Union of
Geological Sciences
Geoindicators  |  GEOIN  |  Publications  |  Events  |  Links
Applying geoindicators  |  Checklist  |  Image gallery  |  Scientific Contributors  |  Contact Us
  home  |  back  |  all documents   mail to friend  |  printer friendly version

Landslides and Avalanches

NAME: Landslides and Avalanches

BRIEF DESCRIPTION: There are many ways in which slopes may fail, depending on the angle of slope, the water content, the type of earth material involved, and local environmental factors such as ground temperature. Mass movements (landslides, mass wasting) may take place suddenly and catastrophically, resulting in debris and snow avalanches, lahars, rock falls and slides, flows (debris, quick clay, loess, and dry or wet sand and silt). For example, the initial velocity of mudflows can reach 30m/second in a few seconds, slowing to several m/day. Slower movements result in slides (debris, rock blocks), topples, slumps (rock, earth), complex landslides and creep. Landsliding is commonly regarded as one of the most predictable of geological hazards. Three parameters are particularly important for monitoring all kinds of mass movements.

Ground cracks are the surface manifestation of a variety of mass movements. In plan, they are commonly concentric or parallel, and have widths of a few centimeters and lengths of several meters, which distinguishes them from the much shorter desiccation cracks [see desert surface crusts and fissures]. The formation of cracks and any increase in their rate of widening is a common measure of impending slope failure.

The appearance of and increases in ground subsidence or upheaval is also a good measure of impending failure.

The area of slope failure is a measure of the extent of landsliding in any region. Changes over time may both reflect significant environmental stresses (e.g. deforestation, weather extremes) and provide important clues about landscape and ecosystem degradation.

Special conditions and processes exist in permafrost terrains. Landslides and mudflows of permafrost regions are mobilized and shaped by the freezing and thawing of pore water in the active layer, the base of which acts as a shear discontinuity. Failure here can occur on slopes as low as 1 . Gelifluction (a form of solifluction, the slow downslope movement of waterlogged soil and surficial debris) is the regular downslope flow or creep of seasonally frozen and thawed soils. Gentle to medium slopes with blankets of loose rock fragments overlying frozen ground may be subject to mass movements such as rock glaciers and rock streams or kurums [see frozen ground activity]. Catastrophic slope failure here can expose new frozen ground, setting off renewed mass wasting.

Climate change may accelerate or slow the natural rate of slope failure, through changes in precipitation or in the vegetation cover that binds loose slope materials: wildfires can also promote mass movements by destroying tree cover. However, it is difficult to generalize where information is lacking on the present distribution and significance of landslides, and because many parameters, in addition to climate change, contribute to slope stability.

SIGNIFICANCE: Thousands of people are killed each year by landslides: in China and Peru, tens of thousands of deaths have resulted from single landslides. Annual property damage from landslides worldwide is estimated in the tens of billions of dollars, with more than $1.5 billion in annual losses in the USA alone. There are innumerable small to medium-size slope failures that cumulatively impose costs to society as great or greater than the large infrequent catastrophic landslides that draw so much attention. Damage to ecosystems has not generally been documented, but landslides may destroy habitats, for example by blocking streams and denuding slopes.

HUMAN OR NATURAL CAUSE: Slope failure is a natural process which can be induced, accelerated or retarded by human actions. Many causes are involved including:

Removal of lateral support through the erosive power of streams, glaciers, waves, and longshore and tidal currents; through weathering, and wetting, drying and freeze-thaw cycles in surficial materials; through land subsidence or faulting that creates new slopes; and through human actions such as cutting slopes for roads and other structures, quarrying, removal of retaining walls, and lowering of reservoirs.

Adding weight naturally to slopes by rain, hail, snow and water from springs, by accumulation of talus or volcanic debris, and by human actions such as landfills, stockpiles of ore or rock, waste piles, construction of heavy building and other structures, and water leaking from pipelines, sewers, canals, and reservoirs.

Earthquakes, thunder, vibrations from nearby slope failures, and human activities such as vibrations from explosions, machinery, road and air traffic.

Regional tilting that increases slope angles.

Decrease of underlying support by subrosion and removal of granular and soluble materials [see karst activity], mining, loss of strength or failure and/or squeezing out of underlying material.

Lateral pressure from water in cracks and caverns, freezing of water in cracks, hydration of minerals, and mobilization of residual stress.

Volcanic processes that modify ground and rock stresses, such as inflation or deflation of magma chambers, fluctuations in lava-lake levels, and increase in ground tremors.

ENVIRONMENT WHERE APPLICABLE: Landslides are most common on moderate to steep slopes worldwide, but even gentle and flat-lying slopes may fail where adjacent to steep slopes, rivers, and other bodies of water. The risk of failure is generally greater where rocks are highly fractured, or where there are surficial soils, clays and silts that are liable to liquefaction. Many pre-existing landslides are re- activated, even under conditions that the original slope, prior to first failure, could have resisted.

TYPES OF MONITORING SITES: The highest part (crown) of landslides and other potential failures is generally the most important place for monitoring cracks, subsidence and sagging. Upheaval or buckling generally begins in the toe area. As failure progresses and the slide or flow develops, cracks and ground subsidence may form at any point including the toe: one landslide in Japan formed cracks and uplifted a railroad tunnel in the toe area over 1 km away and on the other side of a river from the crown.

SPATIAL SCALE: patch (cracks, subsidence and upheaval) to landscape (areal extent) / mesoscale to continental

METHOD OF MEASUREMENT: Surface methods for measuring the development of cracks, subsidence and uplift include repeated conventional surveying, installation of various instruments to measure movements directly, and tiltmeters to records changes in slope inclination near cracks and areas of greatest vertical movements. Subsurface methods include installation of inclinometers and rock noise instruments to record movements near cracks and other areas of ground deformation, bucket auger holes large enough to accommodate a person who locates, records and monitors cracking and deformation at depth, and geophysical techniques for locating shear surfaces throughout the landslide area. The areal extent of landsliding over large areas is most effectively determined using air photos. Satellite images may be helpful in identifying large landslides and in noting changes in soil and vegetation cover that may be associated with landsliding. Conversion of landslide extent from photos to a digital database will permit easy measurement of areal changes for each type of landslide deposit. At higher latitudes, color infrared photos at scales of 1:25,000 to 1:50,000 are best for most landslides, if taken in early spring or late fall (no snow or deciduous leaves) when sun angle is high and shadows are at minimum.

FREQUENCY OF MEASUREMENT: The frequency of monitoring will largely be dictated by changes in rate of crack propagation and ground deformation and by the degree of potential damage if the slope fails. For example, urban or industrial areas of high risk and rapid change will require continuous monitoring and, probably, the installation of automatic warning devices such as sirens and barriers. Areas of low risk and very slow development of cracks and ground deformation can be examined much less frequently. Critical periods for monitoring are during and immediately after intense rains and rapid snowmelt: real-time monitoring of rainfall from telemetered rain gauges and precipitation forecasts may be very important. Once a comprehensive and reliable baseline of past and present landslide activity has been established, the measurement should be repeated after hurricanes and other significant snowfall/snowmelt and rainfall events, or after fires, deforestation or human activities that extensively modify the land surface. If little activity has taken place in a particular area, re-assessment can be delayed for several years or more.

LIMITATIONS OF DATA AND MONITORING: The factors influencing slope stability are numerous and often complex, and monitoring of cracks may provide little insight in dealing with particular landslides. Many cracks form and are followed by landsliding within seconds, as do many areas with active subsidence and upheaval, so that monitoring these slopes will not help in warning local residents. Other landslides may form cracks, subside or buckle over long periods of time and then fail suddenly with little warning. Other methods may be more useful for predicting failure, such as regional landslide hazard mapping, rainfall monitoring, and monitoring of pore water pressure. In assessing changes in areal extent, the skill of the interpreter, the quality of the air photography, and factors such as cloud and vegetation cover, haze, and sun angle may greatly limit the usefulness of the data. Many landslides cannot be detected on available air photos, and if their scale is too large, the expense and time required to analyze them may be prohibitive. Radar images may be needed for tropical area where vegetation cover is extensive. Most of the features used to identify landslides are too small be to recognized on currently available satellite images.

APPLICATIONS TO PAST AND FUTURE: Dating cracks and their subsurface extension, subsidence and upheavals, and other features associated with landsliding can provide clues about past climatic changes and associated denudation rates, and the relative significance of rainfall, earthquake, and volcanic triggers for past slope failures. This knowledge, if aided by regional and temporal monitoring of crack development, can be used to forecast future landslide activity.

POSSIBLE THRESHOLDS: Slope failure takes place when the critical slope angle is exceeded. The angle depends on the frictional properties of the slope material and increases slightly with the size and angularity of the fragments. Dry, cohesionless material will come to rest on similar material when the angle of repose ranges generally between 33 and 37 . For wet, cohesive materials underlain by frozen ground, downslope movement may occur on slopes as low as 1 . A related threshold for permafrost is the freeze-thaw transition [see frozen ground activity]. In humid areas of unstable slopes, the cumulative and anticipated rainfall may reach empirical threshold values, at which time warnings of impending slope failure should be issued.


Berger, A.R. & W.J. Iams (eds) 1996. Geoindicators: Assessing rapid environmental changes in earth systems. Rotterdam: A.A. Balkema (see papers by Nesje, Rasch et al., Romanovskii et al.).

Brabb, E.E. 1984. Innovative approaches to landslide hazard and risk mapping. Proceedings of IV International Symposium on Landslides 1: 307-324.

Brabb, E.E. and B.L. Harrod 1989. Landslides - extent and economic significance. Rotterdam: Balkema.

Casale, R., R. Fantechi & J.C. Flageollet 1995. Temporal occurrence and forecasting of landslides in the European Community. European Commission, 2 vols.

Jones, D.K.C. & E.M. Lee 1994. Landsliding in Great Britain . London: HM Stationery Office

Novosad, S. & P. Wagner (eds) 1993. Landslides. Proceedings of 7th International Conference on Landslides. Rotterdam:Balkema. (especially papers by Glawe et al. and Moriwaki).

Keefer, D.K. 1987. Real-time landslide warning during heavy rainfall. Science , 238: 921-925.

Selby, M.J. 1993. Hillslope materials and processes . 2nd edition. Oxford: Oxford University Press.

Schuster, R.L. & R.J. Krizek (eds) 1978. Landslides, analysis and control . U.S. National Academy of Sciences Transportation Research Board Special Report 176. (especially papers by Varnes, and Wilson & Mikkelson - N.B. A new edition is in preparation).

OTHER SOURCES OF INFORMATION: Geological surveys, World Data Center-A for Natural Hazards, International Landslide Working Group (Dr. E. Brabb, U.S. Geological Survey MS 975, 345 Middlefield Rd., Menlo Park, CA 94025, USA), IAEG, IGA, IPA, ISSMFE (Dr. R.H.G. Parry, Secretary-General, International Society for Soil Mechanics and Foundation Engineering, University Engineering Department, Trumpington Street, Cambridge CB2 1PZ, UK), UNDRO, UNESCO.

RELATED ENVIRONMENTAL AND GEOLOGICAL ISSUES: Removal of forest and plant cover (fire, deforestation); land clearing and accelerated soil erosion in sloping terrain; potential destruction of ecosystems affected by slope failure.

OVERALL ASSESSMENT: Slope failure is one of the most widespread causes of land disturbance, so that the initiation and development of landslides should be closely monitored.

 © Geoindicators Initiative GEOIN