The CANEW Project

Lecture

ENVIRONMENTAL SETTING AND EVOLUTION FROM THE 9th TO THE 5th MILLENNIUM CAL BC IN CENTRAL ANATOLIA. An introduction to the study of relations between environmental conditions and the development of human societies

Catherine KUZUCUOĞLU
MAE/IFEA, Nuru Ziya Sok n° 22, PK 54, 80072 Beyoglu, Istanbul, TR. UMR 8591-CNRS/Paris I University, 92195 Meudon cedex, F.
kuzucuogl@netcourrier.com

The geographic definition of the CANeW area (Fig. 1)

The CANeW map covers an area defined by a water divide including four endoreic basins: the Tuz Gölü to the north, the Beysehir Lake to the west, the Konya Plain in the centre, and the Cappadocian plateau (including the Sultansazligi Plain) to the east. Although naturally endoreic, the Beysehir Plain discharges today to the Konya Plain by an old, if not always active, river network system. The whole CANeW territory is thus limited to the north by the Kizilirmak River valley and to the south by the Taurus range.

Annual precipitation amount is roughly similar in these regions when considering: 1) the flat, semi-arid central plateaus and plains of the Tuz Gölü (saline) and Konya areas (freshwater) (<350mm/yr); 2) the more humid and more mountainous Beysehir Plain, representing a transition to the Lake District (>400mm/yr); 3) the Cappadocian highlands and plateaus, highly eroded and somewhat wetter (400mm/yr).

With regard to the Central Anatolian endoreism, this region is included into a larger region that includes also the Lake District (Pisidian lakes). Geographically, the CANeW area also belongs to a second very large area corresponding to the Central Anatolian Plateau, where altitude is commonly 1000–1200m and the climate is continental with cold winters and warm–dry summers; while an annual precipitation of <500mm/yr is concentrated in winter and spring. On the Anatolian Plateau a steppe-like vegetation covers the lowlands and limestone plateaus, while residual forests are contracted on non-limestone heights. Saline lakes occupy water-depleted areas.

How was the limit of ‘Central Anatolia’ drawn to serve as the space base in which the CANeW archaeological approach was to be set? The CANeW territory possesses several aspects of the Central Anatolian continentality: endoreism; plateau relief with largely developed plains at 1000m altitude; a continental climate with a yearly precipitation <400mm and with a subsequent steppe vegetation. Unlike the Lake District, the CANeW territory is concerned with semi-arid environments. As such, it is highly sensitive to any change in climate, whether the trends in water availability are positive or negative. Exclusion of the Lake District from the CANeW territory is thus due to contrasting differences in the water budget, water availability, alimentation and routes. In the Lake District, annual precipitation is higher (>500mm/yr). Lower summer evaporation has a reduced impact. Slopes are more densely vegetated and river basins are highly water-fed by winter snow reservoirs of the Taurus Mountains. Underground water discharges may occur towards the Mediterranean, whereas the underground water discharges in the basins covered by the CANeW area are mainly directed towards evaporative plains on the Central Plateau.

I. INTRODUCTION

Palaeoenvironment studies in lakes

Lake budget elements and mechanisms of variations

Whatever the depth and extension of lakes, the water budget of wetlands in sensitive areas such as the closed depressions of Central Anatolia depends on a balance between water input and output. In Central Anatolia water input consists mainly of precipitation and runoff. Importance of runoff in the budget depends on 1) the amount of precipitation on the drainage area (climatic indicator); 2) the capacity of soil and rock outcrops to infiltrate the precipitated water, also related to the density of vegetation cover on the slopes; and 3) the total area of the lake drainage area. Water output consists of water losses in soils (infiltration) and of evaporation (climatic indicator, mainly of temperature vs. precipitation amounts and seasonal distribution, especially summer ones, but also of wind). Variations in water input are dominated by variations in precipitation and in runoff. Variations in runoff depend on the amount of precipitation in the lake drainage area and on the vegetation cover density over the slopes (whether grass or forest). Variations in water output are triggered by variations in evaporation which are related to variations affecting the temperatures, the winds, the seasonal distribution pattern of temperatures versus precipitation and the vegetation cover density over the slopes.

In lake environment responses to climatic changes, lake depth is a major element. When shallow, lake or marsh level will react quickly to any change in the water budget, with expansion (marsh becoming lake; a saline, shallow lake becoming a more dilute lake) or shrinkage (concentrated water) or disappearance (stratigraphic hiatus). When deeper, the lake will persist whether higher or shallower, and climatic conditions will be continuously registered, without hiatus, within the bottom sediment.
Thus, the interpretation of proxy analyses results must take into account the type of lake whether shallow or not, and with large drainage areas (Beysehir; Konya wetlands; Tuz Gölü; Sultansazligi), deep and with small drainage areas (crater lakes in Cappadocia such as Eski Acigöl), deep and with large drainage areas (Van Lake, Burdur Lake).

Starting from the period of the development of agricultural activities, humans negatively intervene in the water budget of closed depressions sensitive to any variations in the water budget by: 1) diverting runoff water when irrigating (in river valleys and alluvial fans); 2) increasing output losses through water consumption (humans, animals, crops, irrigation) and irrigation water evaporation; 3) reducing the vegetation cover density on slopes by deforestation for energy needs, cultivation and animal grazing. Humans will positively intervene in the water budget by: 1) releasing water from previously diverted runoff courses; 2) abandoning irrigation networks; 3) abandoning range lands; 4) decreasing agricultural activities.

Lake indicators used for studying palaeoenvironmental and palaeoclimatical records

Researches on palaeoenvironments are based on indicators recording the physical and biological characteristics of the lake waters. When studied in cored sequences, such indicators allow reconstruction of the major variations in the lake budget and, consequently, the climatic conditions.

The lake depth, water temperature and dilution of chemicals in water, which are indicators of variations in precipitation-evaporation ratio, in the evaporation stress and in the runoff discharges are indicated by mineralogical (authigenic and/or precipitated minerals), sedimentological and lithological (lake sediment depositional characteristics and post-deposition transformations), biologic (diatoms, organic carbon content, molluscs) and stable isotopic indicators. The composition of the vegetation cover in the surrounding areas is reconstructed using pollen assemblages (landscape and vegetation cover) and plant macro-remains while microcharcoal inform about fire occurrences. The amount of runoff over the surrounding drainage area (more sensitive in large drained areas than in small ones) is studied using the grain content of imported sediment material. The setting of a chronological frame is usually provided by 14C measurements on organic sediments, pollen and charcoal. Dates may also be provided by U-Th measurements on carbonates.

Climatic context in the northern Hemisphere and in the Mediterranean at the lateglacial/Holocene transition and during the early Holocene

(Palaeoenvironmentalists and climatologists used to work in a BP frame; I apologise for my paper being given in cal BP and not in cal BC except for discussion of the ECA maps. I would also like to stress that the precision of archaeological dating is rarely obtained in environmental continental records (except for high-resolution records such as laminated lake sequences, speleothems and tree rings). For example, the ECA III starting date (6700–6600 cal BC) will be translated in sediment sequences into c. 6500 cal BC.)

Climatic phasing in the northernhemisphere (global data from ice, marine and continental records)

The succession is the following:

17,000–16,000 BP: start of the lateglacial warming

16,000–14,500 BP: Heinrich Event 1 (HE1), which is a complex event registered in the marine sediments of the northern Atlantic ocean, showing high cold freshwater discharges from the melting northern ice caps (Heinrich 1988). During HE1, a 5–8° cooling of the Mediterranean surface water may have occurred, which was accompanied by vegetation changes inland

14,500–12,700 BP: Bölling–Alleröd warm phase

12,700–11,600 BP (10,800–9700 cal BC) is the Younger Dryas cold episode, also due to a major threshold related to the warming and melting of the ice cap modifying the thermohaline circulation in the Atlantic (Alley and Clark 1999).

Brutal events during the Holocene

The Holocene starts at ca. 11,600 BP (c. 9700 cal BC). It corresponds to an interglacial period that is still going-on today. From its onset on, it has been shown that yearly temperatures started to rise, followed by an increase in precipitation, inducing rapid changes in the vegetation cover inland. Some ‘brutal events’ evidenced in marine cores (Cortijo et al. 2000) and in an increasing number of continental records in the northern hemisphere produced dry spells in the northern monsoon domain and a deterioration in the cyclonic temperate domain (Alley and Clark 1999). They are pointed out here because of their possible impacts on the climatic trends of the early Holocene in Central and Southeast Anatolia where they might be recorded as sudden and heavy droughts. Such events occurred at 8400–8000 BP (c. 6500–6100 cal BC), at 6600 BP (c. 4700 cal BC), and at 4200–4000 BP (c. 2100–1900 cal BC, cf. Dalfes et al. 1997). The 8200 BP (Gasse 2000) and 4100 BP (Dalfes et al. 1997) events are similar in spread, the 6600 BP event being more rarely evidenced.

General scheme of Holocene vegetation evolution around the Mediterranean

During the last glacial maximum (LGM), all arboreal vegetation formations, including those with typical summer-dry taxa (e.g., Olea, evergreen Quercus) are contracted and most of the northern littoral of the Mediterranean Sea is occupied by herb-dominated steppe in which Artemisia and Chenopods are dominant (Rossignol-Strick 1993). While in the Levant evergreen oak, Pistacia and olive have survived, interior uplands in the Balkans, in northern Greece and in the Black Sea region were probably more important as refugia for northern European deciduous trees. From the early to the mid-Holocene, the herb-steppe is replaced by sub-humid forests, dominated by broad-leaved deciduous trees. During the late Holocene, typically Mediterranean formations as xeric evergreen forests, shrub and heathland, develop around the Mediterranean (Jalut et al. 2000).

Early Holocene climates in the northern hemisphere

In northwestern Europe, the last major shift in climate occurred around the Pleistocene/Holocene boundary. In these regions, the Holocene variations have been of a lower magnitude than those associated with the lateglacial. In low-latitude regions such as intertropical Africa and South Asia, the last permanent climate change occurred more recently, during the mid-Holocene. In these areas the early Holocene is marked by a greatly strengthened monsoon circulation leading to a northward shift of summer rainfall into the North African–Arabian arid zone. On the contrary, during the mid-Holocene (at 6000 BP) the enhanced monsoon weakened, inducing mid-Holocene climatic desiccation. (Gasse 2000). Thus there are marked climatic differences between the two halves of the Holocene in both parts of the hemisphere.

In the Mediterranean latitudes, the early Holocene is marked by mixed contrasts. In some parts of the eastern Mediterranean region, forest re-advance during the Holocene is rather slow, tending to show that early Holocene climate was drier than that of today. For example, in eastern Turkey and western Iran, arboreal pollen only achieved ‘modern’ values between 7500 and 5500 BP (Van Zeist and Bottema 1991). Other parts of the Mediterranean basin show an opposite tendency: in Crete (Bottema 1980) and Israel (Horowitz and Gat 1984), the early Holocene pollen spectra have a greater proportion of deciduous compared to evergreen oak, probably due to an increase in summer rainfall. Furthermore, many sites of the western Mediterranean record a Quercus ilex-type or Q. suber pollen expansion, indicating summers drier than those of today. Thus, a complex patterning of Holocene climate change appears to have occurred across the circum-Mediterranean region during the early Holocene (Roberts et al. 2001a).

The onset of the Holocene (9600 cal BC) and the early Holocene (9600-4600 cal BC) in Central Anatolia according to lake level and lake sediment records

In the eastern Mediterranean deep marine cores (18 records), sapropel layers S1 cover the 10,500–6000 BP timespan. Pollen included in the sapropel layers show climate conditions most favourable for temperate deciduous trees: oak pollen is much higher than that of Artemisia, indicating that annual precipitation was at least 550mm without summer drought in mid-elevation borderlands, while the pollen of Pistacia indicates mild winters in lowlands. According to these records, the early Holocene in the eastern Mediterranean corresponds to the postglacial climatic optimum with the highest moisture and mildest winters (Rossignol-Strick 1995, 1999).

In Cappadocia (Eski Acigöl), the onset of the Holocene is marked by a rapid humidity rise. This rise lasts during the early Holocene reaching a maximum at c. 6500 BP, the late Holocene climate being drier (Roberts et al. 2001b). Such a record is similar to other records from South Spain, Sicily, Israel (Soreq Cave) and Lake Zeribar in the southeastern Taurus (Roberts et al. 2001b). They show also similarities with the Mediterranean sea cores.

In the Konya Plain (Akgöl) and on the Anatolian Mediterranean coast, while the onset of the Holocene is still dry, the early Holocene shows a steady but rather slow ‘wetting up’, also reaching a maximum humidity at c. 6500–6000 BP (Roberts 1983; Bottema and Woldring 1984; Kuzucuoglu et al. 1999; Kuzucuoglu et al. 2001). Here, also, the late Holocene is marked by climatic desiccation (Kuzucuoglu et al. 1998; Fontugne et al. 1999). Such trends are also evidenced in North Africa (Morocco; Lamb et al. 1995; Cheddadi et al. 1998), Israel (Ghab) and Lake Mirabad in the southeastern Taurus (Griffiths et al. 2001).
In conclusion, Anatolian regions seem to have experienced a humidity rise from 10,500 to 6500 BP, within distinctive regional patterns (Kuzucuoglu and Roberts 1997).

II. THE GEOMORPHOLOGICAL MAP: PRESENTATION OF THE LEGEND

Today’s landscapes and climate

Today’s landscapes in Central Anatolia are ‘artificial’ as far as the vegetation cover, soil erosion and water depletion are concerned. They result from the geomorphologic evolution during geological times, determining relief patterns and surface soil characteristics; and during Holocene times, with 1) erosion and accumulation of rocks and soils on the slopes and lowlands, and 2) vegetation distribution. Among the most important factors triggering this evolution are climatic variations, the history of human use of soils and water with the specific importance of heavy transformations occurring during the last century because of an increasing human impact on vegetation, soil and water bodies.

The CANeW region has a continental climate: cold winters, warm summers, strong winds. Amounts of total precipitation vary according to altitude, the mountain ranges receiving more snow and rainfall. On the plateaus and inner plains, precipitation occurs mainly as winter snow and spring rain. Humid air masses mostly come from the north, originating from diverted Mediterranean cyclonic depressions or from Black Sea cold and humid air masses. Northern winds are frequent; southern winds are rarer but violent. High summer temperatures lead to high evaporation rates on bare soils and free water: 800mm/yr in the Konya Plain and 1000mm/yr in the Tuz Gölü plain.

The CANeW region can thus be regarded as incorporating two semi-arid plains and surrounding plateaus (average rainfall <300mm/yr), viz. the Konya Plain and the Tuz Gölü plain, where small variations in the annual precipitation amount and/or seasonal distribution pattern of rain have heavy impacts on non-irrigated crop productivity. Because of such sensitiveness to climatic variations, they belong to what is called ‘climatic margins’. In addition, the CANeW area includes two wetter regions: the Cappadocian plateaus where rainfall reaches 400mm/yr, and the Beysehir region with 500mm/yr. Both regions are less sensitive than the previous ones to variations in precipitation amounts and seasonal patterns. The Beysehir Lake water budget reaches higher amounts because of melt water input through superficial and underground routes.

The legend basis

In agreement with CANeW objectives, the construction of the legend was aimed at obtaining a landscape classification based on the ability of land:

To integrate the evolution of the climate and its effects on water, soil, plants and fauna resources. Area distinction is based here on sensitiveness towards climatic variations. For example, wetlands are distinguished according to their salt content, the dilution/concentration of salt responding to differences in water rarity or abundance.

To offer opportunities to human societies for the use, management and profit of natural resources in a given environment. Classification will thus highlight resources availability for agriculture (light and wet soils), animal husbandry (grasslands) and hunting grounds (forested zones, wetlands). Distinction is thus based on the water and plant availability resulting from either hydro-geological or soil characteristics of the area.

To give access to convenient routes and areas suitable for settling, for production and exchanges. Settlements and activity patterns will rely on access to freshwater and soil-rich areas (for agriculture), grass-rich areas (for animal husbandry) and forest-rich areas (for hunting). Routes may be looked for in the valleys, the highlands and the undulating highlands or plateaux occupying the space between resource-rich areas.

The classification is thus based on two main elements:

1. The relief, allowing the plains that are capable of attracting population involved in agriculture to be distinguished from the highlands, which offer possible pasture and hunting ranges. Within the ‘highlands’, plateaus (which are often water sinks in the CANeW area) and residual old reliefs (which are easily forested) are distinguished from the Taurus range, also on the basis of the access to routes.

2. The hydrology of the outcrops, with reference both to their lithology, whether impermeable, meaning more humid and forested (volcanics), or permeable, meaning dry and steppe-covered (karstic limestones); and to soil coverage, whether impermeable, meaning difficult to cultivate (lacustrine marls), or permeable, meaning more favourable for cultivation (alluvial fans, other fluviatile deposits).

The classes

Wetlands

Freshwater wetlands (which may become slightly brackish in time of decreasing feeding discharge) are located in karstic areas. Perennial lakes (Beysehir Lake, Akgöl Lake, although today becoming seasonal) are distinguished from seasonal lakes (Sugla Lake, although today dried). Freshwater marshes are located at the apex of alluvial fans or in the vicinity of springs (Konya marshes and Sultansazligi plus Develi marshes).

Salt water wetlands (evaporative plains) are either perennial (Tuz Gölü, Yaygölü) or seasonal lakes (Sultaniye, Tuzla Gölü in Sultansazligi).

Plains

They correspond to the largest poljes (i.e., karstic, seasonally inundated plains: Konya, Tuz Gölü, Beysehir, Sugla), their large size being related to tectonic subsidence and faulting. All poljes and plains are covered with impermeable lacustrine marls or clay, except at the mouth of rivers constructing expansive alluvial fans.

Plains not suitable for crops. These are the lake bottom marls (lastglacial in age). Impermeable, nutrient-poor, becoming salty at time of heavy evaporation (De Meester 1970), they are often used as steppe-covered rangelands.

Plains suitable for crops, especially when water is available by flooding. Such plains include, 1) alluvial fans (early Holocene) bordering the lacustrine plains that they tend to fill. In the south of the Konya Plain, the Çarsamba alluvial fan was built in two phases (Roberts et al. 1997): one from 9500–9000 to 7000 BP (from 7600–7100 cal BC to 5100 cal BC), thus corresponding to all ECA periods except ECA I; the second phase is dated c. 5600 BP (c. 3700 cal BC); 2) Quaternary depressions filled with sand and gravel material, possibly weathered. These depressions are either volcanic or karstic in origin.

Highlands

In this class, extensive karstic Neogene lacustrine soft-limestone plateaus are dominant. Because of their karstic nature, they correspond to dry steppe-covered rangelands. On the Obruk Plateau, their surface is hollowed by poljes filled with clay suitable for grass growth and pasture, and by avens, some of which being filled by underground water lakes.

Some old limestone or volcanic residual mounds outcrop out of the Neogene limestone and break the monotony of the landscape. Their slopes were covered by forests during the last century, as can be seen from forest-brown soil trapped in non-eroded spots.
Bordering highlands are naturally forested when not cleared by humans for energy needs and by goat and sheep flocks. They correspond to:

- karstic limestone-rich (mainly) high ranges: the Taurus, from Beysehir to Eregli;

- volcanic heights: the volcanoes and pyroclastite plateaus of Cappadocia; the volcanoes around the Konya Plain; the volcanic complex between Konya and Beysehir.

III. EARLY HOLOCENE CLIMATIC AND ENVIRONMENTAL EVOLUTION IN THE CANEW AREA

Compared to the regional climatic evolution context, the development of the Neolithic in Central Anatolia (c. 10,800–6900 BP) fits into the first half of the Holocene, excluding its onset (11,500–10,800 BP) and its last phase (6900–6500/6000 BP):

ECA I 9th to late 8th mill. cal BC c.10800 – c. 9700 cal BP

ECA II late 8th to 6700/6600 cal BC c.9700–8600/8500 cal BP

ECA III 6700/6600–6000 cal BC 8600/8500–7900 cal BP

ECA IV 6000–5500 cal BC 7900–7400 cal BP

ECA V 5500–5000 cal BC 7400–6900 cal BP

In the following section, description of the environmental context for each period will be based on the characteristics of what is known about 1) evolution of the climate; 2) its effects on environment, i.e. on water, soil, plants and faunal resources; 3) opportunities and constraints for human societies to use, manage and take profit – or suffer – from existing and/or changing natural resources.

Early Holocene environmental evolution in Cappadocia (Eski Acigöl)

The postglacial spread of woodland across interior parts of southwest Asia was relatively slow, with a timing of the first AP maximum appearing to vary between c. 7500 and c. 5500 BP. The delayed advance of arboreal vegetation has generally been attributed to moisture deficiency (Van Zeist and Bottema 1991).

As with many other pollen diagrams in Central and Eastern Anatolia, Eski Acigöl shows a rather slow increase in AP (mainly oak) during the first 3–4 millennia of the Holocene. Some important tree taxa are, however, under-represented palynologically as a result of low pollen production (Woldring, in press). Such taxa are terebinth (Pistacia) and juniper in the early Holocene, Ulmus and Tilia towards the mid-Holocene (Woldring and Bottema, in press).

In the Eski Acigöl record, isotope, diatom, mineralogical and lithological data are all in agreement in showing that the onset at 11,000 BP is extremely rapid and synchronous in both the lake ecosystem and in the composition of non-arboreal regional vegetation. The shift from herb (Artemisia-chenopod) to grass-steppe occurs at precisely the same time as negative shift, both responding to substantial increase in humidity by a more favourable water balance (Roberts et al. 2001b).

Deep and dilute lake conditions remain until 6500 cal BP when lake level starts to fall, with maximum salinity levels probably being achieved around 3000–2000 cal BP. During this period a decline in Pistacia is followed by a permanent reduction in mesophilous trees (elm, hazel). Steppic herbs (Artemisia, chenopods) increase, partly at the expense of grasses. Oak, however, continues at similar levels as before (Woldring, in press). Such an evolution shows a depletion in water availability for trees originating in drier climatic conditions leading to the loss of drought-tolerant trees.

In conclusion, during the ECA I and II stages, the landscape, according to the regional vegetation reconstruction, is that of an oak-terebinth-juniper-grass land, showing optimal conditions for the establishment of permanent settlements like Asikli Höyük. The AP maximum occurs at 8000 BP (c. 6100 cal BC), 3000 years after the start of increase in effective moisture availability at the beginning of Holocene. By the time of the ECA IV and ECA V stages, the landscape has become a mosaic of woodland with mesic trees and shrubs (e.g., hazel) and more open grassland.

Early Holocene environmental evolution in the Konya plain

From 10,800 to 8000 cal BP (i.e. ECA I–III), no lake nor marsh stage has been identified in the available records (Roberts 1983; Karabiyikoglu and Kuzucuoglu 1998; Kuzucuoglu et al. 1999; Fontugne et al. 1999). This absence shows a delay in the reaction to moisture increase also recorded in the pollen curves of the Akgöl Adabag core (Bottema and Woldring 1984). Although the onset of the Holocene shows no soil formation nor marsh and lake appearance, the detailed study of the sediments forming the base of the Çarsamba fan at locations close to Çatalhöyük shows, however, that the first settlements (dated 7500 cal BC, i.e. 9400 cal BP) occurring during the first part of the early Holocene took advantage of a previous start of alluvial deposition by the river (Roberts et al. 1999).

From 8000 to 6200 cal BP (i.e. 6100–4300 cal BC = ECA IV and V), humidity increases in the plain and on the southern slopes as shown by alluvial fan construction and palaeosol developments, favouring vegetation growth and river run-off. This evolution responds to enhanced rainfall recorded also in the eastern Mediterranean sapropel layer (Fontugne et al. 1994; Rossignol-Strick 1995). The rise in forest coverage in the area culminates at 6200 cal BP (4300 cal BC) when AP ratio reaches its maximum in the Adabag pollen record (Bottema and Woldring 1984). The delay in the pollen signal when compared to the start of humidity rise recorded in sediments, is to be interpreted in the light of the discussion above on the Eski Acigöl case. In the Konya Plain, some marshes and small shallow lakes occurred around 6000 cal BP (c. 4000 cal BC) (Kuzucuoglu et al. 1998; Fontugne et al. 1999).
In conclusion, during the ECA I, II and III stages,. according to sediment and vegetation records, the environment reacts slowly to Holocene onset temperature increase. Lake and marsh sediment hiatus indicates some restricted water availability. The increase in runoff and spring water discharges, due to early Holocene increasing precipitation in the Taurus, allows the start of river fan construction at the northern foot of the Taurus range. In the mean time, vegetation expands over the highland slopes; its maximum coverage and variety is reached only at 6200 BP cal. (4300 cal BC), i.e. after the end of the ECA periods.

ECA IV and ECA V are periods of further humidity increase in the plain and on the southern slopes. Growing vegetation allows soil formation. This shows that, even if climatic amelioration started first on the Taurus heights, its effects on the Konya Plain and inland lowlands occurred later with a local precipitation increase delayed in time.

Early Holocene in the Lakes Region (Gölhisar, Sögüt, Karamik, Beysehir)

Pollen results (Bottema and Woldring 1984) also show that the maximum AP curves happened during the early Holocene and not at its onset. This delay in woodland expansion is usually explained by differences in the migration of trees or by differences in the composition of tree species (Eastwood 1999).

During ECA I and II, local environments are probably dominated by a low herb cover (grass) with arboreal taxa most likely confined to restricted locations. During ECA III, IV and V, i.e. from 8500 cal BP (6700 cal BC) on, arboreal pollen show a true woodland, composed of mixed forest comprising oak, pine and juniper. Only at 3500 cal BP does human occupation become visible in the spectra (Bottema and Woldring 1984; Eastwood 1999).

IV. Comparing the geomorphological and archaeological maps

The aim of this comparison is to point out the possible relations between the settlement distribution patterns of ECA periods and the available environmental data on early Holocene evolution in Konya, Cappadocia and other areas, in order to suggest possible environmental constraints on the evolution of human societies during each ECA period. (I will not refer here as to the question of how confident we can be when looking at the density and location of sites with regard to possible insufficient data collection, especially because, for example, of site burial by younger sediments (see Summers, this volume). In this respect, the following discussion may end up raising non-pertinent questions.)

ECA I – 9th to late 8th mill. cal BC / c. 10800 – c. 9700 cal BP.

In the Konya plain the period is locally dry; however, running and spring water is discharged from exogenic sources (Taurus range) while vegetation slowly increases. In Cappadocia, humidity rises substantially and regional vegetation is an oak-terebinth-juniper-grass land. The map shows two distinctive settlement types:

- one type is related to springs and river mouths along the edges of the mountain ranges:
a.Suberde (Sugla); Pinarbasi (South Konya) and Canhasan III (South Konya),
b.Hacibeyli and Toparin Pinar (Sultansazligi);

- another type is related to the obsidian production centres in the Göllüdag area. (Some of the sites in northern Cappadocia could potentially be linked to routes reaching the Göllüdag obsidian workshops northwards along a network organised along the Melendiz and Karasu–Demirci river valleys towards the Tuz Gölü and the Kizilirmak basins. Another point is that, according to the distribution of sites, the absence of sites at some locations is surprising, viz. at the mouth of the Melendiz entering the Tuz Gölü plain where salt must have been attractive to Neolithic people, and in the Beysehir Lake area, since there is a site identified in the Sugla Lake plain (i.e. Suberde).)

ECA II – late 8th to 6700/6600 cal BC / c. 9700–8600/8500 cal BP.

Climate and environment show no striking differences from the previous period, except that, at least in the Konya plain, the humidity rise starts to affect also local grounds (growth of endogenic resources). In Cappadocia regional humidity rises in continuity with the previous period. The map shows a settlement distribution pattern distinct from the ECA I one, with:

1. the expansion of river and spring-related settlements in the Konya–Beysehir region:
new sites located at the main outputs of underground water (springs): Beysehir and Sugla,
new sites located close to the mouth of rivers in the southern plains;

2. the decreased number of obsidian workshops in Cappadocia;

3. site locations at the edges of the Cappadocian plateaus, possibly related to ‘routes’ towards neighbouring areas:
three sites on the route northwards from the Melendiz to the Tuz Gölü
one remaining site northwards to the Kizilirmak valley.

ECA III – 6700/6600–6000 cal BC / 8600/8500–7900 cal BP.

In the Konya region as well as in Cappadocia the forests expand, as does also humidity availability. The map evidences a sharp increase in site numbers, with new and more diversely orientated human occupation:

1. In the Beysehir Lake surroundings all new sites are located on karstic spring spots or at the mouth of rivers on the alluvial fans.

2. The only sites abandoned on the map are the obsidian workshops; the location of the Tepecik and Çiftlik sites may be interpreted as related to the development of agriculture more than to the proximity to the obsidian sources.

3. Some new sites can be possibly interpreted:

- either as new ‘route’-type locations, possibly showing human activities expansion in the highlands as well as the growth of exchanges between territories,

- or as expansion of exploited land due to climatic long-term trends towards amelioration.
This observation concerns sites north and south of the Alacadag volcano, between Beysehir and Konya; in the south of the Konya Plain between the Çarsamba and the Kilbasan fans; between the Bor corridor and Cappadocia; over the Obruk Plateau; and between the Sultansazligi Plain and the Kizilirmak valley.

In conclusion, expansion of site numbers during this period shows the opening of new territories in the highlands and lowlands. These new settlements seem to take advantage of increasing water availability: as the climatic conditions became more and more advantageous, the societies were also able to take advantage of them. This also means that together with the increase in resources a population increase and a human mobility expansion seem to have occurred.

ECA IV – 6000–5500 cal BC / 7900–7400 cal BP.

In the Konya Plain humidity still increases, favouring vegetation growth (as shown by dated palaeosol) on forested slopes. In Cappadocia, the landscape has evolved to form a mosaic of woodland with mesic trees and shrubs and more open grassland. Climatic conditions, whether regional or local, are approaching the mid-Holocene climatic optimum. The map shows a site concentration affecting both occupation density and type of territory used by the various human communities. Settlement distribution remains regular but it is more scattered than in the previous period. Possible causes of such a change in pattern are most probably not mainly due to climatic and environmental constraints, but to changes in socio-economic practices. Isolation of communities together with abandonment of routes may be related to more autarkic agricultural practices in a context of limited exchange. Territories controlled by a single site may have grown larger and more complex (complementary uses?) possibly concentrically controlled and managed around a centre-site, leaving wide open spaces in-between site-territories (pasture or hunting grounds?).

ECA V – 5500–5000 cal BC / 7400–6900 cal BP.

This period corresponds to the Holocene climatic optimum in Anatolia. It precedes the start of the mid-Holocene climatic change (desiccation), which starts in palaeoenvironmental records at ca. 6200 cal BP in the Konya Plain and at 6500 cal BP in Cappadocia. The map shows the ‘abandonment’ of the land with the last ‘colonised land’ being abandoned first (Beysehir, Sugla and Beysehir–Sugla corridors). A few remaining sites are located on alluvial fans at the mouth of Tauric rivers with high (spring?) discharge (2 sites), and at the northern edge of the central plateaus (3 sites).

According to palaeoenvironmental records, such newly occurring site decrease is not a response to climatic stress since the period corresponds to the Holocene climatic optimum. The end of ECA V even occurs before the beginning of the mid-Holocene climatic deterioration (desiccation). In this context, human societies have reacted to triggering factorsother than the rarity of resources.

CONCLUSIONS

Environment provides mainly three types of resource-rich elements usable by human economy and settlements, activities and movements : air, water and soil, interacting in turn with fauna and flora. The quantity and availability of such resources depend directly on climatic factors, as much as on surface and subsurface characteristics of rock and soil covers.

However, the availability – and thus the ‘value’ – of water and soil resources also depends on the techniques and social organisation involved for their exploitation and management. In relation with their social organisation, human societies are able or unable to take advantage, react, adapt themselves, or ignore changing environments. This means that the constraints within which human exploitation and management of resources develop, depend as much on the environment as on the socio-economic organisation of the human societies concerned.

Keeping in mind that water and soil availability, as well as plant expansion also depend on human choices, techniques and organisation involved for their agricultural exploitation, a dynamic interpretation of the CANeW maps can only rely on the confrontation of both environmentalists’ and archaeologists’ knowledge of what may have happened during each period. The CANeW meeting is thus a unique and exciting opportunity to discuss the evolution and relationships between elements that are fundamental to our understanding of the history of Neolithic societies in Central Anatolia in times of changing climate.

References

Alley, R. and P. Clarck, 1999. The deglaciation of the northern hemisphere: a global perspective. Annual Reviews of Earth and Planetary Sciences 27, 149–182
Bottema, S., 1980. Palynological investigations in Crete. Review of Palaeobotany and Palynology 31, 193–217
Bottema, S. and H. Woldring, 1984. Late Quaternary vegetation and climate of southwestern Turkey, part II. Palaeohistoria 26, 123–149
Bottema, S. and W. van Zeist, 1981. Palynological evidence for the climatic history of the Near East, 50,000–6,000 BP. In J. Cauvin and P. Sanlaville [eds.]. Préhistoire du Levant. Chronologie et organisation de l’espace depuis les origines jusqu’au VIe millénaire. Paris:Éditions du CNRS, 111–132
Cheddadi, R., H. Lamb, J. Guiot and S. van der Kaars, 1998. Holocene climatic change in Morocco: a quantitative reconstruction from pollen data. Climate Dynamics 14, 883–890
Cortijo, E., L. Labeyrie, M. Elliot, E. Balbon and N. Tisrenat, 2000. Rapid climatic variability of the North Atlantic Ocean and global climate: a focus of the IMAGE program. Quaternary Science Reviews 19, 1–5, 227–242
Dalfes, N., G. Kukla and H. Weiss [eds.], 1997. Third millennium BC climate change and Old World collapse, NATO ASI Series 1, volume 49
Eastwood, W., N. Roberts, H. Lamb and J. Tibby, 1999. Holocene environmental change in southwest Turkey: a palaeoecological record of lake and catchment-related changes. Quaternary Science Reviews 18, 671–696
Fontugne, M., M. Arnold, L. Labeyrie, M. Paterne, S. Calvert and J.-C. Duplessy, 1994. Palaeoenvironment, sapropel chronology and Nile river discharge during the last 20,000 years as indicated by deep-sea sediment records in the eastern Mediterranean. In O. Bar-Yosef and R. Kra [eds.]. Late Quaternary chronology and paleoclimates of the eastern Mediterranean. Radiocarbon, 75–88
Fontugne, M., C. Kuzucuoglu, M. Karabiyikoglu, C. Hatté and J.-F. Pastre, 1999. From Pleniglacial to Holocene: a 14C chronostratigraphy of environmental changes in the Konya Plain, Turkey. Quaternary Science Reviews 18, 573–591
Gasse, F., 2000. Hydrological changes in the African tropics since the Last Glacial Maximum. Quaternary Science Reviews 19, 189–212
Griffiths, H., A. Schwald and L. Stevens, 2001. Environmental change in southwestern Iran: the Holocene ostracod fauna of Lake Mirabad. Holocene 11, 757–764
Heinrich, H., 1988. Origins and consequences of ice-rafting in the Northeast Atlantic ocean during the past 130,000 years. Quaternary Research 29, 142–152
Horowitz, A., and J. Gat, 1984. Floral and isotopic indications for possible summer rains in Israel during wetter climates. Pollen and Spores 26, 61–68
Jalut, E., A. Esteban Amat, L. Bonnet, T. Gauquelin and M. Fontugne 2000. Holocene climatic changes in the western Mediterranean from south-east France to south-east Spain. Palaeogeography, Palaeoclimatology, Palaeoecology 160, 255–290
Kallel, N, M. Paterne, J.-C. Duplessy, C. Vergnaud-Grazzini, C. Pujol, L. Labeyrie, M. Arnold, M. Fontugne and C. Pierre, 1997. Enhanced rainfall in the Mediterranean region during the last sapropel event. Oceanologica Acta 20, 697–712
Karabiyikoglu, M. and C. Kuzucuoglu, 1998. Late Quaternary and environmental evolution of the Konya basin. MTA and TÜBITAK Report, Ankara
Kuzucuoglu, C., A. Emery-Barbier, M. Fontugne and S. Kunesch, 2001. The Öküzini marshes: a new palaeoenvironmental record on the Anatolian Mediterranean coast. PAGES conf. The PEP II transect: past variability through Europe and Africa. Aix-en-Provence, 107–108
Kuzucuoglu, C., M. Fontugne, M. Karabiyikoglu and C. Hatté, 1999. Evolution de l'environnement dans la plaine de Konya (Turquie) pendant l'Holocène. In M. Otte [ed]. Anatolian prehistory: at the crossroads of two worlds. Liège University, 605–624
Kuzucuoglu, C., R. Parish and M. Karabiyikoglu, 1998. The dune systems of the Konya Plain (Turkey). Their relation to the environmental changes in Central Anatolia during the late Pleistocene and Holocene. Geomorphology 23, 257–271
Kuzucuoglu, C. and N. Roberts, 1997. Évolution de l’environnement en Anatolie de 20,000 à 6,000 BP. Paléorient 23/2, 7–24
Lamb, H., F. Gasse, A. Benkaddour, N. el-Hamouti, S. van der Kaars, W. Perkins, N. Pearce, and N. Roberts, 1995. Relation between century-time scale Holocene arid intervals in tropical and temperate zones. Nature 373, 134–137
Meester de, T. [ed.]. 1970. Soils of the Great Konya Basin, Turkey. Wageningen: Centre for Agricultural Publishing and Documentation
Roberts, N., 1983. Age, palaeoenvironments, and climatic significance of late Pleistocene Konya Lake, Turkey. Quaternary Research 19, 154–171
Roberts, N., S. Black, P. Boyer, W. Eastwood, H. Griffiths, H. Lamb, M. Leng, R. Parish, J. Reed, D. Twigg and H. Yigitbasioglu, 1999. Chronology and stratigraphy of late Quaternary sediments in the Konya Basin, Turkey: results from the KOPAL project. Quaternary Science Reviews 18, 611–630
Roberts, N., M. Meadows and J. Dodson, 2001a. The history of mediterranean-type environments: climate, culture and landscape. Holocene 11, 631–534
Roberts, N., J. Reed, M. Leng, C. Kuzucuoglu, M. Fontugne, J. Bertaux, H. Woldring, S. Bottema, S. Black, E. Hunt and M. Karabiyikoglu, 2001b. The tempo of Holocene climatic change in the Eastern Mediterranean region : new high-resolution crater-lake sediment data from central Turkey. Holocene 11, 721–736
Rossignol-Strick, M., 1993. Late Quaternary climate in the eastern Mediterranean region. Paléorient 19, 135–152
Rossignol-Strick, M., 1995. Sea-land correlation of pollen records in the eastern Mediterranean for the Glacial–Interglacial transition: biostratigraphy versus radiometric time-scale. Quaternary Sciences Reviews 14, 893–915
Rossignol-Strick, M., 1999. The Holocene Climatic Optimum and pollen records of Sapropel 1 in the eastern Mediterranean, 9000–6000 BP. Quaternary Science Reviews 18, 515–530
Woldring,, H., in press. Early-Holocene vegetation of Central Anatolia and the impact of farming. In R. Cappers and S. Bottema [eds.]. The dawn of farming in the Near East. Berlin: ex oriente
Woldring, H. and S. Bottema, in press. The vegetation history of east-Central Anatolia in relation to archaeology: the Eski Acigöl evidence compared with the Near Eastern environment. Palaeohistoria 43–44
Zeist, W. van and S. Bottema, 1991. Late Quaternary vegetation of the Near East. Beihefte zum Tübinger Atlas des Vorderen Orients, Reihe A18. Wiesbaden: Dr. Ludwig Reichert Verlag

__________________________

Go to the related debate (Panel : Henk Woldring, Douglas Baird, Geoffrey Summers, Sevil Gülçur)