Morphology-Adaptive Curvature Analysis of Fluid–Fluid Interfaces in Porous Media

Deep within the earth, in the tiny, tangled pores of rocks like sandstone and limestone, fluids such as oil, water, and gas constantly jostle for space. These fluids do not mix; instead, they meet at sharp boundaries where they push against one another. The force that governs this struggle is called capillary pressure, a physical push that determines whether a drop of oil will stay trapped in a rock or be swept away by flowing water. To understand how much pressure is needed to move these fluids, scientists must measure the curvature of the boundary where the two fluids meet. If the boundary is tightly curved, like a small bubble, the pressure is high. If it is flatter, the pressure is lower. This relationship is a fundamental law of physics, but measuring the shape of these invisible boundaries inside a rock is incredibly difficult. The rock is imaged using powerful X-rays, which create a digital picture made of tiny cubes. When scientists try to trace the fluid boundary from these cubes, the result is often a rough, jagged surface full of digital noise, making it nearly impossible to tell if a curve is real or just an artifact of the measurement.

A team of researchers has developed a new way to solve this problem, creating a method that adapts to the specific shape of the fluid boundary rather than forcing every shape into the same mold. In a study published in Research Article, Faisal Aljaberi, Martin Blunt, and Mohammed Al Kobaisi demonstrate that no single mathematical tool works best for every situation. Just as a carpenter might use a chisel for straight lines and a rasp for curves, the researchers found that different fluid shapes require different ways of calculating their curvature. By testing their approach on perfect geometric shapes and then on real rock samples, they showed that their adaptive method reduces measurement errors by 45 percent compared to the best single method currently available. This improvement allows scientists to see the true pressure inside the rock with much greater clarity, revealing how oil and water behave in ways that were previously hidden by the noise of the data.

The researchers began by acknowledging that the standard tools used to measure curvature were failing because they treated all fluid interfaces the same. In the complex world of porous rock, fluid boundaries can take on three distinct shapes. Sometimes, a trapped blob of oil forms a smooth, rounded cap, like a bubble. Other times, the fluid clings to the corners of the rock, forming a long, thin arc. In more complex situations, particularly in rocks that have been altered by crude oil, the boundary can twist into a saddle shape, curving up in one direction and down in another. The team tested various mathematical techniques on computer models of these three shapes. They found that a method based on averaging the surface worked well for the rounded caps and the corner arcs, smoothing out the digital noise to reveal the true shape. However, that same averaging method ruined the measurement of the saddle shapes, flattening their unique curves and hiding their true nature. Conversely, a different technique that fitted a smooth patch to the surface worked better for the saddles but was too sensitive to noise for the other shapes.

To fix this, the team built a smart workflow that acts like a shape detector. When the computer analyzes a fluid interface, it first identifies whether the surface is mostly rounded, mostly arced, or mostly saddle-shaped. Once the shape is identified, the system automatically selects the best mathematical tool for that specific geometry and adjusts how much it smooths the data. It also learned to ignore the edges where the fluid meets the solid rock and the other fluid, a region where the digital image is often most distorted. By excluding these problematic edges and applying the right tool to the right shape, the researchers could extract a clean, accurate picture of the curvature. They tested this new system on two different types of sandstone: one that naturally repels oil (water-wet) and one that has been altered to hold onto oil (mixed-wet).

The results were striking. In the water-wet rock, where the fluid boundaries are mostly smooth and rounded, the new method produced results that matched independent measurements taken with pressure sensors outside the rock. The data points clustered tightly together, showing a clear and consistent relationship between how much oil was left in the rock and the pressure holding it there. However, the difference became even more dramatic in the mixed-wet rock. In these complex conditions, the fluid boundaries twist into those difficult saddle shapes. The standard commercial software, which uses a fixed method for all shapes, calculated a much higher pressure than actually existed, failing to capture the true, low-pressure state of the connected oil. The new adaptive method, by contrast, correctly identified that the pressure was dropping toward zero as the water pushed the oil out. It revealed that the standard tools were overestimating the force needed to move the oil, simply because they could not handle the complex geometry of the mixed-wet rock.

The study also looked at the tiny, disconnected blobs of oil that get trapped during the process. The researchers found that these isolated droplets have their own distinct pressure states, different from the main flow of oil. When the new method measured all the oil interfaces together, including these trapped droplets, the calculated pressure jumped up, showing that the system was mixing two different physical realities. This distinction is crucial because it proves that the pressure governing the flow of oil is different from the pressure holding the trapped droplets in place. Furthermore, by analyzing thousands of these trapped droplets, the team discovered a clear pattern: smaller droplets, which get trapped early in the process, have highly curved surfaces and high pressure, while larger droplets, trapped later, have flatter surfaces and lower pressure. This detailed view of the microscopic world inside the rock provides a much deeper understanding of how fluids move through the earth, offering a more reliable way to predict where oil can be found and how it can be recovered.

The success of this work lies in its refusal to apply a one-size-fits-all solution to a problem that is inherently diverse. By recognizing that the shape of the fluid interface dictates the best way to measure it, the researchers have created a tool that sees the rock more clearly than ever before. This clarity is not just an academic exercise; it translates directly into a better understanding of the forces at play deep underground. Whether for managing water resources, storing carbon dioxide, or extracting energy, knowing the true pressure inside the rock is essential. The new method provides a way to see that truth, cutting through the noise of the data to reveal the physical reality of the fluids moving through the earth's pores.