Imp.D'Alessandro RREESSPPOONNSSEE OOFF MMAACCRROOBBEENNTTHHOOSS TTOO CCHHAANNGGEESS IINN PPAALLAAEEOOEENNVVIIRROONNMMEENNTTSS IINN TTHHEE LLOOWWEERR--MMIIDDDDLLEE PPLLEEIISSTTOOCCEENNEE ((LLUUCCAANNIIAA BBAASSIINN,, SSOOUUTTHHEERRNN IITTAALLYY)) AAssssuunnttaa DD’’AAlleessssaannddrroo11,, RRaaffaaeell LLaa PPeerrnnaa && NNeerrii CCiiaarraannffii Dipartimento di Geologia e Geofisica, Università di Bari Campus Universitario, Via Orabona 4, 70125 Bari, Italia 1Corresponding author e-mail address: tina@geo.uniba.it ABSTRACT The Lucania Basin is part of the Bradano Trough, whose internal sectors were subject to continuous subsidence during the Early Pleistocene. This led to the deposition of a muddy sequence with a thickness in excess of 500 m in the Montalbano Jonico area (Lucania). During the Middle Pleistocene, a general shallowing took place, starting from the more internal sectors of the Lucania Basin. The regressive succession exposed in the badland area of Montalbano is candidate of the Early/Middle Pleistocene boundary stratoty- pe. The restored sequence was obtained from selected partial sections correlated by means of nine volcaniclastic layers, each of them characterised by distinct macrofauna assemblages. A gap of unknown thickness (probably a few tens of metres) divides the succession in two intervals: the lower one (A) consists of muddy deposits, the upper one (B) of muddy to sandy deposits, overlain by organogenic calcarenites in the uppermost part. Integrated palaeoecologic, taphonomic and ichnologic analyses were performed in order to recon- struct bathymetric fluctuations and environmental changes, based on biotic responses. Palaeocommunities from the lower interval point to background environments with moderate sedimentation rate, low hydrodynamic level and low oxygen content, punctuated by nume- rous events of mass deposition. Palaeodepths range from the upper bathyal to the shelf break. Through interval B, palaeodepths range from the shelf break to the inner shelf-nearshore. This interval includes genetically different shell beds. The inferred background envi- ronments are characterised by sedimentation rates and water-energy ranging from low to high, under fully aerobic conditions. Like in the lower interval, blanketing events are numerous. Within the general regressive trend, several fifth- and sixth-order cyclothems are recognised throughout the sequence. Vertical changes in the fossil communities within individual cycles indicate changes in water depth that parallel climatic fluctuations, as proved by climatic indicators in interval B. Further, the sea-level curve proposed in the pre- sent work matches well the local oxygen isotope curve, but the bathymetric range of some of the cycles suggests a tectonic control superimposed on the climatic influence. RIASSUNTO - Il bacino lucano fa parte della Fossa Bradanica i cui settori interni furono interessati, nel corso del Pleistocene inferiore, da una subsidenza pressoché continua che determinò l’accumulo di depositi prevalentemente fangosi, affioranti per uno spessore di oltre 500 m nell’area di Montalbano Jonico (Lucania). Nel Pleistocene medio ebbe inizio una fase regressiva a partire dai settori più interni del bacino. La successione regressiva di Montalbano è candidata come stratotipo per il limite Pleistocene inferiore/medio. La sequenza è stata ricostruita sulla base di numerose sezioni parziali, correlate attraverso nove livelli guida vulcanoclastici, a loro volta caratterizzati da associazioni faunistiche diverse. Un’interruzione di entità non valutabile (probabilmente di poche decine di metri) divide la successione in due intervalli: quello inferiore (intervallo A) è rappresentato da depositi fangosi, quello superiore (intervallo B) è costi- tuito da depositi da fangosi a sabbiosi, passanti a calcareniti organogene nella parte sommitale. Analisi paleocologiche, tafonomiche ed icnologiche, svolte in maniera integrata, sono state utilizzate per dedurre fluttuazioni batimetriche ed evoluzione paleoambientale. Le paleocomunità dell’intervallo inferiore, permettono di dedurre ambienti di background caratterizzati da tasso di sedimentazione general- mente moderato, bassa energia e basso contenuto di ossigeno, sottoposti a numerosi eventi di deposizioni in massa e localizzati a profondità variabili fra il batiale superiore ed il margine di piattaforma. Nell’intervallo B, le paleobatimetrie variano tra il margine di piat- taforma e la transizione al nearshore. La presenza di shell beds di varia genesi è tipica di questo intervallo. Gli ambienti di background sono caratterizzati da tasso di sedimentazione e livello di energia idrodinamica da basso ad alto e da buona ossigenazione. Come nel- l’intervallo inferiore, sono numerosi gli eventi di blanketing. Nell’ambito di una generale tendenza regressiva, sono stati riconosciuti diversi ciclotemi di quinto e sesto ordine; tale ciclicità è evidenziata dai cambiamenti, in verticale, delle associazioni fossili. Il controllo climatico è confermato dalla presenza di indicatori climatici fra le faune dell’intervallo B. Esiste, inoltre, un buon accordo fra i cicli dedotti sulla base del presente studio e la locale curva isotopica dell’ossigeno, anche se l’ampiezza batimetrica di alcuni cicli suggerisce la sovrapposizione di un controllo tettonico a quello climatico. Keywords: benthic communities, sea level changes, oxygen changes, Early-Middle Pleistocene, Lucania Basin. Parole chiave: comunità bentoniche, fluttuazioni del livello marino, fluttuazioni nell’ossigenazione, Pleistocene inferiore e medio, Bacino lucano. Il Quaternario Italian Journal of Quaternary Sciences 1166(1Bis), 2003, 167-182 11.. IINNTTRROODDUUCCTTIIOONN The Bradano Trough (Fig. 1) is part of a Foredeep Basin, extending from Molise to the Gulf of Taranto. During the Lower Pleistocene a high sedimentation rate and an almost continuous downlift led to the deposition of thick clastic sequences. Particularly, along the Apenninic margin of the Lucania Basin (i.e. the southernmost part of the foredeep), active subsidence caused the deposition of hemipelagic clays interfingered with muddy turbiditic deposits. A generalized shallowing phase started in the Middle Pleistocene resulting in a relative shallowing of the basin, testified by a shift from upper slope to shelf depositional environments. The Pleistocene marine succession cropping out in the bad- lands area of Montalbano (Fig. 2) contains fossil assem- blages, that clearly indicate short-term cyclic palaeo- bathymetric changes. As a whole, the sequence shows 168 A. D’Alessandro, R. La Perna & N. Ciaranfi a general regressive trend, from upper slope to inner shelf deposition (Ciaranfi et al, 1997). This succession is over 500 m thick and mostly silty-clayey. Except for an isolated section about 15 m thick, referred to the “large” Gephyrocapsa biozone, the sequence is referred to the “small” Gephyrocapsa biozo- ne and to the Pseudoemiliania lacunosa biozone, provi- ding a reference for the selection of the Lower/Middle Pleistocene boundary stratotype (Ciaranfi et al. 1997, Maiorano et al., in press). The present study investigates the palaeoenviron- mental evolution inferred from the fossil assemblages in order to discriminate between the cyclically modulated climatic changes, and the tectonic control. A palaeoeco- logical analysis of the macrofauna has been combined with ichnological and taphonomic observations, with the aim of clarify changes in depth, oxygen content, and sedimentation rate. Molluscs have been primarily used due to their good distribution and abundance in the stu- died succession; nevertheless, environmental data have been also obtained from other groups (e.g., echino- derms, decapods, bryozoans, planktonic gastropods). Such an integrated approach enables a more detailed stratigraphic resolution within a framework of predictable relations between biota and sedimentation dynamics. 22.. SSTTRRAATTIIGGRRAAPPHHYY The “Argille Subappennine” Formation crops out in the Montalbano Jonico area in a monoclinal, gently SE- dipping structure. In spite of several mainly N-S and NE- SW trending faults, a continuous succession (Fig. 3) was reconstructed by numerous partial sections over an area of about 2.5 km2 (Ciaranfi et al., 2001, fig. 5). Nine volcaniclastic beds are present (Fig. 2), ran- ging from a few centimetres to 50 cm in thickness. They consist of pure ashes or volcani- clastic-rich sands, occasionally with pumice clasts as redeposi- ted material. This volcanic mate- rial is referred as have been generated by an alkaline under- saturated volcanism (De Rosa in Ciaranfi et al., 1996). The volca- niclastic beds (V1-V9) are asso- ciated to benthic palaeocommu- nities pointing to different bathy- metric trends, and this allows them to be used as marker beds (Ciaranfi et al., 2001, fig. 7). The recognition of alterna- ting darker and lighter intervals (3-6 m thick) all through the sec- tion and the presence of a dark, laminated horizon rich in Delec- topecten vitreus (GMELIN) provi- ded further tools for correlation (Ciaranfi et al., 2001, figs. 5, 7). A gap of unknown thick- ness (probably a few tens of metres) divides the restored succession in two intervals. The lower interval (AA), about 180 m thick, consists of dark grey, massive to locally laminated mud (i.e., silty clay to clayey silt), commonly bioturbated and with dispersed macrofossils. The upper interval (BB), over 300 m thick, consists of muddy to sandy shelf deposits, generally with abundant macrofauna, and including thick shell- beds. This interval is capped by transgressive continen- tal conglomerates. 33.. MMEETTHHOODDSS Three main approaches have been used to unra- vel the depositional evolution of the study sequence. Fig. 1 - Structural map of Southern Italy and location of the Montalbano Jonico sequence. Fig. 2 - View of the Montalbano Jonico badlands with the thickest volcaniclastic layers V5 (lower arrow) and V7 (upper arrow). 169Response of macrobenthos to changes ... Taphonomic observations allowed to infer the main depositional patterns (i.e., background versus event depositions, sedimentation rate, etc.). Palaeoichnologic analysis has been useful to infer fluctuations in the oxy- gen content, and palaeoecology has been mainly applied to reconstruct bathymetric changes. The main difficulty encountered during this study was the extremely time-consuming field work, due the need of recording in continuum taphonomic features, trace fossil distribution, faunal composition and distribu- tion, as well as sedimentological features (raw field data and the list of the taxa are available from the authors on request). The outcrop surface was first cleaned from the weathered part to expose fresh mudrock over a belt 50- 100 cm wide throughout the succession. Mudrock was then removed down to a depth of 10-30 cm and exami- ned. Most aragonitic skeletons and, to a less extent, even the calcitic ones, had been weakened by dissolu- tion, frequently fractured by compaction or are preser- ved as deformed steinkerns. A number of bulk-samples (50-100 kg of sediment) has been taken for a quantitati- ve palaeoecological analysis, although it often turned out that the benthic palaeocommunities (i.e associa- tions) suffered more or less preservational bias. Accordingly, most analyses were carried on a qualitative or semi-quantitative base. Whenever possible, fossil associations were paralleled to Recent Mediterranean biocoenoses or to the Atlantic communities. The Mediterranean biocoenoses were defined (Pérès & Picard, 1964) qualitatively and each of them is identified by a group of characteristic species, irrespective of their abundance. This bionomic approach, successfully adop- ted in the Mediterranean area, is the most suitable tool for the identification of biotopes (Basso & Corselli, 2002) and for a more precise bathymetric location of palaeo- communities. 44.. TTAAPPHHOONNOOMMYY 44..11 DDeessccrriippttiioonn IInntteerrvvaall AA. The main taphonomic feature of this interval is highly dispersed autochthonous fossils. Except for larger specimens, fossils are partly decalci- fied and preserved as more or less flattened steinkerns, often internally lined by pyrite (Pl. 1, Fig. 1). Pyrite linings, or pyrite steinkerns of the smaller components (Pl. 1, Figs. 4, 7), can occur in several unrelated taxa, e.g. molluscs, echinoids, decapods (Pl. 1, Fig. 10), and foraminifers (Pl. 1, Fig. 14). Fossils, often articulated and without signs of abrasion or encrustation, lack any preferred orientation in plane view and are concordant to bedding plane in transversal view. Numerous loosely packed concentrations have been regarded as intrinsic biogenic concentrations (Kidwell et al., 1986) including three thin, dark, mostly laminated intervals rich in Delectopecten vitreus. The mud-pecten shells, concen- trated in discontinuous pavements and layers, are pristi- ne, concordant, densely packed, commonly preserved as closed or slightly shifted valves, mostly with an inter- nal pyrite lining, single, and rarely in butterfly position. Extrinsic biogenic concentrations are recorded by seve- ral small accumulations of small angular bioclasts due to biological activity (Pl. 2, Fig. 13). Several obrution events are inferred by completely Fig. 3 - Reconstructed sequence of Montalbano Jonico. A and B are the stratigraphic intervals considered in the present work. 170 A. D’Alessandro, R. La Perna & N. Ciaranfi 171Response of macrobenthos to changes ... Fig. 1 - Abra longicallus (SCACCHI). Articulated, compressed and partially fragmented specimen with internal pyrite lining. Interval A, scale bar 0.5 cm. Fig. 2 - Abra nitida (MÜLLER). Specimen preserved in butterfly position. Interval A, scale bar 0.5 cm. Fig. 3 - Bathyspinula excisa (PHILIPPI). Articulated, compressed and partially fragmented specimen. Interval A, scale bar 0.5 cm. Fig 4 - Delicate erect bryozoan colonies with pyrite infilling. Interval B, scale bar 0.5 cm. Fig. 5 - Undetermined “small shrimp” preserved as a muddy steinkern with minor pyrite mineralization. Interval A, scale bar 2 mm. Fig. 6 - Amphiura chiajei FORBES. Fully articulated specimen with arms closed around the disc. Interval B (top part), scale bar 2 cm. Fig. 7 - Ebalia nux NORMAN & MILNE EDWARDS. Pyritized carapace steinkern with skeletal remains in dorsal (a) and ventral (b) view. This specimen was fully articulated but appendages were lost during the bulk-sample treatment. Interval A, scale bar 2 mm. Fig. 8 - Brissopsis lyrifera (FORBES). Undeformed muddy steinkern. Interval B, scale bar 1 cm. Figs. 9, 11 - Brissopsis lyrifera (FORBES). Strongly compressed and fragmented tests with attached raised spines. Minor pyritization in Fig. 11. Interval A, scale bars 1 cm. Fig. 10 - Goneplax rhomboides (LINNAEUS). Spectacularly preserved and fully pyritized specimen. Note the eyestalks in the cephalic region (arrows). Interval B, scale bar 0.5 cm. Fig. 12 - Geryon sp. Fully articulated specimen with pyrite infilling in the smaller appendage cavities. Interval A, scale bar 1 cm. Fig. 13 - Funiculina quadrangularis (PALLAS). Long fragment with inner pyrite lining. Interval A, scale bar 1 cm. Fig. 14 - Discospirina italica (COSTA). Complete test with pyritized chamber infilling. Interval A, scale bar 2 mm. Fig. 15 - Brissopsis lyrifera (FORBES). Detail of articulated and raised spines. Interval A, scale bar 0.5 cm. articulated crustacean skeletons and by strongly flatte- ned infaunal echinoid tests, with some diagenetic featu- res, such as pyritization, due to decay of organic matter. Another evidence of obrution events comes from obli- quely embedded and deformed echinoid tests, occasio- nally preserved with articulated and raised spines, which are signs of escape attempts (Pl, 1, Figs. 9, 11, 15). Erosional, reddish silt-bearing surfaces with or without dispersed shells indicate episodic rise of water- energy. Rare, poorly evident, thin lenses of dispersed to densely-packed biogenic elements occur as sedimento- logical concentrations, mainly consisting of hard parts of benthic organisms (e.g., small bivalves and large fora- minifers, Pl. 2, Fig. 3), pteropods (Pl. 2, Fig. 1), terre- strial leaves (Pl. 2, Fig. 6) and Posidonia oceanica (LINNAEUS) leaves (Pl. 2, Figs. 2, 5). Two, 10-20 cm thick, graded beds with erosional base occur in the uppermost part of interval A. They are mud-supported, rhodolith-rich accumulations including large and abra- ded shell fragments of shelf origin, testifying mass depo- sition events. IInntteerrvvaall BB. Fossils are more abundant and better preserved than in the lower interval, but are still disper- sed. The hardparts of epifaunal organisms may be encrusted, moderately bioeroded and locally abraded. The degree of disarticulation is generally low to medium, except for some shell beds in the upper part of the sequence, where bivalves are highly disarticulated and exhibit a moderate degree of fragmentation. Burial events by muddy plumes are numerous and well docu- mented by the same biostratinomic signatures as in the lower interval associated with common pyrite steinkerns (e.g. minute gastropods, bryozoans, foraminifer tests) or patinas (e.g. pyrite coatings of echinoid spines, internal linings of bivalves, etc.). Different genetic types of shell beds occur. Some intrinsic biogenic concentrations, such as the associa- tions of erect bryozoans-small pectinids, Protula worms- celleporiform bryozoans, and Neopycnodonte, record burial episodes of sessile palaeocommunities. Commonly, the appearance of these concentrations is abrupt, without any evident basal erosional surface. In rare cases, the erect-bryozoan shell beds are complex, showing a thin, sharp-based layer consisting of densely packed fragments of delicate branching, mud-tolerant bryozoans, overlain by clumps of erect bryozoans. Below V3 (Fig. 3), the erect bryozoan-small pectinid shell beds alternate with less fossiliferous intervals con- taining loosely packed to dispersed Ditrupa tubes or, more rarely, articulated shells of Corbula (in both cases without preferred orientations). Rare extrinsic biogenic concentrations, produced by scavengers or deposit-fee- ders, were encountered. In other cases the original concentration has been partially distorted by high-energy events (e.g. Hiatella arctica (LINNAEUS) and Aequipecten opercularis-bearing beds), resulting in mixed concentrations. Bivalve remains are densely packed and concordant to bedding, both articulated (closed shells to butterfly position) and disarticulated, complete, and the pectinids are also par- tially encrusted. A single case of diagenetic concentration was encountered about 15 m above the V2 volcaniclastic layer. It consists of a population of Isocardia, whose shells form a loosely packed pavement. This concentra- tion resulted from compression, as indicated by the strong deformation and fragmentation of closed valves. Pure sedimentologic concentrations, formed by storm-induced currents, are represented by small lenses of bioclasts, sharp-based silty layers and, upwards, by discontinuous shell pavements (Fig. 4) and by few com- plex shell beds resulting from multiple events. These ➧ 172 A. D’Alessandro, R. La Perna & N. Ciaranfi thick shell beds, occurring in sandy-silty intervals, mainly consist of Aequipecten opercularis (LINNAEUS) valves which are closely-packed, highly disarticulated, convex-upward, sometimes stacked, broken and encru- sted. Moreover, scattered Arctica islandica (LINNAEUS) and ophiuroid lenses occur in the coarser interval. The bivalves consist of articulated and empty valves, and the Amphiura chjaiei (FORBES) skeletons are preserved per- fectly articulated and with arms closed around the disc (Pl. 1, Fig. 6). 44..22 IInntteerrpprreettaattiioonn In most cases, background and episodic proces- ses can be clearly distinguished (discrete signatures of Speyer & Brett, 1991). However, in the upper interval, the taphonomic features reflect more complex interac- tions, mostly due to higher energy settings (on average) and to more diverse biostratinomic responses to dyna- mics of sedimentation. The biostratinomic signatures in the lower interval allow to infer low-energy background palaeoenviron- ments located well below the maximum storm wave base and characterized by low to modera- te sedimentation rates, soft and, less commonly, soupy substra- tes. This is clear evidence of fre- quent mud blanketing events by deposition of muddy plumes and rarer winnowing episodes by weak bottom currents. The dia- genetic features, particularly the common pyrite linings and rarer pyrite cores, suggest low dysae- robic conditions within the sedi- ment (Brett & Baird, 1986), not always related to burial events. In the upper interval, the biostra- tinomic signatures prevail over the diagenetic ones, thus allowing a greater accuracy when inferring palaeobathyme- try. Low- and high-energy tapho- facies - the latter indicative of settings located above the maxi- mum storm wave base - alterna- te. Deeper settings, where taphonomic signatures are less evident and distinctive, are bet- ter defined on palaeoecological grounds. The higher energy 173Response of macrobenthos to changes ... Fig. 1 - Winnowed concentration, mostly consisting of pteropod shells. The dark spots are plant remains. Interval A, scale bar 2 cm. Fig. 2 - Winnowed concentration of Posidonia oceanica (LINNAEUS) leaves. Interval A, scale bar 1 cm. Fig. 3 - Winnowed concentration of Cyclammina cancellata BRADY tests. Interval A, scale bar 0.5 cm. Fig. 4 - Chondrites targionii (BRONGNIART). Interval A, scale bar 1 cm. Fig. 5 - Leaf fragment of Posidonia oceanica (LINNAEUS). Interval A, scale bar 1cm. Fig. 6 - Quercus sp. leaf. Interval A, scale bar 1 cm. Figs 7, 10 - Cross-sections of Zoophycos spreiten. Interval A, scale bar 1 cm. Fig. 8 - Teredolites isp. in a carbonized wood. Interval A, scale bar 1 cm. Fig. 9 - Chondrites patulus FISCHER-OOSTER. Interval A, scale bar 1 cm. Fig. 11 - Track-like trace. Interval B, scale bar 1 cm. Fig. 12 - Cladichnus isp. (detail of a polished surface). Interval A, scale bar 0.5 cm. Fig. 13 - Burrow filled with shell debris (extrinsic concentration). Interval A, scale bar 0.5 cm. Figs. 14, 15 - Problematica. Interval A, scale bars 0.5 cm. ➧ Fig. 4 – Interval B (upper part). Discontinuous pavements of convex-up valves of Aequipecten opercularis. Some of them are heavily encrusted by corallinaceans and/or serpulids (arrows). taphofacies are indicative of middle and inner shelf set- tings. Numerous winnowing episodes led to firmer sub- strates suitable for the settlement of sessile fast-growing bryozoans and their epibionts, which characterise this community. In few cases, event-concentrations of mud- related delicate bryozoans provided the firm substrate colonized by the erect bryozoans (taphonomic feed- back). In the uppermost part of sequence, taphonomic features suggest inner shelf, maybe transitional to sho- reface, environments. In particular, the Arctica and Amphiura lenses testify reworking of both dead and living benthic organisms and their catastrophic burials during transgression culminating in a marly sediment with Neopycnodonte clumps. The stratigraphic distribu- tion of taphofacies points to clear short-term bathymetric fluctuations, within a general shallowing trend. Levels with abundant pyrite steinkerns within small skeletal cavities suggests episodic dysaerobic microen- vironments due to rapid burial of organic material. 55.. PPAALLAAEEOOIICCHHNNOOLLOOGGYY 55..11 DDeessccrriippttiioonn IInntteerrvvaall AA. This interval is generally bioturbated with a highly variable bioturbation index (nil to mottled). Distinctive trace fossils are visible only when they con- trast in colour with the host rock. Where the sediment is massive, mottling is doubtful, though more than likely. Distribution of ichnotaxa and bioturbation index (BI) are reported in Appendix. All through interval A, a few centi- metres-thick layers, highly bioturbated by Chondrites intricatus (BROGNIART) may occur near silty surfaces with taphonomic evidence of burial events. In the lower part, some intervals up to few metres thick are characterised by dispersed and diverse ichnotaxa (2-4 BI): Chondrites (Pl. 2, Figs. 4, 9), Cladichnus (Pl. 2, Fig. 12) and subor- dinately Alcyonidiopsis are dominant. In the middle part, Chondrites-Zoophycos intervals may alternate with mas- sive (mottled?) beds, or with Phycosiphon-bearing strata (BI 5). Zoophycos (Pl. 2, Figs. 7, 10), which has a discontinuous stratigraphic occurrence, disappears a few metres above the V2 layer (Appendix). Poorly bio- turbated intervals (BI 2 to 3) with small Thalassinoides, Planolites and undetermined track-like traces (Pl. 2, Fig. 11) are also present. In the upper part, the degree of bioturbation is generally lower and ichnotaxa are mostly represented by different Chondrites ichnotaxa. In three cases, Chondrites intricatus is markedly thin and dispersed until its disappearance, which hap- pens in dark, laminated Delectopecten vitreus-rich sedi- ment (Fig. 5). The thickest of these intervals (Fig. 6) coincides with the first occurrence of Gephyrocapsa sp. 3 (upper part of the interval A). IInntteerrvvaall BB. Bioturbation is poorly defined, except for the volcaniclastic beds, where Scolicia (BI 3-5) or Thalassinoides (BI 2-3), due to the highly contrasting colour, are clearly evident. More generally, Thalassinoides and Planolites, sometimes associated with track-like traces, are present, although uncommonly. Bioerosion is abundant in the middle-upper part of the sequence and, more generally, in the shallower settings. 55..11 IInntteerrpprreettaattiioonn Ichnologic features of interval A (i.e., high abun- dance and diversity of chemichnia) point to general dysoxic conditions in the muddy sediment, except for the well oxygenated Phycosiphon-rich massive packa- ge. An increasing oxygen content in the bottom water and/or in the pore water may be inferred by the decrea- se of chemichinia diversity, as well as by the increased size of traces in the middle-upper part of the sequence. Zoophycos has been found in regressive tracts (inferred by the palaeoecological analysis), thus confirming its relation with raising sedimentation rates (Brett, 1998). The thin layers with minute Ch. intricatus that occur throughout the muddy interval, are interpreted as coloni- zation episodes triggered by abrupt burial of organic matter. The aforementioned, laminated, Delectopecten vitreus-rich sediment are an interesting case, since this situation records the exaerobic biofacies of Savdra & Bottjer (1987): i.e. depletion of oxygen and sealing of the mud bottom by biomats allow specialized chemo- symbiotic epifauna to live in an abenthic regime (Savdra, et al., 1991). The bacterial mats might have provided a suitable bottom for the epibyssate life habit of D. vitreus. This hypothesis envisages D. vitreus as a facultative chemosymbiotic organism, although no data are found in the literature on this habit. However, Hickman (1984) reports a deep-water facies, characteri- sed by mud-pectens, in laminated, organic matter-rich sediments. A second hypothesis suggests colonization by opportunistic shelled organisms during short oxic pul- ses (Sageman et al., 1991), but the absence of any kind of infauna remains unexplained. For interval A, an oxygen curve (Fig. 5) has been reconstructed, based on the vertical distribution of oxy- gen-related ichnoassociations. For the D. vitreus interval containing the first occurrence of Gephyrocapsa sp.3, a more detailed curve (Fig. 6) was constructed based also on macrobenthic organisms sensitive to oxygen condi- tions. The ichnologic features of interval B point to fully aerobic conditions. Rare cases of moderate dysaerobic conditions are suggested by the occurrence of abundant Alcyonidiopsis and small concentrations of faecal pel- lets, which may be related to local obrution events (blanketing). 66.. PPAALLAAEEOOEECCOOLLOOGGYY 66..11 DDeessccrriippttiioonn IInntteerrvvaall AA. Most of the invertebrate macrofauna consists of dispersed molluscs and, subordinately, of burrowing echinoids, macroforaminifers, decapods, and octocorals. The most typical bathyal molluscs are Bathyspinula excisa (PHILIPPI) (Pl. 1, Fig. 3), Katadesmia confusa (SEGUENZA), Delectopecten vitreus, Abra longi- callus (SCACCHI) (Pl. 1, Fig. 1), Dentalium agile SARS, and Entalina tetragona (BROCCHI), taxa also known from other coeval bathyal sequences (Di Geronimo & La Perna, 1997, Di Geronimo et al., 1997). In the short basal section, these species are particularly common, but they become more and more dispersed upwards until they are replaced by a monotypic Chondrites intri- catus (BI 4-5) association. In the remaining part of the interval, the strictly bathyal species and their palaeo- communities exhibit a somewhat discontinuous vertical 174 A. D’Alessandro, R. La Perna & N. Ciaranfi 175Response of macrobenthos to changes ... Fig. 5 - Interval A. Palaeobathymetric curve, main palaeocommunities and oxygen curve. Each palae-ocommunity is named after the most typical faunal component(s). Oxygenation levels in the bottom water are according to Bromley (1996, fig. 12.1): Aer = aerobic, Dys = dysaerobic (Upper and Lower), Ex = exaerobic. Log abbreviations: Dv = Delectopecten shell-beds, m = mottling, obr = obrutions (main events), t = turbidites, tl = thin muddy turbidites (main events), w = winnowing (main events). distribution, alternating with taxa characteristic of shal- lower slope facies such as Aporrhais uttingerianus (RISSO) , A. serresianus (MICHAUD), Nassarius cabrieren- s i s (F O N T A N N E S ), Kelliella abyssicola (F O R B E S ), Parvicardium minimum (PHILIPPI), Hyalopecten similis (LASKEY), and Fissidentalium rectum (GMELIN). D. vitreus exhibits a peculiar distribution. It is uncommon throu- ghout interval A, but forms closely packed concentra- tions in three cases (see taphonomy), giving rise to a monotypic palaeocommunity preserved in laminated layers. Shells mostly belong to fully grown individuals. This species belongs to a group of thin-shelled pectinids with a deep-water distribution on muddy bottoms and facultative epibyssate habits. There is a certain syste- matic confusion about D. vitreus (GMELIN, 1791) due to the supposed existence of a distinct species, D. abysso- rum (SARS, 1878), which would have slightly different ecologic needs. Robba (1996) kept these two species distinct, but there is no general agreement about this distinction. In the present paper, both taxa are conside- red as a single mud-tolerant species referred to as D. vitreus. Echinoids are almost exclusively represented by Brissopsis lyrifera (FORBES). Decapods include the bathyal species Ebalia nux NORMAN & MILNE EDWARDS and the eurybathic Goneplax rhomboides (LINNAEUS). In some thin intervals (1-2 m), a monospecific association of the macroforaminifer Discospirina italica (COSTA) (Pl. 1, Fig. 14) occurs as loosely packed concentrations. Bathyal octocorals are recorded by rare remains of Funiculina quadrangularis (PALLAS) (Pl. 1, Fig. 13) and Isidella elongata (ESPER), the latter being locally com- mon. IInntteerrvvaall BB. Fauna from this interval is richer and referable to several palaeocommunities of clearly diffe- rent bathymetric settings. However, changes in palaeo- community composition are mostly gradual, making any clear-cut separation of different palaeocommunities rather difficult. Moreover, the biotic response to environ- mental changes, particularly to edaphic factors, is so marked in shelf settings, that fossil associations highly differ from each other, and it is neither possible to outli- ne general compositional features, nor to describe each community one by one. In this paper, a few case studies 176 A. D’Alessandro, R. La Perna & N. Ciaranfi Fig. 6 - Interval A. Oxygen curve for the Delectopecten vitreus package containing the FO (arrow) of the calcareous nannofossil Gephyrocapsa sp. 3 (sensu Rio, 1983). The curve is based on the vertical distribution of oxygen-related trace fossils and on other fauni- stic and taphonomic features. 177Response of macrobenthos to changes ... Fig. 7 - Interval B. Palaeobathymetric curve and main associations. Each association is named after the most typical faunal compo- nent(s). Log abbreviations: Dv = Delectopecten shell beds, m = mottling, obr = obrutions (main events), t = turbidites, w = winnowing (main events), BS = sequence boundary, TS = transgressive surface. Symbols: sun = interglacial peak, snowflake= glacial peak. 178 A. D’Alessandro, R. La Perna & N. Ciaranfi are reported to characterise the main environments, related to sea-level fluctuations and encompassing diffe- rent system tracts of sequence stratigraphy. A polyspecific middle-inner shelf molluscan palaeocommunity (between V2 and V3) contains Aequipecten opercularis (LINNAEUS), Venus multilamella (LAMARCK), Plagiocardium papillosum (POLI), Turritella ex gr. communis RISSO, T. mediterranea MONTEROSATO, tro- chids (commonly encrusted by bryozoans), and small rhodoliths. Below V5, another similar community is represented by Arctica islandica, Isocardia cor (LINNAEUS), large cardiids, and A. opercularis. Near the top of the sequence, a sharp environmental change is recorded by a middle shelf muddy bottom palaeocom- munity (Turritella ex gr. communis association) abruptly replaced by an inner shelf-shoreface sandy bottom palaeocommunity (A. opercularis-A. islandica associa- tion). The latter is recorded by parautochthonous ele- ments packed in a thick shell bed with complex internal structure, overlying an irregular erosional surface. A Protula-encrusting bryozoan palaeocommunity is referred to outer shelf low-stress biotopes with cohesi- ve substrates and low sedimentation rates. Upwards, near the V5 layers, the Protula-bryozoan palaeocommu- nity evolves into a palaeocommunity characterised by large-sized Hiatella arctica, a weak borer in stiff sedi- ments. This ecological succession culminates with the Neopycnodonte cochlear palaeocommunity. Nassarius edwardsi (P.FISCHER) is an associated species which, in this upper part of the succession, replaces the deeper taxon N. cabrierensis. Another case study are erect bryozoan-small pecti- nid palaeocommu- nities, dominated by adeoniform and r e t e p o r i f o r m growth forms. They alternate with opportunistic soft- bottom palaeocom- munities characte- rised by Ditrupa arietina (MÜLLER) or Corbula gibba (OLIVI). 66..22 IInntteerrpprreettaattiioonn In the Mon- talbano succes- sion, palaeocom- munities and their stratigraphic suc- cessions provide the best tool to identify cyclic sea- level changes and key surfaces of the short-time cyclo- thems. Such chan- ges are recorded throughout the succession. Benthic pa- laeocommunities from interval A are clearly indicative of upper slope envi- ronments, with a maximum depth of about 500 m. Only near the middle part of this interval, the fauna suggests a shelf-margin setting (Fig. 5). Within this bathymetric range, several cyclic fluctuations involving changes in sedimentation rate and substrate consistency, are recor- ded. In these deep water settings, the fluctuations are mainly recognised basing on the relative dominance or abundance of typically bathyal species vs shelf margin and deep shelf species in the associations. Palaeocommunities from interval B record outer- to inner-shelf palaeocommunities, except for a Nassarius cabrierensis-dominated palaeocommunity, located a few metres above the volcaniclastic layer V4, which suggests a transitional-to-slope setting, as is also indicated by the presence of rare bathyal species. Like in the lower interval, palaeoecological evidence indica- tes cyclic sea-level changes (Fig. 7), by a distinct chan- ge of shelf palaeocommunities. The occurrence, in the shallower phases, of the so-called “Boreal Guests” Arctica islandica and Pseudamussium septemradiatum (MÜLLER) together with numerous Hiatella arctica speci- mens, whose size is similar to that of the Recent North Atlantic ones, suggests a climatic control for these fluc- tuations. Sea-level lowstands are recorded by the polyspe- cific molluscan palaeocommunities occurring in thin horizons between V2 and V3 and below V5. Near the top, the lowstand is represented by an erosional surface (TS, Fig. 8) capped by a thick shell-rich bed. During deepening phases, Protula worms-cellepo- riform bryozoan palaeocommunities flourished in outer Fig. 8 – Interval B (top). A = silts with Turritella ex gr. communis. B = simple shell bed with Arctica islandica, Aequipecten opercularis and T. ex gr. communis. SB = sequence boundary. TS = transgressive surface. C = complex shell bed. D = sands with A. islandica and Amphiura chiajei lenses. shelf low-stress biotopes characterised by cohesive substrates and a low sedimentation rate. Near the V5 layers, the ecological replacement starting with faunas indicative of inner shelf environments, culminates with the Neopycnodonte cochlear palaeocommunity that indi- cates the maximum flooding surface. The inversion of the trend up section is pointed out by gradually shal- lower palaeocommunities related to higher sedimenta- tion rates. The Neopycnodonte palaeocommunity is recurrent in interval B and highlights the positions of the maximum flooding surfaces in several cyclothems. Conversely, shallowing phases are highlighted by the Ditrupa arietina or Corbula gibba opportunistic com- munities that flourish in condition of high turbidity (Di Geronimo & Robba, 1989) and high sedimentation rate and inhabit soupy-soft substrates. These conditions are common during the late highstand system tracts. The frequent intercalations within these communities of erect-bryozoan communities (that have contrasting eco- logical needs) can be explained by periodic winnowing that removed the soupy boundary layer and exposed fir- mer substrates suitable for such sessile communities of fast-growing organisms. The vertical succession of com- munities confirms the upwards shallowing trend. 77.. DDIISSCCUUSSSSIIOONN Changes in the taphonomic and palaeoecological features allowed to infer a sea-level curve for the Early- Middle Pleistocene (Figs. 5, 7). The curve is composed of short-term fifth- (100 Ka) and sixth-order (40Ka) cycles, as evaluated from the local oxygen isotope curve (Brilli in D’Alessandro et al., 2002). The ba-thymetric range of cycles is about 100 m, although in so-me cases it considerably exceeds (over 200 m) this value. Throughout interval B, the climatic control is pro- ved by the occurrence of “cool water” molluscs indica- tors in the shallower phases and, occasionally, by “warm water” indicators in the deeper ones, i.e. two teleosteans (Girone & Varola, 2001) and a serpulid (Sanfilippo, in press). Conversely, no palaeoclimatic indicator occurs in the lower interval. However, there is a good match between the sea level curve and the oxy- gen isotope curve, which at present does not cover all the sequence. The same type of control can be then supposed for the whole sequence of Montalbano. However, in some intervals, the bathymetric range is too high to be exclusively ascribed to climatic effects; therefore, a tectonic control must be an additional factor. The best example occurs near the volcaniclastic layer V4, where the bathymetric range exceeds 200 m and the kind of palaeocommunities - i.e., palaeocommunities dominated by attached forms preserved as loosely to closely packed concentrations versus palaeocommuni- ties of mostly vagile elements, occurring as dispersed assemblages - are indicative of a low and high sedimen- tation rate respectively. The high sedimentation rate events occur during abrupt deepening phases (as infer- red by the palaeocommunity succession) or follow them, thus contrasting with the expected sedimentation pat- tern during the transgressive system tract. This can be attributed to uplift of the continental areas surrounding the foredeep basin and resulting in increased terrige- nous supply. Hence, the “cycle” is exalted in magnitude and broken down in tectonically controlled steps. In interval A, palaeobathymetric ranges of sea level fluctuations are less detailed due to the intrinsic features of both deep-water fauna and biotopes. Anyway, in analogy of the upper interval, the largest cycles are also regarded as partially tectonic-controlled. The ichnological assemblages from the lower interval allowed to draw a curve of changes in the bot- tom water oxygen content (Figs. 5, 6). It is worth noti- cing that the dysaerobic/anaerobic events occur during maximum flooding phases, as a result of the decreased oxygen supply to the bottom water during interglacial periods. The present study also allows to elucidate the palaeoecological significance of two types of associa- tions hitherto unknown in the literature, i.e. the Delectopecten vitreus and the Discospirina italica com- munities. In the studied sequence, D. vitreus is com- monly present in muddy bottom, bathyal palaeocommu- nities, as is known in the literature (D’Alessandro & De Marco, 1993; Robba, 1996; Di Geronimo & La Perna, 1997), but this species, when recorded in monotypic palaeocommunities, seems to be related to exaerobic conditions, as suggested by ichnofossil analysis. The “exaerobic model” may be thus adequate to explain the absence of all kind of infauna, including the Chondrites- makers. This hypothesis would entail a facultative che- mosymbiotic feeding mode for D. vitreus. Discospirina italica is an unusual deep-water macroforaminifer (Adams, 1973, 1976; Hottinger, pers. comm., 2001), whose ecology is poorly known. Monotypic associations of Discospirina, in the studied sequence, have been found during the early highstands, in slope (interval A) and outer shelf/slope settings, cha- racterising an increased sedimentation rate related to shallowing. There is an overall good match between the infer- red cycles and those obtained from other palaeontologi- cal studies, although at different degree of resolution. Benthic foraminifers provide a detailed reconstruction of bathymetric changes in particular for the upper interval, while the same group allows a better resolution of oxy- gen changes through the interval A (Stefanelli, in press), giving a curve more or less equivalent to that inferred by the ichnofossils. Fish otoliths are a fairly good bathyme- tric indicators in the deeper settings (Girone, 2000). Tapho-facies analysis was particularly useful in shal- lower environments (Soldani, 2000; Ciaranfi et al., 2001). The integrated palaeontological analyses applied to the Montal-bano sequence support the statement by Brett (1998) that fossils provide a good tool to identify key surfaces and sedimentation dynamics within se- quences, while the sequence stratigraphy model provi- des a predictive framework to interpret probable causes of biotic changes. AACCKKNNOOWWLLEEDDGGMMEENNTTSS We thank the referees B.M. Cita (Milano University) and F.T. Fürsich (Würzburg University) for improving of the manuscript and helpful comments, and A. Girone (Bari University) for determination of crusta- ceans. The authors were funded by the Italian Ministry of University and Scientific Research (grant MIUR 40%, prof. N. Ciaranfi). 179Response of macrobenthos to changes ... RREEFFEERREENNCCEESS Adams C.G., 1973 - Some Tertiary Foraminifera. In: Hallam A. (ed.), Atlas of Palaeobiogeography. Elsevier, Amsterdam, p. 455-468. Adams C.G., 1976 - Larger Foraminifera and the Late Cenozoic history of the Mediterranean region. Palaeogeography, Palaeoclimatology, Palaeoeco- logy, 2200, p. 47-66. Basso D. & Corselli C., 2002 - Community versus bio- coenosis in multivariate analysis of benthic mollu- scan thanatocoenoses. Rivista Italiana di Paleontologia e Stratigrafia, 110088: p. 153-172. Brett C.E., 1998 - Sequence Stratigraphy, Paleoeco- logy, and Evolution: Biotic Clues and Responses to Sea-Level Fluctuations. Palaios, 1133, p. 241-262. Brett C.E. & Baird G.C., 1986 - Comparative taphonomy: a key to paleoenvironmental interpretation based on fossil preservation. Palaios, 11: p. 207-227. Bromley R.G., 1996 - Trace fossils. Biology, Taphonomy and applications. Chapman & Hall, p. 361. Ciaranfi N., D’Alessandro A. & Marino M., 1997 - A can- didate section for the Lower-Middle Pleistocene boundary (Apennine Foredeep, South Italy). In: W. Naiwen and J. Remane (Eds), Proceedings ot the 30th International Geological Congress, 1111, p. 201-211. Ciaranfi N., D’Alessandro A., Girone G., Maiorano P., Marino M., Soldani D. & Stefanelli S., 2001 - Pleistocene sections in the Montalbano Jonico area and the potential GSSP for Early-Middle Pleistocene in the Lucania Basin (Southern Italy). Memorie di Scienze Geologiche dell’Università di Padova, 5533, p. 67-83. Ciaranfi N., Marino M. Sabato L., D’Alessandro A. & De Rosa, 1996 - Studio geologico-statrigrafico di una successione infra e mesopleistocenica nella parte sud-occidentale della Fossa bradanica (Montalbano Jonico, Basilicata). Bollettino della Società Geologica Italiana, 111155, p. 379-391. D’Alessandro A. & De Marco A., 1993 - Bionomic analy- sis of two Upper Pliocene communities from Southern Italy. Beringeria, 88, p. 109-137. D’Alessandro A., Brilli M., Ciaranfi N., Girone A., La Perna R., Maiorano P., Marino M., Soldani D., Stefanelli S. & Turi B., 2002 - Lower-Middle Pleistocene sea-level changes in the Lucania basin (Southern Italy) as recorded by fossil asso- ciation. In: Ciaranfi N. (Ed.), Abstract Convegno AIQUA “Elementi cronostratigrafici per una corre- lazione fra successioni marine e successioni conti- nentali: stato delle conoscenze e nuovi dati, p. 17- 18. Di Geronimo I., D’Atri A., La Perna R., Rosso A., Sanfilippo R. & Violanti D., 1997 - The Pleistocene bathyal section of Archi (Southern Italy). Bollettino della Società Paleontologica Italiana, 3366, p. 189- 212. Di Geronimo I. & La Perna R., 1997 - Pleistocene bathyal molluscan assemblages from Southern Italy. Rivista Italiana di Paleontologia e Stratigrafia, 110033, p. 389-426. Di Geronimo I. & Robba E., 1989 - The structure of benthic communities in relation to basin stability. Memorie dell’Accademia dei Lincei, 8800,, p. 341-352. Girone A., 2000 - The use of fish otoliths for paleobathy- metric evaluation of the lower to Middle Pleistocene deposits in Southern Italy. Bollettino della Società Paleontologica Italiana, 3399 (2): 235- 242. Girone A. & Varola A., 2001 - Fish otoliths from the Middle Pleistocene deposits of Montalbano Jonico (Southern Italy) . Bollettino della Società Paleontologica Italiana, 4400, p. 431-443. Hickman C.S., 1984 - Composition, structure, ecology and evolution of six Cenozoic deep-water mollusk communities. Journal of Paleontology, 5588 (5): p. 1215-1234. Kidwell S.M., Fürsich F.T., Aigner T., 1986 - Conceptual Framework for the Analysis and Classification of Fossil Concentrations. Palaios, 11, p. 228-238. Maiorano P., Marino M., Di Stefano E. and Ciaranfi N., submitted - Calcareous nannofossil events in the Lower-Middle Pleistocene transition at Montalbano Jonico section (Southern Italy) and ODP Site 964 (Ionian Sea), and their calibration with oxygen iso- tope and sapropel stratigraphy. Rivista Italiana di Paleontologia e Stratigrafia. Pérès J.M. & Picard J., 1964 - Nouveau manuel de bio- nomie benthique de la Mer Méditerranée. Recuil des Travaux de la Station Marine d’Endoume, 3311 (47), pp. 137. Rio D., 1982 - The fossil distribution of Coccolithophore Genus Gephyrocapsa Kamptner and related Plio- Pleistocene Chronostratigraphic problems. In: Prell, W.L., Gardner, J. V. et al. (Eds.), Initial Reports of the Deep Sea Drilling Project, 6688, p. 325-343. Robba E., 1996 - Autoecology of some Pliocene thin- shelled pectinids. In Cherchi A. (ed): Autoecology of selected fossil organisms: Achievements and problems, Bollettino della Società Paleontologica Italiana, spec. vol. 33, p. 159-174. Sageman B.B., Wignall P.B., Kauffman E.G., 1991 - Biofacies models for oxygen-deficient facies in epi- continental seas: tool for paleoenvironmental analysis. In: Einsele G., Ricken W., Seilacher A., (Eds.): Cycles and Events in Stratigraphy Springer-Verlag, Berlin, 542-564. Sanfilippo R. (in press) - Climatic response in the genus Plagostegus Philippi, 1844 from Plio-Pleistocene to Recent Mediterranean, with description of a new species (Polychaeta, Serpulidae). Bollettino della Società Paleontologica Italiana. Savdra C.E. & Bottjer D.J., 1987 - The exaerobic zone, a new oxygen-deficient marine biofacies. Nature, 327: p. 54-56. Savdra C.E., Bottjer D.J., and Seilacher A., 1991 - Redox-Related Benthic Events. In Einsele G., Ricken W., Seilacher A. (Eds.): Cycles and Events in Stratigraphy Springer-Verlag, Berlin, p. 524-541. Soldani D., 2000 - Utilizzo e limiti di applicabilità delle tafofacies nelle analisi di bacino. Tesi di Dottorato, Università di Bari, unpublished, pp. 114. Speyer S.E. & Brett C.E., 1991 - Taphofacies controls. In: Allison P.A. & Briggs D.E.G. (Eds.) Taphonomy: Releasing the Data Locked in the Fossil Record. Vol. 99 of Topics in Geobiology, , p. 501-545. Stefanelli S. (in prep.) - Early-Middle Pleistocene benthic foraminifera microhabitats as response to oxygen variations: proxy of oxygen content? 180 A. D’Alessandro, R. La Perna & N. Ciaranfi 181Response of macrobenthos to changes ... A p p e n d ix 182 A. D’Alessandro, R. La Perna & N. Ciaranfi A p p e n d ix A p p e n d ix - B io tu rb a tio n in d e x (B I) a n d d is tr ib u tio n o f ic h n o ta xa in in te rv a l A . - In d ic e d i b io tu rb a zi o n e ( B I) e d is tr ib u zi o n e d e g li ic h n o ta xa n e ll’ in te rv a llo A