Consequences of radioactive releases into the sea resulting from the accident at the Fukushima Dai-ichi nuclear power plant − Evolution of expert investigation according to the data available

International audienceThe accident at the Fukushima Dai-ichi Nuclear Power Plant (FDNPP) in March 2011 led to an unprecedented direct input of artificial radioactivity into the marine environment. The Institute for Radioprotection and Nuclear Safety was requested by the French authorities to investigate the radioecological impact of this input, in particular the potential contamination of products of marine origin used for human consumption. This article describes the close link between the responses provided and the availability of the data, as well as their nature and ability to meet the requirements of expert investigation. These responses were needed: (i) to evaluate the inputs of radionuclides into the marine environment, (ii) to understand their dispersion in seawater, and (iii) to estimate their transfer to the biota and sediments. Three phases can be distinguished which characterise these processes during the accident and post-accident periods. The first phase corresponds to an emergency phase during which no measurements were available on samples from the marine environment. It involved the formulation of hypotheses based solely on the expertise of the Institute for Radioprotection and Nuclear Safety. The second phase started when the Japanese authorities provided measurements of the concentrations of radionuclides in seawater. Although these data were not yet adapted to addressing the problems of radioecology, the scenarios could then be refined and the estimates developed in more detail. During the third phase, the accumulation of data over the course of time made it possible to study the phenomena in an appropriate way. The chronology of the events shows that it is essential to have (i) significant measurements of concentration from samples collected in the various matrices of the marine environment, regularly updated and sufficiently well-documented, (ii) samples of seawater collected at the earliest opportunity as close as possible to the damaged site to characterise the source term, and (iii) a numerical tool allowing rapid modellin