SUMMARY

INTRODUCTION

Presently, scientists, politicians and mass media discuss diverse problems of geo-ecolo­gy, a discipline studying relationships of human societies and the environment.

The book demonstrates that both individual human communities, and the society as a whole behave as complex dissipative systems (as understood by I. Prigogine [Пригожин, Стенгерс, 1986]) of interaction between not only social, political, economic, and technolog­ical developments, but also endogenous and climatic environmental processes. A systematic approach to the estimation of the role natural processes play in the development of a society may, in our view, help to avoid lapsing into «geodynamic determinism».

We have limited our studies of influence of natural events on individuals and society both in time (the final stage of human history characterized by rise and evolution of productive economy, and civilized societies) and in space (history of societies in the central part of the Alpine-Himalayan orogenic belt and its surroundings). In the ancient time, the region we con­sider was known as Eastern Oykumena; it gave birth to the earliest productive economies and first civilizations. It encompassed a territory stretching from Egypt, Greece and Black Sea coasts in the west up to India and Central Asia in the east.

The goals and tasks of our studies determined the need for an integrated analysis and cor­relation of data of Quaternary geology, neotectonics, seismology, volcanology, climatology, hydrology, as well as history, archaeology, biology, and medicine. Such a multi-disciplinary approach requires some terminological explanations. «Active tectonics» and «active fault» are meant to understand tectonic manifestations that take place now and are expected in near future. For the studied region, we use these terms referring to tectonic events of the Late Pleistocene and Holocene, i.e., the last 10(0-150 thousand years [Трифонов, 1983; Trifonov, Machette, 1993]. To estimate recent activity in stable regions, like the East European plat­form, we also ought to take into account Middle Pleistocene events (within the last 700 thou­sand years).

The meaning of “civilization” in this work is two-fold. In a general sense, we use it to define a society development stage opposed to the preceding stage of primitive savagery. A civilization is based on a productive economy that precedes all other civilization attributes and represents a pre-requisite for the latter. In a narrower sense, civilization is understood as «a socio-cultural community formed on the basis of universal, i.e., supra-local values reflect­ed in global religions, morality systems, law, and art» [Сравнительное изучение..., 1999, p. 25]. This understanding means that civilizations can, in particular, differ in social relations. Although today’s globalization of economy and infrastructure draws civilizations together, differences between them remain.

The Holocene is assumed to last for 10 thousands years. It is subdivided into Early (Boreal), Middle (Atlantic and Sub-Boreal), and Late (Sub-Atlantic) Holocene with bound­aries set respectively at 8-7,7 ka BP (early 6^th millennium BC), 5-4.6 ka BP (the first half of the 3^rd millennium BC), and 2,5 ka BP (middle of the 1^st millennium ВС) [Нейштадт, 1985].

In this book, we mostly use age estimates derived by historical (preferably, «high chronology» approach), archaeological and radiocarbon dating methods. Estimating the age of soil horizons by the radiocarbon method, we consider the time span of their formation with due regard to the fact that top soil layer contemporary with a natural, or technological impact is.randomly preserved in the section. As the soil layer preserving on top can be hundreds or even thousand years older than the considered impact, it can be used merely to constrain the earliest possible age of the latter.

Chapters 1 to 4 (Part I) examine influence of individual natural processes on social deve­lopment in the following order: Climatic Changes and Associated Oceanic Level Variations, Active Tectonics, Strong Seismicity, and Volcanism. Chapters 5-8 (Part II) devote to interac­tion between natural and social processes. Principles of such interaction are discussed in Chapter 5. In Chapter 6, we base on our studies of the Armenian Upland testsite to describe a relationship between natural and social processes in detail. Chapter 7 presents some results from the studies of specific nature-society complex systems. Sources and social repercussions of the Late Holocene fluctuations of the Caspian Sea and Black Sea levels are discussed. Correlation between changes of climate and geodynamic activity and the largest social-po­litical crises during the Middle and Late Holocene is carried out. Grounds are provided in sup­port of the orbital-astronomic regulation of the synchronous climatic and seismotectonic vari­ations with a frequency from few years up to tens of thousand years. Chapter 8 addresses general regularities of humanity-nature interaction, problems of sustainable development considering effects of geodynamics and natural changes, and implications these regularities have had in the history and will have for the future of Russia. The supporting catalogues of new radiocarbon age estimates and strong earthquakes in the Eastern Oykymena are present­ed in Annexes 1 and 2.

The main part of this book was written by V.G. Trifonov and A.S. Karakhanian. E.R. Sen’ko and T.P. Ivanova had a leading role in writing Chapter 3 and Section 7,1, respec­tively. E.R. Sen’ko and V.N. Balassanian helped to compile the Catalogue of Strong Earthquakes in the Region. A. Avagyan, R. Agamirzoev, J. Adjemian, O. Azizbekian,

D. M. Bachmanov, A. Bagdasaryan, M.S. Bayraktutan, A. Chatzipetros, Yu. El-Hair, P.V. Florensky, F. Jamaly, M. Khademi, Kh. Hessami, D. Khondkarian, M.L. Kopp, A.I. Kozhurin, V.K. Kuchay, V.I. Makarov, I. Mariolakos, S. Pavlides, E.A. Rogozhin, T. Zaza, L.M. Rastsvetaev, S.F. Skobelev, and E. Vittori participated in the field works that provided findings used in this book. Field trips to Israel and China were organized by I. Carch and A. Mart, and Prof. Ding Guoyu, respectively. The radiocarbon dating of collected sam­ples was conducted by L.D. Sulerzhitsky (Geological Institute of the Russian Academy of Sciences). Kh.A. Arslanov conducted U-Th dating of Pleistocene shells from. the Zagros, and experts from the Museum of Ancient History in Tehran estimated the age of samples by the thermoluminescent method. P. Avetissian, R. Badaljan and A. Pilipossyan (Armenia) and Kh. Salibi (Syria) characterized the collected samples of ceramics. E.M. Vangengheim iden­tified the collected fossil mammalia and their images in petroglyphs. Consultations and rec­ommendations of Yu.G. Leonov, V.M. Masson, E.E. Milanovsky, S.N. Roerich, D.V. Rundquist, A.L. Yanshin, N.N. Vorontsov, E.A. Lyapunova, R.T. Jrbashyan, A.E. Dodonov, V.I. Zhegallo, K.Kh. Kushnareva, Yu.A. Lavrushin, D.V. Lopatin, K.S. Losev, A.A. Nikonov, S. Stiros, S.K. Tatevian, V.I. Ulomov, V.N. Kholodov, A.L. Chepalyga, N.M. Chumakov, and S.S. Shults, Jr. were extremely useful. E. Abgaryan helped to translate the Summary and S. Arakelian, D.M. Bachmanov, S.V. Oskolkova and R.V. Trifonov helped to illustrate the book. The authors are grateful to all these persons.

Parti

NATURAL PROCESSES INFLUENCING DEVELOPMENT OF CIVILIZATIONS

Chapter 1

CLIMATIC CHANGES AND ASSOCIATED OCEAN LEVEL VARIATIONS

1.1.

The last glaciation in Europe (often referred to as Wurm, Valday, or Visla glaciation), was preceded by the Mikulino interglacial period that started about 130 ka BP. By that time, the average January temperature in Central Russia was higher and precipitation was more abundant than now. The previous deglaciation caused transgressions in the north (Boreal) and in the south (Karangat in the Azov-Black Sea basin and Khazar in the Caspian). The Early and Late Valday glacial epochs and the Middle Valday interstadial are differentiated within the Valday glaciation period. The Early Valday lasted from 90-70 ka till 47—8 ka BP, the Middle Valday from 47-48 ka till 24 ka BP, and the Late Valday from 24 ka till 10 ka BP. The Late Valday glaciation was the most inclement time for the whole Pleistocene. The glaciation had two centers: Scandinavian and Novaya Zemlya. For a short period (no more than 4 ka) at the peak of glaciation, the glacial tongue was reaching the Valday Height. Study of a layer with a ¹⁴C-measured age of 17460 ± 10 yr in the Western Dvina River yielded esti­mates of average January and July temperatures that are respectively 10-14 °C and 2-3 °C lower than the respective values today. The margin of permafrost lowered to 49-50° N [Раз­витие..., 1993].

The peak of the Late Valday glaciation was immediately followed by deglaciation inter­rupted with episodes of glacier advance. The Holocene started with warming that was more pronounced in the south-west of the Russian Plain [Klimanov, 1995; Изменение..., 1999]. The last half of Atlantics, referred to as the Atlantic Optimum (4—3 ka BC), was the warmest time in Eastern Europe and other regions of Northern Eurasia. The start of the Sub-Boreal period was marked by considerable cooling and drying of the climate, followed by the next warming. At that stage, temperatures were higher than those observed presently, but lower than those of the Atlantic Optimum.

1.2. Climatii chhtges in the EEsieer Oykumenn and surrounding areas during the Late Pleistocene and the Holocene

1.2.1. Plains in Kazakhstan and the Middle Asia. There was general synchronism of climate changes in this region and in Eastern Europe during the Late Pleistocene and the Holocene. Palynology of the Mikulino interglacial in the northwest of Kazakhstan is attrib­uted to a forest-steppe type. During that pluvial time, the Amu-Darya River fell into the Caspian Sea and later formed three deltas in the Southern Aral area. The Zeravshan River

reached the Amy-Darya. The Tedjen and Murgab deltas were located northwards of the their recent location and from time to time these rivers became tributaries of the Amy-Darya. Dry vegetation of steppe and cryogenic structures were typical for the main stage of the last glaciation (20-18 ka BP) in the northwestern Kazakhstan. Climate in the Middle Asia became dry and cool. Supply of water to the southern and eastern Aral region decreased. In the east­ern Caspian region, the average annual temperature was 4,5 °C lower than today [Разви­тие..., 1993].

Humidification of climate in the Middle Asia started in the end of the Late Pleistocene. The Atlantic Optimum manifested itself in general humidification, which, at least in places, was accompanied by slight decrease of summer temperatures and appearance of hygrophilous and arboreous plants, including broad-leaved species. Drying of die climate in die first half of the 2^ndmillennium BC changed to certain humidification in the end of it. The subsequent aridization was interrupted by epochs of relative humidification in the second half of the 1^st millennium BC and in the 10*-15* centuries AD in particular. The latter could be accom­panied by certain decrease of temperatures (2-2,5 °C of the annual average values) in the l3*-14* centuries. The changes of climate settings through the Holocene were reflected in the history of the Amu-Darya-Aral-Sarykamysh-Uzboi-Caspian system. The level of Aral was always higher than the level of the Caspian. The flow of Amu-Darya through Sarighamish to the Caspian began not until a +55 m level was reached (the highest point of Uzboi near Mount Kughenek). By the changes of flow in the Middle and Late Holocene his­tory of this system, it is possible to identify phases of humidification (Atlantic Optimum, the second half of the 1st millennium BC- early 1^st millennium AD; 11* - early 16* centuries) and aridization (the first half of the 2nd millennium and, perhaps, 8th-7* centuries BC, 3rd_10* centuries AD.) In the meantime, within some periods the Amu-Darya River dis­charged the largest part of its water either to Aral, or to Sarychamysh and Uzboi. Most prob­ably, this was determined by tectonic events (see Chapter 5.)

1.2.2. Mountainn in ns entral Aria and the nohthweshern Hindustan. The eariy strge of the Late Pleistocene was characterized by warm climate. Manifestations of the earlier Late Pleistocene glaciation are represented in the Pamirs and Tien Shan. During the epoch, corre­sponding to the Middle Va^ay (Wurm), warm and humid climate was predominant in Central Asian mountains and in adjacent parts of Hindustan. The last Late Pleistocene glacia­tion, corresponding (perhaps, with a delay in some areas) to the Late Valday glaciation, was accompanied by cooling and general arization. The latter was caused by the Central Asian baric maximum and reduction of influence of the Indian Ocean monsoons because of the tec­tonic uplift of the mountains, particularly the Himalayas and Tibet. The southern slopes of the Himalayas and the northwestern Hindustan were not taken up by cooling, but aridization was significant there.

Warming in Boreal, and, in places, pre-Boreal, occurred against preservation of relative­ly dry conditions. The humidification of the climate covered the Outer Himalayas and the northwestern Hindustan. Almost everywhere, manifestations of the climatic optimum includ­ed increase of either temperatures, or humidity, or both. The north of Tibet was the only cold and dry area. Warming and humidification in most of the mountainous systems lasted from the second half of the 6th till the middle of the 2nd millennium BC with its peak around the 3rd millennium BC. The warming then changed to cooling and hridizhtion. In the Indus Valley and in the NW of Hindustan, climatic optimum of the 6* - early 2nd millennia BC was later followed by arization about 1800 BC.

In the mountains, glaciers appeared and grew again in the Sub-Atlantic period, being, nevertheless, considerably inferior to the glaciers of the Late Pleistocene. Peaks of glaciation corresponded approximately to a period about 1000 BC and the 17^t^-19* centuries AD. In Northern India, relative improvement of climate conditions against the background of incip­ient arization took place late in the 1^stmillennium BC and in the first half of the 1^s* millen­nium AD; the conditions worsened in the second half of the 1^stmillennium AD and improved relatively in the first half of the 2nd millennium (the Sub-Atlantic Optimum, or Vikings’

Epoch). The climate again became worse in the 16*-19* centuries, which corresponds to the Lesser Scandinavian Glaciation.

1.2.3. The Caucasus and Iran. In the Caucasus, the Late Pleistocene began with a warm and humid interglacial period. This was followed with a two-stage glaciation in the Caucasus and other high mountains, setting in cold and arid climate [Развитие..., 1993; Саядян, 1985]. Certain humidification was recorded for the interstadial. In a periglacial area, drastic aridiza- tion in the second glacial phase manifested itself everywhere, including the Red Sea, while cooling of that period was more pronounced in latitudes farther to the north. Lake Van was under considerable transgression, which, however, could be caused by tectonic motions and volcanism.

Deglaciation and associated gradual warming and humidification started in different places at different time [Мурзаева, 1991; Развитие..., 1993; Ван Зейст, Боттема, 1985; Саядян, 1985; Taviani, 1995]. In the Eastern Tran-Caucasus and Northwestern Zagros, these changes became evident as early as 14 ka BP, when the Persian Gulf and the Red Sea still had the dry climate of the glacial epoch. The warming continued in the Boreal, accompanied by humidification everywhere except of the Black Sea slope of the Northwestern Caucasus and, possibly, Lake Van region. Warm and humid conditions continued to exist during the Atlantic period, and in places became even more pronounced. Transgressions of the Black Sea and the Caspian Sea took place that time. Favorable conditions in some areas (Northwestern Caucasus, Northwestern Zagros, Urmiye Lake and Persian Gulf regions) lasted till the end of the Atlantic, but in other areas they changed in the second half of the period. Early in the Sub­Boreal, temperatures and precipitation start to decline slightly in the Northwestern Caucasus and in the Van region, although in the latter the climate kept to be highly humid due to reduced evaporation loss. In the same period, glaciers in the Greater Caucasus moved. The climate in the Persian Gulf region became more arid in the Sub-Boreal, while in the Sub­Atlantic it humidified slightly. Climatic conditions approached the today’s pattem as early as in the Sub-Boreal, or later in this region. However, there were episodes of relative cooling and humidification in the 10* century BC (Sevan, Van and Sistan), 1^sl-2^nd centuries AD (Sevan, Sistan) and 16*-19* centuries AD (the Gagry Ridge and Sevan), which were almost coincident with the time of glacier motions in the Caucasus and partly in Taurus.

1.2.4. The Eastern Mediterranean Region. Climatic changes in the last glacial and post-glacial epochs were complex and varied from area to area. The epoch of Late Valday glaciation was characterized by considerable aridity almost everywhere. Relatively pluvial conditions set in since 18 ka BP in Israel only [Horowitz, 1979, 1987]. Detailed palynologi­cal studies and radiocarbon estimations of ages for this area indicate [Leroi-Gourhan, Darmon, 1987] that initially surges of humidification were, however, comparatively minor, but reached a more noticeable rate around 14,5-11 ka BP and particularly within 10,3-9,5 ka BP (Fig. 1). A similar pattern, although with specific deviations, can be inferred for other places as well, although the scales and start times of the humidification were different. Relatively low in the southern Anatolia (10-8 ka BP), the rate of humidification was higher in Lebanon and Western Syria (since 11 ka and particularly 10 ka BP) and in Central and Northern Anatolia (since 13 and particularly 10 ka BP). In Northern Africa, the humidifica­tion began 10 ka BP and reached its maximum 8 ka BP [Petit-Maire, 1992].

Later changes developed also differently. In Northern Africa, humidity peaks corre­sponded to 9-7 ka and 5,9-4,9 ka BP, and were followed with aridisation. In Israel, in con­trast, most part of the Early and Middle Holocene was arid. There were a few surges of sub­sequent humidification related to certain cooling about 5 ka, 3,5-3 ka, and about 2 ka BP, again followed by aridization [Мурзаева, 1991; Issar, 1996]. Aridization in the Early Holocene was less pronounced in Lebanon and Western Syria, since climate conditions were milder there. As a result of extended period of the Early-to-Middle Holocene aridity in Southern Anatolia, present-day humid conditions had set in there only about 3 ka BP, while in an area northward, the' humidification reached its peak at about 7 ka BP with aridization signs showing up 5 ka BP.

1.3. Climate and development ofthe earliest agriculture in hhv Enthved Oykrmvdt

1.3.1. Gansert evmnekt. Synchronism between the spread of Late Paleolithic, Mesolithic and Neolithic cultures and favorable climate conditions, and, on the contrary, between culture degradation up to settlement interruption signs and unfavorable climate, has been established for various parts of Eastern Oykumena. The origination of agriculture and domestication of animals were critical steps in the human history, often defined as Neolithic Revolution. This important event is related to the very beginning of the Holocene and to the area of so-called «Fertile Crescent» that represents a north-convex arc bounding the Arabian plate and including Israel, Lebanon, the western Syria, the southeastern Turkey, the northern Iraq, and the western Iran (Fig. 2). A.J. Toynbee [Тойнби, 1991: Russian version of «А Study of History», 1934-1961] developed the challenge-and-response concept, which suggests that the change to agriculture was a response of early hunters and collectors to the abrupt hrinizh- tion concurrently with melting of Late Pleistocene glaciers. Our opinion is that the situation was right the opposite: agriculture originated during relatively humid phases. We will try to prove this.

1.3.2. Prtvteinv, Lvbrnpn, rnd Syeir. In Israel and adjacent part of Jordan, the Mesolithic Kebarien was changed by Natoufien, a Late Mesolithic culture, in late 1 l^th-early 10^ttl millennia BC. Settlements of this culture were excavated in Jerico, Einan (Ain Mellaha) and Beida. The artifacts attest to intense collecting of wild eatable plants and, per­haps, to the first, but never developed attempts of their reproduction. The beginning of Natoufien corresponded to moderate humidification (Fig. 1), which was later followed by aridizatioe, a possible cause of stagnation in the Natoufien culture. A major step towards agri­culture was made during the following epoch of the pre-ceramic Neolithic, along with sig­nificant humidification in the second half of the 9^ttl and first half of the 8^til millennia BC. Wheat and barley were cultivated side by side with collecting wild plants and hunting [Мел­ларт, 1982] (Table 1). Jerico became a big settlement (Fig. 3 and 4). In Lebanon and the northwestern Syria, cultivation of cereals started in the 9^1|1-8^Лmillennia BC, again concur­rently with humidification (Fig. 5).

1.3.3. In Sprueven Anrhptir, the climate of Early Holocene was relatively dry, but prob­ably more humid than in the Late Pleistocene. Evidence of the onset of agriculture was found in Layer V in Hajilar (the 8^й1millennium ВС) [Мелларт, 1982]. Moderate humidity and water streams in the Konya Valley enabled agriculture without irrigation, and a large settle­ment of Chatd-HuyuR flourished there in the second half of the 8^tl' - first half of the 7^ttl mil­lennia ВС [Мелларт, 1982] (Fig. 6). The economy of the settlement was based on versatile agricultural activities, cattle-breeding and hunting.

1.3.4. In the Innve Zrgept, humidification started in the 10^thmillennium BC. The earli­est signs of sheep domestication were found in the lower layer of the settlement of Zavi- Chemi (about 11 ka BC) in the north-west part of the region. Evidence of intense collecting of wild plants and, possibly, of agricultural activity was found in the higher layers of Zavi- Chemi and in the adjacent Shankar cave. More evident signs of agriculture and domestica­tion of sheep and dog were found in the lower layers of the settlement of Jarmo dated back to early 7th millennium BC. In Gandj-Dere, a settlement located to the southeast, evidences of wheat and barley cultivation were dated back approximately to the boundary of the 9* and 8th millennia BC. Ceramics of the oldest age was found in a layer located a little up from this [Антонова, 1982]. In Huzistan, a period of non-irrigative agriculture and domestication of goat continued till the middle of the 6* millennium BC and was changed later by a period of irrigative agriculture and domestication of cow.

1.3.5. Plains and low hills of the NpeUhven Mvtppphrmir remained dry longer than the rest of considered areas. The region included some areas with temporary streams. An agri­cultural settlement of Jebel Magsalia (8th-7^m millennia BC) situated in one of such areas was, probably, founded by migrants from a more humid region. Manifestations of irrigative agri­culture are attributed to the Samarra culture epoch (6^ttl millennium BC).

All these data indicate that humidification, often associated with certain warming, was one of the most important factors for generation of agriculture in the Fertile Crescent (Fig. 5). Agricultural technology originally developed at foot-hills and within intra-mountain basins, where no irrigation was required for agriculture. At a later stage, agricultural evolution lead to the discovery of irrigation; irrigative system spread into Mesopotamia, where agriculture had not been productive without irrigation.

1.3.6. Eariy agricullure in the Eastern Oykumena beyong the «Feetiie

Among other directions, early agriculture spread out of the Fertile Crescent to Susianna and Southern Mesopotamia. Irrigative agriculture and cattle-breeding arrived to this area in the second half of the 6th millennium BC. The early agricultural economy reached its peak in the Obeid culture, which developed from the second half of the 5th millennium BC, laying foun­dation for later rise of the first Sumerian city-states.

Foci of the earliest agriculture reached areas to the east of Zagros (Syalk) and spread fur­ther to the east along the Alborz into the Southern Turkmenia, where warm and rather humid climate settled early in the Atlantic period (7^ft millennium BC) along with superposi­tion of the western cultural influence and domestic traditions was quite beneficial. The agri­cultural Jeton culture formed here in the 6^m millennium BC across a foothill plain, bounded from the southwest by the Main Copet Dagh fault zone [Массон, 1971]. The settlements were located near streams, which were then full-flowing all the year round. According to N.I. Vavilov [Вавилов, 1965], wild ancestors of cultivated wheat belonged to the Iranian- Afghan group. Meat food was provided by hunting and cattle-breeding, and the share of the latter was increasing in time. The earliest signs of domestication of goats in the southeastern Caspian region were found in the Mesolithic layers of the 7th millennium BC. Mostly, or entirely, cattle-breeding was the basis for transition to productive economy in other regions of the Middle Asia [Массон, 1971].

The first signs of agriculture and cattle breeding in the Trans-Caucasus were recorded in the Shulaveri-Shomutepeh culture, belonging to the ceramic Neolithic (the second half of the 7* - early 5th millennia ВС) [Кушнарева, 1993]. Absence of signs of any transitional Mesolithic, or Early Neolithic cultures may indicate that a period of rather cool and dry cli­mate retarded the advent of productive economy, which formed not earlier than in the Atlantic optimum under the influence of the Fertile Crescent cultures.

The rise of agricultural and cattle-breeding cultures in the southeastern Europe is dated back to the 6^th, or, perhaps, late 7th millennium BC. The earliest evidence found in Macedonia and Thessaly is attributed to the Pre-ceramic period, and attests to cultural influence, or even direct migration of the Anatolian population into the region [Мелларт, 1982]. An important fact is that wild ancestors of small cattle and some species of cultivated cereals have not been found in Greece, and stem from Anatolia. Macedonian cultural achievements spread both to the south (Thessaly and Central Greece) and to the north, into Bulgaria and further up to the Carpathians. In a short term, transition to the productive economy spread over a large forest­steppe area of the Middle and Eastern Europe, characterized by the culture of linear-band ceramics. The spread of agricultural practice is proved by finds of seeds of soft wheat, spelt, and peas; bones of domestic cows, pigs, sheep, and goats have been found, too. Quick tran­sition to the new economy system may indicate that aborigines had been ready to adopt it and that climatic conditions of the Atlantic Optimum also favored the transition.

Manifestations of the culture of linear-band ceramics have been found in the Northern Black Sea region up to the valleys of the Prut River and the Dnieper River in the east. The Bug-Dniester culture with signs of agriculture formed side by side to it in the 6th millennium ВС [Даниленко, 1969]. Likeness of the ceramics proves the Macedonian and finally Anatolian influence. Presence of wild bull and boar inhabiting the Northern Black Sea region conditioned the important role of cattle breeding [Массон, 1971]. Pig in Crimea and cow in the Lower Dnieper area were domesticated as early as in the pre-ceramic period. Cattle’s share in the meat food was increasing throughout the period and become particularly large in the Ceramic Neolithic epoch (the 5* millennium BC). Cattle-breeding became the economic

base of the Sura-Dnieper culture. Tribes of the Dnieper-Donets culture, which inhabited the forest-steppe and forest zones of the Upper Dnieper area and Belarus mastered cattle-breed­ing as well.

Domestication of horse was an important step in the development of cattle-breeding in the Northern Black Sea region. Evidences indicating presence of domestic horses were found on the right bank of the Dnieper River, in the settlement of Dereivka dated to the second half of the 4^tfl millennium ВС [Бибикова, 1967]. Horse breeding is associated with the Chalcolithic Srednestogovskaya culture of the second half of the 4* - early 3^rd millennium BC. It had formed on the basis of the Sura-Dnieper and Dnieper-Donets cultures (the 5* - first half of the 4th millennia BC), when the first attempts of horse domestication could take place. Anyway, the cult of a horse-head sceptre as a sign of power began to spread across the steppe between the Middle Danube and Volga Rivers since the second half of the 4* millen­nium BC; it manifested the origination of a cavalry [Клейн, 1990].

Most of ancient settlements in the valley of Nile, whose population intensively prac­ticed the collecting, are related to the 13^th-12t^h millennia ВС [Заблоцка, 1989]. However, they never turned to agriculture, which possibly could be explained by the lack of a neces­sary set of wild plant species and adequate climate conditions. During the last glaciation epoch, climate in Northern Africa was arid and rather cold. Humidification started in the 8t^h millennium BC; in the 7*-6^m millennia occasional rains changed to more regular water­ing. In favorable areas, population turned to intensive collecting, which facilitated and increased stability of settlements [Barich, 1995]. It is probable that even since then residents of those settlements domesticated sheep and made their first attempts to keep agriculture. However, definitive transition to a productive economy system was again retarded, quite likely due to certain aridization in the 5th millennium BC, which changed to a new humid­ification period late in the 5th and in the 4* millennia BC. Fayum A settlement is related right to this period, at the boundary of the 5^th and 4th millennia BC. The population was engaged in pig, goat, sheep, and, may be, cattle breeding, hunting, and collecting, and start­ed to grow barley, wheat and flax, although 'that growing was still playing a secondary role. First stable settlements in Delta appeared about that time, or a little later, and were proba­bly also based on cattle breeding. The earliest evidence of productive economy in the Nile valley was found in the Badara culture, dated to the early 4th millennium ВС [Заблоцка, 1989]. It was rather an agricultural, than a cattle-breeding culture. That stage peaked at the Gherze culture (late 4th millennium BC) that spread all over Egypt. The progress in irriga­tive agriculture, trades, ideology and social relations the culture achieved, pre-netsrminsn creation of the Ancient Egyptian state.

Therefore, the productive economy of Egypt formed at a relatively late stage, and was probably a product of interaction and eventually mergence of cultures native of African regions located southerly that inhabited the Nile valley, on one hand, and semi-nomadic hunters, collectors, and early stock-breeders of the southern near-Mediterranean area, on the other hand. Plants and animals were, at least partly, adopted from Western Asia. Therefore, it is not improbable that immigrants from this area contributed to the formation of productive economy in Egypt.

1.3.7. Climatic affeuds m the taehs гіеи^те^ af dericnrtuue. tler^, wecD^ider juse a few typical examples (see detailed discussion of the problem in Section 7.3). After a gap in the Chalcolithic, which was coincident with an arid phase, an Earlier Bronze Age culture began to develop in Jerico since about 3200 BC, during the humid phase of 3,5-2,3 ka BC [Issar, 1996; Marchetti, Nigro, 1997a, b]. Later decay of this town, as well as of other Palestinian towns in 2,3-2,0 ka BC was concurrent to a phase of abrupt aridizatiee. A new epoch of activity is recorded in the Middle Bronze Age (2,0-1,55 ka BC) in parallel with humidification, the main phase of which is dated to 1,5-0,8 ka BC. Although it is possible to suggest that such humidification could provoke the Exodus of Jews from Egypt to Palestine, it was the time when Jerico lost its significance, which was not restored until the humid Hellenic-Roman period [Issar, 1996].

The Chalcolithic Cucuteni-Tripolie culture formed in the south of Ukraine, and in Moldova and Romania in the second half of the 5th millennium BC (the humid and warm Atlantic period). That was the peak of early agriculture and cattle breeding development in the region. Fast degradation of the Tripolie culture in the second half of the 3^πi millennium BC coincided with cooling early in the Sub-Boreal period. The considered climatic change influenced development of an agricultural Kura-Arax culture that formed in the Trans­Caucasus in the middle of the 4* millennium BC (Section 6.1). Since the second half of the 3rd millennium BC, the Kura-Araks population had concentrated in large river valleys, occu­pied high foot-hills and intra-mountain valleys, and migrated to the southwest (Eastern Anatolia, Syria, and Palestine) and to the southeast (the northwestern Iran). In the same peri­od, crisis-ridden Sumerian towns-states were conquered by the Western Semitic tribes led by Sargon the Akkadean (2371-2316 BC).

A humid period in the plain along the northeastern foot of Copet Dagh started in the 7th millennium BC and continued till early 2^mj millennium BC; correspondingly, the earliest agricultural culture of Jeitun covered an area spreading up to the Pra-Tedjen valley during the Chalcolithic and the Bronze Age. Later aridization dried up the Tedjen valley and determined degradation and final elimination of these cultures in the second half of the 2^mi millennium BC. The next bloom of agriculture in the Middle Asia in the antique epoch again corre­sponded to a humid phase. In the 6th century BC - 2^mj century AD, the area of irrigated lands in the Aral water system reached 3,5-4 million ha [Толстов, 1962; Андрианов, 1991]. Subsequent aridization since the end of the 1^s* millennium BC reached its peak in the 3 ^πi-5 th centuries AD and caused economic degradation, which facilitated the Arabic con­quest.

The next phase of humidification in the ll^th-15^th centuries BC was accompanied by a slight decrease of temperature and determined a new prime of agriculture, growth of existing towns and foundation of new ones. The area of irrigated lands was restored to its antique peri­od size [Клиге и др., 1998]. Such favorable climate facilitated even the fast recovery of destruction caused by the Mongolian conquest and Timur’s wars. Stagnation, taking in the Middle Asia states from the second half of the 16th and, particularly, from the 17th century AD, corresponded to the aridization during the Lesser Scandinavian glaciation epoch.

Humidification in the northwestern India and in adjacent areas within 3-1,8 ka BC pro­moted the rise of Harappa civilization, while subsequent aridization in many regions of Central, Eastern and Southern Asia caused degradation of the civilization, and led to its total elimination 2 to 3 centuries later. Arriving to the northwestern India, Aryans found only local relicts of the civilization near the oceanic coast and in areas farther to the south of the sub­continent. The next humid phase in the northern India settled in the last centuries BC ( early centuries AD and favored development of the states of the classical period in the Indian his­tory, namely, the kingdoms of Mauri, Sungas, Satavayana, Kushan, and Gupta; later aridiza­tion in the 5*-9* centuries was the time of considerable social and economic degradation.

1.4. Occss ієпєПcheages

1.4.1. The baengraund oU net levea vmriatians rn nhe Holeceoe. А aetween

the sea level rise in the late glacial and post-glacial time and intensity of thawed water influx into the ocean (Fig. 7) indicates that deglaciation was the main factor of the sea level rise. The total magnitude of the post-glacial rise is estimated at 120 ± 50 m, with most estimates falling between 100 and 130 m [Природные условия..., 1986; Селиванов, 1996]. The greatest peak of ocean rise rates coincided with the time of the most intensive deglaciation in Fennoscandia and North America (the middle of the 10^fll millennium BC), while the second peak was at about 7 ka BC (Fig. 7). Although at a decreasing rate, the rise continued till the middle of the 5* millennium BC and in a later period. The studies of the Japan Sea and Okhotsk Sea coasts indicate that during the Atlantic optimum the World ocean level could be 2-3 m higher than now [Изменение..., 1999]. This can be explained by quick destruction of

a part of the West Antarctic glaciers at 6-7 ka BP [Hughes, 1987]. The ocean level became more or less stable in the second half of the 4^ft millennium BC (Fig. 8) and its later variations were minor; they partly coincided with the epochs of warming and could result from changes of volumes of the Antarctic and Greenland glaciers.

1.4.2. Generation oi tne Semerlra civrlization and ОПіIegent rbout the Delvge ir upeestneiρn with Uhv pptertnuint eitv pS бшtvevt. The familiar biblical narration about the Deluge can be traced back to Akkadian sources, where its most comprehensive description is found in the Story about Atralchasis and in Gilgamesh Epics[Гильгемеш, 1919; Я открою тебе..., 1981]. These in tum stem from a preserved version of a Sumerian prototype [От на­чала начал, 1997] dated back to the rule of Khamourapi (1792-1750 BC), but representing, most probably, a more ancient source. According to the Sumerian-Akkadian version, supreme divinities and, first of all, Ellile (Bel, Sumerian Enlille) decided to destroy the humankind with a flood, but Eah, the god of water and wisdom (equivalent of Sumerian Enki) warned Atrakhasis, a king, priest and a righteous man (Uotnapshitim, Sumerian Ziusoudra) and instructed him how to build an ark, where he saved his kin, and domestic and wild animals. The flood lasted for 7 days. The high waters covered the Mesopotamian lowlands up to moun­tain spurs in the northeastern edge, where the uncontrolled vessel was washed ashore Mount Nizir (Nitsir).

We interpret the layer of silt discovered by L. Wulley [Вулли, 1961] during the excava­tion in Uor (the Euphrates) as flood deposits. The layer covers Obeid culture horizons dated back to the second half of the 5^rtl millennium BC, which contain moulded ceramics, without any signs of copper, and are overlain by the Uoruk culture layers of the 4^dl millennium BC that contain ceramics prepared with a potter’s wheel. Tablets with ancient letters found in upper layers are dated approximately to 3000 BC. Therefore, the flood occurred early in the 4th millennium BC against the background of fast rising level of the World Ocean and, cor­respondingly, of the Persian Gulf. As follows from the Sumerian-Akkadian legends, that extreme rise during the Flood could be caused by a combination of thundershowers with a surge of gulf water generated by a south gale-force wind. Direction of the wind is attested to by the leeway of the ark that moved to the north of the town of Shourouppak toward Mount Nizir [Фрейзер, 1989].

The mentioned natural phenomena were important for the formation of both the Sumerian ethnos, and ancient Sumerian city-states. Not later than early in the 4^ttl millennium BC Sumerians appeared in the Lower Mesopotamia. The analysis of archaeological and lin­guistic data and Sumerian legends indicates that their ethnos comprised two sources: one of them is Iranian (the Khadji Mohamed culture) [Мелларт, 1982; Заблоцка, 1989]. Representatives of the other source could have migrated from the south, from the part of the Mesopotamian foredeep that appeared under the rising waters of the Persian Gulf and was later overlain with prograded delta deposits of the Tigris and Euphrates Rivers. Archaeological artifacts found on the gulf coasts are similar to the Eredou culture in the Lower Mesopotamia and can bear witness to this source. The Sumerian tradition of building temples on platforms can be traced back to this culture; most likely, it was originally intend­ed to protect temples against floods, but later transformed into building ziggurats. Migration forced by the surge of water could serve the basis for the Lost Paradise legend, which pre­cedes the Flood story in the Sumerian-Akkadian texts likewise in the Bible.

The efficiency of irrigative measures increased and the system of ancient Sumerian civ­ilization city-states formed by late 4th century BC, after the sea level became relatively stable [D. Kennet, J. Kennet, 1996]. The largest cities of Uorok, Uor, Lagish, and N'girsu were located in estuaries and served seaports and international trade centers of the considered peri­od. Later on, the intensity of sedimentation has exceeded the scale of eustatic variations and tectonic subsidence of the piedmont basin, so that today mins of these cities are 300 km far from the gulf.

Most of ancient peoples’ legends about the Flood were evoked by local disasters of actu­ally later period. Among the others, the ancient Indian and ancient Greek legends take spe-

rial place [Фрейзер, 1989]. Indo-Aryans could adopt the ancient Indian legends from earli­er dravid inhabitants of India, who kept sustainable relations with Mesopotamia at the time of the Harappan civilization. The second legend (the myth about Deucalion) most probably reflects the effects of the Great Minoan eruption on Santorini (see Section 4.4), including tsunami that covered the Attic part of Greece during a severe earthquake, which preceded the eruption, or heavy showers during the eruption itself.

Chapter 2

ACTIVE TECTONICS

2.1. Teetooic zoone aad Нієпгngnterthpic єпо^ііи

Orographic variety of the central part of the Alpine-Himalayan belt is determined by het­erogeneity of the alpine tectonics and nnotnctonic (Oligocene-Quatemary) manifesiaiions (Fig. 9) [Трифонов, 1999]. They control characteristics and pattern of active faulting. Such heterogeneity of the alpine tectonics depends on the longitudinal and transverse zoning. The longitudinal zoning reflects certain distinct features of interaction between the southern Gondwanian plates (the Indian, the Arabian, and the African) and the Eurasian plate. The main of these features is that some fragments of the north-drifting southern plates broke off and started to move quicker than others, so the axis of spreading of the Tethys ocean jumped onto the rear of the fragments. In recent geological transverse sections (from the south to the north) these changes has manifested themselves in a sequence of neo-, meso-, and paleo­Tethys sutures, and corresponding adjoining zones of island arcs, or active continental mar­gins. They all have been more or less renewed by neotectonic movements. As the drifting southern plates were rotating around a western pole, structural manifestations of the drift, including the nnotectonic features, were increasingly more pronounced in the eastern direc­tion [Трифонов и др., 2002].

The considered tectonic zones developed under compression and underwent essential transverse shortening that lasted during a part of, or, in some zones, throughout the neotec­tonic epoch. It started 40-50 million years ago, when the neo-Tethys had been closed and col­lision had taken in the most part of the Alpine-Himalayan belt. In the Pamir-Himalayan seg­ment, the Ьпііrepresents an asymmetric bilateral orogen, the north-northeastern flank of which is extended compared to the opposite one. Tectonic zones of the belt form a sequence of tectonic nappes, whose ages are successively younger both to the north and to the south of the neo-Tethys suture [Трифонов, 1983]. Such sequential rejuvenation of deformation is less evident in the Arabian-Iranian segment, where the belt is narrower. Manifestations of the neo­tectonic compression include folds, thrusts and strike-slip faults not only in the sedimentary cover, but also in the basement. The progressive folding and thrusting in the basement have resulted in general uplift of the area.

The discussed region includes the Adria-Aegean (the eastern part), Arabian-Iranian, and Pamir-Himalayan segments of the Alpine-Himalayan belt [Трифонов и др., 2002]. The seg­ments are bordered from the west by weakly bent systems of NNE-trending left-lateral faults that continue into the southern plates and one way or another join the Middle Indian rift sys­tem. The boundary between the Adrian-Aegean and Arabian-Iranian segments is represented by the Levant sinistral fault zone that is continued by the East Anatolian sinistral zone to the northeast. The Levant zone joins with the Red Sea rift. The recent boundary between the Arabian-Iranian and Pamir-Himalayan segments is represented by a sinistral fault system, the main features of which are the Chaman fault and the Darvaz segment of the Darvaz-Alai zone. The Chaman fault continues to the south with a row of smaller en echelon faults [Wellman, 1966; Tapponnier, Molnar, 1979; Nakata et al., 1991] that are farther continued by the Owen fault to the Indian ocean. The considered transverse fault zones (particularly in the KE-trending parts) have a compression component of motion which reflects in reverse, or thrust offsets and parallel folds. In the meantime, the transverse zones are characterized by en

echelon structure with pull-apart basins forming between some of the segments. They are most typical for the Levant zone (the Aqaba, Dead Sea, Tiberian, and El-Gaab basins), and are identified in the Darvaz fault zone also (the Kokcha basin).

Generally, active structures within the segments strike from the northwest to the south­east with characteristic bends. The southern margin of the central part of each segment forms a gentle arc, bent to the southwest. The northern comer of each segment is rounded by Cenozoic tectonic zones arranged to form a syntaxis represented by an arc convex to the north. The main syntaxes are the Lesser Caucasus and Punjab-Pamir, having common struc­tural features. Their western flanks are formed by sinistral fault zones along the segment boundaries. Dextral active faults strike along the northeastern sides of the syntaxes. The dex- tral faults stretch farther to the southeast, where they are replaced by active thrust-and-fold zones bending to the southwest. The main syntaxes are areas, where the general north-north­eastern drift of the southern plates locally transforms into the northern drift. Smaller syntax­es are identified in the eastern parts of each segment, including the Rhodos syntaxis between the arete-Hsllse and the Cyprus arcs, the Oman syntaxis between Zagros and Makran (the Aladagh-Benalud arc to the north of the Lut block is formed by its drift) and the Assam syn­taxis to the east of the Himalayas.

On the northeastern flanks of the southern plates, tectonics depends on the features of the Earth’s crust. Recent subduction combined with counter thrusting of the northern side takes place in the Crsts-Hellen and Cyprus arcs, where the southern plate has the suboceanic crust [Трифонов, 1999]. In the Himalayas and Zagros, the continental Indian and Arabian plates plunge under the crustal structures of the belt gently because of relatively small average den­sity of rocks. Although the sedimentary cover of the foredeep has little, or no contribution to the underthrusting, it is detached and deformed independently of the basement, forming active thrusts and folds clearly pronounced in the topography. In Zagros, this process is pro­moted by the presence of a Late Precambrian evaporate formation in the lowermost section of the cover. The age of thrusting and folding in Zagros has been determined by paleomag- netic dating of coarse molassa [Бачманов и др., 2000]. These data indicate that both the folding, and the subsequent local thrusting and detachment covered a foredeep area in front of the Main Zagros underthrust. After that area was folded entirely, local detachments joined into a single detachment zone and the area was uplifted. The folding and associated process­es propagated into an adjacent area further to the southwest from the Main underthrust. Eventually it led to the formation of several zones with progressively younger ages of fold­ing, detachment and uplift from the Late Miocene up to the recent time. These zones of dif­ferent age have different features of active tectonics. Active reverse and strike-slip faults are discordant relative to the folded structure in the older zone (the High Zagros). Active tecton­ics of the intermediate zone (the Lesser Zagros) indicates recent continuation of folding, thrusting and development of a marginal flexure, marking detachment propagation boundary. In the Coastal zone, the most distant from the Main underthrust, we observe only local active folds, which represent the initial stage of the process. Similar process is manifested in the Himalayan Foredeep by rejuvenation of course molassa to the south from the underthrust [Yeats, 1986].

Deformation and displacements on the southern flanks of the Alpine-Himalayan belt by no means compensate the drift of the southern plates. By a mechanism of bulldozing, then- motion is, to a considerable extent, transmitted to the northern parts of the belt [Трифонов, 1999], mainly transforming into active offsets and deformation in boundaries of microplates and crustal blocks and partly realizing as intrablock deformation. As intensity of the defor­mation decreases from the south to the north and northeast, the style of active tectonics changes accordingly from combination of faults and folds to faults solely. According to the general increase of deformation from the west to the east of the belt, the bulldozing occupies large areas in the Central and Eastern Asia, being limited only by Iran in the Arabian-Iranian segment, and covers narrower zones to the west of it. The bulldozing is combined with squeezing of rocks out of the syntaxes, which represent areas of maximum compression. This

determines the predominance of strike-slip motion over thrusting and reverse displacements on active faults in the bulldozing areas. Considering the rheological conditions of the conti­nental crust, we explain such predominance of strike-slip motions on active faults in the Alpine-Himalayan belt by the fact that this sense of motion consumes less energy than move­ments on thrusts, reverse and even normal faults [Trifonov, 2000].

2.2. Active faults in the Pamir-Himalayan region

and Central Asia

The region includes the Pamir-Himalayan segment and adjoining structures in Afghanistan and Pakistan that belong to the Arabian-Iranian segment (Fig. 10). Active faults in the structural bounds of the Panjab-Pamir syntaxis play an important role here. Actually, the western boundary of the syntaxis is the boundary between the Arabian-Iranian and Pamir- Himalayan segments, represented by a sinistral fault system, the main features of which are the Chaman fault and the Darvaz segment of the Darvaz-Alai zone. An average slip rate in the Late Quaternary reaches 10 to 15 mm/year [Трифонов, 1983] (Fig. 11). In the E NE- trending Alai segment of the Darvaz-Alai zone, sinistral slip transforms into thrusting. Along the two fault branches, the slip rate reaches 10 to 12 mm/year [Nikonov et al., 1983, 1984]. On the northeastern boundary of the syntaxis, an average rate of the Late Quaternary dextral motion (V_lq) on the Pamir-Karakorum fault reaches 27-35 mm/year [Liu et al., 1991]. It transforms partly into transverse shortening on the Boundary and Frontal thrust zones in the Himalayas (V_lq = 15-25 mm/year [Valdiya, 1986]) and partly into dextral slip on the east­trending en echelon fault system (Vlq = 10-20 mm/year [Armijo et al., 1986; Armijo, Tapponnier, 1989; Molnar, Deng Qidong, 1984]) that strikes along the southern Tibet up to the Red River dextral fault in the East. There are NNE-trending grabens (the Yadong-Gulu is the largest) between the en echelon system and the Himalayan active thrusts.

Active zones with combined trust and dextral components of motion are identified to the west of the syntaxis. They are the Surkhob-Iliak zone [Трифонов, 1983] and Chormak- Andarab and Herat zones [Wellman, 1966; Tapponnier et al., 1981; Трифонов и др., 2002]. In the Tien Shan and Dzhungarian Alatau, north of the syntaxis, WSW-trending active thrusts [Абдрахматов, 1995; Курскеев, Тимуш, 1987] combine with the NW-trending dextral faults, including the Talas-Fergana fault (Vlq increases from 5 mm/year in the SE up to 15 mm/year in the NW; Fig. 12 [Trifonov et al., 1992]) and the Dzhungarian Fault (V_w = 5 mm/year [Трифонов и др., 2002]).

To the east and NE of the syntaxis, major rast-irending sinistral strike-slip zones have been identified in the western China [Ding Guoyu, 1984; Atlas..., 1989], Mongolia [Трифонов, 1985; Трифонов, Макаров, 1988] and in the southwestern part of the Baikal rift system [Лу­кина, 1988]. They are the Amimaqing (Kunlun) (Vlq =1-10 mm/year), Altyn-Tagh (Vlq = 7-9 mm/year), Gobi-Altai (V₁q =6-9 mm/year; Fig. 14), Khangay (Vlq =8-10 mm/year), Baikal-Mondinsky (Vq = 1,5-2 mm/year), and Tunka (Vlq is up to 4,5 mm/year) zones. Eastwards, the system of sinistral faults on the northern flank of Tibet bends first to the SE (the Xianshuihe zone and Changma-Kilian zone of V_lq = 5-20 mm/yr and ^vl_q = 4—6 mm/yr, respec­tively) and then to the south (the Aiming, Zemuhe and Xiaojiang Faults of total Vlq up to 10 mm/yr) [Ding Guoyu, 1984; Molnar, Deng Qidong, 1984; Atlas..., 1989; Allen et al., 1991]. In the western Mongolia, the Gobi-Altai and Khangay sinistral zones join with a NNW-ttending dex^al system, including the Ertai (Vlq = 4-12 mm/yr) [Ding Guoyu, 1984; Molmar, Deng Qidong, 1984; Shi Jianbang et al., 1984], Kobdo (Vlq = 4-5 mm/yr; Fig. 13), and Bidje (Vlq = 2-2,5 mm/yr) faults [Трифонов, Макаров, 1988].

2.3. Active faults in ihe Arabian-Caucasus region

The Levant and East Anatolian fault zones form the western boundary of the Arabian- Lesser Caucasus syntaxis (Fig. 15,19-21). An average rate of the Late Quaternary slip reach­es 7,5 mm/yr in the southern (Israel) segment of the Levant zone [Zak, Freund, 1965], but

decreases to 5-6 mm/yr in the northern (Syrian) segment [Трифонов и др., 1991], where the motion is partly accommodated on the Roum fault along the continental slope [Трифонов, 1999]. The same rates are typical for the East Anatolian zone [Saroglu et al., 1992a], which joins the Pambak-Sevan-Khonarassar Fault Zone in the Lesser Caucasus, forming the North Armenian arc of active faults. Inside the North Armenia arc, there is the second and steeper arc defined by the Akhurian. fault in the west and the Garni fault zone in the east (Fig. 23). To the north of the North Armenian arc, on both slopes of the Great Caucasus (mostly on its southwestern slope) longitudinal active thrusts combine with dextral faults trending to the NNW (Fig. 25). In the northwestern Caucasus, this system is complicated by transverse nor­mal faults.

Two systems of active faults form the northeastern side of the syntaxis. The Pambak- Sevan-Khanarassar fault zone in Armenia represents one of these systems (Fig. 22), where dextral component of the motion exceeds the reverse one repeatedly and reaches 4—5 mm/yr [Trifonov, Karakhanian, Kozhurin, 1994]. The southeastern termination of the Khanarassar fault is continued by the North Tabriz fault (Fig. 24) [Berberian, 1976; Trifonov, Karakhanian et al., 1996]. As a result of bending of its strike to the ESE, the reverse component increases. Fragments of the system under discussion are identified southeastwards, behind the Zagros, where similar regularity is established: reverse, or thrust component of motion increases along with fault bending to the east relative to the general southeastern trend.

The second system follows the Arabian plate boundary. In the northern part, it is repre­sented by the southeastern segment of the North Anatolian dextral fault zone with an average rate of motion of about 9 mm/yr [Saroglu, 1988]. The Main Recent fault of Zagros branches out of it to the SE [Tchalenko, Braud, 1974]. Predominantly, it is also a right-lateral fault with

= ⁵ -10 mm/yr [Trifonov, Hessami, Jamali, 1996]. The main southeastern continuation of this fault is represented by the arched Dena fault, striking to the south and characterized most­ly by dextral displacements; southward it bends to the SE, where thrusting and associated folding predominate on its branches. From the Dena fault, the Kazsrun-Borazjan (Fig. 17) and Ka-eh Bas (Fig. 18) dextral zones branch off to the south [Бачманов и др., 2000]. The Kazerun-Borazjan zone (with Vq = 5 mm/yr of an average rate of the Quaternary motion) continues in the southern direction in proportion with thrust and folded active zones branch­ing out of it to the SE. The Kareh Bas zone strikes to the south with predominance of dextral component of motion. Southerly, the zone forms several step-like bends to the SE, and these southeastern segments are characterized by thrusting. Finally, it turns to the SE forming a flexure-thrust zone with uplifted northeastern side. The described faults demonstrate depend­ence of sense of motion on the fault strike. In the meantime, ^,the system of right-lateral faults is not straight as a whole: it trends to the ESE in the northern part (the North Anatolian zone), then turns to the SE (the Main Recent fault) and finally to the south (the Dena, Kazerun- Borazjan and Kareh Bas faults).

The boundary strike-slip zones converge on the northern flanks of the syntaxes. Along with an expectable increase of compression component, behavior of the strike-slip com­ponent of motion in this area is particularly noteworthy. The north-trending Levant zone continues by the NE-trending East Anatolian zone. The latter bends to the east and joins with the Pambhk-Sevan-Khanhrassar fault zone (that bends to the west) at an angle of only 17°, and both zones keep strike-slip sense of motion up to the junction [Trifonov, Karakhanian, Kozhurin, 1994]. A similar sharp angle between sinistral and dextral faults has been described by A.S. Karakhanian in the Doruneh fault zone to the north of the Lut block in Iran.

From its intersection with the East Anatolian zone, the North Anatolian dextral fault zone continues to the NW and to the west up to the Sea of Marmara, and borders the Anatolian plate from the north [Saroglu et al., 1992b]. An average rate of the Quaternary dextral motion on the North Anatolian zone reaches 13-15 mm/yr in the central part, and 15-20 mm/yr in its eastern and western terminations [Trifonov, Karhkhaeian, Kozhurin, 1994]. According to the GPS data, 20%-30% of the additional dextral deformation is distributed across a 100-km-

wide band around the fault zone and the total rate of the counter-clockwise rotation of the Anatolian plate is up to 24 mm/yr [McClusky et al., 2000].

East-trending faults with sinistral component of motion play a significant role in active tectonics of the northern Iran (Fig. 26-28). These faults include the Dast-e Bayaz, Doruneh, Mosha (Vlq = 2-3 mm/yeao), and Ipak (Vlq = 1-1,5 mm/year) faults and the rupture zone of the Rutbar 1990 earthquake in Albnoz. The associated major north-trending dextoal faults (the Jabbar, Nalband, Ravao, and Kuh Banan) predominate in southern areas of Iran [Wellman, 1966; Tchalenko, Ameraseys, 1970; Tchalenko, Bnrbnriαn, 1975; Mohajnt-AshjaI et al., 1975; Berenrian, 1976, 1977; Berberian et al., 1992; Trifonov, Karakhanian et al., 1996].

In Copet Dagh, the dominant active structure is the NW-trending Main Copet Dagh fault zone (Fig. 29). To the west, the zone is continued en echelon by the Isak-Cheleken and Apsheron-Threshald zones. Dextoal component of motion on the Main Copet Dagh fault zone is several times as great as the reverse one, and reaches 2 mm/yr. Weaker manifestations of active faulting have been identified in the South Aral area of the Turanian plate, where four fault zones join. One of these faults is the NW-trending Central Ustiurt fault [Nikonov, Sholokhov, 1996] that is continued en echelon to the SE by the Northern Border fault of the Bukhara step [Пинхасов, 1884]. The Amu Darya fault can be traced to the NNW up to their junction [Гохберг и др., 1988] and continues northerly by en echelon arrangement of faults near the Amu Darya mouth and in the Aral sea. Small vertical offsets are characteristic of all these faults. Dextoal and sinistral components have been recorded on the Central Ustiurt fault and the Amu Darya fault, respectively.

2.4. Active fauUt in the Аєпгпп єогГпп

The western tnrmination of thr North Anatolian fault zone forms several branches, the northern of which is continued nn echelon by thn North Aegean fault. The pull-apart basin of the Marmata Sea is situated at the junction of these faults. Seismological data allow estimat­ing thn dextral slip rate on the North Aegean fault as 6 tn 24 mm/yr [Papazachns, Kiratzi, 1996]. Southwards from it, both of the coastal areas of the Aegean Sea arr ruptured by numer­ous active normal faults (Fig. 30). Striking mostly to the WSW in Turkey, and to WNW in Gtnrcn, these faults represent fragments of several arcs convex to the south [Seismotectonit map..., 1989; Saroglu et al., 1992b]. The largest normal faults in Greece form thr Teermπpylae-Atalanti zone, for which Stiros and Rondnyanni [1985] suggest an avnrage rate nf vertical movements nf Vlq - 0,8-1,4 mm/yr and the Corinthian Gulf zone, where, by seis­mological data, extension rate comprises 0,8 mm/yr (Tselentis, Makropoulns, 1986).

According to the GPS data, rates of the south-southwestern drift of thn Aegean Sra islands comprise about 30 mm/yr, which is by 5 to 7 mm/yr greater than the rate for which the counter-clockwise rotation nf the Anatolian plate could be responsible [McClusky nt al., 2000]. This additional extension of the Aegean basin manifests itself in normal faulting. We explain this by an influence nf a mantle diapir [Трифонов, 1999]. The drift producrs thrust­ing on thr Crete-Hellen arc, combined with the subduction of the African plate at rates nf 5-7 mm/yr. Therefore, thr rate of total transverse shortrning of the arc exceeds 35 mm/yr. Seismological data indicate that the rate of shortening is 30 mm/yr in some parts of the arc [Papazachos, Kiratzi, 1996]. Its northern side is uplifted at a rate of about 3 mm/yr [Papadopoulos, 1989; Jackson, McKenzie, 1988].

2.5. А^іте faπlte ea environmennta ffacon

2.5.1. ІпЛиедое пПactive cavils an human eeαirnnmenn. Publications HOnKa-sing se⅛r mic hazard active fault zones may cause ate numerous. Ground shaking, tsunamis and lique­faction are the most commonly cited effects of active faulting and related strong seismicity; each of these phenomena can destrny housing and kill people. In the meantime, many other active faulting effects also deserve detailed examination. Among most important factors to consider is specific land suoface topography determined ∙ by active fault movements (Fig. 31,

32). In tectonically active regions, such as the Alpine-Himalayan belt, wide-spread active faults have created numerous relatively steep slopes, linear depressions, closed or semi-closed basins, etc. Since ancient time, local population has used these and other landscape features as areas convenient for agriculture, or as routes for migration facilitating communication between different groups of people, or as natural barriers impeding free passage from one area to another.

Faults and fault zones, and active ones in particular, can serve channels for vertical migration of different chemical elements. It is suggested that such migration can be accom­panied by various geochemical anomalies, or in some way change otherwise regular distri­bution of the elements in grounds, soils, surface or ground waters [Trifonov, 1997]. Some of the evidences in support of this suggestion were obtained in 1988, during the Tien Shan-lntercosmos-88 experiment [Трифонов, Макаров, 1989]. Variable magnetic and electrical conductivity anomalies were detected in active fault zones of the region. A.V. Abdullaev measured abnormally high contents of mercury and radon above the Chon- Kurchak active thrust in the northern Tien Shan, and the Talas-Fergana dextral active fault in the central Tien-Shan (Fig. 33). The most convincing results were obtained for the Surkhob- Ilyak boundary fault zone between the Tien Shan and Pamirs: lucerne in fields crossed by the fault zone contains about three times higher concentrations of Mn, As, Zr, Nb and other heavy metals than lucerne studied further from the zone does (Table 2) [Лукина и др., 1991].

A profile across the Spitak 1988 earthquake fault in the northern Armenia studied sever­al years after the earthquake has provided more versatile data. At the same profile locations, concentrations of several elements were measured in rocks, soil and plants (Fig. 34—36). The recorded concentration variations partly reflect different composition of bedrock on the fault sides: the northern uplifted fault side contains the Meso-Tethys ophiolites, while the southern fault side is composed mostly of the Upper Cretaceous carbonates. The contents of Mg and Fe are higher on the northern side, while the content of Ca is relatively larger on the southern side. These differences are, however, less distinct (and even indistinguishable for Ca) in soils and plants. Concentrations of some elements demonstrate a tendency to increase toward the fault zone and then to drop suddenly within tens of meters from its plane (Na, Mn, Co, Se, Ga and a more gentle curve for Fe and Ti). The shortage of Co decreases from rocks to soils, and that of Na and Si increases in plants. In the meantime, contents of Mg and Ca measured closer to the fault are lower in rocks and soils within the fault zone, but increase in plants. Measurements of V contents provide a reverse pattem. Therefore, fracturing and water cir­culation along the fault zone may be suggested as main factors controlling variations of chemical elements in the studied area.

Some negative ions may penetrate into fault zones together with deep-source fluids. Higher concentrations of “Cl and ^⁴S0,4 were recorded in bottom sediments of Lake Sevan (Armenia) along the trace of the Khanarassar active fault zone (Fig. 37). Intensive CH4 emis­sion was established in lake areas above this and some other active faults. Certain correlation between weak seismicity and amount of plankton in the lake waters shows significance of the fault activity for biota (Fig. 39-41) [Karakhaeian et al., 2001]. Perhaps, the anomalies described above will help to explain how active faulting can influence biota and human health (Fig. 38, 42).

N.N. Vorontsov and E.A. Lyapunova [1984] pointed out that potential influence of active faults on living organisms can be traced even on a chromosomal level. They studied chro­mosomal characteristics of fossorial rodents of the Ellobius talpinus supsrspeciss. The range of this super-species covers an area from the southern Ukraine up to the northern China and from the Kara-Kum desert up to the western Siberia. The two allopatric karyomorphs repre­senting the super-species are E. talpinus s. str. (2n = NF = 54) in the west and E. tancrei (2n = 54; NF = 56) in the east. Near the Surkhob-Ilyak active fault zone in the Pamirs-Tien Shan boundary, the E. talpinus super-species shows an uniquely wide chromosomal variabil­ity with the constant NF = 56: 24 karyomorphs with 2n = 31-54 were identified there (Fig. 43). N.N. Vorontsov suggested this could be a step toward formation of a new species.

Such chromosomal variability could result from a saltatinnal genetic mutatinn and the subse­quent period nf divergence and fast fixation of the changes.

Similar Robertsonian variation was established for mole-rats Microspalax leucodon in active fault zones of Bulgaria, Yugoslavia and Turkey, and fnt M. ehrenbergi in the Levant fault zone. The housr mouse Mus musculus (stable karyotype is 2v = NF = 40) demnnstoated chromosomal variations neat active zones in Italy [Capanna, 1980]; the same phenomenon was established for sub-alpine vnles Pitymys daghestanicus in the Transaaucasus, particular­ly neat the Khαvarassar aatier fault zone [Ляпунова и др., 1988]. A mutant form of Y^hoo- mnsome and some other chromosome anomalies were found among voles inhabiting active fault zones in the southern Italy, Yugoslavia, the Tien Shan, the Altai, the south of the Baikal lift, Kunashit and Shikotan islands of the Kuriles, Hokkaido, Honsu, and the northern Andes. High ceoomnsnmal variability was recorded for pocket gnphers near active fault zones in thr states nf California, Oregon and Washington [Vorontsov, Lyapunova, 1984]. A.S. Karakeaniav found that concentration of endemic plant species neat active zones in Armenia may be interpreted as a mutation effect (Fig. 44).

2.5.2. Importance оіectiea faoeteag ton tfe generatier off narly agricaltuce iu thin Eastern Oykumena. Influences nf active faulting on human life and social development aoe primarily related to its being the source of strong earthquakes and such accompanying phe­nomena as ground defnomatinn, landslides, and, sometimes, volcanism. However, active faulting has been a natural source fno not only disasters. Fig. 2 shows active faults nf the Eastern Mediterranean and the Middle East and the archaeological sites with evidence of rarly agriculture dated back to the second part of the 9*єand 8^m millennia ВС [Мелларт, 1982]. These sites are all (except of Chatal Huyuk in Turkey) lncated near active fault zones, or in basins bounded by active anticlines with blind thrusts in corrs. Good soils in suitable fields, sources of water and seed material for planting, as prerequisites of agriculture, could be associated with active tectonism. Active faults formed steep-slope bounds surrounding ivirrmountain basins and foredeeps with springs and fertile soils on alluvium. The ridges uplifting all around delayed the western cyclones and facilitated raining in the basins. All of the considered early agriculture sites were located in thr reginns, where wild ancestors of cul­tivated plants spread widely [Вавилов, 1965]. Within such regions, N.I. Vavilov distin­guished several arras, where a large number nf different species and fnoms of eatable plants grew conauorenily. These arnas fell within nr clnsr to the active fault zones that may have been sources nf mutation effects. It helped early farmets to find most productive plants.

Therefore, active faulting is accompanied by a eπtiriy geophysical and hydro-geoahem- ical anomalies that can contributr to suppression of thr biota, including cultivated plants, to apprarance of specific human diseasrs, or wider distribution of other illnessrs, depending nn the geπdyvamia, geophysical and geochemical features of the fault zones. On the nther hand, the neotecionIa development of fault zones created specific landscapes rich in underground water sources and suitable fno settlements and agoiculture, especially in arid regions of Asia. Despite threatening the biota and local populations, mutation effects accumulating in active fault zones promoted geneoation of early agriculture. All these features and effects of active faults should be taken into account when planning covtiouation, land use, preventive health caoe measures, and population activities in paoticular arras.

Chapter 3

STRONG EARTHQUAKES

3.1. Cathtopge co estoon cnsOhhueaes in the Eastern Oykumena

The Catalogue of strong earthquakes in the region between 23-50°N and 15-80°E (Appendix 2) was compiled to study the seismic history nf the Eastern Oykumena in the sec­ond part of Holocene. The Catalogue includes M_s> 5,7 events from 3200 Bc till the end of the

20^th century. For this compilation, we used published regional catalogues and earthquake descriptions provided in relevant publications. Moreover, we examined data on historical seis­micity contained in some of Latin, Byzantine, Arabic, and Iranian sources and re-analyzed the Armenian chronicles in detail. To determine intensity and magnitude for historical earthquakes in Armenia, we compared descriptions of their damage with the pattern of destruction caused by the Spitak 1988 earthquake (Fig. 45-47). As a result, we identified previously unreported historical events and defined parameters for some other events with greater precision [Trifonov, Karakhanian, Assaturian, Ivanova, 1994; Караханян, 1999]. The Catalogue (Appendix 2) includes 2,911 earthquakes. Their parameters are grouped under 13 columns.

Column 1 - earthquake ID number.

Column 2 - sources of information. Numbered references for each earthquake are included in full in the list of references enclosed to the Catalogue.

Column 3 - earthquake year; «-» (minus) before the year means BC. A season of the earthquake is given, if it is known without more accurate definition (for example, in ID 379).

Column 4 - Earthquake month.

Column 5 - Day of a month.

Column 6 - Exact time (hours and minutes are separated with « ; »).

Column 7 - Latitude North (degrees and degree decimals).

Column 8 - Longitude East (degrees and degree decimals).

Column 9 - Magnitude value.

Column 10 - Magnitude type (usually Ms). Sign C shows the magnitude is calculated using macroseismic data.

Column 11 - Calculated earthquake energy in J.

Column 12 - Earthquake intensity by the MSK scale.

Column 13 - Hypocentral depth in km.

The selected base magnitude is Ms derived from surface waves and the one approaching in its size and sense to.M_w magnitude referred to in Russian catalogues. In case macroseis­mic data were used to estimate magnitudes we used the relationship suggested by N.V. Shebalin, that links earthquake intensity (/), magnitude (M) and focal depth (Я):

A part of the sources use Mb magnitude derived from body waves. To re-calculate Mb into Ms we used the formula of B. Bolt [Болт, 1981]:

F.T. Aptekayev’s formula was used for earthquake energy estimations:

3.2. Spatial tad temporal variations of siansg seismicity manifestations

3.2.1. Tasks and methods of the study. Majority of strong earthquake foci in the region have been recorded within 20-30 km of the Upper crust. N.V. Shebalin [Шебалин и др., 1991] distinguishes diffused and concentrated seismicity. The concentrated seismicity is located in active fault zones [Ambraseys, 1970, 1988, 1989; Berberian, 1976, 1977; Дотдуев, 1986; По­лякова, 1985; Ulomov et al., 1999; Ulomov, 2000; Shebalin et al., 2000; Trifonov, 2000] (Fig. 52). We studied variations of strong earthquake manifestations in time, including long­term variations for 5000 years (by grouping the earthquakes in 50-years intervals) and short­term variations for the 20^ttl century (by grouping the earthquakes in 5-years intervals).

The first strong earthquakes documented in written can be dated to the fist half of the 8th century BC. These early reports are represented by a well-known passage in the Book of Prophet Zacharia [Zach. 14 : 4-5], according to which the earthquake in Israel is dated to

about 760 ВС, and its magnitude is estimated as M_s ~7,3 [Nur, 1991], and the Khorkhor cuneiform inscription of the Urartian king Arghishti the 1st (787-766 BC), that allows us to date an earthquake in the southern Lake Sevan area (Armenia) to 782 BC presumably, and estimate its magnitude by hrchasossismicity data at M_s ~ 7,2 (Sections 6.1.2 and 6.3.2). These written evidences are unique, considering that strong earthquake reports became more fre­quent only since the classic Greece epoch. Seismic events of the preceding period have been revealed by archaeo- and palsossismological studies that provide known uncertain results, giving a selective characteristic of the studied region. Historical earthquake parameters have been estimated with a relative degree of certainty only if a sufficient amount of macro-seis­mic data is available.

As historical conditions can influence the accuracy of earthquake-related information records and their preservation, these shall be taken into account when assessing representa­tiveness of the catalogue for individual parts of the region and for individual time intervals. To be certain in the important role of this factor we have correlated periods of growth and decline in the Byzantine Empire history [История Византии, 1967] with the numbers of strong earthquakes that were recorded in the same periods across the areas then under the cul­tural influence of the Empire. The areas include the Aegean, Greece, Anatolia, and Eastern Mediterranean. Although the results such correlation provided indicate no simple relation between historical conditions in the Empire and respective numbers of earthquakes recorded, there are grounds to suggest coincidence of stronger seismicity with political and economic decline periods, and potential contribution of earthquake effects to such decline, rather than assume a more consistent record of earthquakes during social stability stages (Fig. 48). The factor of earthquake record incompleteness can, apparently, affect analysis of seismicity vari­ations in time if small areas are considered, but it becomes less sensible if we integrate the areas as shown in Fig. 48. Keeping this in mind, we limited our study of space-time distribu­tion of strong earthquakes to a few selected large areas and zones that are, in the meantime, characterized by distinct seismotectoeic conditions (Fig. 49):

1. The Carpathian-Balkans area;

2. The Eastern Mediterranean and adjacent part of Africa;

3. The Aegean, including Greece and adjacent part of Anatolia;

4. The North Anatolian zone;

5. The Levant and East Anatolian zone;

6. The Lesser Caucasus and Lower Koura basin;

7. The Great Caucasus;

8. The Zagros;

9. The Alborz and Ala Dagh;

10. The eastern Iran and Makran;

11. The Copet Dagh and Binalud;

12. The western surroundings of the Indian plate (Balujistan and the Indus basin);

13. The northern flank of the Indian plate (the Himalayas, Karakorum, Eastern Hindu Kush, Pamirs, Kunlun and adjacent parts of Tibet and Tarim);

14. The western Tien Shan, Afgan-Tadjik basin and adjacent part of the Turanian plate;

15. The southern Tien Shan and adjacent part of Tarim;

16. The central and northern Tien Shan.

The zones are joined into four provinces representing different segments of the Alpine- Himalayan belt. They are:

I. The area of interaction of the African and Anatolian plates, and the European part of the Eurasian plate (Zones 1-3 and the western part of Zone 4);

II. The western flank and the northern front of the area of interaction between the Arabian and Eurasian plates (Zones 5-7 and the eastern part of Zone 4);

Ш. The northeastern part of the area of interaction between the Arabian and Eurasian plates (Zones 8-11);

IV. The western flank and the northern front of the area of interaction between the Indian and Eurasian plates (Zones 12-16).

Comparing the provinces and zones, and analyzing the seismicity time-series, we assumed that systematic earthquake oecnod starts from the first 50 years' mtni-val fno which at least twn seismic events are reported and more oo less unintetrnpted record nf events is kept subsequently. In the Aegean, the onset of systematic record of ealthquakes is dated to the sec­ond half of the 6th century BC, i.e., an eatly stage of the classic Ancient Greece. Lateo in the ancient epoch (second half of the 4*єcentury BC), systematic record of earthquakes was start­ed in the Noote-Avatolian zone and Eastern Mediterranean. In the region of Caucasus, sys­tematic earthquake accounts first appear rarly in the Middle Age: in the middle of the 5* cen­tury and early in the 8* century for the Greateo and the Lesser Caucasus, respectively. In Ioan and Middle Asia oases, this record started during the period nf the Baghdad Caliphate with its magnificent cultural traditions and high level of development in natural and exact sciences. In the Christian Catpathian-Balkan area the same process began since the 12* century. In the areas influenced by the Indian culture, systematic earthquake reports started not before the British authorities had created relevant survey in the middle of the 19* crntuty; probably, the explanation is local mentality. In the same period, administrative authorities of the Russian Empire initiated a systematic record of earthquakes in the mountains of Tien Shan, Pamirs, and Tarim Basin.

3.2.2. Spatial distribution ol strong seismicity. Fig. 50 shows distribution of total ener­gy released in all strong (M_s> 5,7) earthquakes recorded in the region. Of the total energy released, almost 47% fell to Proeivae I, little more than 23% to Province П, about 13% to Province Ш, and almost 18% to Province IV. Such distribution is, to a considerable extent, determined by the duration of systematic earthquake records. For Province IV, this factoo is of particular importance, since in larger part of its area systematic records were started not bnfore the middle of the 19*єcentury. This tendency can be clearly demonstrated if distribution of the total energy released by strong earthquakes in the region is compared to the distribution nf energy released in the 20* centuoy (Fig. 51), for which shares per Provivars are different: the amount of energy released in Province IV increases up to 41,5%, in Provinces I and II it drops to 38,5% and 7%, respectively. The improvement is mostly due to the record nf mantle earth­quakes with hypocentrel depth > 70 km. A greater part of such earthquakes are confined to the Pamios-Hindoukoush seismic source area located in the NE of Zone 13 and in a neighboring part of Zone 14. Earthquakes with focal depths nf >70 km are responsible for almost half of the total energy strong earthquakes in Province IV released in the 20* century, and fnr only 28% of this total in Province I; earthquakes of this depth range are rare in Provinces II and III. This may lead us tn assume that with the same rate nf record completeness for the pre-rnstiu- mental earthquakes, Province IV seismicity would not appear weaker than Province I. Such assumption is supported by tensnr field calculations fnr the oates of recent deformation derived from active fault data [Трифонов и др„ 2002; Trifonov et al.,1999].

3.2.3. Long-term variations of strong seismicity. Early phases nf seismic activation can be constrained only hypothetically, flom the results of palen- and πrchaeo-seismπlogical studies. Signs of strong earthquakes revealed in Provinces I-Ш (Zones 3, 5, 6 and 9) can be dated back to the middle nf the 3^κl millennium BC. Another identified earthquake occurred lateo, just before the Great Minoan eruption of Santorini (Section 4.4). The next supposed surge of seismicity in Provinces I and П is related to the second half of the 2“* millennium BC. Aochaeoseismological data attest [Stitos, 1996] that strong earthquakes destroyed or severely damaged most of Achaean cities nf Greece by the second half of the 13* century BC, which facilitated the success of invading Dorian and Thoaciav-Illyriav tribes and led tπ the fall of the Mycenaean culture (see Section 7.3).

The rhythm of seismic events in individual seismπtectπnic zones of the Eastern Mediterranean can br traced only from the middle of the 1^s* millennium BC. In Province I, such rhythm is determined by the distribution of earthquakes in time across Area 3 and Zone 4 (Fig. 53-58). Seismic activation phases coincided with the first half of the 4* centu- 18. Трифонов В.Г., Караханян A.C. 545

ry ВС, second half of the 3^πi century BC (?), second half of the 1st century AD, second half of the 4th century, second half of the 6* century, first half of the 8th century (?), second half of the 10t^h century, first half of the 14* century, first half of the 16t^h century, second half of the 17th century, and second half of the 19th - first half of the 20th century. These phases repeated in about 200 years intervals, except of those involving the strong earthquakes of 365 and 1303 AD (M_s = 8,0-8,3), for which the preceding intervals increased to 300-350 years. Against this background, individual rare, but rather long epochs of active Seesi-mcny can be related to the periods from the second half of the 4* to the second half of the 6^fll centuries (the Early Byzantine Paroxysm) and from the second half of the 17th to early 20^th century. Both were characterized by seismicity increase (considering improved record of earthquakes since the 18th, and particularly since the 20th century) and seismic activity peaks recorded in the beginning and in the end of each epoch. Although periods preceding these epochs each had a complex pattern with ups and downs in earthquake number and energy, the general tendency was on the increase.

In Province II, the number of earthquakes and amount of released energy are the great­est in Zone 5 (Fig. 59-61), where seismic activation intervals are estimated at 200-250 years, or, at a divisible value of 450-500 years. Similar activation periodicity is observed in Zone 4, but activity peaks in Zone 5 are shifted several decades ahead of the respective peaks in Zone 4 [Trifonov, Karakhanian, Assaturian, Ivanova, 1994]. In the meantime, the peak observed in Zone 5 late in the 9^ttl century changes to a short-term decline followed with a rise reaching its maximum in the second half of the 12* - early 13* centuries, while maximum activity peak in Zone 4 falls to early 11* century, but seismicity indices in the 9* -10* and 12th - first half of the 13* centuries are low. The time-series of seismic phenomena in Province II (Fig. 56-58) reflects an integrated effect of all events that occurred there, but, for larger part, of those related to the most earthquake-prone zones of Levant-Eastern Anatolian- 5 and the eastern part of the North-Anatolian-4, as well as to Zones 6 and 7. The Early Byzantine Paroxysm can be also identified in this Province, although it started earlier than in Province I (the 4^th-5^th centuries AD). Like in Province I, an epoch of activation in Province II coincides with the second half of the 17* - early 20^th century, but activity peak here falls to the second half of the 19^ril century, i.e., again, it precedes the respective peak in Province I. The epoch of activation in the 11*-13* centuries peaks at the first half of the 12* century, while in Province I that was the time of activity decline.

In Province III, the time of seismic activity peaks varies zone to zone (Fig. 62 and 63). In some cases, such differences can be attributed to the wave of activation moving from the south to the north (the first half of the 9* century in Zone 8 - second half of the 9^th century in Zone 9 - first half of the 10th century in Zone 11 and first half of the 12* century in Zone 8 - first half of the 13* century in Zone 11), although such correlation is not established for the rest of activity peaks. A characteristic common for most of the zones in this Province is almost total absence of any signs of the Early Byzantine Paroxysm along with the shift of the last activation epoch to the second half of the 20* century. Zone 11 (Copeth Dag), structurally linked with the Caucasus, has something in common with the western Provinces. The last activity peak in this Zone is related to late 19* century, followed with activity decline up to the end of the 20* century. This, as well as minor activity peak in Zones 8 and 11 in the second half of the 17th century, indicates synchronism of activity variations in all three Provinces. In the meantime, an epoch of activation in the Middle Ages,∙ similar to the one established for Zone 5, can be noticed in Province III, particularly in Zagros. An integral seis­micity pattern for Province III (Fig. 56-58) does not indicate clear periodicity, but 200-250 years’ intervals between activity peaks are more frequent.

Analysis of earthquakes in Province IV (Fig. 64 and 65) can add little to the established regularities, considering that till the 19* century seismic events were recorded only in Zone 14 and in the north of Zone 13, that were under the cultural influence of the Iranian- Islamic world. The number of reported earthquakes was increasing steadily till the second half of the 19th century, mostly due to the accounts of mantle earthquakes with hypocentral

depths of > 70 km. In all zones (except nf Znnr 15) the amount of energy released in thr first half of the 20th century is, however, the greatest.

Unified histograms based on integrated data for all four Provinces (Fig. 56-58) show several epochs of increased seismicity, each including a numbrr of activity peaks. If wr - exclude the peak related tn the specific event of the Great Minoan eaoihquake-eonptiov that happened probably between 1500 and 1550 BC (Section 4.4), it is possible to identify three 200-300 year-long epochs of activation (although with diffrrent individual rates nf certainty considering changed completeness of earthquake record). As we do not consider thr merely hypothetical activity epoch related to the middle of the 3^πi millennium BC, these epochs aoe: the eπtly second half of thr 2nd millennium BC, thr middle of the 1st millennium AD, and the end of thr 17th century - early 20* century AD. Each of these epochs had two main activity peaks. The agr of the second peak in the last epoch is different in individual Zones and Provinces. Its maximum in Provinces I and IV falls tn thr first half of the 20^λ crntuiy, while in Province II and Provincr III to the second half nf thr 19*h and to thr srcond half nf the 20th centuries, respectively. Between these epochs, activity periods were recorded also in the 4th century BC and in the 12*h - eπoly 14*єcenturies AD. We attach less importance to these two, considering that the fiost one had relatively small values of seismic activation indices, and that the second period, the 12*h - early 14*h centuries AD, although releasing seismic ener­gy in the amount commensurable tn the same parameter of the Early Byzantine Paroxysm, was neerrteeless characterized by lower level nf activation, taking into account that the area covered by strong earthquake records had by the time increased essentially.

Against the background of described centuries-long fluctuations, more frequent changes occur as well: these πre identified by alternation of peaks and declines in the number nf rπtth- quakes, just as in the amount nf energy they release. Activation periods of 200-250 years, in a sense similar to seismic cycles defined by S.A. Fedotov [Федотов, 1968], are typical fnr individual zones. Despite certain asynchronicity by individual zones, the timing of these peri­ods is smoothened at the regional scale and most often they take form of activity peaks 11^0- ring each 200-300 years.

3.2.4. Frequent engiatians oo gsms oeigmieiiy Оцпсєuhe ghh ceoth^ Brsed on the Catalogue of Strong Earthquakes (Appendix 2), we analyzed distribution in time of the num­ber of earthquakes and thr amount nf seismic energy they released during the second half of the 19*єcentury and throughout the 20*h century both for the reginn as a whole, and for its individual seismntectonia provinces and zones (Fig. 65-82). Our calculation of intervals separating neighboring seismic peaks reveals a regional-scale rhythm of seismic activation with a period of 10,5 ± 2,5 years for crustal (Я ≤ 70 km) and mantle (Я > 70 km) eaothquakes, similarly. Energy release by the coustal earthquakes in 1920-2000 demonstrates the most clear wave distribution with a period of about 12 years: its autocorrelation function coeffi­cient is clnse to 0,8. In individual provinces and, even mnoe so, in individual zones, seismic energy release rhythm is not that rrgulao, although close time inie-vals. between activation peaks πte recorded in such cases as well.

3.2. Influengc co est'rog ensohhuuaes into the Eastern Oykumena population

Deaths and injuries, destruction nf housing and other niouctures πte the most tangible effects of strong eaothquakes, and theit rate has been always determined by a combination of ∙ natural and social factors. Apart of earthquake power and intensity, natural factors include cli­mate conditions and •seasonal vaoiatinns, as well as time of the day, when earthquake occurs. Direct effects of earthquakes have been often added to by land surface defnrmation, land­slides, rock-falls and mudflows, which led to additional damage and casualties, and rendered large agricultural lands unusable. Such effect has been enhanced by damage caused to irriga­tion systems, and by changes in hydrodynamics and Єу^-п^пє^ігуof underground waters. Seacoasts have been most vulnerable in deformation-related earthquake effects. Apart of pop­ulation density in the epicentral πtea, affected society's stπvdπths of living, sncial secuoity and

health care, its current state, cultural traditions and building practice are among the most important of the social factors. Therefore, in the sense of a disastrous event, earthquake is rather social than natural phenomenon.

Mechanical injuries are responsible for a greater part of damage caused to population’s life and health during strong earthquakes. Dead-to-injured ratios vary in a wide range and approach about 1/3 on average [Келлер, Кувакин, 1998]. Severely wounded and slightly wounded numbers relate one to the other as 1 to 10 on average. Women, children and elders prevail among the dead and wounded. In addition to such direct loss, effects of fires, growth of infectious and neninfectious diseases, and psychological traumas have been always among important damage contributors. Most complete information about earthquake effects is relat­ed to the 20th century. Similar estimates for historical earthquakes and, in particular, for events studied by archae- and paleossismelogical methods are much less exhaustive (Fig. 83). Information about casualties of not less than 5,000 is presented in Table 3. Even with its known incompleteness, the histogram of casualty rate distribution in time (Fig. 84) is to a first approximation comparable to the histograms of released seismic energy distribution. Therefore, this seismicity parameter can be used for a preliminary assessment of plausible seismic impacts on the life and activity of the affected population.

Our analysis of strong earthquake effects indicates that whatever strong single events have been they seldom represented the cause of substantial historical changes. A stable state on its rise managed to recover the damage caused by destruction, while human losses were compensated by migration and, with time, by natality. In a politically unstable state, or in one under excessive economic stress, a natural event of extreme character could disrupt the unsta­ble equilibrium, and lead to irreversible changes at once, or later on. Most probably, the Great Minoan earthquake-eruption played such role for the Minoan state (Section 4.4). Similar sit­uation was more clearly pronounced during the seizure of the city of Behoura (south-west of Lake Sevan) by Urartian troops of Arghishti the 1st early in the 8^a century BC, and during the seizure of Ani by Turkish troops in 1064, when the siege force was much greater than that of the defenders and the natural disaster just speeded up the inevitable events (see Chapter 6).

Apparently, the century-long epochs of increasing number of earthquakes and their total energy had greater historical importance. Earthquake impacts of this kind facilitated the con­quest of Greece by Dorian-Illyrian tribes in the 13* century BC. In a similar manner, the Early Byzantine Paroxysm in the middle of the 1st millennium AD aggravated social and eco­nomic difficulties, and complicated the military state in the Eastern Roman Empire. See detailed analysis of these events in Section 7.3.

Chapter 4

VOLCANISM

4.1. Maniffstetiars vOHoloocns valccnii^ is the Ersteas Oykumesr

In the Eastern Oykumena, there are two main volcanic areas with the Holocene and pres­ent-day volcanoes. One of them is the volcanic zone of the Crsts-Hellen arc in the southern part of the Aegean Sea. Of all the Holocene volcanoes in the zone, the scale of effects gener­ated by eruptions of Santorini (Thera) is the largest. The second area is a discontinuous zone oriented transverse to the Alpine-Himalayan belt (Fig. 85). The zone strikes from the Levant fault zone in the Arabian plate up to Elbruz Mt. in the southern margin of the Eurasian plate and spreads into the inner part of the belt (Armenia and adjacent parts of Turkey and Iran). The zone can represent a thermal anomaly of mantle origin that is discordant relative to the crustal collision structures of the belt. Composition of the volcanic products depended on the composition of rocks in magmatic sources at depths of about 40 km. Local structures of the belt, controlling the centres of areal volcanism, were represented by extension zones in front of the thrust blocks, or in the pull-apart structures or accompanying strike-slip faults.

4.2. The SpunSO Hinghamis in Armeπia

4.2.1. Yousg geologicrl foamrtiods rsS theia rge; nachnnrlrgicrl Srtisg. The Syunik pull-apart structure [K^akhanian et al., 1997] has formed in the Khhnarashr dextral fault zone (Fig. 86). The Middle Eocene basaltic-andesite porphyrites, Neogene rhyolite-dacites, and Lower and Middle Pleistocene basaltic-andesite lavas have provided substrata for this struc­ture, filled up by the Late Pleistocene and three generations of the Holocene basaltic-andesite lava (Fig. 87-89; Table 4). Moraines of the last glaciation cover the Late Pleistocene lavas, but are overlain with the Earlier Holocene ones. The two later generations of the Holocene lava overflowed an area of the Chalcolithic petroglyphs dated to the end of the 5* - early 4t^, millennia BC (Fig. 90 and 91). The upper age limit of the eruption of the third generation lava was dated by the age of a burial mound (3650-3350 BC) built using boulders of this lava. Thus, the second and third generations of the Holocene lava were erupted in the first half of the 4th millennium BC.

4.2.2. Tectonics of the Pyusik stauctuae. The Syunik pull-apart structure formed between the terminal parts of two en 6chelon segments of the Khanarassar dextral fault zone (Fig. 87). There is a component of subsidence associated with normal faulting on the western and eastern boundaries of the structure; nevertheless, the dextral slip component continues to be predominant not only on the terminations of the strike-slip fault zone segments, but even on the western and eastern boundaries of the structure. Dextral component of the slip diminishes southwards on the eastern boundary and increases on the western boundary, but the total slip remains the same as in the other parts of the Khanarasar zone. All of the Holocene and majority of the Late Pleistocene basaltic- andesite volcanoes are within, or on the boundaries of the pull-apart structure. The Holocene volcanoes form chains located either along the continuation of normal and normal-dextral faults, or along overstep zones of their en echelon segments, i.e., the volcanoes mark extension of the pull-apart structure.

4.2.3. Relrtiosshio of rctiae fruits rsS aolcrsism. In the considered area, seismic pro­filing by the refracted waves method detected two areas of seismic wave propagation anomalies at depths of 1,5-2 km, that were also identified by thermal anomalies (Fig. 87). One of these areas is just under the Holocene lava and the other is a little further to the north­west, in the Late Pleistocene margin of the Syunik structure. Perhaps, the anomalies are pro­duced by a deep-seated source of warmed up and partly melted rocks. Faults of the structure bear signs of paleoseismic displacements (Fig. 92). One, or two strong earthquakes occurred in this area in the middle of Holocene, after the petroglyphs had been created. The earlier of these earthquakes is dated to 3960-3650 BC, i.e., it approximately coincided with the erup­tion of the second generation of the Holocene lava.

We suggest that the considered events could start with a strong seismic pulse that renewed the existing faults and created new ones. These faults reached the deep-seated ther­mal magmatic source and had served as channels for magma transport until the source was exhausted. A calm epoch that followed was interrnpten by the next strong earthquake and eruption. There could be two pulses of such associated seismic and volcanic re-activation in the middle of Holocene.

The Syunik structure is of interest not only for this tectonic and probably seismic causal­ity of volcanic eruptions; the Chalcolithic petroglyphs in the area can be hypothetically inter­preted as evidence of the early arrival of the Indo-European cattle-breeder tribes to Armenia and, perhaps, to the Trans-Caucasus (Section 6.1.4).

4.3. 8р^іЄ^гуєпієє^Spria

4.3.1. Locrtios of PoSom rsS Gomoaarh. The Bible provides controversial informa­tion concerning location of Sodom and Gomorrah. On one hand, it tells the cities were in the region of the Siddim Valley, «...where the Salt Sea is now,» [Genesis, 14 : 1-3], i.e., it was the southern part of the Dead Sea. On the other hand, to look at the destroyed cities «Abraham...went to a place, where he stood in the face of God» [Genesis. 19:27], and that was a place

near Sikhem, ihr ruins nf which were identified in Samaria, on the wesirrn bank of the Jordan River [Келлер, 1998]. According to the Bible, the Lord hestrnyeh every living thing in Sodom and Gomnrrah with fior and sulfur, and Abraham saw how «smokr was rising from ihr ground as a smoke from an oven» [Genesis. 19 : 28]. The Bible attests also that the neigh­boring Siegor was no* affected. This evidence allows us to suggest volcanic. eruption as the most probable cause of the described disaster, but if sn, it could hardly happen in Sikhem, or in the Siddim Vallry, since reginnal signs nf eruption as young as the nne supposed have been found only in the southwestern Syria and an adjacent part of Jordan.

4.3.2. Holocene volcanism in the southwestern Syria. To the south and to the SE of Damascus, fields of Pliocene-Quaternary basaltic lava and cinder stretch fπo a distance of 450 km [Geological Map..., 1964; Поникаров и др., 1968] (Fig. 93). In this πteal volcanism, individual centres group to form chains, striking io thr NW, and mark extension πteas that branch from the active left-lateoal strike-slip Levant Fault. The lava flow of Kra looks the most recent; it was erupted by a small volcano neat the today's village of Rdemet Ash- Shakhur and spread for 32 km to the east and to thr NE (Fig. 94). Terminal sections of this flow aneerrh Khirbet-Umbashi and Hebarie, the remains of two settlements with mass accu­mulation of bones [Dueertret, Dunand, 1954-1955; Трифонов, Эль-Хаир, 1988]. In Khirbet-Umbashi, the bones soldered into the lava bottom and firmed, along with other set­tlement remains, an upper culture horizon within the layer πf carbonated pebble-stone and gravel oaau1oivg on the uneven surface of the Late Pleistocene lava (Fig. 95 and 96). Among the bones, those belonging io domestic animals and gazelles are prevailing. The radiocarbon age estimate for the bnnes is 2880-2460 BC, and ceramics found in these ^іі^єє^was dated by H. Saliby io the second half of the 3^κi millennium BC (Fig. 97). We suggest that the seiilemenis and the animals perished from volcanic gases, and the lava coming laieo paitly covered what had remained by that time. A necropolis located nearby contains numerous standard burials of the considered period, and most prπbably, it served the burial place for ihnse whn had no* been covered with lava.

4.3.3. Correlation between the legend about destruction of Sodom and Gomorrah and the eruption of Kra. The biblical division of pastures between Abraham and Lot does not seem io be contrary io locating Sodom and Gomorrah in *he south of Syria and io *heir suggested destruction by the lava flow from Koa.

«...Lot selected an area in the vicinity of Jordan; and Lot moved to the east... and set up his marquees up to Sodom» [Oenesisl3 : 11-12]

Evidence in support of this is contained in the texts of the 3rd millennium BC, found dur­ing *he excavations of Ebla (Tell Mardih, 30 km io the southwest of Aleppo), which mention both Abraham, and the cities of Sodom and Gomorrah lost in fioe; however, the translation of these texts is still in be revised [Келлер, 1998]. In the meantime, assuming that the lost cities wroe in Syria (in contrary io whai the Bible says), we would hardly explain how Lot could reach Sier∏r in few hours (a distance of about 200 km), and how Abraham could see whai happened to *he cities from a mountain. It is not impossible, that the legend about the desioua- iion of Sodom and Gomorrah combines collective memories about two events. Actually lnca*rd in the Dead Sea reginn, Sodom and Gomorτah were destroyed by a natural disaster, most probably, by a strong earthquake, but the fresh memory about two settlements perishing ftom a volcanic eruption caused the population io equate these iwn events, which strongly enhanced the didactic effect of *he legend.

4.4. TTh Grrnt Mingoa aroetiop co fire aaaiheini

4.4.1. Data on the Thera eruption. The Santorini (Thera) volcano is situated in the Cre*e-Hellen calcalkaline volcanic zone (Table 5; Fig. 98 and 99). Today, Santorini is repre­sented by ruins of a caldeoa ring up io 13 km in diameter. According io *he original views [Mπrinaios, 1939] the caldera formed duoing the Great Minπan eruption in the middle of the 2nd millennium BC. However, laieo studies proved the Great Minoan eruption just completed

formation of the caldera, which started about 21 ka BP (Fig. 100) [Sbrana, Vougioukalhkis, 1996]. Therefore, estimates of actual volume and weight of eruptive material provided by the Great Minoan shall be essentially less than considered previously, although they still com­prise considerably large figures.

S. Marinatos [1939] found ruins of a Mi^an town buried by the Great Minoan eruption products in the south of the island, near the present-day village of Akrotiri. The town was dated to the Late Minoan epoch (LM-IA). As indicated by the excavations, a strong earth­quake preceded the eruption, but its early precursors permitted inhabitants to leave, taking along the necessaries. They had time to come back and to start the repairing, but the eruption made them to leave hastily. It happened probably in early summer, before gathering, and pro­duced a 6m-thick layer of pumice and tephra in Akrotiti. It was a single eruption of the Plinian type [Doumas, 1990].

4.4.2. IsSicrtioss of the eauotios rsS r oaeceSisg erathqurke is the Aegers aegios. Indications of seismic damage and ashes of the Great Minoan eruption were found in the ruins of the Knossos palace in Crete [Evans, 1902, 1930, 1935; Warren, Puchelt, 1990]. The dam­age was so severe that complete recovery was not possible throughout the LM-IB epoch till the Achaeans invasion in the middle of the 15^ft century BC [MacDonald, 1990]. Judging by the thickness of ashes on the Aegean and Mediterranean ssa-bottem, the principal share of ash blow was directed towards East and SE. Its traces have been revealed in many settlements on Crete and on neighboring islands. Excavations of Seraglio in Kos and Trianda in Rodes uncovered signs of two strong earthquakes within the layers corresponding to the LM-IA epoch. There are indications that immediately after the second earthquake the ruins were covered with ash [Marketou, 1990]. The casualty rate and decrease of settlement area after the disaster were particularly high in Tian^.

4.4.3. Eruotios Srtisg oaoblems. There is much debate about the age of eruption among high and low chronology advocates, who date the eruption to the second half of the 17t^h, and to the second half of the 16°¹ centuries BC, respectively. The archaeological dating is based on the correlation between the Minoan culture and dynasties of the Ancient Egypt. According to the «high» chronology, the LM-IA epoch corresponded to the 13* Dynasty and the period of the Hyksos, while the «low» chronology correlates it with the late Hyksos peri­od and the onset of the 18^dl Dynasty. P.P. Betancourt [1990] considers there are no decisive archaeological arguments in favor of either high, or low chronology estimates, while A. Renfrew [1990] gives preference to the estimates of «high» chronology. We think that chronological correlations of P.M. Warren [1990] are more convincing; this author considers that the LM-IA epoch continued from 1600-1580 to 1504-1480 BC, and Akrotiri was destroyed inl560-1550 or 1535^1525 BC. This is supported by the description of unprece­dented storm, which is read on the stele of Ahmose the P^t, the 18* Dynasty founder (Fig. 101) in the Kamak Temple at Thebes. The storm could be provoked by ashes penetrating into the atmosphere during the Great Minoan eruption. Mentioning of the Pharaoh’s name on the stele allows dating it to the first 22 years of his reign. Considering that the stele was built some time after the storm, it can be dated to 1550-1528, or 1539-1517 BC [Davis, 1990]. The Ancient Greek myth about Deucalion and the Flood could be related to a tsunami during the Great Minoan earthquake-eruption. According to the Paros Chronicles of 265 BC, it took place in 1530 [Фрейзер, 1989].

Samples taken in Akrotiri were most informative for radiocarbon estimation of the age of the Great Minoan eruption. Measurements conducted in different laboratories [Friedrich et al., 1990; Nelson et al., 1990; Housley et al., 1990; Hubberten et al., 1990] (Table 6) provid­ed a series of age estimates, ranging from 1674-1606 to 1554—1534 BC and indicating that respective probabilities of eruption date as the 17th and 16th centuries BC are distributed as 70% to only 30%. The high date was supported by the evidence of acid eruption in an ice borehole Dye3 in Greenland [Hammer, Clausen, 1990] and by tree ring growth anomalies in California [LaMarche, Hirschoeck, 1984] and in the Old World [Baillie, 1990]. However, these findings can be related rather to the grand eruption of Aniayakchak Volcano in Alaska

[Мелекесцев, Миллер, 1997]. Calibration cuoee eαriaiiovn enable both high and low inter­pretations of age estimates obtained in Akoπiioi (Fig. 102) [Weningeo, 1990]. Finally, with the remaining uncertainty of the eruption date estimate, we neve1teelenn base on the archaeologi­cal data io give preference io the lπw estimate.

4.4.4. The Gheat Minoan eeantsou and ttιe Aegean World. ШєGreat Міпоші піп^ iiπn happened in the epoch of the highest blππm of the Mmoan civilization (LM-IA), whan its centers in Crete and oihao islands ware, most probably, united undao the power of Rossis. A severe eπoihqu∏ka and subsequeni eruption undermined iha powao of this stain, which, in iha fight of Crete against Achaeans who earliao had depended on Minoans, weighed in the favor of Achaeans. This happened 50-100 years afiao the eruption, in 1470-1450 BC. A natural disaster could be reflected also in Plato’s legend about the daaih of Atlantis, narrated in his Thymeus and Creteus dialogues.

Part II

INTERACTION OF GEODYNAMIC AND HISTORICAL PROCESSES

Chapter 5

INTERACTION OF NATURAL PROCESSES WITHIN TECTONIC SYSTEMS AND THEIR INFLUENCE ON HUMAN LIFE AND ACTIVITY

V.S. Ponomarev and V.G. Trifonov [Пономарев, Трифонов, 1987; Трифонов, Поно­марев, 1990] introduced a concept of tectodynamic or tectonic system meant iπ understand an integrity of natural processes inia1-rela*eh in a geolngical medium, which, aiiher dioacily, or indirectly, causa the lithosphere io move and tectonic structures io drvelnp in it. Assessing impact of these processes on human activity, we consider open systems, based on social, tech­nological, economic, and political conditions in human communities (community of culture and ethnicity, state, sei*lamani, etc.), invπlving ivieoaaiion with a naiuoal environment. The latter determines io a considerable extent *he chπtacier of economic activity and the way nf living. Such systems are exposed in the effects of external factors, i.e., relationships with neighbors, and changes in nature, including those caused by geohynamic processes.

This interaction between natural and social processes in the aouose of history is attisied by all case studies discussed in the chapieos above. Generation of agriculture in iha Fertile Crescent was prepared by the appearance of permanent settlements and by the development of necessary implements during the preceding stage of plant aolleaiing. The natural factors played their roles in dur cnurse, with climate improvement following io the las* glaciation, and with active tectonics, creating sui*abla landscapes, and contributing io formation of fao- *ile soils, water sourcrs, and diverse seed stock. Thr sama fac*nos (perhaps, except of seed material diveosiiy and pre-existing experience) acted in adjacent regions, whera agriculture spread under *he influence of iha Fertile Crescent. Combination nf climatic changes with active tectonics reflected in *he development of the Khoresm Oasis and Saryghamysh basin in the Aral region. Although periodic rise of agriculture in this area was faciliiaied by humid­ification, iha re-distribution of the predominant flow into Aral or Sπtighamine was mostly dependent on the behavior of active faults.

The foredeep basin of Mesopotamia has bean under the constant impact of Єп^п^sub­sidence accompanied by formation and growth nf anticline folding on the northeastern side of iha basin, and by its filling with sedimeniaoy material, delivered by Euphoaies, Tigris and *heir tributaries. During the historical period, *he shif* of this balance towards sedimentation reflacied in iha prograding of the Tigris and Euphrates deltas. During *he first half of the Holncane, these processes evolved concurrently with *he poni-rlacial growth of the oceans, iha last suoge of which siπtied early in iha 4*h millennium Be and flooded a pπti of the basin. The flooding forced the local population io leave the place and saived iha basis foo the Lost Paradise legend. The transgression added io with heavy showers and surge of water from the

gulf caused grand flooding that was reflected in the Deluge legend. Intense rise of settlements that started not until the sea level had stabilized concentrated at several centers. These cen­ters became the earliest Sumerian city-states and seaports, but lost this function later because of the delta prograding.

The role played by the severe earthquake and the following Great Minoan eruption in the history of the Minoan state and its relations with the Achaean Greece is one further example of interaction between natural and social phenomena.

Chnotnr 6

INFLUENCE OF GEODYNAMIC PROCESSES ON THE DEVELOPMENT OF PRE-HISTORICAL SOCIETIES AND STATE INSTITUTIONS IN ARMENIA

6.1. Histerical-anahennlooical ffamewook

6.1.1. Main hrisrii^itl aata. Sn іПє Є*—Є^th milleιmia Bq there were were MesolitSSc cen­tres in the Trans-Caucasus and the eastern Caucasus (Fig. 103) [Кушнарева, 1993]. Scholars relate the dissolution of a single proto-Caucasian linguistic community into its western and eastern branches to the 6^t--5^,h millennia BC. Representatives of this community populated the Trans-Caucasus, Northern Mesopotamia and the northeastern Mediterranean. As it is possible to build an evolutionary series of cultures developing in the western Trans-Caucasus from the Chalcolithic to the Late Bronze epoch, we may assume that today’s population of the western Caucasian represents relics of the western language branch. They spread from Central Anatolia and the south-eastern Black Sea region (k-ati) up to the nort-western Pre-Caucasus, where similar features were revealed in the Maikop culture dated to the second half of the 4th - early 3^rd millennia ВС [Андреева, 1977; Клейн, 1990]. No traces of any influence from the western Caucasian area on the formation of cultures in Armenia have been revealed.

The Ceramic Neolithic in Armenia is represented by artifacts found in Hatunarh and Tsakhnunk settlements identified wit- the Shulaveri-Shomutepe culture in the middle course valleys of the Kura River and its tributaries (Fig. 104). The evidence of early agri­culture was found in the settlement of Shomu-Tepe in the Akstafa River. Large settlements in the Kura River basin are multi-layered; some of them bear signs of transition from the Neolithic to the Chalcolithic [Кигурадзе, 1975; Мунчаев, 1982]. By archeological corre­lation, the Chalcolithic and the Neolithic in the middle course of Kura are dated back to t—e 51——.th millennia BC and to the 6* and even late 7^m millennia BC, respectively [Кушнаре­ва, Чубинишвили, 1970]. The ^mc age estimate for Shomu-Tepe is 6430-6250 BC. The age of the Sioni unit given by the radiocarbon dating of a sample from the Berikldeebi set­tlement in Shida Kartli (3820-3640 BC by 2a) takes an intermediate position between the Chalcolithic and the Earlier Bronze age [Badaljan et al., 1992]. In Armenia, t—is period is represented by settlements of Tekhut, Ksyah-Blur and probably by lower horizons of Shengavit and Mohra-Blur.

Concurrently with agricultural cultures of the Neolithic and Chalcolithic, absolutely dif­ferent cultures were developing in mountains south of Lake Sevan. These cultures are repre­sented by a part of petroglyphs in the Syunik, Vardenis, and Ghegam highlands. There are two generations of petroglyphs in Syunik [Мкртчян и др., 1969; Караханян, Сафян, 1970; Karakhanian et al., 1997; Караханян и др., 1999]. Generation I petroglyphs are hypotheti­cally dated to the Neolithic and include a relatively small number of animal pictures archaic by performance technique. Petroglyphs of Generation II (in Syunik and probably Ghegam- Vardenis) were made by a community of cattle-breeders, who combined cattle breeding wit­hunting (Fig. 90, 105). Arming and hunting outfit of these cattle-breeders was typical for the Chalcolithic; they domesticated dogs, goats, and caws, and also horses and could use them as transportation means. The petroglyphs are dated to the end of the 5^λ - early 4^ft millennia BC.

T—e flowering of productive economy and primarily agriculture, which started in the cen­tral and eastern Trans-Caucasus since the early Bronze Age, is reflected in t—e Kura-Araks

culture (KAC). Having occupied the areas nnw coeareh with πtiifaais of the Laie Eneolithic agriculture, this culture spread beynnd, iπ the Easiarn Azerbaijan, Daghestan and eastern Tuokry (Fig. 106). Later on, KAC spread into the region nf Lake Urmia, as well as into the eastern Mediie1oavaan, wherr similar ceramics and Kirbet-Kerak-type πtchiiaciure had been developing. This all gave rise in a hierarchy of settlements with larga populated areas (6-10 ha and up *o 12 ha in Arich, the Shioak Plateau) each surrouvhed by sevaral medium­size (3-5 ha) and numerous small (up io 1,5 ha) ones. Tha settlements contained religious sirnaiures, or central squares, with sanctuaries, paved s*reeis and public granaries. The prin­cipal KAC domain developed in three stages, including formation on the local Eneolithic basis, maturity, and the lair stage. Laie in *he maturity stage, settlements tended in anvaen- irair along high-watao rivers. Occupation of high foothills and mountain valleys started in the same period and continued laiar. Sri*lemenis, including large ones, appeared in difficult access places, protected by natural landscape fea*uoes [Кушнарева, 1993].

According io R. Badalyan’s data [Бадалян, 1996], *he archaic KAC phase is most pre­cisely dated by ihr age nf a sample from *ha lower horizons nf Khornm (35*Є-30^шcenturies BC); its adult phase can be related in the 29^t*-26*h aan*uries BC (age nf a sample from the uppro layers of Mi^aHur), and finally, the late phase can be constrained by ihr 26th-22nd centuries BC (age of samples from Seenravii unit in Kπmui and Shengavi*) (all dates πte by 2π). The study of *he Finleiovn site in the east of the city of Vanadzor, allowed us io date ihr final (Beden) stage of KAC io 2500-2300 BC (Section 6.2.1). The aga of ihr late KAC coin­cides with the age of the eπtlier Trialeti (Beden) kuogans related io the Middle Bronze epoch.

The Trialeti culture (TC) formrd within the principal domain of the KAC ai tha boundary of ihr Early and Middlr Binnzr. Age (Fig. 107). A common fraiuor πf *ha TC and later Karmir Berd, Sevan-Uzerlic, and Kyzil Vank cultures was predominance nf burial grounds (necropolis) over settlements, *hr number and thickness of layers in which were sharply decreasing in time. This suggests decrease of population and indicates that cattle breeding played an increasingly more important oole. Among burial grounds of the consid­ered period, kurgans dominated; sometimes they were complex sitnciures and contained rich sei of implemrn*s, which aiies* io social and material siraiialaaiion of the society that was maximum ai *he final, «prosperity» stage nf the TC. Painted ceoamics of diverse styles pro­duced with potter's wheel is ano*her common feature of these cultures, along with progress in manufacturing of metal products and jewelry.

Drspiir different opinions cnvaernivg the age and aorrelaiiovs between the Middle Bronze cultures [Кушнарева, 1993; Аветисян, Бадалян, 1996] (Fig. 108), all authors agree that in the 16*h топішуBC these cultures were changed and in the middle of the 15*h crniury BC completely replaced by the Late Bonnzr Lchashen-Metsamor cul*ure. This culture is rep­resented by stnne stonatnres in settlements and cemeteries; a small pile of stona is built over each grave and ^1101^^ with a cromlech. Sometimes, cromlech surrounds several graves. We assume that having claar orientation to cattle-ereedivr in the Sevan region, the Lchashen- Metsamor culture was spreading *0 areas in *he west of Aomenia, where agriculture and var­ious crafts added to the range of its activities and where one of the richest monufvevis of this culture has preteoeeh (Meisamor). In ihr Early Iron epoch, Lchashen-Metsamor settlements were still populated, and wire hrstroyed, at least in the south, by the Uraotian conquests. To break the oesis*ancr, invaders ftom Urartu had to undeo*ake several campaigns, victorious results of which are reportrh by *he texts of Aogishti the 1^s* (787-766 BC) and Sπtdouri the 2nd (765-733 BC).

The culture of Urπotu represented a more developed society, which was reflected in building practice, quality of ceramics and the very fact *ha* a single Uraotian state came to change the population grouping around faw settlemen*s by *he tribal principle. Urπotu, essen­tially weakened by Assyrians' militaoy campaigns of Tirlatpalassar the 3^rf in 735 BC and Sargon the 2nd in 714 BC, was aruseeh in 590 by Mydia, which had defeated Assyria in 612 BC. After loosing their hegemony in Armenia, Urartians meoged with the local popula­tion *0 form an ancient Aomenian ethnic community. In 550, this community first claimed

itself to be a state, which later developed during several centuries under the conditions of actual or formal dependence on the Ac-aemeeians, and later the Seleucians.

Under the rule of Artashes the 1^st (200-159 BC) the new state became independent and spread across the entire area of the Armenia Highlands. It reached its greatest power in the reign of Tigran the 2^πd (95-56 BC), but later turned to an arena of lingering conflict between Rome and Parthians [Абаза, 1990]. In 224-226 AD, after the rule of the Parthian dynasty in Iran was overt—rown by the Sassanians, a two century-long struggle began between Iran and Armenia, and Armenia was several times occupied by Persians. In 301, king T^at the 3^rd was baptized and shortly after the whole of his country adopted Christianity. According to the treaty of the year of 387 between Feonesius t—e 1st and Sha—pur the 2∏s, western Armenia became a province of Constantinople, while the eastern one was joined to Persia ie 428. Under the Byzantine protectorate, the western Armenia gained a considerable degree of inde­pendence, but lost it in 1064 under the pressure of Osman Turks. It was freed of that heavy dependence early in t—e 19^t- century, when both parts of Airarat became part of the Russian Empire. In 1918, Armenia gained independence, and in 1920 it joined the USSR as one of its Republics. The present-day Republic of Armenia exists sincel991.

6.1.2. Archгertenical Геа^т of the ^оі^the depres­sion (Fig. 135. and 136). The southern slope of the central hill is d^mpted with a normal fault (Fig. 137). Trenching in this area revealed signs of two earthquakes, *he fiost of which occurred in the La*e KAC epoch, while the other happened somewhat later [Avagyan, 2001; Philip et al., 2002]. In tha Garni Fault Zona, sources of strong earthquakes are confined *o the identified sites of similar in-iahalon arrangement of strike-slip fault segments, стіпішіaddi­tional compression (Fig. 139-141). Correlation of epicentres with in-echelon arrangimiv* sites has been reported also foo the zona of *he right-lateral North-Anatolian fault, characteri­zed by similar n*ouciural features of *he Erzincan basin located in the aast (Fig. 142). We pro­posed the term of push-inside structures for such compressive strike-slip duplex-basins [Trifonov et al., 1995]. Their tendency tn sagging is hetermineh by higher density of the Earth's cous* in *he fault zone.

6.2.2. Almond-shaped structures. In a larger scale, *he tendency of duplex structures forming at sites of en-ёaerlon πrrangement of strike-slip segman*s is repeated as almond­shaped combination of large fault zones with a strike-slip component of motion. In the nno*h- rrn Aomeviav Highlands, we studird two almnnd-shaprd stouatures of this kind: Ghegam- Vardenis and Aoarat (Fig. 138). These stouatures are defined by faults with predominant dex- *ral component combined with a vertical one. On the northim and southern sides of *he almond structures, vertical motion component is represev*ed by reverse faults. In the Ghegam-Vardenis structure, these arr the Pame∏k-Seean fault in the north and the Arpa- Zangezour branch of *he Gaoni Fault in the south (Fig. 143). Along thr wrstern and stepped eastern rdges of the stouc*ure, wherr faults arr striking to the NNW, *he vertical component is, in places, contrastingly rrpresen*rd by normal faults. In *he wrs*eon pat* of the Gergam- Vardrnis stoua*ure, we пЄщі^єan embedded «almond contour» of Geera1n with vnoth-trend- ing chains of Late Quaternary volcanoes (Fig. 144—146).

The Ararat stouature is (difmad by binding fault zones, including the Bπlikghel fault zone in the SW (continued eastward by *he Noo*h Tabriz fault) and thr Sarda^ae-Nakhichevan Fault in the NE (Fig.. 147). In this πtea, we observe transition from dextral strike-slip faults with reviise component in *he northern and southern edges of the almond contour to dex*oal faults with normal component in its western and eastern edges. Stiaightly linrar righi-latroal strike-slip Maku fault strikes along the axis of the almond con*nur. In *he nnr*hwast part of *hr structure, the Balikghel and Maku faults branch nut into numerous vnrmal-dex*.ral faults that shape a horsetail-splay structure (Fig. 148). These faults control locations of parasitic cones on Aoarat and on *he volcanic oiCfstempfrar∏ shfyO-poyifУ variatioss. The results Siscussed above inSicate synchronism of different perioS variations in t—e manifestations of seismotectonic aeS cli­matic processes, which cannot be explained by t—eir cross-influence. This puts forward a sug­gestion about t—eir potential regulation by orbital astronomic factors, playing t—e role of trig­gers of endogenous processes. V.I. Kaftan and S.K. Tatevian [Кафтан, Татевян, 1996] sub­stantiated t—e linkage between the short-eeried level fluctuations in the Caspian and changed solar activity index and angular Earth rotation velocity (Fig. 192). Variations in the angular rotation velocity correlate with earthquake frequency changes both in large seismic region, aeS on the Earth as a whole [Горькавый и др., 1994,1999] (Fig. 194). A correlation is found between the 11 years loeg cycles (aeS longer cycles divisible by 11) of solar activity and magnetic disturbances oe one hand, and climatic variations (Fig. 193), associated phenome­na, and seismic manifestations, on the other [Чижевский, 1973; Сытинский, 1987; see also Section 3.2.4]. Therefore, at the scales of years and tens of years climatic and tectonic c—anges correlate wit— parameters of the Earth rotation, its magnetic disturbances and solar activity, which are, possibly, interrelated.

7.4.2. Mid-period variatirss is the Middle asd Late Holocose. Section 7.3 presents variations of climatic aeS seismotectoeic activity ie 12(^0-1800 years periods. By duration, these variations are similar to the century-long climatic rhythms identifies by A.V. Shniteikov, which, as he thought, could correspond to the period of constellation of the Moon, Eart—, anS Sun, Seterminiπg tidal force changes. Iπ t—e spectrum of geomagnetic field fluctuation parameters, S.P. Bourlatskaya [Бурлацкая, 1987] iSentified fluctuations in peri­ods of 1200 aπS 1800 years (Fig. 195 and Table 10), related to changes in the parameters of the Earth rotation.

7.4.3. Lisg-period variatioss is the Ploistocese. Cycles of climatic changes identified for the Quaternary period comprise 41 ka in the Early Pleistocene and 100 ka ie the later peri­od [Imbrie et al., 1984; RuSSimae et al., 1986; Bassiπot et al., 1994]. Cycles lasting for 19-23 ka, and for about 400 ka, are pronounced less clearly. Such rhythm is typical for acti­vation of tectonic processes in mountainous areas [Костенко, 1972; Макаров, 1980]. M. Milaekovich [Миланкович, 1939] was the first, who substantiated link of these varia­tions with changes of solar activity intensity resulting from changes in the orbital parameters of the Earth. His interpretations were later made more precise [Berger, 1984; Большаков, 2000]. Suc— cyclic recurrence is determined by changes of the Earth rotation parameters, including eccentricity (100 ka cycle), angle of inclination of the rotation axis to the ecliptic plane (41 ka) anS precessions - equinox advances (19-21 ka).

7.4.4. LimiOatioss of the astrosomic regulates of the variatioss. The discussed orbital-astronomic regulation of synchronism between tectonic and climatic events is assumed manifesting itself up to t—e scale of tectonic episodes (variation iπ periods of 1-2 million years). Variations in the manifestations of tectonic processes in periods of tens and —unSreSs of million years (phases of Schtille, cycles of Bertrand anS Wilson) are deter­mined by auto-fluctuations of endogenous activity of the Eart- anS influence climate changes of t—e same low frequency.

Chapter 8

HUMAN BEINGS AND the NATURE

8.1. Human Sei∏ss in 10єeosirynmeoS

Most of natural anS social systems described in t—is book belong to the class of open and dissipative systems that are capable of self-organization (self-perfection), w—ich makes t—em more complex and reSuces the entropy they bear. T—is process is realized through destruction

of primitive sys*ems and increase of entropy in the medium volume encompassing all of thisr systems [Пригожин, Стенгере, 1986]. Assuming such self-ooraniza*ion ability of open sys­tems to be the soul of the system (which is consistent with its metaphysical definitions), wr support the thesis proposed by P. Teilhπod de Chaodin [Тейяр де Шарден, 1987], stating that the spioitual elemen* is i^eoeni not only in human beings, bu* also in poimitive systems of anima*e nature and even of africaem Considering that an en*ioa series of such systems can be built by the ivareasivg rate of theio complexity (rate of order, anti-entoopy), we have no grounds to regard human community as its upper limit and thus reject possibility of more complex systems.

If we assume that the ma*erial Universe as a whole, which is a closed system by its Гегадзор, кс 2004/4 ГИН-10997 2680±30 833-805 ВС 897-801 ВС 31 6.1 Арм., Гегадзор, кс 2004/5 ГИН-11657 1090±40 898-996 AD 886-1020 AD 32 6.1 Арм., Гегадзор, кс, у 2050/3-4 ГИН-12321 3200±120 1623-1371 ВС 1744-1206 ВС 33 6.1 Арм., Ахпюрадзор, кс ГИН-10998 2580+100 827-543 ВС 897—412 ВС 34 6.1 Арм., Канагех, мкс, кости ГИН-11662 2690+40 894-805 ВС 910-799 ВС 35 6.1 Арм., Зорац-Карер, кс ГИН-11654 2220±90 386-197 ВС 413-36 ВС 36 6.1 Арм., Зорац-Карер, кс ГИН-11668 2240±60 383-207 ВС 402-166 ВС 37 6.1 Арм., Мецамор ГИН-9342 3300±360 2030-1129 ВС 2504-786 ВС 38 6.1 Арм., Мецамор, пожар, у ГИН-9340 2750±40 915-836 ВС 977-819 ВС 39 6.1 Арм., Мецамор, пожар, у ГИН-9341 2750+110 1002-802 ВС 1254-764 ВС 40 6.1 Арм., Севан, Норашен, кости ГИН-11665 7800+100 6745-6481 ВС 6953-6455 ВС 41 6.2.1 Арм., Фиолетово, кс, пп ГИН-9916 1050+100 892-1148 AD 773-1213 AD 42 6.2.3 Арм., Ванадзор, пп ГИН-11667 26000±800 43 6.3.2 Арм., Гегадзор, канава, зрв ГИН-9909 6360±140 5476-5208 ВС 5558-4986 ВС 44 6.3.2 Арм., Гегадзор, канава, зрв ГИН-9910 5400±150 4354-4043 ВС 4537-3944 ВС 45 6.4 Арм., Севан, Раздан ГИН-10995а 5540±90 4458-4263 ВС 4550-4200 ВС

Принятые сокращения: зрв - след землетрясения в зоне активного разлома, кс - культурный слой, мкс - культурный слой погребения, пп - палеопочва, у - уголь, щв - щелочная вытяжка, 1 - Trifonov et al., 1992

КАТАЛОГ СИЛЬНЫХ (Ms > 5,7) ЗЕМЛЕТРЯСЕНИЙ ВОСТОЧНОЙ ОЙКУМЕНЫ

Appendix 2

Catalog of strong (M_s> 5,7) earthquakes of Eastern Oykumena

32,38

19. Трифонов В.Г., Караханян А.С.

577

1 2 3 4 5 6 7 8 9 10 11 12 13

10,11, 13,20,

22,23, 30

13,20,23,26,

30,36, 3І