Permafrost (from perma-'permanent' and frost) issoil or underwatersediment which continuously remains below 0 °C (32 °F) for two years or more: the oldest permafrost has been continuously frozen for around 700,000 years.[1] Whilst the shallowest permafrost has a vertical extent of below a meter (3 ft), the deepest is greater than 1,500 m (4,900 ft).[2] Similarly, the area of individual permafrost zones may be limited to narrow mountainsummits or extend across vastArctic regions.[3] The ground beneathglaciers andice sheets is not usually defined as permafrost, so on land, permafrost is generally located beneath a so-calledactive layer of soil which freezes and thaws depending on the season.[4]
Around 15% of theNorthern Hemisphere or 11% of the global surface is underlain by permafrost,[5] covering a total area of around 18 million km2 (6.9 million sq mi).[6] This includes large areas ofAlaska,Canada,Greenland, andSiberia. It is also located in high mountain regions, with theTibetan Plateau being a prominent example. Only a minority of permafrost exists in theSouthern Hemisphere, where it is consigned to mountain slopes like in theAndes ofPatagonia, theSouthern Alps of New Zealand, or the highest mountains ofAntarctica.[3][1]
Permafrost contains large amounts of deadbiomass that has accumulated throughout millennia without having had the chance to fully decompose and release itscarbon, makingtundra soil acarbon sink.[3] Asglobal warming heats the ecosystem, frozen soil thaws and becomes warm enough for decomposition to start anew, accelerating thepermafrost carbon cycle. Depending on conditions at the time of thaw, decomposition can release eithercarbon dioxide ormethane, and thesegreenhouse gas emissions act as aclimate change feedback.[7][8][9] The emissions from thawing permafrost will have a sufficient impact on the climate to impact globalcarbon budgets. It is difficult to accurately predict how much greenhouse gases the permafrost releases because the different thaw processes are still uncertain. There is widespread agreement that the emissions will be smaller than human-caused emissions and not large enough to result inrunaway warming.[10] Instead, the annual permafrost emissions are likely comparable with global emissions fromdeforestation, or to annual emissions of large countries such asRussia, theUnited States orChina.[11]
Apart from its climate impact, permafrost thaw brings more risks. Formerly frozen ground often contains enough ice that when it thaws,hydraulic saturation is suddenly exceeded, so the ground shifts substantially and may even collapse outright. Many buildings and other infrastructure were built on permafrost when it was frozen and stable, and so are vulnerable to collapse if it thaws.[12] Estimates suggest nearly 70% of such infrastructure is at risk by 2050, and that the associated costs could rise to tens of billions of dollars in the second half of the century.[13] Furthermore, between 13,000 and 20,000 sites contaminated withtoxic waste are present in the permafrost,[14] as well as naturalmercury deposits,[15] which are all liable to leak and pollute the environment as the warming progresses.[16] Lastly, concerns have been raised about the potential forpathogenic microorganisms surviving the thaw and contributing to futurepandemics.[17][18] However, this is considered unlikely,[19][20] and ascientific review on the subject describes the risks as "generally low".[21]
Permafrost temperature profile. Permafrost occupies the middle zone, with the active layer above it, whilegeothermal activity keeps the lowest layer above freezing. The vertical 0 °C or 32 °F line denotes the average annual temperature that is crucial for the upper and lower limit of the permafrost zone, while the red lines represent seasonal temperature changes and seasonal temperature extremes. Solid curved lines at the top show seasonal maximum and minimum temperatures in the active layer, while the red dotted-to-solid line depicts the average temperature profile with depth of soil in a permafrost region.
Permafrost issoil,rock orsediment that is frozen for more than two consecutive years. In practice, this means that permafrost occurs at a mean annual temperature of 0 °C (32.0 °F) or below. In the coldest regions, the depth of continuous permafrost can exceed 1,400 m (4,600 ft).[22] It typically exists beneath the so-calledactive layer, which freezes and thaws annually, and so can support plant growth, as theroots can only take hold in the soil that's thawed.[2] Active layer thickness is measured during its maximum extent at the end of summer:[23] as of 2018, the average thickness in theNorthern Hemisphere is ~145 centimetres (4.76 ft), but there are significant regional differences. NortheasternSiberia,Alaska andGreenland have the most solid permafrost with the lowest extent of active layer (less than 50 centimetres (1.6 ft) on average, and sometimes only 30 centimetres (0.98 ft)), while southernNorway and theMongolian Plateau are the only areas where the average active layer is deeper than 600 centimetres (20 ft), with the record of 10 metres (33 ft).[24][25] The border between active layer and permafrost itself is sometimes called permafrost table.[26]
Around 15% ofNorthern Hemisphere land that is not completely covered by ice is directly underlain by permafrost; 22% is defined as part of a permafrost zone or region.[5] This is because only slightly more than half of this area is defined as a continuous permafrost zone, where 90%–100% of the land is underlain by permafrost. Around 20% is instead defined as discontinuous permafrost, where the coverage is between 50% and 90%. Finally, the remaining <30% of permafrost regions consists of areas with 10%–50% coverage, which are defined as sporadic permafrost zones, and some areas that have isolated patches of permafrost covering 10% or less of their area.[27][28]: 435 Most of this area is found in Siberia, northern Canada, Alaska and Greenland. Beneath the active layer annual temperature swings of permafrost become smaller with depth. The greatest depth of permafrost occurs right before the point where geothermal heat maintains a temperature above freezing. Above that bottom limit there may be permafrost with a consistent annual temperature—"isothermal permafrost".[29]
Permafrost typically forms in anyclimate where the mean annual air temperature is lower than the freezing point of water. Exceptions are found inhumid boreal forests, such as in NorthernScandinavia and the North-Eastern part ofEuropean Russia west of theUrals, where snow acts as an insulating blanket. Glaciated areas may also be exceptions. Since all glaciers are warmed at their base by geothermal heat,temperate glaciers, which are near thepressure melting point throughout, may have liquid water at the interface with the ground and are therefore free of underlying permafrost.[30] "Fossil" cold anomalies in thegeothermal gradient in areas where deep permafrost developed during the Pleistocene persist down to several hundred metres. This is evident from temperature measurements inboreholes in North America and Europe.[31]
The below-ground temperature varies less from season to season than the air temperature, with mean annual temperatures tending to increase with depth due to the geothermal crustal gradient. Thus, if the mean annual air temperature is only slightly below 0 °C (32 °F), permafrost will form only in spots that are sheltered (usually with a northern or southernaspect, in the north and south hemispheres respectively) creating discontinuous permafrost. Usually, permafrost will remain discontinuous in a climate where the mean annual soil surface temperature is between −5 and 0 °C (23 and 32 °F). In the moist-wintered areas mentioned before, there may not even be discontinuous permafrost down to −2 °C (28 °F). Discontinuous permafrost is often further divided into extensive discontinuous permafrost, where permafrost covers between 50 and 90 percent of the landscape and is usually found in areas with mean annual temperatures between −2 and −4 °C (28 and 25 °F), and sporadic permafrost, where permafrost cover is less than 50 percent of the landscape and typically occurs at mean annual temperatures between 0 and −2 °C (32 and 28 °F).[32]
In soil science, the sporadic permafrost zone is abbreviatedSPZ and the extensive discontinuous permafrost zoneDPZ.[33] Exceptions occur in un-glaciatedSiberia andAlaska where the present depth of permafrost is arelic of climatic conditions during glacial ages where winters were up to 11 °C (20 °F) colder than those of today.
At mean annual soil surface temperatures below −5 °C (23 °F) the influence of aspect can never be sufficient to thaw permafrost and a zone of continuous permafrost (abbreviated toCPZ) forms. A line of continuous permafrost in theNorthern Hemisphere[35] represents the most southern border where land is covered by continuous permafrost or glacial ice. The line of continuous permafrost varies around the world northward or southward due to regional climatic changes. In thesouthern hemisphere, most of the equivalent line would fall within theSouthern Ocean if there were land there. Most of theAntarctic continent is overlain by glaciers, under which much of the terrain is subject to basalmelting.[36] The exposed land of Antarctica is substantially underlain with permafrost,[37] some of which is subject to warming and thawing along the coastline.[38]
A range of elevations in both theNorthern andSouthern Hemisphere are cold enough to support perennially frozen ground: some of the best-known examples include theCanadian Rockies, theEuropean Alps,Himalaya and theTien Shan. In general, it has been found that extensive alpine permafrost requires mean annual air temperature of −3 °C (27 °F), though this can vary depending on localtopography, and some mountain areas are known to support permafrost at −1 °C (30 °F). It is also possible for subsurface alpine permafrost to be covered by warmer, vegetation-supporting soil.[39]
Alpine permafrost is particularly difficult to study, and systematic research efforts did not begin until the 1970s.[39] Consequently, there remain uncertainties about its geography As recently as 2009, permafrost had been discovered in a new area – Africa's highest peak,Mount Kilimanjaro (4,700 m (15,400 ft) above sea level and approximately 3° south of theequator).[40] In 2014, a collection of regional estimates of alpine permafrost extent had established a global extent of 3,560,000 km2 (1,370,000 sq mi).[34] However, by 2014, alpine permafrost in theAndes had not been fully mapped,[41] although its extent has been modeled to assess the amount of water bound up in these areas.[42]
Changes in subsea permafrost extent and structure between the Last Glacial Maximum and 2020[6]
Subsea permafrost occurs beneath theseabed and exists in thecontinental shelves of the polar regions.[2] These areas formed during the lastIce Age, when a larger portion of Earth's water was bound up inice sheets on land and when sea levels were low. As the ice sheets melted to again become seawater during theHolocene glacial retreat, coastal permafrost became submerged shelves under relatively warm and salty boundary conditions, compared to surface permafrost. Since then, these conditions led to the gradual and ongoing decline of subsea permafrost extent.[6] Nevertheless, its presence remains an important consideration for the "design, construction, and operation of coastal facilities, structures founded on the seabed,artificial islands,sub-sea pipelines, andwells drilled forexploration and production".[43] Subsea permafrost can also overlay deposits ofmethane clathrate, which were once speculated to be a majorclimate tipping point in what was known as aclathrate gun hypothesis, but are now no longer believed to play any role in projected climate change.[44]
At theLast Glacial Maximum, continuous permafrost covered a much greater area than it does today, covering all of ice-free Europe south to aboutSzeged (southeasternHungary) and theSea of Azov (then dry land)[45] and East Asia south to present-dayChangchun andAbashiri.[46] In North America, only an extremely narrow belt of permafrost existed south of theice sheet at about the latitude ofNew Jersey through southernIowa and northernMissouri, but permafrost was more extensive in the drier western regions where it extended to the southern border ofIdaho andOregon.[47] In theSouthern Hemisphere, there is some evidence for former permafrost from this period in centralOtago andArgentinePatagonia, but was probably discontinuous, and is related to the tundra. Alpine permafrost also occurred in theDrakensberg during glacial maxima above about 3,000 metres (9,840 ft).[48][49]
Permafrost extends to a base depth where geothermal heat from the Earth and the mean annual temperature at the surface achieve an equilibrium temperature of 0 °C (32 °F).[51] This base depth of permafrost can vary wildly – it is less than a meter (3 ft) in the areas where it is shallowest,[2] yet reaches 1,493 m (4,898 ft) in the northernLena andYana River basins inSiberia.[22] Calculations indicate that the formation time of permafrost greatly slows past the first several metres. For instance, over half a million years was required to form the deep permafrost underlyingPrudhoe Bay, Alaska, a time period extending over several glacial and interglacial cycles of thePleistocene.[50]: 18
Base depth is affected by the underlying geology, and particularly bythermal conductivity, which is lower for permafrost in soil than inbedrock.[51] Lower conductivity leaves permafrost less affected by thegeothermal gradient, which is the rate of increasing temperature with respect to increasing depth in the Earth's interior. It occurs as the Earth's internalthermal energy is generated byradioactive decay of unstableisotopes and flows to the surface by conduction at a rate of ~47terawatts (TW).[52] Away from tectonic plate boundaries, this is equivalent to an average heat flow of 25–30 °C/km (124–139 °F/mi) near the surface.[53]
Labelled example of a massive buried ice deposit inBylot Island, Canada[54]
When the ice content of a permafrost exceeds 250 percent (ice to dry soil by mass) it is classified as massive ice. Massive ice bodies can range in composition, in every conceivable gradation from icymud to pure ice. Massive icy beds have a minimum thickness of at least 2 m and a shortdiameter of at least 10 m.[55] First recorded North American observations of this phenomenon were by European scientists atCanning River (Alaska) in 1919.[56] Russian literature provides an earlier date of 1735 and 1739 during the Great North Expedition by P. Lassinius andKhariton Laptev, respectively. Russian investigators including I. A. Lopatin, B. Khegbomov, S. Taber and G. Beskow had also formulated the original theories for ice inclusion in freezing soils.[57]
While there are four categories of ice in permafrost – pore ice, ice wedges (also known as vein ice), buried surface ice and intrasedimental (sometimes also called constitutional[57]) ice – only the last two tend to be large enough to qualify as massive ground ice.[58][26] These two types usually occur separately, but may be found together, like on the coast ofTuktoyaktuk in westernArctic Canada, where the remains ofLaurentide Ice Sheet are located.[59]
Buried surface ice may derive from snow, frozen lake orsea ice,aufeis (stranded river ice) and even buried glacial ice from the formerPleistocene ice sheets. The latter hold enormous value for paleoglaciological research, yet even as of 2022, the total extent and volume of such buried ancient ice is unknown.[60] Notable sites with known ancient ice deposits includeYenisei River valley inSiberia, Russia as well asBanks andBylot Island in Canada'sNunavut andNorthwest Territories.[61][62][54] Some of the buried ice sheet remnants are known to hostthermokarst lakes.[60]
Intrasedimental or constitutional ice has been widely observed and studied across Canada. It forms when subterranean waters freeze in place, and is subdivided into intrusive, injection and segregational ice. The latter is the dominant type, formed after crystallizational differentiation in wetsediments, which occurs when water migrates to the freezing front under the influence ofvan der Waals forces.[56][55][58] This is a slow process, which primarily occurs insilts withsalinity less than 20% ofseawater: silt sediments with higher salinity andclay sediments instead have water movement prior to ice formation dominated byrheological processes. Consequently, it takes between 1 and 1000 years to form intrasedimental ice in the top 2.5 meters of clay sediments, yet it takes between 10 and 10,000 years forpeat sediments and between 1,000 and 1,000,000 years for silt sediments.[26]
Cliff wall of a retrogressive thaw slump located on the southern coast ofHerschel Island within an approximately 22-metre (72 ft) by 1,300-metre (4,300 ft) headwall.
Permafrost processes such asthermal contraction generating cracks which eventually becomeice wedges andsolifluction – gradual movement of soil down the slope as it repeatedly freezes and thaws – often lead to the formation of ground polygons, rings, steps and other forms ofpatterned ground found in arctic, periglacial and alpine areas.[63][64] In ice-rich permafrost areas, melting of ground ice initiatesthermokarst landforms such asthermokarst lakes, thaw slumps, thermal-erosion gullies, and active layer detachments.[65][66] Notably, unusually deep permafrost in Arcticmoorlands andbogs often attracts meltwater in warmer seasons, which pools and freezes to formice lenses, and the surrounding ground begins to jut outward at a slope. This can eventually result in the formation of large-scale land forms around this core of permafrost, such aspalsas – long (15–150 m (49–492 ft)), wide (10–30 m (33–98 ft)) yet shallow (<1–6 m (3 ft 3 in – 19 ft 8 in) tall)peatmounds – and the even largerpingos, which can be 3–70 m (10–230 ft) high and 30–1,000 m (98–3,281 ft) indiameter.[67][68]
A group ofpalsas, as seen from above, formed by the growth of ice lenses.
Only plants with shallowroots can survive in the presence of permafrost.Black spruce tolerates limited rooting zones, and dominatesflora where permafrost is extensive. Likewise, animalspecies which live in dens andburrows have their habitat constrained by the permafrost, and these constraints also have a secondary impact on interactions between species within theecosystem.[69]
Cracks forming at the edges of theStorflaket permafrost bog in Sweden
While permafrost soil is frozen, it is not completely inhospitable tomicroorganisms, though their numbers can vary widely, typically from 1 to 1000 million per gram of soil.[70][71]Thepermafrost carbon cycle (Arctic Carbon Cycle) deals with the transfer of carbon from permafrost soils to terrestrial vegetation and microbes, to the atmosphere, back to vegetation, and finally back to permafrost soils through burial and sedimentation due to cryogenic processes. Some of this carbon is transferred to the ocean and other portions of the globe through the global carbon cycle. The cycle includes the exchange ofcarbon dioxide andmethane between terrestrial components and the atmosphere, as well as the transfer of carbon between land and water as methane,dissolved organic carbon,dissolved inorganic carbon,particulate inorganic carbon andparticulate organic carbon.[72]
There are only two large cities in the world built in areas of continuous permafrost (where the frozen soil forms an unbroken, below-zero sheet) and both are in Russia –Norilsk inKrasnoyarsk Krai andYakutsk in theSakha Republic.[75] Building on permafrost is difficult because the heat of the building (orpipeline) can spread to the soil, thawing it. As ice content turns to water, the ground's ability to provide structural support is weakened, until the building is destabilized. For instance, during the construction of theTrans-Siberian Railway, asteam engine factory complex built in 1901 began to crumble within a month of operations for these reasons.[76]: 47 Additionally, there is nogroundwater available in an area underlain with permafrost. Any substantial settlement or installation needs to make some alternative arrangement to obtain water.[75][76]: 25
A common solution is placingfoundations on woodpiles, a technique pioneered by Soviet engineerMikhail Kim in Norilsk.[77] However, warming-induced change offriction on the piles can still cause movement throughcreep, even as the soil remains frozen.[78] TheMelnikov Permafrost Institute in Yakutsk found that pile foundations should extend down to 15 metres (49 ft) to avoid the risk of buildings sinking. At this depth the temperature does not change with the seasons, remaining at about −5 °C (23 °F).[79]
Globally, permafrost warmed by about 0.3 °C (0.54 °F) between 2007 and 2016, with stronger warming observed in the continuous permafrost zone relative to the discontinuous zone. Observed warming was up to 3 °C (5.4 °F) in parts ofNorthern Alaska (early 1980s to mid-2000s) and up to 2 °C (3.6 °F) in parts of the Russian European North (1970–2020). This warming inevitably causes permafrost to thaw:active layer thickness has increased in the European andRussian Arctic across the 21st century and at high elevation areas in Europe and Asia since the 1990s.[83]: 1237
Between 2000 and 2018, the average active layer thickness had increased from ~127 centimetres (4.17 ft) to ~145 centimetres (4.76 ft), at an average annual rate of ~0.65 centimetres (0.26 in).[24]
InYukon, the zone of continuous permafrost might have moved 100 kilometres (62 mi) poleward since 1899, but accurate records only go back 30 years. The extent of subsea permafrost is decreasing as well; as of 2019, ~97% of permafrost under Arctic ice shelves is becoming warmer and thinner.[84][10]: 1281
Based on high agreement across model projections, fundamental process understanding, and paleoclimate evidence, it is virtually certain that permafrost extent and volume will continue to shrink as the global climate warms, with the extent of the losses determined by the magnitude of warming.[83]: 1283
Permafrost thaw is associated with a wide range of issues, andInternational Permafrost Association (IPA) exists to help address them. It convenes International Permafrost Conferences and maintainsGlobal Terrestrial Network for Permafrost, which undertakes special projects such as preparing databases, maps, bibliographies, and glossaries, and coordinates international field programmes and networks.[85]
Permafrost peatlands (a smaller, carbon-rich subset of permafrost areas) under varying extent of global warming, and the resultant emissions as a fraction of anthropogenic emissions needed to cause that extent of warming.[86]
As recent warming deepens the active layer subject to permafrost thaw, this exposes formerly storedcarbon to biogenic processes which facilitate its entrance into the atmosphere ascarbon dioxide andmethane.[11] Because carbon emissions from permafrost thaw contribute to the same warming which facilitates the thaw, it is a well-known example of apositive climate change feedback.[87] Permafrost thaw is sometimes included as one of the majortipping points in the climate system due to the exhibition of local thresholds and its effective irreversibility.[88] However, while there are self-perpetuating processes that apply on the local or regional scale, it is debated as to whether it meets the strict definition of a global tipping point as in aggregate permafrost thaw is gradual with warming.[89]
Feedback processes related to land and subsea permafrost.
In the northern circumpolar region, permafrost contains organic matter equivalent to 1400–1650 billion tons of pure carbon, which was built up over thousands of years. This amount equals almost half of all organic material in allsoils,[90][11] and it is about twice the carbon content of theatmosphere, or around four times larger than the human emissions of carbon between the start of theIndustrial Revolution and 2011.[91] Further, most of this carbon (~1,035 billion tons) is stored in what is defined as the near-surface permafrost, no deeper than 3 metres (9.8 ft) below the surface.[90][11] However, only a fraction of this stored carbon is expected to enter the atmosphere.[92] In general, the volume of permafrost in the upper 3 m of ground is expected to decrease by about 25% per 1 °C (1.8 °F) of global warming,[83]: 1283 yet even under theRCP8.5 scenario associated with over 4 °C (7.2 °F) of global warming by the end of the 21st century,[93] about 5% to 15% of permafrost carbon is expected to be lost "over decades and centuries".[11]
The exact amount of carbon that will be released due to warming in a given permafrost area depends on depth of thaw, carbon content within the thawed soil, physical changes to the environment, and microbial and vegetation activity in the soil.[94] Notably, estimates of carbon release alone do not fully represent the impact of permafrost thaw on climate change. This is because carbon can be released through eitheraerobic oranaerobic respiration, which results in carbon dioxide (CO2) or methane (CH4) emissions, respectively. While methane lasts less than 12 years in the atmosphere, itsglobal warming potential is around 80 times larger than that of CO2 over a 20-year period and about 28 times larger over a 100-year period.[95][96] While only a small fraction of permafrost carbon will enter the atmosphere as methane, those emissions will cause 40–70% of the total warming caused by permafrost thaw during the 21st century. Much of the uncertainty about the eventual extent of permafrost methane emissions is caused by the difficulty of accounting for the recently discovered abrupt thaw processes, which often increase the fraction of methane emitted over carbon dioxide in comparison to the usual gradual thaw processes.[97][11]
Permafrost thaw ponds on peatland inHudson Bay, Canada in 2008.[98]
Another factor which complicates projections of permafrost carbon emissions is the ongoing "greening" of the Arctic. As climate change warms the air and the soil, the region becomes more hospitable to plants, including largershrubs and trees which could not survive there before. Thus, the Arctic is losing more and more of itstundra biomes, yet it gains more plants, which proceed to absorb more carbon. Some of the emissions caused by permafrost thaw will be offset by this increased plant growth, but the exact proportion is uncertain. It is considered very unlikely that this greening could offset all of the emissions from permafrost thaw during the 21st century, and even less likely that it could continue to keep pace with those emissions after the 21st century.[11] Further, climate change also increases the risk ofwildfires in the Arctic, which can substantially accelerate emissions of permafrost carbon.[87][99]
Nine probable scenarios ofgreenhouse gas emissions from permafrost thaw during the 21st century, which show a limited, moderate and intense CO2 andCH4 emission response to low, medium and high-emissionRepresentative Concentration Pathways. The vertical bar uses emissions of selected large countries as a comparison: the right-hand side of the scale shows their cumulative emissions since the start of theIndustrial Revolution, while the left-hand side shows each country's cumulative emissions for the rest of the 21st century if they remained unchanged from their 2019 levels.[11]
Altogether, it is expected that cumulative greenhouse gas emissions from permafrost thaw will be smaller than the cumulative anthropogenic emissions, yet still substantial on a global scale, with some experts comparing them to emissions caused bydeforestation.[11] TheIPCC Sixth Assessment Report estimates that carbon dioxide and methane released from permafrost could amount to the equivalent of 14–175 billion tonnes of carbon dioxide per 1 °C (1.8 °F) of warming.[83]: 1237 For comparison, by 2019, annual anthropogenic emissions of carbon dioxide alone stood around 40 billion tonnes.[83]: 1237 A major review published in the year 2022 concluded that if the goal of preventing 2 °C (3.6 °F) of warming was realized, then the average annual permafrost emissions throughout the 21st century would be equivalent to the year 2019 annual emissions of Russia. Under RCP4.5, a scenario considered close to the current trajectory and where the warming stays slightly below 3 °C (5.4 °F), annual permafrost emissions would be comparable to year 2019 emissions of Western Europe or the United States, while under the scenario of high global warming and worst-case permafrost feedback response, they would approach year 2019 emissions of China.[11]
Fewer studies have attempted to describe the impact directly in terms of warming. A 2018 paper estimated that if global warming was limited to 2 °C (3.6 °F), gradual permafrost thaw would add around 0.09 °C (0.16 °F) to global temperatures by 2100,[100] while a 2022 review concluded that every 1 °C (1.8 °F) of global warming would cause 0.04 °C (0.072 °F) and 0.11 °C (0.20 °F) from abrupt thaw by the year 2100 and 2300. Around 4 °C (7.2 °F) of global warming, abrupt (around 50 years) and widespread collapse of permafrost areas could occur, resulting in an additional warming of 0.2–0.4 °C (0.36–0.72 °F).[88][101]
As the water drains or evaporates, soil structure weakens and sometimes becomes viscous until it regains strength with decreasing moisture content. One visible sign of permafrost degradation is therandom displacement of trees from their vertical orientation in permafrost areas.[102] Global warming has been increasing permafrost slope disturbances and sediment supplies to fluvial systems, resulting in exceptional increases in river sediment.[103] On the other hands, disturbance of formerly hard soil increases drainage of water reservoirs in northernwetlands. This can dry them out and compromise the survival of plants and animals used to the wetland ecosystem.[104]
In high mountains, much of the structural stability can be attributed toglaciers and permafrost.[105] As climate warms, permafrost thaws, decreasing slope stability and increasing stress through buildup ofpore-water pressure, which may ultimately lead to slope failure androckfalls.[106][107] Over the past century, an increasing number of alpine rock slope failure events in mountain ranges around the world have been recorded, and some have been attributed to permafrost thaw induced by climate change. The 1987Val Pola landslide that killed 22 people in theItalian Alps is considered one such example.[108] In 2002, massive rock and ice falls (up to 11.8 million m3), earthquakes (up to 3.9Richter), floods (up to 7.8 million m3 water), and rapid rock-ice flow to long distances (up to 7.5 km at 60 m/s) were attributed to slope instability in high mountain permafrost.[109]
Permafrost thaw can also result in the formation of frozen debris lobes (FDLs), which are defined as "slow-moving landslides composed of soil, rocks, trees, and ice".[110] This is a notable issue in theAlaska's southernBrooks Range, where some FDLs measured over 100 m (110 yd) in width, 20 m (22 yd) in height, and 1,000 m (1,100 yd) in length by 2012.[111][112] As of December 2021, there were 43 frozen debris lobes identified in the southern Brooks Range, where they could potentially threaten both theTrans Alaska Pipeline System (TAPS) corridor and theDalton Highway, which is the main transport link between theInterior Alaska and theAlaska North Slope.[113]
Map of likely risk to infrastructure from permafrost thaw expected to occur by 2050.[114]
As of 2021, there are 1162 settlements located directly atop the Arctic permafrost, which host an estimated 5 million people. By 2050, permafrost layer below 42% of these settlements is expected to thaw, affecting all their inhabitants (currently 3.3 million people).[115] Consequently, a wide range of infrastructure in permafrost areas is threatened by the thaw.[12][116]: 236 By 2050, it's estimated that nearly 70% of global infrastructure located in the permafrost areas would be at high risk of permafrost thaw, including 30–50% of "critical" infrastructure. The associated costs could reach tens of billions of dollars by the second half of the century.[13] Reducinggreenhouse gas emissions in line with theParis Agreement is projected to stabilize the risk after mid-century; otherwise, it'll continue to worsen.[114]
InAlaska alone, damages to infrastructure by the end of the century would amount to $4.6 billion (at 2015 dollar value) ifRCP8.5, the high-emissionclimate change scenario, were realized. Over half stems from the damage to buildings ($2.8 billion), but there's also damage to roads ($700 million), railroads ($620 million), airports ($360 million) andpipelines ($170 million).[117] Similar estimates were done for RCP4.5, a less intense scenario which leads to around 2.5 °C (4.5 °F) by 2100, a level of warming similar to the current projections.[118] In that case, total damages from permafrost thaw are reduced to $3 billion, while damages to roads and railroads are lessened by approximately two-thirds (from $700 and $620 million to $190 and $220 million) and damages to pipelines are reduced more than ten-fold, from $170 million to $16 million. Unlike the other costs stemming from climate change in Alaska, such as damages from increasedprecipitation and flooding,climate change adaptation is not a viable way to reduce damages from permafrost thaw, as it would cost more than the damage incurred under either scenario.[117]
In Canada,Northwest Territories have a population of only 45,000 people in 33 communities, yet permafrost thaw is expected to cost them $1.3 billion over 75 years, or around $51 million a year. In 2006, the cost of adaptingInuvialuit homes to permafrost thaw was estimated at $208/m2 if they were built at pile foundations, and $1,000/m2 if they didn't. At the time, the average area of a residential building in the territory was around 100 m2. Thaw-induced damage is also unlikely to be covered byhome insurance, and to address this reality, territorial government currently funds Contributing Assistance for Repairs and Enhancements (CARE) and Securing Assistance for Emergencies (SAFE) programs, which provide long- and short-term forgivable loans to help homeowners adapt. It is possible that in the future, mandatory relocation would instead take place as the cheaper option. However, it would effectively tear the localInuit away from their ancestral homelands. Right now, their average personal income is only half that of the median NWT resident, meaning that adaptation costs are already disproportionate for them.[119]
By 2022, up to 80% of buildings in some Northern Russia cities had already experienced damage.[13] By 2050, the damage to residential infrastructure may reach $15 billion, while total public infrastructure damages could amount to 132 billion.[120] This includesoil and gas extraction facilities, of which 45% are believed to be at risk.[114]
Detailed map of Qinghai–Tibet Plateau infrastructure at risk from permafrost thaw under the SSP2-4.5 scenario.[121]
Outside of the Arctic,Qinghai–Tibet Plateau (sometimes known as "the Third Pole"), also has an extensive permafrost area. It is warming at twice the global average rate, and 40% of it is already considered "warm" permafrost, making it particularly unstable. Qinghai–Tibet Plateau has a population of over 10 million people – double the population of permafrost regions in the Arctic – and over 1 million m2 of buildings are located in its permafrost area, as well as 2,631 km ofpower lines, and 580 km of railways.[121] There are also 9,389 km of roads, and around 30% are already sustaining damage from permafrost thaw.[13] Estimates suggest that under the scenario most similar to today,SSP2-4.5, around 60% of the current infrastructure would be at high risk by 2090 and simply maintaining it would cost $6.31 billion, with adaptation reducing these costs by 20.9% at most. Holding the global warming to 2 °C (3.6 °F) would reduce these costs to $5.65 billion, and fulfilling the optimisticParis Agreement target of 1.5 °C (2.7 °F) would save a further $1.32 billion. In particular, fewer than 20% of railways would be at high risk by 2100 under 1.5 °C (2.7 °F), yet this increases to 60% at 2 °C (3.6 °F), while under SSP5-8.5, this level of risk is met by mid-century.[121]
Graphical representation of leaks from various toxic hazards caused by the thaw of formerly stable permafrost.[14]
For much of the 20th century, it was believed that permafrost would "indefinitely" preserve anything buried there, and this made deep permafrost areas popular locations for hazardous waste disposal. In places like Canada'sPrudhoe Bay oil field, procedures were developed documenting the "appropriate" way to inject waste beneath the permafrost. This means that as of 2023, there are ~4500 industrial facilities in the Arctic permafrost areas which either actively process or store hazardous chemicals. Additionally, there are between 13,000 and 20,000 sites which have been heavily contaminated, 70% of them in Russia, and their pollution is currently trapped in the permafrost.[citation needed]
About a fifth of both the industrial and the polluted sites (1000 and 2200–4800) are expected to start thawing in the future even if the warming does not increase from its 2020 levels. Only about 3% more sites would start thawing between now and 2050 under the climate change scenario consistent with theParis Agreement goals,RCP2.6, but by 2100, about 1100 more industrial facilities and 3500 to 5200 contaminated sites are expected to start thawing even then. Under the very high emission scenario RCP8.5, 46% of industrial and contaminated sites would start thawing by 2050, and virtually all of them would be affected by the thaw by 2100.[14]
Organochlorines and otherpersistent organic pollutants are of a particular concern, due to their potential to repeatedly reach local communities after their re-release throughbiomagnification in fish. At worst, future generations born in the Arctic would enter life with weakenedimmune systems due to pollutants accumulating across generations.[16]
Distribution of toxic substances currently located at various permafrost sites in Alaska, by sector. The number of fish skeletons represents the toxicity of each substance.[14]
A notable example of pollution risks associated with permafrost was the2020 Norilsk oil spill, caused by the collapse ofdiesel fuel storage tank at Norilsk-Taimyr Energy'sthermal power plant No. 3. It spilled 6,000 tonnes of fuel into the land and 15,000 into the water, pollutingAmbarnaya,Daldykan and many smaller rivers onTaimyr Peninsula, even reaching lakePyasino, which is a crucial water source in the area.State of emergency at the federal level was declared.[122][123] The event has been described as the second-largest oil spill in modern Russian history.[124][125]
Another issue associated with permafrost thaw is the release of naturalmercury deposits. An estimated 800,000 tons of mercury are frozen in the permafrost soil. According to observations, around 70% of it is simply taken up by vegetation after the thaw.[16] However, if the warming continues under RCP8.5, then permafrost emissions of mercury into theatmosphere would match the current global emissions from all human activities by 2200. Mercury-rich soils also pose a much greater threat to humans and the environment if they thaw near rivers. Under RCP8.5, enough mercury will enter theYukon River basin by 2050 to make its fish unsafe to eat under theEPA guidelines. By 2100, mercury concentrations in the river will double. Contrastingly, even if mitigation is limited to RCP4.5 scenario, mercury levels will increase by about 14% by 2100, and will not breach the EPA guidelines even by 2300.[15]
Some of the ancient amoeba-eating viruses revived by the research team of Jean-Michel Claverie. Clockwise from the top:Pandoravirus yedoma;Pandoravirus mammoth andMegavirus mammoth;Cedratvirus lena;Pithovirus mammoth;Megavirus mammoth;Pacmanvirus lupus.[17]
Bacteria are known for being able toremain dormant to survive adverse conditions, andviruses are not metabolically active outside of host cells in the first place. This has motivated concerns that permafrost thaw could free previously unknown microorganisms, which may be capable of infecting either humans or important livestock andcrops, potentially resulting in damaging epidemics orpandemics.[17][18] Further, some scientists argue thathorizontal gene transfer could occur between the older, formerly frozen bacteria, and modern ones, and one outcome could be the introduction of novelantibiotic resistance genes into thegenome of current pathogens, exacerbating what is already expected to become a difficult issue in the future.[126][16]
At the same time, notable pathogens likeinfluenza andsmallpox appear unable to survive being thawed,[20] and other scientists argue that the risk of ancient microorganisms being both able to survive the thaw and to threaten humans is not scientifically plausible.[19] Likewise, some research suggests that antimicrobial resistance capabilities of ancient bacteria would be comparable to, or even inferior to modern ones.[127][21]
In 2012, Russian researchers proved that permafrost could serve as a natural repository for ancient life forms by reviving a sample ofSilene stenophylla from 30,000-year-old tissue found in anIce Age squirrel burrow in theSiberian permafrost. This is the oldest plant tissue ever revived. The resultant plant was fertile, producing white flowers and viable seeds. The study demonstrated that living tissue can survive ice preservation for tens of thousands of years.[128]
Between the middle of the 19th century and the middle of the 20th century, most of the literature on basic permafrost science and the engineering aspects of permafrost was written in Russian. One of the earliest written reports describing the existence of permafrost dates to1684, whenwell excavation efforts inYakutsk were stumped by its presence.[76]: 25 A significant role in the initial permafrost research was played byAlexander von Middendorff (1815–1894) andKarl Ernst von Baer, aBaltic German scientist at theUniversity of Königsberg, and a member of theSt Petersburg Academy of Sciences. Baer began publishing works on permafrost in 1838 and is often considered the "founder of scientific permafrost research." Baer laid the foundation for modern permafrost terminology by compiling and analyzing all available data on ground ice and permafrost.[129]
Baer is also known to have composed the world's first permafrosttextbook in 1843,Materialien zur Kenntniss des unvergänglichen Boden-Eises in Sibirien (Materials for the study of the perennial ground-ice in Siberia), written in his native German. However, it was not printed then, and a Russian translation was not ready until 1942. The original German textbook was believed to be lost until thetypescript from 1843 was discovered in the library archives of theUniversity of Giessen. The 234-page text was available online, with additional maps,preface and comments.[129] Notably, Baer's southern limit of permafrost inEurasia drawn in 1843 corresponds well with the actual southern limit verified by modern research.[27][129]
Beginning in 1942,Siemon William Muller delved into the relevant Russian literature held by theLibrary of Congress and theU.S. Geological Survey Library so that he was able to furnish the government an engineeringfield guide and a technical report about permafrost by 1943.[130] That report coined the English term as a contraction of permanently frozen ground,[131] in what was considered a direct translation of the Russian termvechnaia merzlota (Russian:вечная мерзлота). In 1953, this translation was criticized by another USGS researcher Inna Poiré, as she believed the term had created unrealistic expectations about its stability:[76]: 3 more recently, some researchers have argued that "perpetually refreezing" would be a more suitable translation.[132] The report itself was classified (as U.S. Army. Office of the Chief of Engineers,Strategic Engineering Study, no. 62, 1943),[131][133] until a revised version was released in 1947, which is regarded as the first North American treatise on the subject.[130][134]
The annual number of scientific research papers published on the subject of permafrost carbon has grown from next to nothing around 1990 to around 400 by 2020.[11]
Between 11 and 15 November 1963, the First International Conference on Permafrost took place on the grounds ofPurdue University in the American town ofWest Lafayette, Indiana. It involved 285 participants (including "engineers, manufacturers and builders" who attended alongside the researchers) from a range of countries (Argentina,Austria, Canada, Germany, Great Britain, Japan,Norway,Poland, Sweden, Switzerland, the US and theUSSR). This marked the beginning of modern scientific collaboration on the subject. Conferences continue to take place every five years. During the Fourth conference in 1983, a special meeting between the "Big Four" participant countries (US, USSR, China, and Canada) officially created theInternational Permafrost Association.[135]
In recent decades, permafrost research has attracted more attention than ever due to its role inclimate change. Consequently, there has been a massive acceleration in publishedscientific literature. Around 1990, almost no papers containing the words "permafrost" and "carbon" were released: by 2020, around 400 such papers were published yearly.[11]
^Cooper, M. G.; Zhou, T.; Bennett, K. E.; Bolton, W. R.; Coon, E. T.; Fleming, S. W.; Rowland, J. C.; Schwenk, J. (4 January 2023). "Detecting Permafrost Active Layer Thickness Change From Nonlinear Baseflow Recession".Water Resources Research.57 (1): e2022WR033154.Bibcode:2023WRR....5933154C.doi:10.1029/2022WR033154.S2CID255639677.
^abcdefghijklSchuur, Edward A. G.; Abbott, Benjamin W.; Commane, Roisin; Ernakovich, Jessica; Euskirchen, Eugenie; Hugelius, Gustaf; Grosse, Guido; Jones, Miriam; Koven, Charlie; Leshyk, Victor; Lawrence, David; Loranty, Michael M.; Mauritz, Marguerite; Olefeldt, David; Natali, Susan; Rodenhizer, Heidi; Salmon, Verity; Schädel, Christina; Strauss, Jens; Treat, Claire; Turetsky, Merritt (2022). "Permafrost and Climate Change: Carbon Cycle Feedbacks From the Warming Arctic".Annual Review of Environment and Resources.47:343–371.doi:10.1146/annurev-environ-012220-011847.S2CID252986002.
^abNelson, F. E.; Anisimov, O. A.; Shiklomanov, N. I. (1 July 2002). "Climate Change and Hazard Zonation in the Circum-Arctic Permafrost Regions".Natural Hazards.26 (3):203–225.doi:10.1023/A:1015612918401.S2CID35672358.
^Zhang, Caiyun; Douglas, Thomas A.; Anderson, John E. (27 July 2021). "Modeling and mapping permafrost active layer thickness using field measurements and remote sensing techniques".International Journal of Applied Earth Observation and Geoinformation.102.Bibcode:2021IJAEO.10202455Z.doi:10.1016/j.jag.2021.102455.
^abLi, Chuanhua; Wei, Yufei; Liu, Yunfan; Li, Liangliang; Peng, Lixiao; Chen, Jiahao; Liu, Lihui; Dou, Tianbao; Wu, Xiaodong (14 June 2022). "Active Layer Thickness in the Northern Hemisphere: Changes From 2000 to 2018 and Future Simulations".Journal of Geophysical Research: Atmospheres.127 (12): e2022JD036785.Bibcode:2022JGRD..12736785L.doi:10.1029/2022JD036785.S2CID249696017.
^Luo, Dongliang; Wu, Qingbai; Jin, Huijun; Marchenko, Sergey S.; Lü, Lanzhi; Gao, Siru (26 March 2016). "Recent changes in the active layer thickness across the northern hemisphere".Environmental Earth Sciences.75 (7): 555.Bibcode:2016EES....75..555L.doi:10.1007/s12665-015-5229-2.S2CID130353989.
^abcLacelle, Denis; Fisher, David A.; Verret, Marjolaine; Pollard, Wayne (17 February 2022). "Improved prediction of the vertical distribution of ground ice in Arctic-Antarctic permafrost sediments".Communications Earth & Environment.3 (31): 31.Bibcode:2022ComEE...3...31L.doi:10.1038/s43247-022-00367-z.S2CID246872753.
^abBrown, J.; Ferrians Jr., O. J.; Heginbottom, J. A.; Melnikov, E. S. (1997). Circum-Arctic map of permafrost and ground-ice conditions (Report).USGS.doi:10.3133/cp45.
^Robinson, S. D.; et al. (2003). "Permafrost and peatlandcarbon sink capacity with increasing latitude". In Phillips; et al. (eds.).Permafrost(PDF) (Report). Swets & Zeitlinger. pp. 965–970.ISBN90-5809-582-7.Archived(PDF) from the original on 2 March 2014. Retrieved18 August 2023.
^Andersland, Orlando B.; Ladanyi, Branko (2004).Frozen ground engineering (2nd ed.). Wiley. p. 5.ISBN978-0-471-61549-1.
^Zoltikov, I. A. (1962). "Heat regime of the central Antarctic glacier".Antarctica, Reports of the Commission, 1961 (in Russian):27–40.
^Campbell, Iain B.; Claridge, Graeme G. C. (2009). "Antarctic Permafrost Soils". In Margesin, Rosa (ed.).Permafrost Soils. Soil Biology. Vol. 16. Berlin: Springer. pp. 17–31.doi:10.1007/978-3-540-69371-0_2.ISBN978-3-540-69370-3.
^abHaeberli, Wilfried; Noetzli, Jeannette; Arenson, Lukas; Delaloye, Reynald; Gärtner-Roer, Isabelle; Gruber, Stephan; Isaksen, Ketil; Kneisel, Christof; Krautblatter, Michael; Phillips, Marcia (2010). "Mountain permafrost: development and challenges of a young research field".Journal of Glaciology.56 (200). Cambridge University Press:1043–1058.Bibcode:2010JGlac..56.1043H.doi:10.3189/002214311796406121.S2CID33659636.
^Fox-Kemper, B.;Hewitt, H. T.; Xiao, C.; Aðalgeirsdóttir, G.; Drijfhout, S. S.; Edwards, T. L.; Golledge, N. R.; Hemer, M.; Kopp, R. E.; Krinner, G.; Mix, A. (2021). Masson-Delmotte, V.; Zhai, P.; Pirani, A.; Connors, S. L.; Péan, C.; Berger, S.; Caud, N.; Chen, Y.; Goldfarb, L. (eds.)."Chapter 5: Global Carbon and other Biogeochemical Cycles and Feedbacks"(PDF).Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK and New York, NY, USA: 5.doi:10.1017/9781009157896.011.It is very unlikely that gas clathrates (mostly methane) in deeper terrestrial permafrost and subsea clathrates will lead to a detectable departure from the emissions trajectory during this century.
^Ono, Yugo; Irino, Tomohisa (16 September 2003). "Southern migration of westerlies in the Northern Hemisphere PEP II transect during the Last Glacial Maximum".Quaternary International.118–119:13–22.doi:10.1016/S1040-6182(03)00128-9.
^abLunardini, Virgil J. (April 1995). Permafrost Formation Time.CRREL Report 95-8 (Report). Hanover NH: US Army Corps of Engineers Cold Regions Research and Engineering Laboratory.DTICADA295515.
^abOsterkamp, T. E.; Burn, C. R. (2003). "Permafrost". In North, Gerald R.; Pyle, John A.; Zhang, Fuqing (eds.).Encyclopedia of Atmospheric Sciences(PDF). Vol. 4. Elsevier. pp. 1717–1729.ISBN978-0-12-382226-0.Archived(PDF) from the original on 30 November 2016. Retrieved8 March 2016.
^Fridleifsson, Ingvar B.; Bertani, Ruggero; Huenges, Ernst; Lund, John W.; Ragnarsson, Arni; Rybach, Ladislaus (11 February 2008). O. Hohmeyer and T. Trittin (ed.).The possible role and contribution of geothermal energy to the mitigation of climate change(PDF) (Report). IPCC Scoping Meeting on Renewable Energy Sources, Luebeck, Germany. pp. 59–80. Archived fromthe original on 12 March 2013. Retrieved27 September 2023.
^abShumskiy, P. A.; Vtyurin, B. I. (1963).Underground ice. Permafrost International Conference. pp. 108–113.
^abMackay, J. R.; Dallimore, S. R. (1992). "Massive ice of Tuktoyaktuk area, Western Arctic coast, Canada".Canadian Journal of Earth Sciences.29 (6):1234–1242.Bibcode:1992CaJES..29.1235M.doi:10.1139/e92-099.
^Murton, J. B.; Whiteman, C. A.; Waller, R. I.; Pollard, W. H.; Clark, I. D.; Dallimore, S. R. (12 August 2004). "Basal ice facies and supraglacial melt-out till of the Laurentide Ice Sheet, Tuktoyaktuk Coastlands, western Arctic Canada".Quaternary Science Reviews.24 (5–6):681–708.doi:10.1016/S0277-3791(01)00149-4.
^French, H. M.; Harry, D. G. (1990). "Observations on buried glacier ice and massive segregated ice, western arctic coast, Canada".Permafrost and Periglacial Processes.1 (1):31–43.Bibcode:1990PPPr....1...31F.doi:10.1002/ppp.3430010105.
^Li, Dongfeng; Overeem, Irina; Kettner, Albert J.; Zhou, Yinjun; Lu, Xixi (February 2021). "Air Temperature Regulates Erodible Landscape, Water, and Sediment Fluxes in the Permafrost-Dominated Catchment on the Tibetan Plateau".Water Resources Research.57 (2): e2020WR028193.Bibcode:2021WRR....5728193L.doi:10.1029/2020WR028193.S2CID234044271.
^Zhang, Ting; Li, Dongfeng; Kettner, Albert J.; Zhou, Yinjun; Lu, Xixi (October 2021). "Constraining Dynamic Sediment-Discharge Relationships in Cold Environments: The Sediment-Availability-Transport (SAT) Model".Water Resources Research.57 (10): e2021WR030690.Bibcode:2021WRR....5730690Z.doi:10.1029/2021WR030690.S2CID242360211.
^Kudryashova, E. B.; Chernousova, E. Yu.; Suzina, N. E.; Ariskina, E. V.; Gilichinsky, D. A. (1 May 2013). "Microbial diversity of Late Pleistocene Siberian permafrost samples".Microbiology.82 (3):341–351.doi:10.1134/S0026261713020082.S2CID2645648.
^Allen, Robert J.; Zhao, Xueying; Randles, Cynthia A.; Kramer, Ryan J.; Samset, Bjørn H.; Smith, Christopher J. (16 March 2023). "Surface warming and wetting due to methane's long-wave radiative effects muted by short-wave absorption".Nature Geoscience.16 (4):314–320.Bibcode:2023NatGe..16..314A.doi:10.1038/s41561-023-01144-z.S2CID257595431.
^Li, Dongfeng; Lu, Xixi; Overeem, Irina; Walling, Desmond E.; Syvitski, Jaia; Kettner, Albert J.; Bookhagen, Bodo; Zhou, Yinjun; Zhang, Ting (29 October 2021). "Exceptional increases in fluvial sediment fluxes in a warmer and wetter High Mountain Asia".Science.374 (6567):599–603.Bibcode:2021Sci...374..599L.doi:10.1126/science.abi9649.PMID34709922.S2CID240152765.
^Nater, P.; Arenson, L.U.; Springman, S.M. (2008).Choosing geotechnical parameters for slope stability assessments in alpine permafrost soils. In 9th international conference on permafrost. Fairbanks, USA: University of Alaska. pp. 1261–1266.ISBN978-0-9800179-3-9.
^Temme, Arnaud J. A. M. (2015). "Using Climber's Guidebooks to Assess Rock Fall Patterns Over Large Spatial and Decadal Temporal Scales: An Example from the Swiss Alps".Geografiska Annaler: Series A, Physical Geography.97 (4):793–807.Bibcode:2015GeAnA..97..793T.doi:10.1111/geoa.12116.S2CID55361904.
^F., Dramis; M., Govi; M., Guglielmin; G., Mortara (1 January 1995). "Mountain permafrost and slope instability in the Italian Alps: The Val Pola Landslide".Permafrost and Periglacial Processes.6 (1):73–81.Bibcode:1995PPPr....6...73D.doi:10.1002/ppp.3430060108.
^Catastrophic Landslides: Effects, Occurrence, and Mechanisms. Reviews in Engineering Geology. Vol. 15. 2002.doi:10.1130/REG15.ISBN0-8137-4115-7.
^Ramage, Justine; Jungsberg, Leneisja; Wang, Shinan; Westermann, Sebastian; Lantuit, Hugues; Heleniak, Timothy (6 January 2021). "Population living on permafrost in the Arctic".Population and Environment.43:22–38.doi:10.1007/s11111-020-00370-6.S2CID254938760.
^Barry, Roger Graham; Gan, Thian-Yew (2021).The global cryosphere past, present and future (Second revised ed.). Cambridge, United Kingdom: Cambridge University Press.ISBN978-1-108-48755-9.OCLC1256406954.