AMQ83 Baldanza et al ProofCopy 3 Available online http://amq.aiqua.it ISSN (print): 2279-7327, ISSN (online): 2279-7335 Alpine and Mediterranean Quaternary, 31 (1), 2018, 87 - 104 NEW INTEGRATED DATA FROM CLAY LACUSTRINE DEPOSITS OF THE DUNAROBBA AREA (UMBRIA, CENTRAL ITALY) Angela Baldanza 1, Roberto Bizzarri 1, Lucio Di Matteo 1, Marco Lezzerini 2, Luca Mencaroni 1, Stefano Pagnotta 2, Simona Raneri 2, Giuseppe Vinti 1 1Department of Physics and Geology, University of Perugia, Perugia, Italy. 2Department of Earth Sciences, University of Pisa, Pisa, Italy. Corresponding author: R. Bizzarri ABSTRACT: This paper is aimed to illustrate and discuss new data from the clayey deposits collected in the Dunarobba area (Umbria, central Italy), and to better understand some features of the “Fosso Bianco” Unit lacustrine stage inside the South Ti- berino Basin, during its latest (presumably Early Pleistocene) phases. Data come from a newly cultivated quarry area (Cava Nuova), only partly considered in previous works; the new section extends well inside the deep lake depositional stage of Fosso Bianco Unit, previously not directly described in outcrop in the Dunarobba area, but only reported from well logs. Selected samples from the Cava Nuova section were analyzed from sedimentological, geotechnical, chemical, mineralogical and biostratigraphical point of view. Sedimentological and geotechnical analyses, including density measurements, particle-size analysis, Atterberg limits and organic matter content, as well as XRF and XRPD analyses, resulted the most suitable techniques to identify the main fea- tures, at least for prevailing clayey deposits. Facies analysis and sedimentological data lead to recognize a clear depositional tran- sition from relatively deep to shallow lacustrine deposits, which was only rarely documented formerly through a single section. On the other hand, both geotechnical and mineralogical data indicate a compositional homogeneity for clay sediments, which does not correspond with facies lateral and vertical variability nor with palaeoenvironmental complexity. Despite its preliminary nature, this integrated method looks very promising to characterize the paleodepositional context, and some hypotheses on sediment source were also evaluated. Integrated data from Cava Nuova section, as well as minor well-logs in the immediate surroundings, were discussed and compared with existing data outcoming from the whole Dunarobba area. On the heels of the recent literature, this paper is aimed to put a new light on the complexity of the Fosso Bianco paleoenvironment. . Keywords: Dunarobba clays; geotechnical properties; XRF; XRPD; early Pleistocene 1. INTRODUCTION Inside the Pliocene-Pleistocene evolution of the Tiberino Basin (Umbria, central Italy, Fig. 1), the older evolution phases were testified by the diffuse occur- rence of clayey sediments, particularly in the southern sector. Deposits were associated to lacustrine environ- ment, from the ancient idea of the “Tiberino Lake” (Lotti, 1917, 1926), to the definition of the informal “Fosso Bianco” Unit (Ambrosetti et al., 1995a; Basilici, 1997), finally to the recent hypothesis of a “Paleo-ancient Lake” (Medici & Gliozzi, 2008; Spadi, 2018; Spadi et al., 2018a, 2018b), a relic from the late Messinian salinity crises. In this scenario, the Dunarobba area was always considered one of the key-point. The clayey deposits widely crop out in the area and they were studied since the beginning of the 20th century, due to the cultivation of lignite mines until 1950, and the production of bricks from 1950 onwards (Fig. 2; Cerquaglia, 1996). In this area, at about the half of eighties, the Dunarobba Fossil Forest (DFF in Figure 1) was discovered, which is now considered one of the most relevant geosites in Italy, due to the preservation in life’s position of more than forty mummified trunks, buried inside lacustrine clay deposits. Dunarobba is one of the few sites in the world yielding mummified and upright fossil forests, including (regardless to age) Axel Heiberg Island, in the Arctic region of Canada (45-40 Ma: Bigras et al., 1995; Wil- liams et al., 2008), the Bukkabrany swamp cypress for- est in Hungary (7.7-6.3 Ma: Kàzmér, 2008; Csaszar et al., 2009; Erdei et al., 2009; Gryc & Sakala, 2010; Erdei & Magyari, 2011), and the Stura di Lanzo Fossil Forest, in NW Italy (~3.0 Ma: Martinetto et al., 2007; Vassio et al., 2008). Due to its geological and environmental interest, the Dunarobba Fossil Forest is under consideration for inclusion in the World Heritage List (UNESCO), and most of the previous studies in the area converged on it. Previous works, indeed, were mainly focused on sedi- mentological and paleontological features (Manganelli et al., 1989, 1990, 2008; Biondi & Brugiapaglia, 1991; Esu & Girotti, 1991; Martinetto, 1994; Ambrosetti et al., 1995b, 1997; Basilici, 1997; Ciangherotti et al., 1998; Manganelli & Giusti, 2000; Medici & Gliozzi, 2008; Marti- https://doi.org/10.26382/AMQ.2018.14 88 Fig. 1 - Geographic localization and simplified geological scheme for the study area. CND=Cava Nuova; DFF=Dunarobba Fossil Forest; CT=Cava Toppetti; FB=Fosso Bianco (see text for details). Baldanza A. et al. 89 fraction (loam range) or with clay fraction (paleosol lev- els and clay range strata containing the tree trunks). These facies associations are strictly comparable with the sedimentological restoration proposed by Ambrosetti et al. (1995b, 1997) and Basilici (1997). The originary location of boreholes is shown in Figure 2. Since it was discovered (or rediscovered), the Du- narobba Fossil Forest began to show preservation prob- lems, not only directly related to weathering (Baldanza et al., 2009; Martinetto et al., 2014). Among the factors listed by Valentini et al. (1997) as responsible for the preservation of fossil trunks, the main and, unfortu- nately, the only strongly modified by excavation were the sealing properties of the clay mass. Unfortunately, in order to preserve the geosite, new invasive analyses are avoided inside the perimeter of the Dunarobba Fossil Forest from 1995 onwards. None- netto et al., 2014; Spadi, 2018; Spadi et al., 2018a, 2018b), on both the site area and the Fosso Bianco Unit outcropping area. Nonetheless, only few researches have conducted integrated multidisciplinary study. A research, carried out by Valentini et al. (1997), was aimed to define the geological, hydrogeological and geochemical processes, which allowed the preservation of unaltered wood, making the Dunarobba Fossil Forest an outstanding example of natural analogue, in the per- spective of evaluating the confinement conditions of geological sites for radioactive waste disposal. Barbieri et al. (1997) and Bozzano et al. (2000) presented a set of geotechnical and mineralogical data obtained on samples collected from four boreholes and one open pit quarry, defining two main facies: deep lacustrine and swampy lacustrine coast facies associations. The latter was characterized by samples with abundant sandy Fig. 2 - Evolution of the Cava Nuova quarrying area (CND) and of the Dunarobba Fossil Forest area (DFF) through time. Positions of strati- graphic sections and wells are reported. New integrate data from Dunarobba clay deposits theless, due to the enlargement of neighbouring excava- tion area starting from 2011, new outcrops of clayey deposits were discovered in the surrounding area, only partly investigated during previous research activities. In this framework, a new geognostic campaign was carried out between 2011 and 2012, making disposable lithostratigraphic, sedimentological, geological-technical and chemical-mineralogical data in the neighbouring Cava Nuova (CND in Figure 1) area. The present work - by using an integrated multidisciplinary approach - illus- trates and discusses new and old data from the Dunar- obba Fossil Forest area aiming to better understand some of the features of the "Fosso Bianco" lacustrine stage during the latest evolution of the Tiberino sedi- mentary basin, to try to re-discuss age constrains, and to propose some hypothesis on sediment provenance, based on mineralogical and sedimentological analyses. Although the preservation problems and the possible protection interventions are beyond the aims of this paper, all data collectable in this area are crucial not only for the correct paleoenvironmental restoration and geological evolution of the Tiberino Basin, particularly at the beginning of Early Pleistocene, but also in the light of the Dunarobba Fossil Forest preservation strategies. 2. GEOLOGICAL SETTINGS The study area is situated in southern Umbria, central Italy, inside the present-day southern Tiber Val- ley (Fig. 1). Through the Pliocene and the Pleistocene, this area pertained to the southwestern branch of the Tiberino Basin (also indicated as South Tiberino Basin), one of the NW-SE trending extensional basins, roughly parallel to the Apennine chain in lifting which character- ized the evolution of Northern Apennine from Miocene to recent (Barchi et al., 1991; Martini & Sagri, 1993). The geological and stratigraphic model for the South Tiberino Basin evolved through time, from the initial idea of the fill up of a wide lake (the “Tiberino Lake”: Lotti, 1917, 1926) to the definition of a lithostratigraphic succession of in- formal units divided according to sedimentological and biochronological data (Ambrosetti et al., 1995a, 1995b; Basilici 1997, 2000a, 2000b). Now substantial agree- ment lays on the individuation of two main depositional cycles: 1) the lacustrine phase, Pliocene to Early Pleis- tocene in age, and 2) the alluvial plain phase, Early Pleistocene to Middle Pleistocene in age. Relationships among units belonging to the two cycles are both marked by faults and/or by unconformities (Ambrosetti et al., 1995a; Basilici, 1997). The uppermost cycles is largely represented by the “S. Maria di Ciciliano” Unit and, locally, by the “Acquasparta” Unit (Ambrosetti et al., 1995a; Basilici, 1997). For an updated review and the state of the art, as well as for more detailed geological and stratigraphic schemes, we remind to Martinetto et al. (2014, 2017). The Dunarobba area lies inside the lowermost cycle and belongs to the informal Fosso Bi- anco Unit, which outcropping part is attributed to the Piacenzian-Gelasian interval according to magnetostrati- graphy (Abbazzi et al., 1997). The unit was mainly asso- ciated to a deep lacustrine environment with prevailing clay sedimentation (Facies association A: Ambrosetti et al., 1995a, 1995b, 1997; Basilici, 1997). Locally, clayey silts alternated to wavy- and/or cross-laminated silty sand laminae, parallel-laminated silty clay, lignite- bearing clay deposits, and lenticular sand bodies crop 90 Fig. 3 - Panoramic view of the Cava Nuova quarry area, as it appeared in January, 2012. >>>>> Fig. 4 - Cava Nuova sedimentological-lithostratigraphic section. Facies associations refer to Basilici (1997). Baldanza A. et al. 91 New integrate data from Dunarobba clay deposits 92 Fornaci Briziarelli Marsciano quarry, is the accessible outcrop of Fosso Bianco Unit closest (0.5 km) to the Dunarobba Fossil Forest. It was recently measured and sampled, for a total thickness of 44 m. Other 16 m, in continuity with the uppermost part, were previously de- scribed (Ambrosetti et al., 1995b, 1997; Basilici, 1997), while other six surveys wells were realized recently in the same area (Figs. 2, 5, 6). Sedimentological and geo- logical features of some selected sections are shown in figure 5. out. They were interpreted as Gilbert type delta deposits (Facies association B), marshy lacustrine shore deposits (Facies association C1, lithofacies a-e), and sandy lower shoreface deposits (Facies association C2, lithofacies f- i), respectively (Ambrosetti et al., 1995b; Basilici, 1997). A fourth type of deposits (Facies association D), related to distal alluvial fan environment, is somehow associ- ated to the same depositional cycle, although it was firstly indicated as “Ponte Naja” Unit (Ambrosetti et al., 1995a; Basilici, 1997). The Cava Nuova section (Figs. 3, 4), inside the Fig. 5 - Comparison of selected lithostratigraphic columns in the Dunarobba area from recent and further data. Baldanza A. et al. formed, including soil particle density measurement, particle-size analysis, Atterberg limits and organic matter content, following the ASTM D7263, ASTM D422, ISO/ TS 17892-12:2004 and ASTM D2974 standards, respec- tively. All the analyses were performed at the Applied Geology Laboratory (Department of Physics and Geol- ogy, University of Perugia). Although the European Standard (CEN ISO/TS 17892-12 2004) requires the cone penetrometer technique for Liquid Limit determina- tion and indicates the Casagrande cup as an alternative method, the latter was used in the present work. Despite the Casagrande cup method is much operator depend- ing than the cone penetrometer device (e.g., Di Matteo, 2012; Di Matteo et al., 2016), it has been chosen here in order to compare data with those presented by previous works (Candio et al., 1992; Barbieri et al., 1997; Boz- zano et al., 2000). 3. MATERIALS AND METHODS A preliminary facies description was carried out directly on the field, and a meter-stepped sampling was performed, with the collection of 44 samples, indicated as CND both in the stratigraphic section and in the ta- bles (Fig. 4, Tabs. 1-4). All samples were considered for microfossil content. About 500 g of each sample were washed in a hydrogen peroxide solution 30% in water and filtered through a Satylon 63 μm sieve. The retained material was dried and observed under a stereomicro- scope model Nissan optical equipped with optical fibres. Micropaleontological investigations allowed to identify on washed residues the biological component (mainly os- tracod valves and molluscs). On twelve samples, se- lected according to the preliminary facies description, sedimentological and geotechnical analyses were per- 93 Fig. 6 - Schematic distribution of facies across the Cava Nuova and the Fossil Forest areas. Landslide and anthropically-modified sectors are also perimetered. New integrate data from Dunarobba clay deposits The selected twelve samples were also analyzed for determining their chemical and mineralogical compo- sitions. Major, minor and trace chemical components of the selected samples were determined by X-ray Fluo- rescence (XRF) on pressed powder pellets utilizing an ARL 9400 XP+ sequential X-ray spectrometer under the instrumental conditions reported in Lezzerini et al. (2013). Within the range of the measured concentra- tions, the analytical uncertainties are <5% for all the components except for Na2O, P2O5, CaO, TiO2 and MnO which may occasionally attain <10% for very low concentrations (Lezzerini et al., 2013, 2014). The total amount of volatile components was determined as loss on ignition (LOI in 105-950°C temperature range). The CO2 content was measured by using the calcimetry method (Leone et al., 1988) on 300 mg of finely pow- dered samples previously dried at 105±5°C. The differ- ence between LOI and CO2 values was entirely ascribed to structural water (H2O).The qualitative mineralogical composition of the selected samples was obtained by X- ray powder diffraction analysis (XRPD) using a diffracto- meter with a Bragg-Brentano geometry. XRPD patterns were recorded using a Philips PW 1730 automatic dif- fractometer with the following settings: CuKα radiation obtained at 40 kV and 20 mA; slits: 1° divergence and scatter, 0.2 mm receiving; scan speed: 1°2θ/min; step size of 0.2°θ and counting time of 2s per step. Mineralogy of bulk samples was investigated on randomly oriented whole-rock powder. Oriented mounts of the <2μm fraction deposited on glass slides (2 mg/cm2: Lezzerini et al., 1995; Moore & Reynolds, 1997) were used in order to identify the clay mineral assemblages. The clay mineral assemblage was studied on Mg2+ and K+ saturated oriented aggre- gates of the <2μm fraction. The Mg2+ saturated speci- mens were measured in air-dried (AD) and glycolated (EG) states. For some Mg2+-saturated mounts, glycerol solvation was also performed. K+ saturated mounts were measured in air-dried conditions and after heating at 60, 110, 300 and 550°C. The basal reflections of clay miner- als (including mixed-layer phases) were analyzed by DIFFRAC.EVA suite for phase analysis. The XRPD analysis of the main phyllosilicates was carried out fol- lowing the procedures described by Moore & Reynolds (1997). Quantitative mineralogical composition of the selected samples was obtained by combining XRF and XRD data as suggested by Leoni et al. (2008). 4. CAVA NUOVA SECTION INTEGRATED DATA 4.1. Sedimentological and geotechnical features On the bases of lithological features, three main intervals are recognized already starting from the de- scription on the field (Figs. 3, 4): Interval 1 (from the base of the section up to 23 m): silty clay deposits, massive to slightly parallel-laminated. Fossil record is represented by rare gastropods, ostra- cods, bivalves, plant remains (leaves and wood’s frag- ments). Thin horizons of vegetal remains (“lignite” s.l.) and high-organic clay, black to deep brown in colour, also occur. In the interval between 6 and 19 meters, iron -enriched reddish horizons, crusts and nodules are com- mon inside clay, as reported by Gallo et al. (2014). Stud- ies are still in progress, and the preliminary data associ- ated these features to Fe-carbonate (FeCO3, siderite). Deposits in which nodules occur are associated to a relatively deep lacustrine environment, below the wave base; particularly in the lowermost section, features already correspond to these reported by Basilici (1997) for Facies association A. Nonetheless, the presence of siderite crusts and nodules looks here particularly perva- sive, on respect to the local, occasional occurrence al- ready reported (Ambrosetti et al., 1995b, 1997; Basilici 1997). Studies on the origin and meaning of siderite nodules are still in progress, and their complete descrip- 94 Tab. 1 - Sedimentological and geotechnical data of selected samples. Index properties of the CND soil samples are reported from the bottom (CND1) to the top (CND42) of the section. LL = liquid limit; PL = plastic limit; PI = plasticity index. PI is the difference between the liquid limit and the plastic limit (PI = LL-PL). Soils with a high PI tend to be clay, those with a lower PI tend to be silt. In case of PI=0, soils are considered to have little/no clay or silt and called non-plastic soil. A soil is also considered non-plastic when LL or PL cannot be deter- mined (N.P.). Gs = specific gravity (pycnometer test); USCS = Unified Soil Classification System (CH = clay of high plasticity - LL > 50; CL = clay of low plasticity - LL <50). Samples CND42 and CND42b refer to the same bed, and are reported as an example of variability in sand/silt/clay fractions according to lateral facies variation. Baldanza A. et al. tion is beyond the aims of this paper: at the moment, we choose to assign siderite-enriched layers to a local, not yet described variant inside Facies association A of former authors. Interval 2 (from 23 m to 38 m): silty clay deposits, alter- nated to fine sand and/or silty beds. Sands are usually ripple-laminated. The same thin horizons made of vege- tal remains (from charcoal to decimetric wood frag- ments), described in Interval 1, as well as the high- organic darkish clay, are still commonly documented inside clay, as well as the occurrence, although less frequent, of iron-enriched reddish horizons. Deposits are still referable to a lacustrine environment, showing clear shallowing upward trend, intermittent interaction with wave motion and re-sedimentation processes. Also these deposits can be inserted in the facies description of Basilici (1997): sandy/silty laminated beds are organ- ized as coupled lithofacies f-g, inside the Facies asso- ciations C2, while silty clay beds are still referable to Facies associations A. We assume Interval 2 may probably represent a transition between these two Fa- cies associations. Interval 3 (from 38 m to 44 m): in this interval, deposits are characterized by alternations of sand and silt, from parallel-laminated to cross-laminated. Plant remains, frequently including leaves, are common as well as lig- nite horizons. Deposits are associated to a lake margin subjected to wave motion, and can be still assigned to some features of Facies association C2 of Basilici (1997), although only lignite beds were previously re- ported. The same facies variation, as well as the same depositional architecture, are detectable in the six well- logs all-around the Cava Nuova section (Fig. 5), al- though the transition, represented by Interval 2 in the main section, is here less defined. These entire supple- mentary sections grade from clay deposits (presumably referable to relatively deep lacustrine facies) towards sand and silt coastal deposits, locally enriched in vege- tal remains and/or high-organic sandy clay. Minor varia- tions are imputable both to lateral facies variation and tectonics (Fig. 5). Geotechnical data are useful to carry out an inte- grated analysis with sedimentological and mineralogical ones. Table 1 summarizes the collected geotechnical data for the 12 soil samples from the Cava Nuova sec- tion. Grain size analyses confirm the preliminary field description, with occurrence of three main distributions. Lowermost samples (Interval 1), approximately until meter 25 (CND25 in Table 1), show constant percent- ages of clay (about 60%) and silt (about 40%) fractions. The silty component increases in the intermediate sec- tion (Interval 2), while the sand fraction, which is steadily less than 1%, prevails in the last meters (Interval 3). Figure 7 shows the location of soils on Skempton’s ac- tivity chart taking into account also the characteristics of soils published by Candio et al. (1992) and Barbieri et al. (1997), also belonging to fluvial-lacustrine deposits of Dunarobba Fossil Forest area. Soil samples of Cava Nuova falls along the 0.7 Activity (A) line and are char- acterized by high to very high swelling potential; they fall in the transition zone between deep lacustrine associa- tions and swampy coast facies as defined by Barbieri et al. (1997). Moving from the bottom to the top of the Cava Nuova section, PI values tend to gradually in- crease up to the medium part of the section (Interval 1), then they decrease upward reaching about 36% in Inter- val 2 and 24% in Interval 3, where -due to the low silt- clay fraction -a reduction of about 20 percentage points is observed in the last meters. Although minimal variations in grain size distribu- tion are also related to facies variability, this fact does not affect neither the general trend nor the distribution within the chart of Figure 7. Samples 42 and 42b (Tab. 1) are reported as an example: index properties depend from the percentage of clay fraction rather than from a difference in clay mineralogy. 4.2. Fossil content Throughout the section (Fig. 4), the fossil content is very poor and represented by rare freshwater gastro- pods (mainly Emmericia umbra specimens) and bi- valves, and by ostracods. The ostracod assemblages, although very poor, found along the Cava Nuova section are mainly repre- sented by the occurrence of genera belonging to the Candonidae family (Candona neglecta and Caspiocypris group), by the spotted occurrence of Cytherissa lacus- tris, and by the dominance of Cyprideis torosa in upper- most samples. The new species of Caspiocypris, re- cently identified (inside both FB and CT sections, see Figure 1) by Spadi et al. (2018a), and considered as endemic of the grey clays of Fosso Bianco Unit, also occur through the Cava Nuova section. Particularly, 95 Fig. 7 - Skempton’s activity chart (modified after Williams & Donaldson, 1980) for soils in Table 1 and those from Candio et al. (1992). Data from Barbieri et al. (1997) were elaborated from original graph (only boundaries of data were represented). New integrate data from Dunarobba clay deposits 96 Caspiocypris basilicii and Caspiocypris tiberina are documented from the base, followed by Caspiocypris tuderis and Caspiocypris posteroacuta starting from samples CND12 and CND 23 respectively. In the Inter- val 1 (Facies association A), the Caspiocypris group dominates the assemblages, only with few specimens of C. neglecta. A sharp decrease in Caspiocypris spp. abundance is detected from sample CND24 (about 24 m from the base of the section) onwards, in correspon- dence of the progressive increase in the sandy fraction and environmental energy, evidenced by wave ripples and cross laminations structures. From sample CND24 to CND39 (Interval 2), species of the Caspiocypris group occurs sporadically, with very low abundances. From sample CND40 (deposits of Interval 3), a sudden increase in the abundance of the ostracods is noted: assemblages are represented by Cyprideis torosa (90%) and, for the remaining 10%, by C. neglecta and C. basilicii. Thus, C. torosa characterizes assemblages associated to shallow lake/lake shore paleoenvironmen- tal condition. With regard to the taxon Cytherissa lacus- tris, it is found inside Interval 1 and Interval 3 with two isolated presence (samples CND 15 and CND 42); the low frequencies (no more of 5 valves) suggest a passive transport (wind action and/or dispersion by birds). 4.3. Chemistry and mineralogy Major, minor and trace chemical data are reported in Tables 2 and 3, respectively. The observed values indicate a relatively homoge- nous chemical composition for all the studied samples through the stratigraphic sequence, except for CND1, CND4 and CND15. In detail (Tab. 2), at the bottom of the sequence the samples CND1 and CND4 exhibit the highest levels of CO2 and CaO, as well the lowest SiO2 content. As a rule, SiO2 and Al2O3 slightly increase upwards, while contem- porarily CaO tend to decrease. This datum agrees with the increase upwards in the sandy fraction, enriched in quartz and feldspars. On the contrary, the sample CND15, in the middle of the sequence, shows a very low amount of CaO, with consequently high tenor of the other chemical compo- nents. Finally, noteworthy is that Fe2O3, which can be considered slightly constant through the section, ranges between 5.68 (CND15) and 10.60 wt.% (CND25): iron is probably a significant component in sediments and/or originary water, and variations are related to the occur- rence of levels enriched in siderite nodules. According to major and minor element trends, trace element keep comparable levels over the sequence; an exception is represented by sample CND15, characterized by low Sr, and high Cr, Ni and Ba, due to its high clay/silt+sand ratio (see Tab. 1). Nonetheless, barium keeps relatively high values throughout the section. The derived Rb/Sr ratio (Tab. 3), almost show the same trend, allowing to identify a marked peak at sample CND15. Tab. 2 - Major and minor elements of the analyzed samples (wt.%). Tab. 3 - Trace elements of the analyzed samples (ppm). Baldanza A. et al. Based on X-ray diffraction analysis on oriented mounts (Fig. 8), the clay mineral assemblage is due to illite (10 Å), kaolinite (7 Å) and expansible minerals (also detectable from the Skempton’s activity chart: Fig. 7); in particular, the latter ones consist in illite/ smectite mixed layers (according to Ce- sarano et al., 2018). The quantitative mineralogical composi- tion determined by combining XRPD and XRF data (Tab. 4) evidence a slight decreas- ing of calcite with a concomitant increasing of quartz contents over the sequence, from bottom to top. Out of trend is the sample CND15, for which the lower calcite and the higher quartz tenors are calculated. 4.4. Paleoenvironmental restoration Facies associations recognized in the Cava Nuova area are only partly comparable with these formerly described (Ambrosetti et al., 1995b, 1997; Basilici, 1997). On the other hand, swamp lacustrine coastal deposits looks circum- scribed at the Dunarobba Fossil Forest area. All the sedimentological, geotechnical and paleoecological data from the Cava Nuova section confirm the description and interpretation of facies initially made on the field, and the paleoenvironmental transition from relatively deep lake to coastal lacustrine is reliable. The lower- most section is characterized by fine-grained sediments (clay and silt), indicating a very low water energy and prevalence of decantation processes. The oligotypic assemblages lead to reconstruct the physical and chemical paleoenvironmental parameters. The data on ostracofaunas agree with those reported by Spadi (2018) and Spadi et al. (2018b), where the new pro- posed Caspiocypris species are reported as endemic of the “Paleolake Tiberino”. More generally, the common occurrence in assemblages of Caspiocypris group indi- cates deep lacustrine condition, with a minimum depth of at least 50 m (Spadi et al., 2018b), or >40 m (Ruiz et al., 2013). On the other hand, C. neglecta is widespread in springs, brooks and ponds connected to springs, and lakes (from shallow littoral to great depths).The species is distributed throughout the Holarctic (Meisch, 2000). The lake was characterized by cool and relatively deep waters, and by a silty to clayey floor, grading upwards to shallower depths, with higher water energy, relatively warmer temperature, and a coarse-grained (sandy/silty) floor. Candona neglecta, as well as some species of Caspiocypris (as C. pontica), tolerate very low salinity conditions (0.5-6‰: Medici & Gliozzi, 2008; Spadi et al., 2018a; 2018b); C. neglecta can also sustain low oxygen values (Meisch, 2000; Ruiz et al., 2013), and prefers alkaline to weakly acid pH (Ruiz et al., 2013). Basin waters were probably moderately to well oxygenated, showing low salinity (between 0.5 and 6‰), neutral pH and low concentration of CaCO3. As lower- most clay samples are enriched in both CO2 and Calcite (Tabs. 2, 4), this probably reflects the petrology of the mother rocks (source of sediments) rather than the wa- ter chemistry. This is not surprising, and a main prove- nance of fines from Cretaceous to Oligocene marly lime- stone Units (Figs. 1, 5) is highly probable. 97 Fig. 8 - XRD patterns collected on oriented mounts of CND25 sample, saturated with Mg++and K+, as example. Tab. 4 - Mineralogical composition of bulk samples based on XRD and XRF data. New integrate data from Dunarobba clay deposits 98 As the grain-size, the geotechnical parameters and the fossil record are steady throughout the lower sec- tion, stable and long-lasting deep lacustrine conditions can be assumed. Time inferences imply considerations on stratigraphic constrains and sedimentation rate, which are discussed later (see § 5.2). Deposits in the intermediate section (from sample CND23 to CND38) show a coarsening upward trend, with increase of the silt fraction, and the decrease in ostracods specimens. These data are interpreted in terms of shallowing trend and progradation of coastal systems; the prevalence of silty fraction in respect to the clay presumably conditioned the ecology of benthic forms. In the uppermost section (samples CND39- CND43), the coarsening upward trend proceeds, with a prevailing sandy fraction, and a high content in plant remains, both testifying for the increase in energy and the reduction in depth. The fossil content is dominated by C. torosa, species related to transitional waters, shal- lower depths and/or sandy floor. Deposits are associ- ated to a wave-influenced lacustrine coastal environ- ment. As for the lower section, chemical and physical parameters (water temperature, oxygen content, salin- ity, pH and concentration of CaCO3) seem to be steady in these two other portions. On the other hand, the in- crease in sandy portion clearly indicates a coastal shift: in this sense, a higher sedimentation rate is expected. Throughout the section, mineralogical data show high values in trace elements values (Tab. 3), particu- larly for barium, reaching 441 ppm in sample CDN15, but constantly over 200 ppm. As evidenced by Muller (2014), barium appears to play an important role in bac- terial metabolism, as well as iron and other trace ele- ments. Local condition could be related to the develop- ment of freshwater algal, protozoan, and bacteria, all organisms able to produce intracellular barite from geo- logical material like clay (Muller, 2009, 2014, with refer- ences therein). 5. DISCUSSION Inside the wide outcropping area of “Fosso Bianco” Unit (Fig. 1), Cava Nuova section and Dunarobba Fossil Forest areas represent single, although related, obser- vation point, and this bias needs to be considered in generalizing the results of analyses. Nonetheless, some points can be highlighted. 5.1. Relationships between Cava Nuova and Fossil Forest areas A crucial point in every paleoenvironmental resto- ration lies on the originary vertical and lateral relation between deposits pertaining to the two areas of Cava Nuova and Fossil Forest, respectively. Although they are really close (Figs. 2, 5, 6), facies cropping out in the two areas are almost different. A detailed facies descrip- tion and redefinition is beyond the aims of this work, and we choose still to refer to Facies associations proposed by previous authors (Ambrosetti et al., 1995b, 1997; Basilici, 1997, 2000b): nonetheless, some differences were remarked, and, in our opinion, a partial integration is probably needed. A simplified chart of facies cropping out (Fig. 6) reveals how Swampy lake margin deposits and Wave-dominated lake margin deposits (Facies as- sociations C1 and C2 of Basilici, 1997, respectively) seem to be confined to the two areas. Both types of deposits overlay Deep lacustrine deposits (Facies asso- ciation A, with variations: Basilici, 1997), which are rec- ognized only on quarry fronts and well-logs (Fig. 5). In the fields all around (Fig. 6), the occurrence of plant remains (mainly large wood fragments) were commonly documented (F. Famiani, pers. comm. 2018), although they can be alternatively related to both types of lake margin. As firstly proposed by Ambrosetti et al. (1995a) and Basilici (1997), the relations between the two areas at least partly result from Late Pleistocene and Holocene tectonics. As shown in the transect reconstructed across the Cava Nuova and the Fossil Forest areas (Fig. 5), minor faults are documented only in the fresh-exposed quarry fronts, while, according to the facies distribution, deposits all-around the Fossil Forest look undisturbed. Thus, a NE-SW striking fault, with a presumed transten- sive component, associated to the main fault system (Fig. 1) and dividing the two areas (Cava Nuova and Dunarobba Fossil Forest), although reasonable, it is not directly detectable on the field and can only be pre- sumed. Moreover, the direction of movement and the amount of dislocations are not easy to quantify. As dis- cussed later, and according to sedimentation models formerly proposed (Ambrosetti et al., 1995b, 1997; Basilici, 1997; Martinetto et al., 2014; Spadi et al., 2018b), the two types of lake margin could be strictly related, thus tectonics can be considered a secondary factor in present-day distribution of facies on the field. 5.2. Stratigraphic constrains A definitive age attribution for deposits of Fosso Bianco Unit cropping out in the Dunarobba area is still debated. A time range from ~3.1 to ~2.0 Ma is com- monly accepted for the whole unit, although its base was never clearly identified, nor in outcrop or in wells. Ac- cording to Ambrosetti et al. (1995a), the southern branch of the Tiberino Basin accommodated ~1 km of post- Miocene deposits; assuming, for the sedimentation rate, the mean values reported in the previous section, it im- plies a possible ~5 Ma duration for deposition inside the basin. Thus, Fosso Bianco Unit could be extended downwards fairly closer to the Miocene-Pliocene bound- ary. On the other hand, only two sections were cali- brated to magnetostratigraphic scale (Abbazzi et al., 1997; Pontini & Bertini, 2000; Pontini et al., 2002; Marti- netto et al., 2014): the Cava Toppetti section, near Todi, and the Fosso Bianco section, south from Dunarobba Fossil Forest (CT and FB in Figure 1, respectively). As the first one documents the Gauss (successive to Kaena) - Matuyama (nearly successive to Reunion) magnetostratigraphic interval (between about 2.8 and 2.02 Ma), the second only records 87 m-thick reversal interval, calibrated to the Gelasian portion of the Matuyama chron (about 2.5-2.2 Ma). Pontini & Bertini (2000), by meant of pollen record, also remark the oc- currence of four short but clearly recognizable “climatic” phases, in which the climate oscillated between ”temperate-warm” (pollen zones II and IV) and “subtropical humid” (pollen zones I and III). These four Baldanza A. et al. intervals took place between the 102 (or 100) and 82 isotopic stages (i.e.: between 2.56-2.52 and 2.16 Ma). The recent phyletic revision of Caspiocypris group lead on by Spadi et al. (2018 a, 2018b), and their strati- graphic, although local, use, also shade a new light in the study of Fosso Bianco deposits. The tentative of phylogenetic relations within the Caspiocypris species flock of the Palaeolake Tiberino suggested by Spadi et al (2018a) evidenced that the appearance of C. tuderis is a remarkable local event that took place inside the lower (but not basal) Early Pleistocene (Gelasian), in both Cava Toppetti and Fosso Bianco sections. Inside Cava Nuova section, this event is documented in the middle portion of Interval 1 (sample CND12, Fig. 4), where the assemblages are characterized by C. basilicii, C. tiberina and C. perusia, in the same stratigraphic position as in the other two sections. Although this local event could be time-tansgressive, and new confirma- tions are expected, its relevance for local stratigraphy and correlations appears evident. As pointed out by previous authors, the deep lacustrine deposits (Facies association A) inside Fosso Bianco Unit record significant fluctuations in sedimenta- tion rate, depending on the stratigraphic and the pa- leoenvironmental/paleogeographical situation (Abbazzi et al., 1997; Pontini & Bertini, 2000; Pontini et al., 2002; Martinetto et al., 2014). Average values vary from 14-15 cm/kyr at Cava Toppetti, to 21-28 cm/kyr at Fosso Bi- anco (CT and FB sections in Fig. 1, respectively). Ac- cording to these values, the C. tuderis local event should be dated to about 2.5-2.45 Ma, in both the sec- tions. Assuming the same sedimentation rates for the lowermost, deep lacustrine deposits at Cava Nuova, the first 23 m could have been deposited in a time between about 80-100 kyr (sedimentation rate of 21-28 cm/kyr) and about 155-160 kyr (sedimentation rate of 14-15 cm/ kyr). Thus, the deep lacustrine phase cropping out at Cava Nuova should be comprised between about 2.58- 2.55 and 2.4-2.37 Ma. Thus, a Gelasian age can be finally assumed also for the lowermost, deep lacustrine deposits (Interval 1) cropping out at Cava Nuova. As the section shows a continuous facies transition inside a shallowing-upward trend (Fig. 4), an almost continuous decreasing in age is also expected, with no major time gaps and hiatus. As- suming the same sedimentation rate values also for Intervals 2 and 3, the whole section (43 m) should re- cord a 150 to 280 kyr time interval. That is, the top of the section should be dated between 2.4 and 2.3 Ma. On the other hand, throughout that facies transition variations in sedimentation rate are expected. Although the alternation of dryer and wetter intervals recorded in Fosso Bianco section seems not to find a direct sedi- mentological and/or paleontological correspondence at Cava Nuova (Fig. 4), the lowermost section (Interval 1 and part of Interval 2) records cool water conditions, and it could be associated to pollen zone II of Pontini & Ber- tini (2000). This hypothesis looks consistent with the geochemical peak in Rb/Sr ratio noted in sample CND15 (Tab. 3). This ratio may reflect mother rock com- positions, as well as the Sr content could result from different processes, such as primary carbonate produc- tivity, or early diagenetic calcite dissolution. Nonethe- less, it could also be cautiously considered as indicative of the prevalence of physical or chemical weathering in the catchment area (for example, see Xu et al., 2010). Although data are too scarce to reconstruct a climatic trend, in our opinion the hypothesis that this peak could be related to a moment of cool/arid climate, which fa- voured physical weathering, and which falls shortly after the C. tuderis local event (around 2.45 Ma), should be considered in paleoenvironmental restoration. In this way, part of Interval 2 and Interval 3 could be associated to at least part of pollen zone III, and an age about 2.24 for the top of the section could be reli- able. To assign an age to deposits from the Fossil Forest area is even more speculative; nonetheless, some hy- potheses need to be considered. Ambrosetti et al. (1995b), mainly on the base of paleoecologic/ paleoclimatic topics, firstly proposed a Piacenzian age for the Fossil Forest. Assuming the sedimentation rates previously discussed, and its top being not younger than 2.588 Ma, the base of the composite section (extending from top of S2 to substrate in S1-S7 sections: Fig. 5) should be dated between about 3.49 and 3.06 Ma. It implies the Fossil Forest area records the very begin- ning of sedimentation inside the Fosso Bianco Unit, which was never documented otherwise. The older de- posits inside Cava Toppetti section, indeed, are dated to about 2.8 Ma. In central Italy, so ancient deposits were documented in the Valdarno Basin (Napoleone et al., 2003), and in the eastern branch of the Tiberino Basin (fossil record inside the lignites of Morgnano, attributed to Triversa Faunal Unit: Bizzarri et al., 2018, and refer- ences therein). Although this hypothesis cannot be to- tally abandoned, some question arises. The presence of such old deposits in a restricted area, fairly close sur- rounded by younger (Gelasian) deposits, and both older and younger deposits indifferently overlaid by the same lithostratigraphic unit (S. Maria di Ciciliano Unit: Am- brosetti et al., 1995a), can be justified only in terms of main tectonic displacement, occurring in a very short time interval (200 kyr) between Gelasian and Calabrian (Early Pleistocene tectonics). As synthesized in Figure 1, the present geological setting seems to be mainly related to Middle-Late Pleistocene tectonics, and to the occurrence of an important NE-SW oriented strike-slip area between Dunarobba and Acquasparta villages (see also geological schemes proposed by Ambrosetti et al., 1995a, Basilici, 1997, and Spadi et al., 2018b). The same shallowing-upward trend, from deep lacustrine deposits to coastal deposits, is documented in both the Cava Nuova and Fossil Forest areas, and in both cases idealized composite stratigraphic sections do not record the occurrence of a second, superimposed cycle. That is, both situations appear to record a “final”, although local, evolutive step. This cannot be easily justified if the ages are significantly different, and the presence of several overlapping sedimentary cycles should be expected. On the bases of ostracofauna, Spadi (2018) and Spadi et al. (2018b) proposed deposits of the Fossil Forest area are almost coeval to the Ponte Naja Unit (2.2-2.0 Ma: Abbazzi et al., 1997; Pontini et al., 2002). As discussed in Martinetto et al. (2014), the Dunarobba 99 New integrate data from Dunarobba clay deposits 100 Fossil Forest record, particularly pollen record from S2 section, can be also correlated with the pollen zone III of Fosso Bianco section (Pontini & Bertini, 2000). In this way, Cava Nuova Section and Dunarobba Fossil Forest paleofloristic records were recomprised in the persis- tence of “humid thermophilous plant taxa of East Asian affinity” (HUTEA) as “Pliocene relics”, and inside their progressive disappearance through Early Pleistocene in southern Europe, as proposed by Martinetto et al. (2017). As the local trend of progressive decrease in depth, although with different evidences, is recorded in all the sections of the Fossil Forest and Cava Nuova areas (Fig. 5), the hypotheses the two areas are coeval, and that both record the transition between pollen zones II and III, seem to be reasonable. As assumed for the Cava Nuova section, swampy coastal facies interval could be dated to about 2.4-2.2 Ma; thus, the base of the composite section (S1-S7 in Figure 5) should be positioned between about 2.95 and 2.8 Ma. In the Fosso Bianco section, the pollen zone III is associated to deep lacustrine clay deposits enriched in thin mud- or debris-flows deposits, bearing vegetal re- mains (Pontini & Bertini, 2000; Martinetto et al., 2014), which can be justified as resedimentation deposits from coastal areas enriched in vegetation, as documented in both the Cava Nuova and Fossil Forest areas. For all the exposed reasons, the time gap between deposits pertaining to the two areas (Cava Nuova and Dunarobba Fossil Forest) and their related sections is reasonably minimum, and an average age comprised between 2.5 and 2.2 Ma for the outcropping portion of Fosso Bianco Unit in the area can be assumed. 5.3. Comparison with former mineralogical and geo- technical data Considering the distribution in Figure 7, samples collected throughout the Cava Nuova section are com- parable with the data of Barbieri et al. (1997): thus, the belonging to the variability of Fosso Bianco Unit sedi- ments is clearly demonstrated. On the other hand, data from Skempton’s activity chart does not allow a unique assignment to deep lacustrine (A) and/or swampy coast (C1) facies associations. Moreover, our uppermost sam- ples fall in the area of the graphic, although they are clearly associated to a wave-dominated coast. Contrar- ily to what supposed by Barbieri et al. (1997) and Boz- zano et al. (2000), this parameter does not fully express the facies variability inside the Fosso Bianco Unit. Oth- erwise, when considered together with numeric grain size distribution, it has a more direct correlation with sedimentological intervals described in the field. Thus, if the sedimentological interpretation is needed, the method can be an additional tool for the analysis of ei- ther short sections or wide datasets. On regard to the mineralogical composition of clay fractions, significant bias occur when old and new data are compared, as well as in the comparison between “mineralogical” and ”geotechnical” content of clay. Nonetheless, both geo- technical and mineralogical data highlight the uniformity in clay composition, while only the amount of the clay fraction varies throughout the section. This quite homo- geneous composition throughout the section confirms the hypothesis of a steady supply from the same source area through time. However, the slight increasing of sandy fraction (consisting mainly in quartz) from bottom to top indicates a modification of sedimentation condition from deep to coastal environments, with intermittent inlets of terrigenous materials highlighted by the occur- rence of quartz-rich levels in the middle of the sequence (see for example sample CND15), becoming more fre- quent and even dominant upwards. 5.4. Insights for the basin evolution From what discussed in the previous paragraphs, some paleoenvironmental, paleogeographic and geo- logical implication can be proposed. - Measured sedimentological/geotechnical parame- ters, such as grain size and Skempton’s activity for clay, show a linear trend throughout the section ranging from very high to high (Activity ≃ 0.7), thus indicating a coars- ening upward of deposits inside the basin rather than a clear variation of supply. This statement is supported by the overall uniformity in the mineralogy of clays, whose assemblage is due to constant level of illite/smectite, and kaolinite over the sequence. Assuming a time interval of 150 to 280 kyr (for the deposition of the whole Cava Nuova section), or of 80 to 160 ky for sedimentation of the deep lacustrine clay alone, this homogeneity implies a long-lasting supply of sediments from the same feeding area. Nonetheless, discontinuities in the material supply, suggested by the variability in sand-silt enriched levels, especially in the middle and in the upper part of the section, needs to be explained. This could imply the occurrence of occasion- ally climatic and/or tectonic events, determining the inlet of quartz-rich sediments. This trend is particularly clear for uppermost section (Fig. 4). As pointed out in § 5.3, a local source for deposits is inferred: in this way, quartz- rich fraction can derive by erosion of Oligocene-Miocene sandstone Units (Fig. 1). This wide paleoenvironmental and compositional homogeneity also lead to exclude a far source of sediments, such as a structured river envi- ronment. That is, resediments inside the basin derived from local, contiguous environments, such lake coasts and minor fan-delta. - The deep lacustrine phase inside the Fosso Bianco Unit (i.e.: the one documented by Facies association A of Basilici, 1997) can be considered as a stable, invari- ant, and long-lasting environment. Such phase (and the Fosso Bianco Unit as a whole) is characteristic of the South Tiberino Basin, and finds no equivalent in the other sectors of the basin, each one evolving through time in its peculiar way (Ambrosetti et al., 1995a; Coltorti & Pieruccini, 1997; Melelli et al., 2010; Bizzarri et al., 2011, 2018; Pucci et al., 2014; Bucci et al., 2016; Mira- bella et al., 2018). On the bases of evidences provided by the ostracofaunas (Medici & Gliozzi, 2008; Spadi, 2018; Spadi et al., 2018a, 2018b), this lake can be prop- erly considered a relic of late Messinian -Zanclean ma- rine to brackish conditions (“Paleo-ancient lake”), a wa- ter mass with altered salinity imprisoned into a continen- tal basin, with no input-output dynamics. In fact, no data are available about the beginning of this phase, and maybe only real deep perforations could provide defini- Baldanza A. et al. tive information. - On the other hand, the clear transition upwards from deep lake to coastal lake conditions, directly de- scribed in the Cava Nuova section, is also found in all the sections in the area (Fig. 5). This datum needs to be discussed, and three main questions arise. First, this variation could reflect a climatic trend, as well as it may be driven by tectonics. As a second problem, it is not finally clear where, in the stratigraphic succession of Fosso Bianco Unit, this variation takes place, if in the uppermost part or just at the end of its deposition. The third point is if this regressive trend marks the definitive change in environmental conditions (e.g.: the end of depositional cycle 1 and the transition to depositional cycle 2), or if phases of rise and fall of the lake’s level repeated through time. According to what discussed in § 5.2, the Cava Nuova and Dunarobba Fossil Forest ar- eas record the uppermost, but not final, sedimentation inside the Fosso Bianco Unit. As deposits in the whole area can be considered nearly coeval, the outcrop con- ditions mainly derive from the Late Pleistocene- Holocene evolution of the Tiber Valley. In northern sec- tor (Cava Toppetti area: Fig. 1), deep lacustrine sedi- mentation extends until about 2.0 Ma. It is possible that the unconformity between the cycles 1 and 2 reasonably cut a sensible part of the lower cycle, particularly in the southern portion of the basin (Ambrosetti et al., 1995a; Basilici, 1997). Inside the >800 kyr lasting sedimentation of Fosso Bianco Unit, study sections (Fig. 5) are rele- vant but only represent less than the 1/3 of the outcrop- ping Fosso Bianco Unit. They better photograph a single evolution phase, or the transition between two phases, rather than synthesize the whole sedimentological, cli- matic, stratigraphic variability inside the Unit. Shorter phases (high-order cycles) probably repeated through time, although they seem to be clearly detected only at Fosso Bianco (Pontini & Bertini, 2000; Martinetto et al., 2014). The fossil record from plants (Martinetto et al., 2014, 2017), vertebrates (Abbazzi et al., 1997; Pontini et al., 2002), and ostracofauna (Spadi, 2018; Spadi et al., 2018b), as well as the geological-sedimentological interpretation (Ambrosetti et al., 1995a), suggest a gen- eral trend inside the “Fosso Bianco” cycle, from deep lacustrine to coastal facies. Thus, Fosso Bianco Unit (and the related units like Ponte Naja Unit) document at least part of a 3rd order cycle. Regarding what drove the changes, Cava Nuova Section and Dunarobba Fossil Forest somehow insert in the phase of global climatic instability, which characterized the Gelasian stage, but they mainly record local changes. The longer persis- tence of some floristic species in central Italy after the end of the Piacenzian with respect to the northern part of this country, probably due to the intermediate pattern between climatic changes described for the central Mediterranean littoral and the North Apennine ones (Martinetto et al., 2014, 2017), must be considered. According to Martinetto et al. (2014) the remarkable Fossil Forest at Dunarobba was produced by an ancient swamp vegetation dominated by Glyptostrobus eu- ropaeus; trees grew in a wetland with high sedimenta- tion rate, bordering the ancient lake, and both disap- peared due to the establishment of well-drained pa- leoenvironmental conditions, testified by a paleosol pro- file (Basilici, 1997). The successive restoration of wet- land conditions was not sufficient for the preservation of in situ tree trunks (Basilici, 1997). Coastal facies asso- ciations, although different between Cava Nuova and Fossil Forest areas, were probably both related to a lowering phase in the lake’s level. The relative uniformity of the pollen curves (Martinetto et al., 2014) suggests that the extra-local vegetation and climate remained stable for a long time span. All these data fit well with the mineralogical- petrographic homogeneity we record in the Cava Nuova section, so the paleoenvironmental variation was proba- bly better explained by local (or even regional) morpho- logical changes induced by tectonics. Tectonics proba- bly led to a strong activity of faults at the boundary, pro- ducing a large sediment supply from the same feeding area, and causing the progradation of coastal lake sys- tems and the fall in the level of the lake, as proposed by Ambrosetti et al. (1995b) and Basilici (1997). The data discussed herein, and their comparison with old and recent papers, lead to suggest this fall represented the main, somehow definitive, phase closing the lower depo- sitional cycle. Far from a definitive understanding and modelling of timing and evolution for the Tiberino Basin, what dis- cussed in this paper ranks in the wake of the recent literature, which tries to shade new light on such prob- lematic. Furthermore, it is probably time to reconsider and partly revise the general geological, stratigraphic, and paleoenvironmental scheme for the Fosso Bianco Unit. 6. CONCLUSIONS Inside the Pliocene-Pleistocene evolution of the Tiberino Basin, the vertical and lateral relations of de- posits in the “Fosso Bianco” Unit are a key-point and, unfortunately, they are still an open problem. The new collected data, far to dispel all doubts, put new light to better understand the paleoenvironmental features of at least part of the lacustrine phase. Such long-lasting en- vironment, although fairly structured, survived almost invariant for at least 2.1 Ma. Both the Dunarobba Fossil Forest and the Cava Nuova areas record analogous depositional trend, from deep lake to coastal lake, al- though represented by different facies: this phase, last- ing 150 to 280 ky, can find correspondence in the transi- tion between pollen zones II and III documented in the “Fosso Bianco” section by Pontini & Bertini (2000) and correlated with the early Gelasian. Although new evi- dences are needed, this phase could continue towards the Gelasian-Calabrian boundary, until 2.2-2.0 Ma. Dur- ing the Gelasian climatic instability, the system seemed to accommodate local variations rather than a global trend, as expected for the global datum of southern Europe. Starting from a section recording a continuous facies transition, this work proposed a multidisciplinary approach to better understand both sedimentological- geotechnical and mineralogical aspects. The analytical procedures reveal useful, and we expect they can be a tool to characterize other clayey soils outcropping in the 101 New integrate data from Dunarobba clay deposits 102 Baldanza A. et al. Tiberino Basin fluvio-lacustrine context, as well as in other paleolacustrine environments. ACKNOWLEDGMENTS We kindly thank dr. F. Posati and dr. F. Famiani for their support on the field, and dr. F. Cavalcante and an anonymous reviewer for their precious comments and suggestions in improving the manuscript. REFERENCES Abbazzi L., Albianelli A., Ambrosetti P., Argenti P., Basilici G., Bertini A., Gentili S., Masini F., Napoleone G., Pontini M.R. (1997) - Paleontologi- cal and sedimentological record in Pliocene distal alluvial fan deposits at Cava Toppetti (Todi, central Italy). In: Monegatti P. (ed), Proceedings del XII convegno della Società Paleontologica Italiana. Bollettino della Società Paleontologica Italiana, 36, 5-22. Ambrosetti P., Basilici G., Capasso-Barbato L., Carboni M.G., Di Stefano G., Esu D., Gliozzi E., Petronio C., Sardella R., Squazzini E. (1995a) - Il Pleisto- cene inferiore nel ramo sudoccidentale del bacino tiberino (Umbria): aspetti litostratigrafici e biostrati- grafici. Il Quaternario, 8, 19-36. Ambrosetti P., Basilici G., Ciangherotti A.D., Codipietro G., Corona E., Esu D., Girotti O., Lo Monaco A., Meneghini M., Paganelli A., Romagnoli M. (1995b) - La foresta fossile di Dunarobba (Terni, Umbria, Italia centrale): contesto litostratigrafico, sedimen- tologico, palinologico, dendrocronologico e paleo- malacologico. Il Quaternario, 8, 465-508. Ambrosetti P., Basilici G., Di Filippo M., Di Maio R., Gentili S., Lombardi S., Patella D., Toro B., Brozzetti F., Cammarano F., Cattuto C., Ciotoli G.C., Di Giacomo B., Etiope G, Fisauli M., Gregori L., Piscitelli S., Pontini R., Ruspandini T. (1997) - The stratigraphic, depositional, geomorphological and structural characteristics of the sediments which contain the Fossil Forest of Dunarobba (FFD). In: Valentini G., Lombardi S., Bozzano F., Scarascia Mugnozza G. (eds), Analysis of the geo -environmental conditions as morphological evolu- tion factors of the sand-clay series of the Tiber valley and Dunarobba forest preservation, EUR 17479 EN, Final Report, EUR OP, Luxembourg, 9- 104. Baldanza A., Sabatino G., Triscari M., De Angelis M.C. (2009) - The Dunarobba Fossil Forest (Umbria, Italy): mineralogical transformations evidences as possible decay effects. Periodico di Mineralogia, 78, 51-60. Barbieri M., Bozzano F., Duddrige G.A., Grainger P., Mottana A., Scarascia Mugnozza G., Valentini G. (1997) - Studies of the natural barriers in the Du- narobba Forest area. In: Valentini G., Lombardi S., Bozzano F., Scarascia Mugnozza G. (eds), Analy- sis of the geo-environmental conditions as mor- phological evolution factors of the sand-clay series of the Tiber valley and Dunarobba forest preserva- tion, EUR 17479 EN, Final Report, EUR OP, Lux- embourg, 144-211. Barchi M., Brozzetti F., Lavecchia G. (1991) - Analisi strutturale e geometrica dei bacini della media Valle del Tevere e della Valle Umbra. Bollettino della Società Geologica Italiana, 110, 65-76. Basilici G. (1997) - Sedimentary facies in an extensional and deep - lacustrine depositional system: the Pliocene Tiberino Basin, Central Italy. Sedimentary Geology, 109, 73-94. Basilici G. (2000a) - Pliocene lacustrine deposits of the Tiberino Basin (Umbria, central Italy). In: Gier- lowski-Kordesch E.H., Kelts K.R. (eds), Lake ba- sins through space and time. AAPG Studies in Geology, 46, 505-514. Basilici G. (2000b) - Floodplain lake deposits on an early Pleistocene alluvial plain (Tiberino Basin, Central Italy). In: Gierlowski-Kordesch E.H., Kelts K.R. (eds), Lake basins through space and time. AAPG Studies in Geology, 46, 535-542. Bigras C., Bilz M., Grattan D.W., Gruchy C. (1995) - Erosion of the Geodetic Hills Fossil Forest, Axel Heiberg Island, Northwest Territories. Arctic, 48, 342-353 Biondi E., Brugiapaglia E. (1991) - Taxodioxylon gypsaceum (Goppert) Krausel in the fossil forest of Dunarobba (Umbria, Central Italy). Flora Mediter- ranea, 1, 111-120. Bizzarri R., Albianelli A., Argenti, P. Baldanza A., Colacicchi R., Napoleone G., (2011) - The latest continental filling of Valle Umbra (Tiber Basin, central Italy) dated to one million years ago by magnetostratigraphy. Il Quaternario 24(1), 51-65. Bizzarri R., Corrado P., Magri D., Martinetto E., Esu D., Caprai V., Colacicchi R., Napoleone G., Albianelli A., Baldanza A. (2018) - Palaeoenvironmental and climatic inferences from the late early Pleistocene lacustrine deposits in the eastern Tiberino Basin (central Italy). Quaternary Research, 90(1), 201- 221. Bozzano F., Parroni F., Scarascia Mugnozza G., Valentini G. (2000) - Caratterizzazione geologico- tecnica dei depositi continentali della “Foresta Fossile” di Dunarobba (TR). Atti del 5° Conv. Naz. Giovani Ricercatori di Geologia Applicata, Cagliari 1996, 33-41. Bucci F., Mirabella F., Santangelo M., Cardinali M., Guzzetti F. (2016) - Photo-geology of the Monte- falco Quaternary Basin, Umbria, Central Italy. Journal of Maps. Doi: 10.1080/17445647.2016.1210042 Candio S., Damiani A.V., Federico A., Setti M., Veniale F., Vinti G. (1992) - Caratterizzazione geotecnica e mineralogica di unità argillose dell’Umbria. Geolo- gia Applicata e Idrogeologia, 27, 123-133. Cerquaglia Z. (1996) - La Foresta Fossile di Dunarobba. Ediart, Todi, pp. 106. Cesarano M., Bish D.L., Cappelletti P., Cavalcante F., Belviso C., Fiore S. (2018) - Quantitative min- eralogy of clay-rich siliciclastic landslide terrain, Sorrento Peninsula, Italy, using a combined XRPD and XRF approach. Clay and Clay minerals, 66(4), 353-369. Ciangherotti A.D., Esu D., Girotti O. (1998) - Review of the history of the Late Neogene-Early Quaternary non-marine molluscs of Italy. Mededelingen - Ned- erlands Instituut voor Toegepaste Geoweten- schappen TNO, 60, 491-498. Coltorti M., Pieruccini P. (1997) - The southeastern Tiber Basin (Spoleto, Central Italy): geology and stratigraphy of Plio-Pleistocene sediments. Il Qua- ternario, 10, 159-180. Csaszar G., Kazmer M., Erdei B., Magyar I. (2009) - A possible Late Miocene fossil forest Paleopark in Hungary. Notebooks in Geology, 3, 11. Brest, Book 2009/03, Chapter 11, (CG2009_BOOK_ 03/11), 121-133. Di Matteo L. (2012) - Liquid limit of low- to medium- plasticity soils: comparison between Casagrande cup and cone penetrometer test. Bulletin of Engi- neering Geology and the Environment, 71(1), 79- 85. Di Matteo L., Dragoni W., Cencetti C., Ricco R., Fucsina A. (2016) - Effects of fall-cone test on classification of soils: some considerations from study of two engineering earthworks in Central Italy. Bulletin of Engineering Geology and the Environment, 75(4), 1629-1637. Erdei B., Magyari E. (2011) - Late Miocene plant re- mains from Bükkabrany, Hungary. Studia botanica hungarica, 42, 135-151. Erdei B., Dolezych M., Hably L. (2009) - The buried Miocene forest at Bukkabrany, Hungary. Review of Palaeobotany and Palynology, 155, 67-79. Esu D., Girotti O. (1991) - Late Pliocene and Pleisto- cene assemblages of continental molluscs in Italy. A survey. Il Quaternario, 4, 137-150. Gallo G., Baldanza A., Bizzarri R., Cavalazzi B., Comodi P., Frondini F., Rigo M. (2014) - The relevance of biosignatures in siderite-nodules: an example from Quaternary continental deposits (Tiberino Basin, Italy). EANA 2014 Abstract Volume, Edimburg, Scotland. Gryc V., Sakala J. (2010) - Identification of fossil trunks from Bükkabrany newly installed in the visitor cen- tre of the Ipolytarnòc Fossil Nature Reserve (Novohrad- Nograd Geopark) in the northern Hun- gary. Acta Universitatis Agriculturae et Silvicul- turae Mendelianae Brunensis, 58, 117-122 Kàzmér M. (2008) - The Miocene Bükkábrány Fossil Forest in Hungary -field observations and project outline. In: Galácz A. (ed.), 125th Anniversary of the Department of Palaeontology at Budapest University -A Jubilee Volume. Hantkeniana, 6, 229 -244. Leone G., Leoni L., Sartori F. (1988) - Revisione di un metodo gasometrico per la determinazione di cal- cite e dolomite. Atti della Società Toscana di Scienze Naturali - Memorie Serie A, 95, 7-20. Leoni L., Lezzerini M., Saitta M. (2008) - Calcolo com- puterizzato dell’analisi mineralogica quantitativa di rocce e sedimenti argillosi basato sulla combinazi- one dei dati chimici e diffrattometrici. Atti della Società Toscana di Scienze Naturali - Memorie Serie A, 113, 63-69. Lezzerini M., Sartori F., Tamponi M. (1995) - Effect of amount of material used on sedimentation slides in the control of illite “crystallinity” measurements. European Journal of Mineralogy, 7, 819-823. Lezzerini M., Tamponi M., Bertoli M. (2013) - Reproduci- bility, precision and trueness of X-ray fluorescence data for mineralogical and/or petrographic pur- poses. Atti della Società Toscana di Scienze Natu- rali - Memorie Serie A, 120, 67-73. Lezzerini M., Tamponi M., Bertoli M. (2014) - Calibration of XRF data on silicate rocks using chemicals as in -house standards. Atti della Società Toscana di Scienze Naturali - Memorie Serie A, 121: 65-70 Doi: 10.24.24/ASTSN.M.2014.16. Lotti B. (1917) - L'antico Lago Tiberino e le ligniti dell'Umbria. Miniera Italiana, 1, 274-275. Lotti B. (1926) - Descrizione geologica dell'Umbria. Memorie per la Descrizione della Carta Geologica d’Italia, 21, 1-320. Manganelli G., Giusti F. (2000) - The gastrocoptids of the Fossil Forest of Dunarobba (central Italy) and a preliminary revision of the European Tertiary nomi- nal species of Albinula and Vertigopsis (Gastropoda Pulmonata: Gastrocoptidae). Bollet- tino della Società Paleontologica Italiana, 39, 55- 82. Manganelli G., Benocci A., Esu D., Giusti F. (2008): Staurodon cianfanellianus n. sp. (Gastropoda Pul- monata), a new nesopupine vertiginid snail from the Middle-Late Pliocene Fossil Forest of Duna- robba (central Italy). Bollettino della Società Pale- ontologica Italiana , 47, 211-214. Manganelli G., Delle Cave L., Giusti F. (1989) - Notulae Malacologicae, XLIV. Strobilopsidae (Gastropoda, Pulmonata), a family new to the Villafranchian land snail fauna of Apenninic Italy. Basteria, 52, 3-13. Manganelli G., Giusti F., Delle Cave L. (1990) - Notulae Malacologicae, XLVII. Lauriinae (Gastropoda: Pul- monata, Orculidae/Pupillidae) from the Villafran- chian of the peninsular Italy. Basteria, 53, 87-103. Martinetto E. (1994): Paleocarpology and the ‘in situ’ ancient plant communities of a few Italian Pliocene fossil forests. In: Matteucci R., Carboni M.G., Pig- natti J.S. (eds), Studies on ecology and paleoecol- ogy of benthic communities. Bollettino della So- cietà Paleontologica Italiana, Spec. Vol., 2, 189- 196 Martinetto E., Scardia G., Varrone D. (2007) - Magneto- biostratigraphy of the Stura di Lanzo Fossil Forest succession (Piedmont, Italy). Rivista Italiana di Paleontologia e Stratigrafia, 113, 109-125. Martinetto E., Bertini A., Basilici G., Baldanza A., Bizzarri R., Cherin M., Gentili S., Pontini M.R. (2014) - The plant record of the Dunarobba and Pietrafitta sites in the Plio-Pleistocene palaeoenvi- ronmental context of central Italy. Alpine and Medi- terranean Quaternary, 27, 29-72. Martinetto E., Momohara A., Bizzarri R., Baldanza A., Delfino M., Esu D., Sardella R. (2017) - Late per- sistence and deterministic extinction of ”humid thermophilous plant taxa of East Asian affin- ity” (HUTEA) in southern Europe. In: François L, Bruch A.A., Utescher T., Spicer R.A., Spicer T. (eds), Reconstructing Cenozoic vegetation from proxy data and models - a NECLIME synthesis. Palaeogeography Palaeoclimatology Palaeoecol- 103 New integrate data from Dunarobba clay deposits Ms. received: August 17, 2018 Final text received: December 1, 2018 104 ogy, 467, 211-231. Doi: 10.1016/j.palaeo.2015.08.015 Martini I.P., Sagri M. (1993) - Tectono-sedimentary characteristics of Late Miocene - Quaternary ex- tensional basins of the North Apennines, Italy. Earth Science Reviews, 34, 197-223. Medici M.C., Gliozzi E. (2008) - Preliminary data on the freshwater ostracods from the Middle-Late Plio- cene Tiberino “Palaeo-ancient lake” (Umbria, Cen- tral Italy). Atti Museo civico di Storia Naturale Tri- este, Suppl., al. 53, 39-48. Meisch C. (2000) - Freshwater Ostracoda of Western and Central Europe. Spektrum Academischer, Heidelberg, Berlin, pp. 522. Melelli L., Barchi M., Brozzetti F., Lupattelli A., Mirabella F., Pazzaglia F., Pucci S., Saccucci L. (2010) - Morphotectonic evolution of High Tiber Valley (Umbria, Italy) related to an active low angle nor- mal fault segmentation. Rendiconti online della Società Geologica Italiana, 11, 629-630. Mirabella F., Bucci F., Santangelo M., Cardinali M., Caielli G., De Franco R., Guzzetti F., Barchi M.R. (2018) - Alluvial fan shifts and stream captures driven by extensional tectonics in central Italy. Journal of the Geological Society. Doi: 10.1144/jgs2017-138 Moore M.D., Reynolds R.C. (1997) - X-ray Diffraction and the Identification and Analysis of Clay Miner- als. Oxford University Press, Oxford, UK, pp. 378. Muller B. (2009) - Impact of the bacterium Pseudomo- nas fluorescens and its genetic derivatives on vermiculite: effects on trace metals contents and clay mineralogical properties. Geoderma, 152(1- 2), 94-103. Doi: 10.1016/j.geoderma.2009.07.025 Muller B. (2014) - Uptake of Barium from Vermiculite by the Bacterium Pseudomonas fluorescens - New Indications for Barium as a Palaeoproxy. The Open Geology Journal, 8, 118-123. Napoleone G., Albianelli A., Azzaroli A., Bertini A., Magi M., Mazzini M. (2003) - Calibration of the Upper Valdarno Basin to the Plio-Pleistocene for correla- ting the Apennine continental sequences. Il Qua- ternario, 16, 131-166. Pontini M.R., Albianelli A., Basilici G., Bertini A., Napoleone G. (2002) - Pollen record and magne- tostratigrapy of the middle-late Pliocene lacustrine sequence in the Tiberino basin (Central Italy). Bollettino della Società Geologica Italiana, special volume 1-2002, 467-472. Pontini R., Bertini A. (2000) - Late Pliocene vegetation and climate in central Italy: high resolution pollen analysis from the Fosso Bianco section (Tiberino basin). Geobios, 33, 519-526. Pucci S., Mirabella F., Pazzaglia F., Barchi M.R, Melelli L., Tuccimei P., Saccucci L. (2014) - Interaction between regional and local tectonic forcing along a complex Quaternary extensional basin: Upper Tiber Valley, Northern Apennines, Italy. Quater- nary Science Reviews, 102, 111-132. Ruiz F., Abad M., Bodergat A.M., Carbonel P., Rodriguez-Lazaro J., Gonzalez-Regalado M.L., Toscano A., Garcia E.X., Prenda J. (2013) - Fresh- water ostracods as environmental tracers. Interna- tional Journal of Environmental Science and Tech- nology, 10, 1115-1128. Spadi M. (2018) - Ostracod assemblages from the Pa- laeolake Tiberino (Piacenzian-Gelasian, central Italy): a synthesis. Alpine and Mediterranean Qua- ternary, 31, 193-197. Spadi M., Gliozzi E., Medici M.C. (2018a) - A Plio- Pleistocene Caspiocypris species flock (Candoninae, Ostracoda) from the Palaeolake Tiberino (Umbria, central Italy). Journal of Syste- matic Palaeontology, 16(5), 417-434. Doi: 10.1080/14772019.2017.1310143 Spadi M., Gliozzi E., Medici M.C. (2018b) - Early Pleisto- cene (Gelasian) non-marine ostracods from the Dunarobba Fossil Forest (Tiberino Basin, Umbria, Central Italy). Palaeontology. ID Pala-01-18-4140- OA.R2 Valentini G., Lombardi S., Bozzano F., Scarascia Mugnozza G. (1997) - Analysis of the geo- environmental conditions as morphological evolu- tion factors of the sand-clay series of the Tiber valley and Dunarobba forest preservation. EUR 17479 EN, Final Report, EUR OP, Luxembourg, pp. 211. Vassio E., Martinetto E., Dolezych M., Van Der Burgh J. (2008) - Wood anatomy of the Glyptostrobus eu- ropaeus “whole-plant” from a Pliocene fossil forest of Italy. Review of Palaeobotany and Palynology, 151, 81–89. Williams A.A.B., Donaldson G. (1980) - Building on ex- pansive soils in South Africa. 4th International Conference on Expansive Soils, Denver, Proceed- ings, vol. 2, 834-840. Williams C.J., Mendell E.K., Murphy J., Court W.M., Johnson A.H., Richter S.L. (2008) - Paleoenviron- mental reconstruction of a Middle Miocene forest from the western Canadian Arctic. Palaeogeogra- phy Palaeoclimatology Palaeoecology, 261, 160- 176. Xu H., Liu B., Wu F. (2010) - Spatial and temporal varia- tions of Rb/Sr ratios of the bulk surface sediments in Lake Qinghai. Geochemical Transactions, 11, 3. geochemicaltransactions.com/content/11/1/3 Baldanza A. et al.