DEPOSITIONAL ARCHITECTURE, FACIES CHARACTER AND GEOCHEMICAL SIGNATURE OF THE TIVOLI TRAVERTINES (PLEISTOCENE, ACQUE ALBULE BASIN, CENTRAL ITALY)
GIOVANNA DELLA PORTA1*, ANDREA CROCI2, MATTIA MARINI1 & SANDOR KELE3
1*Corresponding author. Dipartimento di Scienze della Terra, Universita’ degli Studi di Milano, via Mangiagalli 34, 20133, Milan, Italy. E-mail:
giovanna.dellaporta@unimi.it; mattia.marini@unimi.it
2 Department of Secondary Teacher Education, Manchester Metropolitan University, Brooks Building, 53 Bonsall Street, M15 6GX,
Manche-ster, UK. E-mail: andrea.croci@stu.mmu.ac.uk
3 Institute for Geological and Geochemical Research, Research Centre for Astronomy and Earth Sciences, Hungarian Academy of Sciences, 45
Budaörsi Street, 1112, Budapest, Hungary. E-mail: keles@geochem.hu
To cite this article: Della Porta G., Croci A., Marini M. & Kele S. (2017) - Depositional architecture, facies character and geochemical signature
of the Tivoli travertines (Pleistocene, Acque Albule Basin, Central Italy). Riv. It. Paleontol. Strat., 123(3): 487-540.
Abstract. Facies character, diagenesis, geochemical signature, porosity, permeability, and geometry of the
upper Pleistocene Tivoli travertines were investigated integrating information from six borehole cores, drilled along a 3 km N-S transect, and quarry faces, in order to propose a revised depositional model. Travertines overlie lacustri-ne and alluvial plain marls, siltstolacustri-nes, sandstolacustri-nes and pyroclastic deposits from the Roman volcanic districts. In the northern proximal area, with respect to the inferred hydrothermal vents, travertines accumulated in gently-dipping, decametre-scale shallow pools of low-angle terraced slopes. The intermediate depositional zone, 2 km southward, consisted of smooth and terraced slopes dipping S and E. In the southernmost distal zone, travertine marshes
dominated by coated vegetation and Charophytes interfingered with lacustrine siltstones and fluvial sandstones and
conglomerates. Travertine carbon and oxygen stable isotope data confirm the geothermal origin of the precipitating spring water. The travertine succession is marked by numerous intraclastic/extraclastic wackestone to rudstone beds indicative of non-deposition and erosion during subaerial exposure, due to temporary interruption of the vent activity or deviation of the thermal water flow. These unconformities identify nine superimposed travertine units characterized by aggradation in the proximal zone and southward progradation in the intermediate to distal zones. The wedge geometry of the travertine system reflects the vertical and lateral superimposition of individual fan-shaped units in response to changes in the vent location, shifting through time to lower elevations southward. The complexity of the travertine architecture results from the intermittent activity of the vents, their locations, the topographic gradient, thermal water flow paths and the rates and modes of carbonate precipitation.
Received: July 18, 2017; accepted: October 02, 2017
Keywords: travertine; Pleistocene; Tivoli; facies; depositional architecture; stable isotope geochemistry.
I
ntroductIonTravertines are defined as terrestrial carbon-ates precipitated by water supersaturated with re-spect to calcium carbonate, typically hydrothermal in origin (Pedley 1990; Ford & Pedley 1996; Capez-zuoli et al. 2014). Travertines form through abiotic (CO2 degassing and evaporation of thermal water outflowing from the spring) and biologically medi-ated precipitation processes; they are characterized by common bacteria and cyanophytes and generally lack macrophytes (Chafetz & Folk 1984; Capezzuoli et al. 2014). Thermal water precipitating travertine is suggested to have temperature over 20°C
(Ped-ley 1990) or 30°C (Capezzuoli et al. 2014). There is, however, general agreement that temperature based classifications of terrestrial spring carbonates are problematic to apply to the fossil record of inac-tive systems (Jones & Renaut 2010; Capezzuoli et al. 2014). Spring carbonates can also be differenti-ated in hydrothermal or thermogene travertines vs. ambient temperature calcareous tufa or meteogene travertine based on the sources of the dissolved in-organic carbon (DIC), geothermal vs. karstic me-teoric respectively, as it can be inferred from the carbon stable isotopic signature of the precipitat-ed carbonates (Pentecost & Viles 1994; Pentecost 2005; Capezzuoli et al. 2014).
The term travertine derives from the Italian
which meant “stone of Tibur” (Chafetz & Folk 1984; Faccenna et al. 2008). Tibur was the Roman name for the present-day town of Tivoli, located nearly 20 km east of Rome (Central Italy) along the Aniene River (Chafetz & Folk 1984; Faccenna et al. 2008). The Romans extensively quarried the traver-tines in the area west of Tivoli for construction pur-poses, since the III-II century BC and quarrying is still active at present. Hence, the upper Pleistocene Tivoli travertines represent the deposits from which the general lithology for this type of terrestrial car-bonates derives and are the focus of this investiga-tion.
Fundamental precursor studies on the Tivoli travertines by Chafetz & Folk (1984) highlighted the variety of precipitated carbonate fabrics and the complexity of physico-chemical and microbial processes involved in the precipitation of carbon-ates from thermal springs. Faccenna et al. (2008) proposed a depositional model for the Tivoli trav-ertine system interpreting it as a plateau subdivided in benches separated by discontinuities controlled by fluctuations of the water table, driven by cli-matic oscillations and tectonic activity.
In the last years, there has been a renewed scientific interest on travertines from the academia and industry. This has been driven by the discov-ery of non-marine carbonate hydrocarbon res-ervoirs in the Lower Cretaceous syn-rift and sag phase sedimentary succession of the South Atlan-tic subsurface, despite the published data about the Pre-Salt carbonate reservoirs so far concern only alkaline lake carbonates (Wright 2012; Wright & Barnett 2015; Saller et al. 2016; Sabato Ceraldi & Green 2017). Nevertheless, dome- and eye-shaped hydrothermal vents have been identified through seismics in the Lower Cretaceous rift section of the Campos Basin, offshore Brazil (Alvarenga et al. 2016). The most studied travertine deposits are those of Pleistocene-Holocene age associated also with present-day active geothermal systems such as those in Central Italy, Hungary, Turkey and Yellowstone (Wyoming, USA). Numerous recent studies have focused on various aspects of these travertine depositional systems ranging from the variety of the fabric types, geochemical signature and diagenesis (Pola et al. 2014; Claes et al. 2015; Anzalone et al. 2017; Cook & Chafetz 2017; De Boever et al. 2017; Erthal et al. 2017; Rodríguez-Berriguete et al. 2017; Török et al. 2017),
petro-physical reservoir properties (Chafetz 2013; Ron-chi & Cruciani 2015; Soete et al. 2015; Brogi et al. 2016; De Boever et al. 2016; Claes et al. 2017ab), to the depositional geometry and evolution of the depositional system through time (De Filippis & Billi 2012; De Filippis et al. 2012; De Filippis et al. 2013ab; Croci et al. 2016; Wang et al. 2016; Della Porta et al. 2017) and to the use of traver-tine deposits as recorder of climatic oscillations and tectonic activity (Özkul et al. 2013; Van Noten et al. 2013; Bertini et al. 2014; Gradziński et al. 2014; Toker et al. 2015; Brogi et al. 2017; Frery et al 2017). All these recent studies highlight the vari-ability and complexity of travertine depositional systems from the centimetre scale of the carbon-ate fabric types to the depositional geometry, af-fected by numerous extrinsic and intrinsic control-ling factors ranging from the regional tectonics, climatic regime to the substrate topography and thermal water physico-chemical properties.
This study aims to propose a revised depo-sitional model of the Tivoli travertines linked to a detailed facies characterization of both the trav-ertines and underlying sedimentary and volcanic substrate deposits based on six borehole cores, complemented with diagenetic, stable carbon and oxygen isotope and porosity and permeability data. Detailed facies information has been integrated in a digital 3D model that combines the centimetre-scale stratigraphic information from the cores with the tens to hundreds metre-scale depositional geometry from excavated quarry walls. This model allows getting better insights of the evolution of the travertine geometry through time and the sedi-mentary dynamics that affect the geometry of a travertine depositional system.
G
eoloGIcalsettInGTectonics and stratigraphy
The Acque Albule Basin (AAB) is a fault-controlled, elongated (nearly 7 x 4 km) south-dip-ping depression located to the west of the town of Tivoli, north of the Pleistocene Alban Hills volcanic district (Fig. 1). To the north and east, the AAB is surrounded by mountain ranges, i.e. the Cornicolani, Lucretili, Tiburtini and Prenestini Mts., belonging to the east-verging thrust sheets of the Central Apennine orogen (Cosentino &
Parotto 1986; Corrado et al. 1992) and consisting of Mesozoic-Cenozoic carbonate platform and basinal pelagic successions, and Neogene forede-ep siliciclastic turbidites (Bollati et al. 2011, 2012). From the late Miocene (Tortonian), this sector of the Central Apennine underwent extensional tec-tonics as a result of the opening of the Tyrrhenian Sea as a back-arc basin, driven by the westward subduction of the Adriatic-Ionian plate under-neath the European plate (Malinverno & Ryan 1986; Doglioni 1991; Patacca et al. 1992; Gueguen et al. 1997). Extensional tectonics was associated with lithospheric thinning, NW-SE normal faul-ting, and seismic, volcanic and geothermal activi-ty (Faccenna et al. 2008 and references therein). NW-striking normal faults, which often reactivated former Apennine thrust planes, and NE-striking
transfer faults controlled the development of nu-merous Late Miocene-Quaternary extensional ba-sins in Central Italy (Patacca et al. 1992; Faccenna et al. 1994abc, 2008; Carminati & Doglioni 2012). These fault-controlled basins are filled by discor-dant marine Pliocene claystone and Pliocene-Plei-stocene to Holocene lacustrine and fluvial-alluvial successions (Mancini et al. 2014; Milli et al. 2017), which at their top are intercalated with Middle-Upper Pleistocene to Holocene volcanic deposits from the Sabatini Mts. and the Albani Hills volca-nic complexes (Fig. 1B) of the Roman magmatic province (Funiciello et al. 2003; Bollati et al. 2011, 2012). This volcanic activity, related to the exten-sional regime affecting the Tyrrhenian Sea and the inner sector of the Central Apennine (Fig. 1), was characterized by explosive volcanism yielding Fig. 1 - A) Synthetic geological map of Central Italy (modified after Bigi et al. 1990) with location of active and fossil hydrothermal springs
ex-tracted from Minissale (2004). B) Synthetic geological map of the area between Rome and Tivoli showing the two volcanic districts of the Roman magmatic province (Albani Hills and Sabatini Mts.), the Tivoli upper Pleistocene travertines and the Apennine Mesozoic-Cenozoic carbonate units in the North and East of the Tivoli travertine deposits in the Acque Albule Basin (redrafted after De Rita et al. 1995, 2002; Gaeta et al. 2000; Karner et al. 2001; Faccenna et al. 2008; De Filippis et al. 2013ab).
mafic K- and high K-rich pyroclastic deposits and lavas (Serri et al. 1992, 1993; De Rita et al. 1995, 2002; Gaeta et al. 2000; Marra et al. 2009). The activity of the Sabatini Mts. and Albani Hills vol-canoes (Fig. 1B) started from nearly 800 kyr and 561 kyr, respectively, and intermittently continued to 36 kyr (De Rita et al. 1995; Karner et al. 2001). The Albani Hills volcano recorded significant ac-tivity between 561 and 250 kyr, followed by a pe-riod of relative volcanic dormancy between 250 and 45 kyr and the final Hydromagmatic Phase da-ted at 36 kyr (Karner et al. 2001). During the Late Pleistocene-Holocene, N-striking right-lateral and NE-striking transtensional to normal faults (Figs 1B, 2) controlled the hydrothermal circulation and the deposition of travertines dated at 115-30 kyr (Faccenna et al. 2008).
Hydrogeology
The hydrogeological system of the AAB (Fig. 3) comprises two connected aquifers: 1) a deep, partly confined aquifer, hosted in fractured Me-sozoic-Cenozoic marine limestone strata, and 2) a shallow, unconfined to semi-confined aquifer within the upper Pleistocene travertine deposits (Petitta et al. 2010; Carucci et al. 2012; La Vigna et al. 2013ab, 2016). These two carbonate aquifers are separated by low permeability discontinuous Pliocene marine claystone deposits overlain by Pliocene-Pleistocene continental siliciclastic sandy clayey sequences and Pleistocene volcanic deposits, which constitute ei-ther an aquitard or aquiclude (Di Salvo et al. 2013; La Vigna et al. 2013a). The deep confined aquifer is influenced by deeply sourced, high salinity, Ca-Mg-HCO3-SO4 and CO2 rich fluids (Carucci et al.
Fig. 2 - A) Synthetic geological map showing the location of the Acque Albule Ba-sin bounded by strike-slip and normal faults and the travertine deposits located North of the Aniene River, with the Tiburtini and Lu-cretili Mts. on the East and the Cornicolani Mts. in the North made of Mesozoic-Cenozoic marine carbonate successions (redrafted after Faccenna et al. 2008; Bru-netti et al. 2013; De Filippis et al. 2013ab). The area of active quarrying west of the town of Tivoli is marked by a blue line. The locations of the Statoil drilled boreholes object of this study (S1 to S7) are reported. B) Google Earth Pro image showing the travertine quarried area, identified by the white co-lour of the travertines, with the location of the six bo-reholes drilled by Statoil (S1 to S7) and the present-day active thermal springs.
2012) as well as by geothermal heat related to the Albani Hills magmatism (Minissale et al. 2002; Bil-li et al. 2006; Faccenna et al. 2008; Di Salvo et al. 2013). The Ca-Mg-HCO3 composition is attributed to decarbonation of the Mesozoic-Cenozoic carbo-nates, whereas the high concentration of sulphate is caused by leaching of the Triassic evaporitic Burano Formation (Minissale et al. 2002; Minissale 2004). The travertine aquifer is presently unconfined in the quarried area (Fig. 2), whereas it is semi-confined where it is still overlain by soils and alluvial-lacustri-ne deposits post-dating travertialluvial-lacustri-ne deposition (Ca-rucci et al. 2012; Di Salvo et al. 2013; La Vigna et al. 2013ab). Groundwater is directly fed by rainfall and seepage from the surrounding carbonate rid-ges, being the travertines the terminal sector of the drainage path feeding into the Aniene River (Fig. 3; Petitta et al. 2010; Carucci et al. 2012; Di Sal-vo et al. 2013; La Vigna et al. 2013ab). The deep carbonate and shallow travertine aquifers are con-nected through tectonic discontinuities that allow the upwelling of deep, high-temperature, CO2-rich mineralized water and the mixing with shallow, am-bient temperature, meteoric water derived from the
carbonate ridges and superficial drainage (Minissale et al. 2002; Carucci et al. 2012; De Filippis et al. 2013a). This mixing can explain the presence of wa-ter with high mineralization but low temperatures feeding the travertine aquifer (Petitta et al. 2010).
The current active thermal springs are located along the NNW-SSE shear zone in the west of the travertine quarry area (Pentecost & Tortora 1989; Minissale et al. 2002) and in the southern area (Figs 2A, B). The Colonnelle and Regina springs dischar-ge 2-3 m3 s-1 of water (Carucci et al. 2012; La Vigna et al. 2013a) with temperature around 23°C, pH 6.0-6.2, Ca 520 mg/kg, Mg 116 mg/kg, HCO3- 1488 mg/kg, SO42- 734 mg/kg, and pCO
2 0.6 atm (Pente-cost & Tortora 1989; Minissale et al. 2002; Minissale 2004; Carucci et al. 2012; La Vigna et al. 2013a; Di Salvo et al. 2013). The springs are generally super-saturated with respect to calcium carbonate, which explains the travertine deposition (Minissale et al. 2002; Carucci et al. 2012).
Previous studies on Tivoli travertines
Chafetz and Folk (1984) interpreted the Tivo-li travertines as lake-fill deposits, made of horizon-Fig. 3 - Hydrogeological model for the travertine Tivoli plain (redrafted after Carucci et al. 2012; Brunetti et al. 2013; La Vigna et al. 2013ab,
2016). Thermal water has low temperature because of the mixing between the deep geothermal fluids from the confined carbonate aquifer and the shallow travertine aquifer fed by superficial drainage.
tally stratified and laterally extensive accumulations, vertically affected by paleokarst due to periodic lake drainage. The travertines were primarily compo-sed of vertically stacked centimetre-thick layers of shrubs, laterally interrupted by zones composed of calcite ray crystals, intraclasts, and pisoids indicative of spring deposits on the lake bottom.
Faccenna et al. (2008) reconstructed a deposi-tional model of the Tivoli travertine unit, interpre-ting it as a tabular plateau made of sub-horizontal benches, separated by erosional surfaces, with a progradational pattern and southward steepening of strata. The plateau covers an area nearly 20 km2 wide with average thickness of 40-50 m, for a total volume of 1 km3 (Faccenna et al. 2008; De Filip-pis et al. 2013ab). The depocentre, with maximum thickness up to 80-90 m, coincided with a main N-striking fault and the associated emergences of thermal water (Faccenna et al. 2008). The five iden-tified erosional surfaces and the alternating stages of bench deposition and erosion were suggested to be controlled by episodic fluctuations of the water table influenced by Pleistocene palaeoclimate, fault-related deformation and nearby volcanic activity (Faccenna et al. 2008; De Filippis et al. 2013a). The uppermost and youngest erosional surface separa-tes the travertine deposits from the overlying “separa-testi- “testi-na” unit, consisting of 3-4 m thick, poorly lithified travertine capping most of the Tivoli plateau and dated at 29 ± 4 kyr (Faccenna et al. 2008; De Fi-lippis et al. 2013b). The presence of a fissure ridge structure was identified by De Filippis et al. (2013a) in the NW corner of the Tivoli travertine deposits. The Colle Fiorito fissure ridge was 2 km long and nearly 15 m high and must have accumulated when the volumetric deposition rate reached its climax for the abundance of fluid discharge and the rise of water table (De Filippis et al. 2013a).
Anzalone et al. (2017) investigated a nearly 30 m thick borehole core drilled in the NW Tivoli quarry area and linked the core stratigraphy to the depositional geometry visible in nearby saw-cut quarry walls. They interpreted the Tivoli traverti-nes as accumulated in shallow lake to gentle slope environments with deposition marked by nume-rous erosional/non depositional discontinuities of different orders and magnitudes. Using sedimen-tological, stratigraphic and geochemical data they developed a cyclostratigraphic model of the drilled travertine core identifying high-frequency cycles
driven by water table fluctuations controlled by millennial scale climatic cycles, medium term sub-Milankovitch and precession-driven sub-Milankovitch climatic fluctuations.
The depositional model proposed for the Ti-voli travertines by Erthal et al. (2017) consists of an extensive water-logged flat setting varying laterally into a slope system. These authors investigated in detail the travertine facies focussing on the shrub fabrics and identified six different types of shrub morphologies, linking them to the depositional conditions, water flow hydrodynamic, CO2 degas-sing rate, evaporation and influence of microbial-ly mediated precipitation. Slow thermal water flow favours greater contribution of microbially media-ted processes to carbonate precipitation that results in arborescent, arbustiform and pustular shrub morphologies, more fragile and made of clotted pe-loidal micrite aggregates (Erthal et al. 2017).
MaterIalandMethods
The investigation of the stratigraphy and facies character of the Tivoli travertines and of the underlying volcanic and sedimentary substrate rocks is based on six research borehole cores (labelled from S1 to S7), drilled by Statoil ASA in 2010, within the active travertine quarry area, west of Tivoli (Fig. 2). The boreholes are located along a nearly 3100 m long transect, extending from well S1 in the north to well S7 in the south. Elevations vary from 68-57 m a.s.l. in the north (wells S1, S2, S3) to nearly 34 m a.s.l. in the south (well S7), close to the Aniene River. Because borehole S5 had limited recovery, corresponding to the uppermost stratigraphic interval of borehole S6, description of core S5 is combined with core S6. The total thick-ness of the six investigated borehole cores ranges from 35 m to 50 m; each core includes 19-43 m of travertines. At each location, the borehole was drilled on the topmost and youngest suitable quarry bench of travertine excavation; therefore the drilled cores do not include the top 10-20 m thick succession overlying the travertines made of the uppermost Pleistocene-Holocene alluvial and fluvial-lacustrine deposits and the so called “testina” unit (cf. Faccenna et al. 2008). Cores were analysed at the millimetre scale through mesoscale core description complemented with petrographic analysis of thin sections (160 thin sections from the travertines; 83 thin sections from the terrigenous and volcanic deposits).
Cathodoluminescence was performed on 12 thin sections with a luminoscope CITL Cambridge Image Technology Limited, Cambridge, UK (model MK 5-2 operating system at 10-14 kV with a beam current between 300-600 µA, and vacuum gauge 50-70 mil-litor) at the Earth Sciences Department, Milan University. Scanning Electron Microscope (SEM) analyses were performed on polished slabs, thin sections and freshly broken surfaces, gold coated, with a Cambridge Stereoscan 360, operating at 20 kV with working dis-tance of 15 mm at the Earth Sciences Department, Milan Univer-sity. Mineralogy of the travertine facies was investigated on 8 powder samples with X-ray powder diffraction (XRD) analytical technique, by means of a X-RAY Powder Diffractometer Philips X’Pert MPD with high temperature chamber at the laboratory of the University of
Milan. For all samples, qualitative analyses were made with Panalytical X’Pert HighScore software to identify the crystalline phases.
Thirty-five plugs (1 inch in diameter) sampled from core S1 were measured for porosity and permeability with a Helium gas expansion porosimeter at the Weatherford Labs by Statoil ASA (Bergen, Norway). In addition to plug porosity-permeability mea-surements, the porosity of each distinguished travertine facies was semi-quantitatively estimated through image analysis by determining the percentage of pore space per area from thin section
photomicro-graphs (areas of nearly 2–4 cm2) with the software ImageJ.
Stable isotope (oxygen and carbon) analyses on 208 ate powder samples were determined using an automated carbon-ate preparation device (Gasbench II) and a Thermo Fisher Scientific Delta Plus XP continuous flow mass spectrometer at the Institute for Geological and Geochemical Research, Hungarian Academy of Sci-ences, Budapest, Hungary. Carbonate powders were extracted with a dental microdrill avoiding the mixing of carbonate components and were reacted with 100% phosphoric acid at 70°C. Standardization was conducted using laboratory calcite standards calibrated against the NBS-19 standard. The carbon and oxygen isotope compositions are expressed in the conventional delta notation against the international
standard V-PDB (for δ13C and δ18O). Reproducibility for both C and
O isotope analyses is better than ± 0.1 ‰.
Facies types, their vertical stacking and key stratigraphic boundaries identified through core logging were integrated with ob-servations of depositional geometries and stratigraphic architectures on saw-cut quarry faces, adjacent to the locations where the borehole cores were drilled, complemented with high resolution georeferenced photographs of the quarry walls provided by Statoil ASA (Durand 2011) and satellite imagery from Google Earth Pro®. The outcrop spatial information and vertical facies stacking from core data were used to develop a 3D digital model of the travertine deposit using the software Petrel 2015® (Schlumberger). Points and polylines be-longing to key stratigraphic boundaries digitized from georeferenced, orthorectified quarry wall photographs at Statoil ASA were used as input data for surface interpolation, which was accomplished by means of either a convergent interpolation algorithm or a functional surface algorithm depending on the nature and expected geometry of each stratigraphic surface. The 3D Petrel digital model allowed visualizing the architecture of the travertine units bounded by the interpolated surfaces and populated with the facies types extracted from core description. This digital model of the Tivoli travertines provided better insights on the evolution of the depositional system architecture through time and space.
r
esultsTravertine, terrigenous and volcanic facies
The analysed cores (Fig. 4) are composed of fourteen travertine facies (T1-T14) made of low Mg calcite, vertically stacked at the centimetre to decimetre-scale. Travertine facies types were distin-guished on the basis of texture, precipitated carbon-ate fabric and components and classified according to the terminology proposed by Della Porta (2015), Croci et al. (2016) and Della Porta et al. (2017). The description of the travertine facies (Figs 5-9A-D) is summarised in Table 1. The post-travertine carbon-ate precipitcarbon-ates and sediment filling that
accumulat-ed within primary voids and secondary dissolution vugs to metre-scale karstic caves (C1, C2) are shown in Figures 9E-K. Additional travertine diagenetic features are reported in Figure 10. The terrigenous and volcanic deposits, mostly underlying the main travertine deposits (Fig. 4), were distinguished in ten facies (F1-F10) on the basis of lithology and com-position (siliciclastic, carbonate, volcanic and volca-niclastic), dominant grain size, sedimentary struc-tures and degree of lithification (Table 2; Figs 11, 12). The terrigenous deposits include black colour mudstone (F1), marl (F5), fine sand, silt and clay (F6) with rare layers with carbonate encrusted Cha-rophytes algae and plant stems (facies T14) and loose
sand, sandstone and conglomerate (F7 to F10). The volcanic deposits include facies F2, F3 and F4 rep-resenting pyroclastites with feldspars and feldspa-toids, and volcanic ashes.
Post-travertine sediment filling and dia-genetic features
The travertine primary and secondary vuggy porosity display various types of sediment fillings ranging from light to dark grey colour, silt- to mud-grade mixed detrital calcite crystals, travertine intra-clasts, terrigenous sediment and yellowish ostraco-de wackestone/packstone (Fig. 9K), laminated or structureless, locally associated with millimetre-thick calcite rafts (C1 in Figure 4). The thickness of these deposits varies from 2 cm up to 20 cm. A particu-lar kind of vug filling deposit (C2 in Figure 4; Figs 9E-J) consists of dark to light brown concentric laminated and vertically oriented columnar carbo-nate cementstone structures adjacent to extraclastic packstone to grainstone and rudstone and pisoidal packstone/grainstone to rudstone. These deposits only occur in core S7 in a nearly 5 m thick interval at core depths of 13.50-18.50 m (Fig. 4), comprised between the underlying T9 facies and the overlying F6 siltstone. The columnar decimetre-size structu-res (Fig. 9G) are formed by discontinuous sparitic laminae with crystal fans with undulose extinction alternating with micrite/microsparite crusts (10-100 µm thick), embedding tufts of possible calcified fi-lamentous cyanobacteria (Figs 9E-F). Some of the micrite laminae show some weak to bright orange luminescence in cathodoluminescence analysis. The extraclastic packstone to rudstone includes angular clasts, up to 2 cm in size, deriving from the Meso-zoic-Cenozoic successions cropping out in the
rid-ges surrounding the Tivoli area (Figs 1, 2). Pisoids forming packstone/grainstone to rudstone are 1-20 mm in diameter and consist of concentric micrite laminae (10-200 µm thick) coating nuclei made of extraclasts (Neogene marine carbonates, plankto-nic and benthic foraminifer wackestone/packsto-ne, detrital quartz and chert) and travertine intra-clasts (Figs 9H-J). The micrite/microsparite matrix around pisoids and extraclasts includes ostracodes, phosphatic and siliciclastic detrital grains. The pisoi-dal grainstone is cemented by limpid equant calcite sparite with drusy mosaics preceded by meniscus micritic crusts.
The diagenetic features observed in the tra-vertine facies types (Table 1) appear to be similar and can be subdivided into: cement types, recrystal-lization features, dissolution and pedogenetic featu-res. Cement types lining the pore walls (Figs 10A-G) include, in order of occurrence: a) micritic to sparitic pendant cement that, when present, is the first cement phase observed; b) equant microspa-rite to spamicrospa-rite forming blocky mosaics; c) prismatic calcite cement with scalenohedral terminations for-ming more than 1 mm long crystals. In some cases primary voids show geopetal ostracode wackestone infill (C1) that postdates the pendant cement and precedes or postdates the prismatic cement preci-pitation (Fig. 10F). Cathodoluminescence analysis shows that most of the precipitated travertines and cement types are non-luminescent. Neverthe-less, the crystalline dendrites (T1), clotted peloi-dal micrite dendrite (T2) and radial coated grains (T6) might show a weak luminescence (Figs 10H-K). The prismatic scalenohedral and blocky calci-te cements show some luminescent growth phases (Figs 10J, K). Evidence of recrystallization of the turbid lozenge shaped crystals (facies T1 and T6) in limpid blocky sparite and of the clotted peloidal micrite (facies T2) into microsparite to sparite are also observed (Figs 10L-M). Dissolution features range from sub-millimetre size vugs to metre-size caves and vertically oriented conduits. Dissolution affected also the clotted peloidal micrite framework and the prismatic calcite cement lining the primary pores (Fig. 10N). Rarely, in facies T9 and T11, pyri-te crystals occur in the inpyri-tercrystalline space within the pore filling blocky cement mosaic. Quarry wall observations indicate that facies T13 detrital traver-tine packstone/rudstone is often associated with decimetre to metre-thick terrigenous clay layers
and paleosols. These deposits were not identified in cores due to lack of core recovery during drilling through clay beds. However, pedogenetic features such as circumgranular cracks and alveolar texture were observed in thin sections (Fig. 10O).
Travertine SEM analysis
SEM analysis of travertine samples shows that the limpid equant microsparite to sparite (20-100 µm) cement lining the pores consists of euhe-dral rhomboheeuhe-dral calcite crystals growing on the micritic primary precipitates (Fig. 13A). The turbid lozenge shaped crystals forming the T1 crystalline dendrites show a nanometre scale internal moul-dic porosity with moulds that can show squared, rounded or dumb-bell cross sections (Fig. 13B). Some broken crystals show that they are internal-ly composed of sub-micron scale clots of micrite (Fig. 13C). This internal structure of the dendrite crystals might explain their turbid micritic-like appe-arance under the polarized light microscope. SEM investigation of clotted peloidal micrite dendrites shows that they consist of clots (5-60 µm in dia-meter) made of submicron-scale calcite from which columnar bladed microsparite crystals (20-100 µm long) depart radially (Fig. 13D). The central micrite clots can consist of either nanometre-scale (0.1-1 µm) anhedral sub-spherical calcite structures (Fig. 13E) or euhedral dodecahedral carbonate crystals (Fig. 13F). T2 and T6 facies show the presence of organic membrane embedding micron-size clots of calcite and coccoid, filamentous and dumb-bell micron-size bacterioform structures (Figs 13G, H).
Travertine porosity and permeability
Mean porosity calculated through image analysis on thin section areas ranges between 2 and 15 %, with the highest values estimated for facies T4 to T8 and T13 (Table 1). Porosity and perme-ability measured on plugs from core S1 show va-riations of both parameters with respect to facies types (Fig. 14). Facies T2, T4, T8, T10 and T13 have the highest porosity (10-25 %). Permeability vari-es from 0.005 to 7038 mD and samplvari-es with the highest permeability (> 100 mD) belong to facies T2, T8, T10 and T14. Facies T2 and T8 show 5-15 % porosity and variable permeability, from 0.1 to 1000 mD, which might even decrease at increasing porosity. Facies T13 shows a positive linear corre-lation between permeability and porosity, which
Fig
. 4 - Stratig
raph
ic logs of
the six
analysed cores along
a transect oriented nearly N-S with ex ce ption of S1-S2 that are along a NE to SW transect. The upper blac k line re presents the topog raphic surface as extracted from the topog raphic map of the Ti voli area. The dashed blue line re presents the present-da y w ater table de pth as identified during core drilling . Detailed facies descriptions are pro vided in T able 1 and 2.
Fa cie s C ore d es cri pti on Pet ro gra ph ic an aly sis Thi ck ne ss a nd sp atia l d istr ib utio n D iag en et ic f ea tu res Po ro sity (% ar ea in t hi n s ect ion) D epo siti on al pr oce sse s a nd env iro nm ent T1 C ry sta llin e de ndr ite cem en tsto ne (Fi gs 5A -F ) W hit e d en se, m ms to fe w c ms thick la yer s o f b ran ch ing de nd riti c cal cite cr yst al s, a djace nt to e ach ot he r,g row ing u pw ard . Su b-p ara llel sta ck ed la yer s f rom ho riz on tal t o in clin ed u p to 4 5° (sm oo th s lop e); c m to dm -size rou nd ed c on ve x-o utw ard m or ph olo gie s (p oo l r im so f ter rac ed slo pe ); few c m swi de , mm sh igh ,st ep pe d m or ph olo gy (m icro -te rra ces ). Crys tals m ay sh ow su b-p ara lle l m m -th ick gro w th la m in ae. Cem en tsto ne o f l ozen ge -sh ap ed tu rbid , in clu sio n-ri ch c ry sta ls (1 00 -60 0 µ m lo ng, 40 -60 µ m w ide ) d ep art ing fro m a ce ntra l elo ng at ed cr yst al act ing as a stal k (un ifor m ex tin ctio n un der c ros sed p ola rizer s). W he n cen tral stal k cr yst al ab sen t, p airs o f loze ng e cr yst al s fo rm ve rtical ly s tack ed “v ” (un ifo rm or und ulos e e xt inc tion ). Th e da rk t urb id ap pea ran ce r es em bles c rys tals co nsi stin g o f a gg rega ted m icr ite a cq uir ing eu he dral ra tio nal cr yst al face s. Tra nsi tio ns of T 1 d en dri tes fro m /in to T2 m ic riti c d en drite . Be ds: 5 m m to 20 cm th ic k ( S5 -S6 ), gen er ally 2 -5 cm ; m ax im um th ick ne ss: 20 c m (S 5-S6 ). T1 in a ll c ores fro m 1.5 to 9% (S4 ).T1 ass oc iat ed lat era lly w ith T6 o r T2 ; ve rtic ally alt ern atin g w ith T5 , T3 , T6 , T2 an d T1 3. T1 d end rite s s urr oun de d b y 1 ) clou dy m icr osp arit e ( 10 -20 µm ); 2) l im pid e qu an t m ic ros pa rite to sp arite (2 0-1 00 µm ); 3 ) fo llo w ed b y p ris m atic sca len oh ed ral c em en t,s yn tax ial w ith sa m e ex tin ct ion p atter n as T1 cr yst al . T1 cr yst al s p ar tly rep lac ed b y l im pi d c alc ite sp ar (m et eo ric d iag en es is). Me te oric alte ratio n m igh t p rod uc eb oth m ic ritiz atio n a nd sp aritiz atio n. m in: 0 .3% ; m ax : 10 % ; a ver ag e: 3%. In ter -den dri te, bet w een a djac en t cr yst al s. Ph ys ico -ch em ica l p rec ipit at ion du e t o rap id C O 2 de ga ssin g in h igh -en er gy tu rbu len t, f ast f low in g th erm al w ate r se ttin gs, clos e t o t he ve nt a nd o n inc lin ed steep ly d ipp ing su rfac es of slo pes , r im s a nd w alls o f p ool s i n ter rac ed slo pes a nd m icro -ter rac ed inc lin ed su rfac es (cf ., Fol k e t a l. 19 85 ; Jo ne s & Re na ut 1 99 5, 20 10 ; Gu o & Rid in g 1 99 8; C ha fetz & G uid ry 1 99 9; D ella P orta 201 5). D en dri tes fo rm due to h igh dis eq ui lib riu m d uri ng ra pid pre cip ita tio n wi th im pu ritie s a nd c rys ta l de fec ts p rom ot ing cr yst al s pli ttin g (Jo ne s & R en au t1 99 5). T2 C lotte d pe lo id al m icri te de ndr ite bo unds tone (Fi gs 5G -K ) W hite to lig ht tan , b ran ch ing bu sh -lik e d en dri tic stru ctu res (fe w mms to c ms th ick , 1 -15 mm w ide ), u pri gh t o rie nte d a nd ad jac en t to ea ch o ther in rows ve rtic ally sta ck ed , sep arat ed b y mm -th ick den se m icr itic o r m icro sp ari tic c ru sts (T 3). Bed s ho riz on tal, pa rall el o r u nd ula ted at th e cm -scal e; T2 can fo rm step ped m icr o-t erra ces or c on ve x up w ard m ou nd ed m or ph ol og ies . Bo un dst on e o fv er tic ally su per im po sed T2 lay er s; den dri tes m ad e of i rre gu lar m icri te clo ts (10 -20 µm in d ia m ete r, r are ly u p to 50 µ m ) e m be dd ed in m icr ite a nd /or c lou dy eq ua nt m icro sp arit e (5 -50 µ m ) f or m ing bra nc he s ( 10 0s µ m s t o m m s l on g) de pa rtin g fro m a ba sal su bst ra te o r fro m a n elo nga ted u pri ght c en tra l st em . T 2 l ay er s are s ep ara ted b y s ub -m m le io lit ic o r pe lo ida l m ic rite la m ina e ( T3 ).R are ost rac od es, g ast rop od s, m m -size b iva lves an d p oss ible in sect lar val cas es. S par se co ated g as b ub bles an d m icr ite l am inae. So m e T 2 d en dri tes tr an sit io na l in to T 1 (b ra nc hes w ith g eo m et ric lo zen ge s hap es w ith sh arp b ou nd aries m ad e o f d en se m icri te). Be ds: su b-h ori zo nta l or u nd ulat ed , 0. 2-5 cm th ic k; cm -dm size c on ve x-upw ard m ou nd ed str uc tur es; m ax im um th ick ne ss: 10 -20 c m ( S2 ). T2 in a ll co res from 0 .5 to 1 0% (S2 ).T2 ver tical ly a lte rna tin g w ith m m -th ick T 3, T2 as so ci at ed w ith T4 , T6 a nd T7 . T2 d end rite s s urr oun de d b y: 1 ) eq ua nt m icr osp arite to sp arite (2 0-1 50 µm in siz e); 2) pri sm at ic s cal en oh ed ral cr yst al s (10 0s µm to 1 mm) ; 3 ) lo cally prec ed ed b y p en dan t v ad ose cem en t an d m ic rite c rus ts. Irre gu lar sp arite su bst itu tin g t he m icr ite for m ing th e b ran ch es (sp aritiz atio n d riv en b y m et eo ric d iag en es is). Mic rite dis so lut ion du e t o m ete or ic dia gen es is prod uc es sec ond ary m atr ix p oro sity . mi n: 1 .5% ; m ax : 15 % ; a ver ag e: 5%. In ter -den drite , u p to a fe w m m s i n size ; m icr op or os ity . Co ated bu bb le, in te r-lam inae po ros ity (1 -2 mm ).V ertic ally sta ck ed T2 lay er s po orl y c on nec ted bec au se o f t igh t T3 lam inae in bet w een . T2 den dri tes (sh rub s se nsu Ch afet z & Fo lk 1 98 4) m igh t b e d ue t o biol og ica lly ind uc ed /in flu en ced p rec ipit atio n ass oc ia te d w ith m ic rob ia l m ats o f cy an ob act eria and su lph ide ox idi zin g ba ct er ia (C ha fet z & Fol k 1 98 4; Gu o & Rid ing 1 99 4; Gu o e t a l.1 99 6; G uo & Rid in g 199 8; C ha fetz & G uid ry 1 99 9; Ert ha l e t a l. 20 17 ), ev en th ou gh so m e au tho rs s ug ge st a n ab ioti c o rig in (P en tec ost 199 0; P en tec os t & C ole tta 20 07 ).T 2 d en dri tes ty pic al of low to m od er ate en er gy fla t p on ds an d s ub -ho riz on tal po ol s of te rra ced slo pe sys tem s w her e t he f low a nd tu rbu len ce of th erm al w ate r a re l im ite d a nd m icr ob ial m at s d rap e the p oo l s ub strat e. T3 M ic rit e-m icro sp ari te cru st bo unds tone (Fi gs 6A -B , D) W hit e d en se l am ina e ( 0.5 -1.5 m m th ic k) f orm in g m m s to cm s th ick un du lated la yer s; T3 m ight fo rm step ped m icr oter rac ed m or ph olo gie s, s ub -ho riz on tal or slig htl y in clin ed . Lam inae m ad e of d en se l eio litic an d clo tte d p elo ida l m ic rit e or m icr osp ar ite . La m in atio n d ue to th e a lte rna tio n o f m icr itic l ay ers , d ar k in co lou r, an d lig ht co lou r m icr osp ar itic lay er s. Bed s: 1 -5 m m u p to 5-11 c m th ic k ( S4 ). T3 in a ll c ores fro m 0.5 to 2 .5% (S4 ). M os t a bun da nt in co re S 4, s im ila rly to T1 a nd T5 . T3 ve rtic ally alt er na tes w ith T1 a nd T2 . Mi crite re cry sta lliz atio n i n m ic ros pa rite . m in : 0 .1% ; m ax : 5%; av er ag e: 2 %. Ra re i nter -lam inae po ros ity . Po ssib le bi olo gic ally in du ced /in flu en ced prec ip ita tio n o f m ic rite /m icr osp arite cru sts in as so ci at ion w ith m icr ob ial ma ts. I n mi cro -te rra ced p ool s a nd rim s of lo w -an gle terra ced sy ste m san d sm oot h s lop es . T 3 d rap es T 2 an d T 1 an d m ark s th e i nte rru pt ion o f T 2 a nd T 1 de nd rit e g row th . Tab
. 1 - Description and inter
pretation of
tra
ver
tine facies
Fa ci es C or e d es cr ip tio n Pet ro gra ph ic an al ys is Thi ck ne ss a nd sp at ia l d ist rib ut io n D ia gen et ic f ea tu res Po ro sit y (% a re a in t hi n s ec tio n) D epo sit io na l pr oc es se s a nd env ir onm ent T4 Coa te d g as bubbl e bo unds to ne to wack es to ne (Fi gs 6A -B , D -E ) Li gh t ta n to g rey , c ar bo nat e film s iso la tin g ro un de d h ol lo w p or e sp ac es w ith sp her ic al to el on ga ted sh ap e ( fro m 0 .5 mm to 5 mm in di am et er ), at tri but ed to co at in gs ar ou nd g as b ub bl es ; ve rti ca lly sta ck ed a nd w el ded to get her bu bb le s u p t o 2-3 cm in h ei ght . Co at ed b ub bl es c an o cc ur sp ar se in ca rb on ate m ud m atr ix (m ic rite an d m ic ro sp ari te ) o r fo rm fra m ew or ks of w el de d bu bb le s. Bou nd ston e t o w ack es to ne w ith sp ar se bu bb le sco at ed b y th in l ay er s ( 5-10 0 µ m ) m ad e o f le io lit ic o r p elo id al m ic rit e a nd m ic ro sp ar ite. O n th e o ut er su rfac e, m ic ron -s ize T 2 m ic rit ic o r T 1 c ry sta lli ne den dr ites ca n d ev el op . T4 b ou nd sto ne in cl ud es sp ar se p el oi ds, cl ot s o f cl ot te d m icr ite , i nt racl as ts, r af ts an d os tra co de s. Be ds : 5 m m to 5 c m ; m ax im um th ic kn es s: 25 c m (S1 ). T 4 in a ll co res fr om 0 .5 to 9% ( S1 ). T4 as so ci at ed w ith T 2, T6 , T8 , T1 , T5 . Sp ar se b ub bl es in T1 0. T4 -T1 0 ove rly in g t he un co nf or m ity T 13 w ith T 8. T 4 a t r im nu cl ei d rap ed by T 1. In te rn al ly a nd e xt er na lly li ne d by c em en t: 1 ) eq uan t m ic ro sp ar ite to sp ar ite (2 0-20 0 µm in siz e) , 2 ) p ris m at ic sc al en oh ed ra l c em en t( 10 0-25 0 µm ). m in : 3 % ; m ax : 35 % ; a ver ag e: 15 %. In tra -b ub bl e, in trap ar ticl e po ro sit y s ev er al mm -s ize, is ol at ed an d u nc on ne ct ed . G as b ub bl es p ro du ce d by w at er tu rb ul en ce o r b y m ic ro bi al m et ab ol ic act iv ity . P hy sico -c he m ic al p re cip ita tio n (C O2 de ga ss in g) or m ic ro bia l m ed ia tio n (C ha fet z et a l. 1 99 1; Fo lk & Ch af et z 19 84 ; G uo & R idi ng 19 98 ; G an di n & Ca pe zz uol i 2 01 4; D el la P or ta 2 01 5; Er th al e t a l. 20 17 )in st ag na nt lo w -en er gy p on ds a nd p oo ls o f t er ra ce d slop e s ys te m s, n uc le i of p oo l r im s of ter ra ced sl op es. T5 Lam ina te d bo unds to ne (Fi gs 6C , F ) W hi te d en se, w av y l am in ae f ro m ten sµm s to 2 m m th ic k. L am in ae ar e u nd ul at ed , f ro m sub -ho riz on ta l t o sub -v er tical ; lam in ae d ev el op c on ve x-up w ar d m or ph ol og ie s, w he n la te ra lly w el ded to get her . S om e o f t he la m in ae fo rm irre gu la r p or ou s fra m ew ork . Bou nd ston e of la m in ae (2 0 µ m to 1 -2 mm th ic k, o n a ve ra ge 1 00 -3 00 µ m ) m ad e o f le io lit ic a nd c lo tte d p el oid al m ic rite as so cia te d w ith m ic ro sp ar ite (5 -2 0 µm cr yst al si ze ). Cl ots o f p el oid al m ic rite , r ar e os ta cod es a nd b iv al ve s. G re y/ gre en in te rn al se dim en t of o str ac od w ac ke ston e in in te rla m in ae p or os ity . Be ds : 2 m m to 13 cm th ic k, av er ag e 4 cm; ma xi mu m th ick nes s: t en s o f cm (S 4) .T 5 in a ll co res fr om 0 .1 to 7% ( S4 ). T5 ve rtic al ly a lte rn ate s w ith T1 , T3 , T4 . T5 ad ja ce nt to rim s o r slo pe s w ith T 1. Ce m en t ty pe s l in in g th e in te r-la m ina e p or os ity a nd e xt er na lly th e lam in ae ar e: 1) li m pi d eq ua nt m os ai cs o f b lo ck y m ic ro sp ar ite to sp ar ite (2 0-10 0 µm ), 2 ) p ris m at ic scal en oh ed ral cem en t g ro w in g p er pen di cu la r to th e la m in ae o rie nta tio n ( up to 0 .5 m m th ic k) ; 3 ) lo ca lly va do se b row n p end an t s pa rit e an d m icr ite cr us ts. m in : 3 % ; m ax : 15 % ; a ve ra ge : 7%. Le ns -s hap ed in te r-l am in ae po ro sit y up to 3 cm wi de .I n S 4 in te r-l am in ae sp ac e i nf ill ed b y sili cic la sti c s ilt an d os tra co de w ac kes to ne. A lso la be lle d a s s tro m at ol ite s ( Rai ney & Jo nes 20 09 ) a nd as so ci at ed w ith m ic ro bia l b io fil m s( G an di n & Cap ezzu ol i 2 01 4) . T 5 o cc ur s b ot h i n su b-ho riz on ta l p ond s, p ool s of te rra ce d slo pe sy ste m s a nd o n in clin ed slo pe su rface s as so ci at ed w ith fas t-f lo w in g T 1 cr yst al lin e c ru sts. T6 R ad ial co at ed g ra in gr ai ns to ne (Fi gs 6G -K ) W hite to lig ht tan mm -t o c m -th ick la yer s o f s ub -ro un de d co at ed g ra in s g ra in sto ne ;i n s om e cas es gr ai ns w el ded to get her to fo rm a fra m ew ork . G rai ns sh ow ra di al c oa tin gs a rou nd a c ent ra l nu cl eu s ( 40 0-70 0 µ m to 2 -8 mm in si ze) m ad e o f m ic rit e o r ho llo w ; w ell s or te d, can sh ow bo th fi ni ng an d co ar se ni ng up w ar d t re nd s( m or e c om m on ). La yer s o ften su b-ho riz on ta l t o lo w -a ng le di pp ing be ds . G ra in sto ne (m ic ro sp ar ite to sp ar ite ), ra re pa ck sto ne ( m ic ro sp ar ite o r c lo tted m ic rit e) , t o bo un ds ton e ( gr ai ns w el de d to get her ). Coa tin gs m ad e o f t ur bi d, lo ze ng e-sh ap ed c ry sta ls ( 15 0 µm -1 mm lo ng ), s im ila r to T 1 c ry sta ls w ith u nif or m or u nd ul os e ex tin ct io n, a rran ged ra di al ly ar ou nd a n uc le us m ad e of c lot te d p el oi da l m ic rite ,i nt racl as t ( co at ed b ub bl es , r af ts) , os tra co de fr ag m en t, or h ol lo w . L es s co m m on ir re gu la r g ra in s w ith w ho le co at in g o r o nl y ce nt ral ar ea m ad e o f cl ot te d pe lo id al m icr ite as T 2. O str aco de s, bi va lv es , ra re d et rit al fe ld sp ars . Be ds : 5 m m to 20 cm th ic k, av er ag e 4 cm. T6 in a ll c or es from 4 to 2 5% (S2 , S1 ). T6 c an ev ol ve in to T 2 o r T 1 den dr ites .T 6 i s as so ci at ed w ith T 8, T2 , T4 ; T6 a dj ac en t to c on ve x-upw ar d rim s, m ad e o f T 1. T6 v er tic ally al te rn ate s w ith a ll faci es ex cep tT9 , T1 1 a nd T1 4. Cem en t t yp es in in ter pa rti cl e po re s pa ce: a ) eq uan t b lo ck y m os ai cs of m ic ros pa rit e ( 20 -15 0 µ m ); b) p ris m at ic scal en oh ed ral c ry sta ls (4 00 -5 00 µm lon g) sy nt axi al on ra dia l cr ys tal co at in g; c) p re ce di ng pe nd an t c em en t ( 10 0-200 µ m irre gu la r d ro pl et s) in so m e ca ses w ith F e o xi de. Me te or ic vu gg y d iss olu tio n, m ic riti za tio n an d s pa ritiz atio n. m in : 1 % ; m ax : 12 % ; a ver ag e: 6%. In te rp ar tic le an d i nt ra par ticl e po ro sit y f rom su b-mm t o 1 c m i n size; tr ap ped g as bu bb le s a nd irre gu la r fe ne stra l-lik e v ug s. Lab el le d as ra di al p iso id so rs ph er ul ites (F olk & C ha fet z 19 83 ; C ha fe tz & F ol k 19 84 ; G uo & R idi ng 19 98 ; C oo k & Cha fe tz 2 017 )o r r ad ia tin g d en dr ite s (R ai ne y & Jon es 2 00 9) p re ci pita te d w he n di se qui lib riu m is h ig he r t ha n du rin g cr ys ta lli ne d en dr ite p re cip ita tio n (Jo ne s & Re na ut 1 99 5) or du e t o CO 2 de ga ssi ng/ eva por at io n fro m q ui es ce nt wa te r( Ra ine y & Jo ne s 200 9) .W he n T 6 gr ai ns a re tr an sit ion al in to T 2 t he y m igh t r ep re se nt st ag na nt c on di tion s a nd in cr eas in g m icr ob ial ly in flue nc ed pr ec ip ita tio n (c f. G uo a nd R idi ng 19 98 ). T6 gr ai ns oc cu r i n s ha llo w p ool s of ter ra ced sl op e s ys tem s, i n s ta gn an t po nd s w ith reed sa nd o n i nc lin ed su rfa ce s o f s lo pe s( D el la P or ta e t al . 20 17 ). T6 rad ial g rai ns ar e a co m m on fa bri c o f t he T iv ol i t ra ve rti ne s. T he y ha ve b een d es cr ib ed in o th er lo ca lit ies in Ce nt ra l I ta ly (D ell a P or ta e t a l. 2 01 7) . Tab
. 1 - Description and inter
pretation of
tra
ver
tine facies
Fa cie s C ore d es cri pti on Pet ro gra ph ic an aly sis Thi ck ne ss a nd sp atia l d istr ib utio n D iag en et ic f ea tu res Po ro sity (% a rea in t hi n s ect ion) D epo siti on al pr oce sse s a nd env iro nm ent T7 R af t gra in sto ne / ruds tone (Fi gs 7A -C) W hit e t o l ight g rey , gr ain -su pp or ted g rain sto ne to r ud ston e m ad e of fla t s ub -mm t hic k fra gm en ted ca rbo na te p lates (ra fts ). Ind ivid ua l r afts co nsis t o f m icr iti c a nd m icr os pa riti c s ub -mm thic k f ilms fro m w hich pr ism at ic cal cite cr yst al s g row per pen dicu lar ly . R af ts c an a ct as nu clea tio n s ub stra te f or T 1 a nd T 2 de nd rite s o rie nte d u pw ard o r d ow nw ard . Bed s: 0 .5 m m to 2 cm th ic k, 1 -10 c m w ide l en s-s ha ped . T7 v er y r are, a f ew cm s i n e ach co re. T 7 ass oc iat ed w ith T 4, T2 , T 6; it o ve rlie s T1 1 a nd T1 3. Int er pa rtic le s pa ce c em en ted by :1 ) e qu an t m ic ros pa rit e; 2 ) pri sm at ic s cal en oh ed ral cr yst al s (200 µ m lo ng ). m in: 5 % ; m ax : 15 % ; a ve rag e: 8%. Inte rpa rti cle po ros ity up to 1 .5 cm in si ze. Raf ts ar e f ilm s o f car bo nat e p reci pitat ed on th e s urf ac e o f p oo ls, a t th e w ater -air inter fac e, in dica tiv e o f s tagn an t wa ter (C ha fet z et a l. 1 99 1; Ga nd in & Ca pe zzu ol i 2 01 4; D ella P or ta 2 01 5) an d for m ru dst on e w he n fra gm en ted .R aft dep osi ts a re l en s s ha ped bec au se acc um ulat ed in la ter ally c on fin ed po nd s. T8 Wh ite co ated reed bo unds tone to gra in sto ne / ruds tone (F ig s7 D -G ) W hit e c ol ou r, el on ga ted up rig ht ve rtic ally o rie nte d or ho riz on tally pro stra ted ca rbo na te c oat ed stem s w ith ro un de d cros s-s ecti on (m ostly re ed s; d iam ete rs 0 .5-5 m m , le ng th u p to 1 0 c m ), fo rm ing bo un dst on e t o gra in sto ne /ru dsto ne . Bou nd sto ne o f c oa ted reed s i n g row th po siti on an d f rag m en ted reed g rain sto ne/ pa ck sto ne /ru dst on e. C oat ing typ es: a ) ri m s (0 .1-1 m m th ic k) o f le ioli tic an d c lo tte d pel oid al m icr ite an d m icr osp arit e, 2 ) tu bes w eld ed b y c lotte d m ic rite ; 3 ) c oa tin g a s T 1 an d T 6 cr yst al s ( 60 -15 0 µ m lo ng ) a rra nge d rad ially , f oll ow ed b y c lotte d m ic rite a nd m icr osp ar ite . R ar e C ha ro ph yte ss tem s, ost rac od es, g ast rop od s a nd ra fts . O str aco de s in c lotte d m icr ite o r i n ge op eta l p ositio n f illin g p rim ary p ore s. Bed s: 1 -24 c m th ick . T8 in a ll c or es f rom 2 t o 8 .5% (S 1, S 2, S7 ).T 8 as so ci at ed later ally w ith T 2, T6 , T 10 o r m ak ing the c ore o f r im s. T 8 alte rna te s w ith T 10 an d u nd er lies or ov er lies T 13 d et rita l lay er s. Cem en t t yp es b et w een o r w ith in s tem m ou ld ic po res : 1 ) eq ua nt b loc ky sp arit e; 2 ) pri sm at ic s cal en oh ed ral c em en t syn ta xia l w ith c rys ta llin e co atin gs ( 400 -800 µ m lo ng ) an d u nd ulo se e xtin ctio n; 3 ) irre gu lar b ro w n-c olo ur c rus ts (8 0-2 00 µ m th ick ) of p en da nt cem en t; l oc ally p yrite in p ore s. m in: 0 .1% ; m ax : 20 % ; a ve rag e: 6%. Int erp ar ticl e, intr ap ar ticl e a nd seco nd ar y m ou ldic p oro sity fill ed b y o stra co de gra in/p ack sto ne an d p yrite . Reed s g row i n f res hw ater or dilu ted an d co oled th er m al w ater s, i n m ar sh y f lat s an d f orm m ou nd s at the b ase of tra ve rtin e s lop es (m ars h p oo l f aci es; G uo & R idin g 199 8). T8 m ark s t he pres en ce o fv eg eta tion du e t o c ool ed th erm al w ate r a nd m ix in g w ith fre sh w ate r in d ista l lo catio ns o r fol low ing su ba eria l e xp osu re (reed s en cru ste d b y c arb on ate w he n t he ther m al a ctiv ity res um es; D ella P ort a e t al. 2 01 7). T 8 o ccu rs i n s ha llo w p on ds an d ter race po ols a ssoc iate d w ith T 6; oft en o ver lies th e d et rita l f ac ies T 13 . T9 B row n co ated reed and C ha ro ph yt es bo unds tone (Fi gs 7H -L ) Ta n t o b row n o r gr ey bo un dst on e to g rains ton e/ru dst on e f orm ed b y car bo nat e e ncr ust ed m m -size C ha ro ph yte sa lga e a nd reed stem s (1 -20 c m in le ng th, 0.5 -2 m m in dia m et er s). T he d ark c olo ur an d the ab un dan tp res en ce o f C ha ro ph yte sd istin gu ish fa cie s T 9 fro m T 8 at th e co re s cal e. Bou nd sto ne w ith i rre gu lar fra m ew ork o f clo tte d p elo ida l m ic rit e a nd m ic ros pa rit e pre cipi tate d a round C ha ro ph yte s an d r eed ste m s a lso in lif e p osi tio n. G rain sto ne co nsi sts of pe loi ds, c lots o f p elo ida l m icr ite, c oat ed ste m so f a lga e a nd reed s. O str aco de s ( w ith in t he c lott ed p el oid al m icr ite an d w ith in p rim ary fra m ew ork po ros ity ), g ast rop od s, c arb ona te i nt rac last s an d r af ts. Bed s: 5 -30 c m thick , up to 50 c m (S 1). T9 in a ll co res from 1 to 11 % (S7 , S3 ). At th e bo tto m o f S 7 it alte rna te s w ith F5 m arls .T9 a sso ci at ed w ith T1 0, T1 1 an d T1 3. Cem en t t yp es : 1 ) p en da nt m icr itic c em en t; 2 ) eq uan t m icr os pa rite to sp arit e ( 10 0-20 0 µ m ); 3 ) s cal en oh ed ral pri sm at ic ce m en t (2 00 -800 µ m lon g); 4 ) fi bro us cry sta l fa ns. Cem en t g row s a rou nd reed m ou lds a nd a bo ve sm all r afts . Ev id en ces o f e arly d iss olu tio n. m in: 1 % ; m ax : 15 % ; a ver ag e: 5%. Fram ew ork , inte rpar ticl e an d intr ap ar ticl e po ros ity (0 .5-5 mm, to 2 cm ), pa rtl y f ille d b y silt , o straco de s an d c em en t. Ph ys ico -ch em ical an d m icr ob ial ly inf lue nce d p roce sse s. Sh al low lacu stri ne po nd s to p alu stri ne e nv iro nm ent w ith ter rig en ou s a nd fres hw ater or c oo ler th erm al w ate r in pu t.T9 o ccu rs i n th e dis ta l S 7 c ore an d w ith in a n in te rm ed ia te u nit oc cu rrin g at 15 m de pth (fr om S 1 t o S 5-S 6) du rin g w hic h th e m ore d ista l-li ke f ac ies p rev ail ac ros s the w ho le t ran sec t. T10 C lotte d pe lo id al m icri te gra in sto ne / bo unds tone w ith C ha ro ph yt es (Fi gs 8A -C ) Li gh t ta n to g rey h om og en eou s tex tur e, m es osc op ica lly a pp ea rin g as c al ci -m ud sto ne /w acke sto ne w ith sp ars e m m -size v ug s an d tub ula r a lga l s tem s w ith 0 .2-1 mm d iame ter , in v ertic al l ife po siti on an d f ragm en ted , lo cal ly ass oc ia te d w ith ra ft f rag m en ts. Clo tte d p elo id al m icr ite g rain sto ne to bo un dst on e m ad e o f a ) irre gu lar m icri te clo ts (1-0.1 m, mo stly 20 0-3 00 µm in siz e) of l eio litic an d c lotte d p elo id al m ic rite (pe loi ds 10 -20 µm in d ia m ete r, w ith 1 0 µm m ic ros pa rite ), p artl y w el ded to get her ; b) i rre gu lar fra m ewo rk of c lotte d m ic rite ; c) p eloi da l gr ain sto ne w ith Ch aro ph yte s ste m s; d ) w ack est on e w ith C ha ro ph yte s an d o stra co de s. S par se c oa ted ga s b ub ble s, ost raco de s, ga stro po ds, fil am ent ou s m icr on -size ba cte riof or m str uct ures . Be ds: 5 m m to 15 cm th ic k; m ax im um 70 c m (S 3). T10 in all c ore s fr om 1 to 18 % (S 3, S 1, S7 ). T1 0 at ons et o f tra ve rtin ed ep osi tion ove rly ing F 1. T1 0 alte rna te s w ith T 2, T6 , T8 an d T1 3. Cem en t t yp es : 1 ) eq uan t ( 20 -10 0 µ m ) m icr os pa rite to s pa rite irre gu lar ri m s a ro un d m icri te clot s; 2 ) e qu an t m ic ros pa rite /sp arite fo llo w ed by p ris m at ic s ca len oh ed ral cem en t (2 00 -25 0 µ m to 1 mm lo ng ); 3 ) lo call y p en da nt va do se c em en t; 4 ) r are fi bro us cem en t w ith und ulo se ex tin ctio n. D iss ol ut ion o f m icr ite an d m icr osp arit e. m in: 0 .5% ; m ax : 7%; a ver ag e: 3 %. Su b-m m to c m -size su b-ho riz on tal fen es tra e-li ke po res a nd irre gu lar vu gs, inte rpar ticl e, m icr op or os ity, co at ed g as bu bb les po ros ity . Faci es T1 0 n ot dif fer en tia te d in T ivo li tra ve rtin es i n p rev io us w ork s. C lotte d pe lo ida l m ic rite fa bric su gg est the inf luen ce o f m icr ob ial m ats .T1 0 rep res en ts the on set of t ra ve rtin e de po sit io n i n a la cu str in e to sta gn an t po ol e nv iron m en t a nd it oc cu rs i n sh allo w fla t p on ds. Tab
. 1 - Description and inter
pretation of
tra
ver
tine facies
Fa ci es C ore d es cri pt io n Pet ro gra ph ic an al ys is Thi ck ne ss a nd sp at ia l d ist rib ut io n D ia gen et ic f ea tu res Po ro sity (% a re a in t hi n s ec tio n) D epo sit io na l pr oc es se s a nd env ir onm ent T11 B row n in tra cl as tic ph yt ocl as tic pa ck st one / gr ai ns to ne / ruds to ne (Fi gs 8D) Br ow n t o d ar k g re y pa ck sto ne /gr ai ns to ne to rud sto ne w ith p el oid s, a ng ul ar in tra cl as ts, rad ial co at ed g rai ns ,v eg et at io n stem fr ag m en ts ( up to a f ew c m s in len gt h, a few m m s i n di am et er ). Pack sto ne: m atr ix o f l eio liti c m ic rite a nd m ic ro sp ar ite. G ra in sto ne: eq uan t b lo ck y sp ar ite as ce m en t. G rai ns : a) C ha ro phy te s ste m s ( up to 5 m m lo ng ,0 .5 m m in dia m et er), b ) re ed fra gm en ts ( up to 1 c m lo ng ); c ) i rre gu la r c lo ts o f p el oi da l m ic rit e (u p to 5 m m in siz e) ; d ) i ntr ac la sts o f m icr ite , co at ed b ub bl e an d r af ts, r ad ia l co at ed g rai ns ; e) p el oi ds an d f aec al p el let s. G as tro po ds , b iv al ve s a nd o str ac od es . Be ds : u p t o 42 c m th ick (S 2). T1 1 in a ll co re s f rom 1 to 12% (S7 ,S3 ,S2 ). T1 1 as so ci at ed w ith T 9, bo th v er tic al ly a nd lat er al ly , an d w ith T1 0, T8 a nd T1 4. Cem en t t yp es : 1 ) eq uan t m os ai cs of m ic ros pa rit e ( 10 -5 0 µm ) to sp ar ite (1 00 -3 00 µ m ); 2 ) sc al en oh ed ra l p ris m at ic c em en t (2 00 -2 50 µ m ). P ed oge ni c al te ra tio n, F e o xid es , p yr ite in fil l o f v oid s, d es ic ca tio n sed im en t f ill ed fi ss ur es . m in : 1 % ; m ax : 2. 5%; a ver ag e: 2%. Inte rp ar ticl e, in trap ar ticl e, mo ul di c, fen es tra l po ro sit y ( 0. 1-3 m m , u p to 1 .5 cm ). D et rita l f ac ie s r ew or kin g T9 a nd T1 0, acc um ul at ed in sh al lo w p on ds , la cu str in e to p al us trin e e nv iro nm en ts w ith te rrig en ou s a nd fre sh w at er i np ut de ve lo pi ng i n di sta l sy ste m (S 7) a nd as T9 in an in ter m ed iat e un it in th e su cc es sio n oc cu rri ng fr om c or e S 1 to S6 ar ou nd 15 m d ept h du rin g w hi ch th e m or e d ista l-l ik e fac ies p rev ai la lso in pr ox im al z on es . T1 2 Wh ite in tra cl as tic co at ed g ra in gr ai ns to ne / ruds to ne (Fi gs 8E -F ) W hite /g re y gr ai ns to ne/ pac ks to ne to ru ds to ne w ith tr av er tin e in tracl as ts (0 .1 m m to 5 cm si ze, su b-ro und ed to su b-an gu la r), sk el et al fr ag m en ts an d r ad ia l co at ed g rai ns . N o s ed im en ta ry str uct ur es . I nt ra cl as ts w ith m ic riti c to m icr os par iti c car bo nat e co at in g. G ra in sto ne /ru ds to ne w ith fra gm en ts o f: cl ot s o f l eio lit ic /c lo tted m ic rit e ( T1 0) , co at ed b ubb le (T 4) , c oa te d reed s (T 8), ra fts (T 7), ra di al c oa te d g ra in s (T 6), cl ot te d p el oi da l m ic rit e (T 2) , c ry sta lli ne den dr ites (T1 ). O str ac od es , g as trop ods , pe lo id s, r ar e p ho sp ha te g ra in s, f ra gm en ts of p ut at ive fi lam en to us cy an ob ac ter ia . Bed s: 0 .5 -2 0 c m th ic k (S 3, S 4), m os tly 5 mm. T1 2 in al l c or es fr om 1 to 6% ( S4 ); it o ver lies T1 3 a nd it is ov er la in b y T 6, T 2 or T8 . Cem en t t yp es in in ter pa rti cl e po res : 1 ) eq uan t b lo ck y m ic ro sp ar ite a nd sp ar ite lin ed by irre gu la r m ic rit e; 2 ) pr ism at ic s cal en oh ed ral ce m en t. Me te or ic d iss ol ut io n. m in : 2 .5 % ; m ax : 9%; a ver ag e: 5 %. In te rp ar tic le (1 -2 mm ); f en es tra l (1 .5 cm ); in tra pa rtic le / m ou ld ic s tem s (5 mm) . D et rita l g ra in y s ed im en t d ue to lo ca l re w or kin g o f tr av er tin e p re cip ita te s. A cc um ula tin g in p oo ls o r i nc lin ed su rfa ces o r a ss oci at ed w ith (p rec ed in g or fo llo w in g) p er io ds o f s ub ae ria l ex po su re ( T1 3) an d in ac tiv ity o f t he hyd ro th er m al sys te m . T1 3 Gr ey in tr ac la sti c ex tr acl as tic w ack es to ne/ flo ats to ne / ruds to ne (Fi gs 8G -J ) D et rita l tr av er tin e i ntr ac la sts (an gu lar cl as ts, su b-mm to d ms be yo nd c or e w id th ) i n g re y m atr ix (cal ci -m ud sto ne /ma rls ); po or ly so rte d gr ai ns ton e, p ac ks to ne , floa tst on e a nd ru ds ton e ( m ic rit e m at rix , o r c em en ted b y s pa rit e) . T1 3 as so cia te d w ith te rrig en ou s de po sit s ( cl ay sto ne an d m ar ls w ith o str ac od es w ith si lic ic la sti c ex tracl as ts) . E ro sion al ba se; gr ai n-su pp or ted tex tu res ar e a t t he ba se an d t op o f T 13 la ye rs . Gr ai ns to ne , p ack sto ne , f lo at sto ne , ru ds to ne tra ve rti ne fra gm en ts o f T 1, T 2, T6 , T1 0, T4 a nd T8 . V ad os e p iso id s; po ss ib le co at ed cy an ob ac te ria fil am en ts. Ex tracl as ts: q uar tz , m ic a an d K -fe ld sp ar cr ys tal s. W ack es to ne to fl oat sto ne fac ie s sh ow a d en se m icr ite an d m icr os par ite m atr ix , w ith c la y co nt en t. M icr ite ca n sh ow a m en isc us fa br ic fo llo w ed b y ce m en t. S om etim es in tra cl as ts s ho w sh ow dis so lu tio n a nd re cr ys ta lliz atio n f ea tu re s. Bed s: from 5 m m to 4-7 m th ick (S 6). T1 3 in a ll c or es from 1 to 3 4% (S 6, S4 ). T1 3 ve rtic al ly fo llo w ed o r pr ec ed ed b y T 12 an d T 8. Cem en t t yp es a re: a ) eq ua nt sp ar ite m os ai c; b ) s cal en oh ed ral pr ism at ic c em en t; c ) v ad os e pe nd an t ce m en t. T he y can al so fo llo w m en isc us -li ke m ic rite bi nd in g. A lv eo la r t ex tu re re la te d t o s oi ls a nd fun gi in be tw ee n T 2 cl as ts. Sp ar iti za tio n of tr av er tin e i ntr ac la sts d ue to m et eo ric w at er d ia gen es is. m in : 1 .5 % ; m ax : 11 .5% ; a ver ag e: 6%. Inte rp ar ticl e, fen es tra l, v ug gy an d m ou ld ic po ro sit y ( 1 mm -2 cm in si ze) Lith oc la st tr av er tin es w er e a ss oc ia ted w ith p ed og en ic cal cr et e an d p al us tri ne lim es to nes in fr es h-w at er m ar sh es b y (G uo & R id in g 19 98 ). T1 3 r ep res en ts un co nf or m iti es : d et rit al fa ci es d ue t o no n de po sit io n a nd e ros io n of tra ve rtin es , r el at ed to p has es o f in te rru pt io n of th e t her m al w at er fl ow o r ch an ges in th e f lo w d irec tio n (G an di n & Cap ez zu ol i 2 01 4; C oo k & C ha fe tz 20 17 ). D et rit al ru bb le a nd te rri ge no us m at rix, so ils a nd ve ge ta tion a re en cr us te d by c ar bo na te p re cip ita tio n w he n t he h ydr ot her m al fl ow is ren ew ed . T14 B ro w n in tra cl as tic ph yt ocl as tic pa ck st one / gr ai ns to ne to bo unds to ne (Fi gs 9A -D ) Ta n to li gh t b row n, gr ai n-su pp or ted p ack sto ne t o gr ain sto ne /ru ds to ne to bo un ds ton e w ith reed s, ph yt ocl as ts, car bo nat e i nt racl as ts an d s ili ci cl as tic d et rit al gr ain s. G ra in si ze fro m su b-m m to 1 -4 mm. No se di m en ta ry st ru ct ure s.. Pa ck sto ne to gr ai ns ton e/ ru ds ton e a nd bo un ds ton e w ith p el oi ds an d m ic rit e c lo ts, in tra cl as ts o f c lo tte d p el oid al m ic rite , r eed an d al gae fr ag m en ts co at ed b y cr ys tal lin e cru sts (u p to a fe w 1 00 µm th ic k) , w ith w av y pa tte rn s a nd un du lo se e xt in ct io n, rar el y co at ed b y m ic ro sp ar ite to c lo tte d m icr ite . O str aco de s, bi va lv es ,g as trop od s, det rit al fel ds pa rs , p ho sp hat e g ra in s. Ra re C ha ro ph yt es , p os sib ly d ia to m s; fil am en to us tu fts of p ut ativ e ca lc ifi ed cy an ob ac te ria co at p la nt st em s. Bed s: av er ag e 6-7 cm th ic k; 9 0 cm th ic k i n S 4. T1 4 i n all co res 0.1 -2 % . T1 4 i nt er cal at ed w ith te rrig en ou s m ar ls a nd sil tsto ne (F5 , F6 ). T 14 in S 1 to S 7 as so ci at ed w ith T9 , T1 0, T1 1. Co at ed st em s s ur ro un ded b y irre gu la r m ic rit e c oa tin gs fo llo w ed b y e qu an t c al ci te sp ar (1 00 -2 00 µ m ) a nd p ris m at ic sc al en oh ed ra l c em en t ( up to 20 0 µ m in le ng th) . Ma trix so m etim es (i n S 2 an d S3 ) s ili cif ie d. m in : 0 .5 % ; m ax : 4%; a ver ag e: 2 %. In te rp ar ticl e, in trap ar ticl e, m ou ld ic ( pl an t stem , m ollu sc s) . Su b-m m si ze ( up to 2 mm, ra re ly 5 mm) . Pa lu str in e t o sh allo w la cu str in e, fres hw at er po nd de po sit w he re pl ant stem s a re en cr us ted b y c ar bo na te pr eci pi tat ed in sh al lo w st ag nan t w at er w ith si lic ic la stic sa nd an d sil t i npu t. O cc urre nc e o f T 14 fa ci es a ) i n t he in ter m ed ia te p ha se ar ou nd 1 5 m d ep th asso ci at ed w ith T9 , T1 0, T1 1, T1 3; a nd b) in th e tr av er tin e s ub str at e s tra ta bet w een d ec am et re -th ic k s ilic ic la sti c an d vo lcan ocl as tic d ep os its . Tab
. 1 - Description and inter
pretation of
tra
ver
tine facies
