the investigation of known facilities and is not capable of providing proof of clandestine plutonium production; environmental sampling is restricted to swipe samples taken at the locations that are anyway routinely visited by inspectors.

In 1997 and 1998, the IAEA convened a technical committee to study the technical possibilities of Wide Area Environmental Sampling (WAES) under the Additional Protocol. Although the printed report[14] remained classified, it became know through publications by committee members[15] that

• the applied simulation methods were only based on Gaussian plume models,
• the requirements for WAES defined by the study were unnecessarily high, and
• the new ultra-sensitive trace analysis technologies based on a atom trap (ATTA) were not known at that time.

Accordingly, the committee concluded that WAES was not feasible. Later activities of the IAEA in exploring novel technologies and verification approaches to detecting clandestine activities brought no progress, either. As a result, further development and implementation of WAES was postponed indefinitely because of the foreseen technical challenges and the need for approval by the IAEA Board of Governors.[16]

However, the IAEA opened the doors for remote environmental sampling when it hosted a Technical Meeting on Noble Gas sampling and monitoring in 2005. Two years ago, the IAEA also established a Novel Technologies Program. There is a new task established in the framework of the German support program for the IAEA.[17] The goal of task C38 is to assess the feasibility of detecting clandestine reprocessing through ⁸⁵Kr sampling. The goal of this project is to provide the IAEA with all the information and technology that is required in order to implement the most effective krypton-85 tracer approach and close the safeguards gap regarding the detectability of clandestine plutonium production.

In the context of the independent Group of Scientific experts (iGSE), this project explores sampling procedures that are more cost-efficient than the classical WAES approach. The following improvements are possible, in particular if mobile sampling devices are employed:[18]

• Shorter sampling periods could reduce the detection thresholds by one order of magnitude.
• Mobile air samplers could be used thus eliminating the need for fixed monitoring sites.
• Mobility would allow the inspection agency to undertake surprise measurements on very short notice.

Atmospheric transport simulations will be used to determine optimum procedures for location-specific sampling and possibly for limited area environmental air sampling approaches in order to achieve a reasonable high detection probability for clandestine reprocessing activities. With methods of inverse modeling and backtracking it should be possible to distinguish a probable source region. The optimization goal is to achieve maximum detection probability with optimum source location precision.

The atmospheric transport simulations will also be used to plan a European demonstration of krypton-85 sampling as a tool to “detect” plutonium production at Sellafield and La Hague. Both the French and the British reprocessing plants publish their reprocessing campaign data. These data can be use for verification of the real sampling results. The existing mini-network of European sampling sites that is operated by the German Federal Office for Radiation Protection (BfS) will be supplemented by strategically located sampling sites. One new site will be set up in Hamburg. During the demonstration exercise, the sampling periods will be reduced from weekly to daily for some weeks. The higher time resolution is a precondition for applying methods to determine the source location.

Conclusions

On the one hand, general circulation models are applicable for the study of spatial distribution of ⁸⁵Kr, and on the other hand, the results (together with other tracer species) allow testing of the transport properties of the atmospheric models.

Historically, the evaluation of atmospheric ⁸⁵Kr had been used to estimate the plutonium production of the Soviet Union. Nowadays, the traceability of single point sources of ⁸⁵Kr through atmospheric transport modeling is going to be investigated as a future instrument for the IAEA to detect and locate unreported plutonium separation.

---

1.   C.E. Junge, Air Chemistry and Radioactivity, Academic Press, 1963.
2.   E. Reiter, Atmospheric Transport Processes. Part 4: Radioactive Tracers, Department of Energy, Technical Information Center, 1978.
3.   World Meteorological Organization, Global Atmosphere Watch (2004): 1st International Atmospheric Expert Meeting on Sources and Measurements of Natural Radionuclides Applied to Climate and Air Quality Studies, Gif sur Yvette, France, 3 -5 June 2003.
4.   See e.g. D.J. Jacob, M.J. Prather, S.C. Wofsy, M.B. McElroy, Atmospheric distribution of ⁸⁵Kr simulated with a general circulation model, Journal of Geophysical Research, Volume 92, Issue D6, 1987, pp. 6614-6626. K. Winger, , J. Feichter, M.B. Kalinowski, H. Sartorius, C. Schlosser, A new compilation of the atmospheric 85krypton inventories from 1945 to 2000 and its evaluation in a global transport model, Journal of Environmental Radioactivity, Vol. 80, 2005, pp. 183-215. P.H. Zimmermann, J. Feichter, H.-K. Rath, P.J. Crutzen, W. Weiss, A global threedimensional source-receptor model investigation using ⁸⁵Kr, Atmospheric Environment, Vol. 23, No. 1, 1989, pp.25–35.
5.   Y.A.Izrael, I.M. Nazarov, A.G.Ryaboshapko, Release of man-made krypton-85 to the atmosphere, Meteorologiya i Gidrologiya, No. 6, 1982, pp. 5–15.
6.   H.H. Loosli, Applications of ³⁷Ar, ³⁹Ar and ⁸⁵Kr in hydrology, oceanography and atmospheric studies. Current state of the art, Technical Report of the Results of a Consultants Meeting ‘Isotopes of Noble Gases as Tracers in Environmental Studies,’ IAEA, Vienna, Technical Report Series No. 332, 1992, pp. 73–85.
7.   F. von Hippel, D.H. Albright, B.G. Levi, Quantities of Fissile Materials in US and Soviet Nuclear Weapons Arsenals, Center for Energy and Environmental Studies, The Engineering Quadrangle, Princeton University, PU/CEES Report No. 168, 1986. K.-H. Rath, Simulation der globalen ⁸⁵Kr- und ¹⁴CO₂-Verteilung mit Hilfe eines zeitabhängigen, zweidimensionalen Modells der Atmosphäre, Inaugural-Dissertation, Ruprecht-Karls Universität, Heidelberg, 1988.
8.   Winger et.al, 2005, op.cit.
9.   M.B. Kalinowski, H. Sartorius, S. Uhl, and W. Weiss, Conclusions on Plutonium Separation from Atmospheric Krypton-85 Measured at Various Distances from the Karlsruhe Reprocessing Plant, Journal of Environmental Radioactivity,Vol, 73, Issue 2, 2004, pp. 203-222.
10.   Y. Igarashi, H. Sartorius, T. Miyao, W. Weiss, K. Fushimi, M. Aoyama, K. Hirose, H.Y. Inoue, ⁸⁵Kr and 133Xe monitoring at MRI, Tsukuba and its importance, Journal of Environmental Radioactivity, Vol 48, Issue 2, 2000, pp. 191–202.
11.   Winter et.al., 2005, op.cit.
12.   D. Youn, D,.H. Wuebbels, and M.B. Kalinowski, Global Modeling of Atmospheric Krypton-85 Concentrations, in this issue of the INESAP Information Bulletin.
13.   Model Protocol Additional to the Agreement(s) Between State(s) and the International Atomic Energy Agency for the Application of Safeguards, IAEA INFCIRC/540 (corrected), Setp. 1997; www.iaea.org/Publications/Documents/Infcircs/1998/infcirc540corrected.pdf.
14.   Member State Support Programs to the IAEA, Use of Wide Area Environmental Sampling in the Detection of Undeclared Nuclear Activities, STR-321, August 1999.
15.   For example P.W. Krey, K.W. Nicholson, Atmospheric sampling and analysis for the detection of nuclear proliferation, Journal of Radioanalytical and Nuclear Chemistry, Vol. 248, No. 3, 2001, pp. 605-620.
16.   Overcoming the technical challenges was viewed as too remote to be achieved soon. In addition, it was assumed that political resistance would be encountered in the IAEA Board of Governors.
17.   See www.fz-juelich.de/ste/joint_programme_tasks/.
18.   See M.B. Kalinowski, H. Daerr, and M. Kohler, Measurements of Krypton-85 to Detect Clandestine Plutonium Production in this issue of the INEAP Information Bulletin.