19–22 May 2026
Europe/Paris timezone

A Hysteretic Aperture Model for Fractured Rocks

20 May 2026, 12:50
15m
Oral Presentation (MS03) Flow, transport and mechanics in fractured porous media MS03

Speaker

Josue Barroso (National Laboratory for Scientific Computing (LNCC))

Description

Fractures play a fundamental role in controlling the hydraulic and mechanical response of geological formations, with direct implications for subsurface energy applications such as hydrocarbon production, CO₂ sequestration, geothermal systems, and underground hydrogen storage (UHS). In particular, UHS operations involve repeated injection and withdrawal cycles that induce successive loading and unloading of the stress field, making hysteresis in fracture aperture evolution a key mechanism governing long-term permeability changes and system performance.

In this work, we propose a fracture aperture model that explicitly accounts for hysteresis under cyclic loading and unloading conditions, with a particular focus on rock joints. The model is based on the Barton–Bandis joint closure law and is designed to reproduce the hysteretic behavior commonly observed in laboratory experiments on jointed rock samples. Two bounding curves describe aperture evolution during monotonic loading and unloading. A general stress–aperture path within the hysteresis loop is then defined by interpolating key parameters such as initial normal stiffness and maximum mechanical closure. When stress decreases after a loading phase, fracture closure follows an intermediate unloading trajectory that depends on the stress history.

The proposed model is especially relevant for hydro-mechanical coupling and fractured-media upscaling, as it introduces a history-dependent relationship between effective stress and macroscopic petrophysical properties. This feature is crucial for applications involving cyclic operations, where neglecting hysteresis may lead to inaccurate permeability predictions and biased reservoir performance assessments.

Laboratory data from cyclic loading–unloading tests with increasing stress levels are used to calibrate the model and to evaluate its ability to predict intermediate closure paths. An analysis of the evolution of model parameters across multiple cycles is also performed, providing insights into model limitations and possible improvements.

Country Brazil
Acceptance of the Terms & Conditions Click here to agree

Author

Josue Barroso (National Laboratory for Scientific Computing (LNCC))

Co-authors

Alexandr Zhemchuzhnikov (PUC-Rio) Euripedes Vargas (PUC-Rio) Marcio Murad (Laboratorio Nacional de Computacao Cientifica) Matheus Peres (PUC-Rio) Tayna Lobo (Laboratório Nacional de Computação Científca)

Presentation materials

There are no materials yet.