Application of a modified aggregated dead zone model to estimate radionuclide transport in running surface water bodies

This study was conducted to determine applicability of the aggregated dead zone model to describe radionuclide transport in running surface water bodies. The model assumes that the hydraulic system is divided into separate sections, and the dissolved substance is transported through each section by piston flow (i.e., transport without dispersion) and then enters the instantaneous and complete mixing box (i.e., dispersion without transport). This results in a box model with an aggregated dead zone and a delay —the transport time. Instead of modeling the dissolved solute concentration continuously in both distance and time along the watercourse, the model uses a “black box” approach and considers the concentration at the box outlet (in the aggregate dead zone) as a function of the concentration at the box inlet and time. The use of the box model significantly reduces the requirements for the necessary initial and boundary data compared to 1-, 2- and 3-dimensional models. The modification of the model accounts for non steady-state flow, radioactive decay and interaction of the radionuclide with suspended sediments and bottom deposits. The mathematical apparatus of the modified model is a system of ordinary differential equations with a delayed argument. The paper compares the modeled values of radionuclide concentrations with the measured data on case of 3H distribution as a result of releases from 5 nuclear power plants in the Loire River channel for six months with hourly discretization and for distribution of 90Sr releases in the Kyiv reservoir that occurred as a result of the Chornobyl disaster for a year with daily discretization. The close match between the model and the measured data the short computer implementation time demonstrate the possibility of parametric identification and adaptation of the model to different types of radionuclides in different running water bodies and highlight the model’s potential for environmental modeling applications.