Abstract-With the most recent seismological and tectonic data, intraplate seismicity of the western Arctic is shown to be caused mainly by the following three factors: intraplate stresses generated at plate boundaries, vertical tectonic movements, and load of fast accumulating sedimentary cover.

 

The western Arctic is assumed here to include water areas of the Norway-Greenland basin and Eurasian subbasin, their continental margins, and adjacent shelf and coast areas. Being the northernmost segment of the global rift system of seismic belts, the mid-Arctic belt of earthquakes traces the divergent boundary between the North American and Eurasian plates and is a single Arctic zone of recent intraplate seismicity. All other seismically active Arctic zones are related to the lithosphere response to intraplate stresses, the role of plate boundaries being insignificant (Fig. I).

As is now commonly acknowledged, earthquakes here are caused mainly by partial stress release in the transarctic interplate zone, glacial rebound, and lithosphere response to the thick overburden. Obviously, these effects differ in scale and are largely dependent on the location and development of a specific area. The first effect being transarctic, one may naturally suppose that its contribution to seismicity is proportional to the distance from the nearest interplate zone. Two other effects are local, and their influence zones can be easily predicted and delineated.

Though higher seismicity is due solely to external forces of geologic-tectonic origin that generate excess stresses in the lithosphere, specific features of seismicity, including earthquake intensity, source depths, and focal mechanisms, are equally dependent on properties of the real geological medium affected by these forces. While external tectonic forces in interplate seismicity zones are strong enough to deform the lithosphere, and initial structure modifies the position and forms of new structural elements, the structural control is considerably more important in zones of intraplate seismicity, where the new structural pattern inherits much from the older one. Therefore, earthquakes occur predominantly in weakened lithospheric zones, a fact important for seismic risk assessment. For this reason, structural-tectonic constraints are essential for geodynamics of intraplate seismicity zones and assessment of tectonic origin of intraplate earthquakes.

It has been suggested that stresses generated at plate boundaries are partially released in intraplate regions: this is confirmed experimentally, in particular, by data on Fennoscandia. Research into this problem is carried in two directions.

First, regional stress fields are compared with predicted ones, with both the position of a region in question relative to the nearest segment of interplate boundary and specific fault tectonics of the region taken into account. Numerous data from direct measurements in holes and openings, fault plane solutions of weak earthquakes, and geological studies of fault zones show unambiguously that the main feature of the regional stress field in Fennoscandia is the presence of a horizontal or subhorizontal compressive component of predominant NW-SE orientation, which complies with the lithospheric dynamics of the Mona Ridge, the nearest segment of the mid-oceanic ridge [e.g., 1-5]. Stability of the general solution, expressed as the spread of individual determinations about the averaged one, depends on the orientation of fault systems in various areas of the region. As is suggested by general considerations and observations, extension in axial zones of basins generates, first of all, shear stresses in weakened zones of adjacent continental margins. These zones are normal to basin axes, so that the compressive component grows wherever this orientation changes. Apart from Fennoscandia, such a pattern is well-pronounced along the Eurasian margin of the Norway-Greenland basin and Eurasian subbasin, where earthquakes are confined to transverse faults and trenches (Figs. 2 and 3). Thus, epicenters concentrate near the Senja fault zone, Lofoten basin and in the Franz-Victoria, Voronin, and Saint Anna trenches [6, 7]. Focal mechanisms here yield evidence for strike-slip and reverse faulting, with one of the subvertical nodal planes being close to the fault plane. Horizontal compression also dominates in back zones of continental slopes (Kola Peninsula and New Land) [8]. Areas of compression on flanks of extension zones occur in the Laptev Sea [9, 10] and New Siberian Islands [11].

The proximity of Spitsbergen to the Mid-Atlantic Ridge also indicates that its seismicity is related to release of stresses generated in the oceanic rifting zone. Sykes and Sbar [12] drew this conclusion on the basis of the absence of uninterrupted zones of seismic activity connected with the mid-Atlantic belt. This hypothesis explains the seismological data acquired since their publication [13-15]. In the eastern Spitsbergen zone of intersection of the nearly east-west trending De Geer fault and a fault of the NNW trending system predominant in the archipelago, earthquake epicenters are confined to the former fault orthogonal to the Knipovich Ridge, the nearest fragment of the midoceanic ridge (Fig. 4). Earthquakes in the northern North East Land group along the predominant system of faults, nearly orthogonal to the nearby Gakkel Ridge. Fault plane solutions indicate the strike-slip mechanism. Mitchell et al. [15] suggested that local seismicity of Spitsbergen is due to weakened zones of fluid inclusions in frac­tures and microfractures. However, their hypothesis is unacceptable, because it implies that higher seismicity should be expected mostly in the Bok-fjord area. North-western West Spitsbergen Island, where Quaternary volcanoes and related thermal springs are known.

Interestingly, Kola geophysicists have found that higher-seismicity zones in the Barents Sea-White Sea region coincide with the areas of minimum heat flow [16]. This can be explained only in terms of induced or «passive» seismicity [9] caused by release of stresses generated outside the region and accumulated by the coldest, more brittle lithospheric blocks.

As yet no data are available that would confirm the presence of interplate stresses in the Greenland margin of the Norway-Greenland basin. Focal mechanisms were determined here only for three offshore earthquakes that occurred nearby the shoreline [17]. All three solutions (one of them is poorly constrained) indicate normal faulting with the extension axis across the shoreline. A similar stress pattern in the eastern offshore zone of the Baffin Island is associated with bottom subsidence and land uplift.

The second direction of research into the influence of tectonic processes in axial spreading zones on the generation of seismogenic stresses in margins consists in comparison of seismicity patterns in these areas. Unfortunately, these studies based on statistical analysis of data are limited, because their results are representative only for areas of fairly high seismicity where data of long-term (a few tens of years at least) reliable seismic monitoring are available. In the region under consideration, only data on Fennoscandia meet these requirements, but even here instrumental data of the first half of the century and especially of the war years are unlikely to be representative. However, Skordas et al. [18] calculated and plotted the seismic energy released in Fennoscandia and the axial zone of the Norway-Greenland basin for over 70 years beginning in 1917. Both plots are similar, with the cross correlation coefficient reaching 0.7-0.8 and time delay for Fennoscandia varying from 0 to 3-4 years.

The contribution of interplate tectonic processes into the higher intraplate seismicity is indubitable, but variability of this effect is also evident. Since the Norway-Greenland basin and Eurasian subbasin are elongated so that their half-width (distance from the axial zone to boundary) is of the same order of magnitude both in the widest and narrowest parts of the basins, one should expect the mean level of intraplate seismicity in their peripheral zones to vary slightly, even if inhomogeneity of medium is taken into account. However, a spotwise pattern of intraplate seismicity is evident from Figs. 1, 2, 3, and its most prominent features are a markedly lower seismicity level at margins of the Eurasian subbasin, even though it is narrower than the Norway-Greenland basin, and Fennoscandia has higher seismicity in contrast to adjacent areas. In my opinion, these facts can be explained by three factors.

The first factor is subjective and related to the aforementioned irregularity of seismic station network. While tens of stations are now operating in Fennoscandia, the number of stations operating over a vast area between the Kola and Chukot Peninsulas did not exceed 5-7 in previous years, and now there are none between stations Apatity and Iul'tin. The deficiency of our knowledge of seismicity in the Eurasian subbasin and adjacent areas was demonstrated convincingly by field observations performed by the Institute of Arctic Research and Research-and-Production Association «Sevmorgeo» [see, e.g., 9], which revealed previously unknown zones of higher seismicity, including that of the New Siberian Islands. Earthquakes recorded north of the islands suggest a nonzero seismicity level at the Lomonosov Ridge, which is corroborated by records of earthquakes in the zone of its junction with the continental slope of North Greenland and Canadian Arctic Islands.

The second possible factor is the lower spreading rate along the subbasin axis due to its proximity to the opening pole. Supposedly, slow accumulation of stresses is largely adjusted by aseismic motions in weakened zones.

The third, and apparently main, factor is the preferred occurrence of higher seismicity zones in areas where regional stresses transmitted from axial zones of basins are complemented with other sources of extra stresses, which may be even more efficient in a given region.

Fennoscandia, exhibiting the highest seismicity within the region considered, provides evidence in favor of this statement (Fig. 5). Presently, both the Baltic shield and related Norway Caledonides to the north-west experience intensive uplift, well documented by historical records. The maximum net amplitudes of this uplift, reaching 250 m for the late postglacial period, are observed in central and northern parts of the Gulf of Bothnia, Baltic Sea (Fig. 6). As is mentioned above, data on focal mechanisms in Fennoscandia give evidence for prevailing strike-slip movements on subvertical planes, indicating that the effect of intraplate processes is still a main factor enhancing seismicity. However, this conclusion does not imply that horizontal movements control the present-day geodynamics of Fennoscandia. Calculated seismic moments are too small to account for the amplitudes of differentiated movements, established from repeated levelings [4]. Therefore, vertical movements are mostly aseismic, which is corroborated by their presence on both seismic and aseismic faults. However, numerous reports on mechanisms of different types, as well as a considerable scatter often observed in azimuths of subhorizontal compression axes, are evidence both for the existence of local areas where excessive seismogenic stresses are controlled by different tectonic factors and for a filtering effect that distorts the observed stress field of fault tectonics in specific areas of the region.

Numerous systems of mutually orthogonal faults crossing the Baltic shield are responsible for its smallblock structure and related kinematic and dynamic nonuniformity of movements. Evidently, in zones of the most intensive vertical movements, the latter are realized partly as earthquakes, which can be demonstrated by the following examples. The most pronounced clusters of epicenters are observed in zones of maximum net amplitudes of late postglacial uplifts. These are the western coast of the central and northern Gulf of Bothnia (an uplift of up to 250 m), southwestern Sweden, southwestern Norway, and coast of northern Norway (local maximums of up to 150-200 m and sharp gradient), Kandalaksha Gulf in northern central Finland and Karelia (up to 200 m and sharp gradient), and zone of differentiated movements near the Lake Venern. In this respect, the seismically active central part of the Kola Peninsula is noteworthy, where the intensively uplifting Khibiny and Lovozero rock masses (including a maximum of up to 200 m) alternate with the subsiding Lovozero and Umbozero basins. In contrast, the entire eastern part of the Kola Peninsula, which is, on the whole, uplifted insignificantly and is now beginning to subside, is virtually aseismic, except for the coastal zone of junction of the Baltic shield and Barents plate (Figs. I and 5). A different focal mechanism can be expected in zones of maximum uplift, as is indicated by the fault plane solution for the Solberg, September 29, 1983, earthquake at the northwestern coast of the Gulf of Bothnia [5, 20]. The solution yielded a predominantly normal fault mechanism characteristic of extension zones including axial and near-axial zones of uplifts.

Fault structure constraints on the development of stress field can also be illustrated as follows. The stress field in southern and northern Sweden being on the whole similar, the scatter in azimuths of horizontal compression axes about the general NW-SE trend is markedly larger in northern Sweden, which is explained by different orientations of mutually perpen­dicular fault systems (S-N and E-W on the south and diagonal on the north [4]). Also, the effect of local fault structure explains the existence of focal mechanisms other than the strike slips typical of Fennoscandia. Obviously, horizontal compression stresses, propagating from northwest and north-northwest and giving rise mostly to strike-slip movements on faults of similar strike, are favorable for development of reverse faulting and earthquake occurrence. This is consistent with numerous determinations of focal mechanisms of weak earthquakes, showing that there are fewer reverse faulting solutions than strike-slip ones [4]. Compression strength of rocks being considerably lower than their tensile and shear strength [21], reverse movements should be expected to occur at strong earthquakes. In fact, six of eight individual determinations of focal mechanisms by the first-break method yielded mechanisms of predominantly reverse-faulting type [8, 22-24]. For the two remaining (southern Sweden), geological data and aftershock patterns yield preferable fault planes coinciding with NE and NNE trending faults orthogonal and suborthogonal to the direction of general compression [5].

Contemporaneous uplift is characteristic of areas of fairly high seismicity, such as Spitsbergen and northeastern Greenland (6-7 mm/yr [25, 26]); there is evidence for earthquakes in New Land (4-5 mm/yr [19]), Franz Joseph Land (3-5 mm/yr [27]), and Northern Land (2-3 mm/yr [19]). Thus, combined analysis of mapped epicenters and recent vertical movements in coast and island areas of Arctic seas shows that all seismically active margin areas of the Norway-Greenland basin and Eurasian subbasin (as well as northeastern Arctic seas) are zones of uplift. However, one cannot state with certainty that all zones of uplift are seismic. Though weak seismicity of areas now experiencing uplift, such as Franz Joseph Land, New Land, North Land, and (aseismic) Taimyr Peninsula, may be apparent due to a deficiency of observations there, one should recognize the effect of uplift on seismicity and consider a value of 5-6 mm/yr as a threshold. Subsiding regions do not exhibit higher seismicity.

In summary, the general conclusion is that higher seismicity of margins of the Norway-Greenland basin and Eurasian subbasin is due, on the whole, to stresses transmitted from the interplate zone of the midoceanic ridge and applied to the lithosphere activated by the contemporary uplift. The observed seismic effect results from combined action of these sources and is proportional to the distance of each concrete region from the nearest segment of plate boundary, spreading rate along this segment, and uplift rate of the region. Similar levels of seismicity of Fennoscandia and Spits­bergen differing greatly in their distances from the plate boundary (500-600 and 150-400 km, respectively) can be explained by the compensating effect of vel1ical movements of different intensity (9- 10 and 5-6 mm/yr, respectively). Other margins, for which their distance from the boundary is nearly the same as that of Fennos­candia and uplift rates are commonly lower as com­pared with Spitsbergen, have a markedly lower level of seismicity.

The recent uplift of margins of the Norway-Green­land basin and Eurasian subbasin, and particularly the uplift of Fennoscandia and Spitsbergen, were first gen­erally acknowledged to be solely of glacio-isostatic ori­gin, but now glacio-isostasy is commonly believed to play a subordinate role, and the leading role is ascribed to deep tectonic processes. The glacio-isostatic hypoth­esis was based on the scheme of De Geer-Hgbom, according to which Fennoscandia experiences dome­shaped uplift, with its rate being maximal at the center of the Baltic shield, northern Gulf of Bothnia and decreasing toward the periphery. This scheme was supported by theoretical estimates [e.g., 28], showing that an ice sheet 500 km in diameter and I km or more in thickness would bend the crust. The dominating role of the glacio-isostatic effect in generating recent vertical movements is at variance with the following main facts.

First, as is evident now, dome-shaped uplift, if any, takes place only in the central part of the shield. The scheme of summary uplifts show that, in addition to the central maximum, there are peripheral ones outside the area of maximal development of the ice sheet [29]. As is shown above, their distribution correlates well with clusters of epicenters.

Second, numerous estimates [30-33] date the uplift maximum rate from a moment not later than one thou­sand years after the complete deglaciation. Then the rates decrease exponentially with time and drop drasti­cally as early as 1-1.5 thousand years after the maxi­mum. In the first thousand years the uplift reaches one third of its maximum, and it is in this period that seis­micity could be controlled by the glacio-isostasy. Thus, Wahlstrom [5] argues that several surface faults of late ­and postglacial origin, as well as some fault scarps, formed either simultaneously or within a short time interval (possibly a few tens of years) as a result of energy release at one or several strong earthquakes. The complete isostatic adjustment is generally estimated to occur 8-9 thousand years after deglaciation, while Morner [32] argues that the glacio-isostatic component became inefficient as early as 900 years after deglacia­tion. In any case, the general complete deglaciation of Fennoscandia 9000-9500 years ago at the latest implies that contemporary glacio-isostatic movements in the region (with the possible exception of the area of maximum isostatic load, coinciding with the central Both­nia-Kandalaksha basin) affect the development of seismogenic stresses insignificantly.

Third, based on geological and paleogeographic evidence, Stille [34] concluded that the Fennoscandia uplift began earlier than the last glaciation; i.e., the recent uplift of Fennoscandia inherits an earlier uplift. Similar results have been obtained for Spitsbergen and Greenland. Thus, Semevskii [25] believes that the uplift of Spitsbergen began on the north of the archipel­ago as early as the Late Devonian and has proceeded with varying intensity for a long time since then. The absence of the glacio-isostatic component in the Holocene (38-10 thousand years ago) uplift of fjord coasts in Spitsbergen and Greenland is constrained by typically marine sedimentation conditions that existed at that time there [26].

In my opinion, the totality of available data and the­oretical results implies that vertical movements in regions that have been subjected to glaciation include both the glacio-isostatic and tectonic components, whose contributions vary considerably (from zero to maximum) depending on region and time interval under consideration. As is shown above, the glacio-iso­static effect may have dominated during the glaciation period and next 1-3 thousand years. Presently its role, varying from region to region, is reduced to modifica­tion of vertical movements. Thus, within Fennoscandia, it is more significant in the area of maximum glaciation (northeastern Gulf of Bothnia) and minimal within the Norway Caledonides. The glacio-isostatic component appears to be appreciable in recently deglaciated coastal Greenland areas that exhibit marked seismicity. The modifying effect of glaciation and deglaciation on recent vertical movements is also corroborated by the aseismicity of inner areas of Greenland and Antarctica, where the recent tectonic uplift is slowed by thick shields of recent glaciation.

In conclusion, a factor second by its seismogenic role in the region considered is the recent vertical movements rather than the glacial rebound. Moreover, their glacio-isostatic component is currently of minor importance.

Finally, the most efficient local factor of recent seis­micity is the load from extremely thick sedimentary cover, applied to the thin oceanic crust in the zone of its junction with the continental crust. Stein et al. [35] esti­mated that a sedimentary layer up to 10 km thick can generate stresses of 100 MPa and more, which is an order of magnitude higher than intraplate stress drops related to deglaciation. In principle, such stresses are sufficient to provide the highest seismicity along the majority of Arctic margins characterized by the devel­opment of thick deposits that cover also offshore sides of the margins. This divergence between theoretical estimates and observed seismicity is related to actual properties of the stressed lithosphere: unlike an elastic layer capable of supporting huge stresses, it fractures well before maximum possible loads are reached. Vis­coelastic and brittle/ductile models have been sug­gested as an alternative to the elastic medium [35]. The former has zero strength to long-term loads, i.e., pro­vides for a higher rate of aseismic relaxation of stresses as compared with their growth and fairly high strength (up to 70-80 % of the strength of elastic layer) at fast growth of stresses. Obviously, in this case generation of excessive stresses is mostly controlled by the thickness growth rate of the sedimentary cover rather than by the thickness itself; i.e., the intensity of recent sedimenta­tion is a main controlling factor. The possibility of earthquake occurrence is admitted to be real at sedi­mentation rates as low as 1:5 mm/yr, which are charac­teristic, for example, of the Gulf of Mexico [36]. Within the margins of the Norway-Greenland basin and Eur­asian subbasin, an appreciable contribution of this fac­tor into seismicity enhancement may be reasonably supposed for the Lofoten basin, particularly, in the zone of its junction with the Norway and Barents Sea (Senja fault) shelves, where the sedimentation rate is found to grow sharply to 1.4 mm/yr in the Pleistocene and to 3­7 mm/yr over the last 100 thousand years [37]. How­ever, general considerations on material properties and data on relatively low seismicity of inner parts of oce­anic basins make the second model preferable. In the upper part, it is characterized by the pressure-induced growth of brittle strength with depth. However, due to high temperatures, the material behavior in the lower part is controlled by the yield strength decreasing with depth. Evidently, such a medium precludes accumula­tion of high stresses regardless of their growth rate, which agrees well with the low seismicity level. The seismogenic effect of the overburden load is also sup­posed for the Lincoln Sea [37, 38].

In conclusion, it may be stated that none of the aforementioned factors, taken alone, can provide a high level of seismicity. At least two of them are required for the observed seismic effect, with the presence of stresses generated in interplate zones being a necessary condition. Moreover, taking into account possible fluc­tuations of absolute values of accumulated stresses, associated with inhomogeneity of the lithosphere, the relative contribution of each of the factors may vary from region to region. An appropriate (in phase) super­position of all of the three seismogenic factors may give rise to the strongest earthquakes. This final conclusion cannot and should not be extended to other zones of intraplate seismicity. Thus, high seismicity of Arctic Canada [39] and New Madrid seismic zone [40] is a result of combined action of regional compressive stresses generated by tectonic processes in the nearest interplate zones and local stresses in the incipient rift­ing zone, superimposed on ancient weakened areas.

REFERENCES

1. Turchaninov, L.A. and Markov, G.A., Neotectonic Impli­cations for the Stress State of Rocks in the Khibiny Apa­tite Mines, Izv. Akad. Nauk SSSR, Fiz. Zemli, 1966, no. 8, pp. 83-86.

2. Bungum, H., Earthquake Occurrence and Seismotecton­ics in Norway and Surrounding Areas, in Earthquakes at North-Atlantic Passive Margins: Neotectonics and Post­glacial Rebound, 1989, pp. 501-519.

3. Glaub, B., Marquart. G., and Fuchs, K., Stress Orienta­tions in the Northern Sea and Fennoscandia, a Compari­son to the Central European Stress Field, in Earthquakes at North-Atlantic Passive Margins: Neotectonics and Postglacial Rebound, 1989, pp. 277-287.

4. Slunga, R.S., Focal Mechanisms and Crustal Stresses in the Baltic Shield, in Earthquakes at North-Atlantic Pas­sive Margins: Neotectonics and Postglacial Rebound, 1989, pp. 261-276.

5. Wahlstrom, R., Seismodynamics and Postglacial Fault­ing in the Baltic Shield, in Earthquakes at North-Atlan­tic Passive Margins: Neotectonics and Postglacial Rebound, 1989, pp. 467-482.

6. Avetisov, G.P. and Golubkov, V.S., Seismotectonic Zon­ing of the Eurasian Basin, Arctic Ocean, and Adjacent Water Areas, in Geologiya i poleznye iskopaemye severa Sibirskoi platformy (Geology and Mineral Resources of the Northern Siberian Platform), Leningrad, 1971, pp. 66-73.

7. Avetisov, G.P., Seismic Zoning of the Franz Ioseph Land Archipelago, in Geofizicheskie metody razvedki v Arktike (Geophysical Exploration Methods in Arctic), Len­ingrad, 1971, issue 6, pp. 128-134.

8. Assinovskaya, B.A., Seismichnost' Barentseva morya) (Seismicity of the Barents Sea), Moscow: Izd. Ross. Akad. Nauk, 1994.

9. Avetisov, G.P., Seismicity of the Laptev Sea and Its Rela­tion to the Seismicity of the Eurasian Basin, in Tektonika Arktiki (Tectonics of Arctic), Leningrad, 1975, issue 1, pp. 31-36.

10. Avetisov, G.P., Some Problems of Geodynamics of the Laptev Sea Lithosphere, Fiz. Zemli, 1993, no. 5, pp. 28-­38.

11. Savostin, L.A. and Drachev, S.S., Cenozoic Compres­sion in the New Siberian Islands Area and Its Implica­tions for the Eurasian Subbasin Opening, Okeanologiya (Moscow), 1988, vol. 28, no. 5, pp. 775-782.

12. Sykes, L.R. and Sbar, M.L., Intraplate Earthquakes, Lithospheric Stresses and the Driving Mechanism of Plate Tectonics, Nature, 1973, vol. 245, pp. 298-302.

13. Bungum, H., Mitchell, B., and Kristofferson, Y., Con­centrated Earthquakes Zones in Svalbard, Tectonophysics, 1982, vol. 82, pp. 178-188.

14. Chan, W.W. and Mitchell, B.J., Intraplate Earthquakes in Northern Svalbard, Tectonophysics, 1985, vol. 114, pp. 181-191.  

15. Mitchell, B.J., Bungum, H., et al., Seismicity and Present-day Tectonics of the Svalbard Region, Geophys. J. Intern., 1990, vol. 102, pp. 139-149.

16. Tsybulya, L.A,        Levashkevich, V.G., and Kremenetskaya, E.O., Heat Flow and Seismicity of the Barents Sea-White Sea Region, in Geotermiya seismichnykh i aseismichnykh zon (Geothermal Study of Seismic and Aseismic Zones), Moscow, 1993, pp. 27­-32.

17. Avetisov, G.P., Seismic Zones in Arctic: Hypocenters, Focal Mechanisms, and Dynamics of the Lithosphere, Doctoral (Geol. and Mineralogy) Dissertation, St. Petersburg, 1995, p. 46.

18. Skordas, E., Meyer, K., Olsson, R., and Kulhanek, O., Causality between Interplate (North Atlantic) and Intra­plate (Fennoscandia) Seismicities, Tectonophysics, 1991, vol. 185, pp. 295-307.

20. Kim, W.Y., Kulhanek, O., et al., The Solberg, Sweden, earthquake of September 29, 1985, Seismol. Dept. Upp­sala. Report, 1985, vol. 85;no. I.

21. Spravochnik fizicheskikh konstant gornykh porod (Hand­book of Physical Constants of Rocks), Moscow: Mir, 1969.

22. Kulhanck, O. et al., The Otterbacken, Sweden, Earth­quake of February 13, 1981, Seismol. Sect. Uppsala. Technical Report, 1981.

23. Arvidsson, R., Gregersen. S., et al., Recent Kattegat Earthquakes-Evidence of Active Intraplate Tectonics in Southern Scandinavia, Phys. Earth Planet. Inter., 1991, vol. 67, pp. 275-287.

24. Avetisov, G.P. and Vinnik, A.A, Arctic Seismological Data Bank, Fiz. Zemli, 1995, no. 3, pp. 78-83.

25. Semevskii, D.Y., Neotectonics of the Spitsbergen Archi­pelago, in Materialy po stratigrafii Shpitsbergena (Stratigraphy of Spitsbergen), Leningrad, 1967, pp. 225­238.

26. Grigor'ev, M.N. and Musatov, E.E., On the Problem of Neotectonic Movements in Western Arctic, in Strati­grafiya i paleogeografiya pozdnego kainozoya Arktiki (Late Cenozoic Stratigraphy and Paleogeography of Arctic), Leningrad: PGO Sevmorgeologiya, 1982, pp. 27-36.

 27. Kovaleva, G.A., Oolubkov, VS., and Gusev, VV, Recent movements on the Alexandra Land Island, Franz Ioseph Land, in Geotektonicheskie predposylki k poiskam poleznykh iskopaemykh na shel'fe Sevemogo Ledovi­togo okeana (Geotectonic Implications for Mineral Pros­pecting on the Arctic Ocean Shelf), Leningrad, 1974, pp. 87-92.

28. Gutenberg, G., Changes in Sea Level, Postglacial Uplift and Mobility of the Earth's Interior, Geol. Soc. Am. Bull., 1941, vol. 52, pp. 721-772.

29. Nikolaev, N.l., Noveishaya tektonika i geodinamika lito­sfery (Neotectonics and Geodynamics of the Lithos­phere), Moscow: Nedra, 1988.

30. Andrews, I.T., Present and Postglacial Rates of Uplift for Glaciated Northern and Eastern North America Derived from Postglacial Uplift Curves, Can. J. Earth Sci., 1970, vol. 7, no. 2, pp. 703-715.

31. Grachev, A.F. and Dolukhanov, P.M., Postglacial Uplift of the Canada and Fennoscandia Crust from Absolute Datings, Baltica, Vilnius, 1970, vol. 4, pp. 297-312.

32. Morner, N.-A., Earth Movements in Sweden, 20000 BR to 20 000 AP, Geologiska Foreningens i Stockholm Forhandlinga,; 1978, vol. 100, part 3, pp. 279-285.

33. Quaternary Geology of Canada and Greenland, Fulton, R.I., Ed., in Geology of Canada, 1989, vol. 1.

34. Stille, H., Recent Deformation of the Earth's Crust in the Light of those of Earlier Epochs, Geol. Soc. Am. Spec. Papers, 1955, vol. 62.

35. Stein, S., Cloetingh, S., et aI., Passive Margin Earth­quakes, Stresses and Rheology, in Earthquakes at North­Atlantic Passive Margins: Neotectonics and Postglacial Rebound. 1989, pp. 231-259.

36. Nunn, J.A., State of Stress in the Northern Gulf Coast, Geology, 1985, vol. 13, pp. 429-432.

37. The Arctic Ocean Region, Grantz, A., Iohnson, L., and Sweeney, I.F., Eds., in The Geology of North America, 1990, vol. 1.

38. Basham, P.W., Forsyth, D.A., and Wetmiller, R.I., The Seismicity of Northern Canada, Can. J. Earth Sci., 1977, vol. 14, pp. 1646-1667.

39. Avetisov, G.P., Seismotectonics of Arctic Canada, Fiz. Zemli, 1995, no. 5, pp. 8-20.

40. Grachev, A.F., On the Nature of the New Madrid High Seismicity Zone, North American Platform, Fiz. Zemli, 1994, no. 12, pp. 12-23.