Hubbry Logo
LandslideLandslideMain
Open search
Landslide
Community hub
Landslide
logo
8 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Landslide
Landslide
Landslide
View on Wikipedia
from Wikipedia
A landslide near Cusco, Peru, in 2018
A NASA model has been developed to look at how potential landslide activity is changing around the world.
Animation of a landslide in San Mateo County, California
Landslips
Noire River (Rivière Noire), Saint-Alban landslide 1894, Quebec, Canada[1][2]

Landslides, also known as landslips, rockslips or rockslides,[3][4][5] are several forms of mass wasting that may include a wide range of ground movements, such as rockfalls, mudflows, shallow or deep-seated slope failures and debris flows.[6] Landslides occur in a variety of environments, characterized by either steep or gentle slope gradients, from mountain ranges to coastal cliffs or even underwater,[7] in which case they are called submarine landslides.

Gravity is the primary driving force for a landslide to occur, but there are other factors affecting slope stability that produce specific conditions that make a slope prone to failure. In many cases, the landslide is triggered by a specific event (such as heavy rainfall, an earthquake, a slope cut to build a road, and many others), although this is not always identifiable.

Landslides are frequently made worse by human development (such as urban sprawl) and resource exploitation (such as mining and deforestation). Land degradation frequently leads to less stabilization of soil by vegetation.[8] Additionally, global warming caused by climate change and other human impact on the environment, can increase the frequency of natural events (such as extreme weather) which trigger landslides.[9] Landslide mitigation describes the policy and practices for reducing the risk of human impacts of landslides, reducing the risk of natural disaster.

Causes

[edit]
Main article: Causes of landslides
The Mameyes Landslide, in the Mameyes neighborhood of barrio Portugués Urbano in Ponce, Puerto Rico, was caused by extensive accumulation of rains and, according to some sources, lightning. It buried more than 100 homes.
The landslide at Surte in Sweden, 1950. It was a quick clay slide that killed one person.

Landslides occur when the slope (or a portion of it) undergoes some processes that change its condition from stable to unstable. This is essentially due to a decrease in the shear strength of the slope material, an increase in the shear stress borne by the material, or a combination of the two. A change in the stability of a slope can be caused by a number of factors, acting together or alone. Natural causes of landslides include:

  • increase in water content (loss of suction) or saturation by rain water infiltration, snow melting, or glaciers melting;[10]
  • rising of groundwater or increase of pore water pressure (e.g. due to aquifer recharge in rainy seasons, or by rain water infiltration);[11]
  • increase of hydrostatic pressure in cracks and fractures;[11][12]
  • loss or absence of vertical vegetative structure, soil nutrients, and soil structure (e.g. after a wildfire);[13]
  • erosion of the top of a slope by rivers or sea waves;[14]
  • physical and chemical weathering (e.g. by repeated freezing and thawing, heating and cooling, salt leaking in the groundwater or mineral dissolution);[15][16][17]
  • ground shaking caused by earthquakes, which can destabilize the slope directly (e.g., by inducing soil liquefaction) or weaken the material and cause cracks that will eventually produce a landslide;[12][18][19]
  • volcanic eruptions;[20]
  • changes in pore fluid composition;[21]
  • changes in temperature (seasonal or induced by climate change).[22][23]

Landslides are aggravated by human activities, such as:

Types

[edit]
Main article: Landslide classification
Types of landslide

Hungr-Leroueil-Picarelli classification

[edit]

In traditional usage, the term landslide has at one time or another been used to cover almost all forms of mass movement of rocks and regolith at the Earth's surface. In 1978, geologist David Varnes noted this imprecise usage and proposed a new, much tighter scheme for the classification of mass movements and subsidence processes.[26] This scheme was later modified by Cruden and Varnes in 1996,[27] and refined by Hutchinson (1988),[28] Hungr et al. (2001),[29] and finally by Hungr, Leroueil and Picarelli (2014).[6] The classification resulting from the latest update is provided below.

Type of movement Rock Soil
Fall Rock/ice fall Boulder/debris/silt fall
Topple Rock block topple Gravel/sand/silt topple
Rock flexural topple
Slide Rock rotational slide Clay/silt rotational slide
Rock planar slide Clay/silt planar slide
Rock wedge slide Gravel/sand/debris slide
Rock compound slide Clay/silt compound slide
Rock irregular slide
Spread Rock slope spread Sand/silt liquefaction spread
Sensitive clay spread
Flow Rock/ice avalanche Sand/silt/debris dry flow
Sand/silt/debris flowslide
Sensitive clay flowslide
Debris flow
Mud flow
Debris flood
Debris avalanche
Earthflow
Peat flow
Slope deformation Mountain slope deformation Soil slope deformation
Rock slope deformation Soil creep
Solifluction
Note: the words in italics are placeholders. Use only one.

Under this classification, six types of movement are recognized. Each type can be seen both in rock and in soil. A fall is a movement of isolated blocks or chunks of soil in free-fall. The term topple refers to blocks coming away by rotation from a vertical face. A slide is the movement of a body of material that generally remains intact while moving over one or several inclined surfaces or thin layers of material (also called shear zones) in which large deformations are concentrated. Slides are also sub-classified by the form of the surface(s) or shear zone(s) on which movement happens. The planes may be broadly parallel to the surface ("planar slides") or spoon-shaped ("rotational slides"). Slides can occur catastrophically, but movement on the surface can also be gradual and progressive. Spreads are a form of subsidence, in which a layer of material cracks, opens up, and expands laterally. Flows are the movement of fluidised material, which can be both dry or rich in water (such as in mud flows). Flows can move imperceptibly for years, or accelerate rapidly and cause disasters. Slope deformations are slow, distributed movements that can affect entire mountain slopes or portions of it. Some landslides are complex in the sense that they feature different movement types in different portions of the moving body, or they evolve from one movement type to another over time. For example, a landslide can initiate as a rock fall or topple and then, as the blocks disintegrate upon the impact, transform into a debris slide or flow. An avalanching effect can also be present, in which the moving mass entrains additional material along its path.

Flows

[edit]

Slope material that becomes saturated with water may produce a debris flow or mud flow. However, also dry debris can exhibit flow-like movement.[30] Flowing debris or mud may pick up trees, houses and cars, and block bridges and rivers causing flooding along its path. This phenomenon is particularly hazardous in alpine areas, where narrow gorges and steep valleys are conducive of faster flows. Debris and mud flows may initiate on the slopes or result from the fluidization of landslide material as it gains speed or incorporates further debris and water along its path. River blockages as the flow reaches a main stream can generate temporary dams. As the impoundments fail, a domino effect may be created, with a remarkable growth in the volume of the flowing mass, and in its destructive power.

The Costa della Gaveta earthflow in Potenza, Italy. Even though it moves at a rate of just a few millimeters per year[15] and is hardly visible, this landslide causes progressive damage to the national road, the national highway, a flyover, and several houses that are built on it.
A rock slide in Guerrero, Mexico

An earthflow is the downslope movement of mostly fine-grained material. Earthflows can move at speeds within a very wide range, from as low as 1 mm/yr[15][16] to many km/h. Though these are a lot like mudflows, overall they are more slow-moving and are covered with solid material carried along by the flow from within. Clay, fine sand and silt, and fine-grained, pyroclastic material are all susceptible to earthflows. These flows are usually controlled by the pore water pressures within the mass, which should be high enough to produce a low shearing resistance. On the slopes, some earthflow may be recognized by their elongated shape, with one or more lobes at their toes. As these lobes spread out, drainage of the mass increases and the margins dry out, lowering the overall velocity of the flow. This process also causes the flow to thicken. Earthflows occur more often during periods of high precipitation, which saturates the ground and builds up water pressures. However, earthflows that keep advancing also during dry seasons are not uncommon. Fissures may develop during the movement of clayey materials, which facilitate the intrusion of water into the moving mass and produce faster responses to precipitation.[31]

A rock avalanche, sometimes referred to as sturzstrom, is a large and fast-moving landslide of the flow type. It is rarer than other types of landslides but it is often very destructive. It exhibits typically a long runout, flowing very far over a low-angle, flat, or even slightly uphill terrain. The mechanisms favoring the long runout can be different, but they typically result in the weakening of the sliding mass as the speed increases.[32][33][34] The causes of this weakening are not completely understood. Especially for the largest landslides, it may involve the very quick heating of the shear zone due to friction, which may even cause the water that is present to vaporize and build up a large pressure, producing a sort of hovercraft effect.[35] In some cases, the very high temperature may even cause some of the minerals to melt.[36] During the movement, the rock in the shear zone may also be finely ground, producing a nanometer-size mineral powder that may act as a lubricant, reducing the resistance to motion and promoting larger speeds and longer runouts.[37] The weakening mechanisms in large rock avalanches are similar to those occurring in seismic faults.[34]

Landslide in Peru after flash flooding in 2013.

Slides

[edit]

Slides can occur in any rock or soil material and are characterized by the movement of a mass over a planar or curvilinear surface or shear zone.

A debris slide is a type of slide characterized by the chaotic movement of material mixed with water and/or ice. It is usually triggered by the saturation of thickly vegetated slopes which results in an incoherent mixture of broken timber, smaller vegetation and other debris.[31] Debris flows and avalanches differ from debris slides because their movement is fluid-like and generally much more rapid. This is usually a result of lower shear resistances and steeper slopes. Typically, debris slides start with the detachment of large rock fragments high on the slopes, which break apart as they descend.

Clay and silt slides are usually slow but can experience episodic acceleration in response to heavy rainfall or rapid snowmelt. They are often seen on gentle slopes and move over planar surfaces, such as over the underlying bedrock. Failure surfaces can also form within the clay or silt layer itself, and they usually have concave shapes, resulting in rotational slides.

Shallow and deep-seated landslides

[edit]
Hotel Panorama at Lake Garda. Part of a hill of Devonian shale was removed to make the road, forming a dip-slope. The upper block detached along a bedding plane and is sliding down the hill, forming a jumbled pile of rock at the toe of the slide.

Slope failure mechanisms often contain large uncertainties and could be significantly affected by heterogeneity of soil properties.[38] A landslide in which the sliding surface is located within the soil mantle or weathered bedrock (typically to a depth from few decimeters to some meters) is called a shallow landslide. Debris slides and debris flows are usually shallow. Shallow landslides can often happen in areas that have slopes with high permeable soils on top of low permeable soils. The low permeable soil traps the water in the shallower soil generating high water pressures. As the top soil is filled with water, it can become unstable and slide downslope.

Deep-seated landslide on a mountain in Sehara, Kihō, Japan caused by torrential rain of Tropical Storm Talas
Landslide of soil and regolith in Pakistan

Deep-seated landslides are those in which the sliding surface is mostly deeply located, for instance well below the maximum rooting depth of trees. They usually involve deep regolith, weathered rock, and/or bedrock and include large slope failures associated with translational, rotational, or complex movements.[39] They tend to form along a plane of weakness such as a fault or bedding plane. They can be visually identified by concave scarps at the top and steep areas at the toe.[40] Deep-seated landslides also shape landscapes over geological timescales and produce sediment that strongly alters the course of fluvial streams.[41]

[edit]
  • An avalanche, similar in mechanism to a landslide, involves a large amount of ice, snow and rock falling quickly down the side of a mountain.
  • A pyroclastic flow is caused by a collapsing cloud of hot ash, gas and rocks from a volcanic explosion that moves rapidly down an erupting volcano.
  • Extreme precipitation and flow can cause gully formation in flatter environments not susceptible to landslides.

Resulting tsunamis

[edit]
See also: Tsunami § Tsunami generated by landslides

Landslides that occur undersea, or have impact into water e.g. significant rockfall or volcanic collapse into the sea,[42] can generate tsunamis. Massive landslides can also generate megatsunamis, which are usually hundreds of meters high. In 1958, one such tsunami occurred in Lituya Bay in Alaska.[43][44]

Landslide prediction mapping

[edit]
See also: Slope stability analysis

Landslide hazard analysis and mapping can provide useful information for catastrophic loss reduction, and assist in the development of guidelines for sustainable land-use planning. The analysis is used to identify the factors that are related to landslides, estimate the relative contribution of factors causing slope failures, establish a relation between the factors and landslides, and to predict the landslide hazard in the future based on such a relationship.[45] The factors that have been used for landslide hazard analysis can usually be grouped into geomorphology, geology, land use/land cover, and hydrogeology. Since many factors are considered for landslide hazard mapping, geographic information system (GIS) is an appropriate tool because it has functions of collection, storage, manipulation, display, and analysis of large amounts of spatially referenced data which can be handled fast and effectively.[46] Cardenas reported evidence on the exhaustive use of GIS in conjunction of uncertainty modelling tools for landslide mapping.[47][48] Remote sensing techniques are also highly employed for landslide hazard assessment and analysis. Before and after aerial photographs and satellite imagery are used to gather landslide characteristics, like distribution and classification, and factors like slope, lithology, and land use/land cover to be used to help predict future events.[49] Before and after imagery also helps to reveal how the landscape changed after an event, what may have triggered the landslide, and shows the process of regeneration and recovery.[50]

Using satellite imagery in combination with GIS and on-the-ground studies, it is possible to generate maps of likely occurrences of future landslides.[51][52] Such maps should show the locations of previous events as well as clearly indicate the probable locations of future events. In general, to predict landslides, one must assume that their occurrence is determined by certain geologic factors, and that future landslides will occur under the same conditions as past events.[53] Therefore, it is necessary to establish a relationship between the geomorphologic conditions in which the past events took place and the expected future conditions.[54]

Natural disasters are a dramatic example of people living in conflict with the environment. Early predictions and warnings are essential for the reduction of property damage and loss of life. Because landslides occur frequently and can represent some of the most destructive forces on earth, it is imperative to have a good understanding as to what causes them and how people can either help prevent them from occurring or simply avoid them when they do occur. Sustainable land management and development is also an essential key to reducing the negative impacts felt by landslides.

A Wireline extensometer monitoring slope displacement and transmitting data remotely via radio or Wi-Fi. In situ or strategically deployed extensometers may be used to provide early warning of a potential landslide.[55]

GIS offers a superior method for landslide analysis because it allows one to capture, store, manipulate, analyze, and display large amounts of data quickly and effectively. Because so many variables are involved, it is important to be able to overlay the many layers of data to develop a full and accurate portrayal of what is taking place on the Earth's surface. Researchers need to know which variables are the most important factors that trigger landslides in any given location. Using GIS, extremely detailed maps can be generated to show past events and likely future events which have the potential to save lives, property, and money.

Since the '90s, GIS have been also successfully used in conjunction to decision support systems, to show on a map real-time risk evaluations based on monitoring data gathered in the area of the Val Pola disaster (Italy).[56]

Prehistoric landslides

[edit]
Rhine cutting through Flims Rockslide debris, Switzerland
  • Storegga Slide, some 8,000 years ago off the western coast of Norway. Caused massive tsunamis in Doggerland and other areas connected to the North Sea. A total volume of 3,500 km3 (840 cu mi) debris was involved; comparable to a 34 m (112 ft) thick area the size of Iceland. The submarine landslide is thought to be among the largest in history.[57]
  • Landslide which moved Heart Mountain to its current location, the largest continental landslide discovered so far. In the 48 million years since the slide occurred, erosion has removed most of the portion of the slide.
  • Flims Rockslide, about 12 km3 (2.9 cu mi), Switzerland, some 10,000 years ago in post-glacial Pleistocene/Holocene, the largest so far described in the Alps and on dry land that can be easily identified in a modestly eroded state.[58]
  • The landslide around 200 BC which formed Lake Waikaremoana on the North Island of New Zealand, where a large block of the Ngamoko Range slid and dammed a gorge of Waikaretaheke River, forming a natural reservoir up to 256 metres (840 ft) deep.
  • Cheekye Fan, British Columbia, Canada, about 25 km2 (9.7 sq mi), Late Pleistocene in age.
  • The Manang-Braga rock avalanche/debris flow may have formed Marsyangdi Valley in the Annapurna Region, Nepal, during an interstadial period belonging to the last glacial period.[59] Over 15 km3 (3.6 cu mi) of material are estimated to have been moved in the single event, making it one of the largest continental landslides.[citation needed]
  • Tsergo Ri landslide, a massive slope failure 60 km (37 mi) north of Kathmandu, Nepal, involving an estimated 10 to 15 km3 (2.4 to 3.6 cu mi).[60] Prior to this landslide the mountain may have been the world's 15th mountain above 8,000 m (26,247 ft).

Historical landslides

[edit]
Main article: List of landslides

Extraterrestrial landslides

[edit]

Evidence of past landslides has been detected on many bodies in the Solar System, but since most observations are made by probes that only observe for a limited time and most bodies in the Solar System appear to be geologically inactive not many landslides are known to have happened in recent times. Both Venus and Mars have been subject to long-term mapping by orbiting satellites, and examples of landslides have been observed on both planets.

  • Before and after radar images of a landslide on Venus. In the center of the image on the right, the new landslide, a bright, flow-like area, can be seen extending to the left of a bright fracture. 1990 image.
    Before and after radar images of a landslide on Venus. In the center of the image on the right, the new landslide, a bright, flow-like area, can be seen extending to the left of a bright fracture. 1990 image.
  • Landslide in progress on Mars, 2008-02-19
    Landslide in progress on Mars, 2008-02-19

Landslide mitigation

[edit]
This section is an excerpt from Landslide mitigation.[edit]
Two retaining walls and other soil slope stabilization methods in Chieti, Italy.

Landslide mitigation refers to human construction activities on slopes undertaken with the goal of lessening the effect of landslides. Landslides can be triggered by many, sometimes concomitant causes. In addition to shallow erosion or reduction of shear strength caused by seasonal rainfall, landslides may be triggered by anthropic activities, such as adding excessive weight above the slope or digging at either the mid-slope or foot of the slope.

Often, individual phenomena join to generate instability over time, which often does not allow a reconstruction of the evolution of a particular landslide. Therefore, landslide hazard mitigation measures are not generally classified according to the phenomenon that might cause a landslide.[65] Instead, they are classified by the sort of slope stabilization method used:

  • Geometric methods, in which the geometry of the hillside is changed (in general the slope);
  • Hydrogeological methods, in which an attempt is made to lower the groundwater level or to reduce the water content of the material
  • Chemical and mechanical methods, in which attempts are made to increase the shear strength of the unstable mass or to introduce active external forces (e.g. anchors, rock or ground nailing) or passive (e.g. structural wells, piles or reinforced ground) to counteract the destabilizing forces.
Each of these methods varies somewhat with the type of material that makes up the slope.

Landslide monitoring

[edit]

The monitoring of landslides is essential for estimating the dangerous situations, making it possible to issue alerts on time, to avoid loses of lives and property, and to have proper planning and reducing measures in place. Currently, there exist different type of techniques aimed to monitor landslides:

Remote sensing techniques

[edit]
  • InSAR (Interferometric Synthetic Aperture Radar): This remote sensing technique measures ground displacement over time with high precision. It is ideal for large-scale monitoring.[66][67]
  • LiDAR (Light Detection and Ranging): Provides detailed 3D models of terrain to detect changes over time by comparison of different point clouds acquired over time.[68][69]
  • Optical satellite imagery: Useful for identifying surface changes, geomorphological features (e.g. cracks and scarps) and mapping landslide-prone areas.[67]
  • UAVs (Unmanned Aerial Vehicles): This technique captures high-resolution images and topographic data in inaccessible areas.[70]
  • Thermal imaging: Thermal images enable to detects temperature variations that may indicate water movement or stress in the slope.[71]

Ground-based techniques

[edit]
  • GPS (Global Positioning System): Tracks ground movements at specific points over time using a constellation of satellites orbiting around the Earth.
  • Topographic surveys: Measures displacements of marked targets on a slope.
  • Ground-based radar (GB-SAR): Continuously monitors surface deformation using a SAR sensor and detects movement in real-time. It follows the same principle than InSAR.[72]

Geotechnical instrumentation

[edit]
  • Piezometers: Monitors groundwater levels and pore water pressure, which are critical triggers for landslides.
  • Load cells: Measures stress changes in retaining structures or anchors.
  • Tiltmeters: Detects small angular changes in the slope surface or retaining walls.
  • Extensometers: Measures displacement along cracks or tension zones.
  • Inclinometers: Detects subsurface movements by monitoring changes in the inclination of a borehole.[73]

Seismic techniques

[edit]

•Geophones and accelerometers: Detect seismic vibrations or movements that might indicate slope instability.

Climate-change impact on landslides

[edit]

Climate-change impact on temperature, both average rainfall and rainfall extremes, and evapotranspiration may affect landslide distribution, frequency and intensity.[74] However, this impact shows strong variability in different areas.[75] Therefore, the effects of climate change on landslides need to be studied on a regional scale. Climate change can have both positive and negative impacts on landslides. Temperature rise may increase evapotranspiration, leading to a reduction in soil moisture and stimulate vegetation growth, also due to a CO2 increase in the atmosphere. Both effects may reduce landslides in some conditions. On the other side, temperature rise causes an increase of landslides due to

  • the acceleration of snowmelt and an increase of rain on snow during spring, leading to strong infiltration events.[76]
  • Permafrost degradation that reduces the cohesion of soils and rock masses due to the loss of interstitial ice.[77] This mainly occurs at high elevation.
  • Glacier retreat that has the dual effect of relieving mountain slopes and increasing their steepness.

Since the average precipitation is expected to decrease or increase regionally,[75] rainfall induced landslides may change accordingly, due to changes in infiltration, groundwater levels and river bank erosion. Weather extremes are expected to increase due to climate change including heavy precipitation.[75] This yields negative effects on landslides due to focused infiltration in soil and rock[78] and an increase of runoff events, which may trigger debris flows.

See also

[edit]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Landslide

View on Grokipedia
from Grokipedia
A landslide is the movement of a mass of rock, debris, earth, or other materials down a slope under the influence of gravity, classified as a form of mass wasting. These events encompass various processes, including falls, topples, slides, spreads, and flows, differentiated by the type of material involved—such as rockfalls, debris flows, or earthflows—and the style of movement, which can be rotational or translational. Landslides vary in scale from localized creep to catastrophic failures displacing millions of cubic meters of material. Landslides are triggered by natural factors like prolonged heavy rainfall saturating slopes, earthquake shaking destabilizing materials, volcanic eruptions melting snow or ejecting debris, and undercutting bases, as well as human activities including slope excavation for roads, reducing soil cohesion, and improper drainage altering water flow. They predominantly occur in steep terrains with fractured rock, deep or thin soils over , or areas prone to rapid saturation, and can propagate underwater as submarine landslides. Globally, landslides result in thousands of deaths annually and inflict billions of dollars in , with the alone recording 25 to 50 fatalities per year. Significant events, such as those induced by seismic activity or , highlight their potential for widespread destruction, underscoring the need for hazard mapping and mitigation strategies like slope stabilization and early warning systems.

Fundamental Concepts

Definition and Physical Mechanics

A landslide is the movement of a mass of rock, , or down a , classified as a form of driven directly by . This encompasses processes where slope-forming materials displace downward and outward, with movement rates varying from millimeters per year to tens of kilometers per hour. The term includes distinct modes such as falls (free-falling detached material), topples (forward rotation of blocks), slides (translational or rotational shear along a defined surface), spreads (lateral extension in weak layers), and flows (fluid-like movement of saturated ). Physically, landslides occur when the downslope component of gravitational force exceeds the shear resistance of the material along a potential failure plane, leading to instability. , which resists failure, follows the Mohr-Coulomb criterion: τ = c + σ' tan φ, where τ represents , c is cohesion, σ' is effective normal stress, and φ is the angle of internal friction. Effective stress σ' = σ - u, with σ as total normal stress and u as ; elevated u from rainfall or saturation diminishes σ', thereby reducing strength and (FS = resisting shear stress / driving shear stress), where FS < 1 triggers failure. Driving forces derive from the slope's geometry, primarily the parallel component of the material's weight, mg sin β, where m is mass, g is gravitational acceleration, and β is slope angle. In rocky slopes, mechanics often involve discrete fractures or joints where tensile or shear failure predominates, while in soil or debris, progressive liquefaction or bulking can enhance mobility during flow. Erosion at the slope toe or loading at the crest can alter stress distributions, lowering FS by increasing driving moments or reducing stabilizing ones. These principles underpin stability analyses, emphasizing material heterogeneity, hydrological conditions, and topographic configuration as key determinants of initiation.

Material Properties and Failure Criteria

The shear strength of slope materials is the primary geotechnical property governing landslide resistance, defined by the Mohr-Coulomb criterion as the maximum shear stress τ\tau that a material can withstand before failure: τ=c+σntan⁡ϕ\tau = c + \sigma_n \tan \phi, where cc is cohesion, σn\sigma_n is effective normal stress, and ϕ\phi is the effective friction angle. Cohesion represents the inherent bonding forces within the soil matrix, typically higher in fine-grained clays (ranging from 0 to 50 kPa or more) and lower in granular sands (often near 0 kPa), while the friction angle reflects interparticle frictional resistance, generally 25°–35° for clays and 30°–45° for sands. Unit weight (γ\gamma), typically 18–22 kN/m³ for soils, contributes to driving shear stresses via gravitational forces, with saturated conditions reducing effective stress through pore water pressure, often leading to reduced shear strength in undrained scenarios. Failure occurs when the mobilized shear stress along a potential slip surface exceeds the available shear strength, as evaluated in limit equilibrium analyses common to landslide assessments. The Mohr-Coulomb envelope approximates brittle failure under increasing normal stress, with the criterion assuming linear behavior and neglecting tensile strength unless modified for rock masses where tension cracks may initiate instability. In landslide-prone glacial deposits, such as those in , properties exhibit wide variability—dry densities from 14 to 22 kN/m³ and friction angles as low as 20° in overconsolidated clays—highlighting site-specific testing needs via direct shear or triaxial tests to derive parameters accurately. Drained strength parameters are preferred for long-term slope stability, while undrained analyses apply to rapid failures, with residual strengths (post-peak, often ϕr<ϕ\phi_r < \phi) critical for reactivated slides. For rock-dominated landslides, aggregate strength integrates discontinuity properties, with effective cohesion solved assuming ϕ=30∘\phi = 30^\circ in some empirical models, revealing near-surface weakening to 0.1–1 MPa in seismically active areas. Pore pressure effects, modeled via Bishop's effective stress principle, further modulate failure criteria, where rising groundwater lowers σn′\sigma_n', potentially triggering slides even in materials with adequate drained strength. These properties are derived from field methods like cone penetration testing, which infers ϕ\phi and cc from tip resistance and sleeve friction, ensuring parameters reflect in-situ conditions over lab ideals. Variability in fine-grained soils—higher cc but lower ϕ\phi—contrasts with coarser materials, influencing failure plane geometry and depth.

Causative Factors

Geological and Morphological Preconditions

Geological preconditions for landslides encompass inherent material weaknesses and structural discontinuities that reduce shear resistance along potential failure planes. These include weak, weathered, sheared, jointed, or fissured bedrock and regolith, such as shales, clays, or sensitive marine deposits prone to liquefaction under stress. Adversely oriented geological structures, including bedding planes, faults, joints, or schistosity dipping parallel or out of the slope face, facilitate planar or wedge failures by aligning with the direction of gravitational shear. Contrasts in permeability or stiffness between overlying and underlying layers, such as stiff soils over plastic clays, can create zones of concentrated stress and pore pressure buildup, further compromising stability. Morphological preconditions involve topographic configurations that amplify gravitational forces and stress concentrations on slopes. Steep gradients, often exceeding 20–45 degrees depending on material cohesion, elevate the component of downslope shear stress relative to normal stress, promoting instability in both bedrock and unconsolidated cover. High relief, resulting from tectonic uplift, glacial rebound, or differential erosion, oversteepens slopes and increases driving forces, as observed in regions with rapid incision by rivers or glacial carving. Concave-up slope profiles or headward erosion of gullies concentrate tensile stresses at the crest and compressive forces at the toe, while undercutting of the slope base by fluvial or wave action removes lateral support, exposing weaker materials to failure. These forms are particularly evident in landscapes with convergent topography, such as valley sides or canyon walls, where depositional loading at the crest can exacerbate imbalances.

Natural Triggers

Intense rainfall is the most common natural trigger for landslides worldwide, as it rapidly infiltrates slopes, elevating pore water pressures and diminishing shear strength along failure planes. This process follows from the principle that saturated soils experience reduced effective normal stress, promoting instability when gravitational forces exceed frictional resistance. Empirical thresholds, such as intensity-duration relationships, indicate that landslides often initiate when rainfall exceeds 50-150 mm over 24-72 hours in susceptible terrains, though these vary by geology and antecedent moisture; for instance, antecedent rainfall accumulation over 15-30 days amplifies risk by preconditioning higher water tables. Seismic shaking from earthquakes constitutes another primary trigger, imparting cyclic shear stresses that exceed static stability in marginally stable slopes, often inducing widespread shallow failures or deep-seated ruptures. Mechanisms include direct ground acceleration amplifying downslope forces and liquefaction in cohesionless, saturated deposits, where excess pore pressures lead to temporary loss of strength; magnitudes above 5.0 on the Richter scale frequently correlate with elevated landslide incidence, as observed in events like the 2008 Wenchuan earthquake, which mobilized over 60,000 landslides across fault-proximal zones. Volcanic eruptions trigger landslides through mechanisms such as pyroclastic flows, ash saturation, or lahar generation from rapid snowmelt on volcanic edifices, where deposited materials overburden slopes or alter hydrology. For example, the 1980 Mount St. Helens eruption initiated debris avalanches exceeding 2 billion cubic meters via blast-induced fragmentation and liquefaction of hydrothermally altered rock. Stream undercutting, coastal wave erosion, and rapid snowmelt or glacial outbursts also serve as triggers by abruptly removing basal support or introducing surge flows that erode and saturate upslope materials, though these are typically localized compared to rainfall or seismicity. Wildfires, through natural ignition, indirectly trigger post-fire debris flows by rendering soils hydrophobic and prone to runoff-driven saturation during subsequent storms.

Anthropogenic Contributors

Human activities alter slope stability through vegetation removal, hydrological modifications, and mechanical disturbances, increasing landslide susceptibility and frequency. In the United States, such activities contribute to annual damages of $1-2 billion and over 25 fatalities from landslides. Globally, records from 2004 to 2016 indicate that human-induced landslides, particularly those linked to mining and construction, have risen, accounting for a notable share of fatal events among nearly 4,900 documented landslides causing 56,000 deaths. Deforestation reduces root reinforcement and evapotranspiration, thereby elevating shallow landslide risks on steep slopes. Studies in Nepal show that deforestation occurring 5-7 years prior to events significantly enhances landslide occurrence by diminishing soil cohesion. Forests mitigate shallow landslides by mechanically stabilizing soils via root networks and facilitating drainage to lower pore water pressure, effects lost upon removal. In Far-Western Nepal, antecedent deforestation combined with agricultural practices directly correlates with heightened susceptibility. Urbanization and construction exacerbate instability by reorganizing surface and subsurface water flows, often through impervious surfaces and slope cutting for infrastructure. In urban settings, precipitation-triggered landslides occur more frequently than in rural areas due to these hydrological alterations. Unregulated development, such as hillside terracing and road building, decreases slope stability, as observed in rapidly urbanizing regions like the Democratic Republic of Congo. In China, human development of steep mountainous areas has intensified landslide density in certain river basins. Mining, quarrying, and excavation remove lateral support and introduce vibrations or water infiltration, triggering deep-seated failures. Human-induced landslides from these activities showed an uptick in fatal incidents between 2004 and 2016, particularly in Asia. Approximately 40% of landslides impacting transport networks stem from human factors, including excavation-related disturbances. Agricultural practices, such as terracing, over-irrigation, and tillage, can saturate soils or erode protective cover, amplifying risks especially when combined with rainfall. Changes in land use for agriculture further destabilize slopes by altering infiltration rates and vegetation. In some regions, human-induced events constitute up to 71.7% of landslides, underscoring the dominance of these activities over natural triggers.

Classification and Types

Movement-Based Classifications

Landslides are classified by their dominant mode of movement into five kinematic categories: falls, topples, slides, spreads, and flows, following the framework established by Varnes (1978) and adopted by the U.S. Geological Survey (USGS). This system emphasizes the mechanics of displacement, independent of material type, though combinations often occur in complex events. The classification aids in hazard assessment by linking movement style to predictive behaviors and mitigation strategies. Falls occur when masses of rock or soil detach from steep slopes and descend primarily through free fall, bouncing, or rolling, with separation along discontinuities like joints or fractures. This rapid movement typically affects discrete blocks and is common in cliffs or steep rock faces, as seen in rockfall events where velocities can exceed 30 m/s. Topples involve forward rotation of rock columns or blocks about a fixed pivot point at or near the base, often leading to subsequent falls or slides if the mass overturns. Tension cracks commonly develop behind the toppling mass, and this type predominates in columnar jointed rocks like basalt, with slow initial pivoting accelerating under gravity. Slides feature downslope movement along a defined surface of rupture, subdivided into rotational slides—where the rupture surface is concave-upward, causing backward tilting and rotation about an axis parallel to the slope—and translational slides, which occur on planar or gently undulating surfaces with minimal rotation. Rotational slides, or slumps, often form spoon-shaped depressions, while translational variants, like block slides, maintain the slide mass's orientation. Spreads entail extensional horizontal displacement of coherent material over a weaker substratum, typically triggered by liquefaction or basal collapse in flat-lying unconsolidated sediments such as or glacial deposits. This results in cracking and lateral fissures, with the upper mass fracturing into blocks that diverge downslope. Flows exhibit fluid-like behavior where the mass moves as a viscous slurry, with little to no intact shearing surface, often involving saturated soil or debris with high water content behaving like a . Velocities range from slow creeps to rapid surges exceeding 50 km/h, and flows can incorporate materials ranging from fine clay to boulders.

Depth and Scale Distinctions

Landslides are categorized by the depth of the failure plane, which influences the mechanics, triggers, and potential impacts. Shallow landslides typically involve movement confined to the surficial soil or regolith layers, with depths generally less than 4.5 meters (15 feet) or up to 3 meters in many soil-mantled slopes. These failures occur above the bedrock interface, often in unconsolidated materials susceptible to rapid saturation, and are commonly initiated by intense rainfall or shallow seismic shaking, leading to quick mobilization as debris flows or slides with high velocities exceeding 10 m/s. In contrast, deep-seated landslides feature rupture surfaces extending into competent bedrock, with depths exceeding 4.5–10 meters and sometimes reaching hundreds of meters. These involve larger shear zones along geological discontinuities, progressing more slowly—often at rates of centimeters to meters per year—due to the greater shear strength and drainage variability in bedrock compared to soil. Deep-seated failures demand sustained destabilizing forces, such as prolonged erosion or groundwater rise, and can reactivate over decades, posing chronic risks to infrastructure. Scale distinctions complement depth classifications, primarily through metrics of displaced volume and planimetric area, which correlate with destructive potential and runout distance. Shallow landslides are typically smaller in scale, with volumes under 10,000 m³ and areas less than 0.1 km², reflecting limited material availability in thin surficial layers. Deep-seated events often achieve greater scales, involving volumes from 10^6 m³ to over 10^9 m³ and areas spanning multiple square kilometers, as bedrock involvement allows for extensive block or rotational failures. A logarithmic classification system for scale, applicable across landslide types, delineates categories as follows:
Size ClassVolume Range (m³)Area Range (km²)Typical Characteristics
Very Small< 10³< 0.001Localized soil slips, minimal runout
Small10³ – 10⁴0.001 – 0.01Common in steep, vegetated slopes
Medium10⁴ – 10⁶0.01 – 0.1Debris avalanches, moderate infrastructure impact
Large10⁶ – 10⁸0.1 – 1Valley-blocking potential, regional effects
Giant/Colossal> 10⁸> 1Catastrophic, long-runout flows or rock avalanches
These scales follow power-law relationships between area and (exponents 1.1–1.6), enabling estimation from mapped areas for modeling, though shallow events exhibit lower exponents due to confined depths. Depth-scale interplay arises causally: shallower failures limit material entrainment, constraining scale, while deeper ones amplify through fracturing, though empirical data from inventories show overlaps, with rare giant shallow flows possible in loose .

Other Typologies

Landslides are classified by rate of movement into seven velocity classes, ranging from extremely slow to extremely rapid, as defined by the Working Party on World Landslide Inventory (WP/WLI) in 1995 and adopted in Cruden and Varnes (1996). This typology aids in assessing potential, as higher velocities correlate with greater destructive capacity and lower predictability.
Velocity ClassDescriptionVelocity Range
Extremely slowMotion perceptible only via precise instruments< 16 mm/year
Very slowMotion observable over years to decades16 mm/year to 1.6 m/year
SlowMotion observable over months to years1.6 m/year to 15 m/year
ModerateMotion observable over days to weeks15 m/year to 0.5 m/hour
RapidMotion observable over minutes to hours0.5 m/hour to 3 m/minute
Very rapidMotion observable over seconds to minutes3 m/minute to 20 m/second
Extremely rapidMotion too fast for effective mitigation> 20 m/second
Another typology distinguishes landslides by state of activity, which reflects current or recent movement status and informs reactivation risks. Active landslides are currently moving or have suspended motion within the last seasonal cycle; reactivated ones resume after prior . Dormant landslides show no recent motion but retain failure surfaces and may reactivate under triggers; stabilized ones have been engineered for immobility, while relict types persist from prehistoric events with minimal modern features. This classification, outlined by WP/WLI (1993), emphasizes morphological indicators like fresh scarps or patterns to differentiate states empirically. Additional descriptors include hydrological state (dry, where <5% pore water drives motion; moist, with partial saturation; or wet, involving liquefaction) and complexity (simple, with singular movement; complex, combining multiple types sequentially). These typologies complement primary schemes by quantifying dynamic behaviors and environmental interactions, enabling refined hazard zoning despite challenges in field verification.

Risk Assessment and Prediction

Susceptibility Mapping Techniques

Landslide susceptibility mapping identifies terrain areas prone to failure based on spatial conditioning factors including slope steepness, geological lithology, soil type, land cover, and distance to drainage networks, without incorporating temporal triggers like rainfall. These techniques integrate geographic information systems (GIS) to overlay thematic layers and compute susceptibility indices, producing zonation maps categorized as low, moderate, high, or very high risk. Empirical inventories of past landslides serve as training data to validate models, though susceptibility reflects inherent terrain instability rather than future event prediction. Techniques are broadly classified into knowledge-based (qualitative/heuristic), statistically-based (semi-quantitative), physically-based (deterministic), and data-driven machine learning approaches, each with distinct data requirements and applicability. Knowledge-based methods rely on expert judgment to weight factors heuristically, such as through geomorphological field mapping or the analytical hierarchy process (AHP), which pairwise compares factor importance via eigenvector calculations. These are advantageous in data-scarce regions for their simplicity and interpretability but suffer from subjectivity and poor reproducibility, limiting scalability to large areas. Statistically-based methods quantify relationships between landslides and factors using historical data, divided into bivariate approaches like weights of evidence (WoE) or frequency ratio, which compute class-specific probabilities independently, and multivariate models such as logistic regression (LR), which estimates coefficients via maximum likelihood for a binary outcome (landslide vs. non-landslide). Data-overlay and index-based techniques aggregate weighted layers simply, while WoE applies Bayesian principles to update prior odds with evidence from factor classes. Logistic regression remains the most prevalent, applied in over 20% of reviewed studies from 1983–2016, offering objectivity and handling multicollinearity, though assumptions of factor independence in bivariate methods can introduce bias. Performance evaluations, often via receiver operating characteristic (ROC) curves, show area under curve (AUC) values exceeding 0.8 in many cases, but uncertainty quantification is rare despite data quality variations. Machine learning methods, including support vector machines (SVM), random forests (RF), and neural networks, process non-linear interactions through training on landslide inventories and predictor variables, outperforming traditional statistics in accuracy (e.g., RF AUC >0.85 in ensemble applications). The process involves data preprocessing (e.g., via information gain), model training with cross-validation, and susceptibility indexing via probability outputs. variants like convolutional neural networks handle spatial dependencies effectively in high-resolution data, while adapts pre-trained models to data-limited sites; advantages include reduced subjectivity and higher predictive power, though they demand computational resources and large datasets, risking without proper tuning. Trends since 2010 show ensembles combining multiple algorithms (e.g., RF with LR) yielding superior results over single models. Physically-based models simulate using geotechnical parameters and physical laws, such as the infinite slope equation incorporating (FoS = c / (γ h sin²β) + (cosβ tanφ / tanβ) - 1, where c is cohesion, γ is unit weight, h is depth, β is , and φ is ). Distributed implementations like SHALSTAB or TRIGRS couple with to assess saturation-induced , requiring detailed subsurface from boreholes or . These provide causal insights into mechanisms but are computationally intensive, sensitive to parameter (e.g., strength variability), and less common due to sparse input , often applied only in site-specific contexts rather than regional mapping. Hybrid approaches integrating statistical with physical elements are emerging to balance and realism, though comprehensive assessments remain underrepresented across all techniques.

Probabilistic and Deterministic Models

Deterministic models in rely on physically based calculations to evaluate , typically computing a (FoS) that compares resisting forces against driving forces along potential failure surfaces. These models, such as the infinite analysis or Bishop's simplified method of slices, require detailed inputs including parameters (cohesion and angle), geometry, , and material unit weights, often derived from geotechnical investigations. For instance, the Transient Rainfall Infiltration and Grid-Based Regional -Stability Model (TRIGRS) simulates rainfall-induced changes in to predict shallow landslide initiation on a grid basis. Such approaches yield binary outcomes—stable if FoS > 1, unstable otherwise—and are suited for site-specific analyses where high-resolution data are available, as demonstrated in assessments of shallow landslides in regions like . However, they assume fixed parameter values, potentially overlooking natural heterogeneities and leading to overconfidence in predictions without sensitivity analyses. Probabilistic models address uncertainties inherent in geological and hydrological parameters by incorporating statistical distributions, such as lognormal for properties or simulations for variability in rainfall intensity. These methods output the probability of failure (e.g., annual exceedance probability) rather than deterministic thresholds, enabling regional-scale hazard mapping by integrating empirical landslide inventories with physical laws. Examples include Bayesian networks that update prior probabilities with observed data or frequency-volume distributions calibrated against historical events, as applied in basin-scale assessments using multi-temporal inventories. The Probabilistic Hydrological Estimation of LandSlides (PHELS) model, for instance, estimates global daily hazard by coupling hydrological simulations with probabilistic triggering thresholds derived from and data. Probabilistic frameworks are particularly valuable for data-scarce areas, providing uncertainty bounds that inform decision-making, though they depend on the quality of input distributions and may require validation against independent events to avoid . In comparative applications, deterministic models excel in designs for precise , such as retaining walls, but often underestimate risks at broader scales due to parameter variability, as evidenced in multi-hazard assessments where probabilistic variants revealed higher exposure. Probabilistic models, while computationally intensive, better capture epistemic and aleatory uncertainties, yielding susceptibility maps with quantified confidence intervals that align more closely with observed landslide frequencies in probabilistic validations. Hybrid approaches, combining deterministic physics with probabilistic sampling, are increasingly adopted to leverage strengths of both, such as in machine learning-augmented stability analyses for enhanced predictive accuracy. Limitations persist in both, including sensitivity to input and the challenge of validating , underscoring the need for ensemble methods in operational .

Limitations and Uncertainties

Landslide susceptibility mapping and hazard prediction models are constrained by incomplete and biased historical inventories, which often underrepresent small or undetected events and fail to capture non-stationary environmental conditions. Data collection challenges, including positional errors in landslide locations and inconsistencies in non-landslide sampling strategies, introduce epistemic uncertainties that propagate through and statistical models, reducing predictive reliability. For instance, digital elevation models (DEMs) of varying resolutions can alter and calculations, leading to divergent susceptibility zonations, while sparse conditioning factors like properties exacerbate model sensitivity. Probabilistic models attempt to quantify aleatory uncertainties inherent in triggers such as or seismic shaking, yet they struggle with non-linear interactions and future variability, often overestimating or underestimating hazard in dynamic landscapes. Deterministic approaches, reliant on physics-based thresholds like , overlook spatial heterogeneity and transient pore effects, limiting their applicability beyond site-specific scales. Validation against independent datasets reveals persistent gaps, with models performing adequately on data but exhibiting high false positives in altered terrains influenced by anthropogenic factors. Moreover, epistemic uncertainties from selection—such as friction angles or hydrological inputs—require simulations for propagation analysis, but comprehensive remains computationally intensive and rarely implemented in operational forecasting. Communicating these uncertainties to decision-makers poses additional challenges, as risk assessments must balance conservative assumptions with , yet incomplete propagation of input errors can mislead rankings. In regions with rapid land-use changes, models fail to integrate real-time anthropogenic contributors, amplifying prediction errors during extreme events. Overall, while advances in ensemble techniques and mitigate some limitations, fundamental gaps in data representativeness and model generalizability persist, necessitating hybrid approaches that explicitly account for scenario-based sensitivities.

Monitoring Approaches

Remote Sensing Methods

Remote sensing methods utilize satellite, aerial, and (UAV)-borne sensors to detect landslide initiation, map extents, and quantify deformation over large areas, enabling timely warnings where ground access is hazardous or impractical. These approaches leverage active and passive sensors to capture topographic changes, surface displacements, and spectral alterations, often integrated with geographic information systems for analysis. Empirical studies demonstrate their efficacy in diverse terrains, from mountainous regions to coastal slopes, though atmospheric interference and resolution constraints pose challenges. Synthetic aperture radar (SAR) , particularly differential InSAR (DInSAR) and advanced variants like persistent scatterer InSAR (PSInSAR) and small baseline subset InSAR (SBAS-InSAR), measures sub-centimeter to millimeter-scale ground displacements by comparing phase differences in radar echoes from repeat passes. SAR operates independent of daylight or , penetrating clouds and to monitor slow-moving landslides with velocities below 1 m/year, as validated in alpine valleys where PSInSAR detected deformations of 5-20 mm/year prior to failure. Limitations include in rapidly deforming areas and atmospheric phase , mitigated through multi-temporal stacking and atmospheric correction models. Applications include earthquake-triggered landslide tracking, with InSAR identifying over 1,000 new displacements in the region at resolutions up to 5 meters. Optical imagery from sensors aboard platforms like or Landsat-8 employs change detection via (NDVI) differentials or classifiers to identify landslide scars through exposed soil and disrupted canopy patterns, achieving detection accuracies exceeding 80% in clear-sky conditions post-rainfall events. For example, tasseled cap transformation on PlanetScope imagery rapidly mapped landslides from in 2019, delineating affected areas within hours of image acquisition. Drawbacks encompass obscuration by clouds in humid climates—prevalent in 70-90% of tropical post-event scenarios—and insensitivity to subsurface motion, necessitating fusion with SAR for comprehensive assessment. Light detection and ranging (), typically airborne or UAV-mounted, produces high-resolution digital elevation models (DEMs) with vertical accuracies of 10-15 cm to delineate landslide boundaries, compute volumes, and track morphological evolution through differential . Repeat surveys have quantified displacements in active slides, such as 2-5 meters of headwall retreat in coastal bluffs, supporting volume estimates accurate to within 5% via cloud-to-cloud comparisons. While offering dense point clouds (up to 50 points/m²), 's high cost and weather dependency restrict routine use, though terrestrial variants enable localized, real-time monitoring of scarps and cracks. Multisensor fusion, combining SAR deformation with topography and optical change maps, enhances reliability; for instance, integrated analyses in the yielded 90% detection rates for co-seismic landslides by cross-validating spectral and geometric signatures. Emerging models automate feature extraction across datasets, reducing manual interpretation while addressing biases in training data from under-sampled regions. Despite advancements, validation against remains essential, as overestimates shallow slides and underdetects vegetated failures without ancillary geophysical data.

Ground-Based and In-Situ Techniques

Ground-based and in-situ techniques for landslide monitoring rely on sensors deployed directly on or within the slope to measure parameters such as displacement, tilt, and pore water pressure with high spatial resolution and real-time capability. These methods complement remote sensing by providing subsurface data critical for understanding failure mechanisms and issuing early warnings. Instrumentation is typically installed via boreholes or surface anchors, often automated with dataloggers for continuous monitoring. Inclinometers, installed in vertical or inclined boreholes with slotted casings, detect lateral shear displacements by traversing a probe that measures deviations from a baseline profile using tilt sensors. In-place inclinometers (IPIs) enable automated, continuous readings, identifying shear zones where movement exceeds 5-10 mm annually in active landslides. The (USGS) employs inclinometers to monitor tilting in potential landslide areas, correlating subsurface shifts with surface indicators. Piezometers quantify and levels, which influence by reducing . Vibrating wire piezometers, embedded in boreholes, transmit frequency signals proportional to pressure, offering accuracy within 0.1% of over depths up to 100 meters. Standpipe piezometers provide manual readings via fluctuations in perforated pipes. These instruments have been used in excavations and slopes to track seasonal pressure changes triggering movements, as documented in geotechnical monitoring protocols. Extensometers measure linear extensions or contractions across defined intervals, either on the surface with wire or rod systems or subsurface via multipoint extensometers (MPBX) that anchor at multiple depths. MPBX systems detect differential movements along potential slip surfaces with resolutions of 0.1 mm, as applied in assessments. Surface extensometers, like those with digital transducers, monitor crack widening or bench displacements in open-pit mines and cuts. Additional tools include tiltmeters for angular changes on shallow surfaces, crackmeters for fracture apertures, and strain gauges on anchors to assess stress redistribution. Global Navigation Satellite Systems (GNSS) stations provide precise three-dimensional surface displacements at millimeter accuracy when fixed on stable benchmarks. Integration of these sensors via systems allows threshold-based alerts, though challenges persist in harsh environments where durability is tested against and vandalism.

Seismic and Geophysical Detection

Seismic detection methods exploit the ground vibrations produced by landslide mobilization, which generate characteristic seismic signals distinguishable from earthquakes or other sources. These signals typically include long-period surface waves with durations of seconds to minutes, enabling regional seismic networks to locate the event and estimate its volume using seismogram amplitudes recorded within three minutes of onset. For instance, algorithms process continuous data from multiple stations to identify landslide-specific waveforms, facilitating rapid alerts for secondary hazards like tsunamis generated by coastal slides. Passive seismic techniques, such as ambient noise correlation, monitor subtle precursory activity in unstable slopes by analyzing long-term recordings for changes in seismic velocity, which indicate evolving internal damage or fluid migration prior to . In active slow-moving landslides, increased microseismicity—manifesting as low-frequency events—serves as a precursor, detectable via localized arrays to provide early warnings of toward catastrophic . Geophysical detection encompasses non-seismic subsurface imaging to identify instability precursors, such as shear zones or water accumulation, through repeatable surveys that map material contrasts. (ERT) delineates low-resistivity zones associated with saturated clays or seepage, which weaken slopes; time-lapse ERT tracks resistivity decreases over time, correlating with rising pore pressures in moisture-driven failures. Seismic surveys reveal contrasts at potential slip surfaces, with refracted P-wave speeds dropping in fractured or weathered layers prone to sliding. Multichannel analysis of surface waves (MASW) complements by inverting shear-wave velocities to profile stiffness gradients, identifying zones of reduced rigidity that signal impending movement. (GPR) offers shallow-resolution detection of discontinuities like faults or voids, though its efficacy diminishes in conductive or clay-rich terrains. These methods, often integrated in time-lapse configurations, enable of hydrological triggers by quantifying property changes, such as desaturation effects post-rainfall that stabilize or destabilize slopes based on empirical thresholds observed in field studies. Limitations include signal ambiguity in heterogeneous and the need for site-specific , as ambient or cultural interference can mask subtle indicators.

Notable Examples

Prehistoric and Megalandslides

Prehistoric landslides, those occurring before the advent of written historical records, are primarily identified through geomorphic mapping, stratigraphic analysis, and dating techniques such as radiocarbon and exposure methods. These events often cluster during periods of climatic transition, particularly the Pleistocene-Holocene boundary around 12,000–10,000 years ago, when glacial retreat reduced on slopes, inducing through debuttressing and isostatic rebound. Empirical evidence from sediment cores and erratic boulders supports that such landslides frequently involved massive volumes, reshaping valleys and coastlines with long-lasting morphological signatures. Megalandslides, typically defined by volumes exceeding 1 km³, exemplify the extreme scale of prehistoric mass movements, often triggered by a combination of tectonic stress, seismic activity, and hydrological factors rather than solely climatic ones. The rockslide in , dated to approximately 9,660–9,430 calibrated years via radiocarbon analysis of overlying lake sediments, displaced 10–12 km³ of and dolomite, forming a debris field spanning over 60 km² in the Vorderrhein Valley. This event, the largest known landslide in the European , likely initiated along a basal plane cutting through competent rock layers, with runout facilitated by fragmentation and possible of basal sediments, as evidenced by hummocky and boulder deposits. The , a megalandslide off mid-Norway's coast, involved the retrogressive failure of approximately 3,000 km³ of glacial marine sediments around 8,150 calibrated years BP, as determined from seismic profiling and core samples revealing headwall scars over 290 km long. This collapse generated a with run-up heights exceeding 20 m along adjacent shores, depositing marine sands inland and potentially contributing to the inundation of settlements in the region, though direct human impact attribution remains debated due to sparse archaeological evidence. Causes include oversteepening of continental slopes by post-glacial and possible triggering, highlighting the causal role of sediment loading in submarine instability. In , the Marysvale gravity slide complex in comprises multiple prehistoric megalandslides of volcanic breccias, each exceeding 100 km³ in volume, dated to the Miocene-Pliocene epochs through stratigraphic correlation with dated ash flows. These slides traveled tens of kilometers across low gradients, likely lubricated by or , demonstrating how volcanic edifice weakening can produce cataclysmic failures with volumes rivaling oceanic events. Such examples underscore that megalandslides, while rare, exert profound control on regional , often damming rivers and altering drainage patterns for millennia.

Historical Landslide Events

On December 16, 1920, the Haiyuan (magnitude 8.5) in China's Province triggered widespread landslides that liquefied and flowed, engulfing over 600 villages across 20,000 square kilometers; estimates attribute 200,000 to 240,000 deaths primarily to these landslides, making it one of the deadliest mass movement events recorded. In July 1949, the Khait earthquake (magnitude 7.5) in Tajikistan's Gissar Valley initiated a complex of rock avalanches and mudflows totaling about 145 million cubic meters, which descended slopes up to 2,000 meters high and buried settlements in the Obi-Gissar and Khait rivers, killing approximately 7,200 people. The Vajont landslide on October 9, 1963, in involved 270 million cubic meters of unstable and clay detaching from Mount Toc above the Vajont ; accelerating to speeds of 20-30 meters per second due to reservoir-induced pore pressure rise, it displaced water to create a 250-meter-high wave that overtopped the dam intact but devastated downstream villages like , resulting in 1,917 confirmed deaths. During the Ancash earthquake (magnitude 7.9) on May 31, 1970, in , seismic shaking dislodged 10-25 million cubic meters of ice, rock, and debris from Nevado Huascarán's north peak, forming a high-velocity that traveled 11 kilometers in minutes, incorporating additional material into a that obliterated Yungay and Ranrahirca, contributing to 66,000-70,000 total fatalities from landslides in the event—the highest for a single landslide in the . Other significant 20th-century events include the Tadjik SSR landslides from and , which caused around 5,000 deaths in remote valleys, and the Father Bayley landslide in , a submarine debris flow triggered by an that generated a killing over 27,000, though attribution to the landslide versus seismic shaking remains debated in some analyses.

Case Studies of Contested Attributions

In geological mapping, certain landslide features have been contested as active tectonic faults due to similarities in geomorphic expression, such as scarps, hummocky topography, and linear alignments, leading to erroneous assessments. A 2012 study documented multiple instances in the where landslide-related structures were mapped as faults, resulting in misinterpretations of regional deformation history and underestimation of mass-wasting risks while overemphasizing potential. For example, rotational slumps and earthflows produce fault-like offsets and shearing that mimic dip-slip faults, but trenching and dating reveal landslide ages rather than faulting, highlighting the need for detailed subsurface investigations to distinguish gravitational from tectonic processes. The 2014 Oso landslide in Washington State exemplifies disputes over anthropogenic contributions versus inherent geological instability. On March 22, 2014, approximately 18 million cubic meters of saturated soil and debris mobilized downslope, traveling 1.8 kilometers and killing 43 people. A U.S. Geological Survey report attributed the trigger primarily to prolonged heavy rainfall on a hillside with a history of prehistoric slides dating back thousands of years, emphasizing natural factors like glacial till composition and river undercutting without assigning blame to human activity. However, critics, including lawsuits from survivors, contested this by pointing to 2004 logging operations that removed mature trees—natural slope stabilizers—from 56 acres above the slide path, arguing that reduced root reinforcement and increased surface water infiltration exacerbated vulnerability, a claim supported by post-event analyses linking clear-cutting to heightened landslide frequency in similar terrains. State officials initially avoided studying logging's role directly, fueling accusations of regulatory oversight failures amid industry pressures. The 1963 Vajont landslide in illustrates ongoing debates regarding slide mechanics and triggering mechanisms in reservoir-induced events. On October 9, 1963, about 270 million cubic meters of detached from Mount Toc, surging into the Vajont Reservoir at speeds up to 30 meters per second, generating an overflow wave that killed nearly 2,000 people downstream despite the structure remaining intact. Predisposing factors included deep-seated clay layers prone to shearing, but attribution remains contested between a first-time of intact rock versus reactivation of a prehistoric slide, with evidence from boreholes and geomorphic mapping supporting the latter through identification of ancient shear zones. Reservoir filling raised water levels by 100 meters, inducing pore pressure rises that critics argue accelerated an inevitable slide, while defenders of the project cited underestimation of slide volume in pre-construction surveys; criminal proceedings later convicted engineers for in risk forecasting. This case underscores causal complexities where human interacts with geological preconditions, with persistent scholarly disagreement on the relative weights of natural versus induced triggers.

Extraterrestrial Landslides

Planetary Occurrences

Landslides occur on multiple planetary bodies in the Solar System, with morphological evidence documented on , , Mercury, and the through orbital imaging, radar mapping, and sample analysis. These extraterrestrial mass-wasting events differ from terrestrial ones due to lower , thin or absent atmospheres, and unique surface compositions, often resulting in longer runouts relative to slope height. Geomorphological inventories highlight their prevalence on steep terrains shaped by , impacts, or . On Mars, landslides are abundant, particularly along the walls of and in polar regions, with global mapping efforts identifying controls such as slope angle, material cohesion, and potential fluid involvement. High-resolution images from missions like reveal features like lobate debris aprons and long-runout slides exceeding 100 km, attributed to mechanisms including dry granular flow, trapped subsurface air, or localized melting of ice-salt mixtures in the . For instance, recurring slope lineae and pit landslides in Sisyphi Cavi exhibit scarps and trenches indicative of recent activity, potentially triggered by cryosalt expansion and collapse. Venus hosts landslides primarily associated with volcanic edifices, identified via NASA's Magellan spacecraft radar data from the early 1990s, which penetrated the thick atmosphere to reveal collapse scars and debris flows. In Atla Regio, 29 such features were cataloged, displaying characteristics akin to terrestrial and Martian analogs, including hummocky deposits and runouts from caldera rims. Similar deposits appear in Navka Planitia, linked to structural failures on steep volcanic flanks, with bright radar returns suggesting fresh, blocky material. Mercury's landslides, observed in orbital data from NASA's mission, cluster on walls and tectonic scarps, forming talus-like aprons and elongated slides facilitated by the planet's low and lack of atmosphere, which prevent but allow extensive boulder rolling. The Moon exhibits rare but confirmed landslides, with a 2025 analysis of unopened samples from the Taurus-Littrow valley revealing shocked basalt clasts deposited by a moonquake-induced slide along the Lee-Lincoln fault approximately 90 million years ago. Modeling indicates seismic shaking displaced material downslope, forming boulder fields observable in images.

Comparative Insights

Extraterrestrial landslides, particularly on and , exhibit significantly larger scales compared to those on , with Martian examples often exceeding terrestrial counterparts in area and volume due to lower and the absence of persistent liquid water erosion. Analysis of landslide size distributions reveals that the for area (p(AL)) on Mars is flatter than on , indicating a higher proportion of very large landslides with areas greater than 10^7 m², and deviations from power-law trends for areas over 10^8 m². On , recently identified landslides in Atla Regio, numbering 29 as of 2024, display morphological similarities to both Martian and terrestrial slides but tend to originate from escarpments higher than typical sources, akin to those in Mars' . Causal factors diverge notably from terrestrial norms, where rainfall-induced saturation dominates; on Mars, the thin atmosphere precludes significant , shifting triggers toward seismic activity, impact events, or volatile releases, enabling longer runouts facilitated by reduced (about 38% of 's). Venusian slides, observed via radar imaging, likely stem from tectonic stresses or volcanic influences in a dense, CO₂-rich atmosphere that suppresses but permits massive collapses from steep terrains. Average fall heights for Martian landslides reach 5.3 km, contrasting with 1.2 km on , amplifying mobility and deposit extent without atmospheric drag mitigating flow as on terrestrial subaerial slides. These comparisons underscore 's role in modulating landslide dynamics across planetary bodies, with lower on Mars promoting extended runouts relative to drop height compared to or ( ~90% of 's). Morphological classifications rely on terrestrial analogs, yet extraterrestrial features preserve pristine forms longer absent fluvial modification, aiding reconstruction of paleoenvironments but challenging direct frequency estimates due to incomplete mapping and dating. Such insights inform models of under varied gravitational and atmospheric regimes, revealing that while initiation thresholds may align via similarities, propagation and cessation differ markedly, with implications for hazard assessment on airless or low- worlds.

Mitigation Strategies

Engineering Interventions

Engineering interventions for target slope destabilization by reducing driving forces, enhancing shear resistance, or controlling water, thereby increasing the against failure. Primary methods include drainage to lower pore pressures, geometric modifications like excavation or buttressing, and structural reinforcements such as retaining walls and anchors. These approaches require site-specific geotechnical to identify controlling processes, such as rise or weak basal layers, as improper implementation can induce larger slides. Drainage systems form the foundation of many interventions, addressing as a key trigger. Surface measures, including ditches graded at a minimum 2% and wattles for runoff interception on gradients up to 70%, prevent and infiltration while promoting revegetation over 1-2 years. Subsurface techniques, such as horizontal slotted PVC drainpipes installed to intersect failure surfaces, reduce tables, with notable effects in clay soils emerging after 1-5 years of operation. Impermeable covers and minimized further limit water addition, directly countering pore pressure buildup. Structural reinforcements provide mechanical support, particularly at slide toes or scarps. Retaining walls vary by material and scale: baskets, wire mesh enclosures filled with cobbles, offer flexibility and permeability for heights up to 2.5 meters; timber crib walls, interlocking logs backfilled with aggregate, suit small volumes requiring 10-15% equivalent fill; and bin walls, corrugated panels with earth infill, achieve stability through self-weight with height-to-width ratios of 1:2 to 3:5. Piles, typically one per 50 cubic meters of soil, and rock-fill buttresses add toe resistance while aiding drainage in rotational failures. Geometric alterations modify inherent slope instability. Excavation at slide heads removes overburden to lessen driving forces, potentially raising the by 10-15% in deep rotational landslides, though it demands precise modeling to avoid deeper plane activation. Toe buttressing with or engineered fills counters basal shear, often combined with drainage for synergistic effects. Effectiveness data indicate concrete retaining walls reduce recurrence risks, with community surveys in prone areas rating them highly reliable by over 70% of respondents. Overall, integrated applications, informed by limit equilibrium analyses, yield optimal outcomes, though costs and site constraints dictate selection.

Early Warning and Monitoring Integration

Early warning systems for landslides integrate diverse monitoring technologies with predictive modeling to detect precursors such as ground deformation, pore water pressure changes, and rainfall intensity, enabling alerts before failure occurs. Ground-based instruments, including inclinometers, extensometers, and piezometers, provide real-time measurements of slope movement and subsurface hydrology at specific sites. Remote sensing methods, such as Interferometric Synthetic Aperture Radar (InSAR) from satellites like Sentinel-1, offer wide-area deformation mapping with millimeter accuracy over weeks to months, complementing in-situ data for broader coverage. Integration platforms often employ Internet of Things (IoT) networks to aggregate sensor data, which is then analyzed using machine learning algorithms for anomaly detection and forecasting. Localized landslide early warning systems (LEWS) fuse hydrogeotechnical sensor inputs with rainfall forecasts to compute stability thresholds, reducing false positives by incorporating and dynamics. For instance, systems in and utilize rainfall intensity-duration thresholds calibrated against historical events, dynamically adjusted via real-time monitoring to issue tiered alerts. A replicable LEWS design, as implemented in experimental sites, deploys horizontal and vertical sensor lines across slopes alongside autonomous point sensors to track multi-dimensional displacements, feeding data into stability models for probabilistic warnings. Regional systems extend this by overlaying satellite-derived estimates with susceptibility maps, as in USGS prototypes that predict shallow landslides hours in advance during storms. Effectiveness hinges on robust data fusion and model validation; peer-reviewed evaluations show integrated hydrologic monitoring can lower alarm rates by up to 30% compared to rainfall-only triggers, though challenges persist in data latency and sensor durability in remote areas. Fiber-optic distributed sensing emerges as a durable alternative, embedding cables in slopes to monitor strain over kilometers continuously, integrated into European coastal LEWS for early detection. Despite advancements, empirical studies underscore that no system guarantees zero failures, with success measured by evacuation efficacy rather than perfect prediction, as evidenced by reduced casualties in monitored Italian sites post-2010 implementations. Ongoing integration of AI-driven multi-source models promises refined predictions, but requires rigorous field to avoid over-reliance on unverified simulations.

Land-Use Policies and Effectiveness Critiques

Land-use policies for primarily involve regulatory that designates high-susceptibility areas as unsuitable for development, often based on GIS-derived hazard maps incorporating factors like slope angle, , and historical failures. These policies typically mandate geotechnical assessments for proposed , enforce setbacks from steep slopes, and promote vegetation retention or restoration to stabilize soils, as and impervious surfaces exacerbate slope instability through increased runoff and reduced root cohesion. In the United States, state-level strategies, such as those outlined in the National Landslide Hazards Mitigation Strategy, emphasize integrating such into local planning to curb exposure from land-use changes, which empirical analyses link to heightened landslide frequency. Effectiveness of these policies hinges on accurate delineation and ; studies indicate that preserving forested cover can modestly reduce susceptibility compared to alternative land covers like vineyards or bare , with systems providing mechanical reinforcement against shallow failures. For instance, expansions and conservation in vulnerable mountain regions have demonstrated cost-effective risk reductions by limiting human-induced triggers like or . However, direct of zoning's impact remains sparse, with susceptibility models showing policies succeed in averting new exposures but falter against legacy developments predating regulations. Critiques highlight implementation shortcomings, including weak enforcement and political overrides that permit hazardous development despite , as seen in cases where urban expansion outpaces efforts, amplifying exposure beyond rates. Inaccurate or overly generalized mapping—often reliant on data prone to interpretive errors—can miszone properties, either underprotecting through false negatives or imposing undue restrictions that invite legal challenges over property takings without commensurate safety gains. Regulatory bodies face liability pitfalls if zones fail to prevent events like the 2014 Oso landslide, where pre-existing conditions and triggers evaded controls, underscoring how policies mitigate static risks but inadequately address dynamic factors such as rainfall or human alterations. In developing contexts, poverty-driven informal settlements routinely bypass , rendering formal policies ineffective absent complementary socioeconomic measures. Broader economic critiques argue that stringent constrains land availability, elevating costs and development pressures that indirectly encourage riskier siting elsewhere, though causal tying this to landslide outcomes is limited. Networks integrating with and monitoring outperform isolated regulatory approaches, as standalone policies often yield compliance gaps due to public skepticism or incomplete hazard communication. Ultimately, while land-use controls address root causes like slope loading from buildings, their variable success reflects enforcement realities over theoretical promise, with calls for refined risk criteria to balance safety against overregulation.

Broader Phenomena and Impacts

Landslides entering coastal or lacustrine waters can displace substantial volumes of water, generating localized tsunamis or megatsunamis with runup heights far exceeding those from tectonic sources. These events differ from earthquake-driven tsunamis in their shorter wavelengths and rapid decay, often confined to fjords or enclosed basins, yet capable of devastating wave heights due to direct mass displacement. landslides, in particular, contribute to open-ocean tsunamis by mobilizing volumes equivalent to thousands of cubic kilometers, propagating waves across basins. The 1958 Lituya Bay event in exemplifies this hazard, where an earthquake-triggered of approximately 30 million cubic meters plunged into the , producing a with a maximum runup of 530 meters on the opposing shore—the highest instrumentally recorded . This single event destroyed and eroded the landscape to up to that , though no fatalities occurred due to the remote . Similarly, the 1929 Grand Banks off Newfoundland, triggered by a magnitude 7.2 , displaced and generated a that killed 28 people along the , with waves up to 13 meters in some areas. In 2015, the Taan Fiord in released about 180 million cubic meters of material into the , yielding a with runups exceeding 120 meters and waves propagating at speeds over 100 km/h. Beyond tsunamis, landslides frequently form temporary by blocking channels, impounding upstream lakes that may breach catastrophically, unleashing outburst with peak discharges orders of magnitude greater than typical flows. These landslide-dammed lake outbursts (LLLOs) erode downstream channels rapidly, amplifying flood peaks through entrainment and posing risks to hundreds of kilometers away. The mechanism involves progressive overtopping or failure of the debris , often within days to months of formation, depending on inflow rates and dam permeability. Notable examples include the 2018 Baige landslides on the Jinsha River in China, where two successive events each impounded over 300 million cubic meters of water; the first dam breached on October 11, producing a flood wave that traveled 100 km downstream at 20-30 m/s, eroding valleys but causing no reported deaths due to evacuations. The Yigong River in Tibet experienced similar superfloods in 1902 and 2000 from landslide dams, each releasing over 10^10 cubic meters of water with peak flows exceeding 100,000 m³/s, reshaping 80 km of valley and depositing thick sediment layers observable in modern geomorphology. These secondary floods highlight how landslides compound hazards through delayed but high-magnitude releases, distinct from immediate debris flows.

Socioeconomic and Human Costs

Landslides exact severe human tolls through direct fatalities, injuries, and long-term displacement. Globally, these events have caused tens of thousands of deaths over recent decades, with 55,997 fatalities recorded in 4,862 distinct landslide events from 2004 to 2016, averaging approximately 4,571 deaths per year. In the United States, landslides kill an average of 25 to 50 people annually. Injuries number in the thousands worldwide over similar periods, as seen in analyses of 3,876 landslides from 1995 to 2014 that resulted in 11,689 reported injuries alongside 163,658 deaths. Such casualties disproportionately affect vulnerable populations in steep, rainfall-prone terrains, often compounded by rapid-onset debris flows or rockfalls that provide little warning. Displacement from landslides disrupts communities, forcing evacuations and resettlement that strain local resources and social structures. In addition to immediate losses, survivors face heightened risks of secondary hazards like outbreaks in temporary shelters or , though quantitative data on these indirect human impacts remains limited compared to fatality statistics. Empirical records indicate that landslides trigger mass migrations in hazard-prone areas, with events destroying homes and livelihoods, as evidenced by socioeconomic assessments in regions like the where development in unstable slopes amplifies exposure. Socioeconomic damages encompass direct costs to , , and , alongside indirect burdens such as halted and elevated reconstruction expenses. Annual global economic losses from landslides are estimated at $20 billion, reflecting damages to roads, bridges, and settlements that impede connectivity and development. In the , these losses range from $2 to $4 billion yearly, often underreported due to unquantified indirect effects like productivity declines. Nations including the , , , and each sustain over $1 billion in annual landslide-related costs, primarily from repairing transport networks and mitigating recurrent threats. These costs extend to agricultural sectors through soil erosion and crop destruction, reducing and export revenues in rural economies, while urban areas grapple with halted industrial operations and depreciated values. Peer-reviewed evaluations highlight that rehabilitation efforts post-landslide often exceed initial damages due to persistent , underscoring the need for of failures over generalized attributions. Overall, landslides' economic footprint reveals inefficiencies in land-use practices, where unaddressed geological risks perpetuate cycles of loss in both developed and developing contexts.

Global Disparities in Vulnerability

Vulnerability to landslides manifests profound global disparities, with human fatalities concentrated in low-income developing countries, particularly in Asia and Latin America, where thousands perish annually despite comprising a smaller share of worldwide economic damages. Between 2004 and 2016, Asia dominated global fatal landslide occurrences, accounting for the bulk of 55,997 deaths across 4,862 events. In contrast, wealthier nations experience fewer deaths—such as 25–50 per year in the United States—but incur higher monetary losses due to the concentration of valuable infrastructure in at-risk zones. Annual economic damages in countries like the United States, Japan, Italy, and India each surpass $1 billion, reflecting the higher asset values exposed rather than elevated frequency or lethality per event. These patterns stem from geographic and socioeconomic factors: endures the highest fatality rates, propelled by monsoon-driven rainfall on densely populated steep terrains, while Central and South American cordilleras amplify risks through seismic activity and tropical storms. Developing regions suffer elevated human exposure because informal settlements encroach on unstable slopes amid rapid , compounded by and agricultural practices that erode soil stability. Low —manifest in absent building codes, deficient early warning systems, and limited engineering interventions—further heightens mortality, as populations lack resources to relocate or reinforce vulnerable sites. Global risk mapping reveals that about 13% of Earth's land surface qualifies as very high susceptibility, primarily along the Andean chain in and associated cordilleras, where tectonic uplift and heavy intersect with human habitation. An estimated 39 million people face annual exposure to rainfall-triggered landslides alone, with disproportionate burdens on tropical and temperate mountainous zones housing 12% of the across 24% of land area. In low-GDP nations, urban clustering in hazard hotspots exacerbates outcomes, as evidenced by clustering of deadly events in areas with minimal per-capita in resilience measures. While natural triggers like intense dominate universally, anthropogenic modifiers such as unregulated construction disproportionately impede recovery and prevention in resource-scarce settings, perpetuating cycles of vulnerability.

Environmental Influences and Debates

Natural Climatic Variability

variability constitutes a primary natural climatic driver of landslides, as episodic heavy rainfall or prolonged wet periods elevate levels and pore water pressures within slopes, thereby diminishing and along failure planes. Empirical studies establish intensity-duration thresholds beyond which landslides initiate, with short-duration intense storms (>50 mm/hour) often triggering shallow debris slides, while antecedent cumulative rainfall over weeks to months predisposes deeper translational failures by saturating . For instance, regional analyses reveal that landslides correlate strongly with rainfall exceeding 100-200 mm in 24-72 hours in tectonically active terrains, reflecting inherent hydrological responses independent of long-term trends. Interannual oscillations, such as the El Niño-Southern Oscillation (ENSO), modulate landslide frequency through alterations in regional precipitation regimes, with El Niño phases linked to heightened activity in equatorial Pacific margins. During strong El Niño events, like those in 1997-1998 and 2015-2016, anomalous wet conditions in and parts of amplified landslide exposure, surpassing seasonal norms and affecting millions via infrastructure disruption and fatalities. Conversely, La Niña phases may suppress events in those areas but elevate risks elsewhere through drought-rain sequences that crack and then saturate soils. These patterns underscore causal linkages via rainfall anomalies rather than uniform global intensification. Seasonal climatic cycles further influence landslide timing, particularly in temperate and alpine environments where spring infiltrates thawed upper soils, raising hydrostatic pressures and reactivating dormant slides. In the , for example, USGS monitoring documents peak velocities in earthflows during March-May, coinciding with temperatures rising above 0°C and , as observed in sites like the Slumgullion landslide where velocities surge 2-5 times baseline rates. In periglacial zones, recurrent freeze-thaw cycles mechanically disintegrate and aggregates, fostering progressive instability; each cycle expands ice lenses by up to 9% volume, generating microfractures that culminate in block failures after 10-50 iterations under natural winter-spring transitions. Such processes dominate pre-instrumental records, evidencing landslides as intrinsic geomorphic responses to orbital and atmospheric variabilities.

Human Landscape Modifications

Human modifications to landscapes, including , , and , compromise by diminishing vegetative reinforcement, altering subsurface support, redirecting hydrological flows, and imposing additional loads on inclined . These interventions elevate the baseline susceptibility to , particularly when combined with natural precipitants such as intense rainfall. Analyses of global fatal events from 2004 to 2016 reveal that approximately 15% of 4,800 documented landslides were directly triggered by human activities like and , with an observable upward trend in their incidence relative to natural-only events. Deforestation erodes the mechanical anchoring provided by root systems, which can constitute up to 30-50% of soil on vegetated slopes, while also accelerating and soil saturation. Modeling of rainfall-induced shallow landslides across multiple sites indicates consistent increases in occurrence following clearance, with conversion to or amplifying erosion-prone failures. In the Far-Western of , episodes 5-7 years antecedent to events were associated with a 16% enhancement in landslide risk, attributable to diminished cohesion in exposed . Empirical inventories further link harvesting to elevated landslide frequencies in steep, humid terrains, where roads exacerbate incision and upslope drainage concentration. Mining operations destabilize slopes through selective material removal and accumulation, often inducing progressive deformation over years. Underground extraction in karstic regions of has precipitated large-scale rocky landslides, as evidenced by case studies where prolonged void creation and pillar reduced resistance, culminating in failures displacing millions of cubic meters. Open-pit methods similarly contribute via bench undercutting and dump saturation, with documented instances in and districts showing failure planes propagating along weakened discontinuities. These anthropogenic excavations lower the below unity under modest hydrological loading, contrasting with intact geological profiles. Urbanization intensifies risks via cut-and-fill earthworks, impervious surfacing that promotes rapid infiltration contrasts, and structural surcharges on marginally stable hillslopes. Panel regressions applied to precipitation-landslide datasets in the quantify that urban development amplifies hazard rates for equivalent storm intensities, with developed zones exhibiting 2-5 times the failure probability of forested equivalents due to modified pore pressures and reduced permeability. Community-scale modeling of expansion scenarios identifies slope excavation followed by vegetation stripping as the dominant destabilizers, reducing stability margins by up to 20-30% in simulated profiles. Informal settlements on peripheral slopes further compound vulnerabilities, as unengineered grading and poor drainage empirically correlate with heightened rainfall-triggered debris mobilization in tropical urban fringes.

Climate Change Hypotheses: Evidence and Skepticism

Hypotheses positing a link between anthropogenic and increased landslide frequency primarily focus on altered patterns, such as more intense rainfall events, which exceed infiltration capacities and trigger slope failures. Additional mechanisms include thaw in high-latitude and alpine regions, destabilizing frozen ground and leading to deeper-seated slides, as well as glacier retreat exposing unstable bedrock. Sea-level rise is also invoked for coastal areas, where undermines cliffs, though this interacts with local and wave action. These claims often rely on global circulation models projecting future extremes, but empirical attribution to historical trends remains contested due to sparse global inventories predating systematic monitoring around the . Supporting evidence is regionally variable and often correlative rather than causally definitive. In the , studies document rising winter rainfall and elevated snowlines since the 1980s, correlating with heightened shallow landslide risks during rain-on-snow events, as observed in increased activity from 2008 to 2014 across select countries. In Alaska's discontinuous zones, analysis of inventory data from 1985–2020 shows a 5–12% expansion in landslide-prone areas tied to warmer temperatures and shifting freeze-thaw cycles, with winter months exhibiting the strongest trends. High Mountain Asia projections indicate intensified rains could amplify deep-seated landslides by mid-century, based on downscaled climate models integrated with assessments. However, these findings are drawn from limited datasets, with global reviews noting underrepresentation of tropical and arid zones, where over 70% of historical fatalities occur but long-term records are inadequate. Peer-reviewed syntheses emphasize that while extreme precipitation has risen in some mid-latitude bands (e.g., 7% intensity increase per degree of warming per Clausius-Clapeyron relation), landslide responses depend on antecedent , , and , complicating direct attribution. Skepticism arises from the paucity of unambiguous global trends and variables overshadowing climatic signals. Comprehensive reviews of landslide-climate interactions find no consistent worldwide increase in or magnitude attributable to warming, with observed upticks in reporting often reflecting improved detection technologies and encroachment into zones rather than geophysical shifts. In drier regions like parts of , wet-year accelerations in slow-moving landslides contrast with dry-year stabilizations, but overall sensitivities mirror pre-industrial variability, suggesting natural cycles dominate over linear warming effects. Human modifications—, road-building, and —account for up to 90% of rain-induced triggers in inventory-based analyses, dwarfing modeled climate contributions in quantitative assessments. Projections of risks frequently extrapolate from regional models without validating against paleo-records, which indicate mega-landslides during cooler epochs, underscoring that causal chains are nonlinear and modulated by non-climatic forcings like seismic activity. Academic biases toward alarmist narratives in climate-impact literature may inflate perceived linkages, as evidenced by selective geographic focus and underemphasis on stabilizing feedbacks like enhanced in moderately warmer regimes. Thus, while localized enhancements are plausible, claims of a dominant global escalation lack robust, disaggregated empirical support.

References

  1. https://www.usgs.gov/faqs/what-a-landslide-and-what-causes-one
  2. https://www.usgs.gov/media/images/types-landslides
  3. https://www.usgs.gov/programs/landslide-hazards/where-do-landslides-occur
  4. https://www.usgs.gov/faqs/how-many-deaths-result-landslides-each-year
  5. https://gpm.[nasa](/page/NASA).gov/applications/landslides
  6. https://www.usgs.gov/programs/landslide-hazards/science
  7. https://www.usgs.gov/programs/landslide-hazards/landslide-basics
  8. https://pubs.usgs.gov/circ/1325/pdf/C1325_508.pdf
  9. https://elementaryengineeringlibrary.com/civil-engineering/soil-mechanics/shear-strength-of-soil-and-mohr-coulomb-theory/
  10. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019JF005461
  11. https://www.sciencebuddies.org/science-fair-projects/project-ideas/EnvEng_p035/environmental-engineering/landslides-what-causes-rocks-to-slide-down-a-slope
  12. https://pubs.geoscienceworld.org/gsa/geology/article/49/4/392/593340/Mechanistic-insights-from-emergent-landslides-in
  13. https://www.nature.com/articles/s41467-021-26959-5
  14. https://pubs.usgs.gov/publication/70033710
  15. https://www.geological-digressions.com/mohr-coulomb-failure-criteria/
  16. https://www.sciencedirect.com/science/article/pii/S2468227624000152
  17. https://pubs.usgs.gov/of/2000/0228/report.pdf
  18. https://ascelibrary.org/doi/abs/10.1061/%2528ASCE%25291090-0241%25282005%2529131%253A5%2528575%2529
  19. https://www.sciencedirect.com/topics/engineering/mohr-coulomb-failure
  20. https://pubs.geoscienceworld.org/gsa/geology/article/43/1/11/131646/Coseismic-landslides-reveal-near-surface-rock
  21. https://www.youtube.com/watch?v=a33rmtNZXKg
  22. https://pmc.ncbi.nlm.nih.gov/articles/PMC3776388/
  23. https://www.sciencedirect.com/science/article/abs/pii/S0341816221000552
  24. https://pubs.usgs.gov/fs/2004/3072/fs-2004-3072.html
  25. https://nhess.copernicus.org/preprints/nhess-2020-87/nhess-2020-87.pdf
  26. https://www.usgs.gov/publications/landslide-triggers-and-types
  27. https://www.sciencedirect.com/science/article/pii/S2405844023104555
  28. https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2018RG000626
  29. https://onlinepubs.trb.org/Onlinepubs/sr/sr247/sr247-004.pdf
  30. https://www.sciencedirect.com/science/article/pii/S0012825216302227
  31. https://www.usgs.gov/faqs/do-human-activities-cause-landslides
  32. https://blogs.agu.org/landslideblog/2018/08/24/global-fatal-landslide-occurrence-1/
  33. https://www.sciencedirect.com/science/article/pii/S0341816222006130
  34. https://www.fao.org/4/i3245e/i3245e.pdf
  35. https://news.agu.org/press-release/urban-areas-more-likely-to-have-precipitation-triggered-landslides/
  36. https://www.nature.com/articles/d44148-022-00192-8
  37. https://www.sciencedirect.com/science/article/abs/pii/S2095927324004833
  38. https://www.researchgate.net/publication/369437132_An_Overview_of_Causes_of_Landslides_and_Their_Impact_on_Transport_Networks
  39. https://www.mdpi.com/2073-4441/17/9/1320
  40. https://pubs.oregon.gov/dogami/fs/landslide-inv-factsheet.pdf
  41. https://www.wfpa.org/news-resources/blog/deep-seated-landslides-shallow-landslides-washington/
  42. https://pubs.usgs.gov/mf/1406/report.pdf
  43. https://pubs.usgs.gov/fs/2005/3107/pdf/FS-3107.pdf
  44. https://www.dnr.wa.gov/washington-geological-survey/geologic-hazards-and-environment/landslides
  45. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019JF005163
  46. https://link.springer.com/article/10.1007/s10346-023-02131-6
  47. https://www.sciencedirect.com/science/article/abs/pii/S0169555X23003057
  48. https://www.nature.com/articles/s41598-023-35088-6
  49. https://www.researchgate.net/publication/263340148_The_Varnes_classification_of_landslide_types_an_update
  50. https://moodle2.units.it/pluginfile.php/454811/mod_resource/content/1/02_A%2520suggested%2520method%2520for%2520describing%2520the%2520rate%2520of%2520movement%2520of%2520a%2520landslide.pdf
  51. https://www.aybu.edu.tr/bolumroot/contents/muhendislik_insaat/files/Landslides_types_and_causes%281%29.pdf
  52. http://www.geology.cz/projekt681900/vyukove-materialy/4_Lanslide_glossary.pdf
  53. https://www.researchgate.net/figure/Scheme-of-the-state-of-activity-of-a-landslide-modified-from-WP-WLI-1993-1-Active_fig4_303377557
  54. https://onlinepubs.trb.org/Onlinepubs/sr/sr247/sr247-003.pdf
  55. https://www.sciencedirect.com/science/article/pii/S0012825217305652
  56. https://www.sciencedirect.com/science/article/pii/S167498712030205X
  57. https://www.sciencedirect.com/science/article/pii/S2405844023051800
  58. https://onlinelibrary.wiley.com/doi/10.1002/gj.4666
  59. https://link.springer.com/chapter/10.1007/978-3-031-56591-5_4
  60. https://www.sciencedirect.com/science/article/pii/S1674775524000714
  61. https://www.nature.com/articles/s41598-024-57964-5
  62. https://www.science.org/doi/10.1126/sciadv.adt1541
  63. https://www.sciencedirect.com/science/article/pii/S277323042300001X
  64. https://www.sciencedirect.com/science/article/pii/S1674987124001105
  65. https://nhess.copernicus.org/articles/23/205/2023/
  66. https://pubs.usgs.gov/of/2001/0324/report.pdf
  67. https://rosap.ntl.bts.gov/view/dot/72895/dot_72895_DS1.pdf
  68. https://www.mdpi.com/2072-4292/14/16/3879
  69. https://dspace.mit.edu/handle/1721.1/33690
  70. https://ntrs.nasa.gov/api/citations/20230015388/downloads/PHELS_NHESS.pdf
  71. https://ui.adsabs.harvard.edu/abs/2025AuGmJ..60...61P/abstract
  72. https://www.researchgate.net/publication/287797320_Landslide_risk_assessment_The_challenge_of_communicating_uncertainty_to_decision-makers
  73. https://pmc.ncbi.nlm.nih.gov/articles/PMC11844721/
  74. https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024AV001214
  75. https://www.nature.com/articles/s43017-022-00373-x
  76. https://www.sciencedirect.com/science/article/pii/S0013795223001746
  77. https://www.sciencedirect.com/science/article/abs/pii/S0034425720303539
  78. https://link.springer.com/article/10.1007/s42452-025-07343-0
  79. https://nhess.copernicus.org/articles/22/1609/2022/
  80. https://www.nature.com/articles/s41598-025-89542-8
  81. https://onlinelibrary.wiley.com/doi/full/10.1002/esp.5775
  82. https://link.springer.com/article/10.1007/s11069-010-9634-2
  83. https://pubs.usgs.gov/circ/1325/pdf/Sections/AppendixB.pdf
  84. https://www.sciencedirect.com/science/article/pii/S0013795224004903
  85. https://www.mdpi.com/2072-4292/17/6/999
  86. https://pmc.ncbi.nlm.nih.gov/articles/PMC11281128/
  87. https://link.springer.com/chapter/10.1007/978-981-19-7245-4_15
  88. https://www.usgs.gov/programs/landslide-hazards/science/real-time-monitoring-potential-landslides
  89. https://www.geosense.com/product-category/inclinometers/
  90. https://www.engineering.pitt.edu/contentassets/dd2935b40f984059888016728b182527/7b2demirkan_instrumentation-for-landslide-monitoring-presentation.pdf
  91. https://www.encardio.com/blog/landslide-monitoring-methods-instrumentation-techniques
  92. https://rstinstruments.com/applications/infrastructure/landslide/
  93. https://www.geokon.com/Slope-Stability
  94. https://sisgeo.com/video/landslide-monitoring-instrumentation/
  95. https://nhess.copernicus.org/articles/24/3651/2024/
  96. https://www.sciencedirect.com/science/article/abs/pii/S0034425716303789
  97. https://pubs.geoscienceworld.org/ssa/tsr/article/4/1/43/634286/Toward-the-Rapid-Seismic-Assessment-of-Landslides
  98. http://earthquake.alaska.edu/new-detection-method-aims-warn-landslide-tsunamis
  99. https://www.sciencedirect.com/science/article/abs/pii/S0012825221000179
  100. https://www.nature.com/articles/s41598-024-84067-y
  101. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018RG000603
  102. https://www.sciencedirect.com/science/article/pii/001379529090082C
  103. https://library.seg.org/doi/10.1190/1.1440727
  104. https://eos.org/editors-vox/downhill-all-the-way-monitoring-landslides-using-geophysics
  105. https://pubs.geoscienceworld.org/aeg/eeg/article/31/3/193/652753/Prehistoric-Landslides-Significance-Recognition
  106. https://www.sciencedirect.com/science/article/abs/pii/S0277379108002102
  107. https://esurf.copernicus.org/articles/10/1185/2022/
  108. https://www.researchgate.net/publication/227696779_Lake_sediments_deposited_on_the_Flims_rockslide_mass_The_key_to_date_the_largest_mass_movement_of_the_Alps
  109. https://royalsocietypublishing.org/doi/10.1098/rsta.1988.0055
  110. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018JC014893
  111. https://geology.utah.gov/map-pub/survey-notes/utahs-ancient-mega-landslides/
  112. https://www.worldatlas.com/natural-disasters/the-deadliest-landslides-in-recorded-history.html
  113. https://timesofindia.indiatimes.com/etimes/trending/10-deadliest-landslides-in-recorded-history/photostory/112503121.cms
  114. http://people.3sr-grenoble.fr/users/bloret/enseee/ai/vaiont/Genevois-Ghirotti-2005.pdf
  115. https://www.cambridge.org/core/books/landslides/1963-vaiont-landslide-italy/7ADFCB5DB17DA3DA6242782705331276
  116. https://pubs.usgs.gov/circ/1970/0639/report.pdf
  117. https://www.britannica.com/event/Ancash-earthquake-of-1970
  118. https://pubs.geoscienceworld.org/aeg/eeg/article/18/4/313/60454/When-Landslides-Are-Misinterpreted-as-Faults-Case
  119. https://www.nytimes.com/2014/07/23/us/washington-mudslide-report-cites-rain-but-doesnt-give-cause-or-assign-blame.html
  120. https://www.nbcnews.com/storyline/deadly-mudslide/oso-mudslide-lawsuit-blames-state-logging-company-deadly-disaster-n291501
  121. https://theconversation.com/what-weve-learned-from-the-deadly-oso-washington-landslide-two-years-on-56528
  122. https://www.kiro7.com/news/logging-not-even-being-studied-cause-oso-slide/81917115/
  123. https://www.journal-rml.com/2/3/21/pdf
  124. https://damfailures.org/sites/default/files/wp-content/uploads/2019/02/The-prehistoric-Vajont-rockslide-An-updated-geological-model_Paronuzzi_2012.pdf
  125. http://ui.adsabs.harvard.edu/abs/2015EGUGA..17.3456B/abstract
  126. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020JE006675
  127. https://www.sciencedirect.com/science/article/pii/S0019103524003749
  128. https://pmc.ncbi.nlm.nih.gov/articles/PMC7857681/
  129. https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2024JE008453
  130. https://www.jpl.nasa.gov/images/pia00262-venus-landslide-in-navka-region/
  131. https://phys.org/news/2025-08-apollo-moon-sample-years-evidence.html
  132. https://airandspace.si.edu/newsroom/press-releases/shaking-moon-moonquake-triggered-landslides-and-boulder-falls
  133. https://meetingorganizer.copernicus.org/EPSC2015/EPSC2015-776.pdf
  134. https://www.researchgate.net/publication/272149103_Statistics_of_terrestrial_and_extraterrestrial_landslides
  135. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2024JE008453
  136. https://www.msss.com/files/Venus_Landslides.pdf
  137. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2017GL076113
  138. https://www.sciencedirect.com/science/article/abs/pii/S0019103521004917
  139. https://oxfordre.com/planetaryscience/display/10.1093/acrefore/9780190647926.001.0001/acrefore-9780190647926-e-254
  140. https://www.cambridge.org/core/books/landslides/landslides-on-other-planets/46995155FA22A20639526C2BAF051F8E
  141. https://pubs.usgs.gov/circ/1325/pdf/Sections/AppendixC.pdf
  142. https://pubs.usgs.gov/circ/1325/pdf/Sections/Section3.pdf
  143. https://link.springer.com/article/10.1007/s11069-023-05947-6
  144. https://www.encardio.com/blog/landslide-monitoring-and-early-warning-systems-best-practices
  145. https://www.mdpi.com/2220-9964/12/5/198
  146. https://nhess.copernicus.org/articles/25/169/2025/
  147. https://nhess.copernicus.org/preprints/nhess-2023-20/
  148. https://pubs.usgs.gov/publication/70262111
  149. https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2025.1570413/full
  150. https://www.sciencedirect.com/science/article/pii/S0012825219304635
  151. https://www.tandfonline.com/doi/full/10.1080/19475705.2025.2513536?af=R
  152. https://files.dnr.state.mn.us/waters/watermgmt_section/shoreland/landslide-mitigation.pdf
  153. https://link.springer.com/article/10.1007/s10346-022-02020-4
  154. https://pubs.usgs.gov/of/2000/ofr-00-0450/ofr-00-0450.pdf
  155. https://www.sciencedirect.com/science/article/abs/pii/S0341816202001704
  156. https://pmc.ncbi.nlm.nih.gov/articles/PMC8764696/
  157. https://www.sciencedirect.com/science/article/abs/pii/S2212420919313494
  158. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2025GL115170
  159. https://link.springer.com/article/10.1007/s10346-024-02451-1
  160. https://mrsc.org/stay-informed/mrsc-insight/february-2018/regulating-landslide-hazards-can-be-hazardous
  161. https://www.sciencedirect.com/science/article/abs/pii/S026483770900177X
  162. https://data.wvgis.wvu.edu/pub/RA/State/CL/Landslide/Brochures/LandslidesInYourCommunity_Brochure.pdf
  163. https://www.sciencedirect.com/science/article/abs/pii/S0197397520314223
  164. https://pubs.usgs.gov/of/2022/1075/ofr20221075.pdf
  165. https://geoenvironmental-disasters.springeropen.com/articles/10.1186/s40677-022-00205-6
  166. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2009jf001381
  167. https://www.sciencedirect.com/science/article/abs/pii/S002532270400324X
  168. https://www.usgs.gov/programs/landslide-hazards/science/2025-tracy-arm-landslide-generated-tsunami
  169. https://www.nps.gov/articles/aps-18-1-2.htm
  170. https://www.sciencedirect.com/science/article/abs/pii/S0022169421003413
  171. https://esurf.copernicus.org/articles/9/1251/2021/
  172. https://www.nature.com/articles/s41598-021-95194-1
  173. https://nhess.copernicus.org/articles/18/2161/2018/
  174. https://www.sciencedirect.com/science/article/abs/pii/S0048969719314214
  175. https://pubs.usgs.gov/publication/ofr01276
  176. https://conferences.iaia.org/2023/uploads/edited-presentations/35_Sim_an-Overview-of-Socioeconomic.pdf
  177. https://profession.americangeosciences.org/society/intersections/faq/how-much-do-landslides-cost-terms-monetary-losses/
  178. https://pubs.geoscienceworld.org/aeg/eeg/article/xxiii/1/11/137265/Economic-Losses-and-Fatalities-Due-to-Landslides
  179. https://geoenvironmental-disasters.springeropen.com/articles/10.1186/s40677-018-0104-6
  180. https://link.springer.com/article/10.1007/s11069-024-06487-3
  181. https://dornsife.usc.edu/news/stories/global-toll-from-landslides-is-heaviest-in-developing-countries/
  182. https://pmc.ncbi.nlm.nih.gov/articles/PMC7567151/
  183. https://www.sciencedirect.com/science/article/pii/S1674987125000799
  184. https://www.pnas.org/doi/10.1073/pnas.2113416118
  185. https://pmc.ncbi.nlm.nih.gov/articles/PMC10755328/
  186. https://www.usgs.gov/media/videos/impacts-increasing-precipitation-variability-landslide-triggering-san-francisco-bay
  187. https://www.nature.com/articles/s41467-021-22398-4
  188. https://www.researchgate.net/publication/350894485_Global_connections_between_El_Nino_and_landslide_impacts
  189. https://pubs.usgs.gov/of/1997/0027/report.pdf
  190. https://pubs.usgs.gov/of/2000/ofr-00-0102/
  191. https://www.sciencedirect.com/science/article/abs/pii/S0341816222006361
  192. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019WR025233
  193. https://nhess.copernicus.org/articles/15/1311/2015/nhess-15-1311-2015.pdf
  194. https://www.mdpi.com/2072-4292/14/23/6071
  195. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021GL094038
  196. https://www.issmge.org/uploads/publications/51/126/356_B_slope_stability_modelling_of_urbanisation_scenario.pdf
  197. https://www.sciencedirect.com/science/article/pii/S0012825216302458
  198. https://www.nature.com/articles/s41598-023-50314-x
  199. https://eos.org/features/a-slippery-slope-could-climate-change-lead-to-more-landslides
  200. https://www.nature.com/articles/s43247-025-02668-5
  201. https://climate-adapt.eea.europa.eu/en/observatory/evidence/health-effects/landslides
  202. https://research.noaa.gov/climate-change-could-trigger-more-landslides-in-high-mountain-asia/
  203. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2022GL099499
  204. https://www.researchgate.net/publication/222146219_Deciphering_the_effect_of_climate_change_on_landslide_activity_A_review
  205. https://www.jpl.nasa.gov/news/nasa-study-finds-climate-extremes-affect-landslides-in-surprising-ways/
  206. https://www.tandfonline.com/doi/full/10.1080/17486025.2025.2496302?src=exp-la
Add your contribution
Related Hubs
User Avatar
No comments yet.