base PLIO-QUATERNARY GEOLOGICAL EVOLUTION OF THE HIGH SALTO RIVER VALLEY (CENTRAL ITALY): THE MARANO DE’ MARSI UNIT Edi Chiarini1, Marco Giardini2, Massimo Mattei3, Felicia Papasodaro1, Massimiliano Porreca3 & Laura Sadori2 1Dipartimento Difesa del Suolo – Geological Survey of Italy, ISPRA, Roma edi.chiarini@isprambiente.it 2Dipartimento di Biologia Vegetale, Sapienza University of Rome 3Dipartimento di Scienze Geologiche, Roma Tre University ABSTRACT: Chiarini E. et al., Plio-Quaternary geological evolution of the high Salto river valley (Central Italy): the Marano de’ Marsi unit. (IT ISSN 0394-3356, 2009). This paper summarizes the results of multidisciplinary researches carried out in a wide span of time on the earliest continental depo- sits of the high Salto River valley. The study led us to characterize sedimentary environments, to analyze their relationships and to pro- pose a chronological attribution. Facies are referable to a complex depositional system consisting of a Gilbert-type lacustrine delta and of slope-type partly interfingered fan deltas coming from the eastern border of the basin. The geological data collected allowed us to refer the succession to a single sedimentary cycle; in Marano de’ Marsi area the paleodrainage and progradation directions were also recognized. To better constrain the chronological, paleoclimatic and paleoenvironmental context, detailed investigations have been conducted on pelitic facies of the deltaic-lacustrine system in two sampling sites (an artificial outcrop at Marano de’ Marsi village and a borehole near Borgorose village). Paleomagnetic analyses show that Marano de’ Marsi and Borgorose successions have a normal polarity and generally low magnetic susceptibility values. Pollen diagrams from both sites mainly record a mid to high elevation forest pollen rain. The existence of trees from different vegetation belts suggests the presence of a well developed mountain system in the surroundings. In the case of Marano de’ Marsi section, these vegetation phases alternate with sudden, strong and short spreads of temperate and subtropical taxa. These alternations suggest that important climate changes occurred, with cyclical forest variations typical of glacial / interglacial periods of Pliocene and Early Pleistocene. Pollen and paleomagnetic data led us to refer both records to the Olduvai sub- chron, at the end of the Pliocene; nevertheless, the possibility that the records deposited during the Gauss chron cannot be, defini- tely, excluded. Field geological data and magnetic fabric results suggest that the Marano de’ Marsi unit sedimentation took place in a basin develo- ped under extensional tectonic regime. RIASSUNTO: Chiarini E. et al., Evoluzione geologica plio-quaternaria dell’alta valle del Fiume Salto: l’unità di Marano de’ Marsi. (IT ISSN 0394-3356, 2009). Il lavoro sintetizza i risultati di studi multidisciplinari condotti, nel corso di numerosi anni, sui depositi continentali più antichi dell’alta valle del Fiume Salto. Attraverso tali studi è stato possibile caratterizzare gli ambienti di sedimentazione e le relazioni fra essi esistenti, oltre che fornire indicazioni cronologiche più precise. Il rilevamento dettagliato di terreno ha consentito il riconoscimento delle facies che compongono tali depositi e la ricostruzione della loro distribuzione, permettendone l’attribuzione prevalentemente ad un delta lacustre di tipo Gilbert e subordinatamente a delta-conoidi, originati da canaloni o da corsi d’acqua a basso grado di maturità. I dati di terreno raccolti in questo studio fanno ritenere che l’unità di Marano de’ Marsi rappresenti il risultato di un unico ciclo sedimen- tario, sebbene molto articolato perché attribuibile ad un sistema deposizionale complesso. Nell’area di Marano, l’esame dei foreset e delle altre strutture sedimentarie riconosciute ha fornito dati importanti per la ricostruzione del verso del drenaggio e della progradazio- ne del delta. Per precisare il quadro paleoclimatico, paleoambientale e cronologico, sono state condotte indagini paleomagnetiche e palinologiche di dettaglio su due successioni affioranti nelle località di Marano de’ Marsi e Borgorose, in corrispondenza delle facies pelitiche del sistema deltizio/lacustre. Le analisi paleomagnetiche indicano che entrambe le successioni studiate hanno polarità magnetica normale e valori di suscettività magnetica relativamente bassi, in accordo con la composizione dei sedimenti essenzialmente carbonatica. I record pollinici registrano prevalentemente fasi forestali con elementi di media ed alta quota. La presenza di alberi tipici di diverse fasce vegetazionali suggerisce la presenza di un sistema montano ben sviluppato nelle vicinanze. A Marano de’ Marsi queste fasi vege- tazionali si alternano con espansioni improvvise, consistenti e di scarsa durata di alberi temperati e subtropicali. Queste alternanze inducono a pensare che si siano verificati importanti cambiamenti climatici nel periodo considerato, con variazioni forestali cicliche, tipiche dei periodi glaciali/ interglaciali che vanno dal Pliocene Medio fin quasi alla fine del Pleistocene inferiore. L’interpretazione congiunta dei risultati paleomagnetici e palinologici suggerisce l’attribuzione di entrambe le successioni al Pliocene. Pur non potendo del tutto escludere l’attribuzione al crono Gauss (Pliocene Medio), il confronto con lavori palinologici ottenuti da sedi- menti continentali e marini italiani, insieme a considerazioni di geologia regionale, fanno propendere per un’attribuzione di entrambe le successioni al subcrono Olduvai (Pliocene Superiore). I dati geologici e i risultati di anisotropia della suscettività magnetica suggeriscono che la sedimentazione dell’unità di Marano de’ Marsi sia avvenuta in un bacino sviluppatosi sotto un regime tettonico di tipo estensionale. Keywords: Deltaic-lacustrine facies; Pollen; Plant macrofossils; Magnetostratigraphy; Pliocene; Salto Valley; Central Apennines. Parole chiave: Facies deltizie-lacustri; Polline, Macrofossili vegetali; Magnetostratigrafia; Pliocene; Valle del Salto; Appennino centrale. Il Quaternario Italian Journal of Quaternary Sciences 22(2), 2009 - 325-344 326 E. Chiarini et al. 1. INTRODUCTION A complex interaction among tectonic uplift, nor- mal fault activity, erosional events and depositional episodes characterizes the Plio-Pleistocene evolution of the Central Apennines. The main features of the most recent phases of this tectono-sedimentary evolution are generally well known. In the contrast, the definition of the earlier phases needs to be improved, although a general correlation of the sedimentary infill of the main intermountain basins had already suggested (BOSI & MESSINA, 1991; BOSI et al., 2003). To this aim strati- graphic and sedimentological studies of continental deposits can be usefully integrated with analyses of erosional paleosurfaces and multidisciplinary research. In the high valley of the Salto River, near the Latium and Abruzzi regions boundary, wide outcrops of continental deposits testify the early steps of the Plio- Quaternary evolution of the area; the sediments have been deposited on the margin of or within a lake no more recognizable in the present-day landscape. These deposits have been studied in the frame- work of the field survey carried out for the Sheet n° 367 of the new Geological Map of Italy at a scale 1:50.000 (SERVIZIO GEOLOGICO D’ITALIA, 2005) and of the same Sheet of the Geomorphological Map ( S E R V I Z I O GEOLOGICO D’ITALIA, 2008). The study has been integrat- ed with paleomagnetic and paleobotanic investigations to better constrain the age of the deposits and to define the paleoclimatic and paleoenvironmental context. In this paper we report the main results of these studies and some considerations on the geological and tectonic history of the sedimentary basin. 2. GEOLOGICAL AND STRUCTURAL SETTING The studied area is located in the high Salto River valley (Fig. 1), between the Carseolani Mts. to the SW, and the Mt. Velino Chain to the NE. These mountain ranges are mainly formed by Lower Cretaceous- Miocene limestones and dolostones, referable to the Latium-Abruzzi carbonatic platform paleogeographic domain (COMPAGNONI et al., 2005). The siliciclastic deposits overlying carbonatic units, related to the evo- lution of the Apennines Tortonian-Messinian foredeep basin (BELLOTTI, 1992; CIPOLLARI & COSENTINO, 1995), crop out south-eastward (Palentini Planes), northward (Malito Valley) and north-westward (Salto Lake), but are lacking in this valley sector. Along the valley axis and in the nearby area (Fig. 2), continental deposits of gravitative, alluvial, and Fig. 1 - Shaded relief of the Apennine sector between the Fucino and Rieti intermontane basins and location of the study area. Modello ombreggiato del settore appenninico compreso fra la Piana del Fucino e la Piana di Rieti ed ubicazione dell’area in studio. 327Plio-Quaternary evolution of the high Salto river valley ... lacustrine origin occur. They are organized in several depositional cycles and are separated by stratigraphic unconformities related to major erosional events (CHIARINI et al., 2007; 2008). The oldest continental deposits, described in this study, crop out between Marano de’ Marsi (AQ) and Borgorose (RI) villages. In previous studies, they have been attributed to different sedimentary events. In the geological sheet n° 367 “Tagliacozzo” (S E R V I Z I O GEOLOGICO D’ITALIA, 2005), authors described a first allu- vial unit and a following succession recording a pro- gressive evolution from an alluvial to a lacustrine envi- ronment, unconformably covered with breccias crop- ping out in the Spedino and Borgorose areas. All these units have been doubtingly related to the Pliocene. BUSINO (1990) and FEDERICI (1990) recognized two main depositional events. The first event, Messinian or Pliocene in age, produced lacustrine clayey sediments (Fosso S. Martino Formation); the second one consists of two formations (Torano and Marano de’ Marsi forma- tions), which correlated each other and referable to the Pliocene, composed of several members. BO S I & FEDERICI (1993) correlated part of these deposits to the first continental cycle of the Fucino basin (S. Pelino Succession of BERTINI & BOSI, 1976 and Aielli Complex of BOSI et al., 1995), ascribed to Pliocene. For the gene- ral shortage of dating materials, these chronological data are essentially provided on the basis of morpho- stratigraphic observations and analogies with other successions of Central Apennines. The Salto River valley lies just to the west of the Apennines watershed, between the Fucino and the Rieti basins, and follows regional tectonic lineaments, corre- sponding to a NW-SE fault system with a multiphase tectonic activity (NIJMAN, 1970; BOSI et al., 1994; BIGI & COSTA PISANI, 2003). During Plio-Pleistocene a segment of the fault system (Salto valley fault), with a main nor- mal fault activity, controlled the location and the evolu- tion of a continental basin (GALADINI et al. 2000; 2003), in the intermediate segment of the Salto valley, where the artificial homonymous lake is presently located (Fig. 1). This basin has been mostly obliterated by subsequent tectonic and morphogenetic events. In the high valley the tectonic structure shifts eastward. Indeed a morphologically evident fault along the base of Duchessa mountain range (Mt. Velino Chain, see Fig. 1) shows an en échelon geometry (GALADINI & MESSINA, 2001; 2004; MOREWOOD & ROBERTS, 2001) with the Salto valley fault above mentioned. An important activity phase of Duchessa Mts. fault has caused the individualization and the evolution of the small Corvaro intermountain basin (Fig. 2), since the Early Pleistocene according to BOSI & FEDERICI (1993). These tectonic events induced a dramatic change in the paleogeography of the area. They made it difficult to recognize the sedimentary basin of the oldest conti- nental deposits and the faults responsible for its struc- turing. Nevertheless, available data make it possible to assume that: 1) sedimentary events considered in this paper took place inside an extensional basin; 2) the tectonic structures that controlled the evolution of this ancient basin could be the same as that responsible for the Corvaro basin setting. Indeed, the tilting of the succession, generally 10° to 30° toward north-east, suggests the localization of the main fault on the eastern basin flank. Moreover, the several tectonic elements affecting the studied sedi- mentary succession are related to extensional tectonic regime, although referable to different phases. NW-SE and NNW-SSE trending faults confine deposits in a nar- row and elongated belt inside the carbonatic bedrock (Fig. 2). Along the western margin, the contact of the deposits with the Meso-Cenozoic limestones consists of rectilinear and high-angle faults, segmented by minor transversal faults. Along the eastern border, the bound- ary is generally tectonic as well; only locally, deposits unconformably rest on the calcareous bedrock. These faults could belong to the Duchessa Mts. fault system, even if they show a moderate morphological evidence (Fig. 2). E-W trending minor faults also displace the Marano de’ Marsi unit. At present, a wide and flat erosional surface (Colle Vena Francesca paleosurface) almost obliterates the offsetting of the above cited faults; it has been recog- nized at about 950 m a.s.l. on both the marine carbon- atic substratum and the deposits examined in the pre- sent work (Fig. 2). In a preliminary study, CHIARINI et al. (1997) have shown that the Colle Vena Francesca paleosurface had formed before the Corvaro depression setting: in fact it shows considerable displacement across the Duchessa Mts. fault. The paleosurface results from the complex inter- action of depositional events, erosional phenomena and activity phases of faults; CENTAMORE & NISIO (2002a; b) have related it to the Fontanelle surface located in the mid Salto valley. Therefore, the wide extension of the Colle Vena Francesca paleosurface would highlight its importance in the stratigraphic frame of the area and testifies an important change in the basin subsidence trend. 3. STRATIGRAPHIC AND SEDIMENTOLOGICAL FEATURES The succession, named Marano de’ Marsi unit, crops out discontinuously along the left flank of the Salto Valley, close to Marano de’ Marsi village, at alti- tudes comprised between 959 m and 710 m a.s.l.; it also crops out between Torano and Borgorose, from 670 m to 951 m a.s.l. (Fig. 2). The maximum thickness preserved from erosion is of about one hundred meters. It is composed of lacustrine marls, claystones, siltstones and arenites and by arenites, gravels and conglomerates of fluvial, deltaic and slope environment, which overlay the carbonatic substratum. This conti- nental succession ends with the above mentioned Colle Vena Francesca erosional surface (Fig. 2); it is uncon- formably covered with the alluvial fan systems of Teve- Ruara valleys in the Torano area and of Valle Amara in the Borgorose area. Deposits of younger cycles sedi- mented after the infilling and dissection of the Corvaro basin and have been referred to the glacial phases of Upper Pleistocene (SERVIZIO GEOLOGICO D’ITALIA, 2008). To collect stratigraphic and sedimentological data, a geological survey was performed at 1:10.000 scale, supported with photo-geological analysis, obser- vations of natural and artificial outcrop sections and of 328 E. Chiarini et al. 329 laboratory samples. A great facies variability was observed. The lack of important erosional discontinuities and the interbed- ding of lithofacies allow us to refer this succession to a single sedimentary cycle. Depositional structures described in this paper were mainly referred to a lacus- trine Gilbert-type delta for the occurrence of the typical three stratigraphic components (bottomset, foreset and topset; Fig. 3); secondly they were connected to slope- type fan-deltas with a high gradient. Sharp facies varia- tions frequently mark the emergence of faults belonging to the main Apennine system or transversal tectonic elements as observed west of Marano de’ Marsi and near Torano. The recognized facies can be grouped into three main facies associations (A, B, C; Fig. 2) on the basis of their meaning, relationships and spatial dis- tribution. Facies association A This association groups prevailingly ruditic facies referred to the lacustrine Gilbert-type delta, because in many cases their differentiation is made difficult by the bad outcropping conditions. Conglomerates with carbonatic pebbles, suban- gular or subrounded, sub-centimeters to centimeters in size widely crop out in the area; the sediments are clast supported, with a pebbly to muddy matrix, cross-bed- ded to plane-bedded foresets and moderately sorted beds and laminations; massive coarse-grained sandy beds and decimetric to metric pelitic intervals are fre- quently present. Pedorelicts and sequences referable to paleo-channel filling have also been recognized. These facies occur at Mt. Pago Osco and near Marano de’ Marsi and Torano villages (Figs. 2, 4 and 5). Between Torano and Colle Scarciofano (Fig. 5), in the upper portion of the outcropping succession a lenticular body of well stratified sandstones, several meters thick, occurs; it is characterized by parallel and cross bedded laminations, well cemented strata, oxides rich levels, and thin coarse grained horizons from sand- Fig. 3 - Longitudinal interpretative cross section of Marano de’ Marsi area. 1) recent alluvial and colluvial deposits (Upper Pleistocene - Olocene); 2) Marano de’ Marsi unit - sub-rounded to rounded coarse grained conglomerates of Gilbert-type delta topsets; 3) Marano de’ Marsi unit - subangular to subrounded fine grained conglomerates with arenitic levels (a) of Gilbert-type delta foresets; 4) Marano de’ Marsi unit - pelitic and ruditic deposits of Gilbert-type delta bottomsets; 5) carbonatic bedrock; 6) boundary: (a) etheropic, (b) ero- sional of low order, (c) erosional of high order ; 7) fault. Sezione interpretativa longitudinale dell’area di Marano de’ Marsi. 1) depositi alluvionali e colluviali recenti (Pleistocene superiore- Olocene); 2) unità di Marano de’ Marsi – conglomerati con clasti grossolani, da subarrotondati ad arrotondati, dei topset del delta Gilbert; 3) unità di Marano de’ Marsi – conglomerati a granulometria fine, con clasti da subangolari a subarrotondati, con livelli arenitici (a) dei foreset del delta Gilbert; 4) unità di Marano de’ Marsi – depositi pelitici e ruditici dei bottomset del delta Gilbert; 5) substrato carbonatico; 6) contatto: (a) eteropico, (b) erosionale di basso ordine, (c) erosionale di alto ordine; 7) faglia. Plio-Quaternary evolution of the high Salto river valley ... Fig. 2 - Geological sketch map of the study area. 1) Quaternary deposits; 2) Marano de’ Marsi unit (Middle?-Upper Pliocene) - alluvial- deltaic deposits of facies association A: fine to coarse grained conglomerates and gravels of Gilbert-type delta topsets (a) and fore- sets with arenitic levels (b); 3) Marano de’ Marsi unit - lacustrine-deltaic deposits of facies association B: marls, marly claystones, silt- stones and conglomerates of Gilbert type delta bottomsets and frankly lacustrine environment; 4) Marano de’ Marsi unit - slope-type fan deltas deposits of facies association C: breccias with laminated yellow siltstones, calcarenites, marls and claystones; 5) fault; 6) inferred or buried fault; 7) bedding strike and dip; 8) Colle Vena Francesca paleosurface; 9) Marano de’ Marsi (Tore) section; 10) Borgorose (Fornace dei Laterizi) borehole. Carta geologica schematica dell’area in studio. 1) depositi quaternari; 2) unità di Marano de’ Marsi (Pliocene Medio?- Superiore) - de- positi fluvio-deltaici dell’associazione di facies A: ghiaie e conglomerati da fini a grossolani dei topset (a) e foreset con livelli arenitici (b) del delta Gilbert; 3) unità di Marano de’ Marsi – depositi deltizi e lacustri dell’associazione di facies B: marne, argilliti marnose, siltiti e conglomerati dei bottomset del delta Gilbert e di ambiente francamente lacustre; 4) unità di Marano de’ Marsi – depositi di delta conoi- de dell’associazione di facies C: brecce con siltiti gialle, calcareniti, marne e argilliti laminate; 5) faglia; 6) faglia incerta o sepolta; 7) gia- citura degli strati; 8) paleosuperficie di Colle Vena Francesca; 9) sezione di Marano de’ Marsi; 10) sondaggio di Borgorose (Fornace dei Laterizi). Þ 330 stones to fine breccias. In these strata, small remnants of plants in a physiological position are often present. Paleocurrents direction, evaluated using foresets geometry and clasts embrication, varies from NW to NE, suggesting a drainage from south to north. This facies have been referred to the frontal part of the Gilbert-type delta. Coarse grained conglomerates, moderately sort- ed, poorly stratified, with bedding thickness ranging from 0.5 m to 5 m crop out on top of succession; deposits are clast supported with a pebbly-sandy matrix and pebbles centimeters to pluricentimeters in size, from sub-rounded to rounded. Between the ruditic beds decimetric intercalations of finer sediments are present; these consist of thin horizons of claystones and beds of oxides-rich yellow calcarenites, with paral- lel and cross lamination. The transition to the overlying ruditic beds is given by a decimetric level of fine grained breccias. Moulds related to dry up cracks are noticeable between claystones and calcarenites. These deposits extensively crop out in the area between Mt. S. Angelo and Colle Scarciofano (Figs. 2, 4 and 5) and are interpreted as topset of the Gilbert-type delta system, referred to an alluvial or deltaic plain envi- ronment. Gravel lithology of the facies association is exclu- sively carbonatic, whereas finer grain-size classes are also siliciclastic. Facies association B This facies association is composed of prevailing- ly pelites, consisting of marls, carbonatic siltstones and silty claystones, with a grey to dark-grey color, mas- sive, poorly stratified to thinly laminated. Whitish, light, limy-clayey horizons, rich of fresh water diatoms, are also present. Pelitic deposits include conglomerate bodies; in this context the transition from pelitic litholo- gies to the overlying ruditic beds is sharp and consists of an etherometric gravelly horizon with muddy matrix and thin levels of silty claystones with shell fragments. These deposits crop out northwest of Marano de’ Marsi village, in Valle di Tesio locality; they are also pre- sent in a narrow belt between the Rosa Mt. slope and the Laduschio Mt.-Castelvecchio Mt. relief, and then in a wide outcrop in the Borgorose area. At the base of the Saticone Mt. slope, the facies association includes intraformational slumps consisting of a close alternation of carbonatic siltstones and clayey marls, deformed by mesofolds and joints. Ruditic facies also include carbonatic metric boulders and chaotic facies, as visible at Valle di Tesio; these elements probably indicate a tectonic instability along the depositional slope. The observed facies have been ascribed either to a frankly lacustrine environment or to a distal portion of the deltaic system (bottomset). Fresh water ostracods referable to the genus Fig. 4 - Upper portion of facies association A exposed at Mt. Saticone near Marano de’ Marsi. a) Outcrop of cross-bedded fine grai- ned conglomerates; b) Detail of moderately sorted cross-bedded foresets; c) Sedimentological log. 1) cross-bedded fine grained con- glomerates of delta-front facies (foresets); 2: poorly stratified, well rounded coarse grained conglomerates of alluvial-deltaic plain facies (topsets); c: clay; s: silt; sa: sand; g: fine gravel; pe: pebble; co: cobble. Porzione superiore dell’associazione di facies A esposta al Mt. Saticone, nei pressi di Marano de’ Marsi. a) Affioramento di conglome- rati a granulometria fine e a stratificazione incrociata; b) Dettaglio degli orizzonti a stratificazione incrociata moderatamente classati; c) Log sedimentologico. 1) conglomerati a granulometria fine e a stratificazione incrociata del fronte deltizio (foreset); 2) conglomerati a clasti grossolani e ben arrotondati, scarsamente stratificati, di piana alluvionale e deltizia (topset); c: argilla; s: silt; sa: sabbia; g: ghiaia fine; pe: ghiaia media; co: ghiaia grossolana. E. Chiarini et al. Candona s.p. and Erpetocypris s.p., gasteropoda frag- ments and characee thalluses (PAMPALONI L. & PICHEZZI R.M., Servizio Geologico d’Italia, personal communica- tion), are observed inside the pelitic sediments. Facies association C This facies association consists of breccias with metric and plurimetric bedding or poorly defined strata, containing sub-angular to sub-rounded carbonatic clasts ranging in size from 0.5 to 10 cm, within a sandy- pebbly matrix. Breccias are poorly organized, clast sup- ported or, less frequently, matrix supported; inverse gradations also occur. Thin normal graded interbed- dings of small sized breccias, laminated yellow silt- stones, calcarenites, marls and claystones are present. In some cases these facies rest on top of lacustrine claystones via clastic horizons rich of shell fragments. T h e a s s o c i a t i o n m i g h t i n c l u d e r u d i t i c d e p o s i t s r e f e r a b l e t o b o t t o m s e t s o f t h e m a i n Gilbert-type delta, for the etheropic relationships between each other and the close facies similari- ty. It crops out in the area comprised between Torano and Borgorose villages, where it is locally covered with ruditic facies of the association A. According to sedimentary structures, ascribed either to a mass transport or, less frequently, to tractive flows, and because of the general lack of a clear foreset facies, this association has been referred to slope-type fan deltas which was originated from stream channels of a low order with a very small and steep catchment area. The facies distribution is consistent with a sedi- ment supply from the eastern border of the basin. 4. THE STUDIED STRATIGRAPHIC LOGS In order to better define the age of the Marano unit and to have indications on the paleoclimate that conditioned the sedimentary events, paleomagnetic and paleobotanic investigations have been carried out on fine sediments of two successions (SADORI, 2005). A first one, called Marano de’ Marsi section, is constitut- ed by deposits belonging to the facies association B and referred to a distal portion of the Gilbert delta. A second succession has been studied through a 24 m long core, drilled close to Borgorose village and there- fore called Borgorose core. It is characterized by clays and silts referred to a frankly lacustrine environment. 4.1 Marano de’ Marsi section This section is located in a marginal position of the Salto valley, at an elevation of 720 m a.s.l. (Fig. 2). The vertically cut surface is about 11 m high. Deposits are rather homogeneous and for about 10 meters from the bottom lacking in significant discon- 331 Fig. 5 - Upper portion of facies association A exposed at Colle Scarciofano. a) Outcrop of rounded coarse grained conglomerates (topsets). b) Sedimentological log. 1) stratified sandstones with parallel- and cross-bedded laminations of delta-front facies (foresets); 2) cross-bedded fine to coarse grained conglomerates of delta-front facies (foresets); 3) claystones, oxides-rich yellow calcarenites and fine grained breccias; 4) poorly stratified, sub-rounded to rounded, coarse grained conglomerates of alluvial-deltaic plain facies (topsets); c: clay; s: silt; sa: sand; g: fine gravel; pe: pebble; co: cobble. Porzione superiore dell’associazione di facies A esposta al Colle Scarciofano, nei pressi di Borgorose. a) Affioramento di conglomerati a granulometria grossolana (topset). b) Log sedimentologico. 1) arenarie stratificate con laminazioni parallele e incrociate; 2) conglomerati a granulometria da fine a grossolana, a stratificazione incrociata del fronte del delta; 3) argilliti, calcareniti gialle, ricche di ossidi e lami- nate, e brecce a granulometria fine in intercalazioni decimetriche; 4) conglomerati a clasti grossolani da subarrotondati ad arrotondati, mal stratificati, di piana alluvionale e deltizia (topset) c: argilla; s: silt; sa: sabbia; g: ghiaia fine; pe: ghiaia media; co: ghiaia grossolana. Plio-Quaternary evolution of the high Salto river valley ... formities; they consist of pale to dark grey silty marls and marly claystones and siltstones, with massive to thinly laminated beds and horizons (Fig. 6a). Sandy oxi- dized laminae, a few millimetres thick, cross-laminated levels and muddy pebbles, also occur. At the top, a weakly north verging cut surface marks the transition to an intraformational slump of laminated siltstones and marly claystones, including microfolds and fractures. At the base of the section, whitish horizons of particularly light clayey marls with fresh water diatom remnants, crop out. These sediments have been subject of two spe- cific samplings, i.e. paleomagnetic and pollen and macrofossil plant sampling. 4.1.1 Paleomagnetic analyses Paleomagnetic sampling and laboratory methods In the Marano de’ Marsi section, we sampled 154, 2.5 cm in diameter, cores (Fig. 6). Cores have been drilled with a ASC-280 gasoline-powered drill and ori- ented in situ with a magnetic compass. The average stratigraphic distance between the sampled cores was 10 cm, along the whole outcropping section. The uppermost part of the section, below the slumping, was sampled more in detail in order to study the timing rela- tionships between slump and the acquisition of rema- nent magnetization. Most of the paleomagnetic analyses were carried out at the paleomagnetic laboratory of the Department of Geological Sciences of Roma Tre University, some of them at ETH Paleomagnetic Laboratory of Natural Magnetism in Zurich. In particular, low-field magnetic susceptibility (Km) and its anisotropy were measured in the paleomagnetic laboratory of Roma Tre University using an AGICO KLY-3 kappabridge. Successively, the natural magnetic remanence (NRM) was measured using a JR6 spinner magnetometer and a 2G-Enterprise SQUID cryogenic magnetometer at Roma Tre University and ETH Zurich, respectively. Magnetic susceptibility analyses Low-field magnetic susceptibility (Km) was mea- sured along the whole section in 10 cm 3 volume speci- mens. Km values include the contribution of all magnet- ic minerals (diamagnetic, paramagnetic and ferromag- netic grains) present in the sediment in proportion to their susceptibility, abundance and grain size. Generally, if a sediment contains paramagnetic miner- als as common constituents, susceptibility values are not higher than 100-200 x 10-6 SI. Magnetic susceptibili- ty analysis has been increasingly used in the last few years as a valid proxy for magnetic minerals concentra- tion (VEROSUB & ROBERTS, 1995; DEKKERS, 1997) and then can be useful to define the presence of climatically induced cyclicity in sedimentary records (for example alternation of carbonates and clays) (DEKKERS, 1997). In fact, the susceptibility fluctuations is used as a sensi- tive indicator of temporal variations in terrigenous input which are likely caused in changes of lake/sea level and 332 Fig. 6 - Stratigraphic log of Marano de’ Marsi section (a) related magnetic susceptibility (b) and natural remanent magnetization (c) profiles. Colonna stratigrafica schematica della sezione di Marano de’ Marsi (a), profili della suscettività magnetica (b) e della magnetizzazione naturale rimanente (c). E. Chiarini et al. 333Plio-Quaternary evolution of the high Salto river valley ... erosion of the continental subaereal landscape (ROSENBAUM et al., 1996). In the Marano de’ Marsi lacustrine section the magnetic susceptibility values are low in average and range between 14 and 164 x 10-6 SI (Fig. 6b). In general, they display a small variation in that interval, reflecting stable and constant input of diamagnetic (mainly cal- cite) and paramagnetic (clay minerals) grains from sub- strate during the deposition. Moreover, this excludes the presence of any volcaniclastic contamination into the paleolake, such as pyroclastic fall products charac- terized by high content of ferromagnetic grains. Volcaniclastic deposits are typical in Middle Pleistocene lacustrine intermountain basins in central and southern Italy (CAVINATO et al., 1994). In detail, it is possible to see a weak increase in the values from the bottom to the top, in particular from 50 to 600 cm, with values from 14 to 70 x 10-6. This part is also characterized by a weak ciclicity with a frequency of about 20-40 cm for each cycle (Fig. 6b). It is likely that this high frequency reflects the alternance of light and brown layers which have relatively major and minor content in carbonate grains respectively (i.e. minor and major content in magnetic grains). Higher values are measured between 600 and 780 cm with a maximum value in correspondence of 756 cm. The remnant upper part of the section shows a constant trend up to the contact with the slumping, which is characterized by higher values, about 80-100 x 10-6 SI. Natural Remanent Magnetization (NRM) analyses The intensity of the NRM is in the 1-20 x 10-3 A/m range, without any particular variation along the Marano de’ Marsi section (Fig. 6c). The widespread occurrence of diamagnetic and subordinate paramagnetic minerals is also responsible for the relatively low NRM values. As already seen in the susceptibility profile, also NRM pro- file of Fig. 6c shows a constant trend around a mean value of 8.8 x 10-3 A/m and characterized by a ciclicity with a frequency similar to that of susceptibility (see Figs. 6b and 6c). Moreover, relatively higher values are recorded in the central part of the section between 400 and 800 cm, whereas they are lower in the upper and lower part of it. Thermal demagnetisation technique was applied for most of the specimens, whereas alternating field (AF) technique was applied only for some representa- tive specimens. In the first kind of measurements the samples were thermally demagnetised with tempera- ture increments of 30 and 50ºC, up to 440-480ºC, whereas in the second treatment the samples were pro- gressively demagnetised in an alternating field (AF) up to 100 mT. Both measurements revealed valuable, giv- Fig. 7 - Calculated paleomagnetic declination (a), inclination (b), maximum angular deviation (c) and Virtual Geomagnetic Pole (d) for Marano de’ Marsi section. Variazione della declinazione magnetica (a), inclinazione magnetica (b), massima deviazione angolare (c) e del polo geomagnetico vir- tuale (d) ottenuto dai dati di declinazione, inclinazione e latitudine. ing the same results. Demagnetization dia- grams are analysed by principal compo- nent analysis (KIRSCHVINK, 1980), in order to determine the ChRM (characteristic rema- nent magnetization) directions for each sample. The maximum angular deviation (MAD) is generally less than 10°. Samples yielding MAD>15° were rejected for further analyses. All the characteristic compo- nents are plotted on equal area stereo- graphic projection before (in situ) and after correction for bedding tilt (tilting value is 8° toward NNE). Some representative paleomagnetic data obtained by thermal and alternating field (AF) demagnetisations have been shown in SADORI et al. (2009). The declination and inclination of the characteristic component (ChRM) are pro- jected versus stratigraphic height after application of bedding correction as shown in Figures 7a, b. It is evident as the declination of this component is constant in declination around N, whereas the incli- nation is always positive (i.e. normal polari- ty) with a mean value of 55°, in agreement with the expected geomagnetic field (GAD) direction expected for this locality. The small scatter of data both in declination and inclination is interpreted as due to secular variations of the Earth’s magnetic field. No magnetic reversals are present in the investigated section. Magnetostratigraphy The declination and inclination obtained for the ChRM of each sample were used to calculate the virtual geomagnetic pole (VGP) latitude, yielding a magnetic polarity sequence in the studied section (Fig. 7d). Positive VGP latitudes indicate normal polarity for the whole section. In the lower part of the section from 0 to 600 cm, the profile of Figure 7d indicates a weak drift of calculated VGP from 70°N to 84°N. In the topmost part of the section (the last 15 cm), the data are not reliable because the clay samples are disturbed by the slump- ing emplacement. The overall magnetic polarity stratigraphy outlined for Tore section is well within a normal polarity magne- tozone (i.e. Olduvai subchron or Gauss chron of Pliocene age). Anisotropy of Magnetic Susceptibility analyses The measurement of the low-field AMS represents a rapid and non destructive method to define the min- eral fabric of rocks (e.g., HROUDA, 1982). The magnetic fabric has proven to be closely related to the strain ellipsoid in weakly deformed or “undeformed” clay sed- iments both in compressional and extensional tectonic regimes, and, consequently, it constitutes a valid tool in defining the deformative pattern in weakly deformed fine-grained sedimentary sequences (BORRADAILE & TARLING, 1981; KISSEL et al., 1986; LEE et al., 1990; SAGNOTTI et al., 1994; MATTEI et al., 1999; CIFELLI et al., 2004). The AMS is defined by a second rank symmetric tensor and is represented geometrically by an ellipsoid with principal axes Kmax>Kint>Kmin. The shape of the AMS ellipsoids and the degree of the anisotropy can be described by several parameters (see also JELINEK, 1981; HROUDA, 1982). The magnetic lineation and folia- tion (at the sample and site scales) are defined as the Kmax and the plane normal to the Kmin respectively. The AMS at both the specimen and the site levels was eval- uated using the Jelinek statistics (JELINEK, 1978). In the case of a sedimentary magnetic fabric, the magnetic foliation is parallel to the plane of sedimenta- tion whereas the magnetic lineation is dispersed on the foliation plane and reflects the hydrodynamic condi- tions in the depositional basin. In the case of a magnet- ic fabric induced by tectonic processes, the maximum axis of susceptibility (Kmax) is commonly aligned along the direction of maximum stretching in extensional environments (e.g. MATTEI et al., 1997; CIFELLI et al., 2004). In extensional environments, it is well document- ed that the AMS of rocks can be imparted during the very early stages of deformation (e.g. CIFELLI et al., 2004). The AMS results of Tore section are grouped in seven sites, from the bottom to the top of the section. The degree of anisotropy is generally low, with values of P’ well below 1.10, and the fabric is either oblate or prolate as shown in Figure 8 (P vs. T). The shapes of AMS ellipsoid are also not well defined with exception of site VS04 (Fig. 9), characterized by oblate ellipsoid. The other sites show low anisotropy degree and a large variability in the shape parameter T. A poor defined magnetic foliation has been mea- sured in all the sites. A well defined magnetic lineation, NE-SW oriented, has been recognized in only one site (VS04) from the middle part of the section. In all the other sites the magnetic lineation is generally dispersed on the foliation plane (Fig. 9). The overall AMS results demonstrate that the investigated deposit is characterized by a magnetic fabric poorly defined because of the low content in 334 E. Chiarini et al. Fig. 8 - Anisotropy degree (P) versus shape parameter (T) of all the measured sam- ples of Marano de’ Marsi section. Note the scatter of the data and the low value of P values. Grado di anisotropia (P) vs. fattore di forma (T) per tutti i campioni misurati. Si noti il grado di anisotropia molto basso e l’alta dispersione dei dati, che testimoniano un fabric magnetico poco definito. 335Plio-Quaternary evolution of the high Salto river valley ... magnetic grains. In fact, the lacustrine section is com- posed mostly by carbonatic marls which contain extremely few ferromagnetic or paramagnetic grains. The shape of AMS ellipsoids is therefore mostly related to sedimentary processes, where the magnetic foliation tends to reflect the deposition plane whereas the scat- ter of the magnetic lineation is likely related to hydrody- namic conditions of the lake at the moment of deposi- tion. Only site VS04, which has the higher Km values (i.e. major amount of magnetic grains), is characterized by a well defined magnetic fabric with a NE-SW orient- ed magnetic lineation. This seems to suggest a stretch- ing direction oriented parallel to the lineation, i.e. NE- SW, which could be the main extensional direction that has overprinted partially the lacustrine succession dur- ing tectonic deformation. 4.1.2 Plant micro- and macrofossils analyses For plant fossils, a continuous column was taken, sampling big blocks of sediment later on subsampled in laboratory every 10 cm for pollen analyses and inspect- ed for macrofossil ones. Plant macrofossils Preliminary analyses have shown the presence of a number of arboreal taxa: levels with angiosperm fos- sils are more abundant than those with gymnosperm ones (Fig. 10). The fossil compressions/impressions mainly consist of leaves, but also fruits/seeds, woods and cuticles were found. The still provisional fossil list comprehends many angiosperms such as Acer cfr. monspessulanum, Carpinus cfr. orientalis, Engelhardia sp., Fagus cfr. sylvatica, Hedera sp., Liquidambar sp., Quercus sp., Rosa sp., as well as Abies sp. and Pinus sp. Both Engelhardia and Liquidambar genera are extinct in Europe and have species living only in Asia the first, and both in Asia and Northern/ Central America the second. Some fossil specimens are shown in Figure 11. Plant microfossils 61 sediment samples have been analysed every 20 cm (with sampling improving at 10 cm at key depths), but only 57 showed a sufficient pollen content to be plotted in the diagram. Known weights (ca. 2 g) of dry sediment were processed in a sterile laboratory, using a standard procedure consisting in HCl (30%), HF (40%) and NaOH (10%) chemical attacks to remove both the mineral fraction and the organic matter. Lycopodium spore tablets (STOCKMARR, 1971) have been added to estimate the pollen concentration val- ues. The palynomorphs identification was carried out with a transmitted light microscope at a minimum mag- nification of 400X. The level of pollen conservation is low, often very low. The grains appear generally to be corroded and crumpled, slowing and making the analy- sis difficult. The total pollen concentration varied from 1900 to 129,000 pollen grains/g, the arboreal pollen (AP) per- centages mainly ranged between 96 and 100%. 56 pollen taxa (with 11 herbaceous taxa in all) were identi- fied. The number of terrestrial spermatophytes pollen taxa per level is between 15 and 39. The highest amount of pollen taxa matches both very low pollen concentration values (3200 pollen grains/g) and the highest percentage values of NAP (13%) at 9.80 m. The number of extinct arboreal pollen taxa is high (14). Fern spores and algal remains were also identified during routine pollen analyses. Synthetical pollen curves are presented in Figure 10. The basic sum is composed by all terrestrial sper- matophytes (trees, shrubs and herbs), so excluding water plants, ferns and algal remains. For details con- cerning the pollen identification criteria and definition of pollen types, as well as for a complete pollen diagram see SADORI et al. (2009). Among gymnosperms Pinus haploxylon type (plus Cathaya), Pinus sylvestris type, Cedrus, Picea, Abies, Tsuga, and Taxodium type are found with significant values. Among angiosperms, herbs are quite absent. Dominant angiosperm tree taxa are Quercus, Zelkova, Ulmus, Carya and Pterocarya (the last two are no longer spontaneously growing in our continent). The presence Fig. 9 - Equiareal projection of AMS principal axes for each site. Note the well defined magnetic fabric for site VS04. Proiezione equiareale dei principali assi dell’ellissoide dell’AMS. Si noti come il fabric magnetico sia ben definito nel sito VS04. 336 E. Chiarini et al. F ig . 1 0 - M a ra n o d e ’ M a rs i( R ie ti , C e n tr a lI ta ly ). P o lle n d ia g ra m (s e le c te d ta xa a n d g ro u p s) . P o lle n g ro u p s: su b tr o p ic a lh u m id fo re st (T ax o d iu m ty p e , E n g e lh ar d ia , N ys sa , c fr . R h o ip te le a) ; te m - p e ra te b ro a d le a ve d d e c id u o u s fo re st (C ar p in u s b e tu lu s, C ar ya , C as ta n e a, C e lt is , C o ry lu s, H am am e lid ac e ae u n d iff ., Ju g la n s, Ju g la n d ac e ae u n d iff ., L iq u id am b ar , P te ro c ar ya , Q u e rc u s ro b u r ty p e , Q u e rc u s c e rr is ty p e , T ili a, U lm u s, Z e lk o va ); m id to h ig h e le va ti o n fo re st (A b ie s, C e d ru s, P ic e a, T su g a, B e tu la , F ag u s) ; sc le ro p h yl lo u s fo re st (Q u e rc u s ile x ty p e , O le a) ; xe ri c e le m e n ts (h e rb s a n d E p h e d ra ). M ar an o d e ’ M ar si (R ie ti , It al ia c e n tr al e ). D ia g ra m m a p o lli n ic o (t a xa e g ru p p i se le zi o n at i). G ru p p i p o lli n ic i: fo re st a u m id a su b tr o p ic al e (t ip o T a xo d iu m , E n g e lh a rd ia , N ys sa , c fr . R h o ip te le a ); fo re - st a te m p e ra ta d i la ti fo g lie d e c id u e (C a rp in u s b e tu lu s, C a ry a , C a st a n e a , C e lt is , C o ry lu s, H a m a m e lid a c e a e in d iff ., Ju g la n s, Ju g la n d a c e a e in d iff ., L iq u id a m b a r, P te ro c a ry a , Q u e rc u s ti p o ro b u r, Q u e rc u s ti p o c e rr is , T ili a , U lm u s, Z e lk o va ); fo re st a m o n ta n a e su b m o n ta n a (A b ie s, C e d ru s, P ic e a , T su g a , B e tu la , F a g u s) ; fo re st a a sc le ro fil le (Q u e rc u s ti p o ile x, O le a) ; e le m e n ti xe ri c i (e rb e e d E p h e d ra ). 337Plio-Quaternary evolution of the high Salto river valley ... of pollen of subtropical taxa as Nyssa, cfr. Rhoiptelea, Liquidambar, Engelhardia is worth being mentioned (Fig. 12). In the studied pollen record gymnosperms are prevailing, even if four main important, sudden and short angiosperm arboreal pollen oscillations can be observed, the oldest of which being more marked. The importance of these changes can be singled out exam- ining the diagram of Figure 10, where pollen taxa, gen- erally found in very different environments, ranging from mountain to arid ones, passing through temperate, sub- tropical, and Mediterranean types, have also been grouped into five “ecological groups”. The more wide- spread vegetation types are two, one comprehending trees typical of temperate broadleaved deciduous for- est and the other trees of mid to high elevation forest. Due to the wide ecological range of living Pinus species and to the fact that both Pinus pollen types compre- hend other genera (e.g. Pinus haploxylon type includes in this paper also other genera such as Cathaya; in fact, for the low degree of pollen grains, preserving its methodical distinction was not possible), Pinus was not inserted in the ecological groups (SADORI et al., 2009). The main vegetation changes do not consist in alternations between open and close vegetation forma- tions, that is to say between forests and steppe forma- tions, but just in changes in the forest composition, as xeric elements are quite absent. The rapid and consis- tent alternations found in the pollen record, generally speaking between conifer forests and deciduous angiosperm ones, induce to consider that consistent changes in vegetation occurred. In northern Italy arbo- real phases with gymnosperms and angiosperms alter- nately dominating characterized the glacial/interglacial cycles since the Gelasian (BERTINI, 2003). In southern coastal Italian areas, however, such cycles were fea- tured by the alternation between steppic and arboreal phases (COMBOURIEU-NEBOUT et al., 2004). More marked glacials/interglacials oscillations and a difference in their periodicity occurred only in the last 1 million years (LEROY, 2007; RAYMO et al., 1997; RUDDIMANN et al., 1989), when a change in the pattern of the orbitally dri- ven cycles from obliquity to eccentricity occurred. These physiognomic vegetation changes are however very strong, as the forest cover changed from a mountain forest to a temperate/subtropical one. In these forest changes, temperature and not humidity, seems to have been the main limiting factor. This was not the case of the vegetation changes from glacial to interglacial periods, which occurred in the last million of years and that were mainly triggered by changes in water availability. Thus, the record of Marano de’ Marsi appears to have been mainly deposited during a glacial period. As a matter of fact the so called interglacial periods are represented by very thin sediments horizons, in which the clear vegetation succession typical of these phases are not found (SADORI et al., 2009). Another substantial aspect of the vegetation recorded in the pollen diagram is that the pollen rain reflects the presence of arboreal taxa requiring different climatic features, and implying the existence of several vegetation belts. It results from the mixing of mountain, mesophilous, and subtropical elements and implies the presence of mountains in the surroundings. The hypothesis that the sediment record from Marano de’ Marsi has some taphonomical problems has to be advanced: 1) the pollen samples with angio- sperm trees dominating show the highest number of taxa and the lowest pollen concentration values. This is in clear contradiction with what happens in the pollen records in which concentration values have been esti- mated; 2) the vegetation changes between glacials and interglacials are sudden and all intermediate phases are missing; 3) the clear forest vegetation successions typi- cal of interglacial periods are also not found, these phases being recorded, ex abrupto, by mature forests; 4) there is an apparent opposite indication from the macrofossil and microfossil record, as the presence of angiosperm leaves is overwhelming even in phases in which gymnosperm pollen is dominant. The second and the third incongruities were investigated through thin section sediment analyses to check the possibility of important changes in the sedi- mentation pattern, aimed at explaining possible sedi- ment gaps, not evidenced in the wide outcrop. Light lithological changes showed by analyses (COSENTINO D., Roma Tre University, personal communication) have been interpreted as disconformities of a low to very low order, ranging from the 0th-order (lamination surface) to the 1st-order (set bounding surface) sensu MIALL (2006), consistent with the sedimentary environment. The fourth contradiction can be explained with an almost continuous local presence of angiosperms (leaves are too liable to deteriorate if transported far away) by the lake also in periods in which the regional vegetation (and therefore the pollen rain) was dominat- ed by gymnosperms. The vegetation and floristic features of the plant fossils record frame the lacustrine sediment succession between the Middle Pliocene and the Early Pleistocene. 4.2 Borgorose core The Borgorose core has a 24.50 m length and was drilled at an elevation of 725 m a.s.l., close to an outcrop of lacustrine sediments. The sediments, quite homogeneous, are characterized by a higher clay con- tent with respect to those cropping out in the Marano de’ Marsi investigated section, according to a frankly lacustrine origin; lithologies are mainly massive or thinly laminated marls and silty claystones. 4.2.1 Paleomagnetic analyses Paleomagnetic samples have been collected every 25 cm on average in order to reconstruct the magnetic susceptibility profile and magnetic polarity of the core and eventually correlate it to Marano de’ Marsi section. The samples have not been oriented with respect to the geographic north, but they have been marked only with respect to the top of the core. In the Borgorose core, the magnetic susceptibility values are generally low and range between 81 and 249 x 10-6 SI (Fig. 13). These values are relatively higher than Marano de’ Marsi section because the sediments of Borgorose core have a major content of clay minerals (i.e. paramagnetic fraction). The highest values are rela- tive to the upper portion of the core, from 550 to 700 cm, whereas then are constant for the remaining lower portion. 338 Fig. 11 - Marano de’ Marsi (Rieti, Central Italy). Plant macrofossils. 1. Liquidambar sp. 2. Acer cfr. monspessulanum 3. Quercus sp. 4. Fagus cfr. sylvatica 5. Engelhardia sp. Marano de’ Marsi (Rieti, Italia centrale). Macrofossili vegetali. 1. Liquidambar sp. 2. Acer cfr. monspessulanum 3. Quercus sp. 4. Fagus cfr. sylvatica 5. Engelhardia sp. E. Chiarini et al. The paleomagnetic analysis has been performed on 62 out of 84 total samples of the core. All the sam- ples have been thermally demagnetised at Paleomagnetic Laboratory of Peveragno of the University of Turin. The same standard techniques described in the previous section have been used to define the principal component and the characteristic remanent magnetization for these samples. The overall results indicate a positive inclination, with an average inclination of 52°, for all the investigated samples, sug- gesting a normal polarity magnetozone without mag- netic reversals. The mean paleomagnetic inclination has been defined without taking into account the tectonic tilting of the succession, that is of about 20°, as mea- sured on the nearby outcrop of the same lacustrine deposits. 4.2.2 Plant microfossils Pollen samples have been taken every 10 cm from 2.60 to 24.50 m. 41 samples have been analysed, but only 15 (from 3 to 8.60 m) showed a sufficient pollen content to be plotted in the diagram. Known amounts of dry sediment (ca. 2 g) have been chemically processed following the treatment illustrated for the Marano de’ Marsi outcrop. The preservation of the pollen grains was also poor in this record, being generally degraded, corroded and broken, causing problems in the identification. The mean count of only the terrestrial spermatophytes pollen was 304 pollen grains. Problems in the pollen preservation are clearly evidenced in the lower part of the core (9-24.2 m) and in the uppermost 3 m. In the related samples the very low pollen concentration values (from 100 to few thou- sands pollen grains/g) did not allow us to reach signifi- cant pollen counting. The total pollen concentration of the selected pollen samples varied from 9000 to 113,000 pollen grains/g at Borgorose, with AP (arboreal pollen) percent- ages ranging from 97 to 99%. Synthetical pollen per- centage and total pollen concentration curves are pre- sented in Figure 14. 46 terrestrial plants pollen taxa (of which 8 herbaceous ones) were identified at Borgorose. The number of pollen taxa per level varies from 12 (matching the lowest pollen concentration values) to 31. The diagram does not record either important vegetation changes or a clear vegetation development. Gymnosperms pollen always prevails on that of angiosperms. Pinus haploxylon type, always dominant, was mainly accompanied by Picea, Pinus sylvestris type, Taxodium type, Cedrus, and Tsuga. Picea, showing percentages between 8 and 19%, can be considered locally present (HUNTLEY & BIRKS, 1983). Deciduous oaks and Carya show high percentage values among the angiosperms, mainly accompanied by Fagus, Pterocarya, Juglans, Ulmus, Hamamelidaceae, and in minor quantities by Engelhardia, Nyssa and cfr. Rhoiptelea. It is worth mentioning the consistent presence of water demanding trees (riparian plants), possibly indi- cated by the presence of Taxodium type, Nyssa, and Carya. Considering the ecological groups, mid to high elevation forest trees are more abundant than those typical of the temperate broadleaved deciduous forest. Subtropical humid forest elements are always present with significant values, while sclerophyllous and xeric elements appear to be less abundant. The sediments are likely to have been deposited during a cool and wet period, and do not show the important angiosperm tree expansions featuring warmer phases. Notwithstanding, angiosperms show a continuous and similar behaviour, suggesting that they were present at lower elevations and, possibly, in wet areas (see for example Taxodium type) by the lake. The high amount of Taxodium type pollen is the main rea- son for the considerable amount of subtropical humid forest taxa found in the pollen record. Four common features can be singled out in the two investigated successions: 1) the scarce number of pollen taxa, that can be partially explained with the very low amount of herbs; 2) the similarity in the pollen lists of the two sites, with 43 pollen taxa in common; 3) the same list of extinct arboreal pollen taxa; 4) the contem- porary presence of plants with different ecological requirements, indicating the presence of vegetation belts developed along an altitudinal gradient. Similarly for the Marano de’ Marsi outcrop, a Middle/Upper Pliocene or Early Pleistocene age appears to be consistent with the palyno- logical data. 5. DISCUSSION AND CONCLUSIONS The succession cropping out in the high Salto River valley, called Marano de’ Marsi unit in this paper, has been the object of a few works, differently from successions of Fucino and middle Salto valley basins. Previous authors have interpreted the high Salto valley succession as the result of more than one sedimentary event, proba- bly related to Pliocene (SERVIZIO GEOLOGICO D ’I T A L I A , 2005), or to Pliocene and to Messinian (BUSINO,1990; FEDERICI, 1990). Stratigraphical and sedimentological data illustrated in this study, on the contrary, allow us to refer the unit to a single sedi- mentary cycle organized in a complex depositional system. Recognized etherop- ic facies – from alluvial to lacustrine via transitional forms - have been referred to a Gilbert-type lacustrine delta and to small scale fan-deltas originated from the east- ern border of the basin. Anomalous sharp facies variations observed in the field have been reported to tectonic activity which significantly displaced the sedimentary unit. Two main unconformities bound the unit: the lower one marks the erosional contact with the carbonatic bedrock; the upper unconformity corresponds to the Colle Vena Francesca paleosurface. The prevalence of carbonatic lithologies marks a significant difference with earliest continental deposits of Fucino (Aielli com- plex, BOSI et al., 1995) and middle Salto valley (Sabbie di Piagge, BERTINI et al., 1986); in the lower part of the sedimentary successions of these basins, siliciclastic lithologies actually prevail, according to a wide spread of Messinian turbidites in the catchment area (BOSI et al., 2003) during the first erosional phase of the chain (CENTAMORE & NISIO, 2002a; b). Therefore it is not possible to correlate the Marano de’ Marsi unit with the Fucino and the middle Salto valley on the base of lithological affinities. In the framework of a general shortage in chronological data, paleobotanic and pale- omagnetic analyses have been conducted on the finer sediments of the Marano de’ Marsi unit to obtain dating elements, also providing interesting paleoenvironmental and paleoclimatic results. Pollen records of the two sampling sites (Marano and Borgorose) suggest the presence of a well devel- oped mountain system in the surroundings, as indicated by the presence of trees belonging to different vegeta- tion belts. Pollen rain coming from mid to high elevation forest taxa is generally prevailing throughout the investi- gated successions. A pollen assemblage of this kind 339Plio-Quaternary evolution of the high Salto river valley ... Fig. 12 - Marano de’ Marsi (Rieti, Central Italy). Pollen grains. 1. Abies 2. Nyssa 3. Pterocarya 4. Picea 5. Cedrus 6. Tsuga 7. Tilia 8. Engelhardia 9. cfr. Rhoiptelea 10. Carya 11. Zelkova. Marano de’ Marsi (Rieti, Italia centrale). Granuli pollinici. 1. Abies 2. Nyssa 3. Pterocarya 4. Picea 5. Cedrus 6. Tsuga 7. Tilia 8. Engelhardia 9. cfr. Rhoiptelea 10. Carya 11. Zelkova. faults located on the eastern border of the basin. Previously exposed data and the lack of com- pressive features inside deposits allow us to hypothe- size that the Marano de’ Marsi unit deposited inside an intermountain continental basin, like the Pliocene deposits of other basins of Central Apennines. The ero- sion of the wide Colle Vena Francesca paleosurface testifies a significant decrease of the subsidence of this basin; during Early Pleistocene the subsequent settle- ment of Corvaro intermountain basin yields a new sub- sident area, where the following sedimentation takes place. The whole area indeed is progressively captured by the hydrographic system of the Salto River and deposits are consequently dismembered; the outlined evolution is common to that of other basins of the western Apennine flank: endorheic at first and then captured and integrated in the drainage network (D’AGOSTINO et al., 2001). In reconstructing the Fucino basin evolution D’AGOSTINO et al. (2001) supposed that the end of the completely internal drainage phase coincided with the capture, during the Pleistocene, of a portion of the basin by regressive headward erosion of the Salto River. Nevertheless, a sediment supply from the eastern border of the basin has also been recognized, the facies distribution and paleocurrent directions testify a main drainage direction - i.e. the Gilbert-delta progra- dation - from south-east toward north-west, thus trend- ing like the principal fault system. These arguments highlight that at the time of Marano de’ Marsi unit sedi- mentation, the Fucino and the high Salto valley areas belonged to two different, isolated, basins. 340 E. Chiarini et al. suggests ascribing the investigated lev- els to glacial periods (BERTINI, 2003). Nevertheless, pollen typical of temperate forests (and of interglacial periods) is always present in low quan- tity, and, at times, overwhelming at Marano de’ Marsi. The moderate expansion of taxa typical of the sub- tropical humid forest (all extinct at pre- sent in Europe) is noteworthy and use- ful for the chronological framing. The similarity of the pollen assemblages induces us to think that the two sedi- ment records were deposited in the same period or at very close times. The plant fossil assemblages, the high amount of taxa at present extinct in Italy indicate that the sediments were likely deposited during the Middle/Up- per Pliocene or the Early Pleistocene. As no magnetic reversals are pre- sent in the investigated section, the combination of pollen and paleomag- netic data allows to constrain both records to Gauss chron (Middle Pliocene) or Olduvai/Reunion sub- chrones (Late Pliocene). A close com- parison (see SADORI et al., 2009) with other Italian palynological records of Pliocene and Pleistocene age (URBAN et al., 1983; ARIAS et al., 1990; SUC et al., 1995; BERTINI, 2000, 2001, 2003) seems to better frame both successions in the subchrones of Late Pliocene, more likely to Olduvai one, for its longer duration. The possibility that the records deposited during the Gauss chron cannot be, definitely, excluded. The presence of Pliocene continental deposits is widely documented in the Central Apennines and it has been related mainly to extensional tectonic activity: since Pliocene a prevailingly lacustrine sedimentation took place in the main intermountain basins like the Turano valley, the middle Salto valley and the Fucino basins (BOSI et al., 1995; BOSI et al., 2003; GALADINI & M E S S I N A , 2004). Intermountain basins in Central Apennines are thought to be progressively younger eastward, for the effects of extensional tectonic migra- tion from the western sectors to the eastern ones (CAVINATO & DE CELLES, 1999; GALADINI & MESSINA, 2004). The quite complex structural setting, outlined in this paper, neither allows us to define the configuration of the basin where the Marano de’ Marsi unit sediment- ed, nor to certainly establish which tectonic structures guided the basin settlement. Geological field data and AMS results suggest that the Marano de’ Marsi unit was deformed in an extensional tectonic regime, with a stretching direction oriented NE-SW. The development of a Gilbert-type delta also implies the occurrence of a steep margin; additionally the presence of slumps and big boulders in some horizons of succession indicate a tectonic insta- bility along the depositional slope, giving value to the tectonic control on the basin evolution. Finally the structural setting of succession shows a general NE tilt- ing. All these features are consistent with the activity of Fig. 13 - Stratigraphic log (a) and magnetic susceptibility profile (b) for Borgorose core. Colonna stratigrafica schematica della sezione di Borgorose (a) e profilo di suscet- tività magnetica (b). 341Plio-Quaternary evolution of the high Salto river valley ... F ig . 1 4 - B o rg o ro se (R ie ti , C e n tr a lI ta ly ). P o lle n d ia g ra m (s e le c te d ta xa a n d g ro u p s) . P o lle n g ro u p s a re lis te d in F ig . 1 2 . B o rg o ro se (R ie ti , It al ia c e n tr al e ). D ia g ra m m a p o lli n ic o (t a xa e g ru p p is e le zi o n at i). P e r ig ru p p ip o lli n ic iv e d il ’e le n c o in F ig . 1 2 . 342 E. Chiarini et al. ACKNOWLEDGMENTS The authors are very grateful to A. Bertini and P. Messina for the useful discussions and the critical review of the paper and F. Cifelli for paleomagnetic sampling. REFERENCES ARIAS C., BIGAZZI G., BONADONNA F.P., IACCARINO S., URBAN B., DAL MOLIN M., DAL MONTE L. & MARZOLINI M. (1990) - Valle Ricca late Neogene stratigraphy (Lazio region, Central Italy). Paléobiologie conti- nentale 17, 61-68. BELLOTTI P. (1992) - I sedimenti terrigeni della depres- sione del Lago Salto – Tagliacozzo: considerazioni sulla loro successione verticale. Boll. Soc. Geol. It., 103, 311-326. BERTINI A.(2003) - Early to middle Pleistocene changes of the Italian flora and vegetation in the light of a chronostratigraphic framework. Il Quaternario, 16 (1bis), 19–36. BERTINI A. (2000) - Pollen record from Colle Curti and Cesi: Early and Middle Pleistocene mammal sites in the Umbro-Marchean Apennine Mountains (central Italy). Journal of Quaternary Sciences, 15, 825-840. BERTINI A. (2001) - Pliocene climatic cycles and altitudi- nal forest development from 2.7 Ma in the north- ern Apennines (Italy): evidences from the pollen record of the Stirone section (5.1 to 2.2 Ma). Geobios, 34 (3), 253–265. BERTINI T. & BOSI C. (1976) - Sedimenti continentali probabilmente pliocenici nella Valle del Salto e nella conca del Fucino (Rieti, L’Aquila). Boll. Soc. Geol. It., 95, 767-801. BERTINI T., BOSI C., MESSINA P. & SPOSATO A. (1986) - Elementi di tettonica compressiva pliocenica nella zona di Borgo S. Pietro (Rieti). Mem. Soc. Geol. It., 35, 547-553. BIGI S. & COSTA PISANI P. (2003) - The “pre-thrusting” Fiamignano normal fault. Boll Soc. Geol. It., 122, 267-276. BORRADAILE G. J. & TARLING D. H. (1981) - The influence of deformation mechanisms on magnetic fabrics in weakly deformed rocks. Tectonophysics, 77, 151-168. BOSI C., GALADINI F., GIACCIO B., MESSINA P. & SPOSATO A. (2003) - Plio-Quaternary continental deposits in the Latium-Abruzzi Apennines: the correlation of geological events across different intermontane basins. Il Quaternario, 16 (1bis), 55-76. BOSI C., GALADINI F. & MESSINA P. (1995) - Stratigrafia plio-pleistocenica della conca del Fucino. Il Quaternario, 8(1), 83-94. BOSI C. & MESSINA P. (1991) - Ipotesi di correlazione fra successioni morfo-litostratigrafiche plio-pleistoce- niche dell’Appennino laziale-abruzzese. Studi Geol. Camerti, vol. spec. 1991/2, CROP11, 27-31. BOSI V. & FEDERICI V. (1993) - La conca di Corvaro. In “Guida all’escursione alle conche intermontane dell’Appennino laziale-abruzzese” (15-18 settem- bre 1993). Il Quaternario, 6, 393-395. BOSI V., FUNICIELLO R. & MONTONE P. (1994) - Fault inver- sion: an example in Central Apennines (Italy). Il Quaternario, 7(2), 577-588. BUSINO M. (1990) – Evoluzione geologica pliocenico- quaternaria dell’alta valle del Fiume Salto, da Bocca di Teve all’abitato di Marano. Unpublished degree thesis. Università degli Studi “La Sapienza”, Roma. CAVINATO G. P., COSENTINO D., DE RITA D., FUNICIELLO R. & PAROTTO M. (1994) - Tectonic-sedimentary evolu- tion of intrapenninic basins and correlation with the volcano-tectonic activity in Central Italy. Memorie Descrittive della Carta Geologica d’Italia, 49, 63–76. CAVINATO G. P. & DE CELLES P.G. (1999) – Extensional basins in the tectonically bimodal Central Apennines fold-thrust belt, Italy: response to cor- ner flow subducting slab in retrograde motion. Geology, 27, 955–958. CENTAMORE E. & NISIO S. (2002a) - Quaternary geology and morphostructural evolution between the Velino and Salto valleys. Studi Geol. Camerti, vol. spec. 2002, 37-44. CENTAMORE E. & NISIO S. (2002b) - Tettonica e sedimen- tazione (Lias-Pleistocene) nella media Valle del Salto (Rieti, Italia Centrale). Studi Geol. Camerti, 2002/2, 53-70. CHIARINI E., D’OREFICE M., GRACIOTTI R., LA POSTA E. & PAPASODARO F. (2008) - Note illustrative della Carta Geomorfologica d’Italia alla scala 1:50.000. F° 367 “Tagliacozzo”. APAT - Servizio Geologico d’Italia / Dipartimento Difesa del Suolo. CHIARINI E., GIARDINI M., LA POSTA E., PAPASODARO F. & SADORI L. (2007) - Sedimentology, palynology and new geochronological constraints on Quaternary deposits of the Corvaro intermontane basin (Central Italy). Revue de Micropaléontologie, 50 (4), 309-314. Elsevier. CHIARINI E., MESSINA P. & PAPASODARO F. (1997) - Evoluzione geologica e tettonica plio-quaternaria dell’alta valle del F. Salto (Italia Centrale): primi risultati derivanti dall’analisi delle superfici relitte e dei depositi continentali. Il Quaternario, 10(2), 625-629. CIFELLI F., MATTEI M., HIRT A. M. & GUNTHER A. (2004) - The origin of tectonic fabrics in “undeformed’’ clays: The early stages of deformation in exten- sional sedimentary basins. Geophys. Res. Lett., 31(9). COMBOURIEU-NEBOUt N., FOUCAULT A. & MÉLIÈRES F. (2004) - Vegetation markes of palaeoclimate cycli- cal changes in the Pliocene of Punta Piccola (Sicily, Italy), Palaeogeography, Palaeoclima- tology, Palaeoecology 214, 55-56 C I P O L L A R I P. & C O S E N T I N O D. (1995) – Miocene unconformities in central Apennines: geodynamic significante and sedimentary basin evolution. Tectonophysics, 252, 375-389. COMPAGNONI B., D’ANDREA M., GALLUZZO F., GIOVAGNOLI M.C., LEMBO P., MOLINARI V., PAMPALONI M.L., PICHEZZI R.M., ROSSI M., SALVATI L., SANTANTONIO M., RAFFI I. & CHIOCCHINI U. (2005) - Note illustrati- ve del F° 367 “Tagliacozzo” della Carta Geologica d’Italia alla scala 1:50.000. Servizio Geologico Nazionale, Roma. D’AGOSTINO N., JACKSON J.A., DRAMIS F. & FUNICIELLO R. (2001) - Interactions between mantle upwelling, drainage evolution and active normal faulting: an example from the central Apennines (Italy). Geophys. J. Int., 147, 475-497. DEKKERS M. J. (1997) - Environmental magnetism: an introduction. Geol. Mijn., 76, 163-182. FEDERICI V. (1990) – Evoluzione geologica pliocenico- quaternaria dell’alta valle del Fiume Salto nella zona di Borgorose-Valle Amara. Unpublished degree thesis. Università degli Studi “La Sapienza”, Roma. GALADINI F. & MESSINA P. (2001) - Plio-Quaternary changes of the normal fault architecture in the central Apennines (Italy). Geodinamica Acta, 14, 321-344. G A L A D I N I F. & M E S S I N A P. (2004) - Early-Middle Pleistocene eastward migration of the Abruzzi Apennine (central Italy) extensional domain. Journal of Geodynamics, 37, 57-81. GALADINI F., MESSINA P., GIACCIO B. & SPOSATO A. (2003) - Early uplift history of the Abruzzi Apennines (cen- tral Italy): available geomorphological constraints. Quaternary International, 101-102, 125-135. GALADINI F., MESSINA P. & SPOSATO A. (2000) - Tettonica quaternaria dell’Appennino centrale e caratterizza- zione delle faglie nel Pleistocene superiore- Olocene. In: Galadini F., Meletti C., Rebez. A. (a cura di), Le ricerche del GNDT nel campo della pericolosità sismica (1996-1999), CNR-GNDC, 181-192 HROUDA F. (1982) - Magnetic anisotropy of rocks and its application in geology and geophysics. Geophys. Surv., 5, 37–82. HUNTLEY B & BIRKS HJB (1983) - An Atlas of Past and Present Pollen Maps for Europe: 0-13 000 years ago. Cambridge University Press, Cambridge. JELINEK V. (1978) - Statistical processing of anisotropy of magnetic susceptibility measures on groups of specimens. Studia geophisica et geodetica, 22, 50-62. JELINEK V. (1981) - Characterization of the magnetic fabric of rocks. Tectonophysics, 79, T63-T67. KIRSCHVINK J. L. (1980) - The least-squares line and plane and the analysis of paleomagnetic data. Geophys. J. R. Astr. Soc., 62, 699–718. KI S S E L C., BA R R I E R E., LA J C. & LE E T.Q. (1986) - Magnetic fabric in “undeformed” marine clays from compressional zones. Tectonics, 5(5), 769- 781. LEE T.Q., KISSEL C., LAJ C., HORNG C. S. & LUE Y.T. (1990) - Magnetic fabric analysis of the Plio- Pleistocene sedimentary formations of the Coastal Range of Taiwan. Earth Planet. Sci. Lett., 98, 23–32. LEROY S.A.G. (2007) - Progress in palynology of the Gelasian–Calabrian Stages in Europe: Ten mes- sages . Revue de Micropaléontologie, 50 , 293–308. MATTEI M., SAGNOTTI L., FACCENNA C. & FUNICIELLO R. (1997) - Magnetic fabric of weakly deformed clay- rich sediments in the Italian peninsula: relationship with compressional and extensional tectonics. Tectonophysics, 271, 107-122. MATTEI M., SPERANZA F., ARGENTIERI A., ROSSETTI F., SAGNOTTI L. & FUNICIELLO R. (1999) - Extensional tectonics in the Amantea basin (Calabria, Italy): a comparison between structural and magnetic anisotropy data. Tectonophysics, 307, 33-49. MIALL A.D. (2006) - The Geology of Fluvial Deposits. Sedimentary Facies, Basin Analysis, and Petroleum Geology. Springer, 582p. MOREWOOD N. G. & ROBERTS G. P. (2000) - The geome- try, kinematics and rates of deformation within an en échelon normal fault segment boundary, cen- tral Italy. Journal of Structural Geology, 22, 1027- 1047. NIJMAN W. (1970) - Tectonics of the Velino-Sirente area, Abruzzi, Central Italy. Modification of compres- sional structures by subsequent dilatation and col- lapse. Koninkl. Nederl. Akademie Van Wetenschappen. Amsterdam. RAYMO M.E., OPPO D.W. & CURRY W. (1997) - The mid- Pleistocene climate transition: a deep sea carbon isotopic perspective. Paleoceanography, 12, 546- 559. ROSENBAUM J. G., REYNOLDS R. L., ADAM D. P., DREXLER J., SARNA-WOJCICKI A. M. & WHITNEY G. C. (1996) - Record of middle Pleistocene climate change from Buck Lake, Cascade Range, southern Oregon - Evidence from sediment magnetism, trace-element geochemistry, and pollen. Geol. Soc. Am. Bull., 108(10), 1328-1341. RUDDIMAN W.F., RAYMO M.E., MARTINSON D.G., CLEMENT B.M. & BACKMAN J. (1989) - Pleistocene evolution: northern hemisphere ice sheets and north Atlantic Ocean. Paleoceanography, 4(4), 353-412. SADORI L. (2005) – Studio paleobotanico dei depositi continentali plio-quaternari affioranti nell’area del Foglio n. 367 “Tagliacozzo” della Carta Geomorfologica d’Italia in scala 1:50.000. Rapporto inedito. S A D O R I L., G I A R D I N I M., C H I A R I N I E., M A T T E I M., PAPASODARO F. & PORRECA M. (2009) - Pollen and macrofossil analyses of Pliocene lacustrine sedi- ments (Salto river valley, Central Italy). Quaternary International, doi:10.1016/j.quaint.2009.05.008 . SAGNOTTI L., FACCENNA C., FUNICIELLO R. & MATTEI M. (1994) - Magnetic fabric and structural setting of Plio-Pliocene clayey units in an extensional regi- me: the Tyrrhenian margin of central Italy. J. Struct. Geol., 16(9), 1243-1257. SERVIZIO GEOLOGICO D’ITALIA (2005) - Carta Geologica d’Italia alla scala 1:50.000. F. 367 “Tagliacozzo”. S E R V I Z I O G E O L O G I C O D ’I T A L I A (2008) - Carta Gemorfologica d’Italia alla scala 1:50.000. F. 367 “Tagliacozzo”. STOCKMARR J. (1971) - Tablets with Spores used in Absolute Pollen Analysis. Pollen et Spores 13, 615-621. SUC J.-P., BERTINI A., COMBOURIEU-NEBOUT N., DINIZ F., LEROY S., RUSSO-ERMOLLI E., ZHENG Z., BESSAIS E. & F E R R I E R J. (1995) - Structure of West Mediterranean vegetation and climate since 5.3 Ma. Acta Z9ool. Crac., 38(1): 3-16. URBAN B., ARIAS C., BIGAZZI G. & BONADONNA F.P. (1983) - Early Pleistocene palynostratigraphy of Fornace Tini, Valle Ricca (Central Italy). Palaeogeography, Palaeoclimatology, Palaeoecology, 41, 153-164. VEROSUB K.L. & ROBERTS A.P. (1995) - Environmental magnetism: past, present, and future. J. Geophys. Res. B, 100, 2175–2192. 343Plio-Quaternary evolution of the high Salto river valley ... Ms. ricevuto l’8 aprile 2009 Testo definitivo ricevuto il 29 ottobre 2009 Ms. received: April 8, 2009 Final text received: October 29, 2009