Climate changes and human-environment interactions in the Apulia region of southeastern Italy during the Neolithic period

Page 1

The Holocene23(9) 1297 –1316© The Author(s) 2013Reprints and permissions:sagepub.co.uk/journalsPermissions.navDOI: 10.1177/0959683613486942hol.sagepub.com

IntroductionUnderstanding human–environment interactions during the Neo-lithic is critical to identify anthropisation dynamics. A number of studies have addressed these issues with special attention given to settlement dynamics as well as paleoenvironmental and paleoeco-nomic reconstructions and the effects of changes across broad tim-escales. Weninger et al. (2009) explored the impact of rapid climate changes on prehistoric communities in the eastern Medi-terranean during the early and middle Holocene, focusing on the social implications of four major cold climate anomalies.These authors stated that climatic deterioration is a major factor underly-ing social change, although it is constantly at work within a wide spectrum of social, cultural, economic and religious factors.

González-Sampériz et al. (2009) studied the relationship between climate and population dynamics in the Ebro area during part of the Holocene. They recognised an ‘archaeological silence’ caused by the impact of the abrupt global 8.2 ka cold event. Avail-able regional paleoclimate archives contain evidence of an arid-ity crisis that interrupted the humid early-Holocene period and forced hunter-gatherer groups to migrate to areas with more favourable conditions.

Our aim was to analyse the modes of human–environment interaction between 6200 and 3700 bc in Apulia, southern Italy. We focused on the Neolithic period because of the availability of chronological and paleoagricultural data.

Owing to its position in the middle of the Mediterranean Sea, between two zones with different climatic characteristics, Apulia may have experienced substantial changes in anthropic dynamics related to the resilience of the first farmers in the area (Crumley, 1994; Orlove, 2005; Redman, 2005; Redman and Kinzig, 2003), even during the slight climate oscillations that characterised the middle Holocene (Mayewski et al., 2004).

We gathered and analysed paleoenvironmental and paleocli-matic information from natural archives and obtained archaeobo-tanical data (i.e. plant macroremains) from archaeological sites. This interdisciplinary approach allowed us to define the climatic characteristics of the studied time span and their influence on human adaptation strategies, as highlighted by changes in agricul-tural practices.

486942 HOL23910.1177/0959683613486942The HoloceneGirolamo et al.2013

1Università del Salento, Italy2Università degli Studi di Bari ‘Aldo Moro’, Italy3Soprintendenza per i Beni Archeologici della Puglia, Italy

Corresponding author:Massimo Caldara, Dipartimento di Scienze della Terra e Geoambientali, Università degli Studi di Bari ‘Aldo Moro’, Campus Universitario, Via Orabona 4, Bari 70125, Italy. Email: [email protected]

Climate changes and human–environment interactions in the Apulia region of southeastern Italy during the Neolithic period

Girolamo Fiorentino,1 Massimo Caldara,2 Vincenzo De Santis,2

Cosimo D’Oronzo,1 Italo Maria Muntoni,3 Oronzo Simone,1 Milena Primavera1 and Francesca Radina3

AbstractThe objective of our research was to define the main human–environment interactions during the Neolithic period (6500–3700 bc) in the Apulia region of southeastern Italy based on available published and unpublished data. Knowledge of these interactions is crucial to understanding the cultural and social dynamics of the period, particularly concerning the earliest farmers. Using a multidisciplinary approach, paleoenvironmental and paleoclimatological data at the regional and Mediterranean scales were compared with the results of analyses performed on natural deposits and deposits in Neolithic settlements. The following data sets were used: (1) 121 14C dates for settlements, from which probability curves (%) of the Apulian Archaeological Occupation (AAO) were developed; (2) offshore data obtained from analyses performed on two offshore sediment cores drilled in the Adriatic Sea; (3) offsite data from studies conducted in two natural coastal contexts; and (4) onsite archaeobotanical data from 35 settlements. This study allowed us to tentatively define the main climatic features between 6200 and 3700 bc. We identified two dry phases (one between 5000 and 4600 bc and a second that peaked c. 4000 bc) and two wet intervals (one between 6200 and 5500 bc and a second that peaked around 4400 bc). Climate changes appear to have been relatively gradual. The use of archaeobotanical data allowed us to determine a direct link between paleoclimatic and archaeological sequences. These data highlight the variations in agricultural strategies (species used and harvest times) as humans responded to changes in the rainfall regime.

KeywordsApulia, archaeobotany, environmental archaeology, human resilience, mid Holocene, multiproxy analysis, Neolithic

Received 19 July 2012; revised manuscript accepted 14 March 2013

Research paper

Page 2

1298 The Holocene 23(9)

Regional settingGeomorphological districts and settlement patternsThe Apulia region includes a variety of landscapes characterised by a common cultural framework during the Neolithic period (Figure 1).

Various archaeological and geomorphological contexts have been the subject of systematic research in recent years (Berger and Guilaine, 2009; Pessina and Tinè, 2008). The study of these topics enhances our understanding of the Neolithisation process in southern Italy through the comparison of settlement patterns and natural resource exploitation strategies at both large and small scales, over a period of more than 2000 years.

The Gargano (District A) is an almost entirely mountainous headland that protrudes into the Adriatic Sea. Its maximum alti-tude is 1055 m. The promontory is part of the Mesozoic–early Cenozoic Apulian Carbonate Platform. The morphology of the Gargano was shaped by tectonic, karst and fluvial processes (Caldara and Palmentola, 1991). The reconstruction of the Neo-lithic population in this area remains problematic (Calattini and Muntoni, 2005). Settlement appears to have occurred along the coastland from the shores of Lesina Lake to Mattinata, both of which are at the north and eastern ends of the headland, and at small sites located on the hillslopes, within caves located on the sides of deep valleys along the southern edge of the Gargano promontory. In this district the start of the Neolithisation process coincided with the beginning of mining activity for chert exploi-tation (Tarantini and Galiberti, 2011). Chert mining appears to

have been initiated in the first centuries of the 6th millennium bc and to have intensified during the following centuries. Data per-taining to the middle-Neolithic period have been collected along the southern foothills of Gargano overlooking the Tavoliere plain. Evidence relating to the late-Neolithic period has been found at sites along the southern slopes of the promontory and in the north-eastearn area, within perilacustrine and coastal environments (Caldara et al., 2008).

The Tavoliere Plain (District B) is the second largest alluvial plain in peninsular Italy. This plain is surrounded by the Gargano headland, the Daunian mountains and the Murge hills. The smooth topography, which gently slopes seawards, is broken by Plio-Pleistocene marine terraces (Caldara and Pennetta, 1993). From a geomorphological perspective, several zones can be distinguished from west to east: (1) the foothills area, with an average altitude of 350–500 m; (2) the terraced zone (District B3 in Figure 1), characterised by several sandy to pebbly marine terraces incised by streams, the most important of which are the Ofanto and the Fortore rivers; (3) the large ancient alluvial plain located in the area surrounding the towns of Foggia, Cerignola and Ortanova (District B2 in Figure 1); and (4) the Holocene coastal plain (Dis-trict B1 in Figure 1) where ancient coastal water bodies existed until the recent past (Boenzi et al., 2006; Caldara et al., 2002).

A fairly complex picture of the Neolithic period in this area has been reconstructed owing to a long tradition of interdisciplin-ary research (Cassano and Manfredini, 1983, 2005; Delano Smith, 1979; Jones, 1987a; Tinè, 1983). The Neolithisation process in the Tavoliere plain has been defined in general terms, although a

Figure 1. Geological map of Apulia and the locations of the studied archaeological sites, geomorphological districts and offshore cores.

Page 3

Fiorentino et al. 1299

large number of sites are known only from surface surveys. Evi-dence from the early-Neolithic period is found mainly along the coastland, with several sites on terraces, and along river valleys, mainly near the Ofanto (Cipolloni Sampò, 1980, 1982, 1987) and the Fortore (Gravina, 2005; Gravina et al., 2005) Rivers.

Fewer but larger sites characterise the initial phases of the mid-dle-Neolithic period (Cassano, 1993). Settlement reduction in the plain appears to characterise the later Neolithic period, during which a contextual increase in the number of settlements character-ises several elevated areas surrounding the Tavoliere plain, includ-ing the highest sector of the Fortore River valley, the Subapennine Mountains and the caves of the lower Gargano (Gravina, 1988).

The plateau of Murge (District C) is a wide and flat calcareous ridge that extends in the NW–SE direction. Its maximum altitude exceeds 600 m. On the Adriatic side, the plateau is characterised by a succession of terraced surfaces furrowed by short karst can-yons (District C2 in Figure 1), several wetland areas (Caldara and Pennetta, 2002) along the coastland (District C1 in Figure 1), and a widespread presence of coastal karst springs. On the southwest-ern side, the plateau is characterised by a morphotectonic scarp facing the Bradano foredeep (Neboit, 1975).

Between the end of the 7th and the first half of the 6th millen-nium bc, the Adriatic side of Murge experienced a significant Neolithisation process (Radina, 2002). The occupation pattern was characterised by settlements located on terraces within a few kilometres of each other, with small and temporarily inhabited sites situated around large and permanent villages. The latter were located in areas of strategic importance for economic and trade purposes (Muntoni, 2012). A more complex occupation pattern developed in the middle of the 6th millennium bc (Radina, 2006) with a new organisation of spaces within the settlements and evi-dence of ritual activities conducted within caves.

The period between the end of the 6th and the first half of the 5th millennium bc is less apparent in the archaeological record. The reduced archaeological evidence coincides with the occurrence of painted pottery, a remarkable reduction in settlement density and, in the case of previously occupied sites, a reduction in settlement size. At the same time, ritual practices in caves appear to have been a consolidated habit (Grifoni Cremonesi, 2002). Late-Neolithic evidence of the transition between Diana and Macchia a Mare facies has been found at sites along the Ofanto River and in Murg-ian subcoastal environments (Caldara et al., 2011).

The Brindisi plain (District D) is surrounded by the Murge plateau, the Taranto Murge and the Salento hills. The maximum elevation of this area is approximately 150 m.

The morphology of the Taranto–Metaponto plain (District G) is characterised by a series of flat arched surfaces sloping towards the Ionian coastline and bordered by dune belts.

The Taranto Murge district (District E) is characterised by sev-eral marine terraces that step down towards the Ionian sea. The flat surfaces stretch parallel to the coastline and are cut by a num-ber of short, straight, narrow and deep fluvial valleys.

The Salento hills (District F) are located in the southernmost part of Apulia. These are low hills (approximately 200 m in elevation) whose peaks are oriented along the NW–SE direction, alternating with broad and flat depressions. The arrangement of the relief is the result of structural factors and exhibits an asymmetric pattern on its Adriatic and Ionian sides, the former being steeper and the latter being gently sloping (Boenzi et al., 1991). A widespread presence of coastal karst springs also characterise this district.

All of these districts (D to F) exhibit evidence of a common settlement pattern during the Neolithic period (Corrado and Ingravallo, 1988; Gorgoglione, 2004; Guilaine and Cremonesi, 2003) with a prevalent coastal distribution along the Ionian side.

During the early-Neolithic period many settlements were established on hills and along streams. At the beginning of the middle-Neolithic period, the archaeological record shows that

sites were fewer and larger. These sites were located in areas that were strategic for economy and trade purposes (Tiberi, 2007).

A clear trend towards settlement reduction appears to character-ise the later phases, culminating with a gap in the archaeological record at the end of the 5th millennium bc. At this time, the Neo-lithic order also appears to have changed in the Murge, although the organisation of spaces within these settlements is more difficult to infer. However, burials and the performance of other rites appear to have been common in caves (Grifoni Cremonesi, 2002).

Climate and vegetationApulia is part of the Mediterranean area, and its climate belongs to the Cs group of the Köppen classification system (Köppen, 1923), which is characterised by mild to cool winters with little rainfall and hot dry summers (Marsico et al., 2007). The average annual temperature ranges between 15°C and 18°C (Figure 2a). During the coldest month (January), the mean temperature is approximately 6°C, with the lowest temperatures occurring in Gargano mountains (2°C) and the highest temperatures found in Salento (9°C). During the warmest month (July), the average temperatures are approximately 26°C. The Tavoliere plain is one of the warmest areas in Italy (with a mean temperature of approx-imately 26°C and maxima higher than 40°C). The average annual rainfall in Apulia is approximately 600 mm (Figure 2b), with the highest and lowest values recorded in the Gargano mountains (1100–1200 mm) and the Tavoliere (less than 400 mm), respec-tively. The peculiar morphology of Apulia dictates the local cli-mate dynamics; for example, in the Tavoliere plain the climate is generally semi-arid (Figure 2c) due to the rainshadow effect of the surrounding mountains (Gargano and Apennines; Caldara et al., 2005); thus, a small increase in the mean annual tempera-ture or a small decrease in mean rainfall can cause a shift towards drier conditions (Boenzi et al., 2002; Caldara et al., 2002).

In general, marked intra- and interannual climate variability characterise the entire Mediterranean area, which is influenced by the activity of three main pressure systems (Lionello et al., 2005): the Icelandic Low, the Azores High and the Siberian High. The first two systems undergo periodic variations in their strength and position (North Atlantic Oscillation; NAO) that determine the seasonal rainfall regimes.

It is probable that, because of its geographic position, Apulia during the Holocene was at the boundary between two climate regions (eastern and western Mediterranean; Roberts et al., 2011). As a consequence, the climate changes that characterised the middle Holocene (Gladstone et al., 2005) could have led to peculiar modifi-cations in water availability for vegetation growth in this area.

The regional climatic variability is mirrored by a mosaic of veg-etation types corresponding to specific phytocoenoses (Figure 2d): (1) forests whose main taxa are Quercus cerris and, in specific topoclimatic conditions Fagus sylvatica; (2) mesophilous sub-mountain vegetation with Q. pubescens and xeric grasslands in those areas characterised by strong summertime aridity; (3) highly grazed Quercus trojana woodlands; (4) Quercus coccifera scrub and garrigues; and (5) Quercus ilex, which is replaced by Pinus halepensis near the coasts and by thermophilous sclerophyllous Mediterranean scrub or bands of meso-thermophilous species belonging to Quercion ilicis Br-Bl 1936 along the southern coasts.

Materials and methodsA multidisciplinary approach was used to analyse the interaction between the environmental and anthropic variables during the Neolithic period. This analysis compared chronological data with the results of paleoenvironmental analyses performed on natural deposits (offshore and offsite data) and with archaeobotanical data obtained from archaeological sites (onsite data) as follows.

Page 4

1300 The Holocene 23(9)

Figure 2. Climate and vegetation of the Apulian region: (a) mean annual temperature (°C); (b) mean annual rainfall (mm); (c) aridity index (according to De Martonne, 1942); (d) phytocoenosis (modified by Macchia et al., 2000). 1: Dominance of Quercus cerris woodlands; 2: Quercus trojana degraded woodlands; 3: Quercus coccifera garrigue/scrubs and bush; 4: Quercus ilex degraded woodlands; 5: sub-montane mesophilic vegetation with Quercus pubescens, xeric meadows.

(a) The chronological framework is based on a radiocarbon data base for Neolithic Apulia, which currently includes 121 14C dates from 40 sites (Table 1). Radiometric dates with a standard deviation of more than 150 years were not used because of their low accuracy. The calibration of dates, based on the data set published by Reimer et al. (2004), and a multisample probability curve of the Apulia Archaeological Occupation (AAO; Figure 3) were obtained using the Calib 5.0.1 program (http://calib.qub.ac.uk). Four AAO curves were also obtained for the main geographic districts (Gargano, Tavoliere, Murge, and southern Apulia). All of these curves were developed using the largest regional radiocarbon data set in Italy; many dates were obtained by analysing samples collected from settlements characterised by deep and complex stratigraphic sequences. All of the AAO curves have been correlated with the chrono-typological framework based on traditional archaeological analysis of pottery remains (Pessina and Tinè, 2008).We consider these curves as proxies capable of describ-ing the fluctuating Apulian population dynamics. We are aware that the AAO curves could also describe other factors, including the following: (1) choices made by many archaeologists to focus on periods of particular interest (such as the early Neolithic and the beginnings of agriculture in Apulia); and (2) favourable conditions of soil fertility and vegetation that could have characterised the start of the Neolithic and the gradual and on-going negative impact of human pressure on the ecology of a comparatively dry region.However, we observe that a decrease in the number of dated samples generally occurs within multistratified set-tlements; in our opinion, such a decrease is evidence of an actual decrease of frequency. In addition, we believe that in a rain-fed agricultural system, when an area had been totally exploited, ancient communities would have moved into nearby areas to take advantage of virgin soils; this pro-cess would not necessarily imply a population decrease.

(b) Offshore data include different proxies (pollen, dino-cyst and isotopes) sampled from two cores drilled in the southern Adriatic sea: AD91-17 (Sangiorgi et al., 2003) and RF93-30 (Mercuri et al., 2012; Oldfield et al., 2003). These data have been taken into consideration because their chronological resolution for the Neolithic period and the number and sensitivity of proxies found there are higher than for other cores drilled in the south-ern Adriatic Sea (Trincardi et al., 1996). In particular,

the AD91–17 curve has one sample every 178 years on average, in addition to several climatic proxies, and it reveals five intervals that can be correlated to major changes recorded in the eastern Mediterranean Sea, suggesting that the AD91-17 core recorded basin-wide ecozones rather than local events (Giunta et al., 2003).For the RF93-30 curve, the presence of significant con-tributions from northern and central Adriatic sectors is not negligible (Cattaneo et al., 2003), but it was chosen because core location is the closest to the north Apulian coast, and it has one sample every 280 years on average. In addition, no other offshore curve located near the Apulia region presents a more detailed temporal resolution for the studied period.

(c) Offsite data were obtained from two cores drilled near Neolithic settlements that yielded different paleoenvi-ronmental markers (seeds and fruits, charcoal, pollen and aquatic plant macroremains): ARI24 – Ariscianne wet-land (Caldara et al., 2011) and BAT1/2 – Battaglia Lake (Caldara et al., 2008; Caroli, 2005; Caroli and Caldara, 2007). These sequences have certain limitations. For example, both sequences cover only part of the studied period: ARI/24 ranges from 5300 to 3500 bc, and BAT1/2 covers only from 4950 to 3500 bc, with chronological sam-ple resolutions of 72 and 69 yr, respectively. Despite these limitations, ARI/24 and BAT1/2 are the most detailed and chronologically constrained sequences available for the studied period in the Apulia region, with respect to both the number of proxies considered and the number of dat-ings performed.

(d) Onsite archaeobotanical data were obtained from the 35 multistratified or monophasic settlements that yielded 14C-dated samples (Figure 1, Tables 2, 3). All plant remains considered were charred and derived from ancient human activities (e.g. wood used as fuel in domestic hearths and residues of food processing). These data were used to define the following indicators.

(1) A paleo-vegetational indicator, i.e. the meso-thermophi-lous/thermo-xerophilous ratio, inferred from plants used for fuelwood production (where the prevalence of higher proportions of thermo-xerophilous taxa indicates drier conditions).

(2) Paleo-agronomical indicators (based on seeds and fruits of edible plants) that take into account the traditional agronomi-cal patterns in the Mediterranean areas (Azzi, 1930; Percival, 1943; von Bothmer et al., 1995): (a) the wheat/barley ratio

Page 5

Fiorentino et al. 1301

Table 1. Available Neolithic radiocarbon dates (14C BP, cal. 2σ bc) from Apulian archaeological settlements. Some dates have been published without indicating any Lab code: *Radina (2002); **Gorgoglione (2002); ***Radina (2013). The date # is published here for the first time.

Site/level Ref. lab. 14C yr BP Calibrated 2σ bc Geomorphological district

Scamuso, phase IIIb Gif-6339 7290±110 6400–5984, 5936–5934 C1

Terragne, US5 Beta-67093 7260±60 6231–6019 EPulo di Molfetta, US10 LTL-142A 7134±60 6203–6143, 6106–5883 C1

Mass. Giuffreda, ditch ‘g’ MC-2292 7125±50 6073–5898 B2

Scaramella A R-350 7000±100 6058–5709 B2

Defensola A Utc-1342 6990±80 6010–5727 ATorre Sabea TAN-88066 6960±130 6063–5628 FTerragne, US5 Beta-59934 6930±70 5983–5705, 5688–5675 ETorre Canne Gif-6725 6900±80 5978–5947, 5922–5645 C1

Torre Sabea TAN-88247 6890±130 6017–5608, 5594–5561 FGrotta Sant’Angelo, lev. 9 Gif-6724 6890±70 5971–5954, 5911–5645 C2

Ripa Tetta CAMS-2681 6890±60 5964–5958, 5901–5661 B3

Coppa Nevigata OxA-1475 6880±90 5978–5947, 5922–5630 B1

Terragne, US3 Beta-59933 6870±70 5961–5961, 5899–5634 ETorre Sabea LJ-1448 6860±45 5843–5658 FVilla Comunale di Foggia MC-2290 6850±130 5990–5540 B2

Coppa Nevigata OxA-1474 6850±80 5967–5956, 5903–5621 B1

Scamuso, phase IIIb Gif-7055 6810±80 5880–5611, 5591–5564 C1

Serra Cicora, T7 LTL-221A 6785±40 5731–5629 FOria Sant’Anna R-322 6780±90 5874–5861, 5847–5526 ESerra Cicora LTL-026A 6762±55 5742–5610, 5592–5563 FDefensola A LTL-4500A 6741±50 5728–5609, 5593–5562 ADefensola A Beta-71143 6740±80 5778–5510, 5499–5492 APorto Badisco LTL-427A 6739±60 5734–5548 FScaloria, tr. 3, lev. 3 LJ-4649 6720±100 5805–5478 ALagnano da Piede UCLA-2148 6700±100 5788–5474 B3

Serra Cicora KIA-10287 6679±54 5705–5491 FDefensola A Beta-71144 6650±70 5702–5692, 5673–5478 ADefensola A Beta-161933 6650±50 5642–5487 AMass. Candelaro, phase II OxA-3684 6640±95 5727–5466, 5439–5424, 5405–5383 B1

Mass. Candelaro, phase II OxA-12062 6638±34 5629–5511, 5496–5496 B1

Defensola A Utc-1411 6630±40 5626–5507, 5504–5490 AMass. Candelaro, phase II OxA-9988 6605±45 5618–5484 B1

Balsignano, T3 KIA* 6602±30 5616–5582, 5573–5485 C2

Mass. Candelaro, phase III OxA-12063 6601±37 5617–5580, 5575–5484 B1

Scamuso, phase IIIb Gif-7345 6600±120 5720–5328 C1

Oria Sant’Anna, lev. 3 LTL-1061 6600±60 5630–5475 ETorre Sabea Ly-4002 6590±140 5755–5295, 5255–5229 FPalata 1 LTL-5187A 6576±45 5616–5582, 5573–5476 B3

Defensola A Beta-80604 6570±70 5631–5463, 5448–5378 ABalsignano, T3 KIA* 6565±29 5606–5595, 5560–5478 C2

La Torretta LTL-2657A 6562±50 5621–5468, 5399–5392 B2

Palata 2 LTL-5188A 6561±50 5621–5468, 5400–5391 B3

Grotte Santa Croce, US78 OxA-7596 6555±50 5620–5467, 5432–5427, 5404–5386 C2

Mass. Candelaro, phase II OxA-9990 6555±45 5617–5469, 5397–5395 B1

Defensola A LTL-2717A 6551±40 5614–5586, 5569–5469 ALama Balice, T1 LTL-1212A 6542±45 5615–5585, 5570–5465, 5442–5382 C2

Scaramella A R-351 6540±65 5618–5458, 5454–5374 B2

Defensola A Beta-80603 6540±60 5617–5462, 5451–5376 AGrotta Sant’Angelo, lev. 6-7 Gif-6722 6530±70 5619–5364 C2

Balsignano, struct. 2 LTL-140A 6523±45 5607–5594, 5561–5375 C2

Santa Tecchia BM-2414 6520±70 5617–5357, 5348–5345 B1

Mass. Candelaro, phase III OxA-3685 6510±95 5625–5312 B1

Mass. Candelaro, phase II OxA-9989 6510±45 5557–5370 B1

Mass. Candelaro, phase III OxA-12064 6501±37 5530–5374 B1

Fontanarosa Uliveto BM-2415 6490±150 5711–5206, 5165–5118, 5108–5078 B1

Scaloria, tr. 1, lev. 8 LJ-4650 6490±140 5709–5207, 5158–5154 AMass. Candelaro, phase III OxA-10013 6450±50 5485–5321 B1

Oria Sant’Anna, lev. 3 LTL-1062A 6442±50 5482–5322 EDefensola A LTL-438A 6417±55 5480–5308 AScaloria, tr. 10 LJ-4980 6410±150 5629–5027 AScaloria, tr. 6, lev. 4 LJ 5095 6400±80 5509–5500, 5491–5217 A

(Continued)

Page 6

1302 The Holocene 23(9)

Site/level Ref. lab. 14C yr BP Calibrated 2σ bc Geomorphological district

La Torretta LTL-2658A 6385±50 5476–5297, 5247–5230 B2

Grotte Santa Croce, US72 OxA-7595 6375±50 5473–5295, 5260–5228 C2

San Domenico, US148 Beta** 6360±70 5475–5217 GOria Sant’Anna, lev. 2 LTL-1060A 6352±60 5469–5221 EGrotte Santa Croce, US69 OxA-7594 6345±45 5466–5404, 5384–5221 C2

Defensola A LTL-2716A 6341±55 5468–5401, 5389–5218 ACapo Rondinella ** 6340±40 5465–5406, 5382–5219 GDefensola A LTL-437A 6334±50 5466–5432, 5429–5404, 5385–5216 AScaloria, tr. 1, lev. 8 LJ-4651 6330±90 5478–5197, 5179–5063 AScamuso, phase II Gif-7346 6320±80 5473–5202, 5174–5071 C1

La Torretta LTL-2659A 6320±50 5466–5433, 5428–5404, 5384–5211 B2

Defensola C LTL-1725A 6306±75 5469–5201, 5176–5069 AScaloria, tr. 7, lev. 1 LJ-5097 6290±90 5471–5046 AScaloria, tr. 6, lev. 6-7 LJ-5096 6290±80 5467–5402, 5388–5054 ADefensola A LTL-4577A 6290±45 5369–5206, 5163–5079 ASan Marco LTL-1727A 6277±45 5358–5206, 5166–5117, 5109–5077 AMass. Candelaro, phase III OxA-3683 6200±95 5366–4906, 4863–4856 B1

Defensola A LTL-2720A 6178±55 5297–5239, 5233–4995 ABalsignano, struct. 3 LTL-139A 6148±45 5217–4962 C2

Passo di Corvo R-846 6140±120 5355–5350, 5345–4784 B2

Capo Rondinella ** 6130±80 5294–5249, 5229–4848 GScaloria, tr. 5, lev. 3 LJ-4983 6120±80 5293–5253, 5228–4842 ACarrara San Francesco, T2 Eth-21724 6080±65 5210–4841 C1

Capo Rondinella ** 6060±60 5207–5094, 5081–4800 GTrani, II Spiaggia di Colonna # UGAMS-8552 6060±30 5051–4849 C1

San Domenico, US145 ** 6040±80 5208–4749 GScamuso, phase II Gif-7057 6040±70 5207–4781 C1

Defensola A LTL-2718A 6035±60 5206–5166, 5117–5109, 5077–4778 ABalsignano, struct. 3 LTL-138A 5923±110 5199–5178, 5066–4520 C2

Le Macchie, US118 KIA-13413 5864±47 4842–4602 C1

Porto Badisco R-1225 5850±55 4836–4578, 4574–4554 FValle Guariglia II LTL-2715A 5822±45 4784–4553 AScamuso, phase II Gif-7054 5820±70 4835–4504 C1

Balsignano, T2 KIA-13414 5783±39 4722–4537 C2

Defensola A LTL-2719A 5746±45 4708–4492 ACasino San Matteo, US5 LTL-2961A 5728±45 4688–4484, 4479–4463 B3

Serra Cicora, T3 LTL-010A 5690±44 4683–4632, 4623–4449, 4414–4406 FSerra Cicora, T10 LTL-145A 5686±37 4668–4639, 4618–4449, 4412–4409 FCarpignano Salentino, T1 LTL-048A 5665±30 4582–4571, 4558–4447, 4419–4401 FCarpignano Salentino, T2 LTL-126A 5660±60 4677–4638, 4618–4359 FSerra Cicora, T4 KIA-13293 5648±34 4547–4440, 4425–4370 FCasino San Matteo, US5 LTL-2962A 5618±45 4535–4358 B3

Serra Cicora, T9 LTL-033A 5591±44 4502–4345 FSerra Cicora, T11 LTL-157A 5575±55 4518–4335 FCapo Rondinella ** 5570±40 4487–4477, 4464–4342 GOria Sant’Anna R-321 5550±80 4549–4239 EGrotta della Tartaruga di Lama Giotta Gif*** 5540±100 4649–4225, 4205–4163, 4130–4072 C1

Serra Cicora, T6 LTL-029A 5531±52 4487–4321, 4293–4265 FBalsignano, T2 KIA * 5508±50 4456–4313, 4301–4260 C2

Scaloria Bassa R-349 5480±70 4485–4226, 4204–4165, 4129–4074 AScamuso, AIII lev. 15 Gif-6338 5290±90 4329–3963 C1

Mass. Stevanato, lev. 12 LTL-621A 5244±55 4232–4188, 4181–3964 C2

Pulo di Molfetta, US60 LTL-2836A 5216±40 4226–4203, 4167–4099, 4075–3957 C1

Mass. Stevanato, T1 LTL-622A 5190±35 4216–4216, 4145–4137, 4053–3947 C2

Cala Colombo, lev. II-IV BM-2260R 5180±140 4327–4283, 4271–3702 C1

Martinetti, lev. 6 Beta-80605 5170±70 4228–4200, 4170–3794 AMass. Stevanato, lev. 13 LTL-624A 5149±55 4142–4139, 4052–3791 C2

Mass. Stevanato, lev. 2 LTL-623A 5115±50 4037–4021, 3994–3789 C2

Mass. Stevanato, lev. 13 LTL-620A 5018±60 3957–3694, 3677–3669 C2

(where a higher proportion of barley corresponds to rela-tively drier conditions), (b) hulled/naked cereal ratio (where a higher proportion of the naked form indicates wetter condi-

tions), and (c) the relative abundance of legumes, weeds and minor cereals (where the ecophysiological characteristics of the species are used as cropping period indicators).

Table 1. (Continued)

Page 7

Fiorentino et al. 1303

Archaeobotanical data have a temporal resolution that allows the distinction of the following four periods: 6200–5600 bc, 5600–5000 bc, 5000–4300 bc, 4300–3700 bc.

Data and resultsChronological and chronotypological frameworkIn Apulia, the archaeological record relative to the 7th millennium bc (the Mesolithic period) is poorly represented (Berger and

Guilaine, 2009). The earliest evidence of the Neolithic culture in Apulia dates to the beginning of the 6th millennium bc (a period associated with a significant increase in the number of datings), with the so-called ‘archaic’ impressed wares. The earlier peak appears to the south (Salento), and it is relative to the trans-Adri-atic contributions from the central-eastern Mediterranean regions (Guilaine and Cremonesi, 2003), as one would expect in terms of the demographic processes in human populations during the local Neolithic transition. Archaeological evidence obtained at later

Figure 3. Multisample probability curves of Archaeological Occupation (%)based on the 121 available radiocarbon dates plotted for the entire Apulia area and for the main morphological districts, as compared with the archaeological facies (modified after Pessina and Tinè, 2008). The grey bands indicate the dry periods.

Page 8

1304 The Holocene 23(9)

sites shows the coexistence of ‘evolved’ impressed and painted-incised wares, with which a further increase in datings is associated.

The middle centuries of the 6th millennium bc (the middle-Neolithic period) experienced a peak in occupation. This period is marked by the development of red-painted fine pottery, known as figulina, associated with painted-incised pottery. The produc-tion of figulina appears to have started in the Tavoliere and Salento areas c. 5600–5400 bc. In the Murge district, the diffu-sion of this type of pottery seems to have occurred slightly later, after 5400 bc; e.g. in the coastal area (the B1 district), the sites that date to c. 5600 bc continue to yield examples of ‘evolved’ impressed and/or painted-incised pottery. The complexity of decorative types and the occurrence of pottery assemblages made of wares belonging to different styles hinder a clear chron-ological distinction between the achaeological facies and a chro-nology based on these materials. For example, in the records from the middle and recent Neolithic periods, an association of incised and red-painted or trichrome with red-painted and Serra d’Alto wares, respectively, can be observed (Laviano and Muntoni, 2009; Muntoni, 2003).

During the last centuries of the 6th millennium bc, a general decrease in frequency occurred, as evidenced by a phase of reduc-tion in settlement density. This trend may reflect a reorganisation towards fewer, and likely larger, settlements corresponding to new economic and socio-organisational needs. This phase of reduction in settlement density is clearly visible in the archaeo-logical record. Such a trend continued during the first half of the 5th millennium bc. This period was characterised by the associa-tion of red-painted wares with trichrome wares; examples of these two types, dating to 5100 bc have been found to be associated with Serra d’Alto pottery. The oldest occurrence of the Serra d’Alto style in southern Italy was recorded in the Tavoliere and Murge districts (Laviano and Muntoni, 2009).

The second half of the 5th millennium bc was characterised by a slight increase in occupation, in Salento in particular, where the contextual occurrence of Serra d’Alto and Diana types dates to c. 4500 bc. The same association has only been found to have occurred in the Murge district (B1) beginning in the last centuries of the 5th millennium bc.

The last Neolithic occupation peak occurred c. 4000 bc and has been radiocarbon dated at certain sites in the Gargano and Murge districts where Macchia a Mare and/or Zinzulusa pottery have been found. These findings mark the end of the Neolithic cycle and the transition to the Copper Age.

Paleoclimatic data

Offshore sequences. The AD91-17 core, drilled offshore the Salento, was analysed by Sangiorgi et al. (2003) using a multiproxy approach (dinocysts, oxygen isotopes and pollen). For the time span examined the records exhibit several interesting features (Fig-ure 4). The δ18O data exhibit values between 0.5 and 0.9‰ PBD for the period between 6200 and 5000 bc and between 0.9 and 1.5‰ PBD for the period between 5000 and 4000 bc. Between 5000 and 4500 bc, the curves show a peak of 70% of warm dinocyst species associatied with a decreased abundance of deciduous Quercus and a peak in pollen corresponding to semi-desert species. Between 4500 and 4200 bc the increase in deciduous Quercus was concomi-tant with a decrease in semi-desert species and Pistacia sp. pollen. This trend changed again between 4200 and 3600 bc, when a peak of semi-desert species was associated with a decrease in deciduous Quercus followed by an increase in Pistacia.

Palynological data from the RF93-30 core, drilled off the northern coast of Gargano, reveal that between 4900 and 4700 bc an increase in evergreen oak pollen (Quercus ilex type) occurred along with a simultaneous decrease in deciduous Quercus and the

appearance of the Hordeum group (Mercuri et al., 2012; Oldfield et al., 2003). In the period immediately following, there was a relative increase in deciduous Quercus and a general increase in arboreal pollen. A new peak of evergreen oak pollen appeared c. 4100 bc.

Offsite data. Using the ARI24 core (Ariscianne wetland), an analysis of aquatic plants macroremains allowed us to examine the environmental evolution in the Murge district (Figure 5; Caldara et al., 2011). Until c. 5050 bc (MA24/1a zone), the Aris-cianne marsh was characterised by the absence of aquatic plants, with the exception of a small peak of Characeae c. 5200 bc.

From 5050 to 4500 bc (MA24/1b zone), the coring area was characterised by the presence of submerged plants (Nuphar and Potamogeton), together with charophytes, both of which provide evidence of a period of greater water availability.

From 4500 bc to 3700 bc (MA24/2 zone) the aquatic vegeta-tion was dominated only by emerging species, suggesting a period of less water availability.

Beyond the studied time interval the Ariscianne marsh contin-ued to have been marked by alternating phases characterised by the prevalence of emergent or submerged species (Caldara et al., 2011).

The analysis of the BAT1/2 core (Figure 6) was performed only on selected pollen from the PB1 and PB2 zones (Caroli, 2005; Caroli and Caldara, 2007). In both pollen zones, arboreal taxa, and Quercus in particular, exhibited high values (75–95%): in zone PB1 evergreen Quercus varied from 13% to 42%, Q. suber/cerris varied from 1% to 17% and deciduous Quercus var-ied from 10% to 30%. The proportions of these three taxa were similar in the PB2 zone.

Between 4800 and 4400 bc, data show an initial decrease in deciduous Quercus followed by an increase in evergreen Quercus (after a peak of microcharcoal occurrence) and the appearance of Ericacea. Evidence of an increase in deciduous Quercus, associ-ated with a decrease in evergreen Quercus between 4400 and 4100 bc can be observed.

Between 4100 and 3900 bc a new peak of evergreen Quercus occured. The appearance of Olea europaea and Pistacia may have been related to a shift towards more typical present-day Mediterranean conditions.

Archaeobotanical analysisThe archaeobotanical data base was developed by assessing 35 multistratified sites for which radiocarbon dates were available (Tables 2, 3).

The analysis of plant remains indicates diachronical changes in the exploitation of edible species (cereals and leguminosae) and fuelwood throughout the Neolithic period:

• 6200–5000 bc. Archaeobotanical data reveal high taxo-nomic variability for cereals with a predominance of Triti-cum and legumes and evidence of Avena and Bromus. However, between 5600 and 5000 bc, a slight decrease in barley (wheat/barley ratio increases) was associated with a rise of naked forms (hulled/naked ratio decreases; Table 3).

• 5000–4300 bc. This period continues with the predomi-nance of Triticum. Moreover, there was a small increase in Hordeum (the wheat/barley ratio decreased; Figure 7), and in hulled forms, as evidenced by an increase in the hulled/naked ratio. Change in the preference for certain cereals appears to have also been accompanied by a decrease in the taxonomic richness of cereals.

• 4300–3700 bc. Archaeobotanical data from this period provide evidence of a change in cropping systems. Within

Page 9

Fiorentino et al. 1305

the general framework of a further reduction in the taxo-nomic variability of cereals, Hordeum appears to have been the preferred cereal, and Triticum exhibited a decrease, resulting in a decrease in the wheat/barley ratio (Figure 7). This time span is also characterised by a reduc-tion of legumes, the disappearance of Bromus and Avena and the rise of naked cereals (the hulled/naked ratio decreases; Figure 7).

DiscussionThis study sheds new light on the relationships between certain climatic changes and human occupation in Apulia during the Neolithic period in addition to consequent changes in liveli-hood strategies.

All of the types of evidence used for the paleoenvironmental/paleoclimatic reconstruction of the region have their limitations, as mentioned in the ‘Materials and methods’ section.

Despite these limitations, it was possible to identify two dry periods (one between 5000 and 4600 bc and another peaking c. 4000 bc) and two wet periods (one between 6200 and 5500 bc and another peaking c. 4400 bc). The transition between these periods appears to have been gradual (Figures 3 and 4).

Regional climatic features were then compared (Figure 8) with selected data reported for the Mediterranean and global climate. For the Mediterranean area, we used data from the Accesa lake-level reconstruction (Magny et al., 2007; Vannière et al., 2008) and the standardised normal δ18O ‰ (V-PDB) records from the Soreq cave (Bar-Matthews and Ayalon, 2011; Bar-Matthews et al., 1997), Ioannina Lake (Frogley et al., 2001; Roberts et al., 2008) and Per-gusa Lake (Roberts et al., 2008; Sadori et al., 2008).

Table 2. List of settlements with available archeobotanical data sets and references.

Settlement Geomorphological district Archaeobotanical references

Early Neolithic c. 6200–5600 bc 1 Foggia, ex Ippodromo B2 D’Oronzo and Fiorentino (2006) 2 Foggia, Villa Comunale B2 Nisbet (1982) 3 Coppa Nevigata B1 Sargent (1987) 4 Masseria Candelaro, I phase B1 Ciaraldi (2004) 5 Lagnano da Piede B3 Jones (1987b) 6 Fontanelle C1 Coppola and Costantini (1987) 7 Monte Aquilone B1 Evett and Renfrew (1971) 8 Masseria Valente B1 Sargent (1983) 9 Monte Calvello B3 D’Oronzo et al. (2008)10 Monte San Vincenzo B3 D’Oronzo et al. (2008)11 Terragne E Fiorentino (1995a)12 Le Macchie C1 Costantini (1984)13 Titolo C1 Evett and Renfrew (1971)14 Scamuso, III phase C1 Costantini et al. (1997)15 Torre Canne C1 Coppola and Costantini (1987); Evett and Renfrew (1971)16 Ripa Tetta B3 Costantini and Tozzi (1987)17 Torre Sabea F Marinval (2003a, 2003b)18 Defensola A A Fiorentino (1995a)27 Canosa B2 This study34 Grotta delle Mura C2 Gaglione (2007)35 Pulo di Molfetta C1 Primavera and Fiorentino (2011)Middle Neolithic c. 5600–4800 bc 4 Masseria Candelaro, II phase B1 Ciaraldi (2004) 4 Masseria Candelaro, III phase B1 Ciaraldi (2004)14 Scamuso II phase C1 Costantini et al. (1997)18 Defensola A A Fiorentino (1995b)19 Grotte Santa Croce C2 Boscato et al. (2002); Fiorentino (2002)20 Grotta Sant’Angelo C2 Castelletti (1972); Costantini (1991)21 Fontanarosa B1 Sargent (1983)22 Santa Tecchia B1 Sargent (1983)23 Oria Sant’Anna E Castiglioni and Rottoli (2007)24 Capo Rondinella G Fiorentino (1999)25 San Domenico G Fiorentino (1999)26 Passo di Corvo B2 Follieri (1973, 1983)Recent Neolithic c. 4800–4300 bc14 Scamuso, I phase C1 Costantini et al. (1997)28 Grotta della Tartaruga di

Lama GiottaC1 This study

29 Carpignano Salentino F Primavera (2008); Fiorentino and Primavera (2009)30 Serra Cicora F Merico (2005)Final Neolithic c. 4300–4000 bc31 Cala Scizzo C1 Castelletti et al. (1987); Costantini (1984)32 Cala Colombo C1 Castelletti et al. (1987)33 Madonna delle Grazie C2 Costantini (1984)

Page 10

1306 The Holocene 23(9)

Tabl

e 3.

Gen

eral

arc

heob

otan

ical

dat

a ba

se u

sed

subd

ivid

ed fo

r di

ffere

nt s

ettle

men

ts a

nd N

eolit

hic

phas

es; s

peci

es n

omen

clat

ure

is t

hat

of t

he a

utho

rs.

Earl

y N

eolit

hic

c. 62

00–5

600

bc

Fogg

ia e

x-Ip

podr

omo

Mon

te

Cal

vello

Mon

te S

an

Vin

cenz

oTe

rrag

neM

asse

ria

Vale

nte

Tito

loSc

amus

o,

III p

hase

Font

anel

leTo

rre

Can

neC

oppa

N

evig

ata

Lagn

ano

da P

iede

Mas

seri

a C

ande

laro

, I p

hase

Mon

te

Aqu

ilone

Rip

a Te

tta

Fogg

ia, V

illa

Com

unal

eTo

rre

Sabe

aC

anos

aG

rott

a de

lle M

ura

Pulo

di

Mol

fett

aD

efen

sola

A

Le

Mac

chie

Tota

l

Aven

a54

00

00

00

00

20

00

00

00

00

00

56

Brom

us s

p.4

01

00

00

01

00

00

00

00

00

00

6

Seca

le0

00

00

00

00

00

00

00

00

00

00

0

Triti

cum

mon

ococ

cum

251

31

00

304

011

80

02

121

11

00

05

204

Triti

cum

dico

ccum

255

31

13

163

29

123

41

015

141

21

01

2640

0

Triti

cum

mon

ococ

cum

/di

cocc

um0

01

00

014

712

00

00

170

00

00

02

53

Triti

cum

spe

lta0

00

00

00

00

20

00

00

10

00

00

3

Triti

cum

aes

tivum

/dur

um8

11

00

06

30

00

00

10

15

00

00

26

Triti

cum

com

pact

um0

30

10

07

00

20

00

00

00

00

00

13

Triti

cum

sp.

012

01

00

132

12

414

09

00

00

40

02

1759

4

Hor

deum

vul

gare

381

40

00

213

07

00

00

70

10

11

03

276

Hor

deum

vul

gare

var

. di

stic

hum

160

30

00

05

180

00

00

01

70

01

152

Hor

deum

vul

gare

var

. te

trast

ichu

m0

00

00

00

00

00

00

00

00

00

00

0

Hor

deum

var

. nud

um0

10

00

00

00

00

00

00

00

00

00

1

Hor

deum

sp.

176

71

010

104

05

374

30

00

01

10

23

201

Triti

cum

/Hor

deum

3194

61

00

00

00

00

00

00

200

00

015

2

Pani

cum

milia

ceum

00

01

00

00

00

00

00

00

00

00

01

Cere

alia

00

30

00

00

00

00

00

00

00

00

03

Lath

yrus

sat

ivum

00

00

00

00

00

00

00

00

00

00

00

Lath

yrus

sp.

00

00

00

600

00

00

00

00

00

00

060

Legu

min

osae

06

01

00

370

02

00

00

00

00

00

046

Lens

cul

inar

is5

94

00

013

02

01

00

00

00

00

00

34

Pisu

m s

ativu

m0

00

00

00

00

00

00

00

00

00

00

0

Pisu

m s

p.0

02

00

021

00

09

00

08

00

00

00

40

Vicia

erv

ilia0

10

00

00

00

00

00

00

00

00

00

1

Vicia

faba

00

00

00

00

00

00

00

01

00

00

01

Vicia

sp.

22

21

00

20

00

00

00

00

00

00

09

Tota

l22

32

(Con

tinue

d)

Page 11

Fiorentino et al. 1307

Mid

dle

Neo

lithi

c c.

5600

–500

0 bc

Mas

seri

a C

ande

laro

II

phas

e

Sant

a Tec

chia

Def

en-

sola

AM

asse

ria

Can

dela

ro

II-III

pha

se

Ori

a Sa

nt'A

nna

Gro

tta

Sant

a C

roce

Sant

a Tec

chia

Gro

tta

Sant

’Ang

elo

Font

anar

osa

Scam

uso

II ph

ase

Cap

o R

ondi

nella

San

Dom

enic

oPa

sso

di

Cor

voTo

tal

Aven

a3

00

160

100

20

412

01

48Br

omus

sp.

00

00

13

00

00

00

04

Seca

le0

00

00

00

10

00

00

1Tr

iticu

m m

onoc

occu

m16

270

707

820

172

50

1813

44

535

Triti

cum

dico

ccum

3235

120

434

104

036

84

1410

238

910

53Tr

iticu

m m

onoc

occu

m/

dico

ccum

970

029

810

00

30

20

00

410

Triti

cum

spe

lta0

00

00

00

00

00

03

3Tr

iticu

m a

estiv

um/d

urum

00

010

30

010

22

167

710

2Tr

iticu

m c

ompa

ctum

00

03

061

00

00

10

1176

Triti

cum

sp.

446

210

43

253

656

1616

950

2051

3H

orde

um v

ulga

re1

00

510

121

042

013

65

224

1H

orde

um v

ulga

re v

ar.

dist

ichu

m0

01

27

630

00

02

00

75

Hor

deum

vul

gare

var

. te

trast

ichu

m0

00

10

00

00

00

00

1

Hor

deum

var

. nud

um0

00

00

00

110

00

00

11H

orde

um s

p.1

22

90

524

02

628

04

110

Triti

cum

/Hor

deum

00

00

104

00

00

3321

343

00

780

Pani

cum

milia

ceum

00

00

01

00

00

43

08

Cere

alia

360

011

3314

40

00

00

00

1871

Lath

yrus

sat

ivum

00

00

00

00

00

00

00

Lath

yrus

sp.

10

09

00

00

01

00

314

Legu

min

osae

00

00

20

00

05

260

033

Lens

cul

inar

is0

00

20

00

10

210

00

15Pi

sum

sat

ivum

00

00

10

00

00

00

01

Pisu

m s

p.0

00

00

00

00

10

00

1Vi

cia e

rvilia

00

01

00

00

00

110

012

Vicia

faba

00

00

00

00

00

00

33

Vicia

sp.

40

031

13

03

00

00

042

Tota

l52

79

Tabl

e 3.

(C

ontin

ued)

(Con

tinue

d)

Page 12

1308 The Holocene 23(9)

Rec

ent

Neo

lithi

c c.

5000

–430

0 bc

Car

pign

ano

Sale

ntin

oSe

rra

Cic

ora

Scam

uso,

I ph

ase

Gro

tta

della

Tar

taru

ga d

i Lam

a G

iott

aTo

tal

Fina

l Neo

lithi

c c.

4300

–370

0 bc

Cal

a Sc

izzo

Cal

a C

olom

boM

adon

na d

elle

Gra

zie

Tota

l

Aven

a0

20

35

00

00

Brom

us s

p.0

00

00

00

00

Seca

le0

00

00

00

00

Triti

cum

mon

ococ

cum

05

04

90

00

0Tr

iticu

m d

icocc

um0

06

1016

02

02

Triti

cum

mon

ococ

cum

/di

cocc

um0

00

00

00

00

Triti

cum

spe

lta0

00

00

00

00

Triti

cum

aes

tivum

/du

rum

01

00

17

00

7

Triti

cum

com

pact

um0

00

00

00

00

Triti

cum

sp.

101

046

047

10

1695

111

Hor

deum

vul

gare

106

63

253

70

10H

orde

um v

ulga

re v

ar.

dist

ichu

m0

20

02

00

00

Hor

deum

vul

gare

var

. te

trast

ichu

m0

00

00

00

00

Hor

deum

var

. nud

um0

00

00

00

00

Hor

deum

sp.

04

81

130

155

20Tr

iticu

m/H

orde

um0

07

07

00

00

Pani

cum

milia

ceum

00

02

20

00

0Ce

real

ia0

80

5462

00

00

Lath

yrus

sat

ivum

00

01

10

00

0La

thyr

us s

p.0

00

1414

04

04

Legu

min

osae

00

00

00

00

0Le

ns c

ulin

aris

00

15

60

00

0Pi

sum

sat

ivum

00

00

00

00

0Pi

sum

sp.

00

00

00

00

0Vi

cia e

rvilia

00

00

00

00

0Vi

cia fa

ba0

00

00

00

00

Vicia

sp.

00

00

00

00

0To

tal

634

154

Tabl

e 3.

(C

ontin

ued)

Page 13

Fiorentino et al. 1309

Figure 4. Paleoenvironmental data from cores drilled in the Adriatic Sea: AD91-17 (Sangiorgi et al., 2003) and RF93-30 (Mercuri et al., 2012; Oldfield et al., 2003). Grey bands indicate the dry periods.

Figure 5. Plant macrofossil sequence from core ARI 24, Ariscianne wetland.Source: modified from Caldara et al. (2011).

Figure 6. Pollen percentage diagram for core BAT1/2, Battaglia Lake, with selected taxa and microcharcoal concentrations.Source: modified from Caroli (2005) and Caroli and Caldara (2007).

Page 14

1310 The Holocene 23(9)

These Mediterranean sequences differ for the various proxies that were used, each of which is characterised by its own time-resolution capability. At Soreq, the δ18O was analysed for speleo-thems with an average time resolution of one sample every 138 years. At Ioannina, δ18O was measured for ostracods with an aver-age time resolution of one sample every 125 years. At Pergusa, δ18O was determined for sediments with an average time resolu-tion of one sample every 250 years. At Accesa, the lake level was inferred from several markers (e.g. the lithology, grain size and macroscopic components of lake marl) with an average resolution of a datum every 100 years on average. The Lago Grande di Mon-ticchio pollen sequence (Allen et al., 2002) was examined because of its geographical proximity to Apulia.

These records were selected because of their chronological resolution capabilities, their distribution around the Mediterranean sea, and their position in areas influenced by the different pressure systems/indexes that affect the Mediterranean climate (Lionello et al., 2005), i.e. the North Atlantic oscillation, the South Asian monsoon, the Siberian High, and the Southern oscillation (Xoplaki et al., 2003).

With respect to the global climate, the Gaussian smoothed 200 yr Greenland GISP2 potassium curve (Mayewski et al., 1997, 2004) has been considered as a proxy for the Siberian High fluc-tuations, that dictate winter precipitation and crop seasonality in the Mediterranean area. In fact, the strengthening of the Siberian High drives polar air outbreaks in the northeastern Mediterranean (Rohling et al., 2002), which are associated with increases in win-ter rainfall (Lionello et al., 2005; Weninger et al., 2009).

The comparative data analysis made it possible to reconstruct sig-nificant climatic and environmental changes on a centennial scale.

6200–5000 bc

• The archaeobotanical data provide evidence of the culti-vation of several different cereal species (and thus evi-dence of a high taxonomic richness), among which Triticum was the best-represented taxon (Table 3); evi-dence of the presence of legumes accompanied by other minor cereals (Avena and Bromus) suggests that the har-vest occurred both in the spring and summer (Figure 7). Between 5600 and 5000 bc there was evidence of a

decrease in Hordeum and an increase in naked cereals. These findings suggest a period characterised, on the whole, by good water availability.

• The GISP2 curve (Figure 8g) shows the occurrence of an oscillating trend towards a progressive reduction in winter rainfall for the entire studied period. For the last centuries of the 6th millennium, the curve shows a marked decrease in winter rainfall, when our data show an increase in Hor-deum and a reduction in T. dicoccum.

• The Ariscianne sequence provides evidence of a low abundance of submerged aquatic plants, indicating a low water availability from 5300 to 5050 bc.

• The AAO curve (Figure 8a) exhibits a peak between 5800 and 5200 bc. Later, c. 5000 bc, a sharp population decrease occurred that can be attributed to the first of the two Neolithic dry phases in Apulia. Disaggregated data analysis performed for the main districts (Figure 3) allowed us to examine the sensitivity of each area to cli-mate change. For example, the Tavoliere plain, which experienced an almost total depopulation c. 5000 bc, could be considered a more sensitive area than the Murge and Gargano areas.

• The oxygen isotope curve for the Soreq Cave (Figure 8b) suggests the occurrence of a wet phase between 6000 and 5500 bc followed by a progressive decrease in rainfall; the Ioannina record (Figure 8c) shows a very similar pattern. In southern Italy, the Monticchio pollen sequence pro-vides evidence of a wet period from c. 6200 to c. 6000 bc (as evidenced by an increase in Abies, as reported by Allen et al., 2002).

• In central Italy, the Accesa sequence recorded a high lake level between 6000 and 5300 bc followed by a period of lake level decline from 5300 to 5000 bc (Magny et al., 2007; Vannière et al., 2008; Figure 8d). The Valle di Cas-tiglione sequence recorded an increase in arborean pollen c. 6200 bc following a typical postglacial phase (Celant, 2000; Di Rita et al., 2013). These data are consistent with Pergusa record, whose oxygen curve exhibits evidence of a dry phase starting c. 5400 bc (Figure 8e).

Figure 7. Schematic reconstruction of the relationships between vegetation and paleoagricultural indicators determined using archaeological data.

Page 15

Fiorentino et al. 1311

Figure 8. Interval of 6200 to 3700 cal. yr bc. Comparison between: (a) probability curves of Archaeological Occupation (%) for Apulia (this study) and a suite of regional paleoenvironmental records, (b) standardised normal δ18O ‰ (V-PDB) records for Soreq cave (Bar-Matthews and Ayalon, 2011; Bar-Matthews et al., 1997); (c) standardised normal δ18O records for Ioannina Lake (Frogley et al., 2001; Roberts et al., 2008); (d) Accesa lake-level reconstruction: raw data (Magny et al., 2007) and trends (dotted line, Vannière et al., 2008); (e) δ18O ‰ (V-PDB) at Pergusa Lake (Roberts et al., 2008; Sadori et al., 2008); (f) schematic reconstruction of seasonal paleoagricultural indicators; (g) Gaussian smoothed (200 yr Greenland GISP2 potassium (non-sea salt K+; ppb) ion proxy for the Siberian High; Mayewski et al., 1997, 2004).

Page 16

1312 The Holocene 23(9)

In summary, during this period, Apulia was characterised by relatively wet conditions, until a time between 5600 bc and 5200 bc when dry conditions began to predominate. These dry condi-tions continued to predominate until the end of the 6th millen-nium bc. These findings appear to be consistent with the Mediterranean climate dynamics of the time.

5000–4600 bc

• The available archaeobotanical data show a chronological resolution of 700 yr (5000–4300 bc), thus it is not possible to distinguish the two subphases (5000–4600 and 4600–4300 bc). Archaeobotanical data regarding the whole period provide evidence of a small increase in hulled cere-als (hulled/naked ratio increases; Figure 7). Moreover, the presence of legumes and the absence of Bromus suggest that the harvest occurred in spring (Figure 7). These evi-dences indicate winter precipitation. Given the reduced taxonomic richness, we suppose that only those plants adapted to the changed eco-edaphic conditions were harvested.

• The interpretation of the GISP2 record (Figure 8g) suggests drier climatic conditions, with the exception of a peak c. 4800 bc that suggests a short-term relative increase in winter pre-cipitation with positive effects on spring crop growth.

• The offsite data show two conflicting trends. The BAT1/2 core provides evidence of a progressive strengthening of dry conditions in accordance with the regional climatic framework reconstructed herein. In contrast, the Arisci-anne marsh horizons provide evidence of a prevalence of submerged plants during this time, indicating that the marsh was not dry and that the Murge area was character-ised by different climate conditions than the rest of Apulia. This climatic diversity appears to be consistent with data from the IN68-9 and MD 90-917 cores (Figure 9). In fact, these cores, drilled in the middle Adriatic Sea off the shore of the Ariscianne marsh, provide evidence of a low-tem-perature peak that would suggest a short-term cold-wet phase that differed from the general regional trend.

• The AAO curve provides evidence of a drastic reduction of occupation of the Tavoliere plain (Figure 3). The Gar-gano, Murge and Salento areas appear to have experi-enced a less severe decline, showing differences that we believe were due to both subregional climate diversity (Murge area) and changes of the occupation and land con-trol dynamics (Cassano, 1993).

• At the Mediterranean scale, the paleoclimate data support a strengthening of the arid phase which started at the end of the previous period. In particular, the Soreq curve exhibited its driest phase (Figure 8b), whereas Ioannina (Figure 8c), where the driest phase appears to have occurred several centuries earlier, exhibited a relative improvement of climatic conditions which were nonethe-less predominantly dry. Peninsular Italian data for the Accesa Lake (Figure 8d) indicate that the water-table was lower than it was during the previous phase. The Pergusa oxygen isotope curve (Figure 8e) shows values similar to those recorded during the previous period.

In general, the time span between 5000 and 4600 bc is charac-terised by the onset of the driest climatic conditions with respect

to the entire Neolithic period in Italy. Probably, as highlighted by Ariscianne sequence and archaeobotanical data, this dry period was interrupted by a short-term cold-wet phase peaked c. 4800 bc.

4600–4300 bc

• A decrease in winter rainfall is suggested by the GISP2 record after the 4800 bc peak (Figure 8g). This pulsation could have triggered changes in seasonal crop rhythms and the selection of plants more adapted to summer har-vesting (Hordeum and T. monococcum).

• The offsite data (Figure 4) provide evidence of a rapid increase of the presence of meso-thermophilous plants, peaking c. 4300 bc. Shortly thereafter, the occurrence of these species rapidly diminishes with a contemporaneous increase in evergreen Quercus.

• The AAO curve provides evidence of a relative popula-tion increase in Apulia during this period (Figure 3). The most frequented area appears to have been the Salento district, where a large body of archaeological evidence has been found in caves or in funerary/cultural contexts. A slight increase in occupation was also recorded in the Tavoliere plain. In contrast, the Gargano headland experienced its lowest population levels, per-haps because of a decline in the demand for chert (Tar-antini and Galiberti, 2011).

• The Soreq curve (Figure 8b) provides evidence of an increase in rainfall; a wetter phase appears to have occurred shortly thereafter at Ioannina (Figure 8c). The Accesa curve shows a decrease in the lake level (Figure 8d).

In general, this period was characterised by wetter conditions than to the previous period, although winter rainfalls were low.

4300–3700 bc

• The archaeobotanical data for this period provide evi-dence of a preference for barley (wheat/barley ratio decreases) and an increase of naked cereals (hulled/naked ratio decreases; Figure 7). The presence of legumes indi-cates that the harvest occurred in the spring.

• The GISP2 curve (Figure 8g) exhibits a series of positive peaks between 4300 and 3700 bc, indicating a discontinu-ous increase in winter precipitation.

• The offsite data from Battaglia Lake provide evidence of two phases. The first, which peaked c. 4000 bc, was char-acterised by the presence of xerophilous vegetation asso-ciated with a low-precipitation regime. The second phase, between c. 3800 and 3900 bc, was characterised by a trend towards meso-thermophilous species enrichment (Figure 6). Evidence from the Ariscianne marsh suggests that this time span was characterised by a limited water availability.

• The AAO curve shows that frequentation in Apulia diminished during this period, except in the Murge area (Figure 3) where several settlements, characterised by Neolithic Diana facies evolving into early-Eneolithic culture, have recently been discovered (Radina, 2006; Radina, 2013).

• The Soreq curve provides evidence (Figure 8b) of a rela-tively dry phase between 4200 and 4000 bc and a gradual

Page 17

Fiorentino et al. 1313

increase in wetness between 4000 and 3700 bc. The Acc-esa record provides evidence of a high lake-level phase (Figure 8d) followed by a lowering of the lake level shortly after 4000 bc and another subsequent water level increase. The same trend (i.e. wet-dry-wet succession) is shown by the Ioannina curve (Figure 8c).

In brief, during this period, there was an initial decrease in wetness that peaked c. 4000 bc; thereafter, a new winter rainfall peak led to the onset of a new wet phase at the end of this period.

ConclusionsIn Apulia, the period between 6200 and 3700 bc was character-ised by the development of Neolithic farm-based cultures. We found that this process was significantly correlated to ‘minor’ Holocene climate oscillations.

The climate history was reconstructed using offsite and off-shore data derived from new and published studies based on the analysis of different proxies (e.g. δ18O values, dinocysts, pollen and macrobotanical remains). We found that wet conditions pre-vailed between 6200 and 5500 bc. Subsequently, a progressive drying phase occurred and peaked after 5000 bc. This drying phase was followed by a brief wet period c. 4500–4300 bc. The last phases of the Apulian Neolithic cultures developed under dry conditions that evolved near the end of the studied period into a progressively wetter climate.

The multisample probability curve for the Apulia Archaeo-logical Occupation (AAO) indicates that the settlement density oscillated during the studied time span with a positive peak occur-ring during the early Neolithic (c. 5500 bc) followed by a progres-sive reduction. An additional peak in occupation was recorded during the final Neolithic period (c. 4500 bc).

We found a good correlation between climate changes and cul-tural evolution, especially during the two arid phases that occurred near the end of the 6th millennium bc and at the beginning of the 4th millennium bc. These phases coincided with the most evident gaps in the archaeological record, namely during the middle-recent Neolithic and final Neolithic periods.

Unfortunately, the available data do not permit a definition, within a detailed chronological time frame, of the cause–effect rela-tionships between climatic changes and human responses (Miller

Rosen, 2007) and a clear and unambiguous discrimination of wet and dry periods (Blaauw et al., 2007); the diversity of subregional climatic conditions further complicates this reconstruction.

Comparing the AD91-17 record, the RF93-30 and BAT1/2 sequences, we found that recognised climate phases exhibited certain minor inconsistencies. We believe that these inconsis-tencies are attributable to both the different time resolution capabilities of the sequences and the different time responses of the proxies that were used (see Materials and methods section)

Similar discrepancies emerged in the comparison of regional climate oscillations and variations recognised within the main Mediterranean Holocene records (i.e. Soreq, Ioannina, Accesa and Pergusa).

We propose that these discrepancies are due to two factors. First, the considered records reflect different climatic histories across areas influenced by varied air masses and contributions. The second set of factors are due to the different proxies used and the varied time resolution of the sequences (see the Discussion section). The archaeobotanical evidence allowed us to identify a direct link between the paleoclimatic and archaeological sequences. These data highlight changes in agricultural strategies (e.g. the species used and cropping periods) as human responses to changes in the rainfall regime.

Because several years of drought were likely to be stressful for ancient communities, the productive strategies chosen vary according to the community resilience (Crumley, 1994; Orlove, 2005; Redman, 2005; Redman and Kinzig, 2003), i.e. the capacity to adapt qualitatively as well as quantitatively to severe, unex-pected and continuous changes, such as those reconstructed for the middle Holocene in the Apulia.

AcknowledgementsWe are very grateful to two anonymous reviewers for useful sug-gestions which have made possible a considerable improvement of our manuscript. We are also grateful to Grace Carone and American Journal Experts (AJE) for the English revision.

FundingThis work has been supported by PRIN 2010-11, Operative Unit of Bari: ‘The contribution of earth sciences (geophysics, remote sensing, geomorphology, archaeometry) to landscape archaeol-ogy studies in northern Apulia (Italy)’.

Figure 9. Emergent and submerged plant records from Ariscianne wetland (ARI 24) correlated with modified curves of warm foraminifera specieis in IN68-9 (Casford et al., 2001) and sea surface temperatures in MD90-917 (Siani et al., 2010).

Page 18

1314 The Holocene 23(9)

ReferencesAllen JRM, Watts WA, McGee E et al. (2002) Holocene environmental vari-

ability – The record from Lago Grande di Monticchio, Italy. Quaternary International 88: 69–80.

Azzi G (1930) Le climat du blé dans le monde. Les bases écologiques de la culture mondiale du blé. Rome: Institut International d’agriculture.

Bar-Matthews M and Ayalon A (2011) Mid-Holocene climate variations revealed by high-resolution speleothem records from Soreq Cave, Israel and their correlation with cultural changes. The Holocene 21(1): 163–171.

Bar-Matthews M, Ayalon A and Kaufman A (1997) Late Quaternary paleo-climate in the eastern Mediterranean region from stable isotope anal-ysis of speleothems at Soreq Cave, Israel. Quaternary Research 47: 155–168.

Berger J-F and Guilaine J (2009) The 8.200 cal BP abrupt environmental change and the Neolithic transition: A Mediterranean perspective. Quater-nary International 200: 31–49.

Blaauw M, Christen JA, Mauqoy D et al. (2007) Testing the timing of radiocarbon-dated events between proxy archives. The Holocene 17(2): 283–288.

Boenzi F, Caldara M, Moresi M et al. (2002) History of the Salpi lagoon-sab-hka (Manfredonia Gulf, Italy). Il Quaternario 14(2): 93–104.

Boenzi F, Caldara M and Pennetta L (1991) Alcuni aspetti del rapporto fra l’uomo e l’ambiente carsico in Puglia. Itinerari speleologici s. II(5): 41–51.

Boenzi F, Caldara M, Pennetta L et al. (2006) Environmental aspects related to the physical evolution of some wetlands along the Adriatic coast of Apulia (Southern Italy): A review. Journal of Coastal Research S139: 170–175.

Boscato P, Gambassini P and Ronchitelli AM (2002) Una stuoia in fibre veg-etali del Neolitico Antico nella Grotta di Santa Croce. In: Radina F (ed.) La Preistoria della Puglia. Paesaggi, uomini e tradizioni di 8.000 anni fa. Bari: Mario Adda Editore, pp. 71–76.

Calattini M and Muntoni IM (2005) Il Neolitico del Gargano. In: Galiberti A (ed.) Defensola. Una miniera di selce di 7000 anni fa. Siena: Protagon Editore, pp. 33–40.

Caldara M and Palmentola G (1991) Lineamenti geomorfologici del Gargano con particolare riferimento al carsismo. Itinerari speleologici s. II(5): 53–66.

Caldara M and Pennetta L (1993) Nuovi dati per la conoscenza geologica e morfologica del Tavoliere di Puglia. Bonifica 8(3): 25–42.

Caldara M and Pennetta L (2002) L’ambiente fisico delle Murge durante il Neolitico. In: Radina F (ed.) La Preistoria della Puglia. Paesaggi, uomini e tradizioni di 8.000 anni fa. Bari: Mario Adda Editore, pp. 21–26.

Caldara M, Caroli I and Simone O (2008) Holocene evolution sea-level changes in the Battaglia basin area (eastern Gargano coast, Apulia, Italy). Quaternary International 183(1): 102–114.

Caldara M, Muntoni IM, Fiorentino G et al. (2011) Hidden Neolithic land-scapes in Apulian Region. In: Van Leusen M, Pizziolo G and Sarti L (eds) Hidden Landscapes of Mediterranean Europe. Cultural and Meth-odological Biases in Pre- and Protohistoric Landscape Studies. British Archaeological Reports International Series 2320, Oxford: Archeopress, pp. 183–191.

Caldara M, Pennetta L and Simone O (2002) Holocene evolution of the Salpi Lagoon (Apulia, Italy). Journal of Coastal Research SI 36: 124–133.

Caldara M, Pennetta L and Simone O (2005) L’ambiente fisico nell’area dell’insediamento. In: Cassano SM and Manfredini A (eds) Masseria Can-delaro. Vita quotidiana e mondo ideologico in un comunità neolitica del Tavoliere. Foggia: Claudio Grenzi Editore, pp. 25–38.

Caroli I (2005) Dinamiche climatiche ed ambientali dell’area garganica nel corso dell’Olocene. PhD Thesis, Università degli Studi di Bari.

Caroli I and Caldara M (2007) Vegetational history of Lago Battaglia (eastern Gargano coast, Apulia Italy) during the middle–late Holocene. Vegetation History Archaeobotany 16(4): 317–327.

Casford JSL, Abu-Zied R, Rohling EJ et al. (2001) Mediterranean climate vari-ability during the Holocene. Mediterranean Marine Science 2(1): 45–55.

Cassano SM (1993) La facies Serra d’Alto: Intensificazione delle attività produttive e aspetti del rituale. Origini 17: 221–245.

Cassano SM and Manfredini A (1983) Studi sul Neolitico del Tavoliere della Puglia. Indagine territoriale in un’area campione. Oxford: British Archaeological Report, Int. S. 160.

Cassano SM and Manfredini A (eds) (2005) Masseria Candelaro. Vita quotidi-ana e mondo ideologico in un comunità neolitica del Tavoliere. Foggia: Claudio Grenzi Editore.

Castelletti L (1972) Contributo alle ricerche paletnobotaniche in Italia. Rendi-conti Istituto Lombardo, Acc. Scienze e Lettere, Classe di Lettere e Scienze Moralie Storiche 106(2): 331–374.

Castelletti L, Costantini L and Tozzi C (1987) Considerazioni sull’economia e l’ambiente durante il Neolitico. In: Atti della XXVI Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria. Firenze, 7–10 Novembre 1985, pp. 37–55.

Castigioni E and Rottoli M (2007) Economia e quadro paleoambientale. In: Tiberi I (ed.) Sant’Anna (Oria). Un sito specializzato del VI millennio a.C. Galatina: Congedo Editore, pp. 153–165.

Ciaraldi M (2004) Analisi dei resti vegetali: cambiamenti economici ed evi-denze rituali. In: Cassano SM and Manfredini A (eds) Masseria Cande-laro. Vita quotidiana e mondo ideologico in una comunità neolitica del Tavoliere. Foggia: Grenzi editore, pp. 447–462.

Celant A (2000) Nuovi dati archeobotanici su ambiente e agricoltura nel Neo-litico del Lazio: Un esempio dalla Campagna Romana. In: Pessina A and Muscio G (eds) La neolitizzazione tra Oriente e Occidente. Udine, 23–24 Aprile 1991, pp. 355–364.

Cattaneo A, Correggiari A, Langone L and Trincardi F (2003) The late-Holo-cene Gargano subaqueous delta, Adriatic shelf: Sediment pathways and supply fluctuations. Marine Geology 193 : 61–91.

Cipolloni Sampò M (1980) Comunità neolitiche della valle dell’Ofanto: Pro-posta di lettura di un’analisi territoriale. In: Padula M (ed.) Attività arche-ologica in Basilicata 1964–1977. Scritti in onore di Dinu Adamesteanu. Matera: Meta ed., pp. 283–311.

Cipolloni Sampò M (1982) Ambiente, economia e società dall’Eneolitico all’età del Bronzo in Italia sud-orientale. Dialoghi di Archeologia 4: 27–38.

Cipolloni Sampò M (1987) Il Neolitico antico nella valle dell’Ofanto: Con-siderazioni su alcuni aspetti dell’area murgiana. In: Atti della XXV Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria. Firenze, 7–10 Novembre 1985, pp. 155–168.

Coppola D and Costantini L (1987) Le Néolithique ancien littoral et la diffu-sion des céréales dans les Pouilles durant le VIe millenaire: Les sites de Fontanelle, Torre Canne et le Macchie. In: Guilaine J, Courtin J, Roudil JL and et al. (eds) Premiéres Communautés Paysannes en Méditerranée Occidentale. Actes du Colloque International du CNRS. 26–29 April Montpellier 1983, Paris, pp. 249–255.

Corrado A and Ingravallo E (1988) L’insediamento di masseria Le Fiatte (Manduria) nel popolamento neolitico del nord-ovest del Salento. Studi di Antichità 5: 5–78.

Costantini L (1984) Cereali carbonizzati e impronte del Neolitico pugliese. In: Atti 3° Convegno Nazionale sulla Preistoria – Protostoria e Storia della Daunia. San Severo, 27–29 Novembre 1981, pp. 107–111.

Costantini L (1991) Origen i difusiò de l’agricultura e la Itàlia meridional. Cota Zero 7: 103–114.

Costantini L and Tozzi C (1987) Un gisement à céramique imprimée dans le subapennin de la Daunia (Lucera, Foggia): Le village de Ripa Tetta. Écon-omie et culture matérielle. In: Guilaine J, Courtin J, Roudil JL and et al. (eds) Premiéres Communautés Paysannes en Méditerranée Occidentale. Actes du Colloque International du CNRS. 26–29 april Montpellier 1983, Paris, pp. 387–394.

Costantini L, Costantini Biasini L and Lentini A (1997) Agricoltura e note sull’ambiente dell’abitato neolitico di Scamuso. In: Biancofiore F and Coppola D (eds) Scamuso: Per la storia delle comunità umane tra il VI ed il III millennio nel Basso Adriatico. Roma: Università ‘Tor Vergata’, pp. 199–210.

Crumley C (1994) Historical Ecology: Cultural Knowledge and Changing Landscapes. School of American Research Advanced Seminar Series.

Delano Smith C (1979) Western Mediterranean Europe. A Historical Geog-raphy of Italy Spain and Southern France since the Neolithic. London: Academic Press.

De Martonne E (1942) Nuovelle carte mondiale de l’indice de la aridité. Annales de Géographie 1: 241–250.

Di Rita F, Anzidei AP and Magri D (2013) A Lateglacial and early Holo-cene pollen record from Valle di Castiglione (Rome): Vegetation dynamics and climate implications. Quaternary International 288: 73–80.

D’Oronzo C and Fiorentino G (2006) Analisi preliminare dei resti carpologici rinvenuti nel villaggio neolitico di Foggia (località ex-Ippodromo). In: Atti del 26° Convegno sulla Preistoria – Protostoria e Storia della Daunia. San Severo, 10–11 Dicembre 2005, pp. 33–38.

D’Oronzo C, Fiorentino G and Gaglione L (2008) L’analisi archeobotanica condotta nel sito neolitico di Foggia in località Monte Calvello. In: Atti del 28° Convegno sulla Preistoria – Protostoria e Storia della Daunia: San Severo, 25–26 Novembre 2007, pp. 441–448.

Evett D and Renfrew J (1971) L’agricoltura neolitica italiana: Una nota sui cereali. Rivista di Scienze Preistoriche 26: 403–407.

Fiorentino G (1995a) Analisi dei macroresti vegetali. In: Gorgoglione MA, Di Lernia S and Fiorentino G (eds) L’insediamento preistorico di Terragne

Page 19

Fiorentino et al. 1315

(Manduria - Taranto), nuovi dati sul processo di neolitizzazione del sud-est italiano. Manduria: C.R.S.E.C., pp. 171–184.

Fiorentino G (1995b) New perspectives in anthracological analysis. Palaeo-cological technological implications of charcoals found in the Neolithic Flintmine at La Defensola (Vieste, Puglia, Italy). Quaternaria nova 5: 99–128.

Fiorentino G (1999) Caratteristiche della vegetazione e abitudini alimen-tari durante la preistoria. In: Mastronuzzi G and Marzo P (eds) Le isole Chéradi, fra natura, leggenda e storia. Mottola: Stampasud, pp. 69–78.

Fiorentino G (2002). Alcuni dati archeobotanici sulla Bassa Murgia barese da Grotta Santa Croce. In: Radina F (ed.), La preistoria della Puglia. Pae-saggi, uomini e tradizioni di 8.000 anni fa. Bari: Mario Adda editore, pp. 85–86.

Fiorentino G and Primavera M (2009) Analisi preliminare dei dati archeobo-tanici della sepoltura neolitica di Carpignano Salentino. In: Fabbri PF and Pagliara C (eds) Prima di Carpignano. Documentazione e interpretazione di una sepoltura neolitica. Lecce: Terra, pp. 83–92.

Follieri M (1973) Cereali del villaggio neolitico di Passo di Corvo (Foggia). Annuali di Botanica 32: 49–59.

Follieri M (1983) Resti di piante alimentari: Cereali e leguminose. In: Tinè S (ed.) Passo di Corvo e la civiltà neolitica del Tavoliere. Genova: SAGEP Editrice, pp. 158–160.

Frogley MR, Griffiths HI and Heaton THE (2001) Historical biogeography and Late Quaternary environmental change of Lake Pamvotis, Ioannina (north-western Greece): Evidence from ostracods. Journal of Biogeogra-phy 28: 745–756.

Gaglione L (2007) Indagini archeobotaniche dei livelli neolitici di Grotta delle Mura (Monopoli, Bari). Tesi di laurea, Università degli Studi suor Orsola Benincasa.

Giunta S, Negri A, Morigi C et al. (2003) Coccolithophorid ecostratigraphy and multi-proxy paleoceanographic reconstruction in the Southern Adri-atic Sea during the last deglacial time (Core AD91–17). Palaeogeography, Palaeoclimatology, Palaeoecology 190: 39–59.

Gladstone RM, Ross I, Valdes PJ et al. (2005) Mid-Holocene NAO: A PMIP2 model intercomparison. Geophysical Research Letters 32: L16707.

González-Sampériz P, Utrilla P, Mazo C et al. (2009) Patterns of human occu-pation during the early Holocene in the Central Ebro Basin (NE Spain) in response to the 8.2 ka climatic event. Quaternary Research 71: 121–132.

Gorgoglione MA (2002) Il territorio di Taranto. In: Fugazzola Delpino MA, Pessina A and Tinè V (eds) Le ceramiche impresse nel Neolitico Antico. Italia e Mediterraneo. Roma: Museo Nazionale Preistorico Etnografico ‘L. Pigorini’, pp. 775–781.

Gorgoglione M (2004) Il processo di Neolitizzazione nel Golfo di Taranto: Alcuni dati. In: Ingravallo E (ed.) Il fare e il suo senso. Dai cacciatori paleo-mesolitici agli agricoltori neolitici. Galatina: Congedo Editore, pp. 69–85.

Gravina A (1988) Caratteri del Neolitico medio-finale nella Daunia centro-settentrionale. In: Atti 6° Convegno sulla Nazionale sulla Preistoria – Pro-tostoria e Storia della Daunia. San Severo, 14–16 Dicembre 1984, pp. 22–41.

Gravina A (2005) Il Popolamento neolitico nella Daunia costiera, garganica e nella Valle del Fortore. Rivista di Scienze Preistoriche LV, pp. 489–500.

Gravina A, Mastronuzzi G and Sansò P (2005) Historical and prehistorical evo-lution of the Fortore River coastal plain and the Lesina Lake area (south-ern Italy). Méditerranée 104: 107–117.

Grifoni Cremonesi R (2002) I culti e i rituali funerari. In: Fugazzola Delpino MA, Pessina A and Tinè V (eds) Le ceramiche impresse nel Neolitico Antico. Italia e Mediterraneo. Roma: Museo Nazionale Preistorico Etnografico ‘L. Pigorini’, pp. 209–219.

Guilaine J and Cremonesi G (2003) Torre Sabea: Un etablissement du Neoli-tique Ancien en Salento. Rome: École Française de Rome.

Jones GDB (1987a) Apulia. Volume I: Neolithic Settlement in the Tavoliere. Report of the Research Committee of the Society of Antiquaries of Lon-don, 44 pp.

Jones GDB (1987b) Botanical remains, in Lagnano da Piede I – An early neo-lithic village in the Tavoliere. Origini 13(1984–1987): 175–180.

Köppen W (1923) Die Klimate der Erde; Gundriss der klimakunde. Berlin, Leipzig: Walter de Gruyter & Co.

Laviano R and Muntoni IM (2009) Produzione e circolazione della ceramica ‘Serra d’Alto’ nel V millennio a.C. in Italia sud-orientale. In: Gualtieri S, Fabbri B and Bandini G (eds) Le classi ceramiche. Situazione degli studi. Atti della 10a Giornata di Archeometria della Ceramica. Roma, 2006. Bari: Edi Puglia, pp. 57–71.

Lionello P, Malanotte-Rizzoli P and Boscolo R (eds) 2005 Mediterranean Climate Variability. Developments in Earth & Environmental Sciences 4, Amsterdam: Elsevier, 439 pp.

Macchia F, Cavallaro V, Forte L et al. (2000) Vegetazione e clima della Puglia. Cahiers Options Méditerranéennes 53 : 29–49.

Magny M, de Beaulieu JL, Drescher-Schneider R et al. (2007) Holocene cli-mate changes in central Mediterranean as recorded by lake-level fluctua-tions at Lake Accesa (Tuscany, Italy). Quaternary Science Reviews 26: 1736–1758.

Marinval P (2003a) Les paleosemences carbonisèes de Torre Sabea: Metodolo-gie et resultats. In: Guilaine J and Cremonesi G (eds) Torre Sabea: Un etablissement du Neolitique Ancien en Salento. Rome: École Française de Rome, pp. 228–233.

Marinval P (2003b) Torre Sabea et la premiére agricolture en Méditerranèe centrale. In: Guilaine J and Cremonesi G (eds) Torre Sabea: Un etablisse-ment du Neolitique Ancien en Salento. Rome: École Française de Rome, pp. 316–324.

Marsico A, Caldara M, Capolongo D et al. (2007) Climatic characteristics of middle-southern Apulia (Southern Italy). Journal of Maps v2007: 342–348.

Mayewski P, Meeker LD, Twickler MS et al. (1997) Major features and forcing of high latitude northern hemisphere circulation using a 110000-year-long glaciochemical series. Journal of Geophysical Research 102: 26,345–26,366.

Mayewski PA, Rohling EE, Stager JC et al. (2004) Holocene climate variabil-ity. Quaternary Research 62: 243–255.

Mercuri AM, Bandini Mazzanti M, Torri P et al. (2012) A marine/terrestrial integration for mid-late Holocene vegetation history and the development of the cultural landscape in the Po valley as a result of human impact and climate change. Vegetation History and Archaeobotany 21: 353–372.

Merico M (2005) Gli intonaci come indicatori tecnologici e paleovegetazion-ali nei contesti preistorici dell’Italia meridionali: Il caso studio di Serra Cicora. Tesi di laurea, Università del Salento.

Miller Rosen A (2007) Civilizing Climate: Social Responses to Climate Change in the Ancient Near East. Plymouth: Altamira Press.

Muntoni IM (2003) Modellare l’argilla. Vasai del Neolitico antico e medio nelle Murge pugliesi. Firenze: Istituto Italiano di Preistoria e Protostoria.

Muntoni IM (2012) Circulation of raw materials, final products or ideas in the Neolithic communities of southern Italy: The contribution of archaeo-metric analyses to the study of pottery, flint and obsidian. Rubricatum 5: 403–411.

Neboit R (1975) Plateaux et collines de Lucanie orientale et des Pouilles. The-sis Sc. Univ. Lille, Paris: Librerie Champion, pp. 715.

Nisbet R (1982) Le analisi archeobotaniche del villaggio neolitico della villa Comunale (Foggia). Origini 11(1977–1982): 175–182.

Oldfield F, Asioli A, Accorsi CA et al. (2003) A high resolution late Holocene palaeoenvironmental record from the central Adriatic Sea. Quaternary Science Reviews 22: 319–342.

Orlove B (2005) Human adaptation to climate change: A review of three his-torical cases and some general perspectives. Environmental Science & Policy 8: 589–600.

Percival J (1943) Wheat in Great Britain. London: Duckworth.Pessina A and Tinè V (eds) (2008) Archeologia del Neolitico. L’Italia tra il VI

e il IV millennio a.C. Rome: Carocci Editore.Primavera M (2008) I dati archeobotanici della sepoltura neolitica di Carpig-

nano (LE): Alcune considerazioni. In: D’Andria F, De Grossi-Mazzorin J and Fiorentino G (eds) Uomini, Piante e Animali nella dimensione del Sacro. Bari: Edi Puglia, pp. 193–200.

Primavera M and Fiorentino G (2011) Archaeobotany as an in-site/off-site tool for paleoenvironmental research at Pulo di Molfetta (Puglia, South-Eastern Italy). In: Turbanti-Memmi I (ed.) Proceeding of the 37th Inter-national Symposium on Archaeometry. 12–16 May 2008, Siena. Berlin, Heidelberg: Springer-Verlag, pp. 421–426.

Radina F (ed.) (2002) La Preistoria della Puglia. Paesaggi, uomini e tradizioni di 8.000 anni fa. Bari: Mario Adda Editore.

Radina F (2006) Rapporti e scambi tra le più antiche comunità neolitiche in Puglia sulla base dell’indicatore ceramico. In: Atti della XXXIX Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria ‘Materie prime e scambi nella preistoria italiana’. Firenze, pp. 25–27 Novembre 2004. Firenze: Istituto Italiano di Preistoria e Protostoria, pp. 1049–1059.

Radina F (2013) Alcune osservazioni sull’Eneolitico in Puglia sulla base delle evidenze archeologiche nell’area murgiana adriatica. In: Dally O, Grae-pler D and Lippolis E (eds) Archeologia e Cultura dalla Preistoria al Tar-doantico. Berlino: Deutsches Archaologisches Institut, in press.

Redman C (2005) Resilience theory in archaeology. American Anthropologist 107(1): 70–77.

Redman CL and Kinzig AP (2003) Resilience of past landscapes: Resilience theory, society, and the longue durée. Conservation Ecology 7(1): 14.

Reimer PJ, Baillie MGL, Bard E et al. (2004) Intcal04 terrestrial radiocarbon age calibration, 0–26 cal kyr BP. Radiocarbon 46(3): 1029–1058.

Roberts N, Brayshaw D, Kuzucuoglu C et al. (2011) The mid-Holocene cli-matic transition in the Mediterranean: Causes and consequences. The Holocene 21(1): 147–162.

Page 20

1316 The Holocene 23(9)

Roberts N, Jones MD, Benkaddour A et al. (2008) Stable isotope records of Late Quaternary climate and hydrology from Mediterranean lakes: The ISOMED synthesis. Quaternary Science Reviews 27: 2426–2441.

Rohling EJ, Mayewski PA, Abu-Zied RH et al. (2002) Holocene atmosphere–ocean interactions: Records from Greenland and the Aegean Sea. Climate Dynamics 18: 587–593.

Sadori L, Zanchetta G and Giardini M (2008) Last Glacial to Holocene pal-aeoenvironmental evolution at Lago di Pergusa (Sicily, Southern Italy) as inferred by pollen, microcharcoal, and stable isotopes. Quaternary International 181: 4–14.

Sangiorgi F, Capotondi L, Combourieu Nebout N et al. (2003) Holocene seasonal sea-surface temperature variations in the southern Adriatic Sea inferred from a multiproxy approach. Journal of Quaternary Science 18(8): 723–732.

Sargent A (1983) Neolithic plant remains from the Tavoliere of Apulia. In: Cassano SM and Manfredini A (eds) Studi sul neolitico del Tavoliere della Puglia. Indagine territoriale di un’area campione. B.A.R. International Series 160, Oxford, pp. 250–252.

Sargent A (1987) Relazione sui resti paleobotanici di Coppa Nevigata. In: Atti della XXVI Riunione Scientifica dell’Istituto Italiano di Preistoria e Pro-tostoria. Firenze, 7–10 Novembre 1985, pp. 761–764.

Siani G, Paterne M and Colin C (2010) Late glacial to Holocene planktic fora-minifera bioevents and climatic record in the South Adriatic Sea. Journal of Quaternary Science 25(5): 808–821.

Tarantini M and Galiberti A (eds) (2011) Le miniere di selce del Gargano (VI-III millennio a.C.). Alle origini della storia mineraria europea. Rassegna di Archeologia 24A (2009–2011), Firenze: All’Insegna del Giglio Editore.

Tiberi I (ed.) (2007) Sant’Anna (Oria - Br). Un sito specializzato del VI millen-nio a.C. Galatina: Congedo Editore.

Tinè S (1983) Passo di Corvo e la civiltà neolitica del Tavoliere. Genova: Sagep.Trincardi F, Cattaneo A, Asioli A et al. (1996) Stratigraphy of the late-Qua-

ternary deposits in the central Adriatic basin and the record of short-term climatic events. In: Guilizzoni P and Oldfield F (eds) Palaeoenviron-mental Analysis of Italian Crater Lake and Adriatic Sediments. Memorie dell’Istituto italiano di Idrobiologia 55: 39–70.

Vannière B, Colombaroli D, Chapron E et al. (2008) Climate versus human-driven fire regimes in Mediterranean landscapes: The Holocene record of Lago dell’Accesa (Tuscany, Italy). Quaternary Science Reviews 27: 1181–1196.

von Bothmer R, Jacobsen N, Baden C et al. (1995) An Ecogeographical Study of the Genus Hordeum. 2nd edition. Systematic and Ecogeographical Studies on Crop Genepools 7. Rome: International Plant Genetic Resources Institute.

Weninger B, Clare L, Rohling EJ et al. (2009) The impact of rapid climate change on prehistoric societies during the Holocene in the eastern Medi-terranean. Documenta Praehistorica 36: 7–59.

Xoplaki E, Gonzàlez-Rouco JF, Gyalistras D et al. (2003) Interannual summer air temperature variability over Greece and its connection to the large-scale atmospheric circulation and Mediterranean SSTs 1950–1999. Cli-mate Dynamics 20: 537–554.