Wildlife chronic exposure to environmentally relevant radionuclide concentrations: Experimental results are needed to compensate for the current lack of knowledge

摘要

The debate on the need for a system of radiological protection of the environment drives regulators to urge scientists on conceptualisation of methods demonstrating explicitly that the environment is protected against radioactive contaminants. As regards the FASSET database on radiation effects to non-human biota, one of the major difficulty in the implementation of ecological risk assessment for radioactive pollutants is the lack of data for chronic low level exposure. A general way to deal with situations for which there are no relevant data as regards the actual situation where the risk is to be estimated is to use safety factors. The higher their values, the higher the uncertainty on the risk estimate, but, on the other hand, any safety factor could be reduced as more data become available. Going back to radionuclides for which data are sparse, both concerning fate and effects in ecosystems, derivation of ecologically relevant and scientifically defensible benchmarks become a critical issue in ERA. The scope of this paper is to illustrate the relevance of the development of a greater depth of understanding of radionuclide fate and biological effects at several hierarchical levels to support quantitative risk assessments with defined and acceptable uncertainty bounds. The examination of the Fasset Radiation Effects Database content draw the conclusion that gaps on wildlife chronic internal exposure to α or β emitters are among the most critical for ERA. The following crucial issues are discussed and exemplified for uranium in this paper : (ⅰ) Radionuclide bioavailability is key to an accurate assessment of both exposure and effect : media quality criteria are needed; (ⅱ) Chronicity of exposure is key to dose estimate and induced effects; (ⅲ) Considering different scales for biological effects (from early to delayed, from subcellular to high organisation level) is crucial to evidence ecologically relevant indicators.