- 1Universidad Politécnica de Cataluña, DECA, Barcelona, Spain (sergio.zlotnik@upc.edu)
- 2Centre Internacional de Mètodes Numèrics a l’Enginyeria (CIMNE). Barcelona, Spain
The identification of hydro-mechanical parameters governing earthfill dam behaviour under transient loading conditions is essential for reliable interpretation of monitoring data and predictive analysis. Although coupled flow–deformation models can represent these processes in detail, their direct use in inverse analyses is often prohibitive due to the large number of forward simulations required. This work addresses the efficient estimation of material parameters in earthfill dams by integrating a reduced-order formulation of the problem into an inverse strategy.
A transient, nonlinear hydro-mechanical model for unsaturated soils is considered in the context of a sensor-driven inverse problem, where piezometric measurements are used to constrain model parameters. Reduced-order models based on proper orthogonal decomposition (POD) are introduced to enable repeated model evaluations within the inversion procedure while retaining the key features of the hydro-mechanical response. The framework targets the estimation of relevant soil properties, such as hydraulic conductivity, water retention characteristics, and mechanical stiffness, and is illustrated using both synthetic observations and field piezometer data from the Glen Shira dam during rapid drawdown events.
REFERENCES
[1] Pinyol, N. M., Alonso, E. E., Olivella, S. (2008). Rapid drawdown in slopes and embankments. Water Resources Resarch, 44(5). doi: 10.1029/2007WR006525
[2] Nasika, C., Díez, P., Gerard, P., Massart, T.J., Zlotnik, S. (2022). Towards real time assessment of earthfill dams via Model Order Reduction. Finite Elements in Analysis & Design, 199: 103666. doi: 10.1016/j.finel.2021.103666
How to cite: Zlotnik, S., Schrader, J., Beatrice, J., García González, A., and Muixí, A.: Hydro-mechanical parameter estimation in earthfill dams using reduced-order models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13600, https://doi.org/10.5194/egusphere-egu26-13600, 2026.