by Lev GOLUBCHIKOV, Cand. Sc. (Phys. & Math.), Coordination Center "Controlled Fusion-International Projects", Russian Research Center "Kurchatov Institute", Moscow; Nikolai RYKHLINSKY, Cand. Sc. (Phys. & Math.), Institute of Innovative Methods of Geophysics, Moscow

By statistics over half of the human casualties and destructions due to natural causes result from earthquakes. Certainly an early warning about an earthquake's time and location can help save many lives. For decades geophysicists the world over have been searching for ways of predicting this natural calamity. Unfortunately their efforts have not been successful yet, and this has led to the shrinking federal funding of their research. However, this all-important problem as well as any new related ideas and practical results ought to command closer attention.

Articles in this rubric reflect the author's opinion. - Ed.

стр. 45

Lithospheric plates and blocks of the earth crust.

Iccording to Prof. R. Geller of Tokyo University, Ian earthquake prediction should involve the like-ly timing of a quake of magnitude (M)>7 about three days prior to its occurrence in a radius of 50 km from the epicenter (which is a projection of the earthquake center proper, the hypocenter, on the terrestrial surface). Yet after many abortive attempts at such predictions, at the close of the 12th century, geophysicists resorted to a less obliging term, "forecast". Forecasting can be long-term (for dozens and hundreds of years), intermediate-term (from several months up to several years), and short-term (for several days only).

In 1996 the British Royal Astronomical Society and the International Incorporated Association of Geophysics held a meeting on the "Assessment of Projects on Forecasting Earthquakes". Most of the delegates agreed in their skeptical attitude toward attempts at such forecasting because of the chaotic nature of processes occurring in the earthquake center, lack of knowledge about the mechanism of electromagnetic signals registered hundreds of kilometers away, and due to an inadequate physical basis in forecasting particular quakes. This past decade has not added any optimism to seismologists. Yet for objectivity we should mention the Opinion of Dr. Seiya Ueda, a member of the Japanese Imperial Academy of Sciences: a practical short-term forecast of earth crust tremors (earthquakes) will be available by 2015.

However, an event that took place in China on February 4, 1975, came as a fantastic anticlimax to such pessimism. The governor of Haicheng announced a general alert just 5 hours 16 minutes prior to the natural disaster. Even though the town of more than a hundred thousand was wiped out, there were but few human casualties.* Another event like this took place there, in the Celestial Empire, in Xinjiang, on April 3, 1978: thanks to an early warning 150,000 were moved to safety, and there were no casualties.

Tribute is due to the swift action of the local authorities in China guided by common sense and the experience of generations (the tremors occurred in wintertime, with reptiles getting out of their holes beforehand, a portent of momentous happenings-something was bound to happen anyway). But animal instincts are not science yet. There is a felt need for a world global service that could both report on and forecast such natural calamities. A dependable system of advance warning is also needed.

In theory we can use mathematics so as to "foresee" a seismic event provided that its physics is clear and relies on longtime observations. But this is a remote prospect. What we have now are only suggestions to explain the occurrence of certain types of earthquakes. The same is true of "early indicators", or precursors, that is regular changes in the earth crust and in the atmosphere prior to a seismic event; this is what we call "active diagnostics". We know of something like 200 indicators (symptoms) of an imminent earthquake; of these twenty are used in research only (e.g. when domestic pets sweat profusely a few hours before the event). Such symptoms are occa-

* See: V. Morgunov, "Earthquake Forecasts for Tomorrow", Science in Russia, No. 1. 2004. - Ed.

стр. 46

sionally observed over areas ten times as large as the epicenter.

Let us try to look into the nature of these dangerous seismic phenomena not studied well enough yet. In keeping with the concept of moving continents (its ideas recognized by the world scientific community), the earth is covered by nine large lithospheric plates. The plates move, they interact with one another and with rift zones in the Pacific and Atlantic Oceans (which eventually resulted in the detaching of 33 smaller crustal blocks). Like huge ice fields of the arctic basin, continents drift and rotate. Thus, the Pacific Ocean shrinks by 5 to 7 cm annually, while the Atlantic expands just as much, which means that America and Europe are drifting apart.

Such events are caused by powerful convective processes in the mantle.* Their huge energy works to deform and destroy the colliding plates, a process that triggers seismic shocks. These forces are much more superior to any other ones-say, those related to our planet's proper vibrations, lunar or solar tides, space effects, and so forth.

Statistically, three major seismic belts are identified: the Pacific (85 percent of all earthquakes), the Mid-Atlantic (8 percent) and the Mediterranean (7 percent). Volcanic eruptions, phase transitions in solid crustal formations, migrations of fluids (water, clay and mud deep under) in cracks and fractures-all that can trigger a violent seismic event (these are "triggers", or "starting mechanisms"). The anthropogenic effect, i.e. one related to human activities (the digging of man-made water reservoirs, nuclear tests, mining, and so on), also evolves as a trigger.**

There are two principal kinds of earthquakes. The first one is connected with deep faults (breaks) down to 700 km, and the tectonic movement of lithospheric plates; the other-with the displacement of smaller blocks causing an increase in the distortional (deformational) stress at contact points. Now and then earthquake hypocenters may be at a depth of 10 to 15 km; such seismic shocks are the most unpredictable and dangerous. Close to 75 percent of all violent seismic events are registered in this relatively cold and fragile upper part of the earth crust.

Yet a comprehensive approach to the problem of "surface" quakes is possible after all. It is based on the new active electromagnetic method of diagnostics of their "harbingers", or precursors, on reliable experimental data evaluation making it possible to act upon the seismic focus ("hotbed"). Special mention should be made of what is known as the "Rebinder effect". In 1928 our compatriot Peter Rebinder (elected to the USSR Academy of Sciences in 1946) discovered a new natural phenomenon: if a solid body under stress contacts a liquid or gas adsorptive-active (i.e. capable of intensive interaction) environment, it becomes less durable (hard) owing to physical and chemical processes in its surface layer.

No full mathematical and physical description of the phenomenon is available yet. It might be explained by

* See: V. Trubitsyn, "Global Plate Tectonics: New Turn?", Science in Russia, No. 2, 2003. - Ed.

** See: V. Osipov, "Threats of the Elements", Science in Russia, No. 2, 2005. - Ed.

стр. 47

The pattern of action of fluids in the beaks (tops) of solid rock cracks leading to a decrease in the hardness of rock and its subsequent destruction (the Rebinder effect of strength reduction through adsorption).

intermolecular interaction (in our case, electrically charged atoms or molecules on the walls of a crack in solid crustal rock and fluids sucked in). This mechanism was first described in qualitative terms by the Dutch physicist Johannes Diderick vanderWaals (Nobel Prize, 1910). He assumed there should be a repulsive force between close neutral molecules, and that of attraction-between remote ones; a minimum of potential energy in such relationships is at a distance where both forces counterbalance each other.

We proceed from the fact that the avalanche increase in the volume of cracks before an earthquake is caused by infiltrating fluids, as confirmed by laboratory experiments. These potent surfactants (surface-active substances) - the saturated salt solutions with acid or alkaline properties-are capable of reinforcing the Rebinder effect considerably. This process brings down (by many orders of magnitude) the surface free energy on crack walls, i.e. one connected with oriented, inductive and disperse forces of attraction between molecules, and hence it facilitates the break of their interaction. As a result the tensile strength of bed rock is significantly reduced. Let us stress it again: the Rebinder effect operates only at continuous loading on a body. In the case of earthquakes the load is sustained by the steady global movement of continental plates at different velocities. Thus, the cumulative effect of physical, chemical and mechanical components sets the stage for an earthquake.

Concentrated hot solutions were detected in an ultra-deep well (12,600 m) on Kola Peninsula at a pressure of more than 1,000 bar and temperatures above 200C; these solutions proved much more aggressive in their action on the well's walls than did ordinary water.* But, as shown by laboratory experiments, water decreases the surface energy in cracks by several orders of magnitude! Responsible for this process are largely the active atoms (molecules) of fluids getting to a predustruction place, namely the "beak" (top) of a crack, upon its expansion caused by the movement of separate crustal blocks.

People watching earthquake "harbingers" (precursors) stay merely as passive onlookers of casual phenomena, and they are unable to collect hard scientific evidence. Thus, one of the long-known active methods of the electromagnetic sounding of the earth crust registers changes of specific electric resistance in the event of seismic oscillations. But this method does not produce reliable results for constructing an adequate and consistent of the natural calamity: bounced from the bottom layers of rock, a signal "spreads thin" over a large area and, in addition, it depends on many uncontrollable factors (say, recent rainfall).

Nevertheless, geophysicists will persist in their efforts to find a way out of this impasse. A central element of our approach was advanced and tested experimentally 17 years ago by one of the authors of the present article Nikolai Rykhlinsky, and this is the method of a spatial filtration of "parasitic" scattered electric fields. It provides for a more reliable, though indirect, way of determining changes of specific electric resistance in the pre-assigned volume of the crust. Say, it becomes possible to measure the reflected electric signals registered strictly under the point of sounding if the distance between the receiver and the source of sounding radiation has been fixed. Naturally only valid, substantive data should be picked. Used as a "filter" is a sensor composed of two grounding connections - the single-point and the circular one with a preset radius.

To exclude any dependence of the parameters of the signals on the excitation current they are normalized (reduced to a common denominator)-not by its value, as it is practiced by the conventional electromagnetic sounding technique, but by a certain indifferent parameter proportional to the density of the source current. This signal is invariant with respect to lateral inhomogeneities of the medium (it does not react to the redistribution of the electric field in the medium); the main thing is that it is not affected by the distorting impact of local inhomogeneities just under the point of sounding.

* See: Ye. Kozlovsky, "10,000 Meters of Discoveries", Science in Russia, No. 1, 2001. - Ed.

стр. 48

Coordinates of experiments by the Rykhlinsky method in the rift zone of Lake Baikal: ø - control equipment; ø' - pulse generator; о², о⁴, o⁶ - registered tremors.

This is above all the surface layer of the earth crust the electric resistance of which is continuous and varies quite unpredictably, also in unison with local weather and climatic changes. The latter actually void the orthodox nonseismological methods for earthquake portents.

The new method was checked in 1990 in the rift zone of Lake Baikal (for lack of a complete set of equipment tests were carried out at 0.05 Hz). The sounding was performed daily at 11 h Irkutsk Standard Time, except Sundays and days off. The most remarkable anomaly of signals occurred on January 8, 1990: 21 h 09 min after their registration the earthquake epicenter (with the energy of one trillion J) happened to be near the point of excitation; this event was registered by seismic stations. In the community of Kabansk, where the Selenga (Buryatia) flowing into Baikal forms a delta, its magnitude (M) was equal to 4 points. However, the conventional parameter revealed no visible anomaly against the background of noises (interferences) probably caused by meteorological variations.

Electrophysical methods of sounding are based on the registration of an increase in the amplitude of a reflected signal observed with an increase in the electric conductivity of rock. We associate the physics of this process with the expanding "ensemble" of cracks and their filling with good conductor fluids. Once the measured signal has attained a peak value followed by its drop (attenuation) with a lag of about 24 h, an earthquake occurs*. The hypocenter (central, focal point) and the epicenter are determined by using the conventional measuring systems.

These experiments, mind you, were carried out more than fifteen years ago with the use of primitive, by present standards, equipment, and at one frequency only. Using two frequencies, we shall be able to sift and eliminate information coming from the upper layers of the terrestrial surface devoid of earthquake foci and thereby amplify the bounced signals severalfold. Today state-of-the-art equipment can record such signals in an automatic mode and transmit them via orbital satellites to data-processing and early warning systems reporting the probability of disastrous tremors.

More than that, as shown by experimental data, it is possible after all to "split" a precursor of a violent earthquake into several weaker ones (with low amplitudes) posing no danger. This should be done at the moment as the registered signal starts subsiding upon reaching a peak value with the aid, say, of a powerful electromagnetic pulse from an MHD generator or inductive energy accumulator. And thus we shall both predict and ward off "the fury of the Earth".

The authors wish to express their sincere gratitude to Alexander Malkin of the Institute of Physical Chemistry (Russian Academy of Sciences) for useful information and advice.

* See: G. Sobolev, "Earthquake: From Laboratory to Focus", Science in Russia, No. 5, 2003. - Ed.