Version classiqueVersion mobile

La Cappadoce méridionale de la Préhistoire à l'époque byzantine

 | 
Aksel Tibet
, 
Olivier Henry
, 
Dominique Beyer

I. Environnement

The rise and fall of the Hittite state in Central Anatolia: How, when, where, did climate intervene?

Catherine Kuzucuoğlu

Résumé

Interested in the debate about climate and human societies relationships during the mid-Holocene in the Eastern Mediterranean (EM), this paper discusses climatic and cultural data for one territory in the EM (Anatolia), and one human society (Hittites) on embedded time scales (the IInd mill. BC, with a focus on its end) and space scales (central Anatolia, with a focus on southern Cappadocia). Results demonstrate the importance of such a multi-scaled approach, and of local-scaled data when studying the role of climate in man’s history. First, data on global and regional scales evidence that the second half of the IInd mill. BC was a generally dry period, dryness increasing after 1350-1300 BC, ending in a drought ca 1250-1100 BC. A regional presentation of sequences illustrates well that all environmental signals recorded in central Anatolia and surrounding regions parallel this global climatic trend. However, it also evidences a highly variable number of short alternations, intensities and timings when comparing the “small” environmental/ climatic regions forming central Anatolia. Comparison evidences the importance, beyond global climate and for the human societies, of the environmental variability and variety providing natural resources, all the more reason when a territory is constituted of neighbored “small” environmental systems multiplying resource availability in times of cultural or climatic stress. Besides, management of regular stress in fragile “small” territories may have preserved practices and production from drought accentuation, thus possibly explaining sub-regional cultural continuities and rapid neo-Hittite renewal in such territories (eg. in southern Cappadocia). Finally, archaeological/historical state of the art during the last 150 years of the Hittite Empire (1320-1170 BC) evidences that (i) internal tensions, whether political, religious, dynastic, military, and consequently economic, appear early, ca 1300-1280 BC; (ii) the Empire seems to have performed successful deeds at the end of the 2nd mill. BC; (iii) the narrative of its end between 1190-1170 BC remains veiled by archaeological and historical silence (eg. no “destruction by the Sea People”, as still too often read in the literature). In conclusion, climate degradation which paralleled cultural disorders in the EM acted in Anatolia as an external element allowing, in the context of the disruption of the EM ‘world’ trade, the end of a very (too much?) centralized system and the birth of new cultural (innovative, eg. The Beyşehir Occupation Phase) systems.

Texte intégral

This paper results from more than two decades of research in Anatolia – especially in the Konya plain and in Cappadocia – and of discussions with archaeologist colleagues and friends in Turkey and France, and from other countries (Italy, UK, US, Germany etc.). First, my deepest thanks and reconnaissance go to late and fondly remembered Prof. A. Dinçol. Together with him, some of my colleagues urged me very early to question the Hittites and their vanishing reality (e.g. E. Jean, K. Köroğlu, H. Genz). This vieille garde and also A. Tibet, C. Marro, O. Henry, M. Godon, the Hittitology scholars at Istanbul around Prof. A. Dinçol, the late O. Pelon, D. Beyer and the Porsuk team, and a nouvelle garde in the Bor and Çiftlik plains, formed by L. d’Alfonso, A. Gürel, A. Matessi… all accompanied me in the exciting exploration of Cappadocia and Central Anatolia, especially with regard to the IInd and Ist mill. BC. Thanks to all of them, I much enjoyed here to question scientific topics in the history of the Hittites.
Regarding financial supports, some of the scientific production used here has been financed by the CNRS (mainly the Mistrals/PaléoMex/ ArchéoMed programme), the Laboratoire de Géographie Physique (UMR 8591), the IFEA (French Institute of Anatolian Studies), excavations financed by the French Ministry of Foreign Affairs (eg. Porsuk), many Turkish institutions (MTA, Tübitak, DSI, Köy Hizmetleri), several Universities such as Istanbul and Niğde, and the Turkish Ministry of Culture (including Museums) and their partners in Turkey (Aşıklı Höyük, Öküzini), Italy (Kınık Höyük), UK (Kerkenes), Germany (Tecer). My deep thanks go to Olivier Henry, Aksel Tibet and Martin Godon from the IFEA who, trusting the interest of the present article, have been very supportive, patient and much helpful.

L’effondrement de l’Empire hittite, vers 1200, s’accomplit silencieusement, avec moins de bruit qu’un château de sable qui s’affaisserait sur lui-même.
Et l’on n’aperçoit pas les responsables. Une trentaine d’années plus tôt, vers1230, les palais mycéniens avaient presque tous été détruits, de nombreuses villes abandonnées sur le continent grec et sur certaines îles. Et là encore, pas de responsables visibles : les accusés d’hier, les Doriens, n’arriveront qu’à la fin du XIIe siècle, cent ans plus tard au moins. Quant aux Peuples de la Mer, personnage central de ces temps apocalyptiques, nous ne les voyons vraiment qu’au moment où, par deux fois (vers 1285 et 1180), les Égyptiens les écrasent.
Parallèlement une période longue de sécheresse tourmente la Méditerranée à la fin du IIe millénaire. Ce dernier personnage : le climat, serait-il le plus important de tous ?
Braudel 1969 (1998), 256

Introduction

  • 1 Kuzucuoğlu et al. 2011.

1In Anatolia, between the climatic Optimum (ending before or during the Vth mill. BC) and the Iron Age (first half of the Ist mill. BC) that preceded the Roman Warm Period (RWP), the Mid-Holocene transition exhibits three dry occurrences marked by intense drought spikes: (i) ca 3150/3050 BC; (ii) ca 2300-1900 BC; (iii) ca 1300-950 BC1. The chronology of the third drought raises the question of the relationships between climate conditions and the history of the Hittite Empire, especially during the last decades of the 13th century BC, before and during the end of the Empire. Questioning the relationships between climate and society is indeed a matter of vivid debate in the scientific community, with proposals to link directly climate and human history, featuring catastrophes and civilization collapses that climate change would have triggered in ancient and recent history. Such pictures are often drawn on the main basis of synchronicity between “Rapid Climate Changes” (RCC) and short periods of cultural changes (transitions or collapses), especially when the latter appear to have occurred “world-wide”. Regarding the Eastern Mediterranean (EM), such direct links are evoked for explaining the ends of EBA and LBA region-wide. In this discussion, data used for evidencing the relationship between society and climate factors are often presented in a bipolar approach organizing a face-to-face confrontation that poorly takes into account a complex suite of processes and events which may additionally be poorly known on the local scales.

2Indeed, careful reviews of published data about climatic and archaeological/historical contexts show that the chronology of events (whether climatic or historic) is not well established, especially on the local and smaller regional scales where cultural and climatic variability are expected to impact the timing of events (e.g. with or without delay). Thus, the synchronism must vary according to this time-variability and to the differences in regional responses of environmental systems to climate changes. In addition, uncertainties rise also from the various chronologies used (including historical ones). Consequently, the understanding of the climate-society relationships makes it necessary, if not obligatory, to collect first specific data concerning specific territories and specific timings. Especially, understanding the roles of multiple factors in the collapse or transformation of ancient societies need detailed local/regional approaches for characterizing and reconstructing the succession of events in which climate and related environmental systems intervene, but in which other elements of a cultural and historical nature intervene also.

  • 2 E.g. Weiss et al. 1993.
  • 3 Kuzucuoğlu/Marro 2007 ; Kuzucuoğlu 2012 ; Leroy-Ladurie 2013.

3The pluridisciplinary approach necessary for such a study addresses two main scientific domains: palaeoenvironments and palaeoclimatology on the one hand, archaeology and history on the other hand. Issues from each domain generate their own sets of information and questions which, when integrated in the dialog between archaeology and palaeoclimate, appear often difficult to handle by the specialists of one another domain. Nowadays the scientific interest of this dialog is increasing because the debate now concerns the crucial question of climatic determinism in history, eventually applied to our contemporary epoch. Indeed, climate is increasingly put forward as the cause2 or an additional cause3 of dramatic cultural changes that occurred in the past. In this debate, Anatolia is an interesting field of study because its geographic characteristics provide a very high variety and variability of contrasts in climate and environment, i.e. a high variety of resources on the local and regional scales (fig. 1) and a high variety of possible environmental responses to climate change. Making use of this geographic variety and environmental variability, the present paper addresses the possible role of climate in the history of central Anatolia (including southern Cappadocia), comparing (i) records of past global and regional climate and local environments; (ii) historical and archaeological records during the 2nd mill. BC when the Hittite Kingdom and Empire appeared and developed; with (iii) an ultimate focus on the sudden -and yet rather unexplained-end of the Hittite Empire ca 1190 - 1170 BC or shortly after.

Fig. 1: Physical Geography of Turkey

Fig. 1: Physical Geography of Turkey

A. Relief; B. Climatic regions; C. Vegetation formations

A: http://commons.wikimedia.org/​wiki/​File:Turkey_topo.jpg; B: Meteorological Service of Turkey; C: van Zeist/Bottema 1980

Fig. 2: Palaeotemperature curves from GISP2 ice cores in northern Greenland

Fig. 2: Palaeotemperature curves from GISP2 ice cores in northern Greenland

A. Temperature (C°) curve based on stable oxygen isotopes analyses; B. Snow surface temperature reconstructed using N and Ar isotopes in gaz trapped in ice

A: modified from Alley 2004; B: modified from Kobashi et al. 2011

1. Environmental records of late Holocene climate changes: Space and Time scales

1.1. The Global scale

  • 4 Cuffey/Clow 1997.
  • 5 In ice: Alley 2004; and air bubbles: Kobashi et al. 2011; see also Severinghaus et al. 2003.
  • 6 δ18O is the isotopes 18O/16O ratio; δ15N is the isotopes 15N/14N ratio; δ40Ar is the isotopes 40Ar/(...)
  • 7 Due to unsteady CO2 content in atmosphere through time, all 14C calibrated dates present an uncerta (...)

4Climatic signals recorded in the GISP2 ice cores from central Greenland provide global climate references for the northern hemisphere4. Considerable amount of work by different scientific communities have demonstrated that stable isotopes and other components in these ice cores (as well as in marine cores not used here) are confident proxies recording the evolution of global climate. Figure 2 features two records of climate trends on the global scale: one record concerns the 4 ka to 0 ka BC time-span; the other one only the 2ka to 0 ka BC time-span. Both curves are based on measurements of isotopic analyses in ice and air bubbles from the GISP2 cores5. In spite of being obtained from the same archive (GISP2 ice cores), the trends of 2A and 2B curves are somewhat different both in chronology and magnitude. They are published by two teams who integrate differently the snow accumulation rate and use different proxies (δ18O in 2A; δ15N and δ40Ar in 2B6) analyzed in two different environments (ice; air gases) with two different methods for the calculation of the bore-hole constraints, i.e. for integration of time. The comparison of these curves shows that trends are similar, with differences in the details of the chronology and intensity of changes which suggest that there might be an uncertainty also on the timing and intensity of the global RCCs (see shaded bands in fig. 2). Consequently, the synchronicity of climate and historical records seems to combine several uncertainties attached to the dating methods used on palaeoclimatology, palaeoenvironments and archaeology (14C dates7, comparative chronologies based on archaeological material… and as isotope ratios in ice and marine cores). The resulting uncertainty can be today meanly evaluated ca ± 50 years thanks to the increasing resolution of several types of records.

Millennium-scaled RCCs during the mid-Holocene on the global scale

  • 8 In this paper 14C calibrated BP dates are presented as BC dates. In the curves presented in figs. 4 (...)
  • 9 E.g. Kuzucuoğlu 2009 ; Kuzucuoğlu et al. 2011 ; Roberts et al. 2011.

5On the global scale, the two millennia corresponding to the Bronze Age societies in Eastern Mediterranean (IIIrd and IInd mill. BC) belong to a climatic transition that spans from the IVth to the Ist mill. BC8. During these 4000 years, four millennium-scaled warm phases occur which concern ca the last 250-200 yrs of each millennium (fig. 2A). These signals are well known from several other researches in the Eastern Mediterranean and Anatolia9, where they are recorded as dry phases and drought successions dated (i) ca 3150-2950 BC; (ii) ca 2250-2150 BC; (iii) ca 1350-1050 BC; (iv) ca 200-0 BC. The warm phase centered around 1250 BC (iii) is the most intense and the longest of these four RCCs (fig. 2A).

6Superimposed to these millennium-scaled short warm phases, another cycle seems responsible for two additional warm phases occurring at the end of the 1st half of two millennia ca 3700-3550 BC and ca 1750-1650 BC, with the possible addition of another warm phase ca 500-400 BC (curve 2B).

The second half of the IInd mill. BC and the beginning of the Ist mill. BC on the global scale

7Regarding the sole IInd mill. BC, curves 2A and 2B show that:

  • (i) The global image of the IInd mill. BC is that of a rather warm millennium, with the highest temperatures culminating after 1500 BC;

  • (ii) The first 200 yrs are rather cold in curve 2A, while curve 2B details warm/cold period alternations during the first five centuries.

  • (iii) The second half of the II nd mill. BC is globally warm, temperatures highly increasing towards the end of the millennium with peaks interrupted by two or three short cold episodes;

  • (iv) A time-delay seems to distinguish both curves. In curve 2B, the warm episodes are concentrated during the second half of the millennium BC and extend into the Ist mill. BC until ca 950 BC; in curve 2A, only after a first warm episode during the first half of the IInd mill. BC, do the whole second half of the IInd mill. BC become increasingly warm, the ultimate drought peak ending before the start of the Ist mill. BC.

1.2. Comparing the global curves and the eastern Mediterranean (regional) climatic curves

  • 10 Kuzucuoğlu 2009.
  • 11 Kuzucuoğlu et al. 2011 ; Roberts et al. 2011.

8The comparison between the global signal and those evidenced in the Eastern Mediterranean (EM), shows that the millennium-scaled mid-Holocene events occur also in the EM. The timing of these events varies however from one wide region to the other10. There is an evident synchronism between these climatic signals and important cultural transition phases in the EM. This synchronism raises the question of the role of climate in these cultural transformations and accompanying historical events at (i) the end of the IVth mill. BC: Final Chalcolithic/ EBA transition in Anatolia; (ii) the end of the IIIrd mill. BC: end of EBA, transition to MBA; (iii) the end of the IInd mill. BC: end of LBA, transition to IA; (iv) the end of the Ist mill. BC: the start of the so-called ‘Roman Warm Period’11. Three observations temperate this more or less obvious synchronicity:

    • 12 E.g. Kuzucuoğlu/Marro 2007 ; Kuzucuoğlu 2009 and 2012 ; Roberts et al. 2011.

    Historical as well as climatic records from the different sub-regions of the EM (southern and northern Levant, Syrian lowlands and mid- Euphrates valley, Taurus piedmont and the Mesopotamian territories, Anatolian highlands and plateaus, Aegean and Mediterranean countries) evidence a high regional-scaled variability both in chronologies and in environmental records vs cultural behaviors during the time-scales of the RCCs12.

    • 13 Kuzucuoğlu 2010 ; Kuzucuoğlu et al. 2011.

    The climate instability at the ends of the IIIrd and IInd mill. BC starts in the EM ca 2500 BC (during the EBA) and ca 1350 BC (during the LBA), i.e. before the droughts which numerous researches link to EM “civilization collapses”13.

    • 14 Kuzucuoğlu 2009.

    The mean duration of the most intense droughts is ca 50 to 100 yrs. These droughts occur within a longer dry phase corresponding to a global warming signal, but the droughts are interrupted by wet sub-phases. The number of these alternations depends on the sensibility of the region14.

1.3. Anatolia: Geographic diversity and climatic contrasts

9Regional climate curves referencing Anatolia are provided by (i) cores in regional seas (e.g. Eastern Mediterranean sea, Black sea, Red sea, Persian Gulf); (ii) continuous pollen and multi- proxy records retrieved from seas, lakes, marshes and speleothems. The precipitations over the mountainous barriers circling the peninsula provoke a high humidity contrast with orographic shadow-areas positioned at the inner foot of these coastal ranges. Behind this topographic obstacle indeed, the humidity gradient decreases with the distance from the sea, favoring continentality (humidity deficit) on the central plateaus (fig. 3). Differences in latitude also partly explain some differences between continental sites with regard to their response to global climate change, as sites are impacted by their location in a specific climatic zone. Last but not the least, the different components of environmental systems (e.g. and valley floors, underground water networks, lake and river basins, coastal zones…) and the relationships between these parts, introduce local diversity to the environmental responses to global climate changes.

Fig. 3A: Precipitation distribution of Turkey. Distribution map of average annual precipitation (mm/yr)

Fig. 3A: Precipitation distribution of Turkey. Distribution map of average annual precipitation (mm/yr)

Note: Numbers correspond to the profiles below.

A. Meteorological Service of Turkey

Fig. 3B: Precipitation distribution of Turkey. Four West-East cross-sections of relief (upper graphs) and precipitation (lower graphs)

Fig. 3B: Precipitation distribution of Turkey. Four West-East cross-sections of relief (upper graphs) and precipitation (lower graphs)

Horizontal shaded bands: (a) dark grey: annual P < 400 mm/yr; (b) light grey: annual P= 400-500 mm/yr Vertical shaded bands: (a) dark grey: Central Anatolia; (b) light grey: other dry regions in Anatolia
Note: Stations selected along profiles are located in towns. Consequently, no data concerning mountains crossed by the profiles in fig. 3A are present on the graphs.

  • 15 Kuzucuoğlu 2009 ; Kuzucuoğlu et al. 2011.

10An additional variability is introduced by the time-changing territorial impact of the four air systems entering the peninsula. These four different atmospheric systems are mainly (a) the west-east Mediterranean cyclonic trails and (b) the Siberian High-related air loaded by Black Sea humidity, with incursions of (c) western European temperate and cold air and of (d) Indian summer monsoon regime sometimes reaching the eastern and southeastern mountains. With time and under the influence of global climatic changes, the spatial distribution of these climatic systems over the peninsula (ie of their impacts as far as seasonal and yearly characteristics of temperature, precipitations and humidity are concerned), also changes, affecting the dynamics of water- and temperature-controlled regional and local ecosystems (river valleys, forests/herbs/wetlands etc…)15. Consequently, the characteristics, intensities, and/or durations of climate phases recorded by palaeoenvironmental sites in the peninsula, vary according to the regional distribution of climates which may have been different from today’s, and according to the type of recording ecosystems.

Regional sensitivity to climate change

  • 16 P = Precipitation (rainfall, snowfall).
  • 17 The P limit of 250 mm/yr is also the conventional limit for dry farming practices; below 250 mm/yr, (...)

11The center of the Anatolian plateaus is formed by the Konya plain and southern Cappadocia where annual precipitation varies between 280 and 340 mm/yr (fig. 3B; Table 1). With the limit of semi-aridity conventionally placed at 250 mm P/ yr1617, the Konya plain and its surroundings form a territory which is very sensitive to whatever change of climate triggering a decrease or increase of the average amount and/or the interannual/seasonal distribution of precipitation. In this context, palaeoenvironmental records in the driest Anatolian plateaus have a high probability of recording small and early changes heading to drier (even with a small P decrease) or wetter (even with a small P increase) phases. Moreover in case of droughts, the environmental dynamics may lead locally to stability (no record) in the absence of morphological agent and vegetation cover. Meanwhile, some regions around this center receive today an average annual precipitation 250 ≤ 450 mm/yr (Table 1; fig. 3B). Environmental systems in these regions will also respond rapidly to past climatic changes, although later than in the driest areas and with a lesser sensitivity tempering at places the intensity of impacts.

Table 1: Average annual precipitation (mm/yr) at selected meteorological stations in central Anatolia

Southern Cappadocia

Northern Cappadocia

Konya Plain

Lake district + Mediterranean

North-central

Peat Wetlands

Aksaray

344

Eski Acıgöl (lake)

Acıgöl

378

Palaeosols

Konya

312

 

Beyşehir

537

Eğri dere

Yozgat

611

Niğde

339

Nevşehir

397

Karapınar

289

 

Isparta

501

Sivas

452

Bor

346

Kayseri

400

Karaman

314

Gölhisar

Burdur

424

Lake Tecer

Tecer

420

Ulukışla

345

 

 

Eregli

337

Öküzini

Antalya

1077

Ulas

416

Çiftlik

349

 

 

 

 

 

 

 

 

 

 

Grey shading and bold lettering point to meteorological stations associated to palaeoenvironmental sites referenced in the text and in figs. 7 and 8

12Due to this sensitivity to global change toward drier or wetter trends, the climate in central Anatolia during the Late Holocene is framed, within the high regional diversity of the peninsula, by (i) the global climatic trends and changes impacting the peninsula during the period; (ii) the specificities of local and regional climate conditions in the central plateaus where dryness increases with the distance to the seas and with the extent of the climatic systems; and (iii) the more or less high sensitivity of the environmental systems toward climate changes.

2. Late Holocene climate records in Central Anatolia

2.1. Environmental variability of past climate sequences in Central Anatolia

13The main environmental systems in central Anatolia, used here for illustrating past climate and environment evolution, are: (i) lakes (Tecer, Eski Acıgöl) and marshes (Bor plain, Eğridere marshes, Öküzini); (ii) soils over run-off or wind deposits (Konya plain); (iii) river fills (Altunhisar alluvial fan near Bor, valley fills of the Eğridere and its tributaries near Yozgat, Euphrates terraces near Birecik) (fig. 4). All types are dependent on local factors (geological, dynamics) and on human activities impacts which may have modified them. The chronology is based on as many 14C dates as possible. These records are either continuous or discontinuous, the latter ones being rarely published although they bring information useful for the understanding of the evolution of environmental changes.

Fig. 4: Climatic phases in central Anatolia between 4000 and 0 BC

Fig. 4: Climatic phases in central Anatolia between 4000 and 0 BC

A. Location of palaeoenvironmental sites and records used in Figure 4B, placed on the map of annual precipitation distribution in Turkey.
Legend: Sites on map: 1= Tecer Lake (Sivas); 2= Eski Acıgöl (Nevşehir); 3= Öküzini (Antalya); 4= Eğridere (Yozgat); 5= Karaman and Ereğli (Konya plain); 6= Bor plain (Niğde)
Notes: Ref. average annual precipitation amount at the sites in Table 1.
B. Regional variability of climate changes in central Anatolia and some surrounding areas according to palaeoenvironmental records in closed depressions and watersheds.
Legend of fig. 4B: Grey shading shows increasing-decreasing dryness. Plain grey bands on synthesis show drought peaks.
Note: For detailed reconstruction proposal for central Anatolia in Figure 7

see down each climatic log

14The comparison of these examples (fig. 4) illustrates the influence of geographic variability and of the type of recording environmental systems. It shows also that the climatic interpretations of the palaeoenvironmental proxies and sequences must take into account both this double variability.

2.2. Presentation of climatic sequences from Central Anatolia

2.2.1. Lake sequences (continuous records)

  • 18 Respectively Kuzucuoğlu et al. 2011 and Roberts et al. 2001.

15Lake sequences can be compared to global and regional sequences if their dating is well constrained, even when hiatus occur (Tecer, Eğridere spring marshes, Öküzini marshes before the IIIrd mill. BC and after the Ist mill. BC). In central Anatolia, main available sequences for the IInd mill. BC are Lake Tecer and Eski Acıgöl18; discontinuous records are also available in the Konya and Bor plains.

Lake Tecer (northern plateaus of central Anatolia)

  • 19 Kuzucuoğlu et al. 2011.

16The Lake Tecer record is based on cores retrieved from a 6m deep lake south of Sivas. The sequence integrates global climatic signals and regional climatic components such as the temporary impacts of Black sea/Siberian-High atmospheric system when it enters this region19. In this record (fig. 5), climate trend is reconstructed from sediment components such as (i) aragonite, calcite and gypsum contents; (ii) proportion of coarse particulates in fraction <2mm; and (iii) sedimentation hiatuses evidenced on the basis of grain-size and content in eroded gypsum grains. The time-model is framed by eight 14C dates on pollen. Evaporation increases the concentration of dissolved carbonates in lake water. When reaching a given concentration, the carbonate species (calcite, aragonite, dolomite) and gypsum crystallize successively as a response to increasing evaporation stress. Here (fig. 5), both the occurrence and increase of aragonite content in the sediment respond to humidity depletion which eventually ends in the crystallization of gypsum, also measured in the record. Coarse particulates content at the core spot signals a low lake level yet fed by runoff on the slopes of Tecer Mountains. This erosion input to a shallow lake indicates a phase characterized by seasonal P contrasts (winter snow; irregular rainfall). Occasionally, desiccation of the lake (hiatus) indicates drought events. Thus, the Lake Tecer sequence records the regional humidity (evaporation intensity). The comparison with the temperature changes recorded in Greenland ice (fig. 6) shows a striking parallel between the drying trends and droughts at Tecer on the one hand, and the warm periods indicated by the global curves on the other hand. This is true both for the comparison with curves produced by Alley (2004) and Kobashi et al. (2011).

Fig. 5: 4000-0 BC climatic record at Lake Tecer (Sivas)

Fig. 5: 4000-0 BC climatic record at Lake Tecer (Sivas)

Modified from Kuzucuoğlu et al. 2011

Fig. 6: The Lake Tecer dry phases sequence, compared to GISP2 temperature records (4000-0 BC)

Fig. 6: The Lake Tecer dry phases sequence, compared to GISP2 temperature records (4000-0 BC)

A: modified from Alley 2004; B: modified from Kobashi et al. 2011; C: Kuzucuoğlu et al. 2011

17At Tecer during the IInd mill. BC, four dry spikes are recorded, all rising and ending sharply. Three of these spikes occur during the second half of the IInd mill. BC which is itself a dry phase (as on the global scale). First three spikes are dated ca 1750- 1650 BC, ca 1550-1450 BC and ca 1400-1350 BC. The fourth increase occurs ca 1300 BC; after ca 1250 BC this trend becomes the longest and most intense drought of the IInd mill. BC, with a peak ending ca 1100-1050 BC.

18These peaks correspond to Alley (2004)’s warm periods ca 1800-1750, 1600-1500, 1450-1400 and 1300-1100 BC (figs. 6A and 6C). The timing difference between Lake Tecer and GISP2 records

Fig. 7: Proposed chronology of wet/dry climate alternations during the IInd mill. BC, interpreted from palaeoenvironmental sequences in Central Anatolia

Fig. 7: Proposed chronology of wet/dry climate alternations during the IInd mill. BC, interpreted from palaeoenvironmental sequences in Central Anatolia

1. Humid phase; 2. Wet phase (transition); 3. Dry phase; 4. Drought phase; 5. Regional dry signal.
Note: The chronology presented here is to be taken with cautious due to uncertainties in the dating of sequences, which reach ca 50 years. For sequences in the Bor plain, chronology published here will be ascertained by further dates. The two options presented here, and the question marks associated to some phases, testify for the going-on work

Kuzucuoglu 2007 ; Kuzucuoglu et al. 2011 ; Kuzucuoglu/Gürel et al. 2014 ; Roberts et al. 2001.

19may represent the 50 yrs dating uncertainty in the Lake Tecer record, or the expected time-delay in the responses of continental environments to global changes (see above). In Kobashi et al. (2013), the temperature curve shows a 200 yr-long dry phase between ca 1500-1350 BC which corresponds to droughts nos 2 and 3 in Lake Tecer (figs. 6B and 6C). In addition in this curve, the dry phase ending the millennium occurs later, longer (1230-950 BC), and is interrupted by a short humid phase ca 1120-1050 BC before drought starts again for another 100 yr-long episode. This double drought at the end of the IInd mill. BC is also reported in some archaeological sites on the EM coast.

20In summary, the Tecer curve indicates that, after a 150yr-long humid phase (1900-1750 BC), the climate in central Anatolia during the whole period of Hittite rise and expansion remains dry, with only three short humid phases between 1650-1550 BC, 1450-1400 BC and 1350-1300 BC. Thus, the central Anatolian regions lived under dry conditions during most of the IInd mill. BC, with a few decades -long additional droughts ca 1680-1650 BC, 1530-1490 BC, 1380-1350 BC. The longest dry phase is bracketed ca 1250 and 1050 BC. At the end of this phase, a brine ca 1100-1050 BC formed in the Tecer depression as the result of excessive summer evaporation impact on a shallow lake formed during winter/spring. After 1050 BC precipitation rose again, favoring only seasonal runoff on mountains while summers remained still dry. During this time, eroded sediment was brought to the depression by streams fed by winter snow melt. After 850 BC, humidity increased in north-central Anatolia, and this humid phase lasted for several centuries.

Eski Acıgöl (northern Cappadocia)

  • 20 Bottema/Woldring 2001 ; Roberts et al. 2001. New multiproxy lake sequences in Cappadocia are awaite (...)

21The multiproxy (pollen, minerals, isotopes) sequence from Eski Acıgöl crater lake near Nevşehir provides today the sole published continuous (no hiatus) climatic and environmental record in Cappadocia20. This sequence presents a great interest for our purpose for it defines trends in climate (humidity, evaporation) and environment (pollen, water availability and quality). However, the dating of the sequence presents some uncertainty because of 14C oldering by volcanic gas input in the crater water (the Late Holocene volcanic activity is also indicated by a high water content in talcum).

22In these conditions and in spite of additional U-Th dates on carbonates, the chronology is still low- resolution. Consequently, the precise (≤ 50 yrs) dating of ‘short’ events in the sequence during the period remains somewhat loose. Keeping in mind this 50 yrs uncertainty, we suggest below a climatic scheme for the period from the IIIrd to the Ist mill. BC. The evolution (figs. 4B and 7) is based on the carbonate species which record evaporation intensity and water budget. In this reconstruction, the first part of the IIIrd mill. BC is warm and rather wet, with a limited impact of evaporation. A sudden increase in evaporation (higher T, lower P) ca 2350-2250 BC initiates an alternation of a wet episode (ca 2250-2150 BC) and a drought dated ca 2150-2100 BC. After 2000 BC, warm conditions associated with seasonal availability of humidity last until ca 1000 BC. During this warm/humid phase, drought episodes occur ca 1750-1650 BC, and 1200-1150 BC.

23These peaks are interrupted by wet (also probably cold) episodes ca 1950-1850 BC and ca 1550-1350 BC. After ca 1000 BC, the climate recovers more humid (seasonal) conditions with warm and dry (evaporative) summers. After ca 900 BC, droughts do not occur again but seasonal evaporation is still recorded, in spite of wetter conditions (temperate climate?).

2.2.2. Other sequences (discontinuous records) in Southern Cappadocia and Konya plain

24Other sequences provide periods of distinct environmental systems, sometimes occurring as short-lived wet (soils) or dry (dunes, desiccation) signals interrupted by changes in environmental dynamics in systems such as river fills, terraces and fans (sedimentation, erosion; environmental stability etc.).

  • 21 Kuzucuoğlu et al. 2014.
  • 22 D’Alfonso et al. 2010.

25The Bor plain (southern Cappadocia) is located at the southern foot of the volcanic Hasandağ- Melendiz range. It constitutes the easternmost part of the Konya plain and a marginal region where all environments are transitional from dry (closed depressions at ca 1050m altitude) to wet (mountains up to 2000-3000m high) environments. In-between environmental systems (alluvial fans, marshes, springs) along the area where low depressions meet high mountains are very sensitive to even a slight climatic change. In this region, spoted data in Gürel & Lermi (2010) and Ballatti & Balza (2012) are completed by unpublished results obtained in 2013-201421, collected in the frame of the Kınık Höyük archaeological research program22. These results concern different components of the local environmental system. They enlighten climatic phases during the IInd and Ist mill. BC. Radiocarbon dates are yet too few for constraining the time- windows of these different climatic phases; the few dates of evidenced climatic phases, which are available today suggest however a climatic succession that can be compared to those of Lake Tecer and Lake Eski Acıgöl (figs. 4B and 7).

26The IInd mill. BC witnesses first (i) a seasonally wet phase; second (ii) a humid and temperate (lake) phase; ending in (iii) a drought marked by a general desiccation of the low (impermeable) depressions which, impermeable, concentrate both rainfall water and the water discharged by streams fed by springs and runoff (precipitation, snow melt). We consider here that this drought (accompanied by decrease in water discharge) corresponds to the end of IInd mill. BC. Today available 14C dates bracket this ‘wet-dry-desiccation’ succession younger than ca ≤ 2100 BC and composed (i) first of alluvial-torrential growth (channel incision, torrential fan advances) at the beginning of the IInd mill. BC producing watered areas around the fan (first half of the IInd mill. BC?); (ii) followed by a water decrease producing both some erosion of thin particulates and their accumulation in evaporative depressions (second half of the IInd mill. BC?); (iii) ending in a drought (affecting all watered environments) which may date end of the IInd mill. BC, as expected from all other — regional and global — curves).

  • 23 LGM= Last Glacial Maximum= 28-17kyrs cal BP (ie 26-15 kyrs BC). A LGM lake covered the Konya plain (...)
  • 24 Kuzucuoğlu et al. 1999a and 1999b.
  • 25 Kuzucuoğlu 2007.
  • 26 Optically Stimulated Luminescence dated dunes: Kuzucuoğlu et al. 1997 and 1998.
  • 27 One U-Th aged and two 14C dated marshy sediments and forest-soils: Kuzucuoğlu et al. 1999a; Kuzucuo (...)
  • 28 Fontugne et al. 1999, 583 ; Kuzucuoğlu 2007.

27In the Konya plain, LGM23 lake marls are unconformably and partly covered by Late Glacial and Holocene non-lake sediments such as silty colluvium, run-off transported thin sands, marsh grey clays, backswamps deposits, alluvial fans and aeolian sands. During Early Holocene, there were some wetlands on the plain floor, such as Lake Akgöl near Ereğli which was cored by Bottema & Woldring (1984). Their Adabağ diagram shows the decrease in humidity (decline of Pinus, increase of Artemisia) that ends in the strong desiccation signal ca 6000 BC, of which the lake never recovers afterwards. After 5000 BC however, in the closed depressions north and south of the Konya plain, fresh-water lake and marsh phases respond to humid conditions during the Vth mill. BC and IVth mill. BC24. In addition, 14C dates from soils around the depressions confirm humid conditions in the mid-IVth mill. BC25. These Vth-IVth mill. BC temporary wetlands disappear with a severe drought between 3600 and 2700/2500 BC during which wind reactivates sand dunes near Karapınar26. This drought may correspond to the RCC often evidenced in Anatolia between 3.3/3.1 and 3.0/2.9 ka BC. During the following first half of the IIIrd mill. BC (2.8 to 2.5 ka BC), renewal of marshes and humid conditions are evidenced again in distinct depressions in the Konya plain27. Regarding the IInd mill. BC, humidity is only signaled by a soil 14C dated ca 1500 BC (fig. 4B). It is delivered by a soil developed over silty colluvium and followed by bedded thin sands deposited by surface (seasonal) run-off in a quarry near Kilbasan (Karaman area)28. This review finally shows that, after a humid Early Holocene, (rare) humid / (longer) dry alternations occurred in the Konya plain until ca 1500 BC. Afterwards, the climate seems dominated by dry conditions, like in the Bor plain (see below).

2.3. The Beyşehir Occupation Phase (BoP): a 1500 yr-long, non-climatic and continuous signal

  • 29 Akkemik et al. 2012.
  • 30 Eastwood et al. 1999.

28In spite of the climatic variability during the generally dry context of the second half of the IInd mill. BC, the pollen compositions in central Anatolia do not record any specific signal pinpointing a climatic change. After a strong deforestation signal ca 1500 BC29, man’s imprint on vegetation is indeed ever so important that all further change in vegetation composition illustrated by results of palynological studies, is to be interpreted as human- caused30.

  • 31 Thirgood 1982.

29In 1980, pollen assemblages from SW, W and central Anatolia allowed van Zeist & Bottema (1980), as well as Bottema & Woldring (1984), to define a mixed production associating crop cultivation (Cerealia-type and adventice plants, grasses), animal husbandry (pastures and pastured forests), together with the expansion of specialized fruit-tree cultivation (mainly Olea, Juglans, Fraxinus, with also Corylus, Pistacia, Platanus and Castanea) and fruit-bearing other species (e.g. Vitis). They called this unique agricultural landscape, the ‘Beyşehir Occupation Phase’ (BOP) which is composed of a mosaic of orchards, mixed crops-fruit trees areas, and high animal production in pastures. The reconstructed patchwork of agricultural terrains and landscapes shows a high, conservative and sustainable management and selection, by human societies, of plant species cultivated in rural territories where mosaics of water, soil and forest ecosystems provide natural resources for agricultural production. It also indicates a high anthropogenic impact on the regional vegetation31.

  • 32 Eastwood et al. 1998.
  • 33 Gauthier et al., in preparation.
  • 34 Dörfler et al., in preparation.

30Figure 8 presents the location and duration of an agricultural landscape recorded by pollen assemblages of several sites in Anatolia, from the southwestern to the north-central regions32 up to Cappadocia33 and the Sivas area34. In each pollen record, the association varies in species and in amount of pollen presence, so that there seems to be a high regional variability of the BOP composition which is adapted to local natural and cultural constraints. Besides, archaeological data show that this anthropogenic landscape contributed also to sustain urban settlements and maximize agricultural output in Anatolia during 1500 years. This lasting imprint of man’s agricultural activities transformed deeply the vegetation landscape of Anatolia and altered the natural landscape irreversibly.

  • 35 Such as during the 7th century AD: Haldon 2007.

31Starting at various dates during the second half of the IInd mill. BC (i.e. during the Hittite Empire) and during the end of the IInd mill. BC/beginning of the Ist mill. BC (i.e. during Early Iron Age), the initiation of the BOP system occurred resulted in several places from innovative practices introduced under the rising rule of local elites, and/or from the influence of population movements in Anatolia. During Iron Age and Classic times, the BOP phase flourished until Roman/Early Byzantine periods. It stops suddenly between ca AD 300/400 – AD 550/650 AD (fig. 8), most presumably because of the influx of pastoral elements and destructive attacks35. Thus, the BOP landscape does not end with any climatic change during ca 1500-2000 yrs.

  • 36 for definition of the BOP, see Bottema/Woldring 1984; Eastwood et al. 1998.

Fig. 8: The Beyşehir Occupation Phase36 (BOP) in Turkey. A. Location of sites with pollen studies corresponding to the period from 2000 BC-AD 1000

Fig. 8: The Beyşehir Occupation Phase36 (BOP) in Turkey. A. Location of sites with pollen studies corresponding to the period from 2000 BC-AD 1000

Sites references : 1. Köyceğiz ; 2. Gölcük ; 3. Gölhisar ; 4. Elmalı ; 5. Avlan ; 6. Söğüt ; 7. Öküzini ; 8. Pınarbaşı ; 9. Gravgaz ; 10. Beyşehir ; 11. Abant ; 12. Melen Gölü ; 13. Yeniçağa ; 14. Nar Gölü ; 15. Eski Acıgöl ; 16. Cora; 17. Suppitassu; 18. Tecer; 19. Hoyran; 20. Karamık; 21. Ova; 22. Van
White number on black background= Sites with Beyşehir Occupation Phase // Black number on white background= Sites without Beyşehir Occupation Phase

32From LBA and EIA, a human-made and human-controlled agricultural landscape developed to attain a very high resistance to external factors (including climate) even when these external factors happened to change. This resistance to change shows a high sustainability of the BOP system in the rural countryside and its high resilience towards climatic instability, more efficient than the effects of political and economic disturbances.

Fig. 8B. Chronology of the Beyşehir Occupation Phase (BOP) in sites in sequences dated IInd-Ist mill. BC

Fig. 8B. Chronology of the Beyşehir Occupation Phase (BOP) in sites in sequences dated IInd-Ist mill. BC

2.4. synthesis

  • 37 As in Kobashi et al. 2013.

33In continuous sequences of north-central plateaus (Lake Tecer) and of northern Cappadocia (Eski Acıgöl), climate records show wet the first centuries of the IInd mill. BC (ca 1900-1700 BC). In the Eski Acıgöl, the following 1700-1650 BC dry episode resembles that dated in Tecer ca 1750-1650 BC. In both records, the drying trend well established in Cappadocia after ca 1550 BC, lasts until 900-850 BC, with a few short wet interruptions (fig. 8). In both records, the final drought spike occurs between 1250 and 1100 BC. In Eski Acıgöl however, another intense drought spike occurs between 950 and 900 BC37. In Tecer, humid conditions start ca 1050 BC, lasting afterwards during several centuries.

34In south-central regions of Anatolia (Konya and Bor plains), scarce sediment archives younger than 6.0 Ka BC indicate climate conditions oscillating between steadily dry conditions which may have been the general climatic context during the Vth to Ist mill. BC, sometimes interrupted by very few 150 to 100-yr-long wet sub-phases generating (i) marshes and lakes in depressed parts of the plains; and (ii) dense vegetation on slopes. Such wet sub- phases seem much rarer during the IInd mill. BC than during the preceding and following millennia. This does not mean that humid records do not exist; it only confirms climate was usually dry during the IInd mill. BC (fig. 8). It is interesting to stress, with regard with the aim of this paper that, in spite of the regularly occurring of long-lasting dry conditions in central Anatolia after 1700-1650 BC, the MBA and LBA societies faced successfully this climatic and environmental challenge until the centralized power of the Hittite Empire disappears ca 1190-1170 BC, this end-timing however fitting partly into the most intensely dry conditions of the IInd mill. BC.

3. The end of the Hittite Empire: Historical issues

35I was once asked: “Catherine, what happened with the climate in 1170 BC?” “Why such a precision with the date?” “Because it is the end of the Hittite Empire”. The prerequisites to this initial question were that (i) a direct and immediate chronological link between both ‘events’ is possible, even though the ‘events’ are of a different nature and most probably of a different timing; (ii) occurrences of such ‘events’ are supposed to be datable with a few years precision, even though they occurred 3000 yrs ago; (iii) both communities of historians and palaeoclimatologists are capable of providing a high-resolution answer with regard to the dates and scenarios of both events. Further readings about this type of question evidence another prerequisite: (iv) the ‘real definition’ of both events does not always need to be detailed before discussing the link between them.

36Evaluating the role of a climate change in a historical (cultural) event needs first to collect the most detailed and confident historical and archaeological data related to the event studied. Consequently, the following paragraphs expose the results of an exploration of publications related to the history of Anatolia during the end of the IInd mill. BC (LBA) and the transition between the IInd and the Ist mill. BC when the Hittite Empire disappeared.

  • 38 Braudel 1969.
  • 39 Klengel 2011.
  • 40 Mora/d’Alfonso 2012.
  • 41 Schachner 2011, 109-114.

37Reading historians’s and archaeologists’s contributions to the story, a first surprise from the palaeoenvironmentalist’s point of view is the high amount of uncertain or hypothetic data about (i) the circumstances and nature of the proper ‘event(s)’ ending the Hittite Empire38, which would enlighten its cause (religion, political contest in the court, partition of the Empire under the influence of neighboring States, political claims of local elites, etc.)39; (ii) the post-‘event’ cultural continuity in some regions of the Empire, which would contradict the ‘collapse’ interpretation40. This difficulty to obtain clear evidence from scarce material produces the “silence that veils the end of the Hittite State”41.

38Facing this silence, most hypotheses relate directly the ‘fall’ of the Hittite capital Hattuša in 1175-1170 BC to the destruction of Ugarit (an ally to the Hittites) in 1190 BC by the “Sea People”. A second hypothesis claims, on the basis of letters from the Hittite king to the Pharaoh asking for urgent grain delivery, that a very intense drought in Anatolia brought the Hittite Empire to its end. Since ten or fifteen years however, discussions about the sudden disappearance of the Hittite Empire shortly after 1200 BC are enriched by increasing data and interpretations from new archaeological findings and historical researches. These new dates and narratives about the causes and successions of events help us to evaluate the way climate intervenes in the collapse of the LBA Hittite Empire.

The period previous to the ‘event’ (1250-1100 BC)

  • 42 Schachner 2011, 109-114.
  • 43 Doğan-Alparslan/Alparslan 2013, 41.

39During the last 100-150 years of the Empire, increasing internal tensions between members of the ruling family, between the Empire and its more or less allied neighboring states, between religious and politic administrations… provide new factors that may have caused or contributed to the end of the Empire. New data tell us that between 1290 and 1270 BC, the Great King Muwatalli II transferred the capital of the Empire (with his court, administration, gods, army) from Hattuša in north-central Anatolia to a site in southern Anatolia (Tarḫuntašša) which has not yet been located42. According to Singer (2006), the cause of this unprecedented move was religious, the new city being most probably devoted to a single god’s cult. As a result, temples in Hattuša turned into workshops. After Muwatalli’s successors returned to Hattuša and its gods, “internal politic tensions nevertheless increased during the following decades, ending in a trend towards partition and/or contest of the Empire central power between two distinct families… However, the few texts attributed to the last king (Suppiluliyama, who reigned ca 1210-1190 BC), do not convey a picture of inevitable decline but of rather successful deeds43.

The end of Hittite presence at Hattuša at an unknown date: an abandonment, not a destruction

  • 44 Ibid.

40The last phase of the Hittite capital is characterized by a systematic clearing of the buildings, probably including the most recent documents44, a picture which is that of a demise and not of a destruction. These data do not either sustain the usual connection made with the destruction of Ugarit. Furthermore, recent archaeological evidence points to “an assimilation of remaining inhabitants with newcomers from west and central Anatolia”.

The variety of cultural transitions in Anatolia after the Hittite Empire disappeared

  • 45 Mora/d’Alfonso 2012.
  • 46 Mora 2010.

41Recent papers discussing the turmoil which, at the turn of the IInd/Ist mill. BC, led to the end of the Bronze Age in the Eastern Mediterranean, attribute the end of the Hittite Empire c. 1190-1170 BC to the Sea People. In Anatolia, traces of the Sea People attacks and invasion are mainly concentrated in the Cilician coast. In addition, the replacement of Hittite occupations in western Anatolia by cultural elements from the west, points to a population movement westwards, which originated in northern Greece, or the Aegean, or western Anatolia. In the north of central Anatolia, there are some evidences of conflagrations at sites in the Kızılırmak basin and surrounding areas. Upheavals are also evidenced in the Euphrates area. In the Hittite core land however, three important sites (Yassı Höyük/Gordion, Böğazköy/Hattuša and Kaman Kalehöyük) show traces of settlement continuity after the LBA, with different developments evidenced in the following EIA levels45. In the southern part of the Anatolian plateau (Cappadocia), LBA-EIA continuity is attested also by historical records testifying for a relevant role of the region during both the final phase of the Hittite Empire and the following Neo-Hittite phase46.

After the collapse

  • 47 Mora/d’Alfonso 2012.
  • 48 Bryce 2012.

42During the centuries following the collapse, fragmented territories slowly recovered. This historic period, abusively labelled “Dark Ages” in the history of Greece and Anatolia, corresponds to the rise of local Iron Age chieftains and small kingdoms taking over the control and management of the land and resources. The regional dimension of this recovery is very important as shown by articles in Fisher et al. (2003). In NW and SE Anatolia, destructions occurred and, at places, villages were abandoned to be reconstructed nearby. In these regions, prosperity did not revive fully until the Middle Iron Age. In other regions, local administrative institutions reorganized the territorial systems. For example in Cappadocia where a strong continuity is attested after the disappearance of the central power, imperial Hittite administration survived in some urban centers from which secondary monarchies arose47. Elsewhere Hittite States emerged early in the 12th century BC, some with heirs of the dynasty which had ruled in Hattuša (Karkamış on the Euphrates; Karaman in the western Konya plain). These Neo-Hittite states show strong continuity with some aspects of the previous Hittite social organization and culture. At other locations, other states appeared in filling the vacuum created by the fall of the Hittite empire48.

Looking for an explanation (explanations) of the collapse of the Hittite Empire

43In spite –or because– of the sudden textual silence about Hittites after 1190 BC, causes of the ‘event’ mostly mentioned in the literature are several, but none is attested with force. Today, hypotheses evoke invasions, wars, revolts and reversals of allies; crises of the Hittite economic (very centralized) system, internal political unrest, religion, climate, several causes possibly acting together…

44For example, archaeological records in western Anatolia show population movement (from west) at the end of LBA, which replace Hittite cultural elements by new ones. This movement does not seem however to have penetrated in central Anatolia.

  • 49 Singer 2000, 27.

45Regarding possible impacts of “Sea People” incursions, there is no single indication of any conquest nor destruction by foreign enemies in the whole Hittite Anatolia. The Hatti land is known to have suffered from attacks of enemies at its borders during the late 13th century BC but texts of Suppiluliuma II (the last attested Hittite king) claim victory over Tarhuntašša which had possibly been taken over by the Sea People who had landed on the southern coast of Anatolia and were pushing north49. Thus the “Sea People” incursions did not reach the Hittite State north of the Taurus range.

  • 50 Thus the Hittite king’s call to Egypt for grain need; Klengel 2011 44-45.

46It must however be stressed here that the Sea People’s attacks on sea (islands) and land (harbors) may have interfered with overseas grain shipments from the Levantine coast or Egypt to Hatti, possibly leading to food shortages in Anatolia50.

47Today there are convincing evidences that the possible causes and consequences of the first and temporary abandonment of Hattuša ca 1280, which may be related to tensions between adverse parties in the dynasty, may have lasted during the decades following the southward move of the king’s court, administration and gods. After the return to Hattuša, the religious context of this abandonment and its meaning in terms of power centralization and disturbing contests possibly lasted too.

  • 51 Doğan-Alparslan/Alparslan 2013.

48Meanwhile complications with neighbor allies (revolts of Tarhuntašša, of Karkemiš, also evoked in Hittite texts), and claims from local elites… suggest that internal pressures were reaching the point of paralyzing the central government, thus weakening its hold on the empire (local authority, food production, rural organization etc.) or leading to its partition51.

  • 52 Op. cit., 43.
  • 53 Klengel 2011, 44-45.

49From the historian’s point of view and without yet examining what role the climate may have played in the story, “many of the events usually evoked for having caused the end of the Hittite Empire can also be explained as resulting from the disappearance of all central authority instead of causing its fall. Two examples: population movements may have been possible with no more government able to stop them; the breakdown of a centralized economic system (based on collection and redistribution of food) may have led to famines52. As a matter of fact, “the Hittite system of rule depended largely on the production of royal and individual households: part of their yield was transferred to royal storehouses in the districts to be consumed by the local administration or sent to Hattuša. Without these deliveries, Hattuša, it seems, was unable to feed the court and the central administration, the inhabitants of the city and –last but not least- the gods with their numerous cult places in the City…53.

4. What role for the climate in the end of the Hittite Empire?

4.1. The drought at the end of the IInd mill. BC: Issues regarding scales, environmental systems, and chronology

  • 54 Kuzucuoğlu 2009.

50The drought at the end of the IInd mill. BC in Anatolia (ca 1300/1200 BC to ca 1100-1000 BC) is recorded throughout the peninsula, from the southwestern mountainous areas to the central, north-central and eastern plateaus54 where it is recorded in the proxies of lakes and marshes systems as well as in other environmental records such as changes in river dynamics (in the Eğri dere watershed), soils developed over colluvium or alluvial fills, desiccation of lakes. It is also recorded in western Iran, the Levant (Israel, Lebanon, coastal Syria), the Euphrates terraces in the Taurus piedmont, the Dead Sea, the Eastern Mediterranean marine cores, etc. Its duration is however generally more that of a “phase” than that of an “event” since it is interrupted at places by short wet episodes (depending on the sensitivity of the region concerned by humidity changes).

  • 55 Kuzucuoğlu 2009 ; Roberts et al. 2011.

51This environmental record corresponds to the global records, which exhibit distinctive dry phases framed by drought peaks at the ends of the IIIrd, IInd and Ist mill. BC55. In-between these millennium-scaled warm and dry peaks, humid phases occur during the first half of IIIrd mill. BC, ca 1900-1700 BC, and the early half of the Ist mill. BC. In addition, rapid dry/wet alternations occur during the second half of the millennia and at the turns of the millennia. These humid sub-periods alternate with a dry signals (more or less rapid, more or less intense, more or less numerous). These dry signals present varying characteristics and dating responding to the (local, regional) sensitivity level of sites with regard to climate change.

Uncertainties in chronologies are always present

  • 56 Gasche 1998.

52With dating uncertainties of ± 50 yrs, caution is necessary before establishing a causal relationship between climatic and historic events). As often recalled: “a cause cannot precede the event caused”. This difficulty can be solved with the multiplication of 14C dates for each record (including archaeological contexts), or with the multiplication of high-resolution records (e.g. for climatic records: varved lake sediments, speleothems, tree rings). This uncertainty also pertains – although it is less evoked – to the cross-dating of archaeological material with textual documents (long, middle, short chronologies56).

A high variability of changes is observed in the different territories of Anatolia

53The geographic context specific to the Anatolian peninsula determines a high variability, in space/ chronology and magnitude, of (i) the impacts of global climate change in relation to the differences in sensitivity and vulnerability of territories concerned; (ii) the changes in the extension of specific atmospheric circulations pertaining to the four climatic systems influencing the Anatolian peninsula. As a result, each region presents its own succession of climatic phases within the global frame set by the climatic phasing of the earth climate.

The millennium-scaled droughts of the IInd and Ist mill. BC occur in a context of remarkable agricultural continuity

  • 57 Collective 2007.

54Once born in the southwestern, southern and central regions of Anatolia, in the LBA or during the LBA-EIA transition, the BOP phase flourished during Iron Age and Classic times. This 1500 yr-long continuity demonstrates the capacity of this production system to resist to climatic change. It also shows that the impacts of climate changes on a production system depend on the organization and technics of the rural management. It also evidences the importance of the maintenance of resources diversity. When dependent on a large centralized political authority (in a “capital” such as Hattusa for example), the centralization would have reverse impacts on this sustainability in case climate change modifies the availability vs. types of resources57.

4.2. What role for climate change in the end of the Hittite Empire?

55The silent vanishing of the Hittite Empire between 1190 and 1170 BC provides a good example for the study of the dialectic relationships between climate and civilizations in times of changes. First, it is necessary to keep in mind that the IInd mill. BC chronologies, whether from climatic- environmental or historical sources, rarely reach a resolution higher than 40-50 years. In addition, it is important to take a number of facts into account, which are related to processes varying through time:

  • environmental systems have a ‘memory’ (i.e. accumulation impact) of climate and anthropogenic contexts previous to the stress generated by climate change. For example: erosion sensitivity of sediments is dependent on older contexts of their depositional environments; the alteration of a resource system previously controlled and organized by human groups increases with climatic change; impacts or regular events reach thresholds generating stress peaks …

  • the variability in the sensitivity of environmental and cultural systems towards a climatic change modifies the resources availability;

  • the embedded time and space scales of processes and acting systems generate time- delays in the occurrence of environmental impacts of climate changes in the territories and river watersheds;

  • the role of time-accumulation with regard to the rigidity of cultural systems involved, is also affected by (i) the accumulation, before the climate change, of rising yet non bursting social and economic tensions; (ii) the impending time- lag that political (authority) rigidity generates in face of necessary change, especially when the necessary solutions need reducing the privileges and benefits of the ruling elites.

  • 58 Kuzucuoğlu/Marro 2007; Miroschedji 2009; Kuzucuoğlu 2012.

56The cultural EBA to LBA history in the EM is phased in several long periods running along stable trends corresponding to ‘flourishing’ economic systems ruled by elites (politics, religion). At times, short periods of rapid changes interrupt these ‘quiet’ periods. These interruptions often seem to occur unexpected. Their possible violence and abruptness suggest that their highly disturbing efficiency is caused by the impossibility of cultural systems to evolve smoothly… as if a threshold generates rapid destruction instead of transforming social, economic, environmental contexts58. In other words, the fragility of cultural systems with regard to climate change is the same as with regard to other types of changes, whether these changes are internal (e.g. the social structure) or external (e.g. international trade, migrations). In the case of the end of the Hittite Empire, what is best shown by documents is: (i) population movements from west and south did not reach the heart of the Empire but disturbed its alliances around; (ii) wars at sea in the Mediterranean, islands and harbors, disrupted trade and exchanges feeding the ‘global’ economic system of the time (from the Balkans to Anatolia, from Egypt to the Levant and Mesopotamia)… In the meantime, when the Hittite Great King asks for wheat from the Egyptian Kingdom, the lack of grain may have risen from military difficulties.

57Another fact that must be taken into account is that, whatever its exact chronology, the drought at the end of the IInd mill. BC started at least 100 years if not 200 years before the Hittite Empire disappeared. No historical document allows identifying with certainty a relationship with an unusual dry context.

Does the permanency of a dry context during most of the Empire’s life, suggest that dry conditions may have contributed to the development of a stable politic system, because such a climatic context necessitated adaptive rules allowing agricultural management capable of facing irregularity of resources?

58As shown by global, regional and local curves referencing climate trends and changes in the IInd mill. BC, the history of the Old Kingdom as well as that of the Empire runs in dry climatic conditions. This difficult climatic context must have been integrated into the functioning of the economic, politic and administrative system. Symptomatic to this adaptation is that an original agricultural system (BOP) expanded. This system happened to be very efficient, remaining efficient (adaptive) during external (e.g. climatic) and internal (e.g. politic) changes.

59The chain of events leading to the Empire’s fate between 1190 and 1170 BC, is thus most probably to be found in a set of facts in which climate most probably played the role of the ‘on’ button, not so important as to destroy, with irregular and mostly dry conditions, the agricultural system which was sustaining life in rural areas as well as production partly feeding the Empire... Archaeological and historical data show that social and political actors were also present, such as: (i) a rigid system of resources distribution; (ii) influences of groups seeking to preserve and increase their privileges, thus generating conservatism; (iii) increasingly dilute authority between an increasing number of (politic, religious) groups… etc. These internal factors made the central political power unable to identify and take decisions when facing changes in external conditions. According to this interpretation, climate was the last, possibly decisive but not the first, actor in the tragedy. The challenge generated by the high level of climatic stress (already present since a few decades) added problems in a general difficult context of population movements at the boundaries and in the allied territories of the Empire… events which were also challenging the politic and economic stability of the Empire.

Does the regional contrasts of climate suggest that the parts of the Empire (such as southern Cappadocia) which were the most prepared –by a mixed agricultural production best adaptive- to face drought impacts were also the parts which were most capable of facing its increasing impacts?

60This question rises from the fact that it is in the semi-arid parts of central (and southeastern) Anatolia that archaeologists observe the continuity of settlement occupation, the rise of the local elites, and the development of Neo-Hittite States during the Early Iron Age. In the northern plateaus where the drought was long and intense (see Lake Tecer sequence), the management system may have been, to some degree, more fragile with regard to changing conditions. Another way of asking the question is: “was northern Anatolia plateau where imperial power concentrated possibly more impacted by (more fragile, unprepared, in the face of) the ‘dry’ context of the second half of the IInd mill. BC than the southern Anatolia plateau which was used to face such dry contexts for a longer time”? The answer to this question needs historical data which are yet lacking today.

61At any case, in southern Cappadocia, the environmental conditions allowed to survive the period of the politic and economic collapse with resources made still available both by the conservative measures of the agricultural system, and by the variety of water and soil resources: mountain streams running to the plain in alluvial sediments capable of storing water in soils easy to exploit; animal production taking advantage of complementary mountain and plain ecosystems, etc.

The importance of geographic scale (from global, regional to local) in the timing and intensity of environmental impacts of climate changes in the EM

62Global, regional and local scales must be taken into account when evaluating the dimensions of cultural responses to climatic change.

  • Local scale may explain some differences in the chronology and in the intensity of the impacts of climatic changes on the environment as well as on the politic and economic systems around the EM, differences which must have induced time delays in the events and processes occurring in the sub-regions of the EM.

  • Depleted humidity conditions in parts of the EM may have weakened too specialized production systems, thus triggering population movements searching for new resources, driving warfare across the sea and surrounding Sates States, i.e. acting on the global scale of the EM.

  • Consecutive ruptures within the ‘global’ economic network linking and enriching all parts of this EM world generated a high degree of disturbances interrupting the EM trade.

63Such disputes occurred several times during the second half of the IInd mill. BC, for example when destruction hit Minoan Palaces ca 1450 BC, Mycenian Palaces ca 1300 BC… With attacks against the maritime trade (the “Sea People” evoked by the Egyptian texts) and the following fall of trade-posts, military positions, powerful cities, harbors and alliances in the islands and on the coasts…, the global dimension of the economic and politic degradation at the end of the IInd mill. BC, added to the difficulties arising from droughts and internal weaknesses, thus to the ‘fall’ of the Hittite Empire after that of several other States in the EM.

Conclusion

64In Anatolia, several historical changes in Anatolia (the end of the Hittite Empire, Iron Age states, Achaemenid and Roman Empires, Early Byzance) occur at the same time as (sometimes drastic) climate changes marked by successive droughts. These changes occur also at the same time as a period of remarkable agricultural continuity…. Sustainable rural activities thus long-live with regard to climate change and political turmoil. This scenario contradicts a deterministic view of the climate role in history that would be based on the assumption of the irreversible decline of agriculture during increasing dryness. Such a decline only means that the agricultural system was not respecting diversity and inventiveness which are necessary for the adaptation to changing external conditions.

65It is such a type of rigidity that characterizes the centralized management of food resources pertaining to the Hittite political system. The continuity between the Hittite collapse and start of Early Iron Age societies in central Anatolia clearly shows that, even in times of population movements, the capacity of a production and cultural system to resist, through any kind of adaptation, to climatic change depends clearly on the sensibility of the internal functioning of economic and political systems to climate change. This sensibility may be related, not to the intensity or nature of the climate change, but to the internal factors of vulnerability with regard to change: rigidity, centralization of decision; decision transmission networks, unbalanced distribution of resources (water, soil, means of production –including irrigation equipment- wealth, land ownership…).

66In the conclusive sentences below, I state E. Leroy-Ladurie’s words commenting the role that climate plays in history, a role played as in the succession of events (historical, demographic, religious, military, dynastic etc.) that led to the end of the Hittite Empire:

Bibliographie

Akkemik, Ü / Caner, H. / Conyers, G.A. / Dillon, M.J. / Karlıoğlu, N. / Rauh, N.K. / Theller, L.O., “The archaeology of deforestation in south coastal Turkey”, International Journal of Sustainable Development and World Ecology, 2012. doi:10.1080/13504509.2012.684363

Alley, R.B., GISP2 Ice Core Temperature and Accumulation Data. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series 2004-013. NOAA/NGDC Paleoclimatology Program, Boulder CO, USA.

Ballatti, S. / Balza, M.E., “Kınık Höyük and Southern Cappadocia: Geo-Archaeological activities, Landscapes and Social Spaces”, in R. Hofmann / F.K. Moetz / J. Müller (éds.), Tells: Social and Environmental Space, vol. 3, Bonn, 2012: 93-104.

Bottema, S. / Woldring, H., “Late Quaternary vegetation and climate of Southwestern Turkey. Part II”, Palaeohistoria 26, 1984, 123–149.

Bottema, S. / Woldring, H., “The vegetation history of East-Central Anatolia in relation to archaeology: the Eski Acigöl pollen evidence compared with the Near Eastern environment”, Palaeohistoria 43/44, 2001, 1–34.

Braudel, F., Les Mémoires de la Méditerranée. Préhistoire et Antiquité. Edition of an original manuscript dating 1969. Edited in 1998 by R. de Ayala & P. Braudel, with notes by J. Guilaine and P. Rouillard, Références Histoire, Ed. de Fallois, Paris, 1998.

Bryce, T.R., The World of the Neo-Hittite Kingdoms. A political and Military History, Oxford, 2012.

Collectif, “Characteristics and changes in archaeology- related environmental data during the Third Millennium BC in Upper Mesopotamia”, in C. Kuzucuoğlu / C. Marro (éds.), The crisis at the end of the 3rd mill. BC in the middle Euphrates Valley. A reality?, Paris, 2007, 573-580.

Cuffey, K.M. / Clow, G.D., “Temperature, accumulation and ice sheet elevation in central Greenland through the last deglacial transition”, J. Geophys. Res. 102.26, 1997, 383-396.

D’Alfonso, L. / Balza, M.E. / Mora, C., Geo-archaeological activities in southern Cappadocia, Turkey [Proceedings of the Meeting held in Pavia 20.11.2008], Italian Univ. Press, 2010.

Doğan-Alparslan/Alparslan 2013

Doğan-Alparslan M. / Alparslan M. (éds.), Hititler, Bir Anadolu Imporatorluğu. Hittites, An Anatolian Empire, Istanbul, 2013.

Eastwood, W.J. / Roberts, N. / Lamb, H.F., “Palaeoecological and archaeological evidence for human occupation in southwest Turkey: the Beyşehir occupation phase”, Anatolian Studies 48, 1998, 69-86.

Eastwood, W.J. / Roberts, N. / Lamb, H.F. / Tibby, J.C., “Holocene environmental change in southwest Turkey: a palaeoecological record of lake and catchment-related changes”, Quaternary Science Review 18, 1999, 671-695.

Fontugne, M. / Kuzucuoğlu, C. / Karabıyıkoğlu, M. / Hatté, C. / Pastre, J-F., “From Pleniglacial to Holocene. A 14C chronostratigraphy of environmental changes in the Konya Plain, Turkey”, Quat. Sc. Reviews 18.4-5, 1999, 573-592.

Gasche, H., Dating the Fall of Babylon. A Reappraisal of Second-Millennium Chronology [MHEM 4], Chicago, 1998.

Gürel, A. / Lermi, A., “Pleistocene-Holocene fills of the Bor- Ereğli Plain (central Anatolia): recent geo-archaeological contributions. Geoarchaeological Activities in Southern Cappadocia – Turkey”, in L. d’Alfonso / M.E. Balza / C. Mora (éds.), Geo-archaeological activities in southern Cappadocia, Turkey [Studia Mediterranean 22], Italian Univ. Press, 2010, 55-70.

Haldon, J., “Cappadocia will be given to ruin and become a desert”, in K. Belke / E. Kislinger / A. Külzer / A. Stassinopoulou (éds.), Byzantina Mediterranea, Wien, 2007, 215-230.

Klengel H., “History of the Hittites”, in H. Genz / D.P. Mielke (éds.), Insights into Hittite History and Archaeology [Colloquia Antiqua 2], Louvain/Paris/Walpole, MA, 2011, 31-46.

Kobashi, T. / Kawamura, K. / Severinghaus, J.P. / Barnola, J-M. / Nakaegawa, T. / Vinther, B.M. / Johnsen, S.J. / Box, J.E., “High variability of Greenland surface temperature over the past 4000 years estimated from trapped air in an ice core”, Geophysical Research Letters 38.21, 2011, doi:10.1029/2011GL049444.

Kuzucuoğlu, C., “Le dernier glaciaire et l’Holocène en Anatolie centrale : apports de la géomorphologie et de la sedimentology”, in M.F. André / S. Etienne / Y. Lageat / C. Le Coeur / D. Mercier (éds.), Du continent au bassin versant. Théories et pratiques en géographie physique (Hommage au Professeur Alain Godard), Clermont-Ferrand, 2007, 495-505.

Kuzucuoğlu, C., “Climate and environment in times of cultural changes from the 4th to the 1st mill. BC in the Near and Middle East”, in A. Cardarelli / A. Cazzella / M. Frangipane / R. Peroni (éds.), Reasons for Changes of Societies Between the End of the IV and the Beginning of the Ist Millennium BC (Le Ragioni del Cambiamento) [Scienze delle Antichità, 15], Roma La Sapienza, 2009, 141–163.

Kuzucuoğlu, C., “Climate change and anthropogenic signals in Holocene sequences of North Anatolia”, Abstract Book, 7 ICAANE, Londres, Avril 2010.

Kuzucuoğlu, C. / Dörfler, W. / Kunesch, S. / Goupille, F., “Mid-Holocene climate change in central Turkey: the Tecer lake record”, The Holocene 21/1, 2011, 173-188.

Kuzucuoğlu, C., “Le rôle du climat dans les changements culturels, du Vè au 1er millénaire avant notre ère, en Méditerranée orientale”, in J-F. Berger (éd.), Des climats et des hommes, Paris, 2012, 239-256.

Kuzucuoğlu, C. / Karabıyıkoğlu, M. / Fontugne, M. / Pastre, J-F. / Ercan, T., “Environmental changes in Holocene lacustrine sequences from Karapınar in the Konya Plain (Turkey)”, in N. Dalfes / G. Kukla / H. Weiss (éds.), Third Millenium BC Climate Change and Old World Collapse [NATO ASI Series 149], 1997, 451-163.

Kuzucuoğlu, C. / Parish, R. / Karabıyıkoğlu, M., “The dune systems of the Konya Plain (Turkey). Their relation to the environment changes in Central Anatolia during Late Pleistocene and Holocene”, Geomorphology 23, 1998, 257-273.

Kuzucuoğlu, C. / Fontugne, M. / Karabıyıkoğlu, M. / Hatté, C., “Evolution de l’environnement dans la plaine de Konya (Turquie) pendant l’Holocène”, in M. Otte (éd.), Anatolian Prehistory. At the Crossroads of Two Worlds, Liège, 1999, 605-624.

Kuzucuoğlu, C. / Bertaux, J. / Black, S. / Denèfle, M. / Fontugne, M. / Karabıyıkoğlu, M. / Kashima, K. / Limondin- Lozouet, N. / Mouralis, D. / Orth, P., “Reconstruction of climatic changes during the Late Pleistocene, based on sediment records from the Konya Basin (Central Anatolia, Turkey)”, Geology Journal 34, 1999, 175–198.

Kuzucuoğlu, C. / Emery-Barbier, A. / Fontugne, M. / Kunesh, S., “Öküzini marcshes: A new Upper Pleistocene record on the Anatolian Mediterranean Coast”, in I. Yalçınkaya / M. Otte / J. Kozlowski / O. Bar-Yosef (éds.), Öküzini: Final Paleolithic Evolution in Southwest Anatolia, ERAUL 96, Liège, 79-82 and 88-89.

Kuzucuoğlu, C. / Fontugne, M. / Mouralis, D., “Holocene terraces in the Middle Euphrates valley, between Halfeti and Karkemish (Gaziantep, Turkey)”, Quaternaire 15.1/2, 2004, 195-206.

Kuzucuoğlu, C. / Marro, C. (éds.), Sociétés humaines et changement climatique à la fin du Troisième Millénaire : une crise a-t-elle eu lieu en Haute Mésopotamie ?, Paris/Istanbul, 2007.

Kuzucuoğlu, C. / Gürel A. / d’Alfonso L. / Gauthier A. / Robert V. / Cetoute J., Impacts of climate change on hydrological dynamics and water-related landscapes in South-Cappadocia (Turkey) during the Holocene. Preliminary results in the Bor-Eregli plain [QuickLakeH 2014, an International Workshop on Lakes and Human Interactions, 15-19 September], Ankara, Sept. 2014.

Le Roy Ladurie, E., Naissance de l’histoire du climat, Paris, 2013.

Miroschedji de, P., “Rise and Collapse in the Southern Levant in the Early Bronze Age”, in A. Cardarelli / A. Cazzella / M. Frangipane / R. Peroni (éds.), Reasons for Changes of Societies Between the End of the IV and the Beginning of the Ist Millennium BC (Le Ragioni del Cambiamento) [Scienze delle Antichità, 15], Roma La Sapienza, 2009, 101-129.

Mora, C., “Studies on Ancient Anatolia at Pavia University, and the Hittite Low Land”, in L. d’Alfonso / M.E. Balza / C. Mora (éds.), Geo-Archaeological Activities in Southern Cappadocia, Turkey, Pavia University, 2010, 13-25.

Mora, C. / d’Alfonso L., “Anatolia after the end of the Hittite Empire. New evidence from southern Cappadocia”, Origini 24, Nuova Serie V, X-XX, 2012, 385-398.

Roberts, N. / Reed, J.M. / Leng, M.J. / Kuzucuoğlu, C. / Fontugne, M. / Bertaux, J. / Woldring, H. / Bottema, S. / Black, S. / Hunt, E. / Karabıyıkoğlu, M., “The tempo of Holocene climatic change in the eastern Mediterranean region: new high resolution crater-lake sediment data from central Turkey”, The Holocene 11.6, 2001, 721-736.

Roberts, N. / Eastwood, W. / Kuzucuoğlu, C. / Fiorentino, G. / Caracuta, V., “Climatic, vegetation and cultural change in the eastern Mediterranean during the mid-Holocene environmental transition”, The Holocene 21.1, 2011, 147-162.

Schachner, A., Hattuscha. Auf der Suche nach dem sagenhaften Grossreich der Hethiter, Munich, 2011.

Severinghaus, J.P. / Grachev, A. / Luz, B. / Caillon, N., “A method for precise measurement of argon 40/36 and krypton/argon ratios in trapped air in polar ice with applications to past firn thickness and abrupt climate change in Greenland and at Siple Dome”, Antarctica. Geochimica et Cosmochimica Acta 67.3, 2003, 325–343.

Singer, I., “New Evidence on the end of the Hittite Empire”, in E.D. Oren (éd.), The Sea peoples and Their World: A Reassessment, Philadelphia, 2000, 21-33.

Singer, I., “The Failed Reforms of Akhenaten and Muwatalli”, BMSAES 6, 2006, 37-58.

Sullivan, D.G., Human-induced vegetation change in western Turkey: Pollen evidence from central Lydia (unpub. PhD Thesis), University of California, Berkeley, 1989.

Thirgood, J.V., “Man and the Mediterranean Forest: A History of Resource Depletion”, Journal of Applied Ecology 19.3, 1982, 980-981.

Vermoere, M. / Bottema, L. / Vanhecke, L. / Waelkens, M. / Paulissen, E. / Smets, E., “Palynological evidence for late- Holocene human occupation recorded in two wetlands in SW Turkey”, The Holocene 12.5, 2002, 569-584.

Weiss, H. / Courty, M.-A. / Wetterstrom, W. / Guichard, F. / Senior, L. / Meadow, R. / Curnow, A., “The genesis and collapse of Third Millennium North Mesopotamian Civilization”, Science 261, 1993, 995-1004.

Zeist van, W. / Woldring, H. / Stapert, D., “Late quaternary vegetation and climate of southwestern Turkey”, Palaeohistoria 17, 1975, 53-144.

Zeist van, W. / Bottema, S., “Vegetational History of the Eastern Mediterranean and the Near East during the last 20,000 years”, in J.K. Bintcliff / W. van Zeist (éds.), Palaeoclimates, palaeoenvironments and Human Communities in the Eastern Mediterranean Region in Later Prehistory [BAR 133], Oxford, 1980, 277-323.

Notes

1 Kuzucuoğlu et al. 2011.

2 E.g. Weiss et al. 1993.

3 Kuzucuoğlu/Marro 2007 ; Kuzucuoğlu 2012 ; Leroy-Ladurie 2013.

4 Cuffey/Clow 1997.

5 In ice: Alley 2004; and air bubbles: Kobashi et al. 2011; see also Severinghaus et al. 2003.

6 δ18O is the isotopes 18O/16O ratio; δ15N is the isotopes 15N/14N ratio; δ40Ar is the isotopes 40Ar/36Ar ratio.

7 Due to unsteady CO2 content in atmosphere through time, all 14C calibrated dates present an uncertainty of 50 yrs minimum. In specific records where 14C dates are numerous enough for highly constraining the time-model, the precision of the chronology may increase to 30-40 yrs.

8 In this paper 14C calibrated BP dates are presented as BC dates. In the curves presented in figs. 4B, 5, 7 and 8B, the BC scale has been obtained by substracting 1955 yrs out of all 14C cal. BP ages of time-models. Elsewhere in the text, when occurring, the “kyrs” notation means “1000” years.

9 E.g. Kuzucuoğlu 2009 ; Kuzucuoğlu et al. 2011 ; Roberts et al. 2011.

10 Kuzucuoğlu 2009.

11 Kuzucuoğlu et al. 2011 ; Roberts et al. 2011.

12 E.g. Kuzucuoğlu/Marro 2007 ; Kuzucuoğlu 2009 and 2012 ; Roberts et al. 2011.

13 Kuzucuoğlu 2010 ; Kuzucuoğlu et al. 2011.

14 Kuzucuoğlu 2009.

15 Kuzucuoğlu 2009 ; Kuzucuoğlu et al. 2011.

16 P = Precipitation (rainfall, snowfall).

17 The P limit of 250 mm/yr is also the conventional limit for dry farming practices; below 250 mm/yr, P irregularity limits crop production; reducing the risk then needs technological developments such as irrigation.

18 Respectively Kuzucuoğlu et al. 2011 and Roberts et al. 2001.

19 Kuzucuoğlu et al. 2011.

20 Bottema/Woldring 2001 ; Roberts et al. 2001. New multiproxy lake sequences in Cappadocia are awaited soon publication : Nar Lake crater Nar near Kitreli (north of the Göllüdağ massif : Roberts et al.) ; Büyük Göllüdağ Lake crater (Woldring) ; Çora Lake crater in the Erciyes volcano near Kayseri (Gauthier et al.).

21 Kuzucuoğlu et al. 2014.

22 D’Alfonso et al. 2010.

23 LGM= Last Glacial Maximum= 28-17kyrs cal BP (ie 26-15 kyrs BC). A LGM lake covered the Konya plain bottom from ca 25.2 to 18.2 kyrs cal BP (= 23.5 to 17.5 kyrs uncal. BP) (Fontugne et al. 1999).

24 Kuzucuoğlu et al. 1999a and 1999b.

25 Kuzucuoğlu 2007.

26 Optically Stimulated Luminescence dated dunes: Kuzucuoğlu et al. 1997 and 1998.

27 One U-Th aged and two 14C dated marshy sediments and forest-soils: Kuzucuoğlu et al. 1999a; Kuzucuoğlu 2007.

28 Fontugne et al. 1999, 583 ; Kuzucuoğlu 2007.

29 Akkemik et al. 2012.

30 Eastwood et al. 1999.

31 Thirgood 1982.

32 Eastwood et al. 1998.

33 Gauthier et al., in preparation.

34 Dörfler et al., in preparation.

35 Such as during the 7th century AD: Haldon 2007.

36 for definition of the BOP, see Bottema/Woldring 1984; Eastwood et al. 1998.

37 As in Kobashi et al. 2013.

38 Braudel 1969.

39 Klengel 2011.

40 Mora/d’Alfonso 2012.

41 Schachner 2011, 109-114.

42 Schachner 2011, 109-114.

43 Doğan-Alparslan/Alparslan 2013, 41.

44 Ibid.

45 Mora/d’Alfonso 2012.

46 Mora 2010.

47 Mora/d’Alfonso 2012.

48 Bryce 2012.

49 Singer 2000, 27.

50 Thus the Hittite king’s call to Egypt for grain need; Klengel 2011 44-45.

51 Doğan-Alparslan/Alparslan 2013.

52 Op. cit., 43.

53 Klengel 2011, 44-45.

54 Kuzucuoğlu 2009.

55 Kuzucuoğlu 2009 ; Roberts et al. 2011.

56 Gasche 1998.

57 Collective 2007.

58 Kuzucuoğlu/Marro 2007; Miroschedji 2009; Kuzucuoğlu 2012.

Table des illustrations

Titre Fig. 1: Physical Geography of Turkey
Légende A. Relief; B. Climatic regions; C. Vegetation formations
Crédits A: http://commons.wikimedia.org/​wiki/​File:Turkey_topo.jpg; B: Meteorological Service of Turkey; C: van Zeist/Bottema 1980
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-1.png
Fichier image/png, 2,3M
Titre Fig. 2: Palaeotemperature curves from GISP2 ice cores in northern Greenland
Légende A. Temperature (C°) curve based on stable oxygen isotopes analyses; B. Snow surface temperature reconstructed using N and Ar isotopes in gaz trapped in ice
Crédits A: modified from Alley 2004; B: modified from Kobashi et al. 2011
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-2.png
Fichier image/png, 326k
Titre Fig. 3A: Precipitation distribution of Turkey. Distribution map of average annual precipitation (mm/yr)
Légende Note: Numbers correspond to the profiles below.
Crédits A. Meteorological Service of Turkey
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-3.png
Fichier image/png, 168k
Titre Fig. 3B: Precipitation distribution of Turkey. Four West-East cross-sections of relief (upper graphs) and precipitation (lower graphs)
Légende Horizontal shaded bands: (a) dark grey: annual P < 400 mm/yr; (b) light grey: annual P= 400-500 mm/yr Vertical shaded bands: (a) dark grey: Central Anatolia; (b) light grey: other dry regions in AnatoliaNote: Stations selected along profiles are located in towns. Consequently, no data concerning mountains crossed by the profiles in fig. 3A are present on the graphs.
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-4.png
Fichier image/png, 677k
Titre Fig. 4: Climatic phases in central Anatolia between 4000 and 0 BC
Légende A. Location of palaeoenvironmental sites and records used in Figure 4B, placed on the map of annual precipitation distribution in Turkey.Legend: Sites on map: 1= Tecer Lake (Sivas); 2= Eski Acıgöl (Nevşehir); 3= Öküzini (Antalya); 4= Eğridere (Yozgat); 5= Karaman and Ereğli (Konya plain); 6= Bor plain (Niğde)Notes: Ref. average annual precipitation amount at the sites in Table 1.B. Regional variability of climate changes in central Anatolia and some surrounding areas according to palaeoenvironmental records in closed depressions and watersheds.Legend of fig. 4B: Grey shading shows increasing-decreasing dryness. Plain grey bands on synthesis show drought peaks.Note: For detailed reconstruction proposal for central Anatolia in Figure 7
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-5.png
Fichier image/png, 494k
Titre Fig. 5: 4000-0 BC climatic record at Lake Tecer (Sivas)
Crédits Modified from Kuzucuoğlu et al. 2011
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-6.png
Fichier image/png, 552k
Titre Fig. 6: The Lake Tecer dry phases sequence, compared to GISP2 temperature records (4000-0 BC)
Crédits A: modified from Alley 2004; B: modified from Kobashi et al. 2011; C: Kuzucuoğlu et al. 2011
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-7.png
Fichier image/png, 373k
Titre Fig. 7: Proposed chronology of wet/dry climate alternations during the IInd mill. BC, interpreted from palaeoenvironmental sequences in Central Anatolia
Légende 1. Humid phase; 2. Wet phase (transition); 3. Dry phase; 4. Drought phase; 5. Regional dry signal.Note: The chronology presented here is to be taken with cautious due to uncertainties in the dating of sequences, which reach ca 50 years. For sequences in the Bor plain, chronology published here will be ascertained by further dates. The two options presented here, and the question marks associated to some phases, testify for the going-on work
Crédits Kuzucuoglu 2007 ; Kuzucuoglu et al. 2011 ; Kuzucuoglu/Gürel et al. 2014 ; Roberts et al. 2001.
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-8.png
Fichier image/png, 521k
Titre Fig. 8: The Beyşehir Occupation Phase36 (BOP) in Turkey. A. Location of sites with pollen studies corresponding to the period from 2000 BC-AD 1000
Légende Sites references : 1. Köyceğiz ; 2. Gölcük ; 3. Gölhisar ; 4. Elmalı ; 5. Avlan ; 6. Söğüt ; 7. Öküzini ; 8. Pınarbaşı ; 9. Gravgaz ; 10. Beyşehir ; 11. Abant ; 12. Melen Gölü ; 13. Yeniçağa ; 14. Nar Gölü ; 15. Eski Acıgöl ; 16. Cora; 17. Suppitassu; 18. Tecer; 19. Hoyran; 20. Karamık; 21. Ova; 22. VanWhite number on black background= Sites with Beyşehir Occupation Phase // Black number on white background= Sites without Beyşehir Occupation Phase
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-9.png
Fichier image/png, 1,6M
Titre Fig. 8B. Chronology of the Beyşehir Occupation Phase (BOP) in sites in sequences dated IInd-Ist mill. BC
URL http://books.openedition.org/ifeagd/docannexe/image/3232/img-10.png
Fichier image/png, 511k

© Institut français d’études anatoliennes, 2012

Conditions d’utilisation : http://www.openedition.org/6540

Lire

Open access

Acheter

Volume papier

amazon.fr
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search