Most permeability calculations start with Darcy’s law, which relates the flow rate of a fluid through a porous material to the pressure difference driving that flow, the fluid’s viscosity, and the material’s cross-sectional area and thickness. That single relationship covers a surprising range of situations, from testing a soil sample in a lab cylinder to modeling groundwater moving through an aquifer. But depending on what you’re measuring, the conditions you’re working under, and whether the fluid is a liquid or gas, the right approach can shift considerably.
Darcy’s Law as the Foundation
Henry Darcy published his findings in 1856 after studying water flow through sand filters in Dijon, France. He was the first to demonstrate that significant flow resistance occurs within porous media like aquifers, and he recognized the mathematical similarity between his law and flow through narrow tubes.1Water Resources Research. Henry Darcy and the making of a law The core idea is straightforward: the volume of fluid passing through a porous sample per unit time is proportional to the pressure drop across it and the cross-sectional area, and inversely proportional to the sample’s length and the fluid’s viscosity.
In practical terms, you measure or control four things: the flow rate of fluid coming out of the sample, the pressure difference between the upstream and downstream sides, the dimensions of the sample (area and length), and the viscosity of the fluid. Rearranging the relationship to solve for permeability gives you a value in units of area, typically expressed in darcys or millidarcys in petroleum and geoscience work, or in square meters in SI notation. If you’re working with water through soil, you’ll often see the result expressed as hydraulic conductivity instead, which folds in the fluid’s density and viscosity so the output is a velocity (meters per second or centimeters per second). The two quantities are directly related: permeability is a property of the material alone, while hydraulic conductivity also depends on the fluid.
Laboratory Permeameter Tests
The most direct way to calculate permeability is to push fluid through a sample in a controlled setup and measure what happens. Two classic lab approaches exist, each suited to different permeability ranges.
A constant-head permeameter keeps the water level (and therefore the pressure) at a fixed height above the sample throughout the test. You measure the volume of water that passes through in a given time, note the sample dimensions and the head difference, and plug those into Darcy’s law. This works well for relatively permeable materials like sand and gravel. One practical issue to watch for: if the material is highly permeable, the tubing and fittings in a conventional permeameter can introduce their own resistance, leading to underestimation of the true hydraulic conductivity.2PubMed Central. A modification of the constant-head permeameter to measure saturated hydraulic conductivity of highly permeable media Modified designs that eliminate the bottleneck beneath the sample address this problem.
A falling-head permeameter works better for finer-grained, less permeable materials like silt or clay, where the flow rate is too slow to measure conveniently with a constant head. Here, water starts in a narrow standpipe above the sample and you let it drop. You record how long it takes for the water level to fall from one marked height to another. The calculation accounts for the changing head over time, using the ratio of the starting and ending water levels along with the sample and standpipe dimensions. Both laboratory and field versions of this test consistently show a strong linear relationship between flow rate and hydraulic gradient, confirming Darcy’s law applies well to these setups.3Nature. Improvement of field falling-head test and determination of hydraulic conductivity using Darcy’s equation
Estimating Permeability From Grain Size
You don’t always have a lab sample to test. In many soil and groundwater projects, permeability gets estimated from the size distribution of the grains making up the material. Two approaches dominate here.
The Hazen formula is the simpler one. It uses a single number from a standard sieve analysis: the grain diameter at which 10 percent of the material (by weight) is finer, known as D10. You square that diameter and multiply by an empirical coefficient that depends on the grain shape and sorting. The formula is widely used in practice for clean, uniformly graded sands, though it tends to lose accuracy with poorly sorted or silty materials.4Jurnal Asiimetrik: Jurnal Ilmiah Rekayasa & Inovasi. Analysis of Sand Grain Characteristics and Permeability Using Hazen Formula and Laboratory Test Its appeal is speed and low cost: if you already have sieve data, the calculation takes seconds.
The Kozeny-Carman equation is more physically grounded. Rather than relying on a single grain-size percentile, it links permeability to the porosity (the fraction of the material that’s void space), the specific surface area of the grains, and an empirical constant that captures pore geometry. The classical version assumes the pore space can be modeled as a bundle of tortuous capillary tubes. More recent work has revisited this by assuming a fractal pore-space geometry, yielding updated permeability-porosity relationships that better fit a wider range of natural materials.5Geophysical Research Letters. Permeability‐porosity relationship: A reexamination of the Kozeny‐Carman equation based on a fractal pore‐space geometry assumption The Kozeny-Carman approach works well for granular materials but breaks down for rocks with fractures or with very heterogeneous pore structures.
Field Methods Using Pumping Tests
In groundwater and petroleum engineering, you often need permeability values for rock or sediment that’s buried hundreds of meters underground. You can’t easily bring it into a lab, so instead you stress the formation in place by pumping water out of a well and watching how the water level responds in nearby observation wells.
The Thiem solution is one of the oldest and most practical field approaches. It assumes steady-state conditions, meaning you’ve pumped long enough that water levels have stabilized. By comparing the water levels in two observation wells at different distances from the pumping well, you can solve for the formation’s transmissivity (the product of permeability and aquifer thickness). One advantage of this method is that you don’t necessarily need a formal pumping test: researchers have shown that by analyzing long-term monitoring data from wells with variable but known withdrawal rates, you can approximate steady-state conditions and apply the Thiem solution to estimate transmissivity without ever running a dedicated constant-rate test.6PubMed. Estimation of Aquifer Transmissivity From Analysis of Long-Term Monitoring With the Thiem Solution
For transient conditions, where water levels are still changing, the Theis method and its Cooper-Jacob approximation are the standard tools. The Cooper-Jacob method simplifies the math by working with a straight-line portion of a graph of drawdown versus the logarithm of time, making it practical for hand calculations. Applied to pumping test data, it yields estimates of both transmissivity and storativity (how much water the formation releases from storage as the pressure drops).7Mathematical Modelling in Civil Engineering. Performing Pumping Test Data Analysis Applying Cooper-Jacob’s Method for Estimating of the Aquifer Parameters From transmissivity, you divide by the aquifer thickness to get permeability.
When Flow Isn’t Linear
Darcy’s law assumes a linear relationship between flow rate and pressure gradient. That works at low velocities, but at higher flow rates, inertial effects kick in and the pressure drop starts rising faster than the flow rate. This happens in the region close to a wellbore in gas reservoirs, in coarse gravel beds, and in fractured rock where fluid velocities can be substantial.
The Forchheimer equation handles this by adding a second term to Darcy’s law. Where Darcy’s law has a single resistance term proportional to velocity, the Forchheimer equation adds a term proportional to velocity squared, capturing the kinetic energy lost to inertial effects.8PubMed Central. Reply to Comment “Assessing Porous Media Permeability in Non-Darcy Flow: A Re-Evaluation Based on the Forchheimer Equation” The extra term involves a coefficient (often called the Forchheimer or inertial coefficient, denoted β) and the fluid density. Calculating permeability under these conditions means you need flow-rate data at multiple velocities so you can fit both the Darcy permeability and the inertial coefficient simultaneously.
In gas reservoir engineering, the non-Darcy coefficient β is a key parameter for evaluating how easily gas can flow in the near-wellbore region, where velocities are highest.9Geofluids. Numerical Simulation of the Non-Darcy Flow Based on Random Fractal Micronetwork Model for Low Permeability Sandstone Gas Reservoirs Laboratory measurements of β are typically performed on core samples by running flow tests at a range of rates and fitting the resulting pressure-drop data to the Forchheimer equation.10Journal of Petroleum Science and Engineering. Laboratory measurements of non-Darcy flow coefficients in natural and artificial unconsolidated porous media If you only use Darcy’s law in these high-velocity situations, you’ll overestimate the formation’s resistance (or equivalently, underestimate the flow rate for a given pressure drop), because the extra inertial losses get lumped into an artificially low permeability value.
Gas Permeability and the Klinkenberg Correction
Measuring permeability with gas instead of liquid introduces its own complication. At low gas pressures, especially in tight rocks with very small pore throats, gas molecules slip along pore walls rather than sticking to them the way a liquid does. This “slip flow” makes gas appear to flow more easily than a liquid would through the same rock, inflating the apparent permeability.
The Klinkenberg correction accounts for this. You measure gas permeability at several different mean pore pressures, then plot the results against the inverse of the mean pressure. At high pressures (where slip flow is negligible), the data points fall on a straight line. Extrapolating that line to infinite pressure (where the inverse of pressure equals zero) gives you the equivalent liquid permeability, sometimes called the Klinkenberg-corrected permeability. Research on tight sandstones has shown that slip flow effects become significant when gas pressure drops below about 5 MPa. At those low pressures, the apparent gas permeability can be considerably higher than the true liquid-equivalent value.11Journal of Geophysical Research: Solid Earth. Klinkenberg Effect and Effective Pressure for Gas Permeability of Tight Sandstones If you skip this correction, you risk reporting a permeability that’s systematically too high and not representative of how the rock would behave with oil or water flowing through it.
How Stress and Temperature Shift the Numbers
Permeability isn’t a fixed number for a given rock. It changes as conditions change, and two of the biggest factors are mechanical stress and temperature.
When rock is buried deeper, confining pressure squeezes pore spaces shut. The relationship between permeability and effective stress (confining pressure minus pore pressure) is nonlinear: at low stresses, microcracks close rapidly and permeability drops steeply, while at higher stresses, the remaining pore space compresses more gradually. Experimental work on clayey sandstones has shown that the effective-stress coefficient for permeability tends to be greater than one, meaning pore pressure changes have an outsized influence compared to confining pressure changes of the same magnitude.12Journal of Geophysical Research: Solid Earth. Effective Stress Law for the Permeability and Pore Volume Change of Clayey Sandstones For any lab permeability measurement to be relevant to in-situ conditions, the test should reproduce the effective stress state the rock experiences underground.
Temperature adds another layer. Heating a fluid lowers its viscosity, which by itself increases hydraulic conductivity even if the rock doesn’t change. But the rock does change: thermal expansion can close microcracks under confining stress, reducing permeability, while at the same time thermal cracking can open new flow paths. In soils, an increase in temperature changes groundwater density and viscosity, and thermal loading can alter soil fabric in ways that either increase or decrease intrinsic permeability.13Engineering Geology. Exploring the effects of temperature on intrinsic permeability and void ratio alteration through temperature-controlled experiments In mudstone under triaxial stress, the competition between these mechanisms plays out in stages: at lower temperatures, pore closure from thermal expansion dominates, and permeability drops; at higher temperatures, thermal cracking can partially offset the closure.14Scientific Reports. Effect of temperature on permeability of mudstone subjected to triaxial stresses and its application The upshot is that a single permeability number measured at room temperature and atmospheric pressure can be far from what the material exhibits in its natural environment.
Well Logging and Downhole Estimation
In the oil and gas industry, permeability often needs to be estimated continuously along a wellbore from logging tools rather than from discrete core samples. Nuclear magnetic resonance (NMR) logging is one of the most widely used techniques for this purpose. NMR tools measure the relaxation times of hydrogen atoms in pore fluids, which correlate with pore size: larger pores produce longer relaxation times, and larger pores generally mean higher permeability. Two classical models, the Seevers-Kenyon and Timur-Coates models, translate NMR relaxation-time distributions into permeability estimates.15Petrophysics – The SPWLA Journal of Formation Evaluation and Reservoir Description. Estimating the Permeability of Rocks by Principal Component Regressions of NMR and MICP Data These models require calibration against core measurements, and their accuracy can vary depending on the rock type and pore structure.
More recently, researchers have turned to micro-CT imaging of rock samples combined with numerical simulation. A small core plug is scanned at micrometer resolution to produce a three-dimensional image of the pore space, and then flow equations are solved directly on that digital pore geometry. Physics-informed convolutional neural networks trained on direct numerical simulations of the Stokes equation can predict permeability from the 3D images, offering a faster alternative to the traditional lattice Boltzmann approach while still grounding the predictions in fluid physics.16Computational Geosciences. Estimating permeability of 3D micro-CT images by physics-informed CNNs based on DNS These digital methods are powerful for understanding pore-scale controls on permeability but are limited by the size of the imaged volume, which may not capture larger-scale heterogeneity.
Relative Permeability in Multiphase Systems
Everything discussed so far assumes a single fluid saturating the pore space. In reality, oil reservoirs contain oil, water, and gas occupying the same pores, and each fluid partially blocks the others. Relative permeability describes what fraction of the rock’s absolute (single-phase) permeability is available to each fluid at a given saturation. At full water saturation, the relative permeability to water is one and to oil is zero. As oil saturation increases, its relative permeability rises while water’s drops.
Measuring relative permeability directly is expensive and time-consuming, involving core-flood experiments where two fluids are injected simultaneously at controlled rates. An alternative is to calculate it from capillary pressure data, which is often easier to obtain. The Brooks and Corey model is widely accepted for this purpose: it takes a capillary-pressure-versus-saturation curve and, using the pore-size distribution index derived from that curve, produces relative permeability curves for both the wetting and non-wetting phases.17Water Resources Research. Comparison of methods to calculate relative permeability from capillary pressure in consolidated water‐wet porous media The accuracy depends heavily on how well the capillary pressure data represents the actual pore structure, and the model works best in water-wet systems with reasonably uniform pore geometry.
Permeability Beyond Earth Sciences
The same physics of fluid moving through porous materials shows up in biomedical engineering, where permeability is becoming a key design parameter for bone-replacement scaffolds. A scaffold for bone repair needs to be porous enough that nutrients, oxygen, and waste products can flow through it, and that bone cells can migrate into it and grow. Permeability captures the transport capability of the scaffold’s architecture more comprehensively than porosity alone, because two scaffolds with the same total void fraction can have very different flow characteristics depending on pore connectivity and geometry.18PubMed Central. Determining the Permeability of Porous Bioceramic Scaffolds: Significance, Overview of Current Methods and Challenges Ahead Researchers measure scaffold permeability using pump-based setups (essentially miniaturized permeameters), gravity-driven flow rigs, acoustic methods, and computational simulations. Standardization remains an open challenge: different labs using different methods on similar scaffolds can get substantially different numbers, making it hard to compare results across studies. The underlying math, though, is the same Darcy’s law framework used in geosciences, just scaled down to centimeter-sized samples and adapted for the irregular, often hierarchical pore structures of bioceramics and glass scaffolds.