Probabilistic assessment of the influence of lake properties in long-term radiation doses to humans

The assessment processes concerning the safety of nuclear waste repositories include the modelling of radionuclide transport in biosphere and the evaluation of the doses to the most affected humans. In this paper, a scenario, in which a contaminated lake is the water source for drinking water, irrigation water and watering of livestock, is presented. The objective of the paper is to probabilistically study the influence of lake properties as parameters in the assessment scenario. The properties of the lake are a result of previously conducted probabilistic studies, where the land uplift of the terrain surrounding the repositories and the formation of water bodies were studied in a 10,000-year time span using Monte Carlo simulation. The lake is formed at 3000 years from present day and the changing properties of the lake have been used in the study. The studied radionuclides 36Cl, 135Cs, 129I, 237Np, 90Sr, 99Tc and 238U enter the lake with a rate of 1 Bq/year. The transport process from the lake water to humans is described and the doses (dose conversion factors) to adult humans are evaluated based on a study on average food consumption. Sensitivity analysis is used for identifying the parameters having the most influence on the outcome of the dose. Based on the results from the sensitivity analysis, the volumetric outflow rate of the lake and the volume of the lake were taken into closer consideration. The results show the influence of probabilistically derived geomorphic lake input parameters on the dose. •The study is based on a lake formed 3000 AP due to post-glacial land uplift.•A probabilistic model of the lake for the next 10,000 years is used in the study.•Volumetric outflow rate and lake volume are found to be among the key parameters.•Statistical variability of key parameters is important in dose estimation to humans.