aiwiki.page
English
Geography / soil-erosion

Soil Erosion

Soil erosion is the detachment and movement of soil by water, wind, and cultivation, reshaping landscapes and affecting land productivity and water quality.

31 keywords22 linked from5 not yet writtenWritten by AI
ErosionWeatheringDrainage BasinGeomorphologyInfiltrationSediment Transpo…SaltationAeolian Processe…Soil Erosi…

Soil erosion is the detachment and transport of soil material by flowing water, raindrop impact, wind, and certain human activities, especially cultivation. It is a specific form of erosion, distinguished by its effects on the biologically active surface layer that supports terrestrial vegetation and agriculture. Erosion occurs naturally, but clearing vegetation, disturbing soil, and concentrating runoff can accelerate it far beyond the rate at which new soil develops. The distinction between natural and accelerated erosion is therefore central to understanding its environmental significance.

Erosion redistributes soil rather than simply making it disappear: material removed from one location may accumulate farther downslope, enter a river, settle in a reservoir, or travel through the atmosphere as dust. Its consequences depend on both the source and the destination. Loss of fertile surface soil can reduce agricultural productivity, while transported sediment can alter aquatic habitats, carry nutrients and contaminants, and shorten the useful life of water-storage infrastructure. Understanding soil erosion consequently requires attention to physical processes, land management, measurement scale, and the connections between hillslopes and river systems.

Physical Processes and Sediment Budgets

Soil erosion consists of three closely connected processes: detachment, transport, and deposition. Detachment occurs when an applied force exceeds the resistance holding particles or aggregates in place. Raindrops can break aggregates apart, flowing water can shear particles from a surface, and wind can dislodge dry grains. Transport then moves the detached material, while deposition occurs when the carrying medium loses sufficient energy or when particles encounter an obstacle. These processes can occur within centimeters of one another, so visible soil movement does not necessarily imply that material has left a field or watershed.

The resistance of soil depends partly on soil structure, the arrangement of mineral particles and organic matter into aggregates and pore spaces. Roots, fungal filaments, organic binding agents, and some mineral constituents help stabilize aggregates. Conversely, repeated mechanical disturbance, low organic matter, and rapid wetting of susceptible dry aggregates can weaken them. Individual clay particles are small enough to remain suspended for long periods once detached, but cohesive clay-rich soil may resist initial detachment more strongly than loose sand or silt. Particle size alone therefore does not determine erodibility.

Soil formation and erosion operate simultaneously. Weathering converts parent material into smaller particles and chemically altered products, while biological activity and organic inputs develop recognizable soil horizons. A local soil budget can be expressed schematically as

∂Ms∂t=P+D−E,\frac{\partial M_s}{\partial t}=P+D-E,

where MsM_s is soil mass per unit area, PP is soil production, DD is deposition, and EE is removal. This simplified expression excludes some processes, such as dissolved losses and deliberate soil addition, but clarifies why persistent erosion exceeding production progressively thins a soil profile.

A central distinction is between gross erosion and sediment yield. Gross erosion describes material detached or displaced within a specified area; sediment yield is the mass exported across its boundary, usually measured at a stream outlet. Between these quantities lies storage on footslopes, floodplains, channel beds, and behind barriers. In a drainage basin, sediment removed during one storm may remain stored for decades before later floods remobilize it. The resulting sediment budget is therefore both spatial and historical.

The boundary of soil erosion is not identical to that of all landscape erosion. Riverbank retreat, landslides, and gullying may remove soil together with deeper sediment or weathered rock. These processes can dominate sediment supply in some settings even where surface erosion is modest. Geomorphology treats them as parts of a connected landscape system, whereas agricultural assessments often concentrate on the cultivated soil layer. Clear statements about the material, process, area, and time interval being measured are essential when comparing erosion estimates.

Erosion by Rainfall and Flowing Water

Water erosion commonly begins with rainfall striking exposed ground. A falling drop transfers kinetic energy to the surface, dislodging particles and splashing them short distances. On a slope, repeated impacts produce a net downslope movement. Drop impact also breaks aggregates and can redistribute fine particles into surface pores, forming a seal that reduces infiltration. More rainfall then becomes surface runoff, creating a feedback in which detachment helps establish the conditions for further transport.

Runoff develops through more than one hydrological pathway. Infiltration-excess runoff occurs when rainfall intensity exceeds the rate at which water can enter the soil. It is common on compacted, crusted, or sparsely vegetated surfaces during intense storms. Saturation-excess runoff occurs when the soil becomes saturated and additional water cannot be stored, even if rainfall intensity is relatively low. This is frequent near streams, in convergent hollows, or where shallow impermeable layers restrict drainage. Antecedent wetness can therefore be as important as storm intensity.

Interrill erosion combines raindrop detachment with transport through shallow, relatively unconfined surface flow. It is often described as sheet erosion, although actual flow usually follows a fine network of irregular pathways rather than a uniform sheet. Because removal can be spatially diffuse, substantial cumulative loss may occur without conspicuous channels. Pedestals beneath stones, exposed roots, surface crusting, and the gradual disappearance of darker topsoil can reveal its effects.

As runoff concentrates, it cuts small channels called rills. Rills are generally shallow enough to be removed by normal cultivation, but their transport capacity may greatly exceed that of surrounding interrill areas. Larger channels, or gullies, cannot normally be erased by routine tillage. Some begin at headcuts that migrate upslope as flowing water undercuts a step and the unsupported material collapses. Ephemeral gullies recur in the same topographic hollows after being filled by cultivation, whereas permanent gullies remain as enduring drainage features.

Flow detachment is often related to boundary shear stress:

τ=ρgRhS,\tau=\rho g R_h S,

where ρ\rho is water density, gg is gravitational acceleration, RhR_h is hydraulic radius, and SS is the energy slope. Detachment becomes more likely when this stress exceeds a soil-specific threshold, although turbulence, roots, stones, and changing soil strength complicate the relationship. Sediment transport also depends on whether flow has enough available energy to carry additional material: a sediment-laden stream can behave differently from clear water at the same discharge.

Subsurface pathways create additional forms of erosion. Water moving through cracks, animal burrows, or dispersive soil can enlarge tunnels, a process known as soil piping. Collapse of these tunnels may initiate gullies even where surface runoff initially appears limited. Along streams, bank saturation, toe scour, and changes in water level can combine to cause block failure. Water erosion thus spans processes from the impact of individual drops to the collapse of entire soil masses.

Wind Erosion and Soil Redistribution by Tillage

Wind erosion is most effective where the surface is dry, loose, poorly protected, and exposed across a sufficiently long distance. It occurs in agricultural fields as well as deserts and semiarid rangelands. The relevant force is the shear stress exerted by moving air near the ground, rather than wind speed at an arbitrary height alone. Surface roughness, clod size, moisture, vegetation, and crust strength determine the threshold at which particles begin to move.

Three modes of movement interact. Larger grains roll or slide in surface creep; intermediate grains move in short, repeated hops called saltation; and fine particles enter suspension. Saltating grains are particularly important because their impacts can eject additional material from the surface. Once this bombardment begins, substantial erosion can continue at wind speeds lower than those initially needed to start movement. Abrasion by moving grains also damages seedlings and breaks fragile aggregates, generating additional dust.

Fine dust can travel far beyond the source area, while coarser material usually accumulates closer to it. Aeolian processes therefore sort soil by particle size and can selectively remove fine mineral particles, organic matter, and associated nutrients. Deposited sand may bury seedlings, obstruct drainage channels, or form drifts along fences and roads. Suspended dust becomes an atmospheric aerosol, influencing radiation and supplying minerals to distant ecosystems. Not all atmospheric mineral dust originates from agricultural soil erosion; dry lake beds and natural desert surfaces are also major sources.

Cultivation produces a different, mechanically driven form of redistribution known as tillage erosion. Ploughs and other implements move soil in the direction of operation, but the amount displaced varies with slope, working depth, speed, implement design, and soil condition. Over repeated operations, more material commonly moves away from convex slope crests than is brought back to them. Concave positions and lower slope margins tend to accumulate the displaced soil.

Tillage erosion can be severe even where little sediment leaves the field. A convex hilltop may lose its dark surface horizon and expose paler subsoil, while a nearby hollow develops a thick accumulated profile. Because these changes arise within the cultivated area, monitoring only stream sediment misses them. Subsequent water erosion can further redistribute the material, so the effects of tillage and runoff may reinforce one another without being physically identical.

Other human operations can also export soil. Root-crop harvesting removes soil adhering to potatoes, sugar beet, and similar crops; earthmoving transfers soil during construction; and traffic can carry material away from disturbed sites. These losses are sometimes recorded separately from erosion by water and wind. Their inclusion depends on the purpose of the assessment, but they illustrate why a complete agricultural soil budget cannot always be inferred from rainfall and runoff alone.

Environmental Controls and Human Acceleration

The amount of erosion produced by a storm depends on both the force applied and the susceptibility of the surface. Rainfall erosivity reflects characteristics such as kinetic energy, intensity, and timing, whereas soil erodibility describes the tendency of a soil to detach and be transported under specified conditions. High annual precipitation does not automatically imply severe erosion: frequent gentle rain on well-protected ground may cause less damage than a few intense storms falling on a bare, recently cultivated slope.

Topography influences the depth, velocity, and concentration of overland flow. Longer uninterrupted slopes generally allow runoff to accumulate, while steeper gradients increase its potential energy. Nevertheless, slope length and steepness alone are insufficient. A convergent hollow can gather water from a broad contributing area, whereas a divergent ridge sheds it. Roads, field boundaries, drainage ditches, and terraces can interrupt natural flow paths or create new ones. Small alterations in these connections can strongly affect whether sediment remains on a hillslope or reaches a stream.

Vegetation protects soil through several mechanisms. Leaves and stems intercept rainfall, surface litter absorbs drop impact, and ground cover slows runoff and traps sediment. Roots bind soil and help create pore networks, while organic inputs support aggregation. A tall canopy without litter or understory does not necessarily provide complete protection: water dripping from leaves can still strike exposed ground with appreciable energy. The continuity of cover close to the surface is often more important than vegetation height alone.

Agricultural practices alter these controls seasonally. A field may be well protected when a crop canopy is mature but highly vulnerable between harvest and establishment of the next crop. Overgrazing can reduce plant cover and compact frequently used areas, particularly near watering points. Forest roads and timber extraction routes can concentrate runoff even when the surrounding forest floor remains relatively stable. Construction sites are especially susceptible because they combine stripped vegetation, disturbed soil, stockpiles, and artificially connected drainage.

Fire can produce abrupt changes by removing protective cover and altering soil properties. Some burned soils develop water-repellent layers, though the response varies with fire severity, soil characteristics, and subsequent rainfall. Intense storms following fire can generate rapid surface runoff and large sediment pulses. These responses should be distinguished from debris flows, which involve concentrated mixtures of water and sediment and may entrain material well below the soil surface.

Climate change affects erosion through rainfall intensity, drought, vegetation, fire, and agricultural calendars. More intense rainfall can increase erosive power, but changes in annual rainfall alone do not determine the outcome. Drought can leave soils sparsely covered and vulnerable to wind or to subsequent heavy rain. Regional projections remain sensitive to land management: similar climatic changes can produce different erosion responses under continuous cover, exposed cultivation, or expanding urban development.

Effects on Soil, Water, and Ecosystems

The most direct on-site effect is the removal of surface soil, which often contains more organic matter, nutrients, biological activity, and plant-available water than the underlying horizons. As erosion reduces rooting depth, plants may encounter dense subsoil, gravel, salinity, or chemically unfavorable material. Yield effects are consequently not proportional to the mass lost alone. Removing a thin layer from a shallow soil over rock may be more consequential than the same loss from a deep, uniform deposit.

Erosion also changes soil properties without immediately exhausting soil depth. Selective removal of fine particles can leave a coarser surface, while surface sealing reduces water entry and increases runoff. Exposed subsoil may be harder to cultivate or less able to support seedling emergence. Depositional areas sometimes gain nutrients and moisture-storage capacity, but excessive burial can damage crops and natural vegetation. Redistribution can therefore create strong productivity contrasts within a single field.

Transported sediment affects rivers, lakes, reservoirs, and wetlands. Fine suspended particles increase turbidity and reduce light penetration, influencing aquatic plant growth. Deposition can cover fish spawning gravels, alter channel dimensions, and bury bottom-dwelling organisms. Sediment also carries substances attached to mineral and organic surfaces, especially phosphorus and some pesticides or metals. Nutrient delivery can contribute to eutrophication, although dissolved nutrient losses follow additional pathways that cannot be inferred from sediment measurements alone.

Sediment is not intrinsically undesirable. Rivers require sediment to maintain floodplains, sand bars, and many downstream landforms. A river delta may lose elevation relative to the sea if upstream dams and river engineering sharply reduce sediment delivery. The environmental problem is therefore often an altered sediment regime—too much, too little, or delivery at the wrong time and place—rather than sediment presence itself. Reservoir sedimentation illustrates the contrast: material valuable to downstream land building can reduce upstream water-storage capacity when trapped behind a dam.

Soil erosion also interacts with carbon cycling. It removes organic carbon from source areas, exposes some carbon to decomposition, and buries some in depositional environments where preservation may improve. Vegetation can replace part of the carbon lost from eroding sites, but replacement depends on productivity and nutrient availability. Because transport, oxidation, burial, and new plant inputs operate simultaneously, displaced soil carbon cannot simply be counted as an equivalent release of carbon dioxide to the atmosphere.

Broader effects include damage to roads and drainage systems, reservoir maintenance costs, loss of archaeological deposits, and changes in biodiversity. Wind-blown material can impair visibility and carry fine particles across populated areas. Economic accounting is difficult because losses occur at different locations and times: a landholder may experience declining soil quality, while downstream communities bear treatment or dredging costs. Estimates of erosion damage therefore depend on which effects are included and how long-term soil degradation is valued.

Measuring Erosion across Space and Time

Direct measurements are often made on bounded runoff plots with known dimensions. Water and sediment leaving the plot are collected after storms, allowing soil loss to be expressed as mass per area per event or year. Standardized plots have been essential for comparing soils and management practices, but their boundaries alter natural flow connections. Small plots also cannot represent all processes operating in large fields, such as gullies, channel storage, or sediment supplied from roads.

Surface change can be measured with erosion pins, repeated leveling, photogrammetry, and laser scanning. These methods are particularly useful for rills, gullies, and banks, where measurable changes in elevation or volume occur. Converting volume into mass requires an appropriate bulk density. The smallest detectable change matters: a survey capable of documenting a collapsing gully wall may not resolve a few millimeters of diffuse annual soil loss across an irregular field.

At catchment outlets, sediment monitoring combines water discharge with sediment concentration. Suspended load can be estimated as

L=∫Q(t)Cs(t) dt,L=\int Q(t)C_s(t)\,dt,

where Q(t)Q(t) is discharge and Cs(t)C_s(t) is suspended-sediment concentration. Infrequent sampling can miss a large fraction of transport because concentrations change rapidly during storms. Automated samplers and calibrated turbidity sensors improve temporal coverage, but the relationship between turbidity and sediment concentration changes with particle size, color, and organic content. Bed load requires separate measurement or estimation.

Fallout radionuclides provide evidence of longer-term soil redistribution. Caesium-137, deposited widely during atmospheric nuclear weapons testing, binds strongly to many soil particles. An inventory lower than that at a suitable undisturbed reference site can indicate net erosion, while a larger inventory can indicate deposition. Unsupported lead-210 offers another tracer, and beryllium-7 is useful over shorter timescales. These techniques depend on assumptions about initial deposition, soil mixing, radioactive decay, and subsequent redistribution.

Sediment fingerprinting uses properties such as geochemistry, magnetic characteristics, isotopes, or organic markers to distinguish likely source areas. It can help separate topsoil from subsoil contributions or agricultural land from channel banks. However, source signatures must be distinguishable, and selective transport can modify them. Statistical source attribution is therefore not equivalent to directly observing every contributing erosion event.

Spatial observation increasingly combines remote sensing, a digital elevation model, and a geographic information system. Satellite imagery can identify bare ground, vegetation seasonality, and large gullies, while airborne or drone surveys resolve finer features. These observations usually measure indicators or surface change rather than soil mass loss directly. Comparisons between studies must therefore distinguish modeled erosion, observed redistribution, and measured sediment export, as well as the period represented by each estimate.

Prediction Models and the Interpretation of Erosion Rates

Erosion models range from empirical equations fitted to observations to process-based systems that represent runoff, detachment, transport, and deposition. Their usefulness depends on the question being asked. A model designed to compare average sheet and rill erosion under alternative cropping systems is not necessarily suitable for forecasting gully growth during a single extreme storm. Model complexity also does not guarantee accuracy when essential inputs are poorly known.

The Universal Soil Loss Equation, developed in the United States and consolidated by Walter H. Wischmeier and Dwight D. Smith in a 1965 handbook, estimates long-term average soil loss as

A=RKLSCP.A=RKLSCP.

Here AA is average soil loss, RR rainfall erosivity, KK soil erodibility, LL slope length, SS slope steepness, CC cover and management, and PP support practices. The factors summarize relationships established largely from runoff-plot observations. The equation became influential because it translated extensive field experiments into a practical framework for comparing land-management alternatives.

The Revised Universal Soil Loss Equation, commonly called RUSLE, refined these relationships and their calculation. The USLE family primarily addresses sheet and rill erosion; it does not by itself represent all gully, streambank, wind, or tillage erosion. Nor does a calculated hillslope loss automatically equal sediment delivered to a river. Applying a sediment delivery ratio can estimate export, but that ratio varies with connectivity, storage, watershed size, and event conditions rather than being a universal constant.

Process-based models, including the Water Erosion Prediction Project, or WEPP, simulate more of the physical sequence linking weather, infiltration, runoff, detachment, and deposition. Wind-erosion models use different representations of surface condition, aerodynamic thresholds, and particle movement. Distributed catchment models can connect hillslope sources with stream channels, but require information on soil, topography, vegetation, management, and hydrology. Uncertainty in these inputs can be as important as the mathematical representation of erosion itself.

Rates are commonly reported in tonnes per hectare per year. To translate mass loss into an equivalent depth, bulk density must be specified. At a bulk density of 1.31.3 tonnes per cubic meter, a loss of 1313 tonnes per hectare corresponds to about 11 millimeter of soil. This is an equivalent average thickness, not necessarily a uniform lowering of the ground surface. The annual average can also conceal concentration in a few exceptional storms.

“Tolerable” soil-loss rates used in conservation planning are management thresholds, not universal ecological constants. They depend on soil depth, expected productivity, time horizon, and assumptions about soil renewal; they may not protect downstream water quality. Global erosion totals are similarly sensitive to what is counted. In 2017, Pasquale Borrelli and colleagues published a model-based estimate of about 35.935.9 billion tonnes of annual water-driven soil displacement for 2012. Such figures describe modeled processes within a defined domain, not a direct weighing of all soil lost worldwide, and should not be compared uncritically with estimates that include different sources.

Prevention, Land Management, and Restoration

Soil conservation seeks to reduce detachment, limit transport, and maintain soil functions. The most widely applicable principle is protection of the surface. Crop residues, mulch, living cover, and persistent vegetation absorb raindrop energy and reduce the exposure of loose soil to wind. Their effectiveness depends on coverage during erosive periods, not simply on annual biomass production. A field left bare during the season of strongest storms may remain vulnerable despite supporting a dense crop later in the year.

Conservation tillage and no-till systems reduce disturbance and can retain protective residues. They also reduce tillage-driven redistribution when fewer or shallower operations move less soil. Their hydrological effects vary with soil texture, compaction, drainage, climate, and the time since adoption. Reduced disturbance does not automatically eliminate runoff, nutrient loss, or concentrated-flow erosion. Crop rotation and cover crops can provide complementary benefits by extending ground cover and adding roots and organic inputs.

Contour cultivation and strip cropping interrupt downslope flow, while terraces shorten effective slope length and create more manageable gradients. These measures must accommodate water safely: overtopping or poorly designed outlets can concentrate flow and produce severe local erosion. Grassed waterways stabilize natural drainage routes, and vegetated buffer strips intercept sediment before it reaches streams. Buffers are less effective when runoff bypasses them through a ditch or arrives as highly concentrated flow.

Wind control depends on keeping soil covered, maintaining stable aggregates, and reducing near-surface wind speed. Shelterbelts and windbreaks create protected zones downwind, while standing crop stubble disrupts airflow close to the ground. Their performance varies with height, porosity, orientation, and gaps. Shortening the unsheltered length of a field can limit the development of saltation, although protection declines where vegetation fails or winds approach from unprotected directions.

Gully restoration usually requires treatment of both the channel and its contributing area. Stabilizing a headcut without reducing concentrated runoff may only shift erosion elsewhere. Grade-control structures, check dams, revegetation, and reshaped channels can reduce local instability when matched to expected flows and sediment supply. Accumulated sediment behind structures creates storage, but that storage is finite and may be remobilized if structures fail. On construction sites, sediment traps similarly complement rather than replace control of erosion at its source.

Management at watershed scale must account for connections among fields, roads, channels, and downstream receptors. The most actively eroding site is not always the largest contributor to river sediment if it is isolated by depositional areas. Conversely, a small disturbed area directly connected to a stream may have a disproportionate effect. Conservation performance is therefore evaluated through both local soil retention and downstream outcomes, with continued monitoring needed to distinguish an immediate reduction in surface disturbance from slower changes in channel and floodplain sediment storage.

Historical Development and Regional Patterns

Concern about soil loss predates modern erosion science. Terracing, stone walls, managed vegetation, and irrigation works in many agricultural societies redistributed water and sediment long before quantitative monitoring developed. Some practices deliberately retained transported soil, showing that erosion and deposition were managed together. Historical interpretations nevertheless require caution: abandoned terraces or sediment-filled valleys do not by themselves establish a single cause, because climate variability, population change, land tenure, warfare, and changing agriculture can produce overlapping evidence.

In the United States, the Dust Bowl of the 1930s made wind erosion a major national issue. Severe drought coincided with extensive cultivation and loss of protective vegetation across parts of the Great Plains. Dust storms damaged farms and carried soil far beyond its source region. The Soil Erosion Service was established in 1933, followed by the Soil Conservation Service in 1935. Hugh Hammond Bennett was a central figure in promoting soil conservation and organizing federal responses. The agency became the Natural Resources Conservation Service in 1994.

Water-erosion research expanded through networks of experimental plots and watersheds during the twentieth century. These records connected rainfall intensity, cropping, slope, and management to measured soil loss and supported the development of empirical prediction methods. Later work increasingly examined entire catchments, using sediment budgets, radionuclide tracers, and repeated surveys to distinguish source areas from storage zones. This shift helped explain why reducing erosion on agricultural plots does not always produce an immediate, proportional decline in sediment at a distant river outlet.

The Loess Plateau of China is a prominent example of highly erodible terrain affected by long-term cultivation. Thick deposits of wind-blown silt, steep dissected slopes, intense summer rainfall, and extensive gullies have supplied large quantities of sediment to the Yellow River. Terracing, revegetation, check dams, and changes in land use have altered erosion and sediment delivery. China’s Grain for Green program began in 1999 and expanded the conversion of some cultivated slopes to forest or grassland. Changes in river sediment reflect these measures together with reservoirs, hydrological variability, and other interventions.

In the Ethiopian Highlands, steep cultivated slopes, seasonal rainfall, grazing pressure, and limited protective cover create combinations of sheet, rill, and gully erosion. Soil and water conservation measures include stone bunds, terraces, and managed vegetation, but their outcomes depend on maintenance and local hydrology. In Mediterranean landscapes, long dry periods followed by intense rain make seasonally bare ground vulnerable, particularly in sloping orchards and vineyards. In humid tropical regions, erosion can increase sharply after clearing when exposed soil receives heavy rain, even though intact forest nearby remains well protected.

Present research increasingly treats erosion as a coupled environmental and land-use process rather than a single rate assigned to a soil type. Important unresolved questions concern extreme events, gully initiation, sediment connectivity, carbon redistribution, and the transfer of models between regions with unequal monitoring coverage. Long-term conservation experiments remain especially valuable because soil profiles, vegetation, and sediment stores respond over different timescales. A short record may capture an exceptional storm or a temporary management effect without revealing the trajectory of soil depth and productivity over decades.