The tropical South Pacific Ocean continues to be exposed to the risk of radioactive contamination in the marine environment. Underground nuclear weapons testing in French Polynesia was conducted from 1975 until 1996 at Moruroa Atoll and the closely neighbouring Fangataufa Atoll. The contaminated material resulting from these tests contained within the geological structure of these atolls is a potential source of pollution ( Ribbe and Tomczak; 1990, hereafter RT90).
It is appropriate to determine the extent and magnitude of such potential pollution using the best available methods. To this end, an off-line tracer model was developed which utilised the velocity fields of a high-resolution global ocean model. Previous studies have been underaken of the advection and dispersion of radioactive material from Moruroa Atoll (eg., RT90, Tomczak and Herzfeld; 1998, Mittelstaedt et. al.; 1999, and Lazar and Rancher; 1999; hereafter LR99). However, these have been limited by a number of simplifications. All of the above studies were restricted by a limited ocean domain requiring unrealistic boundary conditions. RT90's study in particular suffered from very low model resolution inadequate for simulating realistic ocean circulation. The study by Tomczak and Herzfeld (1998) was the only study to address seasonal variation, and this was on a limited regional scale. All of the previous studies neglected the possible role of natural interannual variability. In this study we plan to address all three of these limitations.
The aims of this study are: (1) To develop a global off-line tracer model of eddy-permitting resolution, and (2) to predict the potential fate of radioactive material released from Moruroa Atoll, improving on previous studies by using a higher resolution, a global domain, and incorporating seasonal and interannual ocean variability.
To avoid excessive complexity and preserve to the generality of the results of this study, two important assumptions are made. As in all previous studies (except RT90) the decay of radioactive isotopes has not been incorporated into the tracer model. The high number of radionuclides present (32 species were identified by one study; IAEA, 1998) and their relative abundance creates enormous complexity in the incorporation of decay rates and decay products into the study results. Radionuclides are therefore considered as a passive tracer (physically constant in time). This assumption allows the model to be used for any radionuclide with a half-life greater than around one year (over shorter time-scales non-linear mixing effects could be important). The second important assumption is that the background concentration of radionuclides in the South Pacific Ocean is set to zero. Results therefore describe the concentration of the extra radionuclides originating from the release site.
The releases from Moruroa Atoll are simulated at different depths in the water column and at different release rates (instantaneous and gradual). This study also incorporates different climate scenarios to gain some indication of the effects of seasonal and interannual variability on tracer advection. All of the simulations are carried out over ten years. Finally, a comparison of the results of this study is made with two earlier simulations that used coarse resolution (RT90) and limited domain models (RT90 and LR99), as well as constant background flow fields.
Throughout this paper the spelling of the name of the atoll at which the majority of French nuclear tests was conducted is MORUROA, as opposed to MURUROA. The former being more closely associaed with its origins in the local Maohi languages, the later being attributable to a typographical error made by a French cartographer (Maclellan and Chesneaux, 1998).
http://www.maths.unsw.edu.au/~doughaze Page created by Douglas R. Hazell 7/6/2001 Last update of this page: 1/8/2001