Soil liquefaction

Soil liquefaction is the group of processes by which otherwise solid soil particles are shaken apart by earthquakes or collapse away from edge-to-edge contact with one another and become temporarily supported by the pore water contained within them. The resulting fluid mush can allow buildings to sink; this phenomenon is responsible for considerable loss of property and life.

Types of Liquefaction Failure

Soil liquefaction is the abrupt and temporary change of seemingly solid soil materials into liquid mush. The soil can lose all of its cohesion and bearing capacity in a matter of moments. Once the process begins, most commonly when an earthquake strikes, ground that had supported a hospital or a high-rise apartment can suddenly become a fluid into which the buildings sink like rocks into quicksand. Anything built upon such materials can slip or sink into the new liquid. Buried gasoline or septic tanks can suddenly become buoyant and float to the surface.

A number of different sorts of liquefaction failure are recognized; several occur on land and a few are subaqueous. Quick condition failure (not to be confused with sensitive or quick-clay landslips) is the complete loss of bearing capacity caused by liquefaction of sands and silts so that structures sink or rise in material that appears otherwise solid. Flow landslips, or flowslides, can occur on moderate slopes on land or beneath water and involve either sands or clays. These types tend to retrogress, or work their way backward, slice by slice as the material beneath them liquefies. Lateral spread landslips can occur on gentle, or nearly flat, slopes. Many lateral spreads are in quick clays. In extreme cases, the liquefaction slides result from spontaneous subaqueous liquefaction that propagates, or spreads, in all directions.

Water-Saturated Sand Liquefaction

Two classes of soil materials can liquefy: water-saturated sands and silts and the unusual sensitive, or quick, clays. In the first example, layers of loosely packed, well-sorted, fine- to medium-grained sands and coarse silts are subject to liquefaction where groundwater tables are within 10-15 meters of the ground surface. When this water-saturated sediment starts to shake apart in an earthquake or other vibration, the grains temporarily lose rigid contact with one another and collapse inward. Much of the pore water (water in the gaps between soil grains) is then superfluous but does not escape at once, so that the rearranged grains cannot fit close to one another. The particle weight is thereby transferred to the pore water, its pressure increases (which reduces friction between the grains), and the soil becomes liquid for a short time.

When water-saturated sands liquefy during the course of an earthquake, sand boils erupt muddy water and sand from ground fractures and turn the surface into a quagmire. Sand boils tend to be roughly circular in plan and can have a depressed center like a volcano. After a very large earthquake, sand boils may form as much as several hundred meters across and several meters high. Their presence in an area is a reliable sign of past earthquake activity.

Water-saturated earth-fill dams are also subject to soil liquefaction, with obvious disastrous consequences. For example, during the 1971 San Fernando Valley earthquake in California, the Lower Van Norman dam collapsed. The 40-meter-high dam had been built with a core of clay surrounded by a fill of water-saturated sand. After about twelve seconds of strong shaking, a large, wedge-shaped segment of this water-saturated sand fill liquefied. Eight large blocks on the upstream side of the dam slid into the reservoir as parts of the sediment fill liquefied. Before the quake, the crest of the dam was a safe 10 meters above the water level of the reservoir. Afterward, however, only a thin barrier of 1.5 meters was left above the water level. Had this minor amount of remaining dam surface moved down but a fraction, the dam would have quickly failed, because water would have rushed down and eroded the downstream side. The 80,000 people who lived directly below the reservoir thus escaped disaster by a slim margin.

Quick-Clay Liquefaction

In the quick-clay type of soil liquefaction, because of a rather special geologic history, certain clays develop the ability to liquefy, or to become quick, as with quicksand. Two main theories have been advanced to account for the unusual distribution of these quick clays. The saltwater theory holds that sensitive clays are first formed where glaciers erode very fine-grained clay mineral platelets from the soils and bedrock over which they ride. Where the weight of such glaciers depresses the surface of the land close to the sea, the small platelets can be deposited directly into the salty waters.

The small clay mineral platelets of a sensitive clay carry a negative charge, and as they settle in ocean water they tend to pick up positively charged sodium ions from the salt in solution. These oppositely charged, mutually attractive forces act as a glue to cause the particles to clump together into a sort of coagulated honeycomb structure. The platelets thus develop an edge-to-edge, or “house of cards,” structure held together by the electrostatic forces of the salt ions attached to the platelet edges.

Glacial marine clays may remain quite stable as long as the salt water remains in the pore spaces of the open card-house structure. These deposits can later be raised above sea level by rebound of the land following removal of the glacier ice load. Then the salty pore waters may be flushed out of the clays by the fresh waters that normally occur above sea level. Removal of the mutually attractive electrochemical charges then sets the stage for eventual liquefaction.

A different theory explains the freshwater type of quick clays. In this case, glacial erosion also provides extremely fine particles of other than clay minerals. Instead, the fine particles (mostly quartz minerals) are so small that the normally weak van der Waals attraction is sufficient to hold them together. When the particles are small enough, the ratio of these weak attractions to the weight of the particles is greater, and the material will be cohesive, strong but brittle, and sensitive. When the cohesive bonds are broken through shaking, the short-range forces are ineffective, and a total loss of strength results. If there is sufficient pore water, the material will liquefy and flow on gentle slopes.

An example of a well-documented failure in quick clay is the St. Jean-Vianney failure of May 4, 1971, in Quebec, Canada. This landslip was located in the middle of a much larger prior failure that was at least five hundred years old. At about 7:00 p.m., instability first began with no warning on the steep west bank of the small Petit Bras tributary to the Saguenay River. The first actual failure, however, did not occur until 10:15 p.m., when, in fifteen minutes, the failure surface retrogressed 150 meters west into the slope. The resulting debris moved into the river and blocked both the river valley and the opening to the failure. The dam blocked the outflow of the liquefied material from the crater until the pressure became too high. Finally, the dam burst and allowed the remolded fluid to flow down the river valley with a wavefront about 18 meters high and traveling about 26 kilometers per hour. The mass carried with it thirty-four houses, one bus, and an undetermined number of cars. Thirty-one lives were lost as well.

Soil Strength

The strength, or load-carrying capacity, of all soils varies considerably. In addition, the strength of any specific soil can vary under different conditions of moisture and density. Sensitive soils have natural water contents already above the liquid limit, which is the moisture content at which a soil passes from a plastic to a liquid state. Plasticity is a characteristic of clayey soils that allows them to be squeezed and easily deformed without disintegration, whereas being above the liquid limit, as the name implies, allows the soil to flow easily and remold itself.

Sensitive clays are generally considered to be thixothropic in the sense that strength diminishes upon disturbance and is regained when disturbance ceases. Once collapse of the card-house matrix is initiated and flow begins, the clays seem to be remolded by lining up of the platelets and particles in a parallel or linear fashion. Remolded clays slip easily over one another in the watery mixture. Once these instabilities develop, they can spread rapidly throughout an area of retrogressive failure. The regaining of strength is not well understood but is thought to result from the gradual rearrangement of particles into positions of increasing mechanical stability under the action of new bonding forces. Perhaps the compaction and gradual expelling of excess water from the moving mass also are responsible for some restoration of strength. The artificial addition of new salts to such materials can be used in certain cases to regain strength and thereby stabilize hazardous areas.

Remolding

The typical lateral spread and flow landslips in sensitive clays have certain characteristics that make them distinctive. The processes are essentially a gravitational remolding that transforms the clay into a viscous slurry, or mixture of liquid and solid. Overlying sediments can break into strips or blocks that then become separated. The cracks between the blocks fill with either soft material squeezed up from below or detritus (loose material resulting from disintegration) from above.

Such failures commonly start in the lower part of a slope as a result of local oversteepening through stream or other erosion. After the initial slip, the failure spreads retrogressively, slice by slice, farther and farther into the bank. Movement generally begins quickly, without appreciable warning, and proceeds with rapid to very rapid velocity. A large, bowl-shaped crater commonly results.

Soil Liquefaction Study

Analysis of soil liquefaction is not extensive, because the phenomenon has been known as an important hazard for a relatively short period of time and is an unusual event in any case. The rarity of the phenomenon means that scientists have had trouble installing instruments and making measurements of the ground in places that would eventually liquefy and provide a well-documented record. In most cases, studies had to be done on ground that had liquefied in the past but had since become stable or in the rather artificial situations of the laboratory.

After soil liquefaction was first discovered following catastrophic ground failure, interviews with survivors provided most of the early information. Borehole cores and measurements of displaced ground showed types of sediments and characteristic depths and amounts of water involved. In many engineering applications, the undisturbed cores are subjected to various stresses to determine their behavior when shaken or loaded. Stirring of undisturbed quick clays to a liquid state in the lab, followed by resolidification upon addition of salt, is a means of analyzing the condition. Such studies allow the hazardous condition of certain areas to be represented on maps of liquefaction susceptibility that indicate places in which the phenomenon is likely to occur.

Success in observing soil liquefaction in action has occurred where instruments were previously installed in the ground to measure pore-water pressures at various depths during carefully measured earthquakes. Surprisingly, and in contrast to laboratory experiments in which liquefaction occurs at the same time as does strong shaking, pore-water pressures rose only slowly as the shaking intensified, and sandy layers completely liquefied only well after strong earthquake motions had ceased. The delay appears to occur as uneven pore-water pressures are redistributed in the ground.

Ever-Present Earthquake Hazard

The significance of soil-liquefaction potential is enormous. Many of the world's major cities are partly built upon weakly consolidated, water-saturated sediments. Understanding the mechanisms of liquefaction is also important in analyzing earthquakes long past; old sand boils have been used to date and to estimate magnitudes of prehistoric earthquakes. Major earthquakes in Niigata, Japan, and Anchorage, Alaska, brought about a heightened appreciation of the importance of soil liquefaction as a general geologic process. Following these events, geologists, engineers, and planners have come to recognize soil liquefaction as an ever-present earthquake hazard and have associated it with almost all major earthquakes since.

In June 1964, for example, a magnitude 7.3 earthquake occurred 55 kilometers from the city of Niigata on Japan's west coast. In 50 seconds of shaking, the city of 300,000 was subjected to dramatic soil liquefaction that affected thousands of dwellings and industrial structures. Much of the city was built originally upon sand deposits about 100 meters thick along the Shinano River and upon younger lowland sediments and reclaimed riverfront land. During the earthquake, subsurface sand and water flowed up and out of cracks in the ground. The liquefaction caused major destruction of highways, bridges, railroads, utilities, oil refineries, and harbor facilities. There were 3,018 houses destroyed outright and 9,750 damaged moderately or severely because of cracking and unequal settlement of the ground; much of this damage occurred on the newly reclaimed areas. In a most spectacular occurrence, a number of large apartment buildings, which had been designed to be earthquake-resistant, tipped over on their sides to settle at angles of as much as 80 degrees, though the structures themselves remained intact. People were able to escape by walking down the sides of the buildings. Several of these apartment houses were later jacked up, reinforced, and opened for reoccupation.

The Good Friday (March 27) earthquake of 1964 in Alaska was, at 8.4-8.6 magnitude, the largest ever recorded in North America. The Turnagain Heights landslide in Anchorage, Alaska, took place in flat terrain along the steep coastal bluffs that border the Knick Arm of the Cook Inlet there. Before the earthquake, the bluffs rose steeply some 30-35 meters above sea level. Marine clays and silts with layers and lenses of sand of the Bootlegger Cove Clay were exposed in the bluffs. A 6-meter thickness of sand and gravel on the flat terrace above had provided an apparently fine place to build homes. During the 1964 earthquake, giant blocks of Bootlegger Cove Clay and the overlying sands and gravels were set in motion toward the sea as the sand layers in the clay formation were liquefied. The first movements of the blocks began about two minutes after the onset of intense earthquake shaking. In the next five minutes, the previously flat terrain was transformed into a jumble of blocks capped by tilted trees and broken buildings. Seventy-five homes and three lives were lost in the breakup of the ground. More than seven huge blocks became widely separated as they moved toward the sea.

One positive result of soil liquefaction is the tendency for earthquake shaking to be significantly reduced, as liquids do not support shear stress. Once the soil liquefies due to shaking, shear waves are not transferred to buildings at the ground surface.

Incidences of Subaqueous Failure

In a subaqueous example of failure, the sand beds along the coast of Zeeland in the Netherlands periodically liquefy. The coast is located on a thick layer of fine quartz sand that consists of rounded quartz grains. The slope of the beach is only about 15 degrees. Once every few decades, especially after exceptionally high spring tides, the structure of the sand breaks down beneath a short section of the coastal belt. The sand flows out and spreads with great speed in a fan-shaped sheet over the bottom of the adjacent body of water. The tongue of such a flowslide is always much broader than is the source; the flowslides themselves commonly have surface slope angles of as small as 3-4 degrees. Such a failure occurred at Borssele in 1874 and involved nearly 2 million cubic meters of sand.

A liquefaction process also seems to be significant in the fine silts and clays of the Mississippi Delta region, particularly in the formation of collapse depressions and elongate flows. In these sediments, the pore-water pressures are extremely large, and the pore spaces also contain large amounts of methane gas from decaying organic matter. Following initial failure of these materials, softening of the highly pressured clay/water/gas system causes remolding and strength loss similar to a type of liquefaction or quick behavior. Collapse of offshore structures and sinking of pipelines and seafloor monitoring equipment vertically into the sediment may result. Increased pore pressures during storms also cause bottom movement, collapse, and indications that the sediment can become active as a fluid. On very low-angle slopes (0.1-0.2 degree), distinct collapse depressions are formed, whereas on slightly steeper slopes (0.3-0.4 degree), more elongate flows can result.

Principal Terms

bearing capacity: the ability of granular soil materials to support the weight of building structures

clay minerals: the diverse group of very finely crystalline mineral structures that are predominately composed of silicon, aluminum, and oxygen and that have very different water retention capabilities

cohesion: molecular attraction by which the particles of a body are united throughout the mass, whether alike or unlike

electrostatic charge: the fundamental atomic force in which objects that have a similar electric charge repel each other whereas those with unlike charges attract each other

flocculation: the sedimentation process by which a number of individual minute suspended particles are held together in clotlike masses by electrostatic forces

remolding: the property of some sensitive clays upon disturbance to reorient their particles, which softens them, and to flow in a liquid form

sensitive (quick): describes fine-grained deposits that are characterized by considerable strength in the undisturbed condition, but upon disturbance their ability to support themselves declines dramatically

van der Waals force: a weak electrostatic attraction that arises because certain atoms and molecules are distorted from a spherical shape so that one side of the structure carries more of the charge than does the other

Bibliography

Das, Braja M., and G. V. Ramana. Principles of Soil Dynamics. Stamford: Cengage Learning, 2011. Provides background information on physics of vibrations, elastic waves, stress waves, and earthquakes. Chapter 10 covers aspects of soil liquefaction procedures and tests.

Dennen, W. H., and B. R. Moore. Geology and Engineering. Dubuque, Iowa: Wm. C. Brown, 1986. This volume is one of the most nontechnical but accurate books available on general geotechnical subjects. The sections on soil liquefaction and related phenomena are not long but are easy to understand.

Doyle, Hugh A. Seismology. New York: John Wiley, 1995. A good introduction to the study of earthquakes and the earth's lithosphere. Written for the layperson, the book contains many useful illustrations.

Holzer, T. L., T. L. Youd, and T. C. Hanks. “Dynamics of Liquefaction During the Superstition Hills, California, Earthquake.” Science 244 (April 7, 1989): 56-59. This article reports the details of the first record of a natural liquefaction event after an earthquake. An array of instruments had been installed up to 12 meters deep that recorded excess pore pressures generated once horizontal ground acceleration from the earthquake exceeded a certain threshold value.

Idriss, I. M., and R. W. Boulanger. Soil Liquefaction During Earthquakes. Earthquake Engineering Research Institute, 2008. This text covers the basic principles of liquefaction behavior of soils and analysis methods.

Lade, Poul V., and Jerry A. Yamamuro. “Evaluation of Static Liquefaction Potential of Silty Sand Slopes.” Canadian Geotechnical Journal 48 (2011): 247-264. The article compares sand types and their liquefaction potential.

Lundgren, Lawrence. Environmental Geology. 2d ed. Englewood Cliffs, N.J.: Prentice-Hall, 1998. A general book, with several excellent sections and photographs of soil liquefaction phenomena. Not too difficult or technical for the average interested reader.

Penick, J. L., Jr. The New Madrid Earthquakes. Rev. ed. Columbia: University of Missouri Press, 1981. This well-written account covers the most intense earthquakes ever to strike the North American continent and details their effects upon people, animals, waterways, and land. Some of the largest sand boils from soil liquefaction ever recorded are described here. The vivid description of the devastation wrought upon the face of the land gives a picture of the dramatic changes caused by the upheaval of natural forces.

Pipkin, Bernard W., and Richard J. Proctor. Engineering Geology Practice in Southern California. Belmont, Calif.: Star Publications, 1992. This book provides a detailed description of geological engineering in areas prone to earthquakes, including attempts to prepare for the destructive effects of soil liquefaction. Illustrations and bibliographic references.

Plescan, Costel, and Ancuta Rotaru. “Aspects Concerning the Improvement of Soils Against Liquefaction.” Bulletin of the Polytechnic Institute of Iasi (2010): 39-45. Presents options for improving foundation soils to protect against soil liquefaction. Discusses jet grouting as an effective preventative method.

Plummer, Charles C., and Diane Carlson. Physical Geology. 12th ed. Boston: McGraw-Hill, 2007. A college-level introductory geology textbook that is clearly written and wonderfully illustrated. An excellent sourcebook of basic information on geologic terminology and fundamentals of geologic processes. An excellent glossary.

Spangler, M. G., and R. L. Handy. Soil Engineering. New York: Harper & Row, 1982. This book is a detailed engineering text with many equations. Nevertheless, readers will find the discussions on soil liquefaction useful even if they disregard the numerical equations.

Terzaghi, K., and R. B. Peck. Soil Mechanics in Engineering Practice. New York: John Wiley & Sons, 1948. This text is the classic work on soil liquefaction. Although it includes many equations, the authors have also provided plentiful written descriptions that are easy to understand.