Microsoft Word - 00_indice_LM04.docx Il Quaternario IT ISSN 039-3356 Italian Journal of Quaternary Sciences 24 (1), 2011 - 51-65 THE LATEST CONTINENTAL FILLING OF VALLE UMBRA (TIBER BASIN, CENTRAL ITALY) DATED TO ONE MILLION YEARS AGO BY MAGNETOSTRATIGRAPHY 1Roberto Bizzarri, 2Andrea Albianelli, 3Patrizia Argenti, 1Angela Baldanza, 1Roberto Colacicchi & 2Giovanni Napoleone 1Dipartimento Scienze della Terra, Università di Perugia, Perugia 2Dipartimento Scienze della Terra, Università di Firenze, Firenze 3Lab. Paleontologia dei Vertebrati, Dip. Scienze Geologiche, Università di Roma Tre, Roma Corresponding author: R. Bizzarri ABSTRACT: Bizzarri R., Albianelli A., Argenti P., Baldanza A., Colacicchi R. & Napoleone G., The latest continental filling of Valle Umbra (Tiber Basin, central Italy) dated to one million years ago by magnetostratigraphy. Paleogeography of central Italy during Pliocene and Pleistocene is characterized by the on-setting of continental basins, oriented almost parallel to the uplifting Apennine mountain chain and marked by the valleys of main rivers Arno, Chiana, Tevere, together with minor intermountain basins (e.g.: Colfiorito). The internal range, with its complex of intermountain basins, was bounded by coastal marine de- posits of the Tyrrhenian and Adriatic seas. Due to a decidedly established palaeoenvironmental and magnetostatigraphic scenery for Valdarno, Valtiberina and Colfiorito Basins, the short section at Arquata in the Valle Umbra Basin, here presented, may be constrained much more closely to the evolutionary pattern of the former ones by means of a similar magnetostratigraphic characterization. In particu- lar, detailed magnetostratigraphy led to date the earliest deposits in the Valdarno Basin at 3.3 Ma, whereas in Valtiberina the oldest ana- lyzed deposits, which don’t reach the substratum yet, postdate 2.8 Ma. Upward, the two above mentioned main sedimentary profiles reach 2.0 Ma and 2.1 Ma, respectively, and in the shorter section of Colfi- orito only the youngest dates are attained, until 0.78 Ma. The same younger date was detected in the Valdarno, where also the Olduvai chron (1.95-1.77 Ma) is recorded, followed by the Jaramillo chron (1.05-0.97 Ma) after a long sedimentary hiatus. The Valle Umbra profile exposed near Bevagna is dated across the Jaramillo chron, the duration of its deposition extending approx- imately 300 ky. Both results are very close to the previously mentioned final fillings of Valdarno and Colfiorito. RIASSUNTO: Bizzarri R., Albianelli A., Argenti P., Baldanza A., Colacicchi R. & Napoleone G., Il riempimento continentale più recente della Valle Umbra (Bacino Tiberino, Italia centrale) datato a un milione di anni fa attraverso la magnetostratigrafia. La Valle Umbra, nel settore orientale del Bacino Tiberino in Umbria, ha sinora offerto pochi dati per integrare la transizione paleo- geografica, da ovest verso est, che ha interessato sia gli adiacenti bacini continentali più ampi (Valdarno, Val di Chiana, Valle Tiberina), sia i bacini intermontani minori e più orientali (Colfiorito), tutti con asse orientato circa NW-SE, parallelo alla nascente catena appennini- ca. Nel loro insieme, questi bacini intermontani, appartenenti alla zona centrale della catena appenninica, sono confinati tra fasce costie- re peritirreniche e periadriatiche. Il quadro paleoambientale e magnetostratigrafico, già individuato per i suddetti bacini, ha permesso ora di fissare anche quello della Valle Umbra. La successione sedimentaria della porzione sommitale, messa a giorno nella Cava Arquata a Bevagna, permette di integrare il quadro conoscitivo dell’evoluzione paleoambientale più recente dell’Italia centrale, poiché conclude quelle fasi del sollevamento appenninico iniziate alla data di 3.3 Ma nel Valdarno e quelle presumibilmente coeve della Valtiberina, inter- rottesi rispettivamente a 2.0 Ma e 2.1 Ma. Il nuovo ciclo deposizionale è cominciato subito dopo, datato solo nel Valdarno, da 1.9 Ma a 1.7 Ma, mentre il ciclo finale, nel Valdarno e a Colfiorito, contiene il magnetocrono Jaramillo (1.05-0.97 Ma). Nella successione di Colfio- rito, inoltre, viene raggiunta la data più recente, testimoniata dalla registrazione dell’inizio dell’ultimo magnetocrono, a 0.78 Ma. I depositi studiati in Valle Umbra sono marcati molto nettamente dalla presenza del crono Jaramillo e la deposizione della successione analizzata a Bevagna, nei pressi di Foligno, ha avuto una durata stimata di 300 ky. La parte finale del Pleistocene inferiore e il passaggio Pleisto- cene inferiore - Pleistocene medio sono calibrati alla data di Jaramillo e all’inizio dell’ultimo magnetocrono (Brunhes), rispettivamente. Entrambi i segnali sono registrati nella sezione di Colfiorito. Si può ritenere che la deposizione conclusiva fluvio-lacustre della Valle Um- bra sia ragionevolmente coeva con quella dell’Altopiano di Colfiorito. Key Words: Valle Umbra, Magnetostratigraphic dating, Jaramillo chron, Apennine basins, Early Pleistocene, Central Italy. Parole-chiave: Valle Umbra, Datazione magnetostratigrafica, Crono Jaramillo, Bacini Appenninici, Pleistocene Inferiore, Italia Centrale. 1. INTRODUCTION The Northern Apennine intermountain basins represent bench-marks of structural and stratigraphic sig- nificance for updating most of the present environmental studies on continental deposition occurring through the recent geological past. Basin infills were firstly assigned Pliocene and Pleistocene ages according to the lithologi- cal subdivision of blue and dark-green clay underlying yellowish sand, with both the lithologies closely equivalent to marine analogous deposits ascribed to the end of Pli- ocene and the beginning of Quaternary (e.g., see NAPO- LEONE et al., 2003; VAN COUVERING, 1997). Such ages, though, have recently been calibrated to dates by means of magnetostratigraphy, which right in Umbria provided the conceptual and more complete R. Bizzarri et al. 52 tool for defining the numerical timing for marine sedi- mentary successions. The present example is aimed at testing a different approach to a short interval of conti- nental deposits, devoid either of direct lithostratigraphic correlation with adjacent sections or of a significant fos- siliferous record. Actually, biochronologic ages used for classifying mammal fauna stages in the richest non- marine deposits of Italy (AZZAROLI, 1977; AZZAROLI et al., 1988) were made equivalent to the biostratigraphic ages once they were both calibrated to the same global scale of geomagnetic polarity reversals, currently used for geochronology (ALVAREZ, 2009). The process was first verified in the nearby Upper Valdarno, where local lacustrine to deltaic successions had magnetostrati- graphically been dated in million years (Ma) and several classical Villafranchian Faunal Units, from there estab- lished for biochronology, were labelled with an absolute reference (NAPOLEONE et al., 2003). The Upper Valdar- no profile contained also the thickest continuous profile and earliest mammal fossil record, so to become the calibration frame for other Apennine continental succes- sions, beginning with Valtiberina and Colfiorito Basins in Umbria. These intermountain basins, therefore, have assumed the dates in million years of the geomagnetic polarity time scale. In particular, the Valdarno sequence was calibrated since 3.3 Ma to 1.7 Ma, and with a reso- lution of few thousand years (ky) (ALBIANELLI et al., 1997) and even 1 ky (NAPOLEONE & AZZAROLI, 2002). With recently collected data (ALBIANELLI et al., 2002; NAPOLEONE et al., 2003), and following the cali- bration of the Chronostratigraphic Time Scale to the Geomagnetic Polarity Time Scale (GPTS), for the “Plio- Pleistocene Boundary and the Beginning of the Quater- nary” it is possible to relate the ages of continental de- posits to those of marine sequences, both calibrated to the Gauss and Matuyama chrons including Olduvai. To the Olduvai chron the Pliocene/Pleistocene boundary was fixed at 1.796 Ma in the Vrica stratotype (VAN COU- VERING, 1997), albeit, recently, the Pliocene/Pleistocene boundary has been repositioned at 2.58 Ma (GIBBARD et al., 2010). The magnetostratigraphic analysis of the Valle Umbra continental sediments has therefore been done for identifying particular intervals (magnetozones) co relatable to the chrons of the geomagnetic time scale, by which the biochronologic “stages” could be calibrated. The magnetostratigraphic survey has been carried out on the profile exposed in the Arquata quarry, near Bevagna (Perugia), in order to attain the conti- nuous dating for the ages expected in the about 65 m thick deposit, from prominent sedimentological and bi- ochronologic evidences selected in the adjacent basins. For that section, sedimentological features, fossil con- tent and magnetostratigraphic calibration are here con- sidered, and firstly some time constrains for Valle Um- bra deposits are proposed, allowing to insert also the eastern branch of Tiber Basin into a Northern Apennine Basins comprehensive evolutionary scheme. 2. GEOLOGICAL SETTING OF VALLE UMBRA The paleogeographic evolution of central Italy was marked since Piacenzian by the setting and develop- ment of large continental basins along the inner Apen- nine Chain. Basin complex was bordered by coastal ma- rine deposits on both sides, from the Tyrrhenian and Adriatic seas, although direct correlations among conti- nental and marine domains are not always clearly do- cumentable. Such basins, as the major ones of Mugello and Upper Valdarno, Valdichiana and Valtiberina, were structurally aligned NW-SE, almost parallel to the Apen- nines Mountain range, and occur together with the in- termountain minor basins, as Tavernelle-Pietrafitta, Gub- bio, Gualdo, Colfiorito, Norcia etc. (Fig. 1). The structural assessment for all of them lies within tectonic depres- sions bordered by direct faults belonging to the exten- sional phase, which followed the compressional one act- ing since Miocene on the whole area of Tuscan-Umbrian and Latium regions. Right because linked to the W-to-E migration of the Apennine over-thrusting front, ages of the extensional basins and related infill become progressively younger with the same trend (Fig. 1). The master-fault system, mainly dipping to SW and associated to the anti- thetic one, produced half-graben structures with dis- placements of several hundreds of meters (BARCHI et al., 1991; MENICHETTI, 1997). Pliocene and Pleistocene continental deposits cover nearly one third of the Um- brian territory (LOTTI, 1917, 1926) and mainly belong to the Tiber Basin and its split major valleys, Valtiberina (from Sansepolcro to Terni) and Valle Umbra (from Pe- rugia to Spoleto). They are respectively confined by the Meso-Cenozoic mountain belts of Massicci Perugini, Mt. Peglia and Narnese-Amerina chain, due west, and Mt. Catria-Mt. Nerone-Mt. Serano Mountains, due east and separated by the Martani mountain range (CONTI & GIROTTI, 1977; AMBROSETTI et al., 1977, 1978, 1987, 1989, 1995a; BASILICI, 1992, 1995; MARTINI & SAGRI, 1993). The whole basin underwent extensional deforma- tion phases (due to faults reactivation), from Gelasian- Calabrian to Late Pleistocene (GREGORI & CATTUTO, 1986; BARCHI et al., 1991; BONINI, 1998). Contrary to the Valtiberina where several expo- sures are available, Valle Umbra is fairly poor of out- crops, thus hampering an appropriate recognition of its most recent geologic history (Fig. 1.2). In south Valle Umbra (Spoleto area), COLTORTI & PIERUCCINI (1997) identified two important depositional phases referable to Pliocene and Pleistocene, respectively, separated by a “tectono-sedimentary discontinuity”. The earlier depo- sits, unconformably laying on the pre-Pliocene substra- tum, are made of fluvial-lacustrine and alluvial fan-delta sediments. The older are exposed in the Spoleto area, in the lignite-bearing levels outcropping at Morgnano. The mammal fauna there recovered was classified to belong to the Triversa Faunal Unit, and assigned to the earliest stage of the Early Villafranchian Mammal age in the biochronologic classification of AZZAROLI (1977). In the Valdarno Basin a comparable fauna was dated 3.1 Ma in the Kaena magnetochron (ALBIANELLI et al., 1997). Most of these lower deposits from Valle Umbra are made of clays, containing lignite dark levels inter- bedded with thin sandy levels; in the upper part they mainly alternate with banks of coarser sand and fine- grained conglomerates. The younger sedimentary phase, on the contrary, is represented by thick alluvial-fan con- glomerates deposited in the southern valley by a num- ber of streams draining the western flank of the Serano- Brunette ridge and the eastern flank of the Martani Mts. (COLTORTI & PIERUCCINI, 1997). The latest continental filling of Valle Umbra (Tiber Basin, central Italy) … 53 Fig. 1 - 1) Sketch of northern Apennine extensional basins and geographic collocation of the study area (modified after NAPOLEONE et al., 2003). 2) Schematic geological setting of Valle Umbra. 1) Schema dei Bacini estensionali dell’Appennino settentrionale e ubicazione geografica dell’area studiata (da NAPOLEONE et al., 2003, modificato). 2) Assetto geologico schematico della Valle Umbra. Tectonics played an important role in modeling the present morphology of Valle Umbra since Middle-Late Pleistocene, even until last interglacial stage (GREGORI & CATTUTO, 1986). Two major fault systems were acting: one, in the eastern side uplifting the Serano-Brunette ridge along the border fault, while its piedmont accumu- lated in fan deposits; the other one, the antithetic east- dipping fault system was uplifting the conglomerates of the second sedimentary phase (presumably), originally laying at the plain level, up to the present Montefalco hill, nearly 180 meters above the valley. Immediately before this phase of uplift, small ponds were isolated, with local carbonate-siliceous deposition (COLACICCHI et al., 2007). 3. THE ARQUATA QUARRY STRATIGRAPHIC SEC- TION The whole available succession of the Valle Umbra sediments, representative of latest evolution of the basin, is here exposed across a nearly 65 m thick stratigraphic profile (Figs. 2, 3). The examined deposits are mainly formed by clays and sands containing lignite levels, which dip to the SW by 10°-12°. A well visible extension- al fault in the mid-profile, almost E-W striking and dipping southward, suggests a minor displacement of few metres (Fig. 3). Some minor surfaces may be locally recognized, either produced by tectonics or sin-depositional col- lapses, each less than 1 m in displacement. 3.1 Sedimentological features The succession may be referred to a depositional la- custrine and fluvial-lacustrine frame, and two depositional environments are documented: a shallow-water lake and an alluvial plain with distal fluvial supplies (Fig. 3). The facies associations are divided accordingly (COLACICCHI & BIZZARRI, 2005; COLACICCHI et al., 2007), as: Facies Association A (shallow lake environment) - The association groups facies referable to a shallow water lacustrine environment. Facies A1 is mainly represented by massive to thin-laminated, fossils bearing clay beds (Fig. 4.1). Thin fine sand lenses showing wave and cur- rent ripples locally occur. Fossils are both scattered or grouped in thin (up to 5 cm thick) layers, and banks of vegetal matter also occur as thin lignite levels, but more commonly enhancing the lamination. Facies A2 is represented by alternations of thin laminated clay and ripple laminated silt/fine sand, with the latter prevailing (Figs. 4.2, 4.3). Both facies A1 and A2 are referable to a shallow lake environment with stable and presumably faint wave motion. Facies A3 is characterized by wave ripple lami- nated sand (Fig. 4.4), locally with hardened top and bot- toms and polygonal drying cracks (Figs. 4.5, 4.6), do- cumenting subaerial exposure such as of lake beach. Wave ripple laminated sands locally contain clay chips, vegetal matter and root remains. Facies A4 is represented by dune- to ripple-laminated sand resulting from minor turbidites (Fig. 4.7), due to distal fluvial flooding and/or sediments remobilization possibly after seismic shocks or strong wind action. Some of these sandy bodies evolve to a clearer beach deposits, also showing drying polygons and root remains, as a proof of sedimentation in a not deep environment. Facies Association B (alluvial plain environment) - Fa- cies B1 is prevalently composed of laminated silty clay, deposited in an alluvial-plain environment, whereas fa- cies B2 is described as root-bearing massive to lami- nated clay and is interpreted as poorly drained paleo- R. Bizzarri et al. 54 Fig. 2 - Panoramic views of Arquata quarry, along SW-NE (up) and W-E (down) profiles. Panoramiche di cava Arquata: vista SW-NE (in alto), vista W-E (in basso). soils (Fig. 4.10). Facies B3 shows decimeter-to-meter thick levels of massive to slightly laminated silty sand (Figs. 4.8, 4.9), locally associated with facies B4, the latter described as current ripple and/or cross laminated sand (Figs. 4.11, 4.12). The both latter two facies are interpreted as produced by distal fluvial influxes (river flooding). Facies Association C (palustrine or marsh-pond envi- ronment) - Facies C1 (Fig. 4.13) is formed by less than 1 meter-thick lignite levels, usually bearing larger gastro- pods and Unio sp. Grey clayey palustrine paleosoils with rooted radical apparatuses, scattered vegetal ma- terial, fossils, remains of the original lamination and red- dish surfaces are associated in facies C2, which can be indicated as root-bearing massive clay (Fig. 4.13). The lignite levels and the massive clayey paleosoils (facies C1 and C2 respectively) are anyhow constantly asso- ciated, leading to a poorly oxygenated and strongly re- ducing environment, able to maintain organic matter al- most un-altered. Both associations represent swampy lake flanks as well as small ponds and marsh within the alluvial plain. Facies architecture - The lower two thirds of the succes- sion (Unit 1: Fig. 3) are dominated by Associations A of more frankly lacustrine environment, while alluvial plain deposits (Associations B) prevail in the final portion (Unit 2: Fig. 3). The variable thicknesses of lignite and clays of palustrine deposition (Associations C) may be present in both environments, often transitional from A to B (Fig. 3). Rhythmicity is a recurring characteristics in the A associations, both within a single association and vertically shifting from to one another. Long-lasting and stable lacustrine conditions, still with minor variations due to seasonality, supply variations, etc., characterize the lower section, with the predominance of A1 and A2 facies (Fig. 3) whereas the shift from lacustrine to alluvi- al plain conditions may be most probably linked to long- er period of climatic fluctuation. Post-depositional structures, such as convolute la- mination (Fig. 4.14) and micro-faulting (Fig. 4.15) com- monly occur throughout the entire section and are do- cumented in all the described facies. Particularly, struc- tures due to subaqueous remobilization of partially lithi- fied deposits (Fig. 4.16) suggest a probable seismic ori- gin which testifies of tectonic pulses coeval to sedimen- tation. 3.2 Paleontological record Macropalaeontological content is represented by rare vertebrate remains (fragments of carapace and long bones of turtle Emys orbicularis and only one sca- pula fragment of amphibian) and abundant non-marine molluscs, associated to palaeocarpological remains. The malacological assemblage (COLACICCHI et al., 2005; CAPRAI, 2010) collected in the lower portion (Unit 1) is characterized by Melanopsis (Melanopsis) affinis, Theodoxus (Neritina) groyanus, Viviparus (Viviparus) belluccii, Prososthenia cf. P. oblonga, Neumayria priscil- lae, Emmericia umbra, Melanoides (Melanoides) curvi- costa, Unio pillai, Dreissena polymorpha, Pisidium sp., Hauffenia minuta, Valvata interposita, V. piscinalis (Fig. 5). The assemblages are almost comparable with the ones reported by PETRONIO et al. (2000-2002). Oc- currence of terrestrial gastropods, as Carychium (Sara- phia) tridentatum and Vertigo (Vertilla) angustior, al- though previously reported (SIMONETTI, 1996) have not been here documented. Assemblages contain only few species which occur from Middle Villafranchian (M. affi- Fig. 3 - Magnetostratigraphic and sedimentological section in the Arquata quarry deposits. In addition to the calibration to GPTS scale, recognized facies, fossil content and paleoenvironmental restoration are also provided. Sezione magnetostratigrafica e sedimentologica nei depositi di Cava Arquata. Oltre alla calibrazione con la scala GPTS, sono riportati anche le facies riconosciute, il contenuto paleonotologico e l’interpretazione paleoambientale. The latest continental filling of Valle Umbra (Tiber Basin, central Italy) … 55 R. Bizzarri et al. 56 Fig. 4 - Sedimentary facies in the Arquata quarry deposits. 1) Massive and thin laminated clay (facies A1). 2, 3) Alternations of thin lami- nated clay and wave ripple-laminated sand/silt (facies A2). 4) Wave ripple-laminated sand (facies A3). 5) Particular of hardened top of wave ripples in the facies A3. 6) Polygonal drying cracks in facies A3. 7) Dune- to ripple-laminated sand resulting from minor turbidites (facies A4). 8, 9) Massive to slightly laminated silty sand (facies B3); picture 8 also shows channeled base surface. 10) Laminated silty clay (facies B1) associated to root-bearing massive to laminated clay (facies B2). 11) Top hardened surface of current ripple (facies B4). The latest continental filling of Valle Umbra (Tiber Basin, central Italy) … 57 12) Cross laminated sand, bearing clay chips (facies B4). 13) Lignite level (facies C1), at the top of root-bearing massive to poorly lami- nate grey clay (palustrine paleosoils, facies C2). 14) Convolute lamination in clay and silt of facies A2. 15) Micro-fractures showing less than 5 cm of dislocation (“micro-faulting”) probably effect of sediment adjustments. 16) Sin-sedimentary remobilization of partially lithified clay and silt, due to seismic shocks. Facies sedimentarie descritte nei depositi di Cava Arquata. 1) Alternanza di argille massive e sottilmente laminate (facies A1). 2, 3) Alter- nanze di argille sottilmente laminate e sabbie o silt con ripple da onda (facies A2). 4) Sabbie con ripple da onda (facies A3). 5) Particolare della superficie indurita dei ripple da onda nella facies A3. 6) Poligoni di disseccamento nella facies A3. 7) Sabbie laminate, con dune nella parte inferiore passanti verso l’alto a ripple da onda, prodotte da correnti di torbida (facies A4). 8, 9) Sabbie siltose, da massive a sottilmen- te laminate (facies B3); l’immagine 8 mostra anche la superficie erosiva canalizzata alla base del deposito. 10) Argille siltose laminate (fa- cies B1) associate ad argille da massive a laminate, contenenti resti di apparati radicali (facies B2). 11) Superficie sommitale indurita nei ripple da corrente della facies B4. 12) Sabbie a laminazione incrociata, con plasti di argilla (facies B4). 13) Livello di lignite (facies C1), al tetto di argille grigie, massive o debolmente laminate, con apparati radicali (paleosuoli di palude, facies C2). 14) Lamine convolute nelle argille e silt della facies A2. 15) Microfratture con dislocazione inferiore a 5 cm (“micro-faulting”), probabilmente risultanti dall’assestamento del sedimento. 16) Rielaborazione sinsedimentaria di argille e silt parzialmente litificati, dovuta a shock sismici (micro-sismite). Fig. 5 - Fossil record in the Arquata quarry deposits. 1) Unio pillai. 2) Pisidium sp. 3) Dreissena polymorpha. 4) Melanoides (Melanoides) curvicosta. 5) Viviparus (Viviparus) belluccii. 6) Valvata interposita. 7) Melanopsis (Melanopsis) affinis. 8) Emmericia umbra. 9) Proso- sthenia oblonga. 10) Theodoxus (Neritina) groyanus. 11) Cyprideis torosa. 12) Ilyocypris bradyi. 13) Potamogeton sp. 14) Itea europaea. 15) Scirpus isolepioides. 16) Ranunculus pseudoflammula (achene) 17) Mallotus maii (seed). 18) Rubus sp. 19) Selaginella gr. plioceni- ca (macrospore). 20) Thalictrum minimum. Record paleontologico nei depositi di Cava Arquata. 1) Unio pillai. 2) Pisidium sp. 3) Dreissena polymorpha. 4) Melanoides (Melanoides) curvicosta. 5) Viviparus (Viviparus) belluccii. 6) Valvata interposita. 7) Melanopsis (Melanopsis) affinis. 8) Emmericia umbra. 9) Proso- sthenia oblonga. 10) Theodoxus (Neritina) groyanus. 11) Cyprideis torosa. 12) Ilyocypris bradyi. 13) Potamogeton sp. 14) Itea europaea. 15) Scirpus isolepioides. 16) Ranunculus pseudoflammula (achene). 17) Mallotus maii (seme). 18) Rubus sp. 19) Selaginella gr. plioce- nica (macrospora). 20) Thalictrum minimum. R. Bizzarri et al. 58 nis, T. groyanus, E. umbra and, dubitatively, U. pillai) or Olivola F.U. (P. oblonga) to Farneta F.U.; N. priscillae is reported only in the Tasso F.U., whereas V. belluccii, M. curvicosta and V. interposita are still currently docu- mented in the Tasso and Farneta Faunal Units (GLIOZZI et al., 1997). Furthermore, the genera Prososthenia and Neumayria are considered extinct at the end of Early Pleistocene, as well as E. umbra and M. affinis, which do not survive beyond Early Pleistocene (AMBROSETTI et al., 1995b; GLIOZZI et al., 1997; PETRONIO et al., 2000- 2002; GIROTTI et al., 2003). Mollusc assemblages coming from the upper sec- tion (Unit 2) only contain D. polymorpha, lacking strati- graphic relevance. The palaeocarpological analyses evidence two dif- ferent types of assemblages referable to particular edaphic conditions: 1) hygrophilous plants Scirpus isole- pioides, Ranunculus pseudoflammula, cf. Itea europaea, living and accumulated in situ, and 2) mesophilous plants Potamogeton sp., Rubus sp., Melissa elegans, Ficus po- tentilloides, Mallotus maii, Thalictrum minimum of warm- temperate climate, living in the surrounding area and car- ried by rivers into the basin (Fig. 5). These specimens are accompanied by common Characee oogonia and other remains. Among the latter, a much altered bract and a Selaginella gr. pliocenica macrospore (Fig. 5.19), a present-day species of humid and shady environment, tropical to sub-tropical (CAVALLO & MARTINETTO, 2001), are included. Microfossils are only represented by fresh- water ostracods Cyprideis torosa and Ilyocypris bradyi (Figs. 5.11, 5.12), known in Central Italy Basins from Early Pleistocene (Tasso F.U.) onward (AMBROSETTI et al., 1987, 1995b; GLIOZZI et al., 1997). Both forms are very abundant in the taphocenosys, and probably well represent the biocenosys: absence of competition and food richness probably supported a well developed community. Both species prefer clay or silty-clay bottoms and tolerates anomalous salinity, probably finding here ideal conditions to develop, also due to dissolved salts coming from surrounding exposed rocks. Abundance of Cyprideis torosa, together with dark-grey colour of depo- sits, is indicative of low oxygen level at the basin floor, with partial re-oxygenation due to wind and/or rain, which may have remixed the thin water column. Biochronology - Malacological assemblages and the oc- currence of Cyprideis torosa suggest, for the lower sec- tion, a Late Villafranchian age not older than Tasso F.U. Some taxa, such as V. belluccii, M. curvicosta and V. in- terposita, indeed, have been considered as typical of Tasso-Farneta F.U. interval and never documented af- terward (GLIOZZI et al., 1997). On the other hand, the up- per section is characterized by rare fluvial and/or lacu- strine specimens, mainly represented by D. polymorpha, occurring on the whole Late Villafranchian record from Tasso F.U. onward, lacking biochronological meaning. 3.3 Paleoenvironmental reconstruction Facies analysis allows hypothesizing an alluvial plain environment, with contemporaneous occurrence of shallow lakes, with both palustrine margins and beaches, ponds, swamps and creeks, the latter draining from the nearby mountain flanks toward the basin (COLACICCHI & BIZZARRI, 2005; COLACICCHI et al., 2005; COLACICCHI et al., 2007). Such an interpretation is congruent, indeed, with paleocarpologic data, which indicate riverine trans- port from surrounding areas. Facies associations can be compared to the ones described in the western branch of Tiber Basin in the upper portion of Fosso Bianco and S.M. di Ciciliano Units (AMBROSETTI et al., 1995a, 1995b; BASILICI, 1997, 2000). Particularly, the depositional con- text seems the same as in the partially coeval S.M. di Ciciliano Unit, except for the organized river channel de- posits totally missing in the Arquata section. A long- lasting alluvial plain environment, probably endoreic and devoid of an organized drainage, can be proposed for the Early Pleistocene of Valle Umbra. Some inferences on sedimentation rates can be proposed here. In fact, in the adjacent basins, sedimentation rates vary from 20-24 m/100 ky on lacustrine deposits, both in the Tiber Basin and in the Valdarno, to 35 m/100 ky on fluvial sands in Valdarno (NAPOLEONE et al., 2003). As the two thirds of Arquata section are represented by lacustrine deposits, intermediate values of sedimentation rates between la- custrine and river deposits could be expected. Most of the malacofauna has been collected into vegetal re- mains-rich beds, whereas the ostracods characterize only the purely lacustrine facies. As documented in simi- lar contexts (e.g. Fosso Bianco lacustrine clay), absence of vertebrates, and particularly of micromammifers, seems mainly due to taphonomy and lack of preservation rather than to paleoecology. 4. THE MAGNETOSTRATIGRAPHIC PROFILE 4.1 Paleomagnetic sampling The succession of sediments across the lithostra- tigraphic profile appears quite regular, as remarked above, with noticeable repetitions of sedimentary struc- tures as well as of facies. Sampling for magnetic mea- surements has been carried out at an average spacing of one meter (Fig. 3), preferably in the less sandy levels in order to avoid sediment formed under active motion where particle orientation would have been dispersed under a uniform inducing field. The choice helped to better enhance the paleomagnetic vector in sharpening its resolution power for definition of magnetozone boundaries, even in material very weakly magnetized. Each sample was extracted from levels excavated across the few meter thick steps used in quarrying op- erations across the above mentioned profile, and their resulting location was thence accurately fixed in strati- graphic order and of fresh rock material. As remanent magnetizations were presumably weak and no metal protections around, in a nearly 0° declination field, samples were oriented by magnetic compass using manual simple operations. The average spacing of col- lected levels was chosen at one meter distance, as that used in the other intermountain basins previously inves- tigated so far, considering a presumably recorded time span of some hundreds of ky, and having in accor- dance the accumulation rate of 30-40 m/ 100 ky been assumed. A second aim, indeed, led collecting un- oriented samples at 15-20 cm distances to be meas- ured for their magnetic content under the present field, the magnetic susceptibility, which depends on the amount of magnetic particles under magnetic induction from the present Earth’s field. Changes in susceptibility were measured in Um- bria and central Italy for either continental or oceanic The latest continental filling of Valle Umbra (Tiber Basin, central Italy) … 59 Fig. 6 - Alternate-field (AF, left) and Thermal (TH, right) demag- netization procedures applied to the same sample of reversed polarity (Aq27), showing a straightforward magnetization path. The Zijderveld polar diagram in the middle line shows the initial overprint of the present-day normal field, cleaned after the first demagnetization steps and releasing the primary remanence into the reversed characteristic vector. Procedure di smagnetizzazione applicate allo stesso campione, a magnetopolarità inversa (Arq 27): Campo alternato (AF, a sinistra) e Termica (TH, a destra). Entrambe le procedure evi- denziano l’esistenza dello stesso chiaro andamento della ma- gnetizzazione. Il diagramma polare di Zijderveld, nella linea centrale, mostra l’iniziale sovrimpressione del campo magnetico attuale, che viene rimosso dopo i primi gradini di smagnetizza- zione, e restituisce la magnetizzazione residua caratteristica del vettore inverso. rock types outcropping, and correlated with climate changes driven by changes in solar radiation due to the orbital forcing, as accounted by the Milankovitch theory. The effects of all such parameters were tested (in ec- centricity, obliquity and precession modes) either in the former sediments from Valdarno and Valtiberina (NAPO- LEONE et al. 2004) or in the latter ones (NAPOLEONE & RIPEPE, 1989). In the present work such results are not reported. A sampling interruption occurred in the Arquata profile at about its middle portion, where a marked fault is visible, and above which the vegetation cover did not allow sampling for nearly 10 m. The fault had been in- terpreted as relevant at the local scale, possibly cutting a slice of the sequence in the order of some meters. Al- so the top profile was not sampled because covered by reworked material from the quarrying activities, as well as its bottom fully occupied by a pond (Fig. 3). 4.2 Laboratory measurements Paleomagnetic measurements were carried out on cubic or prismatic specimens of nearly one-inch side cut from the oriented hand samples. All samples yielded more than two specimens, so that remanence of each level was measured using demagnetization procedures of both types (Fig. 6), thermal and by alternative fields (AF). The latter using the new built-in degausser of the ETH Magnetic Laboratory at Zurich where all the lacu- strine sediments from central Italy studied in last two decades had been measured. The laboratory proce- dures were the same as summarized for the Upper Val- darno Basin, which represents the magnetostratigraphic reference for rock types of the analogous continental basins bordering the Apennine mountain range (NAPO- LEONE et al., 2003). Stepwise thermal demagnetization was carried out on samples, of each level bed, at close- ly spaced temperature intervals, in order to detect even minor changes occurring in those weakly magnetized sediments that would have affected the original direction of the primary vector (Fig. 6). A complete study on rock magnetism is beyond the aims of the paper; neverthe- less, some consideration can be pointed out. No particu- lar changes were noticed for the presence of sulphide minerals, which on the contrary did affect some signifi- cant intervals in the lignite-bearing levels of the Valdar- no with unstable magnetizations. In the final deposits of Valle Umbra, the paleoenvironmental conditions pro- duced minor oxidation on their magnetic mineralogy. Such a behaviour was mainly enhanced by the smooth trend of the AF curves (Fig. 6), which otherwise as- sumed in several other occasions for the Apennine con- tinental deposits completely inconsistent directions at any successive demagnetization step. This was particu- larly evident in the Valdarno Basin beds closer to the thick lignite-bearing level, due to the anoxic conditions there acting during deposition and presence of sul- phides still pervading the overlying sediment for more than 50 m in thickness. Therefore, the straightforward direction shown by both demagnetization treatments of the paleomagnetic vector to the characteristic primary magnetization is interpreted as the typical behaviour due to the presence of a single component in the mag- netic signature; in fact, the characteristic remanent magnetization (ChRM) was well isolated already at tem- peratures above 100-200 °C or at applied fields above 0.2-0.3 mT (Fig. 6). On our skill, and according to former results on analogous deposits (ALBIANELLI et al., 2003; NAPOLEONE et al., 2003), analyses are reasonably con- sistent with the original paleomagnetic field and en- hance magnetozones with different polarity. 4.3 Results The total of 55 analyzed levels yielded three inter- vals of polarity directions, two reversed (R1 and R2) bracketing a normally magnetized zone (N1), within which the polarity change occurs in an evident continuity of the lithostratigraphic profile (Fig. 3). The three magnetozones have to be correlated to the chrons of the GPTS in order to become calibrated to their dates in million years. That is, they will represent the absolute dating at each limit of a polarity change and a steady time flow throughout the profile, having the GPTS been the time expression, in continuous sequence, of the ocean floor magnetic ano- maly stripes recorded for a steadily expanding oceanic R. Bizzarri et al. 60 crust. At Arquata is applied the same criterion as in Valti- berina, where for the Gauss and Reunion chrons, through Matuyama, the time resolution was obtained for the total time span 2.8-to-2.0 Ma (PONTINI et al., 2002). Here, it has to be taken into account that the Olduvai chron lies in the Valdarno closely above Reunion (200 ky) and in a split mode as in the marine series in Marche (central Italy) and in Calabria (southern Italy) (NAKAGAWA et al., 1997; ALBIANELLI et al., 2003). After it and until Present, only two normal polarity intervals are contained in the GPTS, and of them the only normal chron followed by one of re- versed polarity is Jaramillo. The dates bounding it last 1.05 to 0.97 Ma, and it has therefore to be correlated to the N1 magnetozone, whereas for both magnetozones of reversed polarity, R1 and R2, time values ascending and descending outward maintain open limits (Fig. 3). Thick- ness and characters of the N1 magnetozone, together with sedimentological consideration about sedimentation rate pertinent to depositional context recognized, agree to its attribution to the Jaramillo chron (BIZZARRI et al., 2009; COLACICCHI et al., 2009). 5. DISCUSSION 5.1 Magnetochronology in Northern Apennine basins The role of magnetic stratigraphy is meant at pro- viding a numerical date (or absolute dating) throughout a continuous succession of sediments whenever its magnetic signature enables to identify intervals of de- fined magnetic polarity (magnetozones) that can be re- lated to magnetochrons of the Geomagnetic Polarity Time Scale (GPTS). It seems therefore a quite viable digression here to re-appraise the meaning of the re- sults achieved in the magnetostratigraphic surveys car- ried out over sedimentary sequences in the intra- mountain basins close to Valle Umbra and dated in great detail (AMBROSETTI et al., 1995a, 1995b; FICCA- RELLI et al., 1997; ALBIANELLI et al., 1997, 2002, 2003; COLTORTI et al., 1998; NAPOLEONE & AZZAROLI, 2002; NAPOLEONE et al., 2003). Basins surrounding Valle Um- bra provide therefore the primary source to inferring an age for the pile of a presumably much thicker succes- sion, according to the thicknesses visible or estimated in several such basins (Fig. 7). But, unfortunately, the geo- logic and sedimentary history of Valle Umbra represents the less known portion of the continental deposition in the Tiber Basin. Its south-western branch, in fact, is best known, as the oldest sediments, although not the lo- wermost ones, released a date of 2.8 Ma (ALBIANELLI et al., 2002), whereas the north-eastern sector yields most deposits belonging to the final filling. The basal deposi- tion may be inferred, thence, to have begun largely after Miocene, considering that in the Valdarno Basin the ex- posed contact of the lacustrine facies on the pre- Pliocene substratum was dated 3.3 Ma by magnetostra- tigraphy, as early as in the Mammoth chron (ALBIANELLI et al., 1997). The latest filling of Valdarno, indeed, con- tains the last magnetochrons centered at nearly 1 Ma, but a sedimentation gap lasting at least 0.7 million years (my) has been detected just before it. The mentioned lack of exposures in Valle Umbra does not enable to further hypothesize about earlier ages of its deposits. The closest comparison of its final filling may be done with the equivalent one in the nearby Colfiorito Basin (Fig. 7), where the outcrops of the uppermost deposits, exposed in the small towns of Colfiorito and Cesi, re- leased dates respectively ranging from the well confined Jaramillo chron (1.05-0.97 Ma) to the initial Brunhes chron (across the limit of 0.78 Ma) (FICCARELLI et al., 1997; COLTORTI et al., 1998). It is worth recalling that the final Early Pleistocene and the beginning of Middle Pleistocene are respectively calibrated to these chrons. In the Arquata profile, the less than 80 m exposure reduced by a slight dipping to about 65 stratigraphic me- ters represents a thin cover, whereas the oldest deposits visible in the Morgnano lignite mine (Spoleto in Valle Umbra) seem to equate the Valdarno lignite-bearing le- vels mined at Santa Barbara and dated 3.1 Ma (Fig. 7). Being the Arquata quarry profile the final filling phase, the affinity with the adjacent basins leads to infer that a long-lasting slice of time is unexposed and it might relate to the whole Late Gauss-to-Reunion interval (3.05-2.15 Ma), part of which was calibrated at Toppetti quarry in Valtiberina, and in the Valdarno through the Matuyama chron including Olduvai (1.95-1.77 Ma). As Valdarno and Colfiorito basin covers reach a Jaramillo and Brunhes dating, such values seem perfectly fitting the magneto- stratigraphy results in Valle Umbra. 5.2 Geochronological and stratigraphic implications The magnetochronologic result, of the Jaramillo date (Figs. 3, 7), represents the basic point supporting all stratigraphical considerations to be referred to. Firstly, once this 55 m sampled upper profile would be consi- dered as deposited with constant sedimentation rate, the direct correlation between thickness and number of years of accumulation leads to assign 80 ky duration to the 14 m thick intermediate normal polarity level (Fig. 3). From this assumption it descends a duration of roughly 60 ky for the upper 11 m of reversed polarity and nearly 170 ky for the lower one 32 m thick, which includes the nearly 7 m unsampled section (Fig. 3). The overall dura- tion would be nearly 300 ky, which leads to an accumula- tion rate close to 20 m/100 ky, excluding compaction rate, fairly low when compared to similar lithologies ana- lyzed in Valtiberina and Valdarno basins. In these latter, in fact, the massive clay and/or of deeper lake deposition yielded a value of 24 m/100 ky, progressively reaching in the overlying levels, richer in silty sand, more that of 35 m/100 ky. More generally, the sedimentation rates calcu- lated for the two basins (NAPOLEONE et al., 2004) range from a minimum of 20 m/100 ky to a maximum of 35 m/100 ky, with averages of about 22 m/100 ky (Valtiberi- na) and 30 m/100 ky (Valdarno). Although the unknown variable of compaction rate, these values represent an agreeable datum point for the study section. In the lower levels of the Arquata quarry, sediments range from silty clay to silt and sand in the lacustrine facies, and are not as finely clayey as in Fosso Bianco Unit. On the other hand the alluvial plain sediments are lithologically ana- logous to the uppermost Valdarno deposits. With such a rate, the overall thickness of the Arquata profile would have been sensibly thicker, from 77 to 105 m, in contrast to the measured 65 m, which means that the inferred missing meters need an explanation. Some hypotheses can be provided. 1) It may be a consequence of compaction rates, which may reasonably differ from a basin to another, due The latest continental filling of Valle Umbra (Tiber Basin, central Italy) … 61 Fig. 7 - Magnetostratigraphic dating of Arquata quarry section and the Valle Umbra sedimentary sequences positioned within the Apen- nine Basins’ evolution (redrawn and modified after NAPOLEONE et al., 2003). Calibration of local mammal faunas (lmf) and of main Faunal Units (F.U.) to GPTS is also shown. Datazione magnetostratigrafica della sezione di Cava Arquata, ed inserimento delle sequenze sedimentarie della Valle Umbra nel qua- dro evolutivo dei Bacini Appenninici (da NAPOLEONE et al., 2003, ridisegnato e modificato). È riportata anche la calibrazione delle faune locali a mammiferi (lmf) e delle principali Unità Faunistiche (F.U.) con la scala GPTS. R. Bizzarri et al. 62 to local variations concerning petrology, grain size and organic matter amount (EINSELE, 1992; VAN ASSELEN et al., 2009, with refs.). 2) Contrary to Valdarno Basin, which on-sets on a more erodible sandstone substratum (Macigno s.l. AUCT.), Tiber Basin, particularly Valle Umbra, large- ly set up on a limestone substratum, and sand- stones (Marnoso-Arenacea s.l. AUCT.) only locally crop out. Thus, a primary difference in sediments supply must been considered too. 3) In the case of Arquata, the above reported geo- structural observations stressed the occurrence of the well visible fault cutting the profile, of minor in- terruptions due to the inclination (10°-12° SW) of the strata respect the steps of the quarry, as well as of fractures with a not significant dislocation in the upper profile. The latter have been sampled in de- tail on both sides for a better control. The sampled 55 m can be thence less than those actually depo- sited: although the disturbances seem at a direct view of minor size, two effects implicitly add to one another. On the one hand, the interval of normal polarity should be slightly thicker, leading to a larger accumulation rate and therefore closer to those measured in Valdarno and Valtiberina; on the other hand, duration in the adjacent intervals of reversed polarity is shortened and a certain amount of its thickness removed by this main fault, to become reduced significantly. It therefore appears quite like- ly an accumulation rate of 25-30 m/100 ky, which can become even higher in the final steps of filling - more alluvial and palustrine- as documented in the Valdarno basin (NAPOLEONE et al., 2004). In the lat- ter occasion, the control was made by the cyclostra- tigraphic distribution of the magnetic signature through a 0.5 my long sequence, enhancing at least five steps of increasing sedimentation rates. The consequence would be, for the present profile, a reduction of the erased pile of sediments to less than 25 m, which is significantly shorter than earlier implied but still much more than it appears to a di- rect view. 4) If we consider the lack of about 5 to 15 m of the succession as exclusively due to the fault occurring between lower and upper sections, an average se- dimentation rate of about 25-30 m/100 ky is still ap- propriate. To support the hypothesis, one more consideration may help, changing accumulation rates in the range from 24 m to 30 m/100 ky, with extremes reaching as much as 19 m to 35 m/ 100 ky: the datum is consistent with paleoenvironmental reconstruction, as the Jaramillo record occurs on both lacustrine clay and alluvial plain silty clay/ sand, that would remark a large shift in accumula- tion rates. At last, an average 25 m/100 ky sedimentation rate, as well as values comprised between 19 m/100 ky and 30 m/100 ky, are perfectly comparable to these of other Basins (Valdarno, Valtiberina). It seems plausible to assume for the top reversed interval of 11 m a date shortly younger than the upper limit of Jaramillo at 0.970 Ma, e.g., in the range of 0.940-0.920 Ma; for the lower reversed magnetozone a rate of 30 m/100 ky yields to a date ranging 1.050 Ma (on top) to one not older than 1.130-1.140 Ma (Fig. 3). Also assuming a lower average sedimentation rate (25 m/100 ky), the section bottom cannot be in any case considered older than 1.2 Ma. These time constrains, as well as the occurrence of Jaramillo, allow attributing the Arquata section to the end of Early Pleistocene, to which the Villafranchian-Galerian boundary has been also calibrated (Fig. 7). Mollusc and ostracod assem- blages agree with a Late Villafranchian age, perfectly consistent with the attribution of the long reversal inter- val (R1 and R2 magnetozones) to the Matuyama chron. Furthermore, the fossil assemblages in the lower sec- tion, which have been collected inside the R1 reversal interval, more than 10 m from the base of N1 interval, are considered as not older than Tasso F.U. Moreover, taxa such as V. belluccii, M. curvicosta and V. interposi- ta, as well as the occurrence of C. torosa, restrain the time interval to the Tasso and Farneta Faunal Units (GLIOZZI et al., 1997). Both Tasso and the successive Late Villafranchian F.U. (Farneta, Pirro) are calibrated to the reversal magnetopolarity interval successive to Ol- duvai, whereas to the latter are linked the transitional Olivola-Tasso faunas of Matassino and Poggio Rosso (NAPOLEONE et al., 2003). The biochronological record, indeed, is still consistent with the date of nearly 1.140 Ma at the base and 0.940-0.920 Ma at the top, although clearly Galerian faunas have not been documented. 5.3 Cyclicities in the magnetic signal as produced by climate changes in Valdarno and Valtiberina A short recalling is needed for the reported results from Valtiberina sediments also for their pollen content in cyclostratigraphic distribution, as well as a direct cor- relation with that in the Valdarno by the continuous dat- ing tied between them. Palynological data from the Val- darno Basin have already shown when in its layer beds the vegetation changes marked the onset of a decidedly oscillating pattern from deciduous forest or still subtropi- cal to an open field herbaceous cover of arid climate origin. Dates of such an onset and durations of the steady cyclic patterns were constrained by a high reso- lution accuracy provided by detailed magnetostratigra- phy to have occurred since Pliocene throughout the beginning of Pleistocene, and they tightly correspond to the ones enhanced by the magnetic signal, across a 1.5 Ma record from Kaena through Olduvai chrons. Each limit of the various magnetochrons there present provided tie points to which the cyclic oscillations are tuned, in order to control their rhythms and/or duration in the accumulation hiatuses. Therefore, grouping in few classes the continuously varying cyclicities, major changes range from 21 m/100 ky in the lowermost lacu- strine clay to 27 m/100 ky, 31 m/100 ky, up to 36 m/100 ky for the maximum influx of sand filling the deposit. When the signal processed in cycle length/meter is tuned to precisely timed tie points, the cycle length in years is expressed in intervals strikingly corresponding to the Milankovitch cyclicities. The shorter one among precession, obliquity and short eccentricity, prevails in the lower pile of sediments; at the date of 2.86-2.84 Ma a rather sudden increase of amplitude affects the obliq- uity cycle, while a marked eccentricity is already visible in the short profile of Olduvai time beginning just before the Gelasian-Crotonian boundary. When climate gets prevailingly driven by the short eccentricity cycle at dates close to Jaramillo, the shortness of time series in The latest continental filling of Valle Umbra (Tiber Basin, central Italy) … 63 the Arquata profile would barely allow to detect such a cycle, unless it was strongly recorded as it occurred in the Olduvai profile for the Valdarno. 6. CONCLUSIONS Two major conclusions may be stressed from the present study, both mainly descending from magneto- stratigraphic dating. First, the fluvial-lacustrine deposi- tion in Valle Umbra was over by nearly one million years ago. The inferred presence of sediments equivalent to the Montefalco conglomerates (Fig. 7), tectonically dis- placed in the nearby exposures, would imply a short du- ration for their deposition, which makes still rather long the missing time span until Present, i.e. up to the histor- ic and protostoric times. Secondly, the time length rec- orded at Arquata would not exceed 300 ky, which leads to roughly imply that only 10-15% of the possibly 3-to-1 Ma of the Tiber Valley infilling is here exposed. From the purely geochronologic point of view one may recall another conclusion, i.e., all recorded proxies benefit a date to which their alleged ages are calibrated; having the direct chronologic reference, adjustments may be done for correlation with other exposed sequences. Such is the case of the present biochronologic record, whose assessment in the calibrated frame of the Apen- nine basins may be accomplished: the Tasso F.U. es- tablished in the Valdarno is calibrated to a date lasting 1.6-to-1.7 Ma, right after the end of Olduvai chron. The malacofauna recovered at Arquata, therefore, is dated quite younger than the vertebrate Tasso F.U., and con- firms that biochronology by itself is a fairly weak tool for timing short intervals. A multidisciplinary approach, also involving magnetostratigraphy, if possible is therefore already preferable. Another important result is that the Arquata sec- tion revealed, in the present study, a number of similari- ties in its evolutionary trend with other basins of central Italy, as well as a clearer assessment of the geological events through the Tiber Valley. Its magnetostratigraph- ic dating provides a second constrain for Valle Umbra evolution (Fig. 7), after the attribution of Morgnano lig- nite to Triversa F.U., dated to 3.3 Ma in the Valdarno basin. With such time limits, the two main sedimentary cycles here recognized by COLTORTI & PIERUCCINI (1997) are still confined. The lower cycle can be linked to Pliocene, whereas the top of the second one date to the end of Early Pleistocene (Fig. 7). Thus, this study in Valle Umbra adds a decisive element to define, until the date of 1 Ma, the evolution of Tiber Basin in the paleo- geographic setting of Apennine basins. The interme- diate cycle, well documented in the Valdarno basin and commonly accepted although not dated also for Valtibe- rina (NAPOLEONE et al., 2003), still has not been recog- nized here, being probably buried below exposed sedi- ments. Arquata date across the Jaramillo chron helps en- hancing how facies associations in the eastern branch may be related to those in the western branch, the Valti- berina (Fig. 7). Here the earliest lacustrine deposition in the Fosso Bianco Unit, dated as old as late Gauss chron, constrains the overlying fluvial deposits of S.M. di Ciciliano Unit to be coeval with those of the Valdarno Basin final filling, but, while magnetostratigraphic data presently enable timing in detail the deposition of these two units of the Tiber Basin, it does not occur for the intermediate ones, for which it is desirable a direct time control. The immediate prospect is that several paleo- geographic restorations can be quantitatively controlled by timing their evolutionary extent. It is also worth noting that the measured dates for the final filling of Valle Um- bra Basin, marked also in the surrounding intermountain basins, are recorded in the marine hemipelagic deposits of the eastern Apennines belt of the Umbrian-Marchean realm, i.e. in the basins of the peri-Adriatic platforms (ALBIANELLI et al., 2003). Continental and marine sedi- ments belonging to the same chrons, are then coeval in deposition. Even in the marine conditions, the deposi- tional features followed similar rhythms, with a long con- tinuous phase in the earlier ages of Pliocene since Mammoth and Kaena chrons followed by a shorter in- terval crossing the Olduvai; more displaced are the youngest deposits, but all dates are strikingly corres- ponding on both sides of the Apennine divide (Fig. 7). On such prospects it would be important to pay careful attention on timing the intensive tectonics in the uplifting of peninsular Italy. ACKNOWLEDGEMENTS Authors wish to thanks the property and the work- ers of the Arquata quarry, and especially the responsi- ble of the quarry, Sig. Pietro Antano, for their courtesy and the help in the sampling. This work has also been possible thanks to the project: “Invito al Parco - Lungo i percorsi delle sue acque e della sua storia”, conceived and promoted by Prof. M.R. Trabalza. REFERENCES ALBIANELLI A., AZZAROLI A., BERTINI A., FICCARELLI G., NAPOLEONE G. & TORRE D. (1997) -Paleomagnetic and palinologic investigations in the Upper Valdar- no basin (central Italy): calibration of an early Villa- franchian fauna. Riv. It. Paleont. Strat., 103, pp. 111-118. ALBIANELLI A., MAGI M., MAZZINI M. & NAPOLEONE G. (2002) - The Plio-Pleistocene boundary in the Northern Apennine continental deposits as defined by the Faella magnetostratigraphic section in the Upper Valdarno. Boll. Soc. Geol. Ital., Special Vo- lume, 1, pp. 473-479. ALBIANELLI A., CANTALAMESSA G., DIDASKALOU P., MICA- RELLI A., NAPOLEONE G. & POTETTI M. (2003) - Magneto-stratigraphic dating of the Middle and Late Pliocene sequence in the Marchean Apen- nines, central Italy. Il Quaternario, 16 (1), pp. 171- 183. ALVAREZ W. (2009) - The historical record in the Scaglia limestone at Gubbio: magnetic reversals and the Cretaceous-Tertiary mass extinction. Sedimento- logy, 56 (1), pp. 137-148. AMBROSETTI P., BASILICI G., CAPASSO-BARBATO L., CAR- BONI M.G., DI STEFANO G., ESU D., GLIOZZI E., PE- TRONIO C., SARDELLA R. & SQUAZZINI E. (1995a) - Il Pleistocene inferiore nel ramo sud occidentale del bacino tiberino (Umbria): aspetti litostratigrafici e biostratigrafici. Il Quaternario, 8 (1), pp. 19-36. R. Bizzarri et al. 64 AMBROSETTI P., BASILICI G., CIANGHEROTTI A. D., CODI- PIETRO G., CORONA E., ESU D., GIROTTI O., LO MONACO A., MENEGHINI M., PAGANELLI A. & ROMA- GNOLI M. (1995b) - La foresta fossile di Dunarobba (Terni, Umbria, Italia centrale): contesto litostrati- grafico, sedimentologico, palinologico, dendrocro- nologico e paleomalacologico. Il Quaternario, 8(2), pp. 465-508. AMBROSETTI P., CONTI M.A., PARISI G., KOTSAKIS T. & U. NICOSIA (1977) - Neotettonica e cicli sedimentari plio-pleistocenici nei dintorni di Città della Pieve (Umbria). Boll. Soc. Geol. Ital., 96, pp. 605 - 635. AMBROSETTI P., CARBONI M.G., CONTI M.A., COSTANTINI A., ESU D., GANDIN A., GIROTTI O., LAZZAROTTO A., MAZZANTI R., NICOSIA U., PARISI G. & SANDRELLI F. (1978) - Evoluzione paleogeografica e tettonica nei bacini Tosco - Umbro - Laziali nel Pliocene e nel Pleistocene inferiore. Mem. Soc. Geol. Ital., 19, pp. 573-580. AMBROSETTI P., CARBONI M.G., CONTI M.A., ESU D., GI- ROTTI O., LA MONICA G.B., LANDINI B. & PARISI G. (1987) - Il Pliocene ed il Pleistocene inferiore del bacino del Fiume Tevere nell’Umbria meridionale. Geogr. Fis. Dinam. Quat., 10, pp. 10-33. AMBROSETTI P., CATTUTO C. & GREGORI L. (1989) - Geo- morfologia e neotettonica nel bacino di Tavernel- le/Pietrafitta (Umbria). Il Quaternario, 2, pp. 57-64. AZZAROLI A. (1977) - The Villafranchian stage in Italy and Plio-Pleistocene boundary. Giorn. Geol., 41, pp. 61-79. AZZAROLI A., DE GIULI C., FICCARELLI G. & TORRE D. (1988) - Late Pliocene to Early Pleistocene mam- mals in Eurasia: Faunal succession and dispersal events. Palaeogeogr. Palaeoclimatol. Palaeoecol., 66, pp. 77-100. BARCHI M., BROZZETTI F. & LAVECCHIA G. (1991) - Anali- si strutturale e geometrica dei bacini della media Valle del Tevere e della Valle Umbra. Boll. Soc. Geol. It., 110, pp. 65-76. BASILICI G. (1992) - Il bacino continentale tiberino (Plio- Pleistocene, Umbria): analisi sedimentologica e stratigrafica. Tesi di Dottorato, Università degli Studi di Bologna, 323 p. BASILICI G. (1995) - Sedimentologia della parte distale di un conoide alluvionale del Pliocene superiore (Ba- cino Tiberino, Umbria). Il Quaternario, 8, pp. 37-52. BASILICI G. (1997) - Sedimentary facies in an extension- al and deep - lacustrine depositional system: the Pliocene Tiberino Basin, Central Italy. Sedim. Geol., 109, pp. 73-94. BASILICI G. (2000) - Pliocene lacustrine deposits of the Tiberino Basin (Umbria, Central Italy). in E. H. Gierlowski - Kordesch, and K. R. Kelts, eds., Lake basins through space and time, AAPG Studies in Geology, 46, p. 505-514. BIZZARRI R., ALBIANELLI A., ARGENTI P., BALDANZA A., CO- LACICCHI R. & NAPOLEONE G. (2009) - Primi dati ma- gnetostratigrafici dalla Valle Umbra: la sezione di Cava Arquata (Bevagna, Umbria). Atti del Conve- gno “Ricerca, applicazione e comunicazione delle Scienze della Terra” (Preci, 27-28 Marzo 2009). BONINI M. (1998) - Chronology of deformation and analo- gue modelling of the Plio-Pleistocene “Tiber Basin” implications for the evolution of the Northern Apen- nines (Italy). Tectonophysics, 285, pp. 147-165. CAPRAI V. (2010) - Il Pleistocene continentale in Valle Umbra (Umbria Orientale): malacofaune ed ostra- cofaune dalla Cava Arquata - variazioni paleoeco- logiche e paleoclimatiche. Tesi di laurea, Universi- tà di Perugia, 132 p. CAVALLO P. & MARTINETTO E. (2001) - Flore carpologi- che del Pliocene di Castelletto Cervo (Biella). Boll. Mus. reg. Sci. nat. Torino, 18 (2), pp. 277-343. COLACICCHI R. & BIZZARRI R. (2005) - Late Pliocene - Early Pleistocene lacustrine facies assotiations in the Arquata quarry (Valle Umbra, Central Italy). Geoitalia Epitome, 1, p. 224. COLACICCHI R., ARGENTI P., BALDANZA A., BIZZARRI R. & PAUSELLI O. (2005) - Geological, lithostratigraphi- cal and paleontological analysis of Plio-Plei- stocene deposits of the Arquata quarry (Bevagna, Perugia, central Italy). Geoitalia Epitome, 1, p. 77. COLACICCHI R., BIZZARRI R., ARGENTI P., BALDANZA A. & PAUSELLI O. (2007) - Studio geologico multidisci- plinare della cava dell’Arquata. In: Trabalza M.R. & Colacicchi R. (Eds.) - Invito al Parco - Lungo i per- corsi delle sue acque e della sua storia. Dell’Ar- quata ed., Foligno, 293 pp. COLACICCHI R., BIZZARRI R., NAPOLEONE G., ALBIANELLI A., ARGENTI P. & BALDANZA A. (2009) - Magneto- stratigraphic update of Valle Umbra continental deposits from the Bevagna section (Perugia, Cen- tral Italy). Geoitalia Epitome, 3, p. 397. COLTORTI M. & PIERUCCINI P. (1997) - The southeastern Tiber Basin (Spoleto, Central Italy): geology and stratigraphy of Plio-Pleistocene sediments. Il Qua- ternario, 10, pp. 159-180. COLTORTI M., ALBIANELLI A., BERTINI A., FICCARELLI G., NAPOLEONE G. & TORRE D. (1998) - The Colle Cur- ti mammal site in the Colfiorito area (Umbrian- Marchean Apennines): stratigraphy and palynolo- gical analysis. Quat. Intern., 47, pp. 107-116. CONTI M.A. & GIROTTI O. (1977) - Il villafranchiano nel “Lago Tiberino”, ramo sud-occidentale; schema stratigrafico e tettonico. Geol. Rom., 16, pp. 67-80. EINSELE G. (1992) - Sedimentary Basins. Evolution, Fa- cies, and Sediment Budget. Springer-Verlag, Ber- lin, 628 p. FICCARELLI G., ABBAZZI L., ALBIANELLI A., BERTINI A., COLTORTI M., MAGNATTI M., MASINI F., MAZZA P, MEZZABOTTA C., NAPOLEONE G., ROOK L., RUSTIO- NI M. & TORRE D. (1997) - Cesi, an early Middle Pleistocene site in the Colfiorito Basin (Umbrian- Marchean Apennines), central Italy. J. Quat. Sc., 12, pp. 507-518. GIBBARD P. L., HEAD M. J., WALKER, M. J. C. & THE SUB- COMMISSION ON QUATERNARY STRATIGRAPHY (2010) - Formal ratification of the Quaternary Sys- tem/Period and the Pleistocene Series/Epoch with a base at 2.58 Ma. J. Quaternary Sci., 25, pp. 96-102. GIROTTI O., CAPASSO BARBATO L., ESU D., GLIOZZI E., KOTSAKIS T., MARTINETTO E., PETRONIO C., SAR- DELLA R. & SQUAZZINI E. (2003) - The section of Torre Picchio (Terni, Umbria, Central Italy): a Vil- lafranchian site rich in vertebrates, molluscs, ostra- cods and plants. Riv. It. Paleont. Strat., 109 (1), pp. 77-99. GLIOZZI E., ABBAZZI L., ARGENTI P., AZZAROLI A., CALOI L., CAPASSO-BARBATO L., DI STEFANO G., FICCA- RELLI G., KOTSAKIS T., MASINI F., MAZZA P., MEZ- The latest continental filling of Valle Umbra (Tiber Basin, central Italy) … 65 ZABOTTA C., PALOMBO M.R., PETRONIO C., ROOK L., SALA B., SARDELLA R., TORRE D. & ZANALDA E. (1997) - Biochronology of selected Mammals, Mol- luscs and Ostracods from the Middle Pliocene to the Late Pleistocene in Italy. The state of the art. Riv. It. Paleont. Strat., 103, pp. 368-388. GREGORI L. & CATTUTO C. (1986) - Elementi geomorfo- logici ed episodi di tettonica recente nei dintorni di Spoleto (Umbria). Boll. Soc. Geol. Ital., 105, pp. 173-184. LOTTI B. (1917) - L’antico Lago Tiberino e le ligniti del- l’Umbria. Miniera Ital., 1, pp. 274-275. LOTTI B. (1926) - Descrizione geologica dell’Umbria. Mem Descr. Carta Geol. d’It., 21. MARTINI I. P. & SAGRI M. (1993) - Tectono-sedimentary characteristics of Late Miocene - Quaternary ex- tensional basins of the North Appennines, Italy. Earth Sc. Rev., 34, pp. 197-223. MENICHETTI M. (1997) - Analisi geologico-strutturale dell’area di Pietrafitta-Tavernelle (Umbria occiden- tale, Italia centrale). Il Quaternario, 10, pp. 3-14. NAKAGAWA H., NIITSUMA N., TAKAYAMA T., MATOBA Y., ODA M., TOKUNAGA S., KITAZATO H., SAKAI T. & KOIZUMI I. (1997) - The magnetostratigraphy of the Vrica section, Italy, and its correlation with the Plio-Pleistocene of the Boso peninsula, Japan. In: Van Couvering (Ed.), The Plio-Pleistocene Boun- dary and the Beginning of the Quaternary. World and Regional Geology Series, 9, XI-XVII. Cam- bridge University Press, pp. 46-56. NAPOLEONE G. & AZZAROLI A. (2002) - A collection of fossil vertebrates from the Upper Valdarno (central Italy) calibrated in the end Pliocene to span 220,000 years across the Olduvai magnetochron. Riv. It. Paleont. Strat., 108, pp. 479-492. NAPOLEONE G. & RIPEPE M. (1989) - Cyclic geomagnetic changes in Mid-Cretaceous rhytmites, Italy. Terra Nova, 1 (5), pp. 437-442. NAPOLEONE G., ALBIANELLI A., AZZAROLI A., BERTINI A., MAGI M. & MAZZINI M. (2003) - Calibration of the Upper Valdarno Basin to the Plio-Pleistocene for correlating the Apennine continental sequences. Il Quaternario, 16 (1Bis), pp. 131-166. NAPOLEONE G., ALBIANELLI A. & FISCHER A.G. (2004) - Magnetic susceptibility cycles in Upper Pliocene la- custrine deposits of the Northern Apennines, Italy. In: Cyclostratigraphy: Approaches and Case Histo- ries. SEPM Special Publication, 81, pp. 263-274. PETRONIO C., ARGENTI P., CALOI L., ESU D., GIROTTI O. & SARDELLA R. (2000-2002) - Updating villafranchian molluscs and mammal faunas of Umbria and La- tium (central Italy). Geol. Rom., 16, pp. 369-387. PONTINI M.R., ALBIANELLI A., BASILICI G., BERTINI A. & NA- POLEONE G. (2002) - Palynology and magnetostrati- graphy of the Middle - Late Pliocene lacustrine se- quence in the Tiberino basin (central Italy). Boll. Soc. Geol. It., Vol. spec. 1 (2), pp. 467-472. SIMONETTI I. (1996) - Analisi paleontologica dei mollu- schi continentali del Pleistocene inferiore della Cava di Bevagna (Perugia). Tesi di laurea, Univer- sità di Roma “La Sapienza”, 54 p. VAN ASSELEN S., STOUTHAMER E. & VAN ASCHET TH.W.J. (2009) - Effects of peat compaction on delta evolu- tion: A review on processes, responses, measur- ing and modeling. Earth-Science Reviews, 92, pp. 35-51. VAN COUVERING J.A. (1997) - Preface. In: Van Couver- ing (Ed.), The Plio-Pleistocene Boundary and the Beginning of the Quaternary. World and Regional Geology Series, 9, XI-XVII. Cambridge University Press. Ms. ricevuto il 12/05/2010 Testo definitivo ricevuto il 5/09/2010 Ms. received: May 12, 2010 Final text received: September 5, 2010