Nonisothermal global-pressure exactness in fractured multiphase flow with evolving fracture aperture

This paper derives exactness criteria for nonisothermal global-pressure formulations in fractured multiphase flow with evolving apertures, demonstrating how thermal forcing and aperture evolution influence the equivalence to phase-pressure systems and providing a conservative least-squares surrogate when exactness fails.

The Gist

Imagine you are trying to navigate a complex, hilly landscape with a map. In the world of oil and gas (or geothermal energy), this "landscape" is the underground rock, and the "hills" are the pressures and temperatures of the fluids flowing through it.

For a long time, scientists have used a clever shortcut to map this terrain. Instead of tracking every single fluid (oil, water, gas) separately, they invented a "Global Pressure"—a single, master number that tells you how the total fluid is moving. It's like having one GPS coordinate that tells you the direction of the entire traffic jam, rather than tracking every individual car.

The Problem:
This shortcut works perfectly when the weather (temperature) is constant. But in the real world, especially in geothermal projects or oil reservoirs, things get hot and cold. When temperature changes, the fluids behave differently: they expand, get thinner, or stick together more.

The paper asks a critical question: Does our "Global Pressure" shortcut still work when the temperature is changing?

The authors, Christian Tantardini and Fernando Alonso-Marroquin, discovered that the answer is "sometimes, but not always." They found that for the shortcut to remain mathematically perfect, the underground system must follow very specific, hidden rules.

The Three Scenarios (The "Regimes")

The paper identifies three possible worlds, which they tested using computer simulations of fractured rock (rock with cracks):

1. The Perfect World (Globally Exact):
   Here, the fluids and the rock play by the rules. The "Global Pressure" map is 100% accurate, even as the temperature changes. You can trust your single GPS coordinate completely.

   • Analogy: It's like driving on a highway where every car moves at the same speed relative to the wind. One speedometer tells you everything.

2. The "Slice" World (Slice-wise Exact):
   This is the tricky middle ground. If you freeze time and look at the system at one specific temperature, the map works perfectly. But if you watch the temperature change over time, the map starts to drift. The rules that worked for "Hot" don't quite match the rules for "Cold."

   • Analogy: Imagine a map that is perfect for summer and a different map that is perfect for winter. If you try to use the "Summer Map" while it's snowing, you get lost. The paper shows that in fractured rocks, the temperature often changes fast enough to break the map, even if the rock looks fine at any single moment.

3. The Broken World (Fully Nonexact):
   Here, the fluids are so chaotic that the "Global Pressure" shortcut fails completely. You can't use a single number to describe the flow; you have to go back to tracking every fluid individually.

   • Analogy: The traffic is so chaotic that no single speedometer works. You have to watch every car individually to know where they are going.

The Twist: The Cracks in the Rock

The paper adds a fascinating layer: Fractures.
Think of the rock as a sponge with cracks running through it. These cracks are the "super-highways" for fluid.

• The Aperture Effect: The width of these cracks (aperture) can change. If you pump hot water in, the rock expands or contracts, making the cracks wider or narrower.
• The Feedback Loop: When the crack widens, fluid rushes through faster. This changes the temperature and pressure even more, which changes the crack width again. It's a domino effect.
• The Result: This feedback loop can push the system from the "Perfect World" into the "Broken World" very quickly. The paper shows that just by changing the width of a crack, you can break the mathematical rules that usually make the simulation easy.

The Solution: The "Best Guess" Map

So, what do engineers do when the perfect map breaks?
The authors propose a clever "repair kit." When the system enters the "Broken World," they use a mathematical technique called a Least-Squares Projection.

• Analogy: Imagine you are trying to draw a straight line through a messy cloud of dots. You can't draw a perfect line through every dot, but you can draw the line that comes closest to all of them on average.
• This "Best Guess" map isn't perfect, but it is conservative. It ensures that mass and energy aren't magically created or destroyed, even if the flow direction is slightly off. It also gives engineers a "defect meter" to tell them exactly how far off the map is, so they know when to be extra careful.

Why This Matters

This research is a bridge between pure math and real-world engineering.

• For Geothermal Energy: It helps us understand how to drill and pump hot water without breaking the mathematical models we rely on to predict energy output.
• For Oil Recovery: It helps predict how fluids move in complex, cracked rocks when we inject steam to heat the oil.

In a nutshell: The paper proves that when things get hot and the rock cracks, the simple "one-number" rule for fluid flow often breaks. But, by understanding why it breaks and using a "best-fit" repair method, we can still navigate the underground world safely and accurately.

Technical Summary

1. Problem Statement

The paper addresses a fundamental challenge in modeling multiphase flow in porous media under nonisothermal (temperature-varying) conditions, specifically within fractured systems.

• The Core Issue: Global-pressure formulations are a powerful reduction technique where multiphase Darcy flow is recast into a single pressure equation driving total flux. In isothermal (constant temperature) settings, this formulation is exactly equivalent to the full phase-pressure system only if constitutive data (mobilities and capillary pressures) satisfy specific "Total-Differential" (TD) compatibility conditions.
• The Gap: When temperature varies, viscosities, densities, mobilities, and capillary pressures become temperature-dependent. The state space expands from a saturation simplex to an augmented manifold of saturation and temperature $(s, T)$.
• The Challenge: It is unclear whether a single scalar global pressure can still exactly represent the system when thermal evolution is active. Furthermore, in fractured media, thermal forcing and thermo-poroelastic effects can alter fracture apertures, changing permeability and the path the system takes through the state space. If exactness fails, standard global-pressure models become approximations without a rigorous error metric.

2. Methodology

The authors employ a rigorous mathematical framework combining differential geometry, thermodynamics, and reduced-order modeling.

A. Mathematical Derivation (Theoretical Framework)

• Augmented State Space: The authors define the state space as $\Delta \times \Theta$ (saturation simplex $\times$ temperature range).
• Differential One-Form: They construct a mobility-weighted capillary one-form $\omega$ on this augmented space:
  $$\omega = \sum A_i(s, T) ds_i + B(s, T) dT$$
  where $A_i$ represents saturation-direction contributions and $B$ represents temperature-direction contributions.
• Exactness Criterion: Using the Poincaré lemma, they prove that a global pressure potential $\Pi(s, T)$ exists (making the formulation exact) if and only if $\omega$ is closed ($d\omega = 0$). This yields two sets of compatibility conditions:
  1. Saturation-Sector Condition: $\partial_{s_j} A_i = \partial_{s_i} A_j$ (The classical isothermal condition).
  2. Mixed Saturation-Temperature Condition: $\partial_T A_i = \partial_{s_i} B$ (A new condition coupling thermal and saturation variations).

B. Reduced Matrix-Fracture Model

To test the theory, they embed the exactness criterion into a minimal mixed-dimensional model:

• Continua: A full-dimensional matrix and a lower-dimensional fracture manifold.
• Physics: Coupled mass and heat transport with local thermal equilibrium (LTE).
• Feedback Loop: A reduced thermo-hydro-mechanical law where temperature changes induce thermal stress, altering the fracture aperture ($b$). Since fracture transmissivity scales cubically with aperture ($T_f \propto b^3$), small thermal changes significantly alter flow paths and state trajectories.

C. Conservative Projection for Non-Exact Regimes

When the exactness conditions fail, the authors propose a fiberwise projection strategy:

• Instead of abandoning the global-pressure approach, they perform a least-squares projection on each fixed-temperature slice.
• This extracts the nearest gradient component of the capillary field, providing a conservative scalar pressure surrogate.
• They introduce defect diagnostics ($D_{ss}$ for saturation-sector failure, $D_{sT}$ for mixed thermal failure) to quantify the error.

3. Key Contributions

1. Derivation of the Nonisothermal Exactness Theorem: The paper establishes that global-pressure exactness in nonisothermal flow is not merely a family of isothermal conditions but requires a mixed compatibility condition between saturation and temperature derivatives.
2. Identification of Three Regimes: The theory predicts and validates three distinct behaviors:
   • Globally Exact: Both saturation and mixed conditions hold.
   • Slice-Wise Exact: Saturation conditions hold, but mixed conditions fail (exact at fixed $T$, but not globally).
   • Fully Non-Exact: Both conditions fail.
3. Conservative Surrogate Formulation: A novel method to handle non-exact regimes by projecting the capillary field onto a gradient field on isothermal slices, preserving conservation laws while explicitly tracking the "defect."
4. Integration with Fracture Dynamics: The work demonstrates how aperture evolution driven by thermal stress acts as a mechanism that pushes the system between exact and non-exact regimes by altering the trajectory through the state space.

4. Numerical Results

The authors validate the framework through three benchmarks:

• Benchmark A (Constitutive Verification): Direct evaluation of the exactness criteria on prescribed data. It successfully recovers the three predicted regimes, confirming that a system can be exact on every temperature slice but fail globally due to mixed incompatibility.
• Benchmark B (Fixed Aperture): A fractured thermal front simulation with fixed aperture. Results show that even without mechanical feedback, thermal forcing drives the fracture and matrix along different trajectories. The mixed saturation-temperature defect dominates the error, particularly in the fracture where thermal excursions are sharper.
• Benchmark C (Evolving Aperture): Adds the thermo-hydro-mechanical feedback. The results show that aperture evolution amplifies the loss of exactness. The changing transmissivity reshapes the state trajectory, causing the system to enter regions of high defect more rapidly. The defect history is qualitatively different from the fixed-aperture case, proving that mechanical feedback is a critical factor in exactness.

5. Significance

• Theoretical Rigor: The paper resolves a long-standing ambiguity regarding the validity of global-pressure formulations in nonisothermal fractured flow, providing a precise mathematical closure condition.
• Practical Application for Geothermal/Reservoir Engineering: It offers a unified framework for simulating geothermal reservoirs where thermal injection causes fracture opening/closing.
• Robust Simulation Strategy: By providing a conservative projection and quantitative defect diagnostics, the work allows engineers to use computationally efficient global-pressure models even when exactness fails, while knowing exactly how and where the approximation deviates from the full physics.
• Mechanism Insight: It highlights that thermal forcing alone can drive transitions between exact and non-exact regimes, and that fracture aperture evolution acts as a dynamic amplifier of these transitions.

In summary, this work unifies nonisothermal exactness theory, fractured-flow dynamics, and conservative reduced closures, providing a robust foundation for next-generation simulators for geothermal and multiphase fractured reservoirs.