Imp.Cardara& GEOMORPHOLOGICAL CHANGES DUE TO HUMAN ACTIONS AT COPPA NEVIGATA (TAVOLIERE DI PUGLIA, SOUTHERN ITALY) RECONSTRUCTED THROUGH CORE ANALYSES Massimo Caldara1, Ilena Caroli2 & Oronzo Simone1 1Dipartimento di Geologia e Geofisica, Sezione di Geografia Fisica e Geomorfologia – Università degli Studi di Bari, Campus Universitario, Via Orabona 4, 70125 Bari, Italia; 2Dottorato di Ricerca in Geomorfologia e Dinamica Ambientale – Università degli Studi di Bari Correspondence should be addressed to MASSIMO CALDARA: - caldara@geo.uniba.it ABSTRACT: M. Caldara et al., Geomorphological changes due to human actions at Coppa Nevigata (Tavoliere di Puglia, southern Italy) reconstructed through core analyses. IT ISSN 0394-3356, 2004 The area surrounding the Coppa Nevigata settlement is one of the best places for the reconstruction of the relationship between human activities and the environment since the first Tavoliere di Puglia population phases. The settlement is the best studied in the Tavoliere plain, it is situated between the foot of the Gargano headland and the inner shore of the ancient Salpi lagoon, in a very sensi- tive area from the environmental point of view. Previous studies highlighted the environment evolution and its chronology. The close relationship between the wet area facing the settlement and the population pattern in some cases was documented. The preliminary study of six new cores and considerations made on three previously drilled successions are the base of this research. The studied successions were grouped in three transects and described in function of their distance from the settlement. These are the A (distal), B (in the middle) and the C (proximal) transects. There were recognized eight different phases attributable to two different evolutionary patterns. The environments set under or modified by intentional anthropic action, such as the B (terrestrial phase I), D (terrestrial phase II) and H (historical reclamation) phases, have to be ascribed to the first one. The almost "natural" phases belong to the second one, even if the accumulated material is often related to human actions. These are the phases A (natural phase I), C (natural phase II), E (natural phase III), F (arid terrestrial phase) and G (natural phase IV). The B and D anthropic phases have been recognized beneath the ground surface up to about 90 m far from the nearest exploration ditches dug by archaeologists on the side of the Coppa Nevigata knoll. We were not able to put a boundary between these two phases within the C transect cores, but they are well defined in the A transect. The anthropic phase B corresponds to an urbanization phase that seems to have been started from the most inter- nal settlement areas through the artificial compaction/leveling of marshy/lagoonal deposits, whereas the presence of hearths in distal areas could be attributed to activities carried out outside the defensive wall system. The anthropogenic D phase shows, in the area next to the hillock side, the presence of simple inhabitation structures that lasted for a relatively long time, as testified by numerous tramped and/or leveled surfaces (simple or superimposed) associated to concotto levels and hearths. On the other hand in distal areas the D phase seems to be characterized by an agricultural/pastoral land use. The H phase sediments have locally accumulated under nearly natural conditions, even if they are directly related to reclamation works made during the last two centuries. In conclu- sion, through time man modified and shaped the environment around the Coppa Nevigata knoll, in particular during the Middle Bronze age (B phase), the Iron age (D phase) and during the last two centuries (H phase). Sometimes the sedimentation processes under natural conditions were also strongly influenced by man, as suggested by the presence of the organic deposits accumulated during the A phase or by the C phase sediments. RIASSUNTO: M. Caldara et al., Variazioni dell’ambiente geomorfologico indotte dall’Uomo desunte dallo studio di carotaggi a Coppa Nevigata (Tavoliere di Puglia). IT ISSN 0394-3356, 2004 L’area a ridosso dell’insediamento preistorico di Coppa Nevigata è quella che meglio si presta per la ricostruzione dei rapporti intercor- si tra l’Uomo e l’ambiente a partire dalle prime fasi di popolamento del Tavoliere di Puglia. Difatti, questo insediamento è certo quello più studiato del Tavoliere ed inoltre era collocato ai piedi del Promontorio del Gargano sul margine interno dell’antica laguna di Salpi, in un’area ad alta sensibilità ambientale, in quanto soggetta alle rapide modificazioni dovute alla combinazione di processi diversi (fluviale, marino, eolico ecc.) e all'opera dell'Uomo. Studi precedenti degli Autori hanno delineato l’evoluzione degli ambienti e la loro colloca- zione cronologica. Inoltre, in alcuni casi hanno messo in evidenza la stretta correlazione fra l’area umida antistante l’abitato di Coppa Nevigata e le modalità di vita e di sviluppo dell’insediamento. Lo studio preliminare di sei nuove perforazioni, realizzate ad hoc a ridos- so della collina di Coppa Nevigata, unito ad altre tre fatte precedentemente, è alla base del presente lavoro. Le successioni sono state riunite in tre gruppi e descritte in funzione della loro distanza dall’insediamento (transetto A, distale; transetto B, intermedio e transetto C, a ridosso della collina di Coppa Nevigata). Sono state riconosciute otto diverse fasi riconducibili grosso modo a due differenti modelli evolutivi. Al primo appartengono gli ambienti originatisi o rimodellati intenzionalmente dall’Uomo: fasi B (fase continentale I), D (fase continentale II), e H (colmate storiche). Al secondo appartengono gli ambienti dove il meccanismo di accumulo dei sedimenti è naturale, anche se molte volte i materiali accumulatisi sono riconducibili alle attività dell’Uomo: fasi A (fase naturale I), C (fase naturale II), E (fase naturale III), F (fase continentale arida) e G (fase naturale IV). Le fasi antropiche B e D che non si riescono a differenziare nelle successioni del gruppo C, quelle a ridosso dell’insediamento, e che sono ben separate nelle successioni del gruppo A, si ritrova- no fino a circa 90 metri dai più vicini saggi archeologici. La fase antropica B corrisponde ad una fase di urbanizzazione che sembra procedere a partire dalle aree più prossime all’insediamento compattando i depositi palustri e/o lagunari con battuti, mentre nelle aree più lontane l’attività, limitata a soli focolari, sembra quella svolta all’esterno delle mura. La fase antropica D mostra, nell’area a ridosso della collina di Coppa Nevigata, la presenza, più o meno continua nel tempo, di strutture abitative semplici (abbondanza di battuti sin- goli e/o sovrapposti uniti a livelli di concotto e a focolari), mentre per le aree distali mostra una utilizzazione di tipo agricolo-pastorale. La fase antropica H è riconducibile alle colmate storiche degli ultimi secoli. Si tratta di un deposito realizzato con intenzionalità dall’Uomo anche se i meccanismi di sedimentazione sono praticamente naturali. In definitiva nell’area di Coppa Nevigata l’Uomo ha rimodellato e modificato il paesaggio più volte, durante il Bronzo medio (fase B), l’età del Ferro (fase D) e negli ultimi due secoli (fase H). Comunque è stato fortemente attivo condizionando la sedimentazione anche nella fase A con depositi organici, e nella fase C. Keywords: buried morphology, geoarchaeology, Tavoliere di Puglia, Holocene. Parole chiave: morfologie sepolte, geoarcheologia, Tavoliere di Puglia, Olocene. Il Quaternario Italian Journal of Quaternary Sciences 17(2/2), 2004, 495-508 496 M. Caldara, I. Caroli & O. Simone 1. INTRODUCTION The Coppa Nevigata settlement is one of the best known archaeological sites in the Apulian region. It is located at 41°33’26”N, 15°50’00”E, some five kilome- tres from the modern coast, between the Gargano Promontory and the Tavoliere Plain. The site was situa- ted on the inner shore of the Holocene Salpi lagoon (Boenzi et al., 2001), a coastal waterbody extended from south of Manfredonia to the Ofanto river delta and now totally disappeared because of both natural infilling and land reclamation works (Boenzi et al., in press; Caldara et al., 2002b). The archaeological site is known since the first years of the XX century (Mosso, 1909). Nevertheless, it was subjected to systematic archaeological explora- tions only from the '50s of last century (Puglisi, 1955; 1975; 1982). The investigations started again in 1983 with extensive excavations still in progress (Cazzella & Moscoloni, 1988; 1990; Cazzella et al., 2001). The archaeological research evidenced a complex strati- graphy spanning from the Early Neolithic, throughout the Bronze age and into the Iron age (Cazzella, 1996; Cazzella & Moscoloni, 1999). During the various inhabi- tation phases the mode of occupation changed (Caldara et al., 2003a). Since the '90s the Authors have been investiga- ting the coastal plain of the Tavoliere, but the choice to focalise the multidisciplinary efforts in a sample area occurred only later. The Coppa Nevigata settlement offered good research possibilities, since the good knowledge acquired through decades of archaeological study as well as its location in a high sensitive environ- mental frame. The best available information suggests that the population around the lagoonal area increased and decreased several times. One explanation of this is past changes in the regional environmental conditions (Boenzi et al., 2001; Caldara et al., 2002b). In 1997 five cores were drilled (CN1 to CN5) at various distances from the settlement (Fig. 1). The Fig. 1 - Location of the studied areas. CNn = coring site. Ubicazione dell’area. CNn = posizione dei sondaggi. study of two of the nearest cores to the site (CN2 and CN5) is by now complete (Caldara et al. 1999; 2001; 2003b; in press c), while the most distant ones (CN3 and CN4) are still under examination (Simone 2003; Caldara et al., in press c). The first results obtained were the reconstruction of the environment evolution close to the site (Caldara et al., 1999; 2001) and the understanding of chronology of the events (Caldara et al., 2003a). The idea at the base of our research were that there was a close relationship between the wet area facing the Coppa Nevigata knoll and the way in which the settlement developed (Caldara et al., 2002a). This connection has been documented in various cases. As an example, at the site were found ceramic artifacts, dating back between the Proto-Apennine and the Sub- Apennine periods, containing tephra (Levi et al. 1999). The use of pyroclastic material as a tempering ingre- dient to make ceramics was possible by the presence of this kind of sediments in the lagoon bottom (Caldara et al., 2001). An other example is the use of the cockle Cerastoderma glaucum as a food resource, in particular during the Neolithic and the Bronze age (Late Sub- Apennine), as documented by Minniti (1999). As the study progressed we realized that, espe- cially during the Middle Bronze age, man influenced the local environment evolution (Caldara & Simone 2003; in press), as attested by the presence of deposits directly related to anthropogenic activities only within the cores nearest to the settlement. The aim of this research is to asses how strong was the impact of human actions on the surrounding environment. One of the possible ways to reach this goal was the detailed reconstruction of the geometries of buried sedimentary bodies (both of natural and anth- ropic origin). To verify our hypotheses, six new cores were drilled in 2002 (CN11 to CN16). The coring loca- tions were accurately decided on a detailed map (Fig. 1), in order to get a good distribution of samples. Thus, at the base of this paper are the preliminary results of the study of the six new cores plus the CN1, compared and integrated with data obtained by the analysis of CN2 and CN5 successions. 2. GEOLOGICAL AND MORPHOLOGICAL SETTING The Coppa Nevigata knoll is situated along the south-east transition area (Fig. 2) between the smooth limestone surface of the southern Gargano terraces area, called "Pedegargano" by Caldara & Palmentola (1991), and the alluvial coastal plain of the Tavoliere di Puglia (Southern Italy). The Gargano is a calcareous massif, part of the Mesozoic Apulia Platform. The outcropping rocks in the southern area of the headland are the oolithic, pseudo- oolithic, detritic and micritic limestone of the Portlandian, "Calcari oolitici di Coppa Guardiola" Formation (Luperto Sinni & Masse, 1986). In particular, the lower terrace at the foot of the headland, dips gen- tly eastward starting from elevations slightly above 100 meters a.s.l.. The sub-horizontal surface, in correspon- dence of the terminal section of the Candelaro stream, has been divided in several minor block by NW-SE faults. 497Geomorphological changes ... The Tavoliere plain subsoil is made of a thick suc- cession of Quaternary terrigenous sediments overlaying the Bradanic Cycle deposits, in particular the "Argille subappennine" Formation. Starting from the end of the Lower Pleistocene, due to some stands of the regional uplift combined with glacio-eustatic movements, the Tavoliere landscape was shaped in a number of marine terraces gently sloping down towards the sea. Some of them are now covered by broad alluvial deposits (Caldara & Pennetta, 1993a). During the last glacial phase, because of the lower sea level, the Tavoliere coastland was deeply incised by rivers flowing from the Apennine chain (Caldara & Pennetta, 1996; Caldara et al., 2002b). At the end of the last glacial phase, the eastern margin of the Tavoliere took the characters of a sub- mergence alluvial plain coast. As the sea level raised, several lagoons set on the margin of the submerging plain. Time after time, the eastern area of the Tavoliere was then characterized by a variety of transitional envi- ronments, such as dune belts, lagoons, swamps and a broad alluvial coastal plain. In correspondence of the breaks of slope between the Gargano and the Pedegargano areas and between the Pedegargano and the Tavoliere plain, in particular at the mouths of the streams flowing down the headland, a number of alluvial fans developed during low sea level periods. These are broad sedimen- tary bodies made of limestone breccias whose rounded elements are well cemented in a calcareous matrix (Boenzi & Caldara, 1999). During the middle-late Holocene a broad coastal lake set up, the Salpi Lagoon (Caldara et al., 2002b). Through its history, this lagoon underwent many chan- ges in its shape and extent. Nearly at the end of its exi- stence, the coastal waterbody became a marshy and unhealthy land. Recently, in particular during the last two centuries, the coastal plain was subjected to recla- mation. In fact, several streams have been diverted in order to facilitate the infilling of coastal depressions, such as the Versentino and Salso lakes. As a result, the alluvial sediments covered and concealed all the featu- res that characterized the coastal plain, including the deposits of lagoon-sabkha environments (studied by Caldara & Pennetta, 1993b and Boenzi et al., 2001), the dune belt that separated it from the sea (Schmiedt, 1973) and the historical lake deposits. Nowadays an enclave of the ancient lagoon exists, this is the Palude Frattarolo (Caldara et al., 1994), a broad coastal swamp situated south of Manfredonia, between the Coppa Nevigata knoll and the coastline. 3. RESULTS ACQUIRED SINCE FORMER WORKS So far in the Coppa Nevigata area four cores were studied (Caldara et al., 1999; 2001; 2003b; in press c; Simone 2003). Two of those, the CN2 and CN5, were drilled in vicinity of the archaeological site, next to the lagoon shore. The other two bore holes, CN4 and CN3, were respectively made at 0.4 km and 0.8 km far from the settlement, towards the middle of the basin. The drillings at the foot of the settlement went across to sediments accumulated between before 3600 yr BP ("Avellino eruption" tephra) and recent times. A stop in 498 M. Caldara, I. Caroli & O. Simone F ig 2 - S c h e m a ti c g e o m o rp h o lo g ic a l m a p . K e y to s ym b o ls : 1 = l im e st o n e d e p o si ts ( M e so zo ic ), P lio -P le is to c e n e c a lc a re n it e a n d Q u a te rn a ry s lo p e d e p o si ts ; 2 = T a vo lie re d i P u g lia t e rr a c e d d e p o si ts ( m a ri n e a n d f lu vi a l); 3 = p re se n t d a y sa n d ; 4 = H o lo c e n e a llu vi a l d e p o si ts ; 5 = h is to ri c a l r e c la m a ti o n s e d im e n ts ; 6 = a llu vi a l f a n ; 7 = b u ri e d d u n e b e lt s; 8 = d o lin e ; 9 = e sc a rp m e n t d u e to m a ri n e a b ra si o n ; 1 0 = e sc a rp m e n t d u e t o f lu vi a l e ro si o n ; 1 1 = " V " sh a p e d f lu vi a l v a lle y; 1 2 = f la t b o tt o m e d f lu vi a l v a lle y; 1 3 = f o ss il d ra in a g e , p a la e o -r iv e rb e d ; 1 4 = d ra in a g e d it c h ( re a liz e d to r e c la im a n a re a b y d ra in a g e ); 1 5 = r o a d ; 1 6 = r a ilr o a d . C ar ta g e o m o rf o lo g ic a sc h e m at ic a. L e g e n d a: 1 = d e p o si ti c ar b o n at ic i m e so zo ic i, c al c ar e n it i p lio -p le is to c e n ic h e e d e p o si ti d i ve rs an te q u at e rn ar i; 2 = d e p o si ti t e rr az za ti m ar in i e f lu vi al i d e l T av o lie re d i P u g lia ; 3 = sa b b ie a tt u al i; 4 = a llu vi o n i o lo c e n ic h e ; 5 = c o lm at e s to ri c h e ; 6 = c o n o id e a llu vi o n al e ; 7 = c o rd o n i d u n ar i se p o lt i; 8 = d o lin a; 9 = s c ar p at a d i ab ra si o n e m ar in a; 1 0 = s c ar - p at a d i e ro si o n e f lu vi al e ; 1 1 = in c is io n e a V ; 1 2 = in c is io n e c o n f o n d o p ia tt o ; 1 3 = d re n ag g io f o ss ile , p al e o al ve o ; 1 4 = c an al e d i b o n ifi c a; 1 5 = s tr ad a; 1 6 = f e rr o vi a. 499 deposition, highlighted by an erosive surface, occurred between the conventional 14C ages 2870±40 BP and 370±50 BP. In the middle of the basin (CN3 succes- sion), at the bottom of the core, sediments have an age of 5580±40 yr BP. A hiatus is recorded between the conventional ages 3110±50 BP and 140±40 yr BP (Caldara & Simone, in press). Materials found within the successions close to the settlement suggest the existence of a stable com- munity in the nearby, particularly during the Bronze Age (Caldara et al., 1999; 2001; 2003b). The CN2 and CN5 series show upcore alternating terrestrial and brackish horizons. One could interpret this alternation as the expression of four negative sea level fluctuations, marked by the deposition of terrestrial/marshy sedi- ments (Caldara & Simone, in press). On the other hand, no evidence of sea level fluctuations were found in the middle of the basin (Simone, 2003; Caldara et al., in press c). According to Caldara & Simone (in press), three of the four fluctuations recognized in vicinity of Coppa Nevigata were most likely due to colonization of environments situated at the margin of the wet area, that is to say that the terrestrial/marshy environments should be considered as the expression of an aggra- ding coastline (whatever the cause) rather than a sea level fall. The analysis of the CN2 and CN5 cores made possible the reconstruction of palaeoenvironmental changes occurred in the nearby of the Coppa Nevigata settlement at least from 3600 years BP. On the basis of the fossiliferous assemblages, a descriptive name was given to each reconstructed environment (Fig. 3). These, listed starting from the most ancient, are: 3.1. Hydrobiidae spp. and Cerastoderma Lagoon Mainly represented by clayey sediments, domina- ted by lagoon species, a few marine taxa occur in the lower section (Caldara et al., 1999, 2001). During this first lagoon phase, two episodes con- ditioned the accumulation processes. The first event was the accumulation of a tephra layer containing materials from at least two different eruptions (Caldara et al., 2001; 2003a). One of these has been identified as the "Avellino Eruption" of the Vesuvius, dated at about 3500 ÷ 3600 BP (Terrasi et al., 1999; Albore Livadie et al., 1998; Rolandi et al., 1993; Andronico et al., 1995). The rest of the tephra granules could be referred to one of the eruptions occurred during the last phase (4100 ÷ 3700 yr BP) of the Phlegrean Fields activity (Caldara et al., 2001; 2003a). The second episode characterized the upper part of the Hydrobiidae spp. and Cerastoderma Lagoon. At this level accumulated a horizon characterized by a high organic matter content, whose principal consti- tuents are related to human activities, such as charred seeds and fruits, cereal remains, domestic animal bones, remains of insects living in granaries or breeding in decomposing meat and pottery fragments. Given the excellent preservation state, we can say that these materials underwent a short transportation before the definitive deposition (Caldara et al., 1999; 2001; 2003a). The accumulation of lagoon sediments continued until shortly after the radiocarbon date 3090±40 BP (CN5 core). 3.2. Terrestrial Phase I The successions continue upcore with sediments accumulated under sub-aerial conditions. The first cen- timetres are rich in organic matter and have a similar aspect in comparison to the ones above defined. The rest of the sub-aerial horizon is a soil made of silty- clayey sediment. The findings of pottery fragments, sto- nes and concotto suggest that the settlement expan- ded towards the plain. According to the chronological data available, these structures were built after the radiometric date 3000±80 years BP and before 2870±40 BP (Caldara et al., 1999; 2001; 2003a). 3.3. Cerastoderma Lagoon The terrestrial deposits are overlaid by grey clayey silts. The fauna, characterized by a low diversity, is dominated by Cerastoderma glaucum (Caldara et al., 2001; 2003a). 3.4. Salt Marsh I Gradually, the lagoon environment evolved to a brackish marshy area. The lower part of this interval is characterized by clayey silts with thin arenaceous beds, whereas in the upper part there are calcareous lumps and thin black levels made of organic matter. This envi- ronment was described as a salt marsh (Caldara et al., 2001; 2003a). 3.5. Hydrobiidae spp. and Abra segmentum Lagoon The sediments that constitute this third lagoon episode are clayey silts with pebbles in its lower part, silts with thin black organic matter levels and calca- reous lumps in the upper part. The macrofauna is represented mostly by lagoon taxa. Progressively the environment underwent a relative lowering of the water level. 3.6. Terrestrial Phase II This is a new terrestrial phase whose sediments are clayey silts dark-grey in color, with several thin organic matter levels and small calcareous lumps. The fauna is characterized by a mixed assemblage made of broken terrestrial mollusc shells and young individuals of lagoon species. 3.7. Salt Marsh II This environment is represented by dark gray clayey silts with alternating calcareous levels. The mol- lusc fauna is poor and dominated by Hydrobiidae, few young individuals of Cerastoderma glaucum and Abra segmentum also occur. 3.8. Arid or semiarid Terrestrial Phase It is not clear yet when the salt marsh sediments stop to accumulate. In fact, these are cut by an erosive surface. A sub-aerial period, under arid or sub arid con- ditions, is testified by the formation of calcareous crust levels and lumps. 3.9. Bithynia leachi and Ovatella myosotis wetland Shortly before the radiocarbon date 370±50 BP, the area at the foot of the Coppa Nevigata hillock was submerged by fresh water. The sediments accumulated in this interval are sandy silts rich of Ovatella myosotis and Bithynia leachi. Geomorphological changes ... 500 M. Caldara, I. Caroli & O. Simone F ig . 3 - S tr a ti g ra p h ic c o rr e la ti o n s a m o n g t h e s tu d ie d c o re s. T h e A , B , a n d C tr a n se c ts a re p a ra lle l, w h ile t h e D a lig n m e n t is p e rp e n d ic u la r to t h e f ir st o n e s. B e si d e t h e t ra n se c ts , th e A - H le t- te rs r e fe r to t h e r e c o g n iz e d p h a se s, w h ile t h e t e rm in o lo g y u se d is t h e s a m e a s in li te ra tu re . C o rr e la zi o n i st ra ti g ra fic h e . S o n o d is e g n at i i tr an se tt i A ,B ,C p ar al le li fr a lo ro e D p e rp e n d ic o la re . S o n o r ip o rt at i a fia n c o d e i tr an se tt i le l e tt e re A -H ri fe ri te a lle f as i d a n c h e l a te rm in o lo g ia d e g li am b ie n ti c o m e d e fin it a in le tt e ra tu ra . 3.10. Terrestrial Phase III The very upper part of the CN2 and CN5 cores is made of silty-clayey soil, densely laminated and scar- cely pedogenized. The faunistic content, mostly terre- strial shell fragments, is scarce. 4. DESCRIPTION OF THE STUDIED SUCCESSIONS For convenience the successions were grouped in three transects and described in function of their distance from the settlement. In particular, in the A group there are three successions drilled at the foot of the Coppa Nevigata knoll; the C group consists of four series sampled close to the settlement; in the B tran- sect there are only two cores drilled in the middle between the A and C groups (Fig. 1). 4.1. A transect - CN13, CN2 and CN5 cores These three cores show a great lithological varia- bility (Fig. 3), due to the thick alternation of deposits accumulated in natural conditions or under the more or less direct anthropic influence. The environments for- merly reconstructed for the CN2 and CN5 cores show a good mutual correspondence (Caldara et al., 1999; 2001). On the other hand, only the lower and higher sections of the CN13 core could be correlated in a sati- sfactory way with the homologue environments recon- structed for the CN2 and CN5 cores. At the bottom, the CN2 and CN5 cores show lagoon deposits, mostly clayey (Hydrobiidae and Cerastoderma Lagoon; Caldara et al., 1999; 2001; 2003a; in press a). Between 4.05 m and 3.60 m below m.s.l. (CN2 core) the accumulation of tephra deposits occurred (Caldara et al., 2001; 2003a). In both the CN2 and CN5 cores the upper section of the Hydrobiidae and Cerastoderma Lagoon and the lower part of the overlaying terrestrial sediments are characterized by the accumulation of an organic sedi- ment, dark in colour, mainly made of heterogeneous material originated after the activities carried out at the settlement during the Bronze age (Caldara et al., 2001; 2003a; in press a). In the CN13 core the accumulation of this kind of sediments started under brackish condi- tions and continued in a paludal environment domina- ted by fresh water molluscs. The series continue upcore with terrestrial depo- sits (Terrestrial phase I; Caldara et al., 2003a). These sediment are clayey to sandy, grey or reddish-grey, with small calcareous lumps, calcareous crust frag- ments, pottery fragments and broken domestic animal bones (ox, pig, sheep, dog etc.). In the CN13 core, between 1.49 m and 1.29 m below m.s.l., there is a thin black horizon made of tiny charcoals. Another charcoal level, about one centimetre thick, lies under a leveled surface1. Concotto fragments, limestone and sandstone clasts have been found at several levels within the CN2 and CN5 successions. These findings were attributed to the remnants of a wall structure (Caldara et al., 1999; 2001; 2003a; in press a). In the following section, up to the present day sea level, the sedimentation around the three bore holes occurred under different conditions. In particular, the sediment in the CN2 and CN5 successions is a lagoon clayey silt (Cerastoderma Lagoon, CN2 core, Caldara et al., 2001; Hydrobiidae Lagoon, CN5 core, Caldara et al., 1999). Upcore the lagoon evolved to a more superfi- cial environment in which accumulated clayey silt rich in organic matter characterized by young individuals of brackish taxa (Salt marsh I in Caldara et al., 1999; 2001; 2003a). The upper part of this section is again represen- ted by lagoon clayey silt (CN2 core). The assemblages are dominated by lagoon taxa, the other accompanying species belong to the marine domain (Hydrobiidae and Abra Lagoon in Caldara et al., 1999; 2001; 2003a). On the other hand, the corresponding section of the CN13 core is characterized by sediments virtually similar to the ones that lay directly below, i.e. dark silt with dispersed calcareous crust fragments, charcoals and pottery fragments. Between 0 m and 0.40 m circa above m.s.l. the successions recorded a new terrestrial phase (Terrestrial phase II, Caldara et al., 2003a), with the accumulation of clayey to sandy silt. Several organic levels and dispersed small calcareous lumps also occur. The assemblages are made of terrestrial mollusc fragments and young individuals of brackish taxa. Some domestic and wild animal bones also occur (Caldara et al., 1999; 2001). Burnt bone fragments and marine gastropods have been found within thin dark levels in the CN2 core. The successions continue upcore with dark grey silt accumulated in brackish-marshy environments (Salt marsh II, Caldara et al., 1999; 2001; 2003a). The fossil association is made of scarce brackish molluscs, fora- minifers and ostracods. These sediments are cut by an erosive surface, marked in the CN2 and CN5 cores by a thick accumulation of a secondary calcareous crust, originated by groundwater evaporation. On the erosive surface lays a sandy silt deposit with fresh water fauna. The reconstructed environment is a marshland characterized by a variable fresh water covering (Bithynia and Ovatella Wetland, CN2 core; Bithynia Wetland and Ovatella Wetland, CN5 core. Caldara et al., 1999; 2001; 2003a). The very upper part of these three sequences is made of soil disturbed by the intense ploughing activi- ties (Terrestrial phase III, Caldara et al., 2003a). 4.2. B transect - CN11 and CN1 cores In this group there are only two cores (CN11 and CN1, Fig. 3). Despite the care taken during coring acti- vities, there have been two metres of sample loss in the CN1 core. Thus we were able to correlate these suc- cessions only in a partial way. Further data should be collected with additional cores. The lower part of the CN1 succession, is a grey- greenish clayey sediment, somewhere sandy. There are scattered charcoals, calcareous crust fragments and domestic animal bones (sheep). Between 2.50 and 2.00 501Geomorphological changes ... 1 With "leveled surface" we mean a surface made regular by spreading of fine sediment, whatever it is (generally calcareous or clayey). The use to spread material in order to regularize the ground is well known to the archaeologists that worked on the Coppa Nevigata site. m below m.s.l. the sediment is dark, compressible and rich in vegetal detritus. In the whole this deposit is quite similar to the high organic deposit found in the CN2 and CN5 cores (Caldara et al., 2003a). From the base of the CN11 core and above the two metres gap in the CN1, the two drillings went across to fine sediments, mostly clayey, greyish to dark, hardly stratified, deposited under the direct anth- ropic influence (scattered bone fragments, charcoals, Bronze age pottery fragments). In three occasions the core sediments differ from the general pattern above described. In particular, at the bottom of the CN11 core, there is a rudite horizon made of heterogeneous materials, such as lithic frag- ments (limestone crust and siliceous pebbles), bones, pottery, terrestrial molluscs and charcoals. In the CN1 core, at circa 0.30 m above m.s.l., there is a reddish concotto horizon six centimetres thick. Finally, in the CN11 core, at 0.70 m above m.s.l., a horizon made of calcareous crust fragments suggests the presence of a probable leveled surface. The two successions continue with a brown clayey soil with small lithic fragments. The upper part of this soil is deeply disturbed by ploughing. 4.3. C transect - CN12, CN14, CN15 and CN16 cores In this group there are the successions drilled close to the settlement. The lower part of the sequence accumulated under mostly natural environments (lagoon, salt marsh and fresh marsh), the upper part is made of sediments deposited under the more or less direct anthropic influence. The CN16 core is the only one that reached the limestone substrate in this area. The transgression sur- face, cut on Cretaceous limestone, is marked by holes made by marine molluscs and sponges and covered by a thin calcareous sand with marine shell fragments (Ostrea, Mytilus and Pinna). Rapidly, the assemblage becomes typically lagoonal (Cerastoderma glaucum, Abra segmentum, Cyclope neritea and Ammonia bec- carii and Haynesina germanica). The same fauna cha- racterizes the overlaying clayey deposit. We considered natural the sediments accumula- ted at the base of the CN12 core (up to 1.22 m below m.s.l.). In particular, between the base and 1.66 below m.s.l., the sediments are silty with tiny vertical roots and characterized by lagoon taxa (Abra segmentum, Hydrobiidae, Haynesina germanica and Ammonia bec- cari). In the middle part of this section there are three volcanoclastic horizons (Fig. 4D). The first one, the thicker, has a fairly regular base, whilst the contact between the other two and the sediment below is less evident. Between the horizons with tephra some volca- nic granules occur scattered within the sediment. The faunistic assemblages indicate a lagoon environment. Between 1.40 and 1.22 m below the m.s.l. the sediment is silty-sandy. The collected fauna (mostly Ovatella myosotis and Truncatella subcylindrica) suggests a salt marsh environment with some fresh water input (occur- rence of Bithynia leachi). At the bottom of CN14 succession (Fig. 4A) there is a yellowish sand level with Cerastoderma glaucum, Abra segmentum and Hydrobiidae; several Bithynia lea- chi also occur. A sharp contact separates the sand from a greyish clay layer accumulated in an almost natural environment. The clayey unit could be subdivi- ded in three horizons. The lower and the upper ones are characterized by a thick lamination due to the pre- sence of organic detritus (Fig. 4E). The central part is made of massive clay with subvertical tiny roots. Faunal assemblage is dominated by brackish organisms: Ovatella myosotis and Truncatella subcylindrica, some miliolids (Quinqueloculina spp.). We interpreted this environment as a salt marsh. The natural deposit at the lower end of the CN15 succession is a silty sand made mostly by tephra gra- nules mixed to lagoon molluscs (Hydrobiidae and Abra segmentum), foraminifers and ostracods. The upper part of the natural deposit is represented by a fine grey- greenish sediment, slightly sandy, with fragments of reworked molluscs, perhaps accumulated in a marshy environment. The rest of the sequence accumulated under the more or less direct anthropic influence. In all the cores the contact between natural and anthropogenic sedi- ments is sharp and occurs at similar elevations in the CN12, CN14 and CN15 series, while it is deeper in the CN16 core. The anthropogenic sediments are porous, dark grey, with scattered Bronze age pottery fragments, concotto, lithic tools, domestic animal bones (someti- mes burnt), marine shells (Phyllonotus trunculus) and charcoals. The bedding planes are hardly visible, except when the drillings went across to black hearth material, concotto levels, tramping and/or leveled surfa- ces and structures in use for a considerable time span (Fig. 4A). A number of leveled surfaces have been found in all the drillings in this group. These are well evident in the CN12, CN14 and CN15 cores, because realized with materials very different (in colour and nature) from the rest of core sediments. In some cases the surfaces were leveled by spreading a fine yellowish calcareous sediment, originated from the weathering of the Mesozoic limestone that outcrop in the nearby area (Fig. 4C). In other cases the material spread to create a plain surface is a clayey sediment with high carbonates content. Several tramping surfaces lay around the same elevation. The deeper surface lays at 1.06 m below m.s.l. in the CN14 and CN15 cores (Fig. 4F). There is, in the same two cores, another yellowish surface at 0.19 m above m.s.l., while in the CN12 and CN15 series, drilled at a greater distance, a leveled surface lays around 1.5 m above present day sea level. The upper part of the CN12 core (between 1.44 and 2.08 m above m.s.l.) and the middle of the CN14 core (between 0.17 and 0.49 m above m.s.l.) are cha- racterized by two horizons with many superimposed thin levels, made of alternating fine yellowish and greyi- sh sediments, charcoal levels, brown soil, concotto etc. (Fig. 4B). We think that at those horizons the cores went across to structures in use for a long time span (e.g., a hut). On the other hand, in the CN16 core the leveled surfaces seem to have been realized with a different technique. In fact, these are just a few centimetres thick and realized with limestone, calcarenite, siliceous peb- bles and calcareous crust fragments. The upper part of the CN12, CN14 and CN16 continue with a brown clayey soil with small lithic frag- ments deeply disturbed by ploughing. 502 M. Caldara, I. Caroli & O. Simone 503Geomorphological changes ... Fig. 4 - CN 14 core (A) and close ups of some cores (B-F): B = CN14 core, D phase, alternation of yellow and grey sediments used to level the ground, red concotto levels and black burnt hearth grounds; C = CN12 core, D phase, two beds, yellow and grey, of material spread to level the ground; D = CN12 core, A phase, silt with tiny vertical roots, lagoon taxa and volcanoclastic horizon; E = CN14 core, A phase; from the bottom: yellowish sand with lagoon fauna, thick lamination due to the presence of organic detritus and greyi- sh clay (salt marsh); F = CN15 core, B phase, grey sediment spread to level the ground and a black horizon of burnt vegetal remains. Except for the CN1, CN2 and CN5 cores, got by an hydraulic corer, all the other cores (among which the ones above) were obtained by a percussion drilling machine. Carotaggio CN 14 (A) e particolari di alcune carote (B-F): B = CN14 perforazione, fase D, alternanza di piani battuti gialli e grigi, con livelli rossi di concotto e neri di suoli bruciati; C = perforazione CN12, fase D, due battuti giallo e grigio; D = perforazione CN12, fase A, limi con sottili radici verticali sormontati da un orizzonte vulcanoclastico caratterizzati da faune lagunari; E = perforazione CN14, fase A, sabbie giallastre con fauna lagunare sormontate da argille sottilmente laminate ricche di sostanza organica che passano ad argille grigie massicce con tracce di radici. L’ambiente delle argille è di salt marsh; F = perforazione CN15, fase B, piano battuto di colore gri- gio sormontato da un orizzonte di resti vegetali combusti. Le carote CN1, CN2 e CN5 sono state ottenute con una macchina idraulica, per tutte le altre (tra cui quelle in figura) è stato utilizzato il metodo di carotaggio a percussione. 5. EVENTS RECONSTRUCTION Due to the elements collected after this study we can give a tentative reconstruction of the events at the foot of the Coppa Nevigata knoll. The events succession is described starting from the most ancient one (Fig. 3). 5.1. A phase - natural phase I The natural phase A is characterized by environ- ments with different peculiarities, but under absent or minimal human influence. This phase has been recogni- zed in almost all the successions, except for CN11 (that did not reach those levels) and CN1 (whose bottom is characterized by probably colluvial sediments). The CN16 core is the only one that reached the Cretaceous limestone substrate and presenting eviden- ce of marine sedimentation. Above this episode and within the other drillings, the sediments are dominated by lagoon species. The lagoon environment in the external (distal) successions (A transect) is characteri- zed by salinities ranging between 25 and 18‰ (after molluscan analysis). The correspondent section, in the internal (proximal) cores (C transect), drilled at higher elevations, seem to be conditioned by a lower salinity. All those lagoonal deposits can be related to the "Hydrobiidae and Cerastoderma Lagoon" found in the CN2 and CN5 cores. The cores at the western end of the C transect (CN12 and CN14) show, above the lagoon sediments, salt marsh deposits. On the other hand, in the CN15 core the lagoon sediments are overlain by deposits accumula- ted in a dry environment at the edge of a fresh marsh. Among the A transect drillings, pyroclastic depo- sits were found only in the CN2 core (Caldara et al., 2001; 2003a), whilst in the C transect tephra are pre- sent in the CN12 (Fig. 4D) and CN15 cores. Tephra deposits have been found in marginal lagoon environ- ments, within the proximal cores. The thickness of these deposits, as well as the depth, increases with the distance from the centre of the settlement. This should confirm what hypothesised in Caldara et al. (2003a), that is there was first a phase of tephra accumulation due to direct fall on a large surface characterized by dif- ferent environments, followed by erosion and secon- dary deposition in low areas. The organic deposit (Caldara et al., 2001; 2003a, in press a) has been found in all bore holes in the A transect and in the deeper one in the B transect, while it is not present next to the settlement (C transect). The maximum thickness of this deposit occurs in CN13 core (A transect). At this point it looks more probable that the deposition of the organic horizon is due to washing processes acting along the slopes of the Coppa Nevigata knoll, rather than intentional accumula- tion of material dumped by man next to the lagoon edge (Caldara et al., 2003a; in press a). In conclusion, in the natural phase A there are lagoon (Hydrobiidae and Cerastoderma Lagoon), salt marsh (Ovatella myosotis facies), paludal and colluvial deposits, with abundant input of sediments from the settlement. 5.2. B phase – terrestrial phase I This phase corresponds to the terrestrial phase I (Caldara et al., 2003a) in the A transect cores and pro- bably in the B transect and seems to start during the Early Sub-Apennine (Caldara et al., 2003a; in press a). The erosive surface that separates the A phase from the B phase sediments appears as sloping towards the middle of the lagoon except for the CN16 core, where a depression is outlined. Actually, the CN16 core represents an anomaly; in fact, in this suc- cession the pyroclastic horizon occurs in the B phase and not in the natural interval. This drives us to hypothesize that the area around the CN16 core, no longer lagoonal, were utilized by the Coppa Nevigata community from the Cetina phase (Recchia, 2002), or at least from the Early Bronze Age (before the "Avellino" event). The B phase is characterized by the occurrence of the first leveled surfaces (CN14 and CN15 cores, Fig. 4F) and by a thick accumulation of burnt material (CN13 core). The end of the B phase is well identifiable within the distal cores (CN2 and CN5) because of the return of brackish conditions. In the other cores the B phase seems to be in continuity with the D phase. Tentatively we put the limit between the B and D phases at the base of the yellowish tramping surfaces slightly above the sea level within the C transect cores, and at the base of a tramping surface laying at 0.82 m below m.s.l. in the CN13 core. 5.3. C phase - natural phase II At the moment the C phase has been recognized only for the distal cores (CN2 and CN5), while we hypothesize its continuity in the area around the cores of the B transect (Fig. 3), this should be assessed by means of new drillings. In other words, the setting of the natural phase II seems as occurred only in the northern part of the studied area, facing the present day coastline (Fig. 1). The natural phase II is characterized by the suc- cession of three brackish environments, i.e. Cerasto- derma Lagoon, Salt marsh I and Hydrobiidae and Abra segmentum Lagoon (Caldara et al., 1999; 2001). These environments have been correlated to the Hydrobiidae and Cerastoderma Lagoon, recognised in the middle of the basin (CN4 and CN3 cores; Caldara & Simone, in press). This phase occurred between the Final Bronze and a not well defined Iron Age (Caldara et al., 2003a). 5.4. D phase – terrestrial phase II In the distal area (A transect cores), the D phase corresponds to the terrestrial phase II (Caldara et al., 2003). This one represents a new phase of occupation of the areas freed by lagoon waters. The lagoon retreat seems to have been occurred because of natural cau- ses (Caldara & Simone, in press). We do not know exactly when this phase develo- ped, perhaps between the Iron Age and Roman period (Caldara et al., 2003a; in press a). The erosive surface that separates sediments per- tinent to the C and D phases lays around 0 m in the distal area (A transect), whilst within the B and C tran- sect cores, given the similar characteristics that sedi- ments show, the limit between B and D phases is somewhat undefined. The D phase is characterized by a number of tramping/leveled surfaces. These can be single (CN15) 504 M. Caldara, I. Caroli & O. Simone or superimposed, sometimes there is an alternation of tramping/leveled surfaces, concotto levels and burnt hearth grounds (CN12 and CN14, Fig. 4B,C). As regards to the distal area (CN2 and CN5 cores), given the findings of burnt soils and domestic bone fragments, Caldara et al. (in press a) hypothesized a pastoral-agricoltural land use. Nevertheless, the evi- dence collected with this study suggests that, at the foot of the knoll, several simple inhabitation structures were in existence during the various occupation pha- ses. 5.5. E phase - natural phase III The E phase has been identified only in the distal drillings (A transect) and corresponds to the salt marsh II (Caldara et al., 2003a). Therefore, the recovery of natural environment seems has been occurred in the areas at the external margin of the settlement. The E phase sediments are cut in their upper part by an erosive surface. Their chronological position is not well defined yet. In addition there were not found any anthropogenic indicators, therefore we think that, during the deposition of the E phase sediments, the Coppa Nevigata knoll was not inhabited. The more recent findings are VI century BC pottery fragments (Pallottino, 1951); that could mean that the E phase sediments accumulated successively. 5.6. F phase - arid terrestrial phase The F phase did not leave sediments; it is reco- gnizable because of two elements, i. e. the erosive sur- face that cut the paludal deposits in all the distal cores (A transect) at different elevations and the presence of calcareous crusts and lumps (evaporite) in the upper part of E phase sediments. These features suggest a subaerial phase characterized by a marked pedogene- sis. The thick calcareous crust indicates arid conditions lasting for a long span of time. That is why we called this phase "arid terrestrial". At the moment we are not able to say how long the terrestrial phase lasted and when the erosion occur- red. The only chronological indication comes from the stratigraphic position of this event. 5.7. G phase - natural phase IV The G phase corresponds, in the distal area (A transect), to the Bithynia and Ovatella Wetland (Caldara et al., 2003a). In the lower part of the G phase sediments the environment seems to be fairly brackish, whilst in the upper part the waterbody is exclusively conditioned by fresh water input. Radiocarbon data, 370±50 yr BP (CN2 Caldara et al., 2001) and 140±40 yr BP (Simone, 2003), suggest that this wet area was set around the end of the Middle Ages. Documentary data show that this wet area (part of Salso and Versentino lakes) reached its maximum development around the end of the XVIII century (Caldara et al., 2002b), after the ruinous floods of the Cervaro and Carapelle streams in 1795 (Giustiniani, 1797 - 1805). 5.8. H phase – historical reclamation These sediments correspond to the terrestrial phase III in Caldara et al. (2003a). The H phase is well recognizable within the A and B transects, while is less apparent in the C transect because of a loss of sample occurred during coring operations. Nevertheless, these sediments cover both E phase (natural wet area) and D phase sediments (anthropogenic). The accumulation of the H phase deposits occur- red after reclamation projects carried out through the diversion of the Candelaro stream. The very upper part of this interval is deeply disturbed by ploughing activi- ties. 6. CONCLUSIONS By this research, in the Coppa Nevigata area we reconstructed eight different phases grossly attributa- ble to two different evolution patterns (Fig. 3). The first one includes sediments intentionally accumulated or reworked by man. The latter includes deposits which sedimentation was basically driven by natural proces- ses, even though materials accumulated are often directly due to man activities at the settlement. We ascribed to the "anthropogenic" pattern the B, D and H phases. The "natural" pattern includes the phases A, C, E, F and G. We found the anthropogenic sediments (B and D phases) underground, up to circa 90 m from the nearest trenches dug by archaeologists (Fig. 1). These deposits were not found within the CN4 and CN3 cores, drilled respectively 400 and 800 m far from the settlement (Caldara & Simone, in press). On the contrary, the CN3 and CN4 cores show that the Hydrobiidae and Cerastoderma Lagoon persisted without interruptions from the Neolithic to the Iron age (Caldara et al., 2003b). Therefore, we correlate the Hydrobiidae and Cerastoderma Lagoon found in the CN3 and CN4 cores to the A, B, C, D and E phases identified close to the settlement. At the moment it is not possible to distinguish the limit between the anthropogenic B and D phases within the successions of the C group (next to the settlement), while these are well distinguishable in the distal A group. Even though directly related to anthropogenic activities carried out at the settlement, we defined as "natural" the high organic content material included in the A phase. We consider these sediments as washed out along the side of the knoll and deposited in a hollow at its foot, instead of intentionally disposed by man in a dumping area. The developing of the B phase seems to have been started from the most internal settlement areas (CN14 and CN15 cores). This would be in accor- dance with a urbanization phase occurred within the perimeter described by joining the hypothetical buried parts of defensive wall found by geophysical analysis (Fig. 1). On the other hand, the presence of hearths in distal areas (CN13) could be attributed to activities car- ried out outside the defensive wall system. As regarding the D phase (anthropogenic), by the study of the CN2 and CN5 cores (in the distal area) Caldara et al. (in press b) assumed an agricultural- pastoral land use for those levels. After this study, given the presence of a number of single or multiple (supe- rimposed) tramping/leveled surfaces in association with 505Geomorphological changes ... concotto and hearths levels (CN12, CN14 and CN15), we can hypothesize, at least for the areas surrounding the settlement, the persistence of simple inhabi- tation/occupation units. The B and D anthropic phases recorded within the CN16 sediments are worthy of further analyses. In parti- cular, the first consideration is that around the CN16 drilling site the colonization occurred earlier than other investigated places within the Coppa Nevigata area. In that case the first post Neolithic occupation phases would be occurred at least starting from the Early Bronze age (before the "Avellino" event) or perhaps from the Cetina phase (Recchia, 2002). In addition, the CN16 core suggests a use of the area outside the Apennine defensive system (part of the wall has been found beneath the surface by geophysical surveys; Fig. 1). The H phase considerably differs from the two above mentioned B and D anthropic phases. The H phase deposits accumulated after direct anthropic actions, even though sedimentation processes occur- red under almost natural conditions. In fact, during the XIX century, the Candelaro and Cervaro streams used to wander in the coastal plain without flowing into the sea through a defined mouth. As a result, great part of the Tavoliere coastal area was marshy and unhealthy (Caldara et al., 2002b). For this reason in that period challenging reclamation projects have been conceived (Afan de Rivera, 1838; 1845). To protect their fields from frequent floodings the landowners started to build embankments. Subsequently, the Candelaro and Cervaro streams were partially (sometimes totally) diverted towards the hollows (surrounded by artificial dikes) in order to fill them by siltation. In particular, the Candelaro stream was embanked and its mouth was progressively shifted downstream towards the middle of the former Salso lake. As examples we quote here reclamation works made between the 1869 and the 1909 (Pareto, 1865; historical maps by the Italian Military Geographical Institute), in 1939 (Rotella 1984) and after the World War II. By this study we described the evolutionary mechanisms, both anthopic and natural, that acted in the Coppa Nevigata area since the Neolithic. Nevertheless, several problems have to be fixed in the future. In particular we found that correlations among the group B cores are not reliable, thus should be use- ful to drill more cores around the CN1 site. The chrono- logy of the B and D anthropic phases, well defined for the distal drillings, is not verified for those close to the settlement (C group) and raises perplexities among the archaeologists. In addition, the area around the CN16 core should be investigated, in order to verify the existence of an early colonization phase occurred outside the walls and define the limits of the bulge in the limestone substrate (Fig. 3). An other arousing question is from where the Coppa Nevigata inhabitants supplied material used for ground leveling and to realize the tramping surfaces. In conclusion, it is apparent how man was able, through time, to modify the surrounding environment, in particular during the Middle Bronze (B phase), the Iron age (D phase) and during the last two centuries. Even when the accumulation of sediments occurred under almost natural conditions, anthropogenic activities con- ditioned the environment evolution, such as during the deposition of the high organic content sediments in the A phase and during the C phase. ACKNOLEDGEMENTS This study was made possible thanks to funding from the Bari University Research Project “Genesi ed evoluzione geomorfologica delle piane pugliesi e luca- ne”, led by Prof. F. Boenzi, and the Bari University "Dottorato di Ricerca in Geomorfologia e Dinamica Ambientale". We wish also to thank Dr. Giulia Recchia for chronological attribution of the many pottery frag- ments found within the new cores and for the useful discussions about the anthropogenic sediments. REFERENCES Afan de Rivera C. 1838 - Memoria su i mezzi di ritrarre il massimo profitto dal Lago Salpi, coordinando quest'impresa a quella piu' vasta di bonificare e migliorare la pianura della Capitanata. Stamperia e Cartiere del Fibreno, Napoli. Afan de Rivera C. 1845 - Del bonificamento del Lago di Salpi coordinato a quello della Pianura della Capitanata. Napoli, Stamperia e Cartiere del Fibreno, 638 p. Albore Livadie C., Mastrolorenzo G. & Vecchio G. 1998 - Eruzioni pliniane del Somma-Vesuvio e siti archeologici dell’area nolana. In: Guzzo P.G. Peroni R. (ed.), Archeologia e Vulcanologia in Campania. Napoli, 39-86. Andronico D., Calderoni G., Cioni R., Sbrana A., Sulpizio R. & Santacroce R. 1995 - Geological map of Somma Vesuvius Volcano. Periodico di Mineralogia, 64, 77-78. Boenzi F. & Caldara M. 1999 - Il Gargano: l’uomo e l’ambiente geologico. L’Universo 79(6), 736-754. Boenzi F., Caldara M., Moresi M. & Pennetta L. 2001 - History of the Salpi lagoon-sabhka (Manfredonia Gulf, Italy). Il Quaternario 14(2) 2001, 93-104. Boenzi F., Caldara M., Pennetta L. & Simone O. in press - Environmental aspects related to the phy- sical evolution of some wetlands along the Adriatic coast of Apulia (Southern Italy): a review. Journal of Coastal Research SI 39, (Proceendigs of the 8th Int. Coastal Symp.) March 14-19, 2004, Itajaí, - Brazil. Caldara M., Cazzella A., Fiorentino G., Lopez R., Magri D., Simone O. 1999 - Primi risultati di una ricerca paleoambientale nell’area di Coppa Nevigata (Foggia). Atti del XIX Conv. Naz. Preistoria, Protostoria e Storia della Daunia, S. Severo 28-29 Novembre 1998, 199-236, Foggia. Caldara M., Cazzella A., Fiorentino G., Lopez R., Magri D., Moscoloni M., Narcisi B. & Simone O. 2003a - The relationship between Coppa Nevigata settle- ment and the wetland area during the Bronze Age (south-eastern Italy). In: Fouache E. (ed.) The Mediterranean World Environment and History, 429-438, Elsevier, Paris. Caldara M., Cazzella A., Fiorentino G., Lopez R. & Simone O. 2001 - Nuovi dati sull'evoluzione 506 M. Caldara, I. Caroli & O. Simone paleoambientale nell'area di Coppa Nevigata (Foggia). Atti del XXI Conv. Naz. Preistoria, Protostoria e Storia della Daunia. S. Severo 25-26 Novembre 2000, 87-132, Foggia. Caldara M., Cazzella A., Fiorentino G. & Simone O. in press a - L’insediamento archeologico di Coppa Nevigata (Tavoliere di Puglia) tra uomo e ambiente. Atti della Riunione di lavoro del Gruppo di Lavoro AIGeo “Il contributo della Geografia Fisica e della Geomorfologia alla ricerca Archeologica”, Gonnesa (Ca) 2-4 Settembre 2003. Caldara M., Cazzella A. & Moscoloni M. 2002a - Il rap- porto tra l’insediamento di Coppa Nevigata e l’a- rea umida durante l’età del Bronzo. In: “Paesaggi d’acque. Ricerche e Scavi”, a cura di Nuccia Negroni Catacchio. Atti del Quinto Incontro di Studi sulla Preistoria e Protostoria in Etruria. Sorano-Farnese, 12-14 Maggio 2000, 343-354, Centro Studi di Preistoria e Archeologia Milano. Caldara M. & Palmentola G. 1991 - Lineamenti geo- morfologici del Gargano con particolare riferimen- to al carsismo. International Conference on Environmental Change in Karst Areas, Apulian excursion, September 23th -27th, 1991, “Itinerari speleologici”, s. II, V, 53-66, Castellana Grotte. Caldara M. & Pennetta L. 1993a - Nuovi dati per la conoscenza geologica e morfologica del Tavoliere di Puglia. Bonifica, 8(3), 25-42, Bastogi, Foggia. Caldara M. & Pennetta L. 1993b - Ambienti aridi del tipo "Sabkha" nei sedimenti olocenici della piana costiera fra Manfredonia e Zapponeta. Bonifica, 8(3), 73-82, Bastogi, Foggia. Caldara M. & Pennetta L. 1996 - Influenza del clima sul paesaggio neolitico del basso Tavoliere di Puglia - Atti seminario internazionale "Forme e tempi della Neolitizzazione nell'Italia meridionale e in Sicilia" Rossano Calabro, 29 aprile - 2 maggio 1994, tomo II, 559-568, con bibliografia. Caldara M., Pennetta L. & Simone O. 2002b - Holocene Evolution of the Salpi Lagoon (Puglia, Italy) Int. Coastal Symp., Templepatrick, Northern Ireland, March 25th-29th 2002, Journal of Coastal Research SI 36, 124-133. Caldara M., Pennetta L. & Simone O. in stampa b - L’ambiente nell'area dell'insediamento neolitico di Masseria Candelaro. In: “Il Villaggio Neolitico di Masseria Candelaro” a cura di Cassano S. & Manfredini S., 20pp, figg 8, Grenzi editore, Foggia Caldara M., Pennetta L. & Zito G. 1994 - L'area umida alla foce del Candelaro: osservazioni paleoam- bientali e climatiche. - Atti II Workshop sul tema "Parchi naturali ed aree protette" del Progetto Strategico "Clima Ambiente e Territorio nel Mezzogiorno", S. Maria di Salina (Isole Eolie - ME) 28-30 Maggio 1990, 211-234. Caldara M. & Simone O. 2003 - The area of Coppa Nevigata settlement between history and environ- mental changes. Final Conference “Quaternary coastal morphology and sea-level changes”, Otranto/Taranto – Puglia (Italy) 22-28 September 2003, Field Guide book GI2S Coast, Research publication 5, 139-147. Caldara M. & Simone O. (in press) Coastal changes in the eastern Tavoliere Plain (Puglia, Italy) during the Late Holocene: natural or anthropic? Quat. Sci. Rev. Caldara M., Simone O., Gehrels R.W., in preparation - Holocene environmental changes in the Palude Frattarolo - Lago Salso area (Tavoliere Plain - Southern Italy). To be submitted. Caldara M., Simone O., Porzia S., 2003b - L’aera umida di Coppa Nevigata tra Neolitico e l’Età del Bronzo. Atti del 23° Convegno Nazionale sulla Preistoria, Protostoria e Storia della Daunia, S. Severo 23 - 24 Novembre 2002. Stabilimento Litografico Centrografico Francescano, Foggia, 203-230. Cazzella A. 1996a - Coppa Nevigata. Atti seminario internazionale "Forme e tempi della Neolitizzazione nell'Italia meridionale e in Sicilia" Rossano Calabro, 29 aprile - 2 maggio 1994, tomo I, 108-114 Cazzella A. & Moscoloni M. 1988 - La ripresa degli scavi nei livelli dell’età del Bronzo di Coppa Nevigata. Atti VII Convegno Preistoria, Protostoria e Storia della Daunia, San Severo, Dicembre 1985, 103- 110. Cazzella A. & Moscoloni M. 1990 - Recenti scavi nei livelli dell’età del Bronzo di Coppa Nevigata (cam- pagne 1988-1989). Atti XI Convegno Nazionale sulla Preistoria, Protostoria e Storia della Daunia, San Severo, Dicembre 1989, 99-115. Cazzella A. & Moscoloni M. 1999 - The walled Bronze Age settlement of Coppa Nevigata, Manfredonia, and the development of craft specialisation in southeastern Italy. In: Tykot R.H., Morter J., Robb J.E. (eds.), Social Dynamics of the Prehistoric Central Mediterranean. Accordia Specialist Stu- dies on the Mediterranean 3, 205-216, London. Cazzella A., Moscoloni M. & Recchia G. 2001 - Coppa Nevigata: campagne di scavo 1999 e 2000. Atti XIX Convegno Nazionale sulla Preistoria, Protostoria e Storia della Daunia, 24-26 Novembre 2002, San Severo, 153-170. Giustiniani L. 1797-1805 - Dizionario geografico ragio- nato del Regno di Napoli. Stamperia Giovanni De Bonis, tomo XI (Parte II, tomo I), Napoli. Levi S.T., Cioni R. & Cazzella A. 1999 - Presenza di materiale pomiceo dell’eruzione vesuviana di Avellino nella cerimica dell’età del bronzo di Coppa Nevigata (FG). In: C. Albore Livadie (ed.) “L’eruzione vesuviana delle “Pomici di Avellino” e la facies di Palma Campania, 341-354, Edipuglia, Bari. Luperto Sinni E. & Masse J.P. 1986 - Données nouvel- les sur la stratigraphie des Calcaires de plateforme du Crétacé inférieur du Gargano (Italie méridiona- le). Riv. It. Paleont. Strat., 42, 33-66. Minnitti C. 1999 - Lo sfruttamento dei molluschi nell’età del Bronzo di Coppa Nevigata Atti XIX Conv. Preistoria, Protostoria e Storia della Daunia, San Severo, 28-29 Novembre 1998, 177-197. Mosso A. 1909 - Stazione preistorica di Coppa Nevigata presso Manfredonia. Monumenti Antichi dei Lincei, 19, Roma. Pallottino M. 1951 - Vaso egiziano inscritto proveniente dal villaggio preistorico di Coppa Nevigata. Rend. Acc. Lincei, VI fasc. 11-12, 580-590. Pareto R. 1865 - Sulle bonificazioni, risaie ed irrigazioni del Regno d'Italia. Relazione a S.E. il ministro di agricoltura, industria e commercio. Milano, 507Geomorphological changes ... Tipografia e Litografia degli ingegneri, 270 pp. Puglisi S.M. 1955 - Industria microlitica nei livelli a cera- mica impressa di Coppa Nevigata. Riv. Sc. Preist. 10, 19-37. Puglisi S.M. 1975 - Lo strato neolitico di Coppa Nevigata. Atti Colloquio Internazionale sulla Preistoria e Protostoria della Daunia, Foggia, 24- 28 Aprile 1973. IIPP, Firenze. Puglisi S.M. 1982 - Coppa Nevigata (Manfredonia, Foggia). In: Vagnetti L. (ed.), Magna Grecia e Mondo Miceneo. Nuovi documenti. Atti XXII Conv. Studi Magna Grecia, Taranto 7-11 Ottobre 1982, 45-51. Recchia, G. 2002 - I siti costieri garganici e i loro rap- porti transmarini tra Eneolitico ed età del Bronzo. In: "Paesaggi d’acque" a cura di N. Negroni Capacchio, Sorano-Farnese 13-14 Maggio 2000, 331-341. Rolandi G., Mastrolorenzo G., Barrella A.M. & Borrelli A. 1993 - The Avellino plinian eruption of Somma- Vesuvius (3760 y. B.P.): the progressive evolution from magmatic to hydromagmatic style. Journal of Volcanology and Geothermal Research, 58, 67-88. Rotella G. 1984 - Cinquant’anni di bonifica nel Tavoliere. In: Ciccone S. (ed.) Cinquant’anni di bonifica nel Tavoliere, 31-171, Bastogi ed. Foggia Schmiedt G. 1973 - Contributo della fotografia aerea alla ricostruzione dell'antica laguna compresa fra Siponto e Salapia, Arch. Stor. Pugl., 26, 159-172. Simone O. 2003 - Evoluzione olocenica e dinamica ambientale delle piane costiere pugliesi. Tesi di dottorato inedita. Università degli Studi di Bari - Bari, 171 pp. Terrasi F., Campajola L., Petrazzuolo F., Roca V., Romano M., Brondi A., D’Onofrio A., Romoli M. & Moniot R.K. 1999 - Datazione con la spettrometria di massa ultrasensibile di campioni provenienti dall’area interessata dall’eruzione delle “Pomici di Avellino” . In: Albore Livadie C. (a cura di), L’eruzione vesuviana delle “Pomici di Avellino” e la facies di Palma Campania. Edipuglia, Bari, 139- 146. 508 Ms. ricevuto il 17 luglio 2004 Testo definitivo ricevuto il 28 settembre 2004 Ms. received: July 14, 2004 Final text received: September 28, 2004 M. Caldara, I. Caroli & O. Simone