Imp.Munno-Petrosino NNEEWW CCOONNSSTTRRAAIINNTTSS OONN TTHHEE OOCCCCUURRRREENNCCEE OOFF YY--33 UUPPPPEERR PPLLEEIISSTTOOCCEENNEE TTEEPPHHRRAA MMAARRKKEERR LLAAYYEERR IINN TTHHEE TTYYRRRRHHEENNIIAANN SSEEAA RRoossaallbbaa MMuunnnnoo11 && PPaaoollaa PPeettrroossiinnoo11 1Dipartimento di Scienze della Terra dell'Università di Napoli Federico II L.go S. Marcellino 10, 80138 Napoli ABSTRACT: Munno R. & Petrosino P., New constraints on the occurrence of Y-3 Upper Pleistocene tephra layer in the Tyrrhenian Sea. (IT ISSN 0394 – 3356, 2004). A widespread tephra layer, other than the well known Y-5, has been identified in the Upper Pleistocene marine succession in the Tyrrhenian Sea. Two investigated gravity cores showed, in fact, the presence of two companion pyroclastic tephra layers, separated by a varying thickness of pelagic sediments. The pyroclastic layers are mainly made up of pumice fragments and glass shards together with few K-feldspars and clino-pyroxene crystals. Both layers are alcali-trachytic in composition, even though a sharp difference emer- ges in the K/Na ratio that characterizes the two glasses. 14C dating of foraminiferous shells embedded in the clay layers directly underlying the most recent tephra gave an age of about 26 ka. An accurate review of literature regarding tephrostratigraphy in the Mediterranean area made it possible to correlate the older one to the Y-5 marker layer, joined to the Campanian Ignimbrite eruption, a paroxystic event in the Campi Flegrei area. The younger layer has been correlated with the Y-3 marker layer and probably represents another huge pyroclastic flow event from the Campanian area, whose products have not yet been distinguished in the field from those of typical Campanian Ignimbrite. This work clearly identifies the layer Y-3, firstly recorded by Keller et al. (1978), as the result of a speci- fic volcanic event different from the Campanian Ignimbrite (marker layer Y-5), defines its mineralogical and chemical composition together with its relative age offering an useful support for paleoclimatic and paleoenvinromental reconstruction of the sedimentation in the Tyrrhenian area. RIASSUNTO: Munno R. & Petrosino P., Caratterizzazione di Y-3, un tefra per il Pleistocene superiore nel Mar Tirreno. (IT ISSN 0394 – 3356, 2004). Nella successione marina del Pleistocene Superiore, nel basso Tirreno, è stato identificato e caratterizzato un livello di tefra diverso dall'Y-5, che è il livello marker in genere associato con la grossa eruzione dell' Ignimbrite Campana. Molti gravity cores analizzati hanno infatti mostrato la presenza di una doppietta di livelli piroclastici, separati da uno spessore variabile di sedimenti pelagici. I livelli di tefra sono entrambi costituiti da frammenti di pomici e shard vetrose con una minima percentuale di cristalli di k-feldspato e clinopirosseno. Entrambi i livelli mostrano un chimismo alcalitrachitico, ma sono caratterizzati da una netta differenza nel rapporto K/Na. Datazioni 14C ottenute su gusci di foraminiferi prelevati all'interno dell'argilla alla base del tefra più recente hanno fornito un'età di circa 26 ka. Una revisione accurata della letteratura riguardante la tefrostratigrafia nell'area mediterranea ha permesso di correlare il più antico con il marker Y-5, che rimanda all'eruzione dell'Ignimbrite Campana, un evento parossistico proveniente dall'area flegrea. Il livello più recente è stato correlato con il marker Y-3 e, probabilmente, rappresenta un altro evento di rilievo dell'area campana, i cui prodotti, sul campo, non sono stati ancora distinti da quelli dell'Ignimbrite Campana. Il lavoro identifica in maniera definitiva il livello Y-3, individuato per la prima volta da Keller et al. (1978), come il risultato di un evento specifico diverso dall'Ignimbrite Campana (marker layer Y-5), ne defini- sce la composizione mineralogica e chimica, l'età relativa e, mappandone la distribuzione, offre un utile supporto per la ricostruzione paleoclimatica e paleoambientale della sedimentazione nell'area mediterranea. Keywords: gravity core, tephra layer, tephrocronology, Campanian Ignimbrite Parole chiave: carote, livello di tefra, tefrocronologia, Ignimbrite Campana Il Quaternario Italian Journal of Quaternary Sciences 1177(1), 2004, 11-20 11.. IInnttrroodduuccttiioonn During the last 40000 years several violent volca- nic eruptions took place in the Campanian volcanic pro- vince, southern Italy. In particular, the main volcanic sources in the area for that time-span are Campi Flegrei and Somma-Vesuvio. Although a good knowledge of their activity is now available, and the products are quite well mapped in areas proximal to the sources, some dif- ficulty however persists in the identification of the origin of tephra layers when they are found interbedded to sedimentary sequences in areas many kilometres far from the volcanoes. This is the particular field of interest of tephrostratigraphy and tephrochronology that, together with a detailed biostratigraphy, make it possible to identify marker horizons useful for geological and geomorphological reconstructions. Starting from about 40 ka B.P. the best represen- ted Campanian tephra layer in the Central Mediterranean is Y-5, so labelled according to the nomenclature of tephra layers in the Eastern Mediterranean of Keller et al. (1978). Although a great deal of testified occurrences of this pyroclastic layer are available, no clear attribution to a source has been made until recent. Firstly, Keller et al. (1978), in fact, attributed it to the Ischia Citara-Serrara Fontana Formation, even if this formation, which is the product of several small-volume eruptions, is not widely dispersed in the field (Vezzoli, 1988). Shortly following this attribu- tion, Thunnel et al. (1979) definitely correlated the Y-5 tephra layer with the Campanian Ignimbrite, recorded in the field as a widespread pyroclastic flow deposit (Barberi et al., 1978; Fisher et al., 1993; Rosi et al., 1996) joined to a thick pyroclastic fall layer emplaced 12 R. Munno & P. Petrosino towards the East from the Campi Flegrei source area (Rosi et al., 1999). Most of the recent tephrostrati- graphic studies agree with a Campanian Ignimbrite ori- gin for the Y-5 layer; only Paterne et al. (1986, 1988), in fact, still ascribe the tephra to the Citara-Serrara Fontana Formation of the Ischia Island. Actually, the attribution of Y-5 to the Campanian Ignimbrite rises also some chronological problems. The deposit has been dated by single crystal high precision 40Ar/39Ar method to 37.1±0.4 ka B.P. (Deino et al., 1992,1994), but new 40Ar/39Ar geochronology defines the age of the Campanian Ignimbrite at 39.28±0.11 ka, about 2000 years older than the previous estimate (De Vivo et al., 2002). A great deal of 14C dates on paleosols and charbonized wood, available in literature (Alessio et al., 1971; 1973; 1974), however, gave an age scattering between 25 and 40 ka B.P. Two K/Ar dates (Curtiss, 1966; Cassignol and Gillot, 1982) gave an age of 30 and 34.5 B.P. ka, respectively. According to this data some authors believe that the Campanian Ignimbrite must not be considered as a single eruptive episode, entailing the emplacement of a single pyroclastic flow deposit, but as the result of a series of eruptive episo- des, separated by significant temporal hiatuses (Scandone et al., 1991; Calderoni et al., 1993), that brought to the emplacement of the ignimbrite deposit observed in Campania. More recently, De Vivo et al., 2002, identify on the Campanian Plain volcanic events other than the Campanian Ignimbrite, displaying similar field features. The hypothesis of the existence of more than a single volcanic event representing the Campanian Ignimbrite l.s. is supported by the occurren- ce, in the tephrostratigraphical record of the Central Mediterranean area, of more than one ash layer in the 25-40 ka time-span, attributed by Paterne et al. (1988) to a "Campanian Ignimbrite Series". A second Campanian Tephra layer has never been recorded along the Tyrrhenian coast, but only detected in some deep sea cores in the Ionian Sea. The layer, named Y-3, displays an extrapolated age of about 25 ka B.P. (Narcisi & Vezzoli, 1999), is trachytic in composition and has generically been attributed by Keller et al. (1978) to the Campanian Province. Paterne et al. (1988), on the contrary, ascribed it to the "Campanian Ignimbrite Series". A late glacial Campanian marker tephra layer, labelled C-2 in Paterne et al., 1988, has been recorded in the Central Mediterranean and correlated with the 12 ka B.P. Campi Flegrei eruption of Neapolitan Yellow Tuff. The main Holocene Campanian tephra layers are linked to the Somma-Vesuvio activity. The most diffused marker layers are Z-1, correlated to the Somma-Vesuvio Avellino eruption (ca 3800 B.P., Rolandi et al., 1998) and the tephra related to the A.D. 79 plinian eruption. They both feel the effect of the different direction of emplacement of fall products, so Z-1 is mainly recorded in the East Mediterranean, whereas A.D. 79 tephra is better represented in the Central Mediterranean. The present paper investigates in detail the tephra interbedded to the sedimentary records in two gravity cores of the southern Tyrrhenian Sea. The aims of the research are: - to definitely identify the presence of a single or of more than one tephra layer of Campanian origin in the age range 40-25 ka B.P., that encompasses all the 14C dating obtained on paleosols underlaying Campanian Ignimbrite deposits, - to point out the distribution of the tephra layers and, eventually, to hypothesize a source for the pyroclastic products. In this regard, the work investigates gravity cores located near the Campanian Tyrrhenian coastal area, since the emplacement of Campi Flegrei and Vesuvio products is more probable in this area, and uses sedi- mentological and biostratigraphic data as a support to chronologically locate the identified tephra layers. The present work is part of a multidisciplinary research that involved, through the last five years, geologists, sedi- mentologists, palinologists, palaeontologists and volca- nologists, aiming to investigate environmental and cli- matic evolution of the southern Tyrrhenian Sea through high resolution records. The results of these researches are here reported only when useful to argue tephrostra- tigraphical conclusions. More detailed information about strictly sedimentological and palaeontological aspects can be found in Buccheri et al. 2002a,b. Few chemical data on glass shards and preliminary tephrostratigraphi- cal considerations have been reported by one of the authors (R.M.) in Buccheri et al. 2002a. The present paper represents a step forward, using new tephra and mineral characterization on the two cores and on-land comparison samples, and revising the Upper Pleistocene tephrostratigraphical record in the whole Tyrrhenian area. 22.. EEXXPPEERRIIMMEENNTTAALL 22..11 TTeepphhrraa llaayyeerrss iinn tthhee ccoorreess Two gravity cores located offshore the Campanian Tyrrhenian border have been investigated to reconstruct the tephrostratigraphical record of the area. The first core, named C106 (ϕ = 40°29'32'', λ = 14°42'32'', depht = 292 m) is located in the Salerno Gulf, directly offshore the Sele River mouth, and is 6,40 m long. The second core, named C45 (ϕ = 39°45'10'', λ = 15°20'20'', depht = 1053 m) is located in the Policastro Gulf, in a valley-type depression of the sea floor, about 63 nautical miles SW of Sapri town, and is 4.70 m long. Figure 1 shows the location of investigated cores together with main deep sea morphological contours. Both the cores consist of an alternance of clay, silt, silt-sandy and rare sandy layers, in which the coar- se fraction is made up of organic (planktonic and bento- nic Foraminifers, Pteropods) and inorganic material (pumice fragments, quartz and mica crystals). Along the cored successions some coarse grained layers, mainly made up of sharp edged pumice fragments and crystals, displaying features of a primary volcanic deposition, have been fully investigated to reconstruct the complete tephrostratigraphical record (Fig.2). Along the core C106 the first tephra layer has been identified at 55.5 cm from the top and its lower limit is 110.5 cm. The whole thickness is 55 cm. The layer shows a sharp basal contact with the underlying clay layer. The basal part of the pyroclastic layer (110.5- 88.5 cm) consists of whitish pumice fragments (sample A1(a)), overlain by a 10cm thick layer consisting of gray pumice and rare lava lithic fragments (sample A1(b)). Both the samples display maximum clasts grain-size greater than 2mm and contain pumice fragments, a minor fraction of lava lithic fragments and rare limesto- ne fragments. The second tephra layer of C106 has been identi- fied at 565 cm from the top of the core and its lower limit is 579 cm. The whole thickness is 14 cm. The layer does not show a sharp basal contact with the underlying clay layer. The layer consists of some sublayers contai- ning white and gray pumice fragments (sample A2), and passes towards the top to a dark clay level. All samples display maximum clasts grain-size greater than 2mm and contain pumice fragments and rare lava lithic frag- ments. Juvenile fraction in both samples was represented by elongated pumice clasts, with very well developed elongated vesicles. Along the C45 core three main tephra layers have been recognized. The first one has been identified at 25 cm from the top of the core and its thickness is about 7 cm. The layer is mainly made up of grey pumice frag- ments - sample B1(b), maximum grain size 1 mm and rare lava and limestone lithic fragments, but in the basal part whitish pumice fragments are abundant - sample B1(a). The second tephra layer of C45 core is located at 380 cm from the top and is 3 cm thick. It is made up of pumice fragments and glass shards, maximum grain 13New constraints on the occurrence ... size of clasts is 2 mm; no lithic fraction has been detec- ted (sample B2). The third tephra layer is 460 cm deep and 3.5 cm thick. Juvenile fraction is represented by glass shards and pumice fragments whose maximum grain-size is 1 mm; very rare lava lithic fragments have also been found (sample B3). Mineralogical component percentage evaluation has been performed by counting at least 100 crystals for each level; the results of the analysis are shown in Table 1. K-feldspar, biotite and clinopyroxene are almo- Fig.1 - Map of the Salerno and Policastro Gulfs with main submarine morphological contours and location of investigated cores. Carta batimetrica dei Golfi di Salerno e Policastro con l'ubicazione dei gravity cores studiati. Table 1 - Mineralogical component percentage evaluation for all investigated tephra layers. Valutazione semiquantitativa delle fasi minerali presenti all'in- terno dei livelli di tefra. core sample feldspar leucite biotite cpx garnet Fe-Ti oxides A1 (a) •• •• • C106 A1 (b) •• • • • A2 ••• • • B1 (a) •• •• • C45 B1 (b) •• • • • B2 •• • • • B3 •• • • • • l represents about 20% 14 st ubiquitous phases. Only the topmost layer of both cores shows leucite and garnet among the mineral pha- ses. Chemical composition of glass fragments and cry- stal grains was determined by microprobe analysis using E.D.S. system on loose fragments of 0.5 mm grain-size, at CISAG (Centro Interdipartimentale di Servizio per Analisi Geomineralogiche) at University of Naples Federico II. With regards to the chemical com- position of glasses, the reported one represents a mean composition, obtained from at least ten punctual analy- ses on different glass shards. Chemical composition of all the samples is reported in Table 2. As a reference for correlations with the products of volcanic eruptions recorded in the field, Table 3 reports the representative analyses of mineral grains for C45 core, chosen because it contains all the recognized tephra layers. The whole set of analytical data on mine- rals, both for tephra layers in the cores and for compari- son samples, are at disposal upon request to the authors. 22..22 CCoommppaarriissoonn aannaallyysseess In order to better clarify the possible correlations, chemical analyses other than on the C106 and C45 pyroclastic layers have been performed. The compari- son samples have been selected among the most wide- spread Campi Flegrei and Somma Vesuvio eruptions products ranging in age between about 40 and the pre- sent. The age range has been chosen taking into account the 14C AMS dating that provided the chrono- stratigraphic framework for both cores, as reported in Buccheri et al. (2002a,b) and in Figure 2. The first comparison sample is represented by pumice fragments extracted from the pyroclastic fall layer directly underlaying the Ignimbrite Campana grey tuff at Cologna, NE of Salerno. The other comparison samples are represented by the Somma-Vesuvio Codola and Pomici di Base erup- tions products (Rolandi, 1997), samples courtesy of prof. Giuseppe Rolandi. Apart from this, to better define the characteristics of Y-3, prof. Keller has been asked a little sample of the Fig. 2 - Stratigra- phic succession of C45 and C106 core with interbedded tephra layers and thickness of tephra layers. *data from Buccheri et al., 2002a;**data from Buccheri et al., 2002b; nc = not calibrated. Successione strati- grafica delle carote C45 and C106 con indicazione dei livelli di tefra inter- calati e dei relativi spessori. *dati da Buccheri et al., 2002a;**dati da Buccheri et al., 2002b; nc = età non calibrata. R. Munno & P. Petrosino 15 layer firstly ascribed to Y-3, from the core RC9-191 of the Ionian Sea. The sample has been analysed using the same methods as the unknown layers object of the paper. The analysis of Keller et al. (1978) had, in fact, been performed by X-ray fluorescence on a glass enri- ched sample and, at present, it is well known (Rinaldi, 1979; Narcisi & Vezzoli, 1999) that punctual analytical methods (EDS, WDS, etc) best suit tephrostratigraphy- cal problems. The last comparison sample is split into two sub- samples respectively of white and grey pumice frag- ments from the A.D. 79 eruption, extracted from the Poggiomarino and Zabatta outcrops. It has been analy- sed since its south-west trending diffusion makes its presence expected in the Salerno Gulf recent sedimen- tary record. Table 4 reports SEM-EDS glass compositions for Table 2 - SEM-EDS glass composition recalculated to 100 water free of all the investigated tephra samples. n= number of analyses; standard deviation in parentheses. Composizione SEM-EDS del vetro ricalcolata a 100 per i livelli di tefra esaminati. n= numero di analisi; deviazione standard in parentesi. sample C106 C106 C106 C45 C45 C45 C45 A1(a) A1(b) A2 B1(a) B1(b) B2 B3 n 11 10 21 10 12 11 11 SiO2 55.36 (0.27) 56.54 (0.19) 62.63 (0.20) 54.86 (0.24) 56.04 (0.32) 62.41 (0.32) 61.53 (0.48) TiO2 0.48 (0.05) 0.24 (0.03) 0.34 (0.06) 0.45 (0.08) 0.55 (0.07) 0.33 (0.10) 0.39 (0.06) Al2O3 21.01 (0.17) 22.72 (0.28) 18.24 (0.12) 20.74 (0.20) 21.36 (0.14) 18.30 (0.29) 18.73 (0.17) FeO* 4.07 (0.16) 2.36 (0.21) 3.15 (0.12) 3.93 (0.11) 3.39 (0.12) 3.17 (0.10) 3.16 (0.19) MnO 0.17 (0.08) 0.13 (0.11) 0.15 (0.08) 0.13 (0.04) 0.09 (0.03) 0.15 (0.08) 0.23 (0.07) MgO 0.59 (0.06) 0.11 (0.06) 0.47 (0.08) 1.38 (0.09) 0.48 (0.08) 0.50 (0.09) 0.50 (0.10) CaO 4.23 (0.29) 2.98 (0.15) 2.12 (0.10) 4.47 (0.24) 4.06 (0.31) 2.13 (0.10) 1.74 (0.06) Na2O 5.83 (0.21) 7.39 (0.64) 4.15 (0.19) 5.37 (0.17) 7.13 (0.21) 4.30 (0.15) 6.05 (0.33) K2O 8.28 (0.44) 7.54 (0.76) 8.75 (0.14) 8.67 (0.29) 6.90 (0.30) 8.71 (0.08) 7.67 (0.21) D.I. 79.56 86.03 86.82 77.40 81.75 86.95 89.53 Ne 22.10 24.89 0.00 22.78 22.88 1.07 8.24 FeO* = total Fe as FeO FeO* = total Fe as FeO Table 3 - Representative microprobe analyses of mineral phases extracted from C45 core tephra layers. Analisi in microsonda dei minerali rappresentativi estratti dai livelli di tefra della carota C45. leucite k-feldspar cpx cpx biotite amphibole SiO2 54.50 63.48 51.18 43.96 36.00 33.68 TiO2 0.11 0.10 0.43 1.55 3.53 1.89 Al2O3 22.76 18.81 2.36 8.30 15.50 15.21 FeO* 0.37 0.09 4.65 11.05 11.65 24.84 MnO 0.02 0.05 0.12 0.12 0.08 1.02 MgO 0.08 0.18 16.22 10.26 18.03 4.30 CaO 0.12 0.33 23.85 23.58 0.00 11.55 Na2O 0.83 1.23 0.44 0.46 0.49 1.47 K2O 20.51 15.03 0.00 0.00 10.33 3.37 Total 99.30 99.30 99.25 99.28 95.61 97.33 B1 (a-b) B2 k-feldspar cpx biotite magnetite SiO2 63.60 49.31 35.81 0.29 TiO2 0.13 0.57 4.56 6.23 Al2O3 18.88 3.77 14.50 2.56 FeO* 0.23 8.31 14.49 80.91 MnO 0.03 0.36 0.23 0.87 MgO 0.13 13.53 15.41 1.58 CaO 0.55 23.00 0.00 0.06 Na2O 1.84 0.53 0.56 0.10 K2O 13.95 0.00 9.84 0.04 Total 99.34 99.38 95.40 92.64 B3 k-feldspar cpx biotite amphibole magnetite SiO2 63.58 49.01 35.40 38.15 0.36 TiO2 0.24 0.49 4.79 1.94 5.22 Al2O3 18.87 2.97 15.06 12.20 4.11 FeO* 0.27 11.28 13.89 19.54 79.73 MnO 0.02 0.71 0.17 0.48 0.57 MgO 0.07 11.31 15.70 8.83 2.58 CaO 0.48 22.83 0.00 11.84 0.00 Na2O 1.82 0.79 0.53 1.57 0.40 K2O 14.12 0.03 10.02 2.77 0.00 Total 99.48 99.42 95.56 97.32 92.97 New constraints on the occurrence ... 16 the comparative samples. Figure 3 re- ports the TAS classi- ficative plot (Le Maitre, 1989) for the investigated tephra layers and the com- parison samples, and Figure 4a reports the DI-Ne plot of Armienti et al. (1983), more apt to the slighty saturated and alkali-rich rocks of Campanian area. 33.. RREESSUULLTTSS AANNDD DDIISSCCUUSSSSIIOONN The modal and chemical composi- tions of pyroclastic layers, identified in the two gravity cores, show two compositional fields, phonolitic s.l. and alkali-trachytic. The presence of a phonolitic layer Table 4 - SEM-EDS glass compositions recalculated to 100 water free for the comparative samples. n= number of analyses; standard deviation in parentheses. Composizione SEM-EDS del vetro ricalcolata a 100 per i campioni di riferimento. n= numero di analisi; deviazione standard in parentesi. 1 2 3 4 5 6 7 8 n 10 XRF 10 10 8 11 9 15 SiO2 61.52 (0.20) 61.78 62.76 (0.36) 57.87 (0.30) 62.45 (0.08) 56.46 (0.25) 54.88 (0.21) 61.01 TiO2 0.37 (0.06) 0.44 0.36 (0.03) 0.58 (0.08) 0.35 (0.03) 0.42 (0.05) 0.44 (0.03) 0.40 Al2O3 18.88 (0.15) 18.96 18.21 (0.16) 19.16 (0.20) 18.34 (0.08) 21.35 (0.18) 20.67 (0.11) 18.63 FeO* 3.06 (0.12) 3.23 2.98 (0.21) 3.90 (0.16) 2.74 (0.14) 3.34 (0.17) 4.09 (0.16) 3.55 MnO 0.23 (0.05) 0.17 0.17 (0.12) 0.16 (0.06) 0.13 (0.09) 0.21 (0.10) 0.23 (0.04) 0.05 MgO 0.47 (0.05) 0.59 0.25 (0.12) 0.63 (0.11) 0.35 (0.04) 0.56 (0.09) 0.67 (0.08) 0.77 CaO 1.75 (0.08) 1.98 2.07 (0.15) 4.43 (0.46) 2.75 (0.09) 4.05 (0.28) 5.03 (0.17) 2.47 Na2O 6.01 (0.24) 5.57 4.44 (0.19) 3.30 (0.27) 4.24 (0.08) 7.17 (0.23) 5.55 (0.15) 2.65 K2O 7.71 (0.20) 7.28 8.75 (0.26) 9.96 (0.54) 8.66 (0.06) 6.45 (0.21) 8.45 (0.28) 9.93 D.I. 89.48 86.86 88.18 78.47 85.98 80.81 77.67 80.77 Ne 8.20 3.89 1.28 9.82 1.26 21.24 22.73 0.00 FeO* = total Fe as FeO 11 - Pumice fragment from Campanian Ignimbrite pyroclastic fall deposit at Cologna (Salerno). 22 - Y-5 tephra from core RC9-191, in Keller et al. (1978), recalculated water free. 33 - Y-3 tephra from core RC9-191, sample courtesy of prof. J. Keller. 44 - Pumice fragment from Somma Vesuvio Codola eruption pyroclastic fall products, sample courtesy of prof. G. Rolandi. 55 - Pumice fragment from Somma Vesuvio Pomici di Base eruption pyroclastic fall products, sample courtesy of prof. G. Rolandi. 66 - White pumice fragment from Somma Vesuvio AD 79 eruption pyroclastic fall deposit from the Poggiomarino outcrop. 77 - Grey pumice fragment from Somma Vesuvio AD 79 eruption pyroclastic fall deposit from the Zabatta (San Giuseppe Vesuviano) outcrop. 88 - Unit C Ignimbrite matrix glass from Scarafea (Signorelli et al., 1999). Fig. 3 - TAS classificative plot for investigated tephra layers and for comparison samples. For comparison samples symbols legend see Table 4. Diagramma classificativo TAS per i livelli di tefra e per i campioni di riferimento. Per la legenda dei simboli dei campioni di riferimento vedi Tabella 4. R. Munno & P. Petrosino in the higher part of the C106 (samples A1(a) and A1(b), Tab.2) and C45 (samples B1(a) and B1(b), Tab.2) core makes it possible to hypothesize the rela- tionship with the Somma-Vesuvio A.D. 79 eruption (samples 6 and 7, Tab.4). The fall phase of this erup- tion, in fact, was dispersed towards the Salerno Gulf. Besides this, the change in colour (from white to gray) and in composition (from phonolitic to tephry-phonolitic) of pumice fragments is a diagnostic feature for A.D. 79 eruption products. Moreover, garnet is a typical minera- logical phase for white pumice fragments as leucite is for gray. The tephra layers alkali-trachytic in composition, found both in C106 and C45, represented by samples A2 and B2 (Tab.2), respectively, could be related to Somma-Vesuvio first activity, namely to Codola (25.1±0.4 ka B.P. - Alessio et al., 1974 – sample 4, Tab.4) and Pomici di Base (18.3±0.18 ka B.P. - Andronico et al., 1995– sample 5, Tab.4) eruptions, but also repeat the general chemical features of Campi Flegrei products. The dispersal of Somma-Vesuvio Codola eruption fall products is, at present, not well known, but the comparison of Si/Ca ratio in Codola and in sample A2 and B2 juvenile fraction makes it possible to exclude the correlation (Fig.3, 4). As regards the Pomici di Base eruption products, even if the chemical composition is quite comparable, reconstructed fall pro- ducts distribution is towards the NE from Vesuvio and not in good agreement with the presence of the layer in the Salerno Gulf. Besides this, the age of Pomici di Base eruption does not fit the 14C age of 26030±150 y B.P (Buccheri et al., 2002a) and of 25570±110 y B.P. (Buccheri et al., 2002b) obtained on the microfossils in the clay layer directly underlying the tephra, respectively in the C106 and in the C45 core. This is very reliable since: - in C45 core a 14C AMS age of 19490±110 y B.P. has been recorded in the clay 70 cm above the pyroclastic layer, - in C106 core a calibrated 14C AMS age of about 19.7 ka B.P. has been recorded in the clay 100 cm above the pyroclastic layer (see Fig.2) - no evidence of hiatuses in sedimentation is present along these clay layers. A2 and B2 tephra, however, find a very good cor- respondence in mineralogical content and chemical composition (Tab.4) of Y-3 tephra found in the Ionian Sea. Besides this, in core RC9-191, investigated by Keller et al. (1978), the dating of sapropels gave an age of about 26.000, well fitting the paleontological age obtained for the same layer in the two cores investiga- ted here. As to the third pyroclastic layer in the C45 core, represented by sample B3, it repeats the composition of the fall products of the Campanian Ignimbrite erup- tion sampled at Cologna. Apart from this, it well resem- bles the composition of Y-5 marker layer as reported in K e l l e r e t a l . ( 1 9 7 8 ) , d e f i n i t e l y c o r r e l a t e d t o t h e Campanian Ignimbrite in Narcisi and Vezzoli (1999), and that of the comparison sample. The attribution of lowermost layer to Y-5 is consistent with the presence of 25 cm thickness of pelagic sediments between the layer and the clay level dated at about 26 ka B.P. in C45 core, taking into account a sedimentation rate of about 1,7 cm/ka, as deduced by Buccheri et al. (2002a). 17 Fig. 4 D.I. - Ne plot for investigated tephra layers and for comparison samples. For comparison samples symbols legend see Table 4. Diagramma D.I. - Ne per i livelli di tefra e per i campioni di riferimento. Per la legenda dei simboli dei campioni di riferimento vedi Tabella 4. New constraints on the occurrence ... 44.. CCOONNCCLLUUSSIIVVEE RREEMMAARRKKSS The results of the tephrostratigraphical investiga- tion of C45 and C106 tephra layers are so summarized: - layer B3 of C45, the oldest tephra layer identified, is correlable to the Y-5 marker layer and to the Campanian Ignimbrite basal pumice fall. - Layer B2 of C45 and layer A2 of C106 represent the product of the same volcanic event and have been correlated to the Y-3 marker layer. - Layer B1(a,b) of C45 and layer A1 (a,b) of C106 represent the pumice fall products of the A.D. 79 Somma- Vesuvio eruption. In order to confirm the presence of two different tephra layers in the 40- 25 ka time-span, the tephrostratigraphi- cal literature reporting data for primary tephra layers in the Tyrrhenian Sea has been revised (Paterne 1985; Paterne et al., 1986; Paterne et al., 1988; McCoy and Cornell, 1990; Calanchi et al., 1994). The revision of data entailed the recovering of all stratigraphical rela- tions, modal compositions and chemi- cal analyses of juvenile fraction and, where available, of mineral phases of tephra layers identified in many cores and new correla- tions have been made in the light of the obtained results. Tab. 5 reports these new correlations that con- firm the presence of Y-3 tephra layer in southern Tyrrhenian Sea and along the Campanian coast, whe- reas Y-5 results more widely dispersed (Fig. 5). 18 Fig. 5 - Sketch map of Tyrrhenian Sea reporting the occurrence of Y-3 and Y-5 primary tephra layers, as deduced from the revision of tephrostratigraphic literature regarding the Tyrrhenian Sea and from present paper data. Circles indicate coring sites; black area in the circle indicates the occurrence of Y-3 tephra, dashed area in the circle indicates the occurrence of Y-5 tephra. For core references see Table 5. Schema del Mar Tirreno che riporta la presenza dei livelli marker Y-3 and Y-5, dedotta dalla revisione della letteratura riguardante la tefrostratigrafia del Tirreno e da dati del presente lavoro. I cerchi indicano l'ubicazione delle carote; l'area nera nel cerchio indica la pre- senza all'interno della carota del marker Y-3, l'area tratteggiata nel cerchio indica la presenza all'interno della carota del marker Y-5. Per il riferimento alla carota vedi Tabella 5. Table 5 - SEM-EDS glass composition of Y-3 and Y-5 tephra layers occurring in the Tyrrhenian sea. Composizione SEM-EDS dei vetri dei livelli di tefra Y-3 and Y-5 reperiti in alcune carote del Mar Tirreno. Y-5 Y-5 Y-5 Y-5 Y-3 Y-3 Y-3 core KET 8003 KET 8004 KET 8011 KET 8022 KET 8003 KET 8004 KET 8011 ° ° °° ° ° ° °° SiO2 62.29 62.46 61.90 63.19 60.42 60.59 59.90 TiO2 0.36 0.36 0.34 0.00 0.40 0.23 0.36 Al2O3 19.54 19.59 19.63 19.45 20.18 20.08 20.19 FeO* 2.75 2.49 2.58 2.66 2.34 2.43 2.46 MnO n.d. n.d. n.d. n.d. n.d. n.d. n.d. MgO 0.11 0.11 0.12 0.00 0.08 0.04 0.07 CaO 1.63 1.63 1.88 1.73 3.38 3.16 3.56 Na2O 5.99 6.01 5.86 5.69 3.92 4.39 4.59 K2O 7.33 7.35 7.69 7.27 9.28 9.08 8.87 ° in Paterne et al., 1986 °° in Paterne, 1985 FeO* = total Fe as FeO R. Munno & P. Petrosino The presence of two alcali-trachytic tephra layers offshore the Tyrrhenian coast, namely B2 and B3 in C45, separated by the deposition of pelagic sediments, makes it possible to hypotesize, in the 40-25 ka B.P. time span, the occurrence of two distinct intense volca- nic events with a widely dispersed associated fall phase. The different age, together with the slightly different che- mical composition, suggests that no link exists between them, apart from the source. The lower layer (B3 in C45) is well linkable to the Campanian Ignimbrite eruption, that we can now term First Campanian Ignimbrite, and its presence in the Policastro Gulf core well agrees with the distribution hypothesized for the associated fall phase by Rosi et al. (1999). The sequence cored in C106, offshore the Sele river mouth, where no record of Y-5 has been found, probably stops before encountering this tephra layer. The topmost alkali-trachytic layer (A2 in C106 and B2 in C45) tesifies to the presence of the Y-3 marker layer in the Tyrrhenian area, and no clear attribution to a recognized volcanic event is at present possible. However, it can be interpreted as the record of an erup- tive event from the Campanian area. The absence of a Somma-Vesuvio eruption as the origin of the layer, restricts the field of possible sources to Campi Flegrei. Chemical features of Y-3 are quite peculiar, and distin- guish it from Y-5, here definitely linked to Campanian Ignimbrite. It can be concluded that the presence of this layer accounts for the hypothesis that what is found in the field and ascribed to the Campanian Ignimbrite erup- tion may be the result of at least two distinct eruptive episodes, each with an associated fall phase. Y-3 can be considered the record of another eruptive event, that we term Second Campanian Ignimbrite eruption. The occurrence in volcanological literature of 14C dating ran- ging between 40 and 25 ka can so be read as the result of the sampling of different paleosoil, underlying somewhere the older, elsewhere the younger pyrocla- stic deposit. As to the chemical composition, a slight chemical difference in the composition of Campanian Ignimbrite units has been detected by Signorelli et al. (1999). According to these authors, the last unit of C.I. (sampled in the northern area of Naples, Giugliano municipality, in a quarry at the Scarafea locality) is much richer in K than the first one and seems well to resemble the com- position of Y-3. We hypothesize that this unit can be the pyroclastic flow associated to the fall phase emplacing Y-3. This hypothesis can be supported by the recorded presence of a paleosol between the Campanian Ignimbrite Tuff and the topmost pyroclastic flow deposit at Giugliano and by the 40Ar/39Ar dating of the same deposit, that results much younger than Campanian Ignimbrite (Rolandi et al., 2003). At present more field studies are necessary, aimed at mapping the two diffe- rent ignimbrite deposits linked to Y-5 and Y-3 tephra marker layers. AACCKKNNOOWWLLEEDDGGMMEENNTT This research, within the project "Impact of natural events on biotic communities in the last 20 ka BP Climatic and volcanic events", was supported by PRIN 1998, chief scientist Prof. G. Ciampo. Prof. T.S. Pescatore kindly offered the C106 core. The authors wish to gratefully thank prof. J. Keller, who kindly supplied a tephra sample from core RC9- 191, and prof. G. Rolandi, for supplying some compari- son samples. Prof. L. Lirer is thanked for reading an early draft of the paper. RREEFFEERREENNCCEESS Alessio M., Bella F., Improta S., Belluomini G., Cortesi C. & Turi B., 1971 - University of Rome Carbon-14 Dates IX. Radiocarbon, 1133--22, 395-411. Alessio M., Bella F., Improta S., Belluomini G., Calderoni G., Cortesi C. & Turi B., 1973 - University of Rome Carbon-14 dates X. Radiocarbon, 1155--11, 165-178. Alessio M., Bella F., Improta S., Belluomini G., Calderoni G., Cortesi C. & Turi B., 1974 - University of Rome Carbon-14 dates XII. Radiocarbon, 1166--33, 358-367. Andronico D., Calderoni G., Cioni R., Sbrana A., Sulpizio R. & Santacroce R., 1995 - Geological map of Somma-Vesuvio volcano. Period. Mineral., 6644((11--22)), 77-78. Armienti P., Barberi F., Bizouard H., Clocchiatti R., Innocenti F., Metrich M., Rosi M. & Sbrana A., 1983 - The Phlegraen Fields: Magma evolution within a shallow chamber. J. Volcanol. Geotherm. Res., 1177, 89-311. Barberi F., Innocenti F., Lirer L., Munno R. & Pescatore T.S., 1978 - The Campanian Ignimbrite: a major prehistoric eruption in the Neapolitan area (Italy). Bulletin Volcanologique, 4411,1-22. Buccheri G., Bertoldo G., Coppa M.G., Munno R., Pennetta M., Siani G., Valente A. & Vecchione C., 2002a - Studio multidisciplinare della successione sedimentaria tardo-quaternaria proveniente dalla scarpata continentale del Golfo di Policastro (Tirreno meridionale). Boll. Soc. Geol. It., 112211, 187-210. Buccheri G., Capretto G., Di Donato V., Esposito P., Ferruzza G., Pescatore T., Russo Ermolli E., Senatore MR., Sprovieri M., Bertoldo M., Carella D. & Madonia G., 2002b - A high resolution record of the last deglaciation in the southern Tyrrhenian Sea: environmental and climatic evolution. Marine Geology, 118866,, 447-470. Calanchi N., Gasparrotto G. & Romagnoli C., 1994 - Glass chemistry in volcaniclastic sediments of ODP leg 107, Site 650, sedimentary sequence: provenance and chronological implications. J. Volcanol. Geotherm. Res., 6600,, 59-85. Calderoni G., Vesica P. & Turi B., 1993 - Radiocarbon dating of the Campanian ignimbrite (tufo grigio campano auct.): towards a definition of the main emplacement phase. Symposium "Quaternary stratigraphy in volcanic areas", Roma, September 20-22. Abstracts, 17. Cassignol C. & Gillot P., 1982 - Range and effective- ness of unspiked potassium-argon dating: experi- mental ground work and application. in G.S. Odin (ed.) "Numerical dating in stratigraphy",160-179. Curtiss G.H., 1966 - The problem of contamination in obtaining accurate dates of young geologic rocks. 19New constraints on the occurrence ... In: OA Shaeffer and J Zahringer Eds. Potassium Argon Dating, Springer Berlin. Deino A.L., Courtis G.H. & Rosi M., 1992 - 40Ar/39Ar dating of Campanian Ignimbrite, Campanian Region, Italy. Int Geol Congr Kioto, Japan, Abstracts, vol 3, p 2654. Deino A.L., Courtis G.H., Southon J., Terrasi F., Campaiola L. & Orsi G., 1994 - 14C and 40Ar/39Ar dating of the Campanian Ignimbrite, Phlegraean Fields, Italy. Abstracts ICOG, Berkley, CA, U.S.A. De Vivo B., Rolandi G., Gans P.B., Calvert A., Bohrson W.A., Spera F.J. & Belkin H.E., 2002 - New con- straints on the pyroclastic eruptive history of the Campanian volcanic Plain (Italy). Mineralogy and Petrology, 7733, 47-65. Fisher R.V., Orsi G., Ort M. & Heiken G., 1993 - Mobility of large-volume pyroclastic flow emplacement of the Campanian Ignimbrite, Italy. J. Volcanol. Geotherm. Res., 5566, 205-220. Keller J., Ryan W.B.F., Ninkovich D. & Altherr R., 1978 - Explosive volcanic activity in the Mediterranean over the past 200,000 years as recorded in deep- sea sediments. Bull. Geol. Soc. Am., 8899, 591-604. Le Maitre R.W. Ed, 1989 - A classification of igneous rocks and glossary of terms. Blackwell Scientific Publications, Oxford, 193 pp. McCoy F.W. & Cornell W., 1990 - Volcaniclastic sedi- ments in the Tyrrhenian Basin. Proc. Ocean Drilling Program, 110077, 291-306. Narcisi B. & Vezzoli L., 1999 - Quaternarnary strati- graphy of distal tephra layers in the Mediterranean-an overview. Global and Planetary Change, 2211, 31-50. Paterne M., 1985 - Reconstruction de l'activitè explosive des volcans de l'Italie du Sud par téphrochronolo- gie marine. Thèse Dr. Sciences Paris-Sud., 144 pp. Paterne M., Guichard F., Labeyrie J., Gillot PY. & Duplessy J.C., 1986 - Tyrrhenian sea tephrochro- nology of the oxygen isotope record for the past 60,000 years. Marine Geology, 7722, 259-285. Paterne M., Guichard F. & Labeyrie J., 1988 - Esplosive activity of the south Italian volcanoes during the past 80,000 years as determined by marine teph- rochronology. J. Volcanol. Geotherm. Res., 3344, 153-172. Rinaldi R., 1979 - La microanalisi elettronica: strumen- tazione e applicazioni mineralogico-petrografiche. Rend. Soc. It. Min. Pet., 55((22)), 507-526. Rolandi G., 1997 - The eruptive history of Somma- Vesuvius. In: Volcanism and Archaeology in Mediterranean Area (Cortini M. and De Vivo B. Eds.). Research Signpost, Trivandrum, India 77- 88. Rolandi G., Petrosino P. & McGeehin J., 1998 - The interplinian activity at Somma-Vesuvius in the last 3500 years. J. Volcanol. Geotherm. Res., 8822, 19- 52. Rolandi G., Bellucci F., Heizler M.T., Belkin HE. & De Vivo B., 2003 - Tectonic controls on the genesis of ignimbrites from the Campanian volcanic zone, Southern Italy. Mineralogy and Petrology, 7799, 3- 31. Rosi M., Vezzoli L., Aleotti P. & De Censi M., 1996 - Interaction between caldera collapse and eruptive dynamics during the Campanian Ignimbrite erup- tion, Phlegraean Fields, Italy. Bull. Volcanol., 5577, 541-554. Rosi M., Vezzoli L., Castelmenzano A. & Grieco G., 1999 - Plinian pumice fall deposit of the Campanian Ignimbrite eruption (Phlegraean Fields, Italy). J. Volcanol. Geotherm. Res., 9911((22-- 44)), 179-198. Scandone R., Bellucci F., Lirer L. & Rolandi G., 1991 - The structure of the Campanian Plain and the acti- vity of the Neapolitan volcanoes (Italy). J. Volcanol. Geotherm. Res., 4488, 1-31. Signorelli S., Vagelli G., Francalanci L. & Rosi M., 1999 - Origin of magmas feeding the Plinian phase of the Campanian Ignimbrite eruption, Phlegrean Fields (Italy): constraints based on matrix-glass and glass-inclusion compositions. J. Volcanol. Geotherm. Res., 9911((22--44)), 199-220. Thunell R., Federman A., Sparks S. & William D., 1979 - The age, origin, and volcanological significance of the Y-5 ash layers in the Mediterranean. Quaternary Res., 1122, 241-253. Vezzoli L., 1988 - Island of Ischia. CNR, Quaderni della Ricerca Scientifica 10, Roma, 132 pp. 20 Ms. ricevuto il 15 luglio 2003 Testo definitivo ricevutol’11 novembre 2003 Ms. received: July 15, 2003 Final text received: November 11, 2003 R. Munno & P. Petrosino