By Professor E T Brown FREng FTSE
Background
France began atmospheric nuclear testing at Mururoa and Fangataufa atolls in the South Pacific in July 1966. Following international protest, atmospheric testing ceased in August 1970.
Underground testing began at the Centre d'Experimentations du Pacifique (CEP) with two small tests beneath the southern rim of Fangataufa atoll in 1975. By the time this program of tests was completed in July 1991, 123 underground nuclear tests had been conducted at Mururoa and eight at Fangataufa. France announced a moratorium on nuclear testing in September 1991.
Russia, the United Kingdom and the United States of America had already announced their readiness to adopt the proposed Comprehensive Test Ban Treaty when, in June 1995, President Chirac announced that France would conduct a limited number of further underground tests at the CEP before signing the Treaty. Between 5 September, 1995 and 27 January, 1996, four further tests were carried out under the lagoon at Mururoa and two at Fangataufa. France duly signed the Comprehensive Test Ban Treaty in 1996.
The decision to conduct this final campaign of testing three years after all other countries except China had stopped nuclear testing (regrettably, India and Pakistan have since carried out tests) provoked widespread international concern and anti-French protest, especially in Pacific countries, including Australia. The writer's home city of Brisbane was declared a nuclear free zone. Claims were made, in France and elsewhere, that the underground testing had resulted in lasting and catastrophic damage to the stability and hydrology of the atolls and would lead to dangerous releases of radionuclides into the environment.
In response to these concerns, the French Minister of Foreign Affairs, acting on behalf of President Chirac, invited two separate international groups to conduct fully independent studies of the consequences of the underground tests on Mururoa and Fangataufa. The International Atomic Energy Agency (IAEA) agreed to examine the issue of radiological releases. The report of this study (International Atomic Energy Agency 1998) has been released and was the subject of an international conference held in Vienna from 30 June to 3 July 1998.
In late 1995, the immediate Past President of the International Society for Rock Mechanics, Professor Charles Fairhurst of the USA, was invited to lead an International Geomechanical Commission (IGC) "to assess the short- and long-term effects of underground nuclear testing on the stability and hydrology of Mururoa and Fangataufa". (In this context, "short-term" refers to a period of approximately 500 years from the cessation of nuclear testing and "long-term" to a period in the order of 10,000 years.) Two Australians, Dr Lloyd Townley and the writer, were appointed members of the seven-person Commission. Because the hydrological consequences of the tests have a direct influence on the transport of radionuclides through the rock masses, three members of the IGC, including Dr Townley, served on the IAEA working group which studied geosphere transport.
IGC members visited Tahiti and the atolls in July 1996. The Commission held five full meetings during the course of its work. Additional meetings of two subgroups, one on stability and one on hydrology, were held, often with consultants and/or French experts present. With the aid of its consultants, the Commission sought to develop its own understanding of the mechanics and consequences of the underground nuclear tests. It carried out extensive numerical analyses of shock wave effects, seismic wave propagation, slope stability and pre- and post-test hydrology. However, in its studies, the IGC was constrained to use the data made available to it by the French authorities.
The Commission's report (International Geomechanical Commission 1998) has been submitted to the French Government. This article draws heavily on parts of that report.
Mururoa and Fangataufa Atolls
Mururoa and Fangataufa atolls form part of the Tuamoto archipelago in the southeastern sector of French Polynesia. They are approximately 1200 km from Tahiti. The atolls are usually seen as narrow coral fringes up to a few hundred metres wide, rising a mere 2-3 m above sea level and enclosing a lagoon. Mururoa (Figure 1) is elongated with a maximum dimension of some 28 km and covers an area of 155 km2. Fangataufa, which is 40 km to the south of Mururoa, is more equi-dimensional in shape and covers an area of 45 km2.
Figure 1: Underground Test Areas on Mururoa (DIRCEN/CEA 1998a)
The visible rim of each atoll is, in fact, part of a 300-500 m thickness of accumulated carbonates overlying a volcanic edifice which rises to some 4 km above the surrounding ocean floor. These edifices were formed 10-12 million years ago as the crustal plate passed over a "hot spot" source of molten magma. The magma broke through fractures in the crust and accumulated on the ocean floor creating a seamount or submarine volcano. As the edifice grew, the volcanic lava changed in character from a deep submarine flow to shallower sub-aerial deposits and eventually to aerial explosions of volcanic ash and lava flows over the emerged island (Guille et al 1996).
Volcanic activity ceased after 1-2 million years as the crustal plate moved away from the hot spot. As the volcanic seamount cooled it subsided and a fringing coral reef formed. With further subsidence, erosion and sea-level changes over the ensuing 8-10 million years, several hundred metres of carbonates (limestone and dolomite) accumulated above the volcanic seamount.
Figure 2 shows a schematic geological cross-section of Mururoa based on the logs of a series of boreholes drilled on the rim and in the lagoon. Fangataufa has generally similar geology. In both cases, the geological series consists of submarine and sub-serial volcanics, transitional formations reflecting the discontinuous nature of the build-up of the volcanic edifice, and carbonates consisting of limestones and dolomites. The carbonates contain a great diversity of facies, variable cementation, many depositional unconformities and widespread karstification (Buigues 1996).
Figure 2: Schematic geological cross-section of Mururoa (after Buigues 1996).
(The vertical to horizontal scale exaggeration is approximately 10:1)
Underground Nuclear Testing
The nuclear testing was carried out under the direction of the French Department of Defence through the Directorate of the Centre des Etudes Nucleaires (DIRCEN). Technical support was provided by the Commisariat l'Energie Atomique (CEA).
Because of the relative ease of drilling and servicing the test sites, the early underground nuclear tests were carried out under the atoll rims at Fangataufa in 1975 and at Mururoa from 1976 to 1980. Following a large submarine slope failure in the southwest sector of Mururoa in 1979 (area 4 in Figure 1), some tests were carried out under the rim and some under the lagoon of Mururoa from 1981 to 1986. From 1987 all tests were carried out under the lagoons of both atolls.
Except for a few small "safety tests" which did not generate radioactive products, all nuclear testing at the CEP was carried out at depths of 500-1100 m in the volcanics. The nuclear devices and the associated test equipment were placed in 1.5 m diameter vertical shafts drilled from the surface. The backfilling and effective sealing of the shafts to prevent leakage of radioactive products into the biosphere was a vitally important part of the test preparation process. Drilling such large holes to depths of 1100 m over 40-50 m of water in the lagoons was a major engineering undertaking. Details are provided by Bouchez and Lecomte (1996).
Figure 1 shows the distribution of the tests on Mururoa and their yields. In accordance with an international agreement, the maximum yield of any single test was restricted to 150 kt (i.e. a yield equivalent to that of 150 kilo-tonnes of TNT). The 137 underground tests (including the safety tests) on Mururoa summarised in Figure 1 gave a total yield of 2.4 Mt at an average of almost 20 kt. The 10 tests carried out on Fangataufa had an average yield of 80 kt. It is useful to note that the aerial explosions over Hiroshima and Nagasaki in 1945 averaged 20 kt.
The Mechanics and Effects of Underground Nuclear Explosions
The detonation of an underground nuclear explosion generates quasi-instantaneously an amount of energy, e, which is then converted into other forms of energy through a series of processes.
Firstly, a few microseconds after the explosion, the device and some surrounding rock and water are vaporised. A fraction of the initial energy released by the explosion is expended in this process. Secondly, a few tens of microseconds after the explosion, the cavity expands to a final radius of rc under the influence of the internal energy of the gas in the cavity which is at extremely high temperature and pressure. The gas expands doing work on the surrounding rock generating a shock wave. Part of the energy of the gas is also dissipated in melting some of the rock surrounding the cavity. The volume of the final cavity is proportional to the energy yield, e, so that the final cavity radius may be expressed as
rc = r'c e1/3 (1)
For the test conditions at the CEP, r'c ~ 10 -12 m/kt1/3 so that a 1 kt explosion produces a cavity of radius 10-12 m, depending on the depth of burial. A deep 150 kt explosion would produce a cavity with a radius of approximately 55 m.
With increasing distance from point zero, the shock wave becomes weaker and transforms initially into a plastic wave producing crushing and shear damage of the rock, and then into an elastic wave at a distance from point zero of
rd = r'de1/3 (2)
where the coefficient rd about 100 m/kt1/3.
Figure 3: Damage zones in the rock around a nuclear explosion cavity (IGC 1998)
Thus, at least 90% of the energy liberated by the explosion is dissipated within an approximately spherical volume of rock of radius rd. Depending on the magnitude of the in situ stress field (the IGC has concluded that the horizontal in situ stresses at Mururoa and Fangataufa are quite low), some discrete tensile cracking may develop beyond radius rd. Figure 3 illustrates the damage zones produced by a 1 kt explosion. The elastic or seismic wave produced by the explosion may propagate over huge distances and be detected by seismic monitoring systems in other parts of the world. It can also produce shock loadings on structures and slopes within a few km of the source.
As the gas inside the cavity cools with time and there is some seepage of the gas into the surrounding rock, the pressure inside the cavity falls to well below the original lithostatic pressure. At the same time, the molten rock around the cavity periphery begins to solidify and accumulate at the bottom of the cavity. Under these conditions, and especially when the horizontal in situ stresses are low, the crushed and sheared rock above the cavity will collapse progressively. Over a period of a few minutes to a few hours after the explosion, the caved zone or chimney will propagate upwards until it stabilises naturally. As the blocky rubble accumulates in the cavity and then in the chimney void, it will bulk and occupy a greater volume (say 20-30% more) than it did in situ. This factor will cause the eventual arrest of the upward propagation of the chimney. Post-test drilling carried out by DIRCEN/CEA (Bouchez and Lecomte 1996) and the IGC's calculations indicate that the chimney height can be in the range 4 -10 rc , with values near the lower end of this range (5 - 6 rc ) being most likely.
The IGC has estimated that in test area 4 of Mururoa where the total yield was 750 kt (see Figure 1), the total cavity volume created was about 5,000,000 m3, the total chimney volume about 40,000,000 m3, the total damaged volume about 0.68 km3, and the inelastically strained volume about 2.3 km3. The total volume of the layer in which the testing in this area took place is approximately 5 km3.
Observations and Conclusions
General
The Commission's observations and analyses show that there has been no apparent change, on the atoll scale, to the overall mechanical stability of either atoll as a consequence of the underground nuclear tests.
The main observable consequences of the tests are underwater slope failures, open fractures on the rim surface and surface settlements. The fractures visible on the surface are generally associated with subsurface slope displacements and occur only in the carbonates. There is no evidence that slope failures or settlements have occurred in the underlying volcanics.
There has been no significant change in the long-term (beyond 500 years) hydrology of either atoll. The IGC estimates that the long-term change in the natural groundwater flow will be no more than 1%. There are, however, significant short-term changes locally around the test sites (see below).
Numerical modelling indicates the existence of undamaged volcanic cover above and around most of the tests which, therefore, should have been "contained". CEA scientists acknowledge that the top of the explosion-induced chimney did reach the top of the volcanics in a small number of cases (the so-called CRTV tests). Although the direct venting of gaseous radionuclides to the atmosphere probably did not occur, in about 15 cases there is conclusive evidence of the release of tritium, strontium and caesium to the top of the volcanics within a few years of the test. This indicates that either the full volcanic cover was significantly damaged by the explosion, or the integrity and/or initial permeability of the volcanic cover above these tests was significantly different from that anticipated.
Slope failures
Large underwater slope failures occurred on the southwest rim of Mururoa (area 4 in Figure 1) between 1977 and 1979. They were related directly to the specific high yield tests known as Nestor, Priam and Tydee. The largest such slope failure, estimated to have had a volume of ~ 0.1 km3, occurred in 1979 as a result of the Tyd©!=e explosion. It generated hydraulic waves which submerged the rim to a maximum depth of 2 m for several minutes and a standing wave or "mini-tsunami" which travelled out to sea. This failure was followed by several smaller slope failures. This area was studied by the late Commander Jacques Cousteau during a mission to Mururoa in 1987 (Fondation Cousteau 1988) and subsequently publicised.
No further major slope instabilities arising from the underground nuclear testing program are expected on the southern rim of Mururoa. Microseismic monitoring indicates that there has been essentially no further slope activity since the tests. However, this does not rule out the prospect of future submarine slope failures as a result of natural geological and geomorphological processes. Such slope instabilities are well-documented features of these and other atolls. They are evidenced at Mururoa by the underwater accumulation of debris on the atoll flanks and at depth.
Continuing deformation of the ocean-side slope of the northeast rim of Mururoa has been observed and monitored closely since the 1970s (Figure 4). One or more large submarine slope failures could occur in this region in the future. The total volume undergoing deformation is approximately 0.6 km3 or six times the volume of the major 1979 slide in the southwest.
Measurements show that the movements arise from localised creep-like deformations at a depth of approximately 500 m in a weak, sensitive, "chalky limestone" (see Figure 4). Although there is some indication that the deformation rates are declining and may eventually cease, the accumulated strains in the chalky limestone (approaching 0.5%) may have significantly damaged the limestone so that an acceleration of the deformation to collapse over the next 10-20 years cannot be ruled out. As is illustrated in Figure 4, major surface cracking has resulted from these movements.
Figure 4: Monitoring system and incipient slope instability test area, Mururoa (DIRCEN/CEA 1998b)
(Not to scale)
Surface settlements
Surface settlements generally of 1-2 m, but as high as 3 m, developed under the northeast (test area 1 in Figure 1) and southwest (test areas 2, 3 and 4) rims of Mururoa as a result of tests carried out under the rims. At points in both locations, much of the rim and parts of the pre-existing road were submerged.
IGC calculations indicate that the primary mechanism responsible for these settlements is a constant-volume shearing, rather than the surface compaction suggested by CEA. The shearing mechanism also results in significant lateral deformation and fracturing of the carbonates on the shallow underwater ocean slopes. The IGC calculations support CEA claims that the surface settlements are superficial, being confined to the upper carbonates. There is no direct communication between the surface settlements and the explosion cavities as occurred at the Nevada Test Site in the USA.
Settlements of 3 m or more occurred in test area 4, sections of which remain underwater. Vigorous coral building activity (with accumulations of the order of several cm per year) has been observed in these areas, so that the settlements could be effectively "erased" over a period of 100-300 years. However, as significant coral growth above ocean level is unlikely, the restoration of the rim to its previous level does not seem possible without a change in sea level.
Hydrology
The overall hydrology of Mururoa and Fangataufa is typical of atoll hydrology and is well understood. IGC numerical modelling shows that the underground nuclear tests will have negligible effect on the long-term hydrology of the atolls.
The considerable variations and heterogeneities in rock properties that exist in both the volcanics and the carbonates, have only limited local influence on groundwater flow patterns. However, the hydrology of the carbonates is complicated by the presence of numerous extensive, highly transmissive karstic "layers" at several horizons. These karstic horizons "convey" tidal influences of the ocean to the interior of the carbonates, complicating the flow behaviour in those strata. This complexity of flow in the carbonates introduces difficulties in understanding the upwards flow from the volcanics into the carbonates. Analysis of observed tritium concentrations in the lower carbonates and releases into the lagoons, suggests that considerable mixing of water is taking place in the carbonates.
As noted above, there are significant influences around the test zone in the short-term (up to about 500 years). Temperature increases in the vicinity of each explosion cause a local increase in the upward component of the natural groundwater velocity in the overlying volcanics. This increase declines approximately exponentially as the explosion heat is dissipated. The vertically upward Darcy velocity in the volcanics increases from 8 mm/y prior to the tests, to values varying from 0.1 to 1.3 m/y at one year for all cases in which there is good undamaged volcanic cover above the chimney. This velocity decays to about one-half after 10 years and stabilises to approximately 3 cm/y after 500 years. The Darcy velocity in the carbonates above the volcanics is not significantly affected by the tests. For the few tests with no or damaged volcanic cover, the velocities above the chimney are increased almost 50-fold. The velocities in the carbonates are also increased from 2 m/y prior to the tests, to values in the order of 60 m/y after one year, decreasing to 5 m/y after 500 years.
A period of global glaciation some tens of thousands of years in the future would probably result in the development of substantial lenses of fresh water underground on both atolls, making it possible to support a resident population.
Recommendations
The IGC has recommended that the monitoring system being used to measure deformations in the northeast zone of Mururoa be maintained and monitored for the next 20 years, or until failure, if this occurs earlier. Any acceleration of the creep rate should provide sufficient warning of an impending collapse to enable appropriate precautionary safety measures to be taken on Mururoa and neighbouring atolls likely to be affected by a small "tsunami" such as that which occurred in the southwest zone in 1979.
Little information, other than that on micro-seismic activity, is available concerning the continuing deformations of the northeast rim of Fangataufa. No tests were conducted under this rim, but several large-yield explosions were carried out under the lagoon. It appears that the deformations of the rim have been stimulated by the seismic signals produced by these explosions. Large open fractures, apparently an enlargement and extension of existing fractures, essentially parallel to the rim, are clearly visible in this region.
Although the volume of underwater slope undergoing active deformation in the northeast-rim region of Fangataufa appears to be smaller than that at Mururoa, it seems prudent to attempt to estimate the volume involved, continue to monitor the deformations and assess the potential for a serious underwater collapse.
The current program of measurement of the release of tritiated water into the lower carbonates and observations of groundwater travel through the carbonates into the lagoons of Mururoa and Fangataufa is providing valuable insights into the hydrological processes operating in the carbonates. The IGC has recommended that these observations be continued, and that additional analytical and numerical modelling studies be conducted to better establish the influence of the karstic horizons on flow and the processes whereby extensive mixing of waters flowing into the carbonates take place.
References
- Bouchez, J and R Lecomte 1996. The Atolls of Mururoa and Fangataufa (French Polynesia). II. Nuclear Testing. Monaco: Mus©!=e 0c©!=anographie.
- Buigues, D 1996. Mururoa and Fangataufa: Sea-level changes, karstification and the atoll morphology. M©!=m. Soc. g©!=ol. France, No 169, pp373-382.
- DIRCEN/CEA 1998a. Overall distribution and characteristics of the underground nuclear tests carried out at Mururoa and Fangataufa and their effects on the surrounding media. In Geomechanical and Radiological Impact of Nuclear Tests at Mururoa and Fangataufa. Paris: La Documentation Fran®¢aise.
- DIRCEN/CEA 1998b. Structural integrity and stability of the atolls: data and modelling. In Geomechanical and Radiological Impact of Nuclear Tests at Mururoa and Fangataufa. Paris: La Documentation Fran®¢aise.
- Fondation Cousteau 1988. Mission Scientifique de la Calypso sur le Site d'Experimentation Nucleaires de Mururoa. Technical Report. Paris: Fondation Cousteau.
- Guille, G, G Goutiśre, J-F Sornein, D Buigues, A Gachon and C Guy 1996. The Atolls of Mururoa and Fangataufa. 1. Geology - Petrology - Hydrogeology. Monaco: Muse 0ceanographie.
- International Atomic Energy Agency 1998. The Radiological Situation at the Atolls of Mururoa and Fangataufa: Main Report, International Advisory Committee. Vienna: IAEA.
- International Geomechanical Commission 1998. Stability and Hydrology Issues Related to Underground Nuclear Testing in French Polynesia, Vols I and II. Paris: La Documentation Francaise.
|
