Did prehistoric landscape management retard the post-glacial spread of woodland in Southwest Asia?

Did prehistoric landscape management retard the post-glacial spread of woodland in Southwest Asia? Introduction Archaeological research has revealed the detailed emergence andspread of Neolithic and later farming societies in Southwest Asia.Evidence from excavated seed remains and animal bones indicates that bythe Pre-Pottery Neolithic B (PPNB PPNB Pre-Pottery Neolithic B (era)), c. 7000 cal BC, the main transitionfrom foraging to farming was complete in the core region ofdomestication domesticationProcess of hereditary reorganization of wild animals and plants into forms more accommodating to the interests of people. In its strictest sense, it refers to the initial stage of human mastery of wild animals and plants. (Bar-Yosef & Belfer-Cohen 1992; Harris 1998). Thison-site evidence shows how human action transformed a selected range ofplants (notably cereals and pulses) and animals (sheep, goat, cattle andpig) from wild to domesticated do��mes��ti��cate?tr.v. do��mes��ti��cat��ed, do��mes��ti��cat��ing, do��mes��ti��cates1. To cause to feel comfortable at home; make domestic.2. To adopt or make fit for domestic use or life.3. a. (Zohary & Hopf 2001) On-sitearchaeological data provide abundant evidence for the human exploitationof many other natural resources such as wood, clay, obsidian and otherstone materials; raw materials which were sometimes transported overdistances of >100 km. The size and number of settlement sites alsoincreased significantly during Neolithic/Chalcolithic times, reflectingat least an order of magnitude A change in quantity or volume as measured by the decimal point. For example, from tens to hundreds is one order of magnitude. Tens to thousands is two orders of magnitude; tens to millions is three orders of magnitude, etc. increase in total human population in theregion from the preceding Mesolithic/Epipalaeolithic period. Despite this wealth of archaeological data for greater humanutilization of the natural environment, off-site (i.e.non-archaeological) evidence for prehistoric human impact in this regionhas remained elusive. Willis & Bennett (1994) highlighted thesignificant discrepancy in time between the arrival of Neolithicagriculture, as testified archaeologically, and the first appearance ofcultural indicators in pollen diagrams from Greece and the Balkans, atime delay amounting to >2000 years. In fact, unambiguous,large-scale human impact and the creation of cultural landscapes hasbeen inferred from East Mediterranean pollen diagrams only during BronzeAge and later times (Bottema & Woldring 1990; Eastwood et al. 1998).It has therefore been assumed that the magnitude ofNeolithic-Chalcolithic agriculture was not sufficient to be registeredclearly in pollen diagrams from southwest Asia. Pre-Bronze Age environmental impacts are difficult to identify withcertainty from east Mediterranean pollen records for several reasons.First, the emergence of farming communities early in the Holocene periodmeans that a pre-Neolithic base-line `primaeval' vegetation may nothave existed for comparison, unlike--for example--in northwest Europe.Consequently it is conceivable that significant human impacts began evenbefore vegetation had completed its adjustment to the major shift inclimate at the end of the last glacial period (cf. Pullar 1977). Second,many palynological indicators of cultural activity are present naturallyin southwest Asia. Cerealia-type pollen, for example, which in northernEurope has been used to identify prehistoric crop husbandry (e.g.Simmons & Innes 1996), is also produced by wild cereal and othergrasses in the Mediterranean region, and so does not provide adiagnostic indicator of prehistoric farming activity (Bottema 1992).Precisely because Southwest Asia was a centre of crop domestication,wild and domesticated plants cannot easily be distinguished from pollendata. Third, many fruit trees and other useful plants are notwind-pollinated, and they are consequently invisible in pollen diagrams(Woldring & Cappers 2001). Early Holocene vegetation and climate Pollen diagrams from southwest Asia (FIGURE 1) show that the LastGlacial Maximum The Last Glacial Maximum (LGM) refers to the time of maximum extent of the ice sheets during the last glaciation (the W��rm or Wisconsin glaciation), approximately 20,000 years ago. This extreme persisted for several thousand years. , around 20,000 years ago, was marked by reduced Arboreal arborealpertaining to trees, treelike, tree-dwelling. Pollen (AP), and that tree cover increased during the course of the lastdeglacial transition (van Zeist & Bottema 1991). This expansion tookplace earliest in the coastal montane mon��tane?adj.Of, growing in, or inhabiting mountain areas.[Latin montnus, from m regions of the Levant and westernTurkey, where there was a clear increase in trees during Late-Glacialtimes (van Zeist & Bottema 1991; Kuzucuoglu & Roberts 1998).This was followed by a contraction of woodland probably coeval co��e��val?adj.Originating or existing during the same period; lasting through the same era.n.One of the same era or period; a contemporary. with theEuropean `Younger Dryas' (GS-1) cold stage, and a second expansionduring the early Holocene. However, in interior sites, where the climateis cooler, drier or both, AP shows a significant rise only after theonset of the Holocene. Moreover this increase was rather slow, so thatmaximum AP values were only achieved in mid-Holocene times. In southeastAnatolia and the western Zagros, this process was not complete until c.4000 cai sc (FIGURE 2). The process of woodland expansion wastime-transgressive across interior parts of southwest Asia. Instead ofwoodland, the glacial-age Artemisia-chenopod steppe steppe(stĕp), temperate grassland of Eurasia, consisting of level, generally treeless plains. It extends over the lower regions of the Danube and in a broad belt over S and SE European and Central Asian Russia, stretching E to the Altai and S to was replaced inthese regions during Neolithic times by open grass-dominated savannaParkland. [FIGURES 1-2 OMITTED] Until now, the delayed re-establishment of trees in these regionshas been interpreted primarily as a result of a slow increase inprecipitation during the early Holocene, implying that climaticconditions were drier than at present in Anatolia and Western Iranduring Neolithic times (van Zeist & Bottema 1991; Roberts &Wright 1993). An alternative source of palaeoclimatic data to pollenanalysis comes from water-level and salinity fluctuations in lakebasins. Non-outlet lakes, in particular, respond sensitively to shiftsin regional water balance, with periods of hydrological surplus(precipitation >evaporation) being marked by rising water levels andfreshening waters, and periods of hydrological deficit experiencing lakeregression and salinization. Former lake-level and salinity fluctuationscan be reconstructed from a range of proxy data, many of which (e.g.diatoms, stable isotopes) can be derived from the same sediment coresused for pollen analysis. Several such sequences in southwest Asia havenow also been investigated for their lake-level histories (FIGURE 2). Ifslow Holocene expansion of woodland vegetation was the result ofclimatic conditions drier than today, we would expect them to have beena coeval phase of lower lake levels and/or higher salinities than atpresent at these sites. In fact, these lakes often show the reverse,with marked contrasts between their vegetation and lake-level historiesat this time. At Lake Van, isotopic and geochemical analysis indicates greatesteffective moisture during the early-mid Holocene (6000-2000 varve varve,in geology, pair of thin sedimentary layers formed annually by seasonal climatic changes. Usually found in glacial lake deposits, varves consist of a coarse-grained, light-colored summer deposit and a finer-grained, dark-colored winter deposit formed when fine yearsBC; see below for further discussion of chronology) (Lemcke & Sturm1997). By contrast, oak pollen does not reach maximal Holocene valuesuntil after 4000 varve years BC (van Zeist & Woldring 1978; Wick etal. in press). This contrast is even more marked at Eski Acigol incentral Turkey, which lies close to As Ikli hoyuk and other earlyNeolithic settlements, and to the Ciftlik obsidian source. During theearly Holocene (c. 9500-c. 4500 cal BC) this lake was deep enough to bestratified stratified/strat��i��fied/ (strat��i-fid) formed or arranged in layers. strat��i��fiedadj.Arranged in the form of layers or strata. and form varved sediments, but during the second half of theHolocene the water level fell, varve formation ended and the lake becameperiodically saline. These trends are confirmed by proxy data includingdiatoms, mineralogy and stable isotopes (Roberts et al. 2001). On theother hand, the first maximum in oak values at Eski Acigol was notreached until c. 6000 cal BC. Oak pollen declined sharply during thelater 3rd millennium BC, associated with Bronze Age deforestation deforestationProcess of clearing forests. Rates of deforestation are particularly high in the tropics, where the poor quality of the soil has led to the practice of routine clear-cutting to make new soil available for agricultural use. , atrend probably accentuated by climatic desiccation des��ic��ca��tionn.The process of being desiccated.desic��ca (FIGURE 2). Similar contrasts between pollen and lake geochemistry/isotopesalso apply to the records from Zeribar (van Zeist & Bottema 1977;Stevens et al. 2001) and Urmia (Kelts & Shahrabi 1986; Bottema1986). In contrast, comparable evidence of a discrepancy between earlyHolocene lake level and vegetation cover does not occur in the centraland western Mediterranean, where Neolithic communities only developedafter forests were established. Potential explanations There is a range of potential reasons for the discrepancy betweenthe inferred wetter early--mid-Holocene climatic conditions and thedelayed rise in AP in interior regions of Southwest Asia, compared withareas nearer the coast. While they are evaluated individually below, itis likely that several of them acted in concert. 1 Climatic inference. While any individual lake level/salinityrecord may have been deformed by non-climatic factors (e.g. tectonics)it is much less likely that this applied to a suite of separate lakesediment records. Evidence for a wetter early Holocene climate in theEast Mediterranean region also comes from other climate `archives'such as cave speleothems (Bar-Matthews et al. 1997). Of course, changesin the overall mean annual water balance may disguise changes in theseasonal distribution of precipitation that may have been importantphytologically. 2 Pollen representation and interpretation. Pollen diagrams by nomeans faithfully represent the former vegetation. Pollen assemblages arebiased by differential production, dispersal and preservation processes,by difficulties of separating regional from local pollen sources, and bystatistical artefacts when using percentage data. In the EastMediterranean region, a striking example of these difficulties isprovided by Pistacia (probably P. atlantica or terebinth terebinth(tĕr`əbĭnth)or turpentine tree,small deciduous tree (Pistacia terebinthus) of the family Anacardiaceae (sumac family), native to the Mediterranean region. ). This tree isunder-represented palynologically, and a substantial correction factorwould need to be applied in order to estimate its former abundancecompared, for example, to oak. Pistacia in fact displays a ratherstriking early Holocene maximum in both marine cores (Rossignol-Strick1999) and in lake sediment records from interior regions (FIGURE 2).When the under-representation of Pistacia and also Amygdalus is takeninto account, the early Holocene oak-terebinth-almond parkland of innerAnatolia and the western Zagros may therefore have had a denser treecover than might first appear from summary %AP data. 3 Dating and correlation. The chronologies of the sites discussedhere contain important uncertainties. At Eski Acigol, the [sup.14]Cdates have been `aged' by volcanic out-gassing of C[O.sub.2], andthe core chronology is currently based on U-series dates. The Vanchronology derives from varve counting which places the onset of theHolocene at 8500 BC rather than the expected c. 9500 cal BC, suggestinga significant under-estimation of the true age. Other [sup.14]C lakechronologies may also require future revision. These uncertaintieshinder precise correlation between lake-sediment and archaeological datain Southwest Asia. On the other hand, they do not negate the contrastbetween lake palaeohydrology and pollen-inferred vegetation histories,because they are derived from the same sedimentary archives. 4 Edaphic e��daph��ic?adj.1. Of or relating to soil, especially as it affects living organisms.2. Influenced by the soil rather than by the climate. and migrationary lags. Pollen records extending back tothe Last Glacial Maximum indicate that most Pleistocene forest refugia In the most basic biological sense refugia (singular: refugium) are locations of isolated or relict populations of once widespread animal or plant species. This isolation (allopatry) can be due to climatic changes or human activities such as deforestation and over-hunting. were located in regions near to the Mediterranean and Black Sea coasts(e.g. Ghab), while interior areas of Southwest Asia (e.g. Zeribar) wererelatively treeless at this time. The recolonization Re`col`o`ni`za´tionn. 1. A second or renewed colonization. of inlandlandscapes by trees during the subsequent climatic amelioration maytherefore have involved them spreading up to 1000 km from coastalregions, a process that may have been further delayed by the need forpostglacial post��gla��cial?adj.Relating to or occurring during the time following a glacial period.postglacial?Relating to or occurring during the time following a glacial period.Adj. 1. soil development. On the other hand, the establishment offorest cover was achieved soon after the start of the Holocene in mostof Mediterranean Europe. It is not clear why migrationary lags alonewould have delayed forest re-advance into the interior of Southwest Asiauntil mid Holocene times, when they did not do so in other comparableregions (e.g. Iberia). Human impact While some or all of the above factors may have been responsiblefor the delayed re-afforestation of Southwest Asia's oak-terebinthparkland zone, there are grounds for believing that prehistoric humanimpact may have provided a further and possibly critical role in earlyHolocene landscape ecology. On a priori grounds, oak-terebinth parkland is the most likelyecological zone to have been subject to significant Neolithic impact.This is one of the main natural habitats for wild cereals (Hillman 1996;Harris 1998) and for most of the ungulates that were domesticated.Relatively intensive human exploitation of this landscape zone isconfirmed by the high density of Neolithic settlement sites here, withsite numbers decreasing both towards the arid zone (e.g. Syrian Desert)and towards wetter, mesic mes��ic?adj. EcologyOf, characterized by, or adapted to a moderately moist habitat.mesic?Relating or adapted to a moderately moist habitat. forest regions (FIGURE 1a). The near-absenceof Neolithic sites from the summer-green forests of northern Anatoliahas been noted previously (Roberts 1982), and this pattern has only beenreinforced by subsequent archaeological survey work. These areas mayhave been avoided partly because of the need for clearance of mesicwoodlands that were often well-established before the arrival ofNeolithic communities in northern and western Anatolia (Bottema et al.1994) and partly because agro-climatic conditions were less well-suitedto crop plants adapted to seasonal drought. Maximum human populationdensity therefore appears broadly to have coincided with the savannaparkland zone, across which Neolithic settlement and trees weresimultaneously expanding during the early part of the Holocene. Neolithic impact on the tree cover could have taken several forms,including clearance for crop cultivation, cutting for fuel and timber(while selectively preserving fruit-bearing trees), pressure fromgrazing/browsing animals, and burning. The first of these is probablythe least likely to have had effects visible at a landscape scale. Ithas been proposed that many of the earliest Neolithic farmingcommunities in Southeast Europe and Anatolia practised a relativelyintensive horticulture on alluvial soils (Sherratt 1980; van Andel &Runnels 1995). Such garden-scale cultivation would not have requiredlarge-scale clearance of the pre-existing natural vegetation, and theseactivities would not have been detectable in pollen sequences unlessthey were located in close proximity to prehistoric settlements (Willis& Bennett 1994). Most Neolithic crops (wheat, barley etc.) areseverely under-represented palynologically, and are unlikely to berecognized in pollen diagrams located away from prehistoric sites(Bottema 1992). The use of woodfuel for domestic activities (heat, light, cook.lng)and lime-plaster manufacture is testified by macro-charcoal remains fromNeolithic sites throughout Southwest Asia (e.g. Willcox 1996; Asouti& Hather 2001), with oak most commonly represented. The use ofanimal dung rather than wood as a source of fuel at some sites has beeninterpreted as evidence that the latter was already in short supply(Miller 1996). A lack of suitable available timber might also be arguedto be one reason for the predominant use of mudbrick for Neolithic houseconstruction in the eastern Mediterranean region. Archaeobotanicalmacrofossil mac��ro��fos��sil?n.A fossil large enough to be examined without a microscope. remains further confirm that `wild orchard' trees; suchas hackberry hackberry:see elm. , hawthorn and Prunus Prunusa genus of trees in the family Rosaceae. The seeds of these trees contain cyanogenetic glycosides which are potentially poisonous. The fruit pulp appears to quite safe. The glycosides are amygdalin, prunasin, prulaurasin. , were already common in the centralAnatolian savanna parkland before oak trees reached their maximumdensity (Woldring & Cappers 2001). Pressure on trees may also have come from grazing and browsinganimals, both domestic and wild. Numbers of the latter (e.g. deer,aurochs aurochs:see cattle. aurochsor aurochExtinct wild ox (Bos primigenius) of Europe, the species from which cattle are probably descended. The aurochs survived in central Poland until 1627. It was black, stood 6 ft (1. , equid equidsee equidae. ) are likely to have expanded substantially during theHolocene as plant biomass increased. In the southern Levant,K6hler-Rollefsn & Rollefson (1990) have hypothesized that evenduring the Pre-Pottery Neolithic substantial damage to vegetation wascaused by domestic goats, at least in the vicinity of settlement sites.In the open oak-terebinth parkland zone farther north, the main impactof large herbivores may have been to maintain open habitat vegetationstructure and prevent the encroachment of closed woodland, by preventingthe establishment of tree seedlings. Early Holocene pollen diagrams fromthis zone include several ruderal taxa, such as Rumex, but the typicalweed flora associated with large-scale livestock herding did not emergeuntil the Bronze Age (Behre 1990). The major means by which prehistoric populations were able tomanipulate natural vegetation was by the use of fire for controlledburning. Of the different mechanisms available, this is the one mostlikely to have had an impact at a landscape rather than a local scale.The role of past human groups in altering fire regimes has received muchattention in regions such as Australia, the Americas and even intemperate Europe (e.g. Pyne 1995; Mason 2000) but--apart fromLewis' (1972) study--very little in eastern Mediterraneanlandscapes. The earliest proposed anthropogenic an��thro��po��gen��ic?adj.1. Of or relating to anthropogenesis.2. Caused by humans: anthropogenic degradation of the environment. burning is from at least40,000 BP and has been linked to the impact of aboriginal human arrivalsin Australia, where the vegetation may have been especially sensitive toan additional ignition source (Kershaw et al. 1997). Micro-charcoal andpollen records from sites such as Lynch's Crater have been inferredas indicating not only increased fire frequencies after the first humanarrivals, but also major landscape-scale consequences for forest coverand composition, with fire-adapted eucalypts replacing Araucaria araucariaAny pinelike coniferous plant (see conifer) of the genus Araucaria (family Araucariaceae). Found in South America, the Phoenix Islands, and Australia, araucaria trees are magnificent evergreens, with whorled branches and stiff, flattened, pointed leaves. .Examples such as this should, perhaps, give cause to re-evaluate therole of anthropogenically induced burning in landscapes with a longerantiquity of human occupation. Fire management is most often associated with societies reliant onhunting and gathering, pastoralism PastoralismArcadiamountainous region of ancient Greece; legendary for pastoral innocence of people. [Gk. Hist.: NCE, 136; Rom. Lit.: Eclogues; Span. Lit. or shifting cultivation. However,low-temperature ground fires have advantages for any human group wantingto prevent the encroachment of scrub or woodland, and maintain an opensavanna-type landscape of grassland with trees. This would beparticularly appropriate for achieving a fire sub-climax with a groundstorey of annual and perennial grasses, well-suited for grazing sheepand other livestock. Perennial bunch grasses, such as Stipa lagascae,Oryzopsis holciformis, Hyparhennia hirta and Andropogon distachyium,would have been notable `fire followers', responding particularlyto late-season burning (Naveh 1974; G. Hillman pets. comm.). Burning maynot have assisted annual cereal grasses, as wild emmet and barley grainsare killed by fire (Blumler 1991), but if grains are brought fromelsewhere and sown on burnt ground they grow well. Fire would also havebroken up the soil surface, requiring less use of labour-intensive hoesor digging-sticks, and aiding the growth of seedlings (Lewis 1972: 214).Alluvial horticulture would have been immune to, and unaffected by,dryland fires. Regional fire histories can be recorded through micro-charcoalprofiles in lake sediment sequences (e.g. Clark & Patterson 1997).Although very few such charcoal sequences exist for Southwest Asia,those currently available appear to give support to maximum firefrequencies occurring during the early Holocene (Yasuda et al. 2000;Wick et al. in press; W. Eastwood pets. comm.). In the Ghab record(FIGURE 2) micro-charcoal concentrations increase abruptly after the endof the Pleistocene and then decline gradually towards mid Holocene(Bronze Age) times. This core site lies downwind of denseOro-Mediterranean forest, and so is not ideally situated to establishthe Neolithic fire ecology of the drier oak-terebinth parkland zone.Nor, for the various reasons outlined above, is it easy to accept Yasudaet al.'s conclusion that PPNB farmers carried out large-scaledestruction of already established oak-cedar forests for agriculturalland and timber. Nonetheless these data suggest that fire regimes mayhave changed systematically during the later prehistoric period inSouthwest Asia in response to the changing availability of plantbiomass, the emergence of summer-dry climates (cf. McCorriston &Hole 1991) and fire-management practices. Increased late-season burningin the fire-sensitive oak-terebinth parkland zone during the Neolithicperiod could have played a key role in delaying the re-establishment ofwooded vegetation until the late Chalcolithic or Early Bronze Age. Conclusion The relative roles of human impact and climate change represent along-standing debate about how best to interpret vegetation changesduring mid to late Holocene times in both northern and MediterraneanEurope (e.g. elm decline, beech rise). Matters have been less contestedfor the early Holocene, where it has been assumed that observed changesin forest composition reflected primarily natural agencies (climate,biogeographical bi��o��ge��og��ra��phy?n.The study of the geographic distribution of organisms.bio��ge��og spread, disease) rather than cultural ones. Any humanimpact on Mediterranean vegetation at this time has generally beenassumed to be localized and modest (van Zeist & Bottema 1991). In the western Mediterranean and in central and northern Europe,the `accepted' history is one in which human impact on naturalenvironments increases through time. Even if this includes somereversals and secondary afforestation, the overall trend in most pollendiagrams is one that moves away from primaeval wildwood and towardsculturally created landscapes. The idea that human landscape managementwas implicated during a time of increasing woodland cover and AP risetherefore seems counter-intuitive. However, Neolithic traditions inSouthwest Asia--like Palaeolithic/ Mesolithic hunter-gatherer fire andforest management practices--may have led to modification rather thantransformation of pre-existing ecosystems, lust as the archaeologicalrecord tells us that environment-culture relations in the oak-terebinthparkland zone of Southwest Asia were precisely unlike those in Europe,so the pollen-based vegetation history for the region may require arather different interpretation from that conventionally offered. If correct, this would suggest that in addition to intensivecereal-based cultivation of small portions of the landscape, Neolithicpopulations in Southwest Asia carried out more extensive forms ofrangeland management, involving the systematic use of burning toregulate vegeta;tion structure and composition, in order to improvelivestock grazing and possibly also for dryland cereal cultivation.Given that Neolithic people and trees were advancing to occupy the samelandscape-ecological zone at the same time during the opening millenniaof the Holocene, the two might be considered to have been in competitionfor the same `lebensraum'. Ultimately human action delayed ratherthan preventing the spread of oak to achieve its eventual range. Theidea that a disturbance regime--partly natural, partlyhuman-induced--regulated the vegetation cover and composition ofsouthwest Asia's parkland zone is also in accord with currentecological theory, which largely rejects equilibrium models and theconcept of a climatic climax (Blumler 1996; Meadows 1999). Mediterraneansavanna parkland ecosystems are consequently unlikely to simplyrepresent `degraded forest', but instead evolved as adaptations tolong-term fire, grazing and other disturbances (Trabaud 1981; Grove& Rackham-2001). The case presented here is by no means a proven one, and requiresfurther empirical testing, notably by the study of micro-charcoalprofiles and by examination of records from previous interglacials, whenhuman impact would be expected to have been slighter. It should also beemphasised that it is not proposed that Neolithic-Chalcolithicpopulations engaged in large-scale forest clearance. Rather it issuggested that they acted--in concert with a range of naturalagencies--to retard the establishment and spread of woodland at anecotone e��co��tone?n.A transitional zone between two communities containing the characteristic species of each.[eco- + Greek tonos, tension, tone; see tone. where moisture stress would always have limited tree growth.Palynological data from several lake sites within the samelandscape-ecological zone indicate that the factors which retardedwoodland expansion operated at a landscape, not a local, scale, andpotentially imply a much larger role for human agency in this biome biomeLargest geographic biotic unit, a major community of plants and animals with similar requirements of environmental conditions. It includes various communities and developmental stages of communities and is named for the dominant type of vegetation, such as grassland or atthis time than has previously been recognized. Acknowledgements. Thanks go to D. Harris, A. Rosen, W. Eastwood, H.Woldring, L. Wick, G. Hillman, H.E. Wright and L. Sadori for valuablediscussions and information. Diagrams were drawn by Brian Rogers. References ASOUTI, E. & J. HATHER. 2001. Charcoal analysis and thereconstruction of ancient woodland vegetation in the Konya Basin,south-central Anatolia, Turkey: results from the Neolithic site ofCatalhoyuk East, Vegetation History and Archaeobotany 10: 23-32. BAR-MATTHEWS, M., A. AYALON & A. KAUFMAN. 1997. Late Quaternarypaleoclimate in the eastern Mediterranean region from stable isotopeanalysis of speleothems at Soreq Cave, Israel, Quaternary Besearch47:155-68 BAR-YOSEF, O. & A. BELFER-COHEN. 1992. From foraging to farmingin the Mediterranean Levant, in A.B. Gebauer & T.D. Price (ed.),Transitions to agriculture in prehistory 21-48. Madison (WI): PrehistoryPress. BEHRE, K. 1990. Some reflections on anthropogenic indicators in therecord of prehistoric occupation phases in pollen diagrams from the NearEast, in Bottema et al. (ed.): 219-37. BLUMLER, M.A. 1991. Fire and agricultural origins: preliminaryinvestigations, in S.G. Nodvin & T.A. Waldron (ed.) Fire andenvironment: ecology and cultural perspectives: 351-8. Asville (NC):USDA USDA,n.pr See United States Department of Agriculture. . BLUMLER, M.A. 1996. Ecology, evolutionary theory and agriculturalorigins, in Harris (ed.): 25-50. London: UCL UCL University College LondonUCL Universit�� Catholique de LouvainUCL UEFA Champions LeagueUCL Upper Confidence LimitUCL University of Central LancashireUCL Upper Control LimitUCL Unfair Competition LawUCL Ulnar Collateral Ligament Press. BOTTEMA, S. 1986. A Late Quaternary pollen diagram from Lake Urmia(Northwestern Iran), Review of Palaeobotany and Palynology pal��y��nol��o��gy?n.The scientific study of spores and pollen.[Greek palunein, to sprinkle + -logy. 47: 241-61. 1992. Prehistoric cereal gathering and farming in the Near East:the pollen evidence, Review of Palaeobotany and Palynology 73: 21-33. BOTTEMA, S., G. ENTJES-NIEBORG & W. VAN ZEIST (ed.) 1990.Man's role in the shaping of the eastern Mediterranean landscape.Rotterdam: A.A. Balkema. BOTTEMA, S. & H. WOLDRING. 1990. Anthropogenic indicators inthe pollen record of the Eastern Mediterranean, in Bottema et al. (ed.):231-64. BOTTEMA, S., H. WOLDRING & B. AYTUG, 1994. Late Quaternaryvegetation history of Northern Turkey, Palaeohistoria 35/ 36: 13-72. CLARK, J.S., H. CHACHIER, J.G. GOLDAMMER & B. STOCKS (ed.).1997. Sediment records of biomass burning and global change. Berlin:Springer-Verlag. NATO NATO:see North Atlantic Treaty Organization. NATOin full North Atlantic Treaty OrganizationInternational military alliance created to defend western Europe against a possible Soviet invasion. ASI ASI,n See Anxiety Sensitivity Index. Series, vol. 151. CLARK, J.S. & W.A. PATTERSON. 1997. Background and localcharcoal in sediments: scales of fire evidence in the paleorecord, inClark et al. (ed.): 23-48. EASTWOOD, W.J., N. ROBERTS & H.F. LAMB. 1998 Palaeoecologicaland archaeological evidence for human occupance in southwest Turkey: TheBeysehir Occupation Phase, Anatolian Studies 48: 69-86. GROVE, A.T. & O. RACKHAM. 2001. The nature of MediterraneanEurope. An ecological history. New Haven (CT): Yale University Press. HARMANKAYA, S., TANINDI, O. & M. OZBASARAN. 1997. TurkiyeArkeolojik Yerlesmeleri (TAY) 2: Neolitik. Istanbul: Ege Yayinlari. HARRIS, D.R. (ed.) 1996. The origins and spread of agriculture andpastoralism in Eurasia. London: UCL Press. 1998. The origins of agriculture in Southwest Asia, The Review ofArchaeology 19: 5-11. HILLMAN, G. 1996. Late Pleistocene changes in wild plant-foodsavailable to hunter-gatherers of the northern Fertile Crescent: possiblepreludes to cereal cultivation, in Harris (ed.): 159-203. KELTS, K. & M. SHAHRABI. 1986. Holocene sedimentology sedimentologyScientific discipline concerned with the physical and chemical properties of sedimentary rocks and the processes involved in their formation, including transportation, deposition, and lithification of sediments. ofhypersaline Lake Urmia, northwest Iran, Palaeogeography,Palaeoclimatology Noun 1. palaeoclimatology - the study of the climate of past agespaleoclimatologyarchaeology, archeology - the branch of anthropology that studies prehistoric people and their cultures , Palaeoecology paleoecology, palaeoecologythe branch of ecology that studies the relationship of ancient plants and animals to their environments. — paleoecologic, palaeoecologic, paleoecological, palaeoecological, adj. 54: 105-30. KERSHAW, A.P., M.B. BUSH, G.S. HOPE, K.-F. WEISS WEISS Workshop on Industrial Experience with Systems Software , J.G. GOLDAMMER& R. SANFORD. 1997. The contribution of humans to past biomassburning in the tropics, in Clark et al. (ed.): 413-42. KOHLER-ROLLEFSON, I. & G.O. ROLLEFSON. 1990. The impact ofNeolithic subsistence strategies on the environment. The case of `AinGhazal Ghaz´aln. 1. A kind of Oriental lyric, and usually erotic, poetry, written in recurring rhymes. , Jordan, in Bottema et al. (ed.): 3-14. KUZUCUOGLU, C. & N. ROBERTS. 1998. Evolution del'environnement en Anatolie, de 20000 a 6000 BP, Paleorient 23:7-24. LEMCKE, G. & M. STURM. 1997. [[delta].sup.18]O and traceelement measurements as proxy for the reconstruction of climate changesat Lake Van (Turkey): preliminary results, in H.N. Dalfez, G. Kukla& H. Weiss (ed.), Third Millennium BC climate change and the OldWorld Collapse: 653-78. Berlin: Springer-Verlag. NATO ASI Series, vol.49. LEWIS, H.T. 1972. The role of fire in the domestication of plantsand animals in Southwest Asia: a hypothesis, Man 7:195-222. MCCORRISTON, J. & F. HOLE. 1991. The ecology of seasonal stressand the origins of agriculture in the Near East, American Anthropologist93: 46-69. MASON, S.L.R. 2000. Fire and Mesolithic subsistence--managing oaksfor acorns in northwest Europe, Palaeogeography, Palaeoclimatology,Palaeoecology 164: 139-50. MEADOWS, M. 1999. Biogeography BiogeographyA synthetic discipline that describes the distributions of living and fossil species of plants and animals across the Earth's surface as consequences of ecological and evolutionary processes. : changing places, changing times.Progress in Physical Geography 23: 257-70. MILLER, N.F. 1996. Seed eaters of the Ancient Near East, human orherbivore herbivore:see carnivore. herbivoreAnimal adapted to subsist solely on plant tissues. Herbivores range from insects (e.g., aphids) to large mammals (e.g., elephants), but the term is most often applied to ungulates. ? Current Anthropology 37: 521-8. NAVEH, Z. 1974. The effects of fire in the Mediterranean region, inT.T. Kozlowski & C.E. Ahlgren (ed.), Fire and ecosystems: 401-34.New York (NY): Academic Press. PULLAR, J. 1977. Early cultivation in the Zagros, Iran 15:15-37. PYNE, S.J. 1995. World fire. The culture of fire on Earth. New York(NY): Henry Holt. ROBERTS, N. 1982. Forest re-advance and the Anatolian Neolithic, inM. Bell & S. Limbrey (ed.), Archaeological aspects of woodlandecology: 231-46. Oxford: British Archaeological Reports. Internationalseries S146. ROBERTS, N., J. REED, M.J. LENG LENG Left Engine , C. KUZUCUOGLU, M. FONTUGNE, J.BERTAUX, H. WOLDRING, S. BOTTEMA, S. BLACK, E. HUNT & M.KARABIYIKOGLU. 2001. The tempo of Holocene climatic change in theeastern Mediterranean region: new high-resolution crater-lake sedimentdata from central Turkey, The Holocene 11: 721-36. ROBERTS, N. & H.E. WRIGHT, JR. 1993. Vegetational, lake-leveland climatic history of the Near East and Southwest Asia, in H.E.Wright, Jr., J.E. Kutzbach, T. Webb, III, W.F. Ruddiman, F.A.Street-Perrott & P.J. Bartlein (ed.), Global climates since the lastglacial maximum: 194-220. Minneapolis (MN): University of MinnesotaPress The University of Minnesota Press is a university press that is part of the University of Minnesota. External linkUniversity of Minnesota Press . ROSSIGNOL-STRICK, M. 1999. The Holocene Climatic Optimum The Holocene Climate Optimum was a warm period during roughly the interval 9,000 to 5,000 years B.P.. This event has also been known by many other names, including: Hypsithermal, Altithermal, Climatic Optimum, Holocene Optimum, and PollenRecords of Sapropel 1 in the Eastern Mediterranean, 9000-6000 BP,Quaternary Science Reviews 18: 515-30. SHERRATT, A. 1980. Water, soil and seasonality in early cerealcultivation, World Archaeology 11:313-30. SIMMONS, I.G. & J.B. INNES. 1996. The ecology of an episode ofprehistoric cereal cultivation on the North York Moors, England, Journalof Archaeological Science 23: 613-18. STEVENS, L.R., H.E. WRIGHT, JR. & E. ITO Ito,city (1990 pop. 71,223), Shizuoka prefecture, central Honshu, Japan, on the Izu Peninsula and the Sagami Sea. It is an important fishing port and hot spring resort. See indium. . 2001. Proposedchanges in seasonality of climate during the late-glecial and Holoceneat Lake Zeribar, Iran, The Holocene 11: 747-56. TRABAUD, L. 1981. Man and fire: impacts on Mediterraneanvegetation, in F. di Castri, D.W. Goodall & R.L. Specht (ed.)Mediterranean-type shrublands: 523-37. Amsterdam: Elsevier. Ecosystemsof the World 11. VAN ANDEL, T.H. & C.N. RUNNELS. 1995. The earliest farmers inEurope, Antiquity 69: 489-500. WICK, L., G. LEMCKE, & M. STURM. In press. Evidence ofLateglacial and Holocene climatic change and human impact in easternAnatolia: high-resolution pollen, charcoal, isotopic and geochemicalrecords from the laminated sediments of Lake Van, The Holocene. WILLCOX, G. 1996. Evidence for plant exploitation and vegetationhistory from three early Neolithic pre-pottery sites on the Euphrates(Syria), Vegetation History and Archaeobotany 5: 143-52. WILLIS, K.J. & K.D. BENNETT. 1994. The Neolithic transition:fact or fiction? Palaeoecological evidence from the Balkans, TheHolocene 4: 326-30. WOLDRING, H. & R. CAPPERS. 2001. The origins of the `wildorchards' of central Anatolia, Turkish Journal of Botany 25: 1-9. YASUDA, Y., H. KITAGAWA & T. NAKAGAWA. 2000. The earliestrecord of major anthropogenic deforestation in the Ghab Valley,northwest Syria: a palynological study, Quaternary International 73/74:127-36. VAN ZEIST, W. & S. BOTTEMA. 1977. Palynological investigationsin western Iran, Palaeohistoria 19: 19-85. 1991. Late Quaternary vegetation of the Near East. Wiesbaden: Dr L.Reichart Verlag. Beihefte zum Tubinger Atlas des Vorderen Orients, ReiheA18. VAN ZEIST, W. & H. WOLDRING. 1976. A Postglacial pollen diagramfrom Lake Van in East Anatolia, Review of Palaeobotany and Palynology26: 249-76. ZOHARY, D. & M. HOPF. 2001. Domestication of plants in the OldWorld. The origin and spread of cultivated plants in west Asia, Europeand the Nile valley. 3rd edition. Oxford: Oxford University Press. ZOHARY, M. 1973. Geobotanical foundations of the Middle East.Stuttgart-Amsterdam: Fischer. NEIL ROBERTS, Department of Geographical Sciences, University ofPlymouth The University of Plymouth is the largest university in the southwest of England, with over 30,000 students and is the fifth largest UK university based on student population. (Larger universities are Open, London, Manchester, and Manchester Metropolitan respectively. , Drake Circus, Plymouth PL4 8AA, England.cnroberts@plymouth.ac.uk Received 13 December 2001, accepted 18 February 2002, revised 7July 2002