Bayesian Inference of Coupled Groundwater Flow and Radiogenic Helium-4
Production and Transport at the Catchment Scale
Abstract
Hydrogeological numerical models are crucial for assessing radioactive
waste disposal. State variables are widely used but not quantitatively
during model inversion, except for hydraulic head. This study explores
incorporating radiogenic Helium-4 (4Herad) as an unconventional state
variable in model inversion, addressing parameter solution
non-uniqueness. Employing a formal Bayesian approach, two 3D
physically-based models are used to infer posterior parameter
distributions. Major 4Herad sources include crustal fluxes and in situ
production in the Boom Clay. Coupled inference narrows marginal
posterior distributions, reducing uncertainty in calibration and
prediction of certain parameters. Integrating 4Herad as a state variable
improves model reliability and fidelity, particularly in constraining
solute transport and refining groundwater flow parameters and derived
fluxes. This state-of-the-art study demonstrates the effectiveness of
including 4Herad in combination with classical hydraulic head
observations, showcasing transformative potential for reducing
uncertainties crucial for radioactive waste disposal. It marks the
first-ever attempt at harnessing 4Herad quantitative insights into flow
and transport parameters following a Bayesian approach at a catchment
scale, setting a precedent for future research. The study strongly
advocates for the continued utilization of 4Herad in hydrogeological
modeling, emphasizing its relevance for safety studies.