The Fukushima accident has shown the need to improve prognostic tools designed to assist governments in their decision making in crisis situations. In Japan, the simulation had difficulty in reproducing several radioactive release dispersion episodes and discrepancies were found between the modelling results and the radiological measurements in the environment.
Download "Modelling of atmospheric transport and release fallout emitted during the Fukushima accident" (PDF, 618 KB)
The main input data for the atmospheric dispersion models of entry comprise weather conditions and the characterisation of the releases:
The weather conditions determine the radioactive plume transport in the atmosphere. However, the difficulty of properly taking into account the orography (description of the topographic relief) in the weather forecasts was raised by the IRSN. Thus, precipitation forecasts, which are responsible for soil contamination due to the leaching of the plume, do not always reflect reality.The source term, that is to say the change over time of the rate of each radionuclide released into the atmosphere, is an essential input for atmospheric dispersion models.
However, there is no clear consensus to identify a source term that is more realistic than another. The differences can be attributed to weather conditions and to the types of measurements used (volume activity, dose rate or total deposition).
Since 2011, simulations have become more realistic and the model /measurement differences have now been significantly reduced. Decisive progress has been made through the use of various weather forecasting sources and of more realistic source terms, as well as through a better understanding of environmental contamination episodes.
This work is currently still underway. And for good reason: all studies still indicate weaknesses and often face the same difficulties in modelling some contamination events in Fukushima.
Download "Main contamination events subsequent to the Fukushima accident" (PDF, 375 Ko)
For IRSN, the purpose of improving the operational assessment tools is to enhance the adequacy of the Institute's response to nuclear accidents, particularly regarding the exposure of populations:
The dispersion models used in an operational context generally have very simplified deposition models. Also, the challenge is knowing whether complex models taking into account the physics of aerosols, particle size or even precipitation would
better simulate depositions.A complex deposition model does not solve
the difficulty in reproducing some episodes. These difficulties seem mainly due to remaining uncertainties about the
input data and to the modelling of plume behaviour within the context of a
Coupling environmental measurements with the atmospheric dispersion model
Since 2011, IRSN has been working to improve the understanding of the Fukushima accident and its environmental consequences, by coupling the analysis of environmental measurements with atmospheric dispersion modelling.
This work is largely being carried out as part of international collaborations. This analysis has enabled the path of the plumes in the atmosphere and the periods of leaching by rain leading to the main soil contamination episodes to be specified.
It also revealed areas that were subjected to significant atmospheric contamination with caesium 137, although the depositions measured in these are low and the dose rates showed no significant increase.
Download "Updating of the knowledge relating to the dispersion and deposition of atmospheric releases from the Fukushima Daiichi nuclear accident in March 2011" (PDF, 884 Ko)