Is non-human species radiosensitivity in the lab a good indicator of that in the field? Making the comparison more robust.

Ecological risk assessment has globally become the basis for environmental decision-making within government and industry for chemical substances. Regarding radioactive substances, recently revised International and European Basic Safety Standards are pushing the development of member state policy on environmental regulation in the field of radiological protection. Within this framework, existing derived effect benchmarks for ionising radiation and non-human species need to be more robust to reinforce their credibility when used as levels of exposure considered to be safe for the environment. Actually, the derivation of such benchmarks has mainly relied on laboratory studies from a limited number of species. Moreover lab species would be apparently less radiosensitive than for example terrestrial wildlife chronically exposed to ionising radiation in the Chernobyl Exclusion Zone. Additionally to the results of such comparison that still need to be confirmed, another way to challenge benchmarks is to improve the quality/quantity of radiotoxicity data constituting the basis for a statistically-based comparison. This is the major focus of this paper where we demonstrate through various examples how to make the comparison more robust (i) by analysing the discrepancy between lab and field at the taxonomic level rather than at the ecosystem level, (ii) by extending the knowledge base making use of acute radiotoxicity data, (iii) by identifying environmental factors modifying radiological dose-effect relationship in the field.