Following a review of the history of earthquake prediction, this paper discusses two models of the mechanism responsible for earthquakes that have been proposed in an attempt to put earthquake prediction on a sound theoretical basis. One view, called the dilatancy-diffusion model, was developed mainly in the U.S. The alternative, sometimes known as the dilatancy-instability model, was formulated in the U.S.S.R. Both models begin with a stage in which elastic strain builds up in the earth's crust. In the next stage small cracks open in the strained portion of the crust and dilatancy becomes a dominant factor. In the Russian view the development of cracks "avalanches" in this stage. In both models it is the second stage that marks the real beginning of precursory phenomena, since the open cracks change the physical properties of the rock. Seismic velocity (the ratio of compressional-wave velocity to shear-wave velocity) drops. Electrical resistivity increases if the rock is dry and decreases if it is wet. Water flow through the rock increases (and therefore more radon enters the water from the rock). Volume in the dilatant zone increases. In the American model the number of small tremors decreases in this stage because the cracks become undersaturated in water as they increase in number; as a result sliding friction increases and inhibits faulting. The two models differ markedly in the third stage. In the American model water diffuses into the undersaturated dilatant region. The main effect of this inflow is to increase the seismic velocity and to raise the pore pressure in the cracks, weakening the rock to the point where small earthquakes increase in number and the main shock follows. In the Russian model water plays no role in the third stage. Instead the avalanche-like growth of cracks leads to instability and rapid deformation in the vicinity of the main fault. The stress load drops partially in the region surrounding the zone of unstable deformation, cracks partially close and the rock recovers some of its original characteristics. This sequence of events accounts for the increase in seismic velocity, the decrease in volume and the other changes typically observed in the third stage. The developing instability finally gives way to faulting, and the main shock ensues. In both models stress is released by the earthquake, and the crustal rock recovers most of its original properties. It is pointed out that in spite of some interesting results that have been obtained, the U.S. program is still not sufficiently supported to make prediction a reality within the next decade. Due to the lack of funds potentially important methods cannot be tested. It is estimated that an additional 30 million dollars per year could make prediction within a decade a realistic goal. A brief discussion is made of the programs being conducted in the U.S.S.R., Japan, and China.

  • Availability:
  • Corporate Authors:

    Scientific American Incorporated

    415 Madison Avenue
    New York, NY  USA  10017
  • Authors:
    • Press, F
  • Publication Date: 1975-5

Media Info

  • Features: Figures; Photos;
  • Pagination: p. 14-23
  • Serial:
    • Scientific American
    • Volume: 232
    • Issue Number: 5
    • Publisher: Scientific American Incorporated
    • ISSN: 0036-8733

Subject/Index Terms

Filing Info

  • Accession Number: 00096698
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jul 2 1975 12:00AM