EGU General Assembly location:Vienna date:22-27 April 2012
Radionuclide concentrations in aquifers represent an important indicator in estimating the impact of a planned
surface disposal for low and medium level short-lived radioactive waste in Belgium, developed by the Belgian
Agency for Radioactive Waste and Enriched Fissile Materials (ONDRAF/NIRAS), who also coordinates and
leads the corresponding research.
Estimating aquifer concentrations for individual radionuclides represents a computational challenge because
(a) different retardation values are applied to different hydrogeologic units and (b) sequential decay reactions
with radionuclides of various sorption characteristics cause long computational times until a steady-state is reached.
The presented work proposes a methodology reducing substantially the computational effort by postpro-
cessing the results of a prior non-reactive tracer simulation. These advective transport results represent the
steady-state concentration – source ﬂux ratio and the break-through time at each modelling cell. These two
variables are further used to estimate the individual radionuclide concentrations by (a) scaling the steady-state
concentrations to the source ﬂuxes of individual radionuclides; (b) applying the radioactive decay and ingrowth in
a decay chain; (c) scaling the travel time by the retardation factor and (d) applying linear sorption. While all steps
except (b) require solving simple linear equations, applying ingrowth of individual radionuclides in decay chains
requires solving the differential Bateman equation. This equation needs to be solved once for a unit radionuclide
activity at all arrival times found in the numerical grid. The ratios between the parent nuclide activity and the
progeny activities are then used in the postprocessing.
Results are presented for discrete points and examples of radioactive plume maps are given. These results
compare well to the results achieved using a full numerical simulation including the respective chemical reaction
Although the proposed method represents a fast way to estimate the radionuclide concentrations without
performing timely challenging simulations, its applicability has some limits. The radionuclide source needs to be
assumed constant during the period of achieving a steady-state in the model. Otherwise, the source variability of
individual radionuclides needs to be modelled using a numerical simulation. However, such a situation only occurs
in cases of source variability in a period until steady-state is reached and such a simulation takes a relatively short
The proposed method enables an effective estimation of individual radionuclide concentrations in the frame
of performance assessment of a radioactive waste disposal. Reducing the calculation time to a minimum enables
performing sensitivity and uncertainty analyses, testing alternative models, etc. thus enhancing the overall quality
of the modelling analysis.