France systematically underestimated the devastating impact of its nuclear tests in French Polynesia in the 1960s and 70s, according to new innovative research that could allow more than 100,000 people to claim damages.
France conducted 193 nuclear tests from 1966 to 1996 at Moruroa and Fangataufa atolls in French Polynesia, including 41 atmospheric tests until 1974 that exposed the local population, local workers and French soldiers to high levels of radiation.
Analyzing data from 2,000 pages of recently released French Ministry of Defense documents, analyzing maps, photos and other records, and conducting dozens of interviews in France and French Polynesia, the researchers painstakingly reconstructed three important nuclear tests and their consequences.
The Mururoa Archives, a collaboration between investigative journalism writing, Princeton University’s Global Science and Security Program and an environmental justice research collective, Interprt, suggest that the impact of the 1966 Aldébaran, Encelade and Centaure tests, 1971 and 1974 was much larger than officially recognized.
“The state has been working hard to bury the toxic heritage of these tests,” said Geoffrey Livolsi, editor-in-chief of Disclose. “This is the first truly independent scientific attempt to measure the scale of the damage and recognize the thousands of victims of France’s nuclear experiment in the Pacific.”
France’s national institute of health and medical research (Inserm) published a report last month on the health consequences of the tests, arguing that “it was not possible to conclude with certainty” that there was a link between them and the multiple cases of cancer that emerged on islands, but underlining the need to “refine dose estimates”.
Modeling the Mururoa Archives of radioactive fallout from just the Centaure bomb – the last to explode into the atmosphere before France’s tests went underground – suggests that Paris, in fact, underestimated contamination in Tahiti by up to 40%, potentially allowing dozens thousands of people to be officially recognized as test victims.
Using weather data, military files and scientific records on the size of the weapon’s radioactive mushroom cloud, the team planned their passage over Tahiti – and the 80,000 inhabitants of Papeete, the capital of French Polynesia.
The cloud was predicted to head north, but it never reached the predicted height of 9,000m, instead it remained at around 5,200m. There it was blown inexorably to the west towards Tahiti, where no precautions had been taken to protect the population, arriving on the island at 2 am on January 19, 1974, 42 hours after the explosion.
According to a confidential report by the Polynesian health ministry obtained by the researchers, about 11,000 test victims received radiation doses in excess of 5 millisieverts (mSv): five times the qualified level for compensation, provided that they later contracted certain types of cancer .
Based on documents released in 2013, however, the researchers estimate that the entire population of Tahiti and the Polynesian islands of Leeward – about 110,000 people – were exposed to a radiation dose of more than 1 mSv with the Centaure test alone.
According to the new calculations, the actual radiation doses received by residents of some districts of Papeete were two or three times higher than those recorded in a study by the French Atomic Energy Commission (CEA) published in 2006, researchers say .
The project also reveals a confidential report sent from Paris to the Polynesian government in February last year referring to “a group of thyroid cancers” in the Gambier Islands, which were directly affected by the consequences of France’s first nuclear test, Aldébaran, in July 1966.
The unpublished report, France’s first official recognition of the impact of testing on health, says that the cluster’s location, “focused on islands where radioactive precipitation was most intense … leaves little doubt about the role of ionizing radiation” in cancers.
Thyroid, throat and lung cancer, as well as cases of leukemia and lymphoma and bone and muscle diseases related to strontium and cesium poisoning, remain prevalent on the islands, say the researchers, citing interviews with several inhabitants, many of whom were children at the time. of the tests.
The researchers also cite a confidential exchange of e-mails dated 2017, in which the French army acknowledges, supposedly for the first time, that up to 2,000 of the 6,000 military personnel based in French Polynesia and involved in the tests between 1966 and 1974 have since contracted at least a form of cancer.

Despite widespread concerns, however, France did not establish a compensation council for civilian and military victims until 2010, with claimants – in theory – having to prove only that they lived in French Polynesia at the time and contracted one of the 23 cancers recognized as resulting from radiation to receive a payment.
But the council, known as Civen, has so far paid damages to just 454 people, including just 63 locals, rejecting more than 80% of the claims without having to justify its decisions. Many families, the researchers said, including some with multiple members suffering from different types of cancer, have given up trying to claim.
Prospective claimants do not have the medical information needed to file a complaint, nor the means to accurately establish the level of radiation to which they were exposed, say the researchers: despite 26 “radiological surveillance points” designed to measure the effects of precipitation radioactive, only 20% of the islands’ surface is effectively monitored.
Defective equipment exacerbated the problems: in 1971, the year of the Encelade test, some radiation measuring stations were operating with a 50% margin of error.
In addition, the researchers reveal, the 2006 CEA report on radiation in French Polynesia, on which Civen bases its compensation decisions, was validated by the International Atomic Energy Agency (IAEA) – in a study commissioned and paid for by the Ministry French Defense – “on the assumption that all the data contained therein are correct” – which, his calculations suggest, is far from the case.
The researchers recalculated the effective radiation dose received by the inhabitants based on samples collected by the military of the time and including contamination of the atomic cloud, toxic dust in the soil, the inhalation of contaminated particles and, mainly, the consumption of contaminated rainwater.
For some tests, the difference between recalculations and official figures was insignificant. For others, however, it was dramatic: Aldébaran’s 1966 tests produced levels of contamination three times higher than those recorded so far, they said.