What Is Hydraulic Action in Erosion and How Does It Work?

Hydraulic action is the erosion process in which the force of moving water alone, without the grinding help of sand, pebbles, or dissolved chemicals, breaks apart rock and soil. It works primarily by driving pressurized water into cracks and joints, compressing pockets of trapped air, and generating turbulent pressure fluctuations that gradually pry fragments loose. Coastlines, river channels, and even dam spillways all bear its marks, though the specific mechanics vary depending on whether ocean waves or river currents supply the energy.

How Moving Water Breaks Solid Rock

Rock might seem impervious to water, but no natural rock surface is perfectly smooth or seamless. Joints, bedding planes, and microcracks riddle even the hardest formations. When water under pressure enters those openings, it pushes outward on the walls of the crack, widening it slightly. That alone is not usually enough to snap rock apart in one go. What matters is repetition. Each wave impact or each pulse of turbulent river flow drives another wedge of pressure into the same fracture, extending it a little further every time.

Numerical modeling of high-speed water jets striking rock has shown that the dominant mode of failure is tensile, meaning the rock is pulled apart rather than crushed. Pure shear cracks, where rock slides along a plane, are comparatively rare. Mixed-mode fractures, part tension and part shear, account for a meaningful share of total cracking depending on the boundary conditions, but tension does most of the work.¹ That finding is intuitive once you picture what is happening: water is not so much smashing into rock as it is inflating tiny cracks from the inside, pulling opposing surfaces apart.

The Role of Trapped Air

One of the most powerful amplifiers of hydraulic action is air. When a wave slams into a cliff face, or when river water floods a crack network, pockets of air can become trapped inside. Because air compresses far more readily than water, the initial impact squeezes those pockets to a fraction of their original volume in an instant, and the resulting spike in pressure is far higher than the water alone would generate. Research modeling wave impacts on coastal cliffs has found that intense fluid pressures develop along interior crack surfaces as water rushes in during wave runup. When air pockets within those crack spaces are instantaneously compressed, the pressure climbs sharply, imposing large loads on the overlying rock.²

The bending stresses created by this repeated loading can initiate or propagate microcracks in the rock. Over dozens, hundreds, or thousands of wave cycles, those microcracks link up until an entire slab of bedrock detaches and drops away as a boulder. This is why hydraulic action is sometimes called a fatigue process: each individual wave may do very little visible damage, but the cumulative effect is dramatic. The same study found that boulders produced this way can be surprisingly large, because once crack propagation reaches a critical length the entire block gives way at once.²

Hydraulic Action on Coastlines

Coastal cliffs are where hydraulic action puts on its most visible performance. During high tides and storms, waves strike the base of a cliff with enormous energy. The water exerts several distinct hydraulic forces: wave quarrying, in which pressure differentials pop blocks out of the cliff face; water hammer, the shock transmitted through water when its momentum is stopped abruptly by rock; and the air-compression mechanism described above.³ If sand and gravel are swirling in the surf, abrasion joins the attack, but hydraulic action can erode a cliff even on beaches where there is almost no loose sediment to serve as grinding tools.

Cliff retreat is not a steady, uniform process. Storms do the lion’s share of the work. Modeling of extreme wave events has shown that steep waves around ten meters high impacting a cliff can generate impact pressures sufficient to propagate cracks, detach blocks, and even lift large slabs onto the cliff top. When waves reach the height of the cliff edge, the water that overtops the cliff can itself carry enough energy to dislodge and transport blocks across the plateau above.⁴ The result is that most cliff erosion at a given site can happen in a handful of major storms rather than being spread evenly across decades. Researchers have monitored these impacts using accelerometers mounted on cliff tops, recording vibrations at thousands of readings per second to capture the brief, violent pressure spikes that ordinary instruments would miss.⁵

Hydraulic Action in Rivers

Rivers erode their beds and banks through hydraulic action too, though the process looks a bit different from the wave-slamming drama of the coast. In bedrock channels, the key mechanism is plucking: turbulent flow generates pressure differences across rock blocks bounded by joints, and those pressure differences eventually lever the block free. Computational fluid dynamics work on block entrainment has found that the pressure difference between the upstream face and the downstream face of a block can substantially affect whether that block stays put or gets ripped away.⁶

Research on plucking in steep rivers has identified three contributions to the pressure acting on the underside of a block. The first is flow driven through the crack network by the gradient in water elevation as flow speeds change along an uneven channel bed. The second is pressure fluctuations transmitted into cracks from the unsteady, turbulent flow structures inside hydraulic jumps, those churning zones where fast, shallow flow abruptly transitions to slower, deeper flow. The third is pressure fluctuations carried downstream from turbulent structures generated upstream of the plucking site.⁷ Together, these three mechanisms mean that bedrock erosion in a steep river is not just about how fast the current moves on the surface. The internal plumbing of the crack network matters enormously, because water reaches the underside of a block through cracks and exerts lift from below.

For softer materials like alluvial banks, hydraulic action works differently. The current shears against the bank face, entraining particles and undercutting the slope until a mass of soil collapses into the channel. Interestingly, numerical simulations of bank erosion in river bends have shown that peak bank retreat often happens after the flood peak, not during it. One reason is that in highly curved bends, the fastest flow migrates away from the outer bank as discharge rises, reducing the shear stress on the bank at peak flow. Another is that bank failure is triggered by combinations of pore-water pressure and hydrostatic confining pressure that develop during the drawdown phase between successive flood peaks.⁸ If you have ever walked a river after a flood and noticed a fresh bank collapse that clearly happened after the water dropped, that is the mechanism in action.

How Hydraulic Action Differs from Abrasion, Attrition, and Corrosion

Erosion in real landscapes is a team sport, and hydraulic action rarely works in isolation. Understanding what sets it apart from its partners helps clarify when it dominates and when it plays a supporting role.

Abrasion: Sediment carried by water grinds against rock surfaces like sandpaper. The water is the transport medium; the sediment is the cutting tool. Remove the sediment, and abrasion stops.
Attrition: Sediment particles collide with each other in transit, rounding and shrinking themselves. This breaks down the tools of abrasion but does not directly erode the bed or banks.
Corrosion: Sometimes called solution, this is the chemical dissolution of rock by slightly acidic water. Limestone coastlines dissolve this way, and the process requires no mechanical force at all.
Hydraulic action: Pure mechanical force from water and compressed air. No sediment needed, no chemical reaction required. The water itself is the tool.

The distinction matters because the relative importance of each process varies by setting. On a sandy beach, abrasion may dominate because there is plenty of sediment available to grind the cliff base. On a rocky headland with a clean wave-cut platform and little loose sediment, hydraulic action takes the lead. Coastal research has noted that waves exert both hydraulic forces and mechanical abrasion “if sediment particles are available,” underscoring that hydraulic action can proceed on its own when sediment is absent.³

Cavitation and Its Relationship to Hydraulic Action

Cavitation is a related but distinct mechanism that sometimes gets lumped in with hydraulic action. It occurs when water flows fast enough that the local pressure drops below the vapor pressure of water, causing tiny bubbles to form. When those bubbles are swept into a higher-pressure zone, they collapse violently, generating microscopic jets of water that slam into nearby surfaces at extreme speeds. Over time, these microjets pit and erode even very hard materials like steel turbine blades, let alone natural rock.

Experiments using focused ultrasonic transducers to induce cavitation on copper plates have demonstrated that higher surrounding pressures increase the depth and diameter of the pits left by collapsing bubbles, as well as the overall erosion intensity.⁹ In natural rivers, cavitation is thought to contribute to bedrock erosion in very steep channels and large waterfalls where flow velocities are high enough, but it is difficult to observe directly. The threshold velocity for cavitation in open channel flow is roughly in the range of ten meters per second or more, which limits cavitation to the most energetic settings.

Cavitation can also team up with sediment. A review of erosion in hydraulic turbines found that sediment particles increase the number and size of cavitation nuclei, which are the tiny gas pockets that seed bubble formation. The shape, size, concentration, and hardness of the particles all influence how much worse the cavitation erosion becomes. Changes in local pressure driven by particle motion or rotation can amplify the damage beyond what either cavitation or sediment abrasion would cause alone.¹⁰ While that research focused on engineered systems, the same physics applies anywhere fast-moving, sediment-laden water contacts a surface.

What Makes Rock More or Less Vulnerable

Not all rock erodes at the same rate under hydraulic action. Several properties control vulnerability. The most obvious is rock strength: weaker rocks fail under lower water pressures. Simulations of water-jet erosion have shown that among the rock types tested, failure was observed only in the weakest sandstone, while stronger formations remained intact under the same conditions.¹

Joint spacing and orientation are equally important. A rock mass with closely spaced joints provides more crack surfaces for water to enter, and blocks defined by those joints are smaller and easier to pluck. Research on fluvial plucking has explicitly focused on how pressure is transmitted through joint-bounded crack networks, because the geometry of those cracks determines whether the water can get underneath a block and exert upward lift.⁷ Massive, unjointed rock like fresh granite is far more resistant than the same granite once it has been weathered enough for joints to open.

Back pressure also plays a role. The same water-jet study found that increased water back pressure, essentially the ambient water pressure surrounding the target, suppresses erosion by counteracting the jet’s ability to open and extend cracks.¹ In a coastal context, this means submerged rock at depth experiences less hydraulic erosion than rock at the waterline, where air and water alternate and back pressure is low. It helps explain why the most intense erosion on a cliff is concentrated in the narrow tidal zone.

How Vegetation Slows Hydraulic Erosion on Riverbanks

If you have ever noticed that vegetated riverbanks hold up better than bare ones, the science backs that impression, and the reason is more interesting than simple root reinforcement. Laboratory experiments using jet-erosion tests on soil samples with no roots, synthetic plastic roots, and live switchgrass roots found that both synthetic and live roots reduced soil loss when root density was high enough. But the live roots did something extra. Only the living roots significantly increased the critical shear stress of the soil, the threshold of water force needed to start pulling particles away. Synthetic roots of the same density did not.¹¹

The explanation is biological: living roots produce extracellular substances and promote soil aggregation in ways that dead fibers cannot. Dense fibers of any kind, living or synthetic, also alter the flow patterns near the bank, deflecting and absorbing some of the hydraulic force. So vegetation provides a double defense: it physically disrupts the flow and it chemically strengthens the soil. Losing riparian vegetation, whether to grazing, development, or fire, removes both layers of protection at once and can trigger rapid bank retreat.

Protecting Infrastructure from Hydraulic Action

Engineers have long sought ways to absorb or deflect the hydraulic forces that eat away at coastlines and river channels. One common approach is armoring. Concrete armor blocks deployed along shorelines are designed to break up incoming wave energy before it reaches the structure behind them.¹² The interlocking shapes create a rough, porous surface that dissipates wave energy through turbulence and friction rather than allowing a clean, high-pressure impact against a smooth wall. Seawalls, breakwaters, and revetments all work on variations of this principle.

In rivers, bank protection ranges from simple riprap, loose angular stones heavy enough to resist current forces, to more sophisticated bioengineering approaches that combine structural elements with vegetation. The logic follows directly from the erosion mechanics: if hydraulic action works by pushing water into cracks and generating pressure fluctuations, anything that reduces the velocity or turbulence intensity near the surface under attack will slow erosion. Strategies that maintain or restore riparian vegetation are increasingly popular because they address both the hydraulic and biological components of resistance.

Hydraulic Erosion Beyond Earth

Hydraulic action is not limited to our planet. Both Mars and Saturn’s moon Titan show clear evidence of past or present river channels carved by flowing liquid. On Mars, ancient rivers transported sediment and shaped landscapes in ways that researchers can now reconstruct by analyzing channel widths, slopes, and the grain sizes of sediment found by rovers. Predictions made from channel geometry at Gale Crater and Jezero Crater overlap with grain sizes actually measured by the Curiosity and Perseverance rovers, and the reconstructed flow conditions are consistent with long-lived water flow at both sites.¹³

On Titan, the liquid is not water but methane and ethane, yet the physics of hydraulic erosion still applies. The same research suggests that Titan’s rivers may be wider, slope more gently, and transport sediment at lower flow rates than rivers on Earth or Mars. Predicted sediment fluxes to the coast of Ontario Lacus, one of Titan’s lakes, could build the lake’s river delta in as little as roughly a thousand years.¹³ The details change with the fluid, the gravity, and the rock composition, but the fundamental principle holds: flowing liquid exerts mechanical force on the surface it crosses, and over time that force reshapes the landscape. Hydraulic action, in other words, is not an Earth-specific quirk. It is a consequence of fluid dynamics that plays out wherever liquid meets solid ground.

1
International Journal of Rock Mechanics and Mining Sciences. Numerical simulation of rock erosion performance of a high-speed water jet using an immersed-body method
2
Marine Geology. How does wave impact generate large boulders? Modelling hydraulic fracture of cliffs and shore platforms
3
Geomorphology. The unsteady nature of sea cliff retreat due to mechanical abrasion, failure and comminution feedbacks
4
Marine Geology. Modelling the processes of cliff-top erosion and deposition under extreme storm waves
5
Coastal Engineering. Wave impacts on vertical cliffs: Insights from laboratory experiments and field observations
6
Journal of Geophysical Research: Earth Surface. Toward Entrainment Thresholds in Fluvial Plucking
7
Geosphere. Plucking phenomena in nonuniform flow
8
Water Resources Research. Numerical simulation of hydrodynamics and bank erosion in a river bend
9
PubMed Central. Quantitative evaluation of the microjet velocity and cavitation erosion on a copper plate produced by a spherical cavity focused transducer at the high hydrostatic pressure
10
Heliyon. A review of cavitation and silt erosion in hydraulic turbines: Causes, impacts, and mitigation strategies
11
Geomorphology. Do roots bind soil? Comparing the physical and biological role of plant roots in fluvial streambank erosion: A mini-JET study
12
Water. Numerical Analysis of Wave Interaction with a New Ecological Quadrangular Hollow Block
13
PubMed Central. Reconstructing river flows remotely on Earth, Titan, and Mars