Infiltration is water crossing the boundary from the surface into the soil; percolation is that same water continuing its journey downward through deeper soil layers toward the water table. The distinction sounds like splitting hairs, but it matters because different forces dominate at each stage, different soil properties control the rate, and different things go wrong when either process stalls. Understanding where one ends and the other begins helps explain everything from why your yard floods after a storm to how aquifers get recharged.
Where One Process Ends and the Other Begins
Think of it as a two-part relay. Infiltration is the entry step: rain or irrigation water meets the ground surface and is pulled into the top few centimeters of soil. The rate at which this happens depends heavily on conditions right at the surface, including how porous the topsoil is, whether it’s already saturated, and whether anything (a crust, compacted ground, pavement) is blocking the way in. Once water has entered the soil, it doesn’t just stop. Gravity and pressure differences between wet and dry soil pull it further down through the soil profile. That ongoing downward movement through the unsaturated zone beneath the root layer is percolation.
Field monitoring in Israel’s deep vadose zone showed how this relay works in practice: each large rain event triggered an infiltration wave at the surface, which then pushed a wetting front progressively deeper in a step-like pattern governed by the frequency of big storms, followed by slower drainage between events.1Water Resources Research. Water percolation through the deep vadose zone and groundwater recharge: Preliminary results based on a new vadose zone monitoring system In irrigated agricultural fields, researchers tracked the same sequence: surface water entered the soil during flood irrigation, then percolated below the crop rooting zone at measurable velocities toward the shallow aquifer underneath.2Vadose Zone Journal. Water Movement through a Shallow Vadose Zone: A Field Irrigation Experiment The infiltration and the percolation are physically continuous, but they happen at different depths, at different speeds, and under different dominant controls.
The Forces at Work
At the surface, infiltration is driven by a combination of capillary suction and gravity. Dry soil acts almost like a sponge: the tiny spaces between soil particles create capillary forces that pull water in. Gravity helps, of course, but in very dry soil the capillary pull can be the bigger factor at first. The classic way hydrologists model this is as a sharp “wetting front” being sucked into the soil, sometimes described as piston-type infiltration, where a distinct boundary separates wet soil above from dry soil below.3Soil and Tillage Research. Estimating the parameters of the Green–Ampt infiltration equation from rainfall simulation data: Why simpler is better
As the soil wets up, capillary suction weakens because the pores are already full. Gravity increasingly takes over as the dominant force driving water deeper, and this is where infiltration transitions into percolation. Once the upper soil is fairly saturated, water drains down through connected pore spaces under gravity alone, though the soil’s structure, layering, and how much clay it contains all determine how quickly that happens.
This shift in dominant force matters practically. Early in a storm, infiltration rates tend to be high because capillary suction is strong. As the soil wets, the infiltration rate drops until it settles at a steady value governed mostly by the soil’s saturated hydraulic conductivity. Percolation below continues at a pace set by the deeper soil layers, which may be quite different from the topsoil.
What Controls Infiltration Rate
Several things determine whether water enters the ground quickly or pools on the surface. Soil texture is the big one. Sandy soils have large pore spaces and let water in fast; clay-heavy soils have tiny pores and resist entry. But the surface condition can override texture entirely. Compacted topsoil, biological soil crusts, or physical crusts formed by raindrop impact can seal the surface even if the soil underneath is porous.
Biological soil crusts in dryland environments illustrate this beautifully. These living mats of cyanobacteria, mosses, and lichens increase porosity and surface roughness, which should help infiltration. But they can also secrete water-repellent compounds and clog pores when wet, especially lichen-dominated crusts. During intense rainfall in semiarid landscapes, lichen crusts actually showed lower infiltration rates than cyanobacterial crusts because hydrophobicity masked the structural benefits of lichen cover.4Ecosystems. Soil Loss and Runoff in Semiarid Ecosystems: A Complex Interaction Between Biological Soil Crusts, Micro-topography, and Hydrological Drivers
Vegetation is a consistent positive force. A comprehensive review of bioretention systems found that plants reliably improved infiltration rates, with the effect scaling roughly from turfgrass (modest improvement) up through prairie grasses, shrubs, and trees (largest improvement). Species with thick, deep, or fleshy root systems performed best, and wind-induced movement of stems and stiff foliage helped prevent the surface from clogging.5PubMed. Supporting evidences for vegetation-enhanced stormwater infiltration in bioretention systems: a comprehensive review In heavy-textured soils, grass-planted plots had initial infiltration rates about 63% higher and steady-state rates about 38% higher than shrub-planted plots, with fine root length density emerging as the strongest predictor. The roots didn’t directly increase infiltration so much as they improved soil structure, aggregate stability, and porosity, which in turn let water in faster.6Soil and Tillage Research. Vegetation restoration and fine roots promote soil infiltrability in heavy-textured soils
What Controls Percolation Rate
Once water is in the ground, its downward journey depends on different factors than those at the surface. The layering of soils matters enormously. A permeable sandy topsoil sitting on top of a dense clay layer will infiltrate water quickly but percolate it slowly, because the clay acts as a barrier. In the Amazon basin, researchers found that daily percolation flux varied dramatically with landscape position: roughly 2.4 mm per day on clay-rich plateaus but over 6 mm per day in valleys where clay content was lowest.7Hydrological Processes. Soil water percolation and nutrient fluxes as a function of topographical, seasonal and soil texture variation in Central Amazonia, Brazil Topography and elevation contributed too, with lower-lying areas generally showing higher percolation.
Macropores, the larger channels created by root decay, earthworm burrows, and soil cracks, can dramatically accelerate percolation. These channels allow what hydrologists call preferential flow, where water bypasses the bulk of the soil matrix and moves rapidly downward through connected pathways. The characteristics of the macropore network directly determine how much fast transport occurs.8Hydrology and Earth System Sciences. Relations between macropore network characteristics and the degree of preferential solute transport This is one reason why simple models that assume water moves uniformly through the soil often fail. In irrigated fields, for example, researchers trying to predict deep percolation with standard uniform-flow equations could not simultaneously match soil moisture data and actual percolation volumes, because preferential pathways were carrying water past the soil matrix much faster than the equations allowed.9Vadose Zone Journal. Applicability of Richards’ equation models to predict deep percolation under surface irrigation
Preferential Flow and Why It Complicates Things
Preferential flow deserves its own discussion because it blurs the tidy distinction between infiltration and percolation. In reality, water doesn’t always move in neat fronts. It can shoot down a worm burrow from the surface to a depth of a meter while the soil a few centimeters to the side is still dry. This fast vertical transport can make infiltrated water appear deep in the profile long before you’d expect based on the soil’s average properties.
There’s a common assumption that preferential flow always shortens how long water and dissolved chemicals spend in contact with soil particles, reducing the soil’s ability to filter or absorb contaminants. That assumption turns out to be too simple. Research has shown that while rapid preferential flow right after a chemical application can push pollutants deep before the soil matrix has a chance to interact with them, the effect is not a universal feature of solute transport. Under many real-world conditions, preferential flow redistributes water in ways that don’t necessarily eliminate contact time with the soil matrix.10PubMed Central. A spectrum of preferential flow alters solute mobility in soils
When Infiltration Fails and Runoff Begins
If rain falls faster than the soil surface can absorb it, the excess becomes surface runoff. This is called infiltration-excess overland flow. There’s also saturation-excess flow, which happens when the soil is already full of water and simply can’t hold more, regardless of how fast rain is falling. Both types generate runoff, but they originate from different failures, one at the surface entry point and one from the subsurface being full. Modeling work has emphasized that accurately accounting for the comparison between actual rainfall intensity and the soil’s infiltration capacity is essential for predicting floods in a catchment, and that antecedent wetness often determines which type of runoff dominates.11Hydrological Processes. Modelling infiltration and infiltration excess: The importance of fast and local processes
For anyone managing land, this distinction is practical. If your problem is infiltration-excess runoff, improving the soil surface (reducing compaction, adding organic matter, maintaining vegetative cover) is the fix. If your problem is saturation-excess runoff, the underlying water table or subsurface drainage needs attention instead, because the surface is fine but the soil below is already at capacity.
How Scientists Measure Each Process
Infiltration and percolation are measured with different tools because they happen at different depths. Infiltration rate is typically measured at the soil surface using devices called infiltrometers. The double-ring infiltrometer is the workhorse: you push two concentric metal rings into the ground, fill both with water, and measure how quickly the water level in the inner ring drops. The outer ring exists to prevent water from spreading laterally, ensuring you’re measuring downward movement. These come in various sizes and configurations, with studies comparing 30-cm and 15-cm inner rings under constant-head and falling-head conditions.12Applied Turfgrass Science. Analysis of Double‐Ring Infiltration Techniques and Development of a Simple Automatic Water Delivery System Tension infiltrometers, which apply water at negative pressure to exclude macropore flow, can also be used alongside double-ring devices to tease apart how much infiltration occurs through the soil matrix versus through large channels.13Vadose Zone Journal. Determination of Hydraulic Properties in Sloping Landscapes from Tension and Double-Ring Infiltrometers
Percolation is harder to measure because it happens underground. Researchers use lysimeters, which are essentially buried containers that collect the water draining through a defined soil volume. A drainage lysimeter placed at depth captures the percolating water as it passes a particular horizon, allowing you to measure volume and chemistry. Lysimeter studies in cold environments like Québec have been used to track deep percolation dynamics through seasonal freeze-thaw cycles.14Hydrological Processes. Analysing Deep Percolation Dynamics: A Lysimeter‐Based Study in a Cold Environment Suction lysimeters, a variation that applies vacuum to extract water from unsaturated soil, are used when researchers need to sample the water itself, for instance to track how contaminants migrate downward, because determining the rate of pollutant migration to groundwater requires an independent measure of water percolation.15PubMed Central. Using Suction Lysimeters for Determining the Potential of Per- and Polyfluoroalkyl Substances to Leach from Soil to Groundwater: A Review
Why the Distinction Matters for Groundwater Recharge
Deep percolation is the primary pathway by which groundwater gets replenished in many landscapes. But not all infiltrated water becomes deep percolation. Plants intercept a large fraction through their roots, some water evaporates back to the atmosphere, and some is held by soil particles and never drains past the root zone. The amount that makes it through all these losses to reach the water table is what counts as recharge.
In irrigated agricultural valleys, deep percolation from flood irrigation is a measurable recharge source. Field observations in north-central New Mexico documented irrigation-driven deep percolation ranging from 0 to 157 mm per irrigation event in one field and 0 to 113 mm in another, depending on soil type. Each percolation event caused a transient rise in the shallow water table, confirming a direct connection between surface irrigation, soil water transport, and aquifer replenishment.16New Mexico Journal of Science: New Mexico’s Water Resources. Deep percolation and water table fluctuations in response to irrigation inputs: Field observations The U.S. Geological Survey developed a deep percolation model specifically to estimate long-term groundwater recharge from precipitation on a daily basis, underscoring how central this process is to water resource planning.17U.S. Geological Survey. Documentation of a deep percolation model for estimating ground-water recharge
How Farming Practices Alter Both Processes
Tillage changes both infiltration and percolation, though not always in the same direction. Conventional tillage (plowing, disking) breaks up the surface and initially increases infiltration by loosening the topsoil. But it also destroys macropore networks from roots and earthworms, which can reduce preferential percolation. No-till farming preserves those biological channels, and tracers applied to the surface tend to show up in deep soil water sooner under no-till than under tilled systems.18Vadose Zone Journal. Rainfall Intensification Enhances Deep Percolation and Soil Water Content in Tilled and No‐Till Cropping Systems of the US Midwest
Rainfall intensity adds another layer. In Midwest cropping systems, intensified rainfall (heavier bursts rather than gentle steady rain) increased deep percolation and deep soil water content regardless of whether the field was tilled or not. A surface-applied tracer appeared at 120 cm depth earlier under intensified rain by an average of about 6 days under no-till and roughly 74 days under tilled conditions, and less total tracer was recovered in the soil, meaning more of it had already drained below the monitoring zone.18Vadose Zone Journal. Rainfall Intensification Enhances Deep Percolation and Soil Water Content in Tilled and No‐Till Cropping Systems of the US Midwest As climate change is projected to make rainfall events more intense in many regions, this has direct implications for both nutrient leaching and groundwater recharge under different farming practices.
Soil compaction from heavy machinery complicates things further. In experiments comparing conventional and conservation tillage, compaction loads affected solute movement differently depending on the tillage system. Conservation-tilled soil resisted deformation, while conventionally tilled soil changed its transport properties under the same load.19Soil and Tillage Research. Tillage effects on soil strength and solute transport
Fire’s Dramatic Effect on Infiltration
Wildfire is one of the most dramatic disruptors of infiltration in natural landscapes. High-intensity fires volatilize organic matter at the soil surface, break down soil aggregate structure, increase bulk density, and create a water-repellent layer on or near the surface by coating soil particles with hydrophobic compounds.20Journal of Forestry Research. A review of the effects of forest fire on soil properties The result is a surface that actively resists water entry.
Field measurements in western Montana the year after a wildfire found that burned sites had 88% soil water repellency and only about 10% ground cover remaining. Infiltration capacity dropped to around 30 mm over the measurement period, while sediment concentrations spiked dramatically. The loss of protective ground cover turned out to be even more important than the repellent soil layer itself for generating erosion, though both contributed.21Catena. Infiltration and interrill erosion rates after a wildfire in western Montana, USA
A synthesis across multiple fire studies found that burning reduces the capillary forces that normally help pull water into soil. In unburned soils, gravity and capillarity tend to contribute to infiltration in a balanced way. Fire disrupts that balance, making gravity the dominant remaining force and causing water to pond on the surface much faster during rain. Ash deposits can temporarily mask this effect because ash itself is quite permeable and absorptive, but once the ash washes away or gets compacted, the underlying repellent soil is exposed.22Hydrological Processes. Synthesis of soil‐hydraulic properties and infiltration timescales in wildfire‐affected soils This is why post-fire landscapes are so vulnerable to flash floods, sometimes for years: infiltration is crippled at the surface, and without that entry step, percolation deeper in the profile stops as well.
Contaminants Riding the Water Down
Both infiltration and percolation serve as the conveyor belt for anything dissolved or suspended in water. Pesticides, fertilizers, and emerging contaminants like microplastics all travel downward through these processes. Modeling of microplastic migration has shown that increasing irrigation flux and irrigated area significantly accelerates movement of particles from the soil surface down into deep saturated aquifers. Interestingly, the relationship between particle size and migration speed is not linear: particles smaller than about 20 micrometers reach peak velocities at around 8 micrometers, while larger particles slow down because the soil matrix physically hinders them. Different plastic types behave differently too, with denser polymers showing less downward flux into groundwater.23ACS Publications. Modeling of Microplastics Migration in Soil and Groundwater: Insights into Dispersion and Particle Property Effects
The practical takeaway here is that anything you do to speed up infiltration (say, by improving soil structure on a farm) also potentially speeds up contaminant percolation to groundwater. Conversely, compacted or crusted soils that limit infiltration may generate more surface runoff pollution but offer some unintentional protection to groundwater simply because less water and fewer dissolved chemicals are making the downward trip. Soil and water management always involves trade-offs between surface and subsurface concerns, and the infiltration-percolation pathway is the axis along which those trade-offs play out.
Modeling the Whole Process
Hydrologists have spent two centuries building mathematical models of infiltration, and a historical review cataloged 138 distinct models, spanning simple empirical curve fits to physics-based approaches rooted in flow equations.24Vadose Zone Journal. Review of conceptual and empirical approaches to characterize infiltration Some of these models handle the surface entry well but break down when asked to predict deep percolation through layered, cracked, or biologically active soil. Others capture deep drainage but require soil property data that’s expensive and time-consuming to collect. The ongoing tension in the field is between accuracy (complex models that need many inputs) and practicality (simple models that can be applied widely with available data). For percolation specifically, the challenge of accounting for preferential flow through macropores remains an active research frontier, since that phenomenon defies the assumption of uniform flow that underpins most standard models.