🔢 mathematics

A Second-Order Monolithic Scheme for the Coupled Stokes--Biot Model Using Total Pressure

This paper presents and analyzes a robust, second-order, fully implicit monolithic time-discretization scheme for the coupled Stokes–Biot system using a three-field total-pressure formulation, establishing discrete energy stability and optimal convergence rates that are independent of key physical parameters.

Original authors: Talal Alshehri, Mingchao Cai, Xiaoqin Shen

Published 2026-08-07

📖 3 min read🧠 Deep dive

CC0 1.0

Original authors: Talal Alshehri, Mingchao Cai, Xiaoqin Shen

Original paper dedicated to the public domain under CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the world as a giant, interconnected sponge. Some parts of this sponge are soaked with water that flows freely, like a river, while other parts are made of a squishy, solid material that can stretch and compress, like a wet sponge in your hand. In many real-world situations—from oil being squeezed out of underground rocks to blood flowing through our soft tissues—these two worlds collide. The water pushes on the sponge, and the sponge pushes back, changing the shape of the space the water can flow through. Scientists call this a "fluid-poroelastic interaction." It's a tricky dance because the water moves fast and the sponge moves slow, and they are constantly changing each other's steps. If you try to simulate this on a computer, it's like trying to choreograph a dance where one partner is a hummingbird and the other is a sleeping bear; if you aren't careful, the computer gets confused, the numbers go wild, and the simulation crashes. This happens especially when the sponge is very hard to squish (nearly incompressible) or when the water has a hard time squeezing through the tiny holes (low permeability).

Now, enter a team of mathematicians who decided to build a better dance floor for this hummingbird and bear. They developed a new, super-smart computer recipe—a "monolithic scheme"—to simulate how fluids and sponges interact over time. Instead of trying to solve the water's movement and the sponge's squishing separately and then trying to glue the answers together (which often leads to mistakes), their method solves everything at once, as one giant, unified puzzle. They also introduced a clever new way of looking at the pressure inside the sponge, treating it as a "total pressure" that combines different forces. This trick prevents the computer from getting stuck in a jam, a problem known as "volumetric locking," which usually happens when the sponge is too stiff. Their recipe is designed to be incredibly fast and accurate, using a second-order time-stepping method called BDF2, which is like taking two giant, precise steps forward instead of one small, wobbly one.

The paper presents a rigorous mathematical proof that this new recipe is stable and won't fall apart, no matter how stiff the sponge is or how slow the water flows. They showed that their method is robust, meaning it doesn't break when the physical properties of the materials change drastically. To prove it works in the real world, they ran a series of computer experiments. These simulations confirmed that their method is indeed twice as accurate as older methods when they refine the time steps, and it handles the tricky "nearly incompressible" cases without losing its cool. However, they also found a limit: when the sponge is so tight that the water can barely move through it at all (extremely low permeability), the pressure calculations get a bit messy and require even finer details to get right. But for almost every other scenario, this new approach offers a stable, accurate, and reliable way to watch the fluid and the sponge dance together without tripping over each other.

Technical Summary: A Second-Order Monolithic Scheme for the Coupled Stokes–Biot Model Using Total Pressure

Problem Statement
The paper addresses the numerical approximation of fluid–poroelastic structure interaction (FPSI) systems, which couple the time-dependent Stokes equations (describing a viscous incompressible fluid) with Biot's consolidation equations (describing a saturated poroelastic medium). A primary challenge in simulating these systems is the phenomenon of volumetric locking. In classical two-field displacement–pressure formulations of the Biot equations, standard finite element discretizations often lose robustness when the Lamé parameter is large (nearly incompressible porous skeletons) or when permeability is small. This leads to spurious pressure oscillations and artificially stiff displacement responses that cannot be resolved by mesh refinement alone. Furthermore, existing higher-order time discretization methods for coupled Stokes–Biot systems often rely on operator splitting or impose restrictive stability conditions dependent on physical parameters (e.g., the Beavers–Joseph–Saffman friction parameter or the Lamé parameter), limiting their applicability in challenging regimes.

Methodology
The authors propose a fully implicit, monolithic, second-order time-discretization using the Backward Differentiation Formula of order 2 (BDF2). The core of the methodology involves reformulating the Biot subsystem using a three-field total-pressure formulation. In this approach, the total pressure ( $\xi = \alpha p_p - \lambda_p \nabla \cdot \eta$ ) is introduced as an additional primary unknown alongside the solid displacement ( $\eta$ ) and pore pressure ( $p_p$ ).

Key methodological features include:

Monolithic Coupling: All variables (fluid velocity, fluid pressure, solid displacement, total pressure, and pore pressure) and all interface conditions (mass conservation, stress balance, and Beavers–Joseph–Saffman slip) are solved simultaneously as a single linear system at each time step. This eliminates operator splitting errors and interface extrapolation.
Total-Pressure Variable: The use of total pressure creates a parameter-robust saddle-point structure that suppresses volumetric locking at both continuous and discrete levels, particularly in nearly incompressible regimes.
Discrete Energy Analysis: Stability is established using the BDF2 G-stability identity, a discrete energy argument that avoids the need for CFL-type restrictions or parameter-dependent time-step limits.
Finite Element Discretization: The spatial discretization employs a mixed variational framework based on discrete inf–sup stability, utilizing Taylor–Hood pairs for the fluid and poroelastic regions to ensure stability.

Key Contributions
The paper makes the following specific contributions:

Unconditional Energy Stability: The authors prove that the monolithic BDF2 scheme is unconditionally stable for arbitrary time-step sizes. The stability estimate relies on the G-stability identity and holds without restrictions dependent on the Lamé parameter, Biot–Willis coefficient, or storage coefficient.
Parameter Robustness: The analysis demonstrates robustness with respect to critical physical parameters. The estimates remain valid as the Lamé parameter $\lambda_p \to \infty$ (nearly incompressible limit) and the storage coefficient $s_0 \to 0$ . This is achieved through the total-pressure variable, which ensures the energy remains uniformly bounded.
Optimal Convergence Rates: The authors derive a priori error estimates showing second-order convergence in time and optimal-order convergence in space ( $O(\Delta t^2 + h^k)$ ) in the corresponding energy norms. The convergence constants are independent of the physical parameters in the weighted norms used.
Handling of Differential-Algebraic Structure: The scheme is formulated to handle the differential-algebraic nature of the coupled system, where fluid pressure is determined by incompressibility and total pressure enters through a differentiated storage term, without requiring separate algebraic solvers or splitting.

Results
Numerical experiments were conducted to verify the theoretical findings:

Temporal Accuracy: Tests confirmed second-order convergence in time for all variables (velocity, displacement, pressures) when using a Backward Euler startup step followed by BDF2.
Spatial Accuracy: Spatial convergence studies demonstrated optimal rates consistent with the polynomial degrees of the finite element spaces.
Parameter Robustness Tests:
- Lamé Parameter ( $\lambda_p$ ): The method maintained stability and accuracy as $\lambda_p$ increased by orders of magnitude (approaching the incompressible limit), confirming the locking-free nature of the formulation.
- Storage Coefficient ( $s_0$ ): The scheme remained robust in the singular limit $s_0 = 0$ .
- Fluid Viscosity ( $\mu_f$ ): Variations in viscosity affected error magnitudes but did not alter the asymptotic convergence rates.
- Permeability ( $\kappa$ ): While the method was robust for moderate permeability, the authors observed a deterioration in pore-pressure convergence rates in the extreme low-permeability regime ( $\kappa = 10^{-6}$ ). This is attributed to the singularly perturbed nature of the Darcy operator and the presence of boundary layers, consistent with the theoretical stability estimates which are weighted by permeability.

Significance and Claims
The paper claims that the proposed monolithic BDF2 scheme with total-pressure formulation provides a stable, accurate, and reliable framework for simulating coupled Stokes–Biot problems across a broad range of physically relevant parameter regimes. By avoiding operator splitting, the method eliminates splitting errors, and by utilizing the total-pressure variable, it overcomes the locking issues that plague classical formulations in nearly incompressible scenarios. The authors emphasize that their stability and convergence analysis is parameter-robust, meaning the theoretical guarantees do not degrade as key physical parameters (like stiffness or compressibility) become extreme, a significant improvement over previous methods that required restrictive time-step conditions dependent on these parameters. The work establishes a rigorous foundation for high-order, fully coupled simulations of fluid–poroelastic interactions.

More like this