Lessons learnt on the impact of an unprecedented soil decontamination program in Fukushima on contaminant fluxes.

In the context of elevated concerns related to nuclear accidents and warfare, the lessons learnt from the Fukushima Daiichi Nuclear Power Plant accident in 2011 are important. In particular, Japanese authorities implemented an ambitious decontamination program to reduce the air dose rate in order to facilitate the return of the local inhabitants to previously evacuated areas. This approach contrasts the strategy adopted in Chernobyl, where the most contaminated areas remain off limits. Nonetheless, the effectiveness of the Japanese decontamination strategy on the dispersion of radioactive contaminant fluxes across mountainous landscapes exposed to typhoons has not been quantified. Based on the unique combination of river monitoring and modeling in a catchment representative of the most impacted area in Japan, we demonstrate that decontamination of 16% of the catchment area resulted in a decrease of 17% of sediment-bound radioactive fluxes in rivers. Decontamination operations were therefore relatively effective, although they could only be conducted in a small part of the area due to the dominance of steep forested slopes. In fact, 67% of the initial radiocesium contamination was calculated to remain stored in forested landscapes, which may contribute to future downstream radiocesium dispersion during erosive events. Given that only a limited proportion of the initial population had returned in 2019 (~30%), it raises the question as to whether decontaminating a small percentage of the contaminated area was worth the effort, the price, and the amount of waste generated?

Impacts of farmland decontamination on 137Cs transfers in rivers after Fukushima nuclear accident: Evidence from a retrospective sediment core study.

Following the Fukushima Daiichi Nuclear Power Plant disaster in March 2011, the Japanese government initiated an unprecedented decontamination programme to remediate 137Cs-contaminated soils and allow population return. This programme involved the removal of topsoil under farmland and residential land, and its replacement with "fresh soil" composed of granitic saprolite. However, decontamination was limited to these two land uses, without remediating forests, which cover 70 % of the surface area in the affected region. In this unprecedented context, the specific impact of this unique decontamination programme on 137Cs transfers in river systems remains to be quantified at the catchment scale. In this study, based on the analysis of a sediment core collected in June 2021 in the Mano Dam reservoir draining a decontaminated catchment, the effects of soil decontamination on particle-bound 137Cs dynamics and sediment source contributions in response to a succession of extreme precipitation events were retrospectively assessed. The sequence of sediment layer deposition and its chronology were reconstructed through the analysis of several diagnostic properties (organic matter, elemental geochemistry, visible colourimetry, granulometry) and contextual information. During abandonment (2011-2016), cropland contribution decreased (31 %). Concurrently, 137Cs activity and deposition flux decreased (19 and 29%year-1, respectively). Following decontamination (2017), sediment transfer increased (270 %) in response to increased contributions from decontaminated cropland and "fresh soil" (625 % and 180 % respectively). Meanwhile, forest contributions remained stable. In contrast, 137Cs activity dropped (65 %), although 137Cs deposition flux remained constant. Forests acted as a stable source of 137Cs. Accordingly, 137Cs deposition flux after decontamination (2016-2021) was similar to that observed during the 5-years period of land abandonment (2011-2016), as a result of the regrowth of spontaneous vegetation over farmland, protecting soil against erosion. Future research should further investigate the impact of longer land abandonment that prevailed in some regions decontaminated lately on the 137Cs fluxes in the rivers.