SEISMICITY, NEOTECTONICS, AND RECENT DYNAMICS with special redard to the Territory of Czech Republic 250 66 Zdiby 98, tel. 02/685 7375, fax: 02/685 7056, e-mail: odis@vugtk.cz

1. INTRODUCTION

If the evaluation of shocks reGIStered before, after and during the earthquake belongs to the sphere of seismology, then the investigation of tectonic movements in the hypocentral area and in its wider environs as well as of accompanying nonseismic phenomena and their relations, is in no way concentrated in one branch of science. In the investigation of these phenomena, i.e. predominantly of "precursors" (forerunners) take part in addition to geoloGISts, geophysicists and geodesists even the scientists of very distant branches - namely meteorology, biology, medicine, etc.

The prediction of tectonic, i.e. of the strongest known earthquake based on these "precursors", is undoubtedly a very complex matter. The results still leave much to be desired. That is to say the signals of an approaching earthquake are sometimes reGIStered, but the awaited earthquake does not come. Of course, contrary cases exist, too - an earthquake appears without preceding signals. At times the reGIStered precursor is not too distinct. Therefore, it is not considered important. Only later, after a sudden earthquake its importance comes out into force - as no other, more distinct signal appears.

As it seems, a vague or even incorrect idea or presumption influencing the quality of prediction lies at the root of prognostication. This weak point of prediction was also expressed by RICHTER (1958): "... specific prediction is at best the hope for the remote future".

That is why there are good reasons for reflections. At the same time there emerge doubts whether in the mass of current data and exact measurements a simple reflection can be of any use. If not, we should have to put up with anything. Nevertheless, pondering over old problems is always possible. In my opinion no harm can be done to science.

2. EARTHQUAKE AND TECTONICS

Until now difficulties about prediction have been caused mainly owing to our unclear conception of earthquake origin. To the question "What is an earthquake?" there have been more answers. Some of them are even contradictory. That is the case of Bolt and Stiegeler. In Bolt's view (1981) earthquake is the swinging of the Earth's crust evoked by seismic waves from any source of elastic energy. On the other hand, Stiegeler (1979) argues that earthquake is "a series of shocks which generate seismic waves as a result of fracturing of brittle rocks". Both definitions refer well to the clarification of substance of earthquake, but each of them explicates its principle in a different way. This fact indictates that conceptions of the substance of this process are not definite.

Our notion of the relation of earthquake to the tectonic movement evidently represents a weak point of prognostigation and even of seismology. Naturally, this problem concerns geology as well. Rikitake (1981) quite objectively admits that e.g. the relation of the upheaval of the Earth's surface to future seismicity and to strain accumulation is not at all clear. Equally unacceptable is the conception that the tectonic movement connected with earthquake and with the origin of its precursors can take place only in the hypocentre and in its nearest vicinity.

Speaking about the tectonic movement it is necessary to distinguish primary movements due to global causes from these of secondary origin. Such secondary movements are e.g. the tilting of minor blocks, their moving up and down, etc. Of course, the genesis of earthquakes is just influenced by such local conditions. Tectonic movements in the hypocentral area are surely accompanied by shocks, but on more distant places these movements can occur without shocks, or they can be accompanied by various precursors. Somewhere only changes of strain occur - all according to local conditions.

Therefore, it is impossible to investigate the basis of earthquake in isolation from the surroundings, in which this process is developed. One can hardly raise any objections to that. Nevertheless, it still remains a basic fact, that neither seismicity nor precursors often occuring before shocks, can develop without tectonic activity.

The block movement is perceived as a fundamental and probably the only cause of tectonic earthquakes. Especially the principle of simple shifting on a rough fault plane seems to be quite clear. But what is less clear is the relation of earthquake to the areas of compression and dilatation which both simultaneously occur not only in the same territory (Havskov and Bungum, 1987), but they alternate even on one fault line (Kerr, 1980).

Compression is stated on about 95^o/_o of the Earth's surface (Kropotkin et al. 1987), i.e. also earthquakes generated by compression ought to predominate. On the other hand, earthquake in the region of dilatation (these should be much less numerous) are usually more dangerous not only for lack of "precursors", but especially because they occur in densely inhabited areas. Such places are first of all the rims of rifts and grabens, where contrastly moving blocks join each other. The same situation was discovered on the floor of these depressions (Reilinger and York, 1979) and, in addition, also on the beds of river valleys (Fig 1), which are genetically analogical to grabens (Loyda, 1976, Kats et al. 1986, Milashov and Sokolova, 1987). Armenia can serve as one of the latest examples of a strong earthquake with the epicentre in a river valley (Fig. 2).

The tilting or sinking of minor blocks forming the floor of tectonic depressions causes inevitable wandering of river courses - e.g. of the Indus and Ganges rivers (Snelgrove, 1979). Occasional earthquakes highlight these secondary tectonic movements (Fig. 3). The strongest are naturally earthquakes arising in places of fault lines crossing. Guberman and Rotvayn (1986) stated by means of analyses of 15 strong earthquakes from 1974-1984 that 14 of them developed in such places.

In addition to strong earthquakes which are exclusively considered as "tectonic" there are also weaker "volcanic" earthquakes and local "impact" earthquakes originating in caverns and in deep mines. Whereas movement along a fault plane is considered the only source of tectonic earthquakes then dilatation or rock expansion in the walls and tops of subsurface cavities are the cause of all impact earthquakes. This microdynamic process does in no way resemble mechanical block movements.

The origin of impact earthquakes always depends on the existence of subsurface cavities - caverns and goafs. Their size, shape and genesis may be quite different - these temporary phenomena permanently appear and disappear. Of course, similar underground cavities develop in the course of block movements as well. It is well known that they are filled with magma at a greater depth, and with accretion of water solutions or with the sunken top rocks at a lower depth. In this case, tectonic depressions (collapse sinks, grabens, valleys, etc.) arise on the Earth's surface. Nobody has presumably thought about a possible genetic connection of these cavities with the tectonic earthquake. It is very probable that the way to grasping the origin and mechanism of many earthquakes has thus been made difficult.

Therefore, one must pay more attention to the conception of an uniform genesis of tectonic earthquakes. One cannot but agree with the requirement of Tanaka (1985). He proposed to reinvestigate not only changes of strain in the crust but also all breakdown conditions.

2.1. Tectonic earthquakes

At first, increasing strain in the hypocentral area seemed to be the cause of tectonic earthquakes. But the two main models demonstrating this strain increase are justifiably criticized. Both models, "DD" (dilatancy-diffusion) and "LNT" (lavinno-neustoychiwoye treshchinoobrazovaniye - snowballing of cracks) try to limit stages in the forming of earthquakes. They issue from a progressive build-up of strain, which leads - as far as to transgression of limits of rock cohesiveness - to the generation of cracks, to the breakdown of the former structure. Strain build-up is then replaced by its relaxation. It is at this stage that the development of shocks occurs. The presupposed homogeneous composition and structure of the crust is considered the main defect of both models. The same objections may, of course, be raised to an analogous model (Fig. 4.) constructed by Dobrovolskiy (1980).

Nevertheless, other facts seriously contradict all the above mentioned models. E.g. after Sacks et al. (1978) the strain in the hypocentral area does not at all accumulate a longer time in advance. This happens immediately before the shock .The period of progressive strain increase - being simultaneously a period of the origin of most of all precursors - does not apparently exist. Even more important is Shimazaki's conception (1974): Earthquakes evidently originate under the condition of dilatation and not under compression.

Of course, there is neither a law nor a principle according to which strain increase necessarily must be followed by block movement and hence by an earthquake. This stage does not need to be reached at all. If the direction of tectonic move changes then naturally the earthquake causing situation must disappear. But if tectonic movement only stops then the strain and other conditions necessary for the origin of a shock may remain preserved. In such a case an only impulse is sufficient to raise an earthquake (induced earthquake) quite suddenly. Seismic waves can also serve as an impulse. These alone, of course, are not able to create a situation instrumental in giving rise to a stronger shock. Apparently, earth tides can be an impulse, too (Tamrazyan, 1957, etc.).

So it seems that the state of strain in the period before its relaxation is not at all important. An increase of strain evidently constitutes no condition necessary for the generation of earthquakes (models "DD" and "LNT"). In comparison with its former state it is quite sufficient for strain to decrease. This can happen during the process of cracks and fault opening, the moving of blocks apart, and also during the degassing of magma and its cooling. There are indications that "DD" and "LNT" models, based on the increase of strain, undoubtedly support Shimazaki's conception. According to them, shocks originate only after the culmination of strain, i.e. as late as in the period of starting dilatation. The relaxation of strain as well as dilatation are apparently much more important for the origin of earthquake than compression and increasing of strain.

Nevertheless, it is hardly expedient to reject outside the problem of the strain increase before earthquakes. This was at first recognized, later criticized and rejected. It has brought a relevant piece of knowledge in particular: the relaxation of strain and hence also dilatation initiate, in effect, each earthquake. It is wrong, of course, that more attention has been paid to the long period of increasing strain than to the shorter stage of its relaxation.

2.2. Impact earthquakes

The intensity of the fall-down impact and thus also the intensity of seismic waves depends only on the volume of fallen rocks and on the shape and size of the underground cavity. The range of earthquakes originated in such a way is very limited also due to the fact that the top layers of the cavern or of the goaf usually do not cave in at once. This process occurs in stages. This involves the caving in of some dozens up to hundreds cuODIS metres of rock from the height of few metres. An especially strong earthquake of this type was reGIStered near Recklinghausen in the Ruhr Basin, where shocks were felt in the distance up to 20 km (Köplitz, 1935).

Mines are actually the only open laboratory where it is possible to observe and measure the preliminary phase, course and result of this seismic process. As far as the information accumulated thus far is concerned, the most important finding has been the discovery of the nature of processes, which cause the cave-in of rocks from the roof of the goaf.

Original strain conditions are always changed by rock excavation and the same applies to its solving or to the formation of another underground cavity. The excavated rock has hitherto carried part of the primary strain (the load of the roof) and this part must now be sustained by surrounding rocks. One can imagine that a destructive ellipsoid is created round the new formed cavity (cavern, goaf). On its outer side there is a zone of concurrently strongly increased strain but its inner side is without any strain. This zone reaches from the ellipsoid up to the underground cavity (Fig. 5 A).

Strain decrease inside the ellipsoid leads to the expansion of the until recently compressed rock. Its volume increases and this process results in the generation of cracks. The expanding rock breaks off from the recently compressed rocks outside the ellipsoid (Fig. 5 B). Thus the fallen rock is slowly filling the lower part of the underground cavity. But a new space is necessarily created in the roof. With the repetition of the whole process the underground room, in fact, only migrates slowly upwards until it reaches the Earth's surface (Figs. 5 C,D).

Sinks of various types (sinkholes, troughs, collapse sinks, etc.) arise above goafs in this way. Deeply seated underground cavities are, naturaly, beyond the reach of our reGIStration and research. But also above them, similar sinking depressions are formed on the Earth's surface.

Underground cavities of tectonic origin are always connected with faults and cracks. Every move of one fault plane from another is followed by the formation of a free space - even though of a different shape than that of karst caverns or mine goafs. Like in mines, there must also occur the same strain changes round the destructive ellipsoid and the collapse of the top and walls of the cavity. One should exclude the existence of this process neither in the lower part of the earth's crust nor in the upper mantle, i.e. wherever block movements occur.

Of course, one must take into consideration that during the earthquake of this type a rock layer dozen or hundreds metres thick can suddenly sink. Due to the impact seismic waving of a much greater intensity (amplitude) arises in one moment than it could arise owing to the mere shifting on the fault plane. In the latter case, energy is being released over a longer period.

The investigation of tectonic earthquakes has until now omitted the possibility of the caving in of underground cavity roofs. Thus our knowledge concerning the genetic principle of such earthquakes has remained incomplete. Therefore, replies to the question "What is an earthquake?" have had to remain contradictory, too.

2.3. Combined earthquakes

There certainly exist more possibilities of the formation of underground cavities as a result of tectonic movement. Let us, however, try to single out at least one of them. If the fault line is approximately straight (undisturbed and non-undulated), then during the strike faulting only friction and weak crushing of rocks in areas of lesser asperities should occur. But in the case of local flexure of the fault plane or of its shifting along the transverse fault either crushing or metamorphosis of rocks (Fig. 6 A) should theoretically occur in the area of this disturbance. In case of block movement in the opposite direction, underground cavities should arise (Fig. 6 B).

However, if fault planes on the contact of blocks are undulated and if, in addition, their dip in depth changes, then sectors of compression and dilatation must alternate on one place during the strike faulting (Fig.7). In the compression area, i.e. in the front section of the progressing wave, strain is increased, rocks are crushed, etc. Back parts of these waves are, however, associated with dilatation cavities (Fig. 8).

Since fault lines are not only stright and their dip often changes, conditions necessary for the origin of underground cavities prevail virtually anywhere - even in the case of a simple block shift. Cavities can be filled either with rock crushed during the shift or with rocks released during their expansion in the roof and in walls.

The alternation of compression and dilatation sectors on one fault line has already been mentioned above. Fig. 8 shows that during the shifting on an undulated fault plane this alternation can occur even in the same place - in case of the closing and opening of a cavity.

The size of these tectonically generated underground cavities can be considerably great. With the thickness of a block e.g. of 5 km and with the dip of the undulated fault plane in both directions only 1^o from the vertical, the width of the cavity created on the floor of the lower part of the block can theoretically reach up to 150 m. The dip of fault plane may be, of course, greater and also faults by which blocks are delineated run deeper than to 5 km.

But shifts do not occur at once in the length of hundred metres or kilometres, they run to solely to decimetres up to several metres. Thus involving underground cavities grow only slowly. Therefore they are filled also step by step, just as migration of the roof upwards.

That concerns underground cavities if the amplitude of fault plane undulation increases downwards (Fig.7-I). If undulated fault planes open upwards (Fig.7-II), roof rocks must unevitably sink into such a conically opened cavity with far greater velocity. Provided that this upward migration of the cavity reaches the Earth's surface, terrain depressions of various sizes and shapes evolve. Therefore, the existence of lakes and water streams in these places is quite a natural phenomenon (Niini, 1967-68 ; Scheidegger 1979, 1980, etc.).

Therefore, problems grow not only in relation to earthquake genesis, but they also extend to other geological branches which have seemingly nothing to do with earthquakes.

3. EARTHQUAKE PRECURSORS

The link-up of the precursor to earthquake is namely based only on the temporal and spatial relation of both these phenomena. The prerequisite causal relation of one to the other has been only assumed till now. But any conclusions based only on suppositions must be problematical to say the least. "Precursors" cannot be an exception to the rule. Richter (1958) perceives this problem quite clearly, saying: "No one can justifiably say that an earthquake will affect a named locality in a specific future."

Nevertheless the term "precursor" is so sugestive that it very strongly influences our thinking. The presupposed causal relation of the precursor to the earthquake concerns more or less distant forerunners. Even the classification of precursors as long-term or short-term ones only helps to fix our notion i.e. to adjoin every (often isolated) "precursor" to some earthquake. The readiness with which precursors have been connected with an earthquake strengthens the supposition that earthquke is indeed taken for a dominating process with respect to the precursor. In relation to earthquake the tectonic movement seems to be a less important process. This movement is sometimes regarded as one of the precursors of the earthquake. The causality there is thoroughly inverted. Putting trust only in fixed presuppositions cannot be over-lasting. Our knowledge cannot advance in this way.

Latynina (1963) takes a quite different and probably more correct view of this causality. In her opinion e.g. anomalous tilts of the Earth's surface which appear almost simultaneously with the earthquake, but at a great distance from the epicentre, are only a part of the same planetary mechanism as is the earthquake. Thus shocks are only attributes to a certain stage of the local development of the deformation and are in no causal connection with "precursors".

Similarly Monakhov (1980) states that changes of groundwater regime before earthquake are caused solely by the development of deformation in the point of observation. These changes have nothing to do with the focus. In this sense also the fluctuation of groundwater level in wells in Tokyo Bay region at the time of the 1974 earthquake must evidently be regarded (Fig.9). The independence of the hypocentre is demonstrated by changes of groundwater regime in the area of the geodynamic polygone in Fergana (Kissin, 1988). Before the earthquake, water fluctuation in wells near the epicentre reached cca 3 m, but as much as 16 m further away.

Seen in this light, it is possible to revaluate even the still acknowledged claim that the Lisboa earthquake in 1755 caused the fluctuation of groundwater level in France and in Bohemia, i.e. in a distance of 1000 (2000) km respectively. Similar fluctuation of water level in wells in Florida in 1964 is also causally connected with the earthquake in Alaska which took place at the same time (Coates, 1981). In this case the distance is as much as 5000 km.

After Zubkov (1987) also geophysical precursors are caused by deforming processes independent of the hypocentre. For example, before the earthquake in Gazli in 1984 no changes in the geomagnetic field were reGIStered in the station near the epicentre. But changes differing in time and intensity were established in stations in the distance of 320, 470 and 500 km (Muminov et al.,1986). Many similar examples may be quoted.

The connection of precursors with the hypocentre is not confirmed by the fact that these forerunners are reGIStered in an area which is up to 10 times larger than the area where the earthquake is formed (Myachkin, 1978) either. After Ismail-Zade et al. (1982) the precursor area is even 30 times larger. Single "precursors" appear in the distance of hundreds and thousands km from the epicentre, even many years in advance. Their genetic connection with the earthquake must, therefore, amount to a presupposition but in no way a finding pertaining to the character and course of tectonic movement on the site of a future hypocentre.

But if we regard precursors solely as products of tectonic movement, then several hitherto unclarified problems suddenly disappear, i.e. precursors without subsequent earthquakes, earthquakes without precursors and the emergence of distant and isolated precursors. At the same time, the importance of the hypocentre becomes quite clear. The focus remains to be the centre of the strongests shocks, but not a source of all precursors - with the exception of imminent or alarming precursors. 3.1. Biological precursors

Areas of anomalous animals' behaviour before earthquake are distributed as irregularly as areas of water fluctuation in wells. The map with the occurence of biological precursors in the vicinity of Haicheng in 1975 (Fig. 10) makes it quite clearly - also these "precursors" do not cumulate in the epicentral area (Deng et al. 1981).

Until quite recently it has been taken for granted that the change of animals' behaviour may be influenced by more factors - e.g. by changes of gravitation, by air pressure, etc. Now, the number of possible causes has been basically reduced to two, i.e. air ionization and acoustic emission.

As influencing agents spreading in air are involved, the atmosphere has been investigated, too. Anomalies and disturbances of physical fields of great extent have been detected in the atmosphere above active faults a long time before earthquakes (Milkis, 1986; Morozova, 1988). But most of them are without any influence on animals' behaviour whatever. Therefore, scientists have come to the conclusion that it is first of all air ionization in the hypocentral area, which is the result of cracks generation in earth's crust rocks or of quick deformation and movements of block (Gufeld and Shuleykin, 1988).

Meyer and Ponomarev (1987) suppose that tectonic movement in the subduction zone is the source of these disturbances. It seems then that it is the tectonics (geodynamics) where the genesis of disturbances in atmospheric physical fields must be searched for. During all tectonic movements electric charges arise which cumulate mainly on the walls of fresh cracks, around their rims and, of course, on active faults. It is precisely here where a great number of electric charges may accumulate.

Nevertheless, acoustic emissions, first of all ultrasonic and infrasonic, continue to be problematic. Ultrasound penetrating through rocks (e.g. through granite, gabbro, basalt) generates electromagnetic radiation of the same frequency as that of ultrasound (Khatiashvili and Perelman, 1982). Thus the ionization of atmosphere may be evoked. Both phenomena are evidently correlated. Ultrasonic emissions are connected mainly with the stage of cracking. This is a symptom of accelerated deformation development (Livshits and Gavrilov, 1987).

Compression, of course, occurs with virtually every tectonic movement. Therefore, these acoustic emissions need not arise only before earthquakes. In addition, ionic emissions arise during a stronger movement, too. They are evidently not so intense as to cause panic behaviour of animals. But they still can change their reactions. In this way there originate precursors that are isolated and distant in space and time. The impact of ionization and of ultrasound are thus no longer so mysterious. But animals as well as people can react even to infrasonic waves, becoming restless and frightened (Marikovskiy, 1984).

Thus it is clear that even biological precursors cannot be successfully investigated without studying their relation to the source, i.e. to the tectonic movement which may have indirectly given cause to the generation of ionic and sonic emissions. Precursors can hardly be genetically associated with earthquake. This did not exist at the time of their appearance and possibly did not arise at all.

Precursors evidently depend on 3 different emissions originating at various stages of tectonic movement. The ultrasonic and infrasonic emissions may arise during every more intense tectonic activity. Therefore, they do not depend only on the period and place of earthquake formation. Neither above mentioned emissions, nor ionization caused by them, thus constitute the immediate precursor of an approaching shock. They are only accessory phenomena of the tectonic movement involved. Intense ionization arises only immediately before the shock. The greatest quantity of ions causing the panic of animals is generated during the opening of large cracks, of bedding and fault planes. This opening occurs at the last moment preceding the cave-in of the top wall of any underground cavity.

The changes of geophysical fields are reflected in physiological reactions of animals including man, i.e. first of all in their vascular, cardiac and nervous systems (Nersesov et al.,1988). But not everybody is able to perceive these changes and react to them. It is more likely a matter of exceptions, of people extraordinarily sensitive, nervous, sick, etc. For example, musicians and cardiacs suffer from pulse and breath vibrations but their ECG remains unchanged (Litinetskiy, 1984). In the region of active faults some people may sense a forthcoming earthquake several days in advance (Matsuda, 1984). Surprisingly enough, this also holds true of children suffering from real disorders of the digestive tract and from diseases caused by cold - all the complaints being in a greater extent (Sultankhodzhayev et al., 1982). Of course, some of these complaints appear even before an especially heavy storm, i.e. lassitude, sleepeness. In the same period before storm the change in animals' behaviour is very conspicuous indeed. They are no longer man-shy, eating what they eaten never before, etc.

Therefore, the question remains concerning the physiological influence of atmospheric ions on the organism. Here the main attention was concentrated on the neurohormone serotonin. In greater quantities it causes depressions, ill humour, sleepiness. Atmospheric ions in particular are known to influence changes in serotonin level in blood and brain. Negative ions reduce the serotonin level. This, of course, leeds to the well-being and to a greater activity. In the state of perturbation and anxiety their sedative effect is especially noticeable. In addition, they calm down the migraine, nervousness, nausea. On the other hand positive ions raise the serotonin level. This leads to hypersensivity, ill humour, and to the deterioration of the state of health of pacients, including cardiac complaints (Krueger and Reed, 1976; Tributsch, 1983).

The problem of positive and negative ions' effects has not yet been exhausted. Serotonin is not the only neurohormone that takes part in the transmission of nervous impulses. The same function is discharged by acetylcholine that also shifts over from one neuron to another and evokes reactions in the instinct and emotion sphere. But acetylcholine exercises just the contrary influence on the nervous system. In small doses, it is true, it activates most neurons, but in greater doses strong feeling of anxiety, horror, state of confusion.

Even these are the symptoms of animals' panic, arising immediately before an earthquake and a mine rockburst. It is evident, therefore, that the investigation of animal and human behaviour influenced by ion emissions should be carried out.

3.2. Alarming precursors

As for the forecasting of mountain bumps all long-term and medium-term predictions are useless. The top of the goaf must one day cave in anyway. The only problem remains: people should not be exposed to danger. Changes of stress, of acoustic emmissions, etc. are, of course, monitored and the miners are warned of the danger several hours or days in advance. But changes of stress occur just after the excavation of goafs, namely on both sides of the destructive ellipsoid. Sonic emmissions originate here quite normally. But immediately before a rockburst the genesis of cracks and their opening proceeds much faster. This situation is graphically illustrated by the "DD" and "LNT" models, as well as by Fig.4, even though these do not cover mountains bumps at all. In this stage the evolution of mountain bumps (rockbusts) is evidently the same as the evolution of earthquakes. In both cases it is the accelerated opening of cracks that leads to an intense ionization of the atmosphere, to the panic and flight of animals, i.e. to the origin of alarming precursors. Of course, this was taken into consideration at the time when miners used to take animals to their underground workplaces. Nowadays this no longer happens. Therefore, mountain bumps occur quite unexpectedly from time to time and the cave-in of the goaf roof may be result in by casualties.

The animal accompanying the miner was actually always in the close vicinity of spot the outburst was occured. This process does not form itself in the whole length of the underground corridor but only in its particular segment. In the area of shock the source of ionic emission (associated with crack opening) was actually only several metres distant from the animal that was thus within the hypocentre. Not far from here air was not so much ionized. That is why the animal would not display there either panic or perturbation.

Alarming precursors appears only within a matter of several seconds or minutes before the shock. Therefore, they cannot be reGIStered in an observatory and used for prognoses. Once perceived they render no other possibility but a quick evacuation of the place, i.e. first of all of the buildings and parts of goafs exposed to danger. A later evaluation is, of course, possible but it has not yet been carried out. But from a scientific point of view, it is correct to investigate absolutely all precursors. Considering the practical use it is precisely the alarming precursors that should not be neglected. They can save human lives at the last moment. These alarming precursors should not be neglected only for the reason they can be neither reGIStered nor evaluated in advance.

The answer to the question why animals' behaviour in mines has not been investigated thus far is quite simple. Rockbursts and impact earthquakes are generally considered phenomena or processes differing from earthquakes of tectonic origin not only in size and place, but first of all in terms of quite another mechanism and thus in genesis, too. With respect to this point of view, rockbursts and impact earthquakes have nothing to do with tectonic earthquakes. It is, of course, futile then to investigate the principle of tectonic earthquake in an area of typical impact earthquakes.

But, may be, it is not totally impossible to construct an apparatus that could supply a sensitive animal, i.e. that could react to ionic emissions and be able at the same time to set off the alarm. GeoloGISts and miners do not deal with such a simple problem and its solution is not required from physicists and bioloGISts. One cannot affirm that this investigation would be entirely useless.

4. FURTHER RELATIONS

The formation of underground cavities by dilatation can certainly take place even in great depth so that molten magma can intrude here (Fig 6B, 7). Of course, with the proceeding shift along the undulated fault plane, dilatation must be replaced by compression, and the cavity has to close (Fig. 7,8). At the same time magma, if not solidifies already, must intrude under pressure into places of the minor rock resistence, first of all into the neighbouring cracks and faults.

Even the shapes of some intruded bodies seem to verify the magma intrusion into underground cavities. E.g. the ethmolith seems to be the fill of a cavity that opens upwards. The shifts of blocks with undulated or otherwise disturbed fault planes can evidently contribute to our better comprehension of the genesis of the so called magma chambers, intrusions, and thus of the geodynamic process in general.

The direction of tectonic movement during compression is contrary to that during dilatation. If compression is especially intensive, the crushing of rocks and sometimes also their matamorphosis occur on the contact of undulated fault planes (Fig. 6A). Depending on compression intensity, rocks may even melt. The tilt of fault plane plays an important part, of course. If the tilt is near to the vertical (Fig.7), then the crushing of rocks, their metamorphosis and their melting depend mainly on three factors:

• on the cohesiveness of rocks and on their melting point,
• on geostatic pressure,
• on lateral stress.

Lateral stress may, of course, surpass the value of geostatic pressure several times. If the dip of the fault plane diminishes then the influence of lateral pressure inevitably decreases too. If the dip approaches to the horizontal then its influence disappears altogether. Instead of compression the mechanic movement of blocks comes into being (reverse faults, nappes).

By the melting of rocks during compression primarily light sialic magma should arise. Their different chemical and mineralogical composition is then caused only by the uneven quality of initial rocks which had been melted by compression.

The problem of granite origin has been a truly perennial issue. Although there have always been several different concepts of this genesis we still lack definite, i.e. real knowledge. Nevertheless, some 20 years ago new findings pertaining to the principle of granite origin, as mentioned in this reflection, emerged. Here are some opinions chosen at random:

• the hitherto used theories do not at all elucidate the genesis of granite (Shuber, 1972),
• the igneous origin of granite has not been proved by research (Roonwall, 1968),
• the acid magma is formed by the rock melting in sunken parts of geosynclines (Borchert, 1967). It can be mentioned here that geostatic pressure is determined by the mass of overlying rocks. This mass does not change during the only upward and downward movement.
• in sunken area magma cannot regenerate (Shimadzu and Urabe, 1967),
• in the rocks of the crystalline complex and of the folded fundament there appears horizontal stress which is much more intense than geostatic pressure (Kropotkin and Frolov, 1974).

Of course, other problems are connected with the unsolved problem of granite genesis. It is evident that they remain unsolved too:

• the Californian biotite eyed-gneiss is of Precambrian age (1820 mill years) but its biotites are only 11-14 mill years old (Stern et al., 1968).

The melting of rocks and the genesis of granite are naturally connected with the problem of metamorphosis. Both mentioned processes evidently arise in the some way, i.e. by rocks heating caused by increased pressure. It is only the pressure intensity that makes differences. It is necessary to keep on distinguishing at least two types of solidified granitic bodies:

• concordant bodies when "the igneous rock" continuously transform itself into the neighbouring metamorphic facies,
• discordant bodies when the molten rock intruded to another place and no longer is "in situ". This intrusion seems to be the result of tectonic movement which had to take place here.

Needless to stress the problem of granite genesis is not at all solved by this treatise. A granite massif is sometimes so large that the above mentioned mode of its genesis cannot be taken into consideration, if such a massif arose really at once and not in parts. The influence of heat accumulation released during the disintegration of radioactive elements is often discussed too. This problem has not been definitively solved. That is why this reflection offers just a suggestion and does not insist on its own infallibility.

The principles of geology seemed to be quite clear for a long time. Mainly statistical data, notions and presumptions together serve to establish a coherent system with no apparent weak points. But further investigation slowly changed the views on this comprehensive system, and its weak points are coming out into the open now. It is quite clear that even in geology it is not only suggestions but also real phenomena and processes that are interconnected and that help to form a new scientific system. This is why new investigations are still highly required (Tanaka, Rikitake). The problems of geology are numerous and therefore each new and well-documented idea must be welcome.

5. REFERENCES

BOLT, D. (1981): Zemletryaseniya. Mir, Moskva, 256.

BORCHERT, H. (1967): Vulkanismus und oberer Erdmantel in ihrer Beziehung zum äusseren Erdkern und zur Geotektonik. Boll. geofis.teor.appl., 9, No. 15, 195-203.

BUSKIRK, R.E., FROHLICH, C. and LATHAM, C.V. (1981): Unusual animal behavior before earthquakes:

a review of possible sensory mechanism. New Geophys. Space Phys., 19, No. 2, 247-270.

BYKOVA, V.V., STAKHOVSKIY, P.R., FEDOROVA, T.S., KHRUSTALEV, Yu.A., DERYAGIN, B.F. and TOPOROV, Yu.P. (1987): Elektronnaya emissiya pri razrushenii gornykh porod. Izv. ANSSSR, Fiz. Zemli, No. 8, 87-90.

COATES, D.R. (1981): Environmental Geology.,Wiley, New York, 732.

DENG, Q.,JIANG, P., JONES, L.M. and MOLNAR, P.(1981): A preliminary analysis of reported changes in ground water and anomalous animal behavior before the 4. February Haicheng Earthquake. In: Simpson D.W. and Richards P.G. (Eds.): Earthquake prediction. M.Ewing Ser. IV, Am.Geophys. Union, 543-565.

DOBROVOLSKIY, I.P. (1980): O modeli podgotovki zemletryaseniya. Izv.ANSSSR, Fiz.Zemli, No. 11, 23-31. p>DUNRUD, C.R. and OSTERWALD, F.W. (1980): Effects of coal mine subsidence in the Sheridan, Wyoming area. Geol. Surv. Prof. Paper 1164, 49.

ENGELDER, T.(1978): Aspects of asperity-surface interaction and surface damage of rocks during experimental frictional sliding.In:Rock friction and earthquake prediction. Birkhäuser, Basel-Stuttgart, 705-716.

FAVORSKAYA, M.A. (1967): On the problem of the origin of acid magma. Bull.volcanol. , 30, 93-98.

GOKHBERG, M.B., GUFELD, I.L. and DOBROVOLSKIY, I.F. (1988): Istochniki elektromagnitnykh predvestnikov zemletryaseniy. Dokl.ANSSSR, 250, No.2,323-326.

GOLD, T. and SOTER, T. (1979): Brontides: Natural explosive noises. Science 204,No.4391, 371-375.

GUBERMAN, Sh.A. and ROTVAYN, I.M. (1986): Proverka rezultatov prognoza mest vozniknoveniya silnykh

zemletryaseniy (1974-1984). Izv.ANSSSR, Fiz.Zemli, No.12, 72-74.

GUFELD, I.L. and Shuleykin V.N. (1988): Variatsii atmosfernogo toka v period podgotovki zemletryaseniy. Izv.ANSSSR, Fiz.Zemli, No. 2, 81-84.

HAMILTON, E.I. (1966): The isotopic composition of lead in igneous rocks. I. The origin of some Tertiary granites.Earth Planet. Sci. Lett. 1, No.1, 30-37.

HAVSKOV, J. and BUNGUM, H.(1987): Source parameters for earthquake in the northern North Sea. Norsk Geol. Tidss. 67, No. 1, 51-58.

HUAN, W. and SHI, Z. (1987): Great earthquakes of M 8 in the mainland of China and their evolution. Tectonophysics, 138, No.1, 55-68.

ISMAIL-ZADE, T.A., LUBIMOVA, Ye.A., MUKHTAROV, A.Sh., SULTANOVA, Z.Z. and FIRSOV, F.V. (1982): Variatsii temperatury v seismoaktivnom rayone po nablyudeniyam v gidrologicheskikh skvazhinakh. Izv. AN SSSR, Fiz. Zemli, No. 9, 71-75.

KATS, J.G, POLETAYEV, A.L. and RUMYANTSEVA, E.F. (1986): Osnovy lineamentnoy tektoniki. Nedra, Moskva. 141.

KERR, R.A. (1980): Concern rising about the next big quake. Science, 207, NO.4432, 748-749.

KHATIASHVILI, N.G. and PERELMAN, M.Ye. (1982): Generatsiya elekromagnitnogo izlucheniya pri prokhozhdenii akusticheskikh voln cherez kristallicheskiye dielektriki i nekotoryye gornyye porody. Dokl. AN SSSR, 263, No. 4. 839-842.

KISSIN, I.G. (1988): Vysokoamplitudnye predvestniki zemletryaseniy i "chuvstvitelnyye zony" zemnoy kory. Izv. AN SSSR, Fiz. Zemli, No. 6, 3-13.

KÖPLITZ, W. (1935): Beiträge zur Frage der Gebirgsschläge. Glückauf, Essen, 71, No. 43, 1021-1028, 1054-1057.

KRASILOVA, N.S., NOVIKOV, I.F. and KISELEV, V.A. (1988): Nekotoryye fiziko-khimicheskiye protsesy soprovazhdayushchiye razrusheniye mineralov i gornych porod. Inzh. Geol. No. 4, 109-111.

KROPOTKIN, P.N. and FROLOV, B.N. (1974): Napryazhennoye sostoyaniye i skolovyye deformatsii v kore i verkhney mantii. In: Fiz. svoystva i stroyeniye verkhney mantii. Nauka, Moskva.

KRUEGER, A.P. and REED, E.J. (1976): Biological impact of small air ions. Science 193, No. 4259, 1209-1213.

LATYNINA, L.A. (1963): Osnovnyye predstavleniya o prichinakh i kharaktere izmeneniya rezhima tektonicheskikh dvizheniy pered krupnymi zemletryaseniyami. In: Sovrem. dvizh. zemnoy kory, Moskva, 211-217.

LEHMAN, K. (1919): Bewegungsvorgänge bei der Bildung der Pingen und Trogen. Glückauf, Essen, 55, No. 48, 933-942.

LITINETSKIY, I. (1984): Tajemnství květu smrti. Praha,LN-Moskva, Mir, 176 .

LIU, G. (1987): The Cenozoic rift system of the North China Plain and deep internal process. Tectonophysics 133, No. 3-4, 277-285.

LIVSHITS, L.D. and GAVRILOV, B.G. (1987): Istochniki akusticheskoy emissii i ochag razrusheniya. Dokl. AN SSSR, 292, No. 3, 272-275.

LOYDA, L. (1976) The tectonic origin of river valleys and its geodetic investigation. Rozpravy ČSAV, MPV, 86, No. 11, 83.

MARIKOVSKIY, P.I. (1984): Zhivotnye predskazyvayut zemletryaseniye. Nauka, Alma-Ata, 144 .

MATSUDA, T. (1984): Seismicheskyye shramy. In: Asada, T., et al.: Metody prognoza zemletryaseniy. Nedra, Moskva, 39-66.

MEYER, K., PONOMAREV, A. (1987): Electrotelluric forerunners to earthquakes in Kamchatka. Tectonophysics, 138, No. 3-4. 341-347.

MILASHEV, V.A. and SOKOLOVA, V.P. (1986): Avtonomnost neotektonicheskikh dvizheniy u blokov zemnoy kory s razlichnoy megatreshchinovatostyu. Izv. AN SSSR, Fiz. Zemli, No. 7, 121-126 pp.

MILKIS, M.R. (1986): Meteorologicheskiye predvestniki silnykh zemletryaseniy. Izv. AN SSSR, Fiz. Zemli, No. 3, 36-47.

MONAKHOV, F.I. (1979): Prodolzhitelnost kratkosrochnykh predvestnikov zemletryaseniy. Izv. AN SSSR, Fiz. Zemli, No. 1, 91-94.

MOROZOVA, L.I. (1988): Oblachnyye avtografy zemnych nedr. Priroda, Moskva, No. 5, 55-56.

MUMINOV, M. Yu., ABDULBEKOV, K.N., AZIZOV, G.J. and SIVTSOV, N.M. (1986): Analiz variatsiy geomagnitnogo polya na seti prognosticheskikh stantsiy Uzbekistana v svyazi s Gazliyskim zemletryaseniyem 19. marta 1984 g. Uzbek. geol. zh. No. 5, 6-9.

MYACHKIN, V.I. (1978): Fizicheskiye osnovy prognoza zemletryaseniy. Zemlya i vselennaya, Moskva, No. 6, 15-19.

NERSESOV, M.Yu., PROTASOV, V.R. and SIDORIN, A.Y. (1988): Opyt ispolzovaniya slaboelektricheskikh ryb pri poiske predvestnikov zemletryaseniy. Izv. AN SSSR, Fiz. Zemli, No. 9, 93-99.

NIINI, N. (1967-68): A study of rock fracturing in valleys of Precambrian bedrock. Fennia 97, No. 6, 1-60.

NIKONOV, A.A. (1980): K otsenke vremeni i razmerov oblasti proyavleniya predvestnikov po anomalnomu povedeniyu zhivotnykh pered zemletryaseniyem 2. XI. 1978 g. v Sredney Azii. Izv. AN SSSR, Fiz. Zemli, No. 8, 108-111.

NUR, A. (1978): Nonuniform friction as a physical basis for earthquake mechanics. In: Rock friction and earthquake prediction. Contrib. Curr. Res. Geophys., Birkhäuser, Basel-Stuttgart, 964-984.

RASTVOROVA, V.A. and SHCHERBAKOVA, E.M. (1965): Molodyye dvizheniya po razlomam v Prikazbekskom rayone Tsentralnogo Kavkaza. In: Sovrem. dvizh. zemnoy kory, 2, Tartu, 175-182.

REILINGER, R.E. and YORK, J.E. (1979): Relative crustal subsidence from levelling data in a seismically active part of the Rio Grande Rift, New Mexico. Geology, No. 7, 139-143.

RICHTER, C.F. (1958): Elementary seismology. Freeman, San Francisco, 768.

RIKITAKE, T. (1981): Practical aaproach to earthquake prediction and warning. In: Curr. Res. Earthq. Predict. I., Tokyo, 1-56.

ROONWAL, G.S. (1968): Preliminary note on the geochemistry of Singhbum granite, India. Sci. Cult. 34, No. 3, 121-122.

SACKS, I.S., SUYEHIRO, S., LINDE, A.T. and SNOKE J.A. (1978): Slow earthquakes and stress redistribution. Nature, 275, No. 5681, 599-602.

SADOVSKIY, M.A. (1982): Elektromagnitnye predvestniki zemlestryaseniy. Nauka, Moskva, 88.

SADOVSKIY, M.A. and NERSESOV, I.L. (1978): Voprosy prognoza zemletryaseniy. Izv. An SSSR, Fiz. Zemli, No. 9, 13.

SCHEIDEGGER, E. A. (1979): The principle of antagonism in the Earth's evolution. Tectonophysics, 55, T7-T10.

SCHEIDEGGER, E.A. (1980): Alpine joints and valleys in the Neotectonic stress-field. Rock Mech., Suppl. 9, 109-124.

SHIMAZAKI, K. (1974): Pre-seismic crustal deformation caused by an underthrusting oceanic plate, in eastern Hokkaido, Japan. Phys. Earth Planet. Inter., 8, No. 2, 148-157.

SHUBER, A. (1972): Problema granita. In: Mezhdunar. Geokhim. kongr. 1971, Vol. 3, Book 2. Moskva, 438-472.

SIMADZU, J., URABE, T. (1967): Konvektsiya v mantii i genezis magmy. (res. Engl.). J. Seimol. Soc. Japan 20, No. 3, 192-197.

SNELGROVE, A.K. (1979): Migration of the Indus River, Pakistan, in response to plate tectonic motion. J. Geol. Soc. India, Bangalore, 20, 392-403.

SONDERGELD, C.H., GRANDRYD, L.A. and ESTEY, L.H. (1984): Acoustic emisions during compression testing of rock. In: Proc. 3rd Conf. 1981, Trans Techn. Publ. Ser. Rock Mech., Vol. 8, 131-145.

SPETZLER, H. and MIZUTAMI, H. (1987): Prediction failure in rocks and its implications for predicting eathquakes. Tectonophysics, 138, No. 1, 109-113.

STERN, T.W., NEWELL, M.F. and HUNT, C.B. (1966): Uranium-lead and potassium-argon ages of parts of the Amorgosa Thrust Complex., Death Valley, California. Geol. Surv. Prof. Paper No. 550-B, 142-147.

STIEGELER, S.E. (1979): A dictionary of Earth sciences. Pan Books, Aylesbury, 2nd ed. 312.

SULTANKHODZHAYEV, A.N., SHARMATOV, M.Sh., SUKHOV, R.R. and ISMAILOVA, F.G. (1982): Na puti poiska biologicheskikh predvestnikov Nazarbekskogo zemletryaseniya. Uzbek. geol. zh. No. 2, 37-40.

TAMRAZYAN, G.P. (1957): Razrushitelnyye zemletryaseniya Zakavkazya i fazy Luny. Izv. AN SSSR, Ser. geofiz. No. 12, 1510-1511.

TANAKA, Y. (1985): Crustal stress measurement and earthquake prediction. J. Geol, Soc. Japan, 31, No. 1, 73-85.

TRIBUTSCH, H. (1983): When snakes awake. Animals and earthquake prediction. Cambridge-London, 248.

VOGT, J. (1979): Les tremblements de terre en France. (Abstr.). Mém. du BRGM, No. 96, 220.

VOROBYEV, A.A. (1970): O vozmozhnosti elektricheskikh razryadov v nedrakh Zemli. Geol. i Geofiz. No. 12, 3-13.

WAKITA. H. (1975): Water wells and possible indicators of tectonic strain. Science 189, No. 4202, 553-555.

WEI, G. (1987): The 1983, M 5.9 earthquake of Heze, Shandong Province, with its short-term and imminent precursory phenomena. Tectonophysics 138, No. 1, 25-32.

YE, H., ZHANG, B. and MAO, F. (1987): The Cenozoic tectonic evolution of the Great North China: two types of rifting and crustal necking in the Great North China and their tectonic implications. Tectonophysics 133, No. 3-4, 217-227.

ZUBKOV, S.I. (1987): Vremena vozniknoveniya predvestnikov zemletryaseniy. Izv. AN SSSR, Fiz. Zemli, No. 5, 87-91.

ŽILAVÝ, B. (1986): Vplyvy poddolovanija. Bratislava, Alfa, 268.