Using spectrocolourimetry to trace sediment source dynamics in coastal catchments draining the main Fukushima radioactive pollution plume (2011-2017)

O. Evrard (1, 2) , Roxanne Durand (1) , Anthony Foucher (1) , Tales Tiecher (3) , Virginie Sellier (1) , Yuichi Onda (4) , Irene Lefevre (1) , Olivier Cerdan (5) , J. Patrick Laceby (6)

1 LSCE - Laboratoire des Sciences du Climat et de l'Environnement [Gif-sur-Yvette]
2 GEDI - Géochimie Des Impacts
3 UFRGS - Universidade Federal do Rio Grande do Sul [Porto Alegre]
4 Université de Tsukuba = University of Tsukuba
5 BRGM - Bureau de Recherches Géologiques et Minières (BRGM)
6 EMSD - Environmental Monitoring and Science Division of Alberta

Purpose Spectrocolourimetric measurements provide a relatively inexpensive, quick and non-destructive alternative to the analysis of geochemical and organic matter properties. When used in the analysis of sediments and their potential sources, these colour parameters may provide important information on the dominant processes (i.e. erosion) occurring in the Critical Zone. Here, they are used to investigate whether eroded sediment is derived from forest (i.e. natural), cultivated (i.e. anthropogenic) or subsoil sources in order to assess their potential to monitor the effect of decontamination in regions impacted by fallout from the Fukushima Dai-ichi Nuclear Power Plant (FDNPP) accident. Materials and methods Fifteen spectrocolourimetric properties (L*, a*, b*, C*, h, x, y, z, L, a, b, u*, v*, u', v') were measured in potential source (n = 37) and sediment (n = 400) samples collected during 13 campaigns from 2011 to 2017 after major flood events in two catchments (total surface area of 450 km(2)) draining the main FDNPP radioactive pollution plume. Potential sources included topsoil from forest and cultivated sources along with subsoil material originating from landslides, channel banks and the decontamination of cultivated areas. The optimum set of parameters used in the mixed linear model to calculate the sediment source contributions was obtained through the use of a range test, the Kruskal-Wallis H test and a linear discriminant analysis. Results and discussion Nine selected colour parameters correctly classified 100% of the source samples (i.e. forest, subsoil and cultivated sources). The results illustrate that cultivated landscapes were the main source of sediment to these river systems (mean 56%, SD 34%) followed by subsoil (mean 26%, SD 16%) and forest sources (mean 21%, SD 24%). However, these contributions varied strongly over time, with a peak of subsoil contributions (mean 57%, SD 17%) in Fall 2015, coinciding with the occurrence of a typhoon after the remediation works. These results were consistent with monitoring studies conducted in the same area that showed the major impact of typhoon Etau in September 2015 on sediment and radiocaesium fluxes. Conclusions These original results demonstrate that spectrocolourimetric measurements may contribute to the routine monitoring of the effectiveness of remediation works in this post-accidental context. Owing to the inexpensive, rapid and non-destructive analyses, spectrocolourimetric-based tracing methods have significant potential to provide information on the dominant erosion processes occurring in the Critical Zone.

Mots clés: ACCIDENT; DISPERSION; MANAGEMENT; PREFECTURE; SOURCE DISCRIMINATION; CONTAMINATED PARTICULATE MATTER; RADIOCESIUM; SOILS; ADSORPTION; DEPTH DISTRIBUTION; Colour; FDNPP; Sediment fingerprinting; Soil erosion; Spectroscopy

Domaines: Géologie appliquée

O. Evrard, Roxanne Durand, Anthony Foucher, Tales Tiecher, Virginie Sellier, et al.. Using spectrocolourimetry to trace sediment source dynamics in coastal catchments draining the main Fukushima radioactive pollution plume (2011-2017). Journal of Soils and Sediments, 2019, 19, pp.3290-3301. ⟨10.1007/s11368-019-02302-w⟩. ⟨hal-02380977⟩