Cave clastic sediments and implications for speleogenesis: New insights from the Mugnano Cave (Montagnola Senese, Northern Apennines, Italy)

Geomorphology 134 (2011) 452–460

Contents lists available at ScienceDirect

Geomorphology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / g e o m o r p h

Cave clastic sediments and implications for speleogenesis: New insights from the Mugnano Cave (Montagnola Senese, Northern Apennines, Italy) Ivan Martini Department of Earth Sciences, University of Siena, via Laterina 8, Siena, Italy

a r t i c l e

i n f o

Article history: Received 9 April 2010 Received in revised form 20 July 2011 Accepted 28 July 2011 Available online 4 August 2011 Keywords: Speleogenesis Cave sediments Phantomisation Allostratigraphy Northern Apennines

a b s t r a c t The study of cave clastic sediments has been considered one of the hottest topics during the last years because of their importance in paleoclimatic reconstructions and archaeological surveys. This paper focuses on clastic deposits of the Mugnano Cave, a small cave located in the Siena district (Northern Apennines, Italy), showing unique features regarding the sedimentary fill, mostly made of grey-blue dolomitic silts. The sedimentary succession was investigated through a detailed sedimentological analysis aimed at a better understanding of sedimentary processes active during the deposition. The entire succession was subsequently reinterpreted through an allostratigraphic approach: the recognition of an important erosional surface, associated with a significant change in sedimentation, allowed the distinguishing of two main allounits labelled MG1 and MG2. Furthermore, the different kinds of sediments collected in the cave were analysed using the XRF and XRD techniques, in order to establish their chemical and mineralogical compositions. The integration of lithological, sedimentological, allostratigraphic and mineralogical data permits formulation of an interesting hypothesis about speleogenetic processes that influenced the cave, with particular reference to the processes capable of generating the underground space. In this context, most of the current available space results from a complex interplay between different processes: disintegration of a particular lithofacies of the bedrock, consequent production of sediments and deposition into a subterranean lake. These sediments were removed from the cave during some non-depositional and erosive phases, which led to a positive balance in the available space. © 2011 Elsevier B.V. All rights reserved.

1. Introduction In the last decades several studies on cave sediments have been performed. However, they were generally focused on chemical sediments that are formed in place, precipitated from solution. This kind of deposit is an excellent source of information for paleoclimatic reconstructions. In contrast to the widespread interest in chemical sediments, only in the last years have geoscientists paid attention to clastic sediments (Bosch and White, 2004; Sasowsky and Mylroie, 2004; White, 2007). These deposits are mainly made of clasts or particles that are moved mechanically before the deposition and can be subdivided into autochthonous and allocthonous sediments (White, 2007). The first derive locally within the cave and often correspond to the insoluble component of the bedrock, while the latter are composed of sediments transported into the cave from outside (White, 2007 and reference therein). Clastic sedimentary fills are usually studied because they may give information concerning the evolution of the surrounding territory: for example changes in the deposition in a cave are very sensitive to landscape erosion or flood-related events, which are often directly

E-mail address: [email protected]. 0169-555X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.geomorph.2011.07.024

connected with the climate evolution of the area (Bosch and White, 2004; Springer, 2005; White, 2007; Auler et al., 2009). Furthermore, this kind of study has important implications also in archaeology and palaeontology, whose finds are usually inside cave-fills (Brandy and Scott, 1997; Ghinassi et al., 2009). Some authors hypothesize that clastic sediments in caves may be a source of information about speleogenesis (Springer, 2005), but in spite of the increasing number of papers dealing with these deposits, only few of these have focused on their implications in speleogenetic processes and, in particular, on the creation of the underground space. Only recently some authors (Vergari and Quinif, 1997; Tognini, 1999; Häuselmann and Tognini, 2005; Quinif et al., 2006; Audra et al., 2007; Quinif, 2010) have described a new speleogenetic process (“phantomisation”, or rock-ghost weathering), in which the study of cave clastic sediments provides important information concerning the evolution of karst systems in impure limestone and dolostone rocks. Similar processes have been already described, with different terms, for caves developed in quartz-rich rocks (Piccini and Mecchia, 2009 and reference therein). According to this hypothesis, the bedrock was firstly affected by an incomplete dissolution during a long-term stability of the base level (autochthonous phase sensu Quinif, 2010). This condition allowed the insoluble materials to remain in place,

I. Martini / Geomorphology 134 (2011) 452–460

preserving the structure of the parent rock (phantoms or rockghosts). Generally the dissolution is strongly influenced by rock discontinuities (fractures, joints and faults) and that is the reason why the rock-ghosts are spatially delimited inside the rock massif (Häuselmann and Tognini, 2005; Quinif, 2010). The partial dissolution causes a dramatic increase in rock porosity and consequently in hydraulic conductivity. Residual and unconsolidated materials can be eroded and removed by seepage waters (piping processes), forming small cavities or conduits. These conditions can facilitate the development of an underground drainage network, with a greater capacity to remove the weathered part of the bedrock (fluviatile phase sensu Quinif, 2010). This kind of karst evolution can continue until the unconsolidated material is completely removed; at this point it would not be possible to recognize the evidence of the ancient ghost-rocks and the evolution of the caves could just have occurred through classical karst processes (Audra et al., 2007). This paper deals with the study of the sedimentary fill of the Mugnano Cave (Tuscany, Italy) which shows unique and interesting features. A sedimentological analysis was performed and used in conjunction with an allostratigraphic interpretation of the entire succession, in order to understand the relationships between the sediments and speleogenesis. 2. Geological setting The Montagnola Senese is composed of a series of N–S aligned hills west of Siena (Tuscany, Italy) and is located in the inner Northern Apennines; a fold-thrust chain formed during the Tertiary in response to the interaction between the Adria and Corso-Sardinian microplates (Carmignani et al., 2001 and references therein). The present structural setting of this sector of the Apennines is the result of two major deformation episodes: the first one, characterized by a compressive tectonic regime, was associated with the convergence between the European margin and the Adria micro-plate which occurred between the Cretaceous–Late Oligocene and Early Miocene (Carmignani et al., 1995). A second deformation episode was related to the regional extension that concerned inner Northern Apennines since the Early–Middle Miocene (Carmignani and Kligfield, 1990; Jolivet et al., 1991; Elter and Sandrelli, 1994). This stage was realized through two main extensional events: the older one (Early–Middle Miocene) was characterized by the action of low-angle normal faults and caused a stretching of 120% or more (Carmignani and Kligfield, 1990; Brogi, 2003, 2004) and the exhumation of the originally deeply buried rocks (now outcropping, for example, along the Middle Tuscan Ridge that structurally includes the Montagnola Senese area). The younger extensional event (Early Pliocene–Quaternary) was dominated by the development of high-angle normal and transtensional faults (Brogi, 2003) that caused a lower stretching (about 10%) and gave rise to the present “Horst and Graben” structure of inner Northern Apennines (Carmignani et al., 2001). The Middle Tuscan Ridge is mostly made of Late Paleozoic to Mesozoic formations (Monticiano–Roccastrada Unit) that include clastic (Verrucano group) and carbonate lithologies (e.g. “Montagnola Senese Marbles Formation”) (Giannini and Lazzarotto, 1970), affected by a green-schist facies metamorphism. The Monticiano–Roccastrada Unit was locally overlapped by the non-metamorphic Tuscan and Ligurian Nappes (Giannini and Lazzarotto, 1970). The Tuscan Nappe in the studied area is represented only by the “Calcare Cavernoso” formation that widely crops out on the northern sector of Montagnola Senese. 2.1. The “Calcare Cavernoso” formation The name “Calcare Cavernoso” means vuggy limestone, but the formation mainly consists of clast- to matrix-supported dolomitic limestone with a characteristic vacuolated texture (Passeri, 1979;

453

Gandin et al., 2000; Lugli, 2001; Lugli et al., 2002). This formation crops out diffusely in southern Tuscany where it is present at the same stratigraphic level as the Triassic evaporites of the Burano Anhydrite Fm. that represents the base of the non-metamorphic Tuscan Nappe. During the last century many geoscientists focused their attention on the Calcare Cavernoso in order to understand its origin, which for a long time has been related to the transformation of the original Burano Anhydrite Fm. as a consequence of syn- and post-orogenetic events. These events include the break-up of the brittle dolomitic layer, anhydrite fluidification and hydration–dissolution cycles of the sulphates after exhumation (Signorini, 1946; Merla, 1952; Trevisan, 1955; Giannini and Lazzarotto, 1967). The importance of hydration and dissolution processes on the development of the Calcare Cavernoso was understood a long time ago, since it was observed that the original evaporitic–dolomitic lithotypes are preserved only where an impermeable cover has prevented sulphate dissolution (Signorini, 1946; Signorini and Pieruccini, 1968). This impermeable cover was made of shale of the Ligurian Nappe. Despite this observation for a long time the origin of Calcare Cavernoso was predominantly related to tectonic brecciation. Recently Gandin et al. (2000) demonstrated that the Calcare Cavernoso is a completely new rock that was formed as a consequence of the dissolution of sulphates and destabilization of the original brecciated dolostone. The same authors suggest that it is incorrect to consider this rock as a breccia because it maintains traces of its past brecciated features locally. A particular lithofacies of the Calcare Cavernoso Formation is the so-called “Cenerone” (Passeri, 1975): it is made up dark grey, powdery or sandy dolomite that forms masses or lenses characterized by a poor degree of cohesion (Gandin et al., 2000; Lugli, 2001). The origin of this incoherent lithofacies is attributable to weathering processes that affected the remnants of the original dolomitic clasts (Gandin et al., 2000; Lugli, 2001). Weathering acted in recent times (post-Miocene according to Lugli, 2001) and was favoured by the intense tectonic fracturating of dolomitic layers, which facilitated the percolation of waters. The mechanical features of this lithofacies are similar to the weathering-derived residual sediments, observable in other karst areas (Burger, 1989; Vergari and Quinif, 1997; Tognini, 1999; Quinif et al., 2006; Quinif, 2010). Very similar to the Calcare Cavernoso is the “Breccia di Grotti”, a formation introduced by Signorini (1966) and then described by Giannini and Lazzarotto (1967, 1970), that was made up of continental breccias and sandstones with a dominance of Calcare Cavernoso elements. The deposition of “Breccia di Grotti” occurred during the Neogene in palaeo-morphological depressions in an alluvial fan setting. It is often very difficult to distinguish the two formations, but it is important to note that in the “Breccia di Grotti” lithofacies similar to “Cenerone” are lacking. 3. Cave location and description The Mugnano Cave (T/SI 258 in the Tuscan register of caves) is located in the Siena district (Fig. 1a,b,c), and has its opening at an altitude of 306 m in the Fungaia area (Montagnola Senese). The region is hilly, the highest altitude is represented by Monte Maggio (671 m a.s.l.), and is bounded to the north and south by two plains, respectively “Pian del Lago” and “Pian del Casone” (both about 280 m a.s.l.). In detail, “Pian del Lago” appears as a large and closed depression delimited by karst terrain (Calcare Cavernoso and Breccia di Grotti Fms.), that has generally been interpreted as a polje (Pascucci and Bianciardi, 2001; Pascucci, 2004). The sedimentary infill of this polje is generically attributed to the Quaternary (Costantini et al., 2009), but an older age for the formation of this depression cannot be excluded. A different origin is conceivable for “Pian del Casone” that, according to Capezzoli et al. (2009) and Capezzuoli and Sandrelli (2004), is interpretable as a portion of a palustrine–lacustrine tectonic basin, developed in response to the regional Middle Pleistocene uplift of the Northern Apennines (Bartolini,

454

I. Martini / Geomorphology 134 (2011) 452–460

Fig. 1. (a–b) Location of the Mugnano Cave. (c) Simplified geological map of the area surrounding the cave. Key to the symbols: 1—Recent alluvial deposits; 2—Colluvium; 3— Lacustrine deposits (Quaternary); 4—Marine sandy deposits (Pliocene); 5—“Calcare Cavernoso”. limestone; 6—Landslide; 7—Doline; 8—Cave entrance.

2003). In these flat areas lacustrine sediments occur, testifying to the presence of two important Quaternary lakes (Fig. 1c), which existed until the 18th century when they were drained for agriculture (Pascucci, 2004). This little cave is entirely developed in the Calcare Cavernoso Formation and the entrance consists of a 6 m-deep shaft that leads into the first big chamber. Some narrow passages, often excavated in the sedimentary fill, connect the many rooms of the cavity (Fig. 2). The cave is currently dry and it is characterized by the general absence of dripwater speleothems. Remnants of cave-fill deposits are one of its most striking features: they constitute the floor of the cave and sometimes the walls and the vault; the originally denuded rock floor is observed only on rare occasions (Martini, 2007). The present morphological setting is strongly related to the last vadose stage of the cave, which could have obliterated the evidence of previously developed phases. However, the absence of particular forms, such as maze cave patterns and meso-morphological features resulting from rising flows (MSFR sensu Klimchouk, 2009), allows one to exclude an hypogenetic origin for this cave. 4. Methods Sedimentological analysis of sedimentary fill of the Mugnano Cave is based on the facies associations concept, that consists of assemblages of

Fig. 2. Map of the Mugnano Cave (modified from Fabrizi, 1962).

spatially and genetically related facies. A single sedimentary facies is the expression of a well defined set of lithological, sedimentological (grain size, sorting, structures, etc.) and paleontological features that provide information about the sedimentary processes acting during deposition. Facies associations are the expression of sedimentary environments, rarely represented by a single facies. The descriptive sedimentological terminology used is from Harms et al. (1975, 1982) and Collinson and Thompson (1982). The concept of unconformity-bounded depositional units and allostratigraphy (Walker, 1992) has been used in the stratigraphic analysis of cave sedimentary fill. The integration between facies analysis and unconformity-bounded concepts has recently been used in some studies of cave sediments, especially in those concerning the archaeological investigations (Campy and Chaline, 1993; Pickering et al., 2007; Ghinassi et al., 2009; Hunt et al., 2010). A generally accepted classification for clastic cave sediments is lacking. In this work the suggestion proposed by Bosch and White (2004) and White (2007) will be adopted. Bulk chemical analyses are obtained by X-ray fluorescence (XRF). For X-ray fluorescence; samples have been mechanically crushed in a planetary ball mill and manually ground into powder in an agate mortar. Quantitative analyses have been performed on powder discs obtained by pressing 0.5 g of sample on a boric acid pellet. The XRF apparatus was a Philips MagicX-Pro. Background and mass absorption intensities were calculated using calibrations based on 24 international geological reference materials. Loss on ignition was determined by heating samples to 1050 °C for 2 h. The mineralogical assemblage of sediments is determined by X-ray diffraction (XRD), using randomly oriented powders of the bulk sample. The instrument was a Philips PW 1710 Bragg–Brentano diffractometer, with a copper anode operating at 45 kV accelerating voltage and 25 mA sample current.

I. Martini / Geomorphology 134 (2011) 452–460

5. Clastic sediment deposits In this section the clastic sedimentary fill of the Mugnano Cave is described and divided into four kinds of deposits on the basis of sedimentological features. Each one is representative of a different sedimentary environment. Individual deposits can be composed of a single sedimentary facies (breakdown and entrance talus deposits) or of an association of elementary facies, which indicate the coexistence of different sedimentary processes inside the depositional environment (slackwater and backswamp deposits). Each deposit is coded with capital letters. Single facies are indicated with the abbreviation of the related deposit plus a small letter. 5.1. BD—Breakdown deposit This deposit consists of a monogenic facies which includes unsorted boulders and cobbles without cement binding the clasts. Debris shows a chaotic mixture regarding size and disposition. Boulders range in size from some decimetres to 2–3 m and they are made predominantly of the “Calcare Cavernoso” limestone, which forms the ceiling of the cave. Seldom it is possible to observe some blocks made of cemented fine sediments (fallen from the ceiling and from the walls of the cave). The surfaces of the blocks are slightly or not weathered. This facies is the major thickness in large chambers where it covers the old cave floor. The primary openwork texture is commonly obliterated by secondary, infiltrated silty–sandy matrix. Scattered boulders of this facies are also present within fine-grained sedimentary fill (backswamp and slackwater deposits). The presence of these sediments in large chambers and the sedimentological features indicate deposition by rockfall processes (Nemec and Kazanci, 1999; Fornós et al., 2009; Ghinassi et al., 2009). Traditionally, these types of deposits are called “breakdown” deposits and they are considered as autochthonous sediments (White, 2007 and reference therein). The accumulation of breakdown deposits occurs generally during the air-filled, vadose stage of cave life (Hill, 1999). 5.2. ET—Entrance Talus deposit These deposits are located in a limited area under the current entrance of the cave and are made of very poorly sorted materials ranging from sand to boulder, very irregular in size and shape. These deposits have some similarities to the breakdown sediments. Clasts, blocks and boulders are prevalent and often show a sub-rounded morphology. The matrix is made of clay, silt and sand and shows a very irregular distribution. Plant and bone remains are very common. These deposits create a characteristic cone under the entrance and do not show bedding or gradating. The morphological position and sedimentary features suggest that these deposits represent an “entrance talus facies” (White, 2007; Fornós et al., 2009). They are constituted of a mix of sediments that migrate into the cave from the land surface above (driven by gravity and stream-flow) and fallen fragments of the rock coming from the cave walls. 5.3. BA—Backswamp deposit These sediments are the most distinctive of the cave because of their striking grey colour, that seems blue when it is lit by the lamps of speleologists. This facies association includes two sedimentary facies: the most abundant one (BAa, Fig. 3a,b,d) is made up of fine-grained materials, predominantly silts with subordinate very fine sands, well sorted, not graded and grey-coloured. Beds show a clearly visible plane-parallel lamination. The other one (BAb, Fig. 3c) is similar to the previous facies but shows convolute laminations and water escape pipes and pillars; some beds are characterised by soft-deformation

455

due to slumping processes. Facies BAb is generally associated with blocks of the breakdown facies (BD). Some representative samples were submitted to both chemical and mineralogical investigations. Results obtained by XRF are shown in Table 1. The three samples BA1–3 show a marked compositional homogeneity and consists almost exclusively of CaO and MgO. These results are in perfect agreement with those obtained by XRD, showing only dolomite. Sedimentary structures, the grain size and the absence of grading suggest deposition in a quiet, low energy, subaqueous environment (subterranean lake) for the deposits of facies BAa (Bull, 1981; Collinson and Thompson, 1982; White, 2007). A similar origin is conceivable for the BAb facies. Furthermore, the soft-deformation structures can be regarded as the expression of loading due to fallen blocks (BD) from the ceiling of the cave during deposition (Fig. 3c) (Ghinassi et al., 2009). The entirely dolomitic composition of the sediments is indicative of their completely autochthonous origin (indicating that sediments originated within the cave and possibly moved inside the same cave). Small outcrops of dolomitic rocks (“Cenerone” lithofacies) are also present in the external surrounding areas (Gandin et al., 2000), but their transport into the cave would necessarily imply a contamination from other minerals. Therefore, the only possible source area for sediments of the BA facies association is inside the cave. In particular, these fine-grained, dolomitic sediments could be derived from two main processes: • Erosive processes, which acted on the poorly cohesive “Cenerone” lithofacies (e.g. residual dolomitic powder or sand). The incoherent materials which constitute this lithofacies are prone to the mechanical erosion by flowing or dripping waters, characterizing both the phreatic and the vadose stage of karst evolution. Fine-grained sediments are easily washed out, transported into the subterranean lake and deposited on the lake floor; • Disintegration of “Cenerone” lenses by stable and stagnant waters. This process does not require energetic flows and can be considered as gravity-driven. Water is drawn into the less-cohesive lithofacies by capillary force and along interconnected pores, acting as an intergranular lubricant. The cohesive force vanishes and the granules are moved by gravity (Terzaghi et al., 1996). This process may be important in water-filled chambers and galleries, where this lithofacies constitutes the roof of the cave and the water infiltration can easily cause the collapse of less-cohesive materials; The dolomitic sediments are here considered only as the result of the mechanical disgregation and erosion of a particular and lesscohesive lithofacies of the bedrock. A small amount of dolomitic sediments can actually derive also from the weathering of cave walls due to percolation or condensation waters (Zupan Hajna, 2003). Also in this case, the origin of dolomitic sediments is the consequence of an incomplete chemical dissolution of impure limestone. This process is widespread in caves which are developed in dolostone or in dolomitic limestone, such as the Calcare Cavernoso Fm. outcropping in this area. This kind of process generally produces limited sediments and that is the reason why it is not considered as a significant source of sediments for the large amount of dolomitic deposits observable inside Mugnano cave. This statement is also supported by the evidence that the bedrock (when not hidden by sediments) does not appear weathered; a condition indicating that this process is currently inactive on the cave walls. According to the Bosch and White (2004) classification, based on the transport processes, these sediments can be classified as backswamp deposits. These authors, in fact, define backswamp as “deposits that consist mainly of weathering residue of the bedrock and infiltrate material filtering into the conduit system from overlying soils with little or no lateral transport”. In this case there are no clues of the presence of infiltrate materials from overlying soils.

456

I. Martini / Geomorphology 134 (2011) 452–460

Fig. 3. Principal sedimentary features of Unit MG1. (a) Close-up view of the high-relief erosional surface that separates deposits of Unit MG1 from the overlying Unit MG2. The carabiner for scale is about 10-cm long. (b) Typical aspect of backswamp deposits, carabiner for scale is circled. (c) Block of facies BD collapsed from the ceiling of the cave and resting on silty sediments of backswamp facies association, deformed by the impact (arrows indicate distorted levels). (d) Slackwater deposits (facies SLb) interbedded in backswamp deposits (BAa) of Unit MG1.

5.4. SL—Slackwater deposit This facies association is composed of five different sedimentary facies. The most important one (SLa) consist of fine laminated red clays that contain several intercalations of yellow sands, which can be subdivided into four different facies (SLb-c-d-e) on the basis of the sedimentary features (Fig. 4a,b).

Table 1 Results obtained by XRF. Major, minor and trace element contents are expressed as wt. %. The dashes indicate values below the detection limit.

SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 SO3 Cr2O3 Rb2O SrO Y2O3 ZrO2 BaO F Cl L.O.I.

BA 1

BA 2

BA 3

SL1

SL2

0.24 – 0.12 0.08 – 21.35 31.02 0.04 0.02 0.02 0.05 – – 0.01 – – 0.05 – 0.10 46.90

0.46 – 0.28 0.07 – 21.49 31.13 0.04 0.07 0.01 0.05 – – 0.02 – – 0.05 – 0.12 46.20

0.20 – 0.09 0.04 – 18.96 33.87 0.01 – – – – – 0.01 – – 0.01 – – 46.83

52.00 1.19 11.97 12.20 0.47 6.00 2.03 0.67 6.98 0.24 0.24 0.05 0.05 0.03 0.01 0.09 0.22 1.04 0.04 4.49

36.98 1.34 11.78 13.03 0.47 9.74 7.93 0.43 5.67 0.22 0.18 – 0.04 0.03 0.02 0.05 0.24 0.69 0.05 11.10

Facies SLb forms tabular and normally graded (from medium to fine sand) beds up to 5 cm thick, without internal sedimentary structures and characterized by a sharp and non-erosional base. Sediments of this facies are generally moderately sorted, with a low content of clayey-silty matrix. Facies SLc occurs as isolated flat lenses generally cross-laminated and non-graded (Fig. 4d), the granulometry ranges from medium to very coarse sand and the sediments appear very well sorted. The third sandy facies (SLd) is characterized by normally graded beds, 5 to 10 cm thick, with sharp and erosional bases. Sediments are poorly sorted, with an abundant clayey–silty matrix, and their grain size ranges from fine to coarse sand. Strata are massive (without sedimentary structures) or with a weak plane-parallel lamination which seldom evolves upward into cross-lamination. At the base of the beds abundant red mud-clasts are present. The last sedimentary facies is seldom observable (SLe): it is composed of medium to fine sands, organized in cross-stratified tabular beds, 8-10 cm thick, showing a sharp non-erosional bottom (Fig. 4d). These sandy facies (SLb, SLc, SLd and SLe) often contain remains of fossil shells in a poor state of preservation (Fig. 4c) which does not permit distinguishing between a marine or continental origin. Two samples from the facies SLd have been investigated by XRF and XRD (Table 1, Samples SL1 and SL2). Elevated concentrations of SiO2, Al2O3, Fe2O3 and K2O characterise both samples. Contents of Al2O3 and K2O are comparable in the two samples, while SiO2 weight % concentrations are higher in sample SL1 than in sample SL2. Moreover, SL2 shows higher contents of both CaO and MgO. Consistently, XRD analyses show quartz, feldspars, phyllosilicates (mainly muscovite/illite) and iron oxides (hematite and magnetite) as the main phases. Weak peaks of calcite have also been observed. Sedimentological features and lithology of facies SLa suggest a deposition in a relatively quiet and subaqueous environment, where

I. Martini / Geomorphology 134 (2011) 452–460

457

Fig. 4. Sedimentary features of Unit MG2. (a) The best exposition of Unit MG2 inside Mugnano Cave; the entire succession is more than 3 m thick (the lowermost part is not visible in the figure). Carabiner for scale is about 10 cm. (b) Sedimentological log measured in the succession pictured (a), including the lowermost part not visible. (c) Shell remains included in sandy facies. (d) Close-up view of some typical sandy and muddy facies observable in slackwater deposits.

the sedimentation of red clays is the expression of periods of low sediment supply, when suspended sediments have time to settle out and form thin layers (Bull, 1981). Red clays in caves are generally considered to be autochthonous (Bosch and White, 2004) but, in this case, the complete absence of “Cenerone-related” dolomite silts inside the facies SLa suggests an allocthonous origin for these deposits. This hypothesis is also supported by the widespread presence of similar sediments on the land surface (e.g. terra rossa, Pascucci and Bianciardi, 2001). These quiet hydrodynamic conditions were interrupted by the arrival of sediment-laden flows, which were responsible for the deposition of sandy layers. In detail, the layers of normally graded sands (facies SLb) are the result of rapid sedimentation from relatively highdensity turbulent waning flows preventing the development of tractive structures (Collinson and Thompson, 1982; Harms et al., 1982). Crosslaminated and isolated lenses (facies SLc) are interpretable as lenticular bedding, which forms in environments that fluctuate between high and low energy conditions (Collinson and Thompson, 1982). In this setting the lens-shaped sandy deposits represent discontinuous crests of ripples, deposited during the high energy events which lack sufficient sand to form a continuous layer, while during periods of slack water, mud suspended in the water is deposited on ripples to form a drape. A similar origin is conceivable for the facies SLe that is interpretable as the product of the migration of ripples or small dunes during high energy events with sediment supply is sufficient to form continuous beds. The features of facies SLd (sharp and erosional base, mud-clasts, absence of sedimentary structures) are the expression of deposition from very energetic turbulent flows, which are able to erode previously deposited sediments. In this context the cross-lamination sometimes observable at top of the strata testifies to deposition during the last phase, less energetic, of the flow. With reference to the discussed sedimentological and hydrodynamic specificity, these sediments correspond in broad terms to slackwater deposits (Springer et al., 1997; Springer and Kite, 1997; Bosch and

White, 2004; White, 2007). According to these authors, these sediments range in granulometry from clay to fine sand. However, they reach the size of coarse sand in the Mugnano Cave, indicating more energetic hydrodynamic conditions.

6. Stratigraphic analysis The clastic sedimentary fill of the Mugnano Cave has been analysed in terms of allostratigraphy (Walker, 1992), that is a subdivision of the stratigraphic record into units defined and identified on the basis of their observable bounding discontinuities, generally expressed by erosional surfaces. The sedimentary succession has been divided into two main allostratigraphic units (labelled MG1 and MG2) each bounded by unconformities. The entrance talus deposits (ET) are excluded from this analysis because they are important in archaeological and palaeontological research, but they do not usually provide information concerning cave genesis.

6.1. Allostratigraphic unit MG1 This unit is over 13 m thick (considering the total thickness of sediments in superimposed chambers), and consists predominantly of backswamp (BA) deposits (Fig. 3a,b,c,d). Locally, boulder blocks collapsed from the cave vault (breakdown deposits, BD) are also present and resting on fine-grained sediments of BA (Fig. 3c) that show considerable evidence of deformation by the impacts. Slackwater deposits (SL) are very rare and only represented by thin sandy layers of facies SLb (Fig. 3b). Unit MG1 is the largest in volume of sediment; the basal contact with limestone of the bedrock is rarely observable, while the upper contact with the overlying unit MG2 is generally well exposed (Fig. 3a) and it appears as a high-relief erosional surface (Fig. 5).

458

I. Martini / Geomorphology 134 (2011) 452–460

Fig. 5. View of the high-relief erosional boundary that separates Unit MG1 from the overlying Unit MG2.

6.2. Allostratigraphic unit MG2 Deposits of this unit overlie those of MG1, through a sharp and high-relief erosional boundary (Fig. 5). Another important erosional surface is recognizable at the top of the unit, on which lie recent breakdown deposits. The unit shows a variable thickness, ranging from 50 cm to 5 m, depending on the extent of the erosion. Unit MG2 consists predominantly of slackwater facies (SL), while breakdown (BD) and backswamp (BA) deposits are subordinate (Fig. 4a,b). 7. Discussion The analysis of the Mugnano Cave clastic sedimentary fill allows us to understand the depositional history of the cavity and the relationships between sedimentary and speleogenetic processes. Deposits of unit MG1, and in particular the sediments of the backswamp (BA) facies association, testify to a quiet deposition in a subterranean lake environment, seldom influenced by the arrival of sediment-laden floods, which are responsible for the deposition of sandy facies (SLb). The dolomitic composition of the main sediments (facies association BA) reveals a completely autochthonous origin for these deposits. According to what was written previously (subsection 5.3), an allochthonous origin is not possible. It is important to underline the peculiarity of these autogenic sediments: they are usually made of insoluble residues of limestone dissolution (Bosch and White, 2004; White, 2007), but in this case they entirely derive from the erosion and disintegration of a particular lithofacies of the bedrock (“Cenerone”). This is evidenced by the absence of differences in mineralogy and chemistry between the original rock and the resulting sediments. That is why the only possible source areas are the walls and the ceiling of the water-filled chamber, or at least a nearby area connected to the lake by a conduit. The boundary between units MG1–MG2 (Fig. 5) is represented by a high-relief erosional surface, realistically related to a lake-level fall, that testifies to the development of non-depositional and erosive conditions. These kinds of events can generally be related to climatic changes and/or tectonic uplift but, in the case of study, the absence of age constraints does not allow one to distinguish which is the main factor in controlling the lake-level fluctuations.

Deposition of unit MG2 signifies an abrupt change in the sedimentary setting. Most frequently, sedimentary facies (SLa) are indicative of quiet deposition dominated by external sediment inputs where the suspended load has time to settle out. This situation is frequently interrupted by the arrival of sediment-laden floods, testified by the common sandy facies (SLb-e), that document the occurrence of a channelized underground network connected to the surrounding territory. During periods characterized by the absence of floods, the conditions for autochthonous sedimentation are established again (BA deposits). The present setting is the result of the last important erosive phase, which occurred during the vadose, air-filled stage of the cave history. Two cycles of infilling and erosion are consequently recognizable. Concerning the speleogenetic implications, it is important to note that speleogenesis has recently been defined as “The whole evolution of karst systems, from the origin to their full development up to the present form” (De Waele et al., 2009). This definition includes all the changes that take place between the inception and the possible destruction of the cave. Literally, the term “speleogenesis” means the origin and the mode of formation of caves, with particular attention to the processes capable of forming underground space. This definition, although incomplete, is the most common among geoscientists, in particular among those who do not deal specifically with karst. The origin of an “available” space is mainly related to processes of chemical dissolution of the rock. With regard to the origin of this space, the study of the Mugnano sedimentary fill allows one to hypothesize an interesting scenario (Figs. 6, 7). During deposition of unit MG1, the “Cenerone” lithofacies of the bedrock is removed and subsequently deposited on the floor of the subterranean lake, present inside the proto-cavity (smaller in volume than the present day one). Regarding the origin of the protocavity, it can be realistically related to piping processes (removal of the “Cenerone” lithofacies from the bedrock), which could also be in combination with dissolution of the bedrock. However, the available data do not currently allow one to establish unequivocally how the proto-cavity was formed. Independently from the proto-cavity origin, during this first depositional stage all the eroded materials are deposited without removal from the cavity, and the balance in volume is zero or might be negative if sediments have a higher porosity in comparison to the parent weathered rock. In this case, due to the features of the “Cenerone” lithofacies, it is possible to speculate a similar porosity between the parent rock and the derived sediments. In a second stage, probably related to a lake-level fall and testified by the discontinuity between unit MG1 and MG2, a first erosive phase occurs with the subsequent removal of a little part of previously deposited sediments. This process leads to the formation of space (positive balance). The third stage, corresponding to MG2 deposition, is characterized by an input of allochthonous sediments that partially fill the cave consuming the free space (negative balance). The fourth and final stage corresponds to the last erosive event, which is the most important in terms of space development: sediments are strongly removed from the cave with a consequent volume increase of the cavity. Due to the importance of the mechanical removal of weathered rock in the formation of the underground space, the described situation shows some strong similarities with the aforementioned phantomisation theory (Vergari and Quinif, 1997; Quinif et al., 2006; Quinif, 2010). The main difference, compared to the cases described in literature, is about the origin of weathered materials, which are widespread in the bedrock, and not only located close to evident discontinuities (Häuselmann and Tognini, 2005). This is to be attributed to the intense tectonic fracturing of the Calcare Cavernoso Fm., that favoured the widespread circulation of waters. Another difference compared to the classical model is the absence of sedimentary facies related to an underground drainage network. In any case, the development of a drainage network (fluviatile phase sensu Quinif,

I. Martini / Geomorphology 134 (2011) 452–460

459

Fig. 6. Summary of the depositional history of the Mugnano cave. Stage 0 corresponds to the ancient proto-cavity. During stage 1 the “Cenerone” lithofacies is eroded from the walls and from the ceiling of the cave and deposited in a subterranean lake environment. A first phase of erosion occurs during stage 2 with the subsequent removal of part of the previously deposited sediments. During stage 3 deposition started again with an abrupt change in sedimentation testified by the arrival of sediments originating from outside the cave. Stage 4 represents the current situation: the cave is completely dry and most of the original sedimentary fill is removed by the erosion. Overall, the cycle of disintegration– deposition–erosion produces available space to the detriment of the more incoherent lithofacies of the bedrock.

2010), characterized by erosion without deposition, is to be supposed in order to explain the presence of stratigraphic unconformities in the sedimentary record. Fluviatile phases are often invoked (Quinif et al., 2006; Quinif, 2010) as the main process responsible for the removal of weathered materials and for the subsequent formation of available space. A drainage network can develop in the vadose (unsatured) zone and the flowing of water, driven by gravity and following geological structures (such as fractures, faults, rock layers and so on), can erode the floor of galleries, creating meandering canyons. This is particularly accentuated where channels develop in caves characterized by a poorly consolidated materials floor (e.g. sediments or residues of weathering). In this case, the channels strongly incise the soft substratum and the erosion is generally confined to the restricted area of the channel. This state can readily explain the formation of galleries and meandering canyons, but some problems arise concerning the origin of big chambers. The sedimentary fill of the Mugnano Cave permits understanding of the importance of “lacustrine” settings in the phantomisation theory, and above all the importance of this phase on the enlargement of the proto-cavities, up to the formation of

Fig. 7. Synthesis of the main processes active during each stage and their relative intensity. Note that during stage 1 the balance in volume is equal to zero only if the sediments have a similar porosity in comparison with the parent weathered rock. If this condition is not met, related deposits occupy a larger volume than the parent rock.

big chambers. In this context, the subterranean lake constitutes an accumulation area for bedrock-derived sediments and during this stage the passages are enlarged laterally and vertically. Unconsolidated sediments can easily be removed in the subsequent fluviatile stage, also by diffuse runoff waters. These processes require the presence of an ancestral cavity where sediments can accumulate, but not necessarily of large dimensions: theoretically the recurrence of infill-erosion cycles can generate a lot of space from a small proto-cavity. 8. Conclusion Sedimentological study of the clastic deposits of the Mugnano Cave allows a better understanding of the sedimentary processes active during the deposition. A wide range of sedimentary facies, which generally testify to deposition in a subterranean lake characterized by occasional arrivals of sediment-laden flows, has been recognized. Allostratigraphic analysis of the succession allowed division into two units, separated by an erosional surface and characterized by different depositional settings. Variations in sedimentation or the development of major erosional surfaces are generally related to processes which are directly connected to climatic change and/or tectonic uplift (e.g. landscape erosion, flooding, variations in water availability, etc.) but in this case there is no unequivocal evidence with which to distinguish between these controlling factors. The integration between an allostratigraphic approach and some considerations about sediment mineralogy allow one to draw an interesting scenario concerning speleogenesis. In this context autogenic sediments that derive from a particular lithofacies of the bedrock, play a fundamental role in the formation of available underground space, since they are, respectively, disintegrated, deposited and removed. This process is responsible for the origin of most of the space that today constitutes the Mugnano Cave. Finally, a comparison between the phantomisation theory has been attempted. This highlighted the importance of lacustrine phases on the progressive enlargement of proto-cavities into large chambers. Acknowledgments I am grateful to Dr. Mauro Aldinucci and Prof. Fabio Sandrelli for their constructive comments, and to Dr. Simone Arragoni, Vanessa

460

I. Martini / Geomorphology 134 (2011) 452–460

Martini and Sarah Cox for their help in language reviews. I would like to thank Dr. Simona Puglisi, Dr. Pietro Bartolini and all the members of Commissione Speleologica “I Cavernicoli” (C.A.I. Siena), for their assistance during fieldwork activity. Thanks to the Federazione Speleologica Toscana (F.S.T.) for the permission to publish the map of Mugnano cave. Two anonymous reviewers are thanked for their constructive comments and helpful suggestions that improved the final paper. References Audra, P., Bini, A., Gabrovšek, F., Häuselmann, P., Hobléa, F., Jeannin, P.Y., Kunaver, J., Mombaron, M., Šušteršič, F., Tognini, P., Trimmel, H., Wildberger, A., 2007. Cave and karst evolution in the Alps and their relation to paleoclimate and paleotopography. Acta Carsologica 36, 53–68. Auler, A.S., Smart, P.L., Wang, X., Pilò, L.B., Edwards, R.L., Cheng, H., 2009. Cycling sedimentation in Brazilian caves: mechanisms and palaeonvironmental significance. Geomorphology 106, 142–153. Bartolini, C., 2003. When did the Northern Apennine become a mountain chain? Quaternary International 101–102, 75–80. Bosch, R.F., White, W.B., 2004. Lithofacies and transport of clastic sediments in karstic aquifers. In: Sasowsky, I.D., Mylroie, J.E. (Eds.), Studies of Cave Sediments. Kluwer Academic/Plenum Publisher, New York, pp. 1–22. Brandy, J.E., Scott, A., 1997. Excavations in buried cave deposits: implications for interpretation. Journal of Cave and Karst Studies 59, 15–21. Brogi, A., 2003. Megaboudinage della Falda Toscana nell'area geotermica del M. Amiata e relazioni con la tettonica distensiva miocenica dell'Appennino Settentrionale. Geo Acta, Special Issue 2, 33–36. Brogi, A., 2004. Miocene extension in the Inner Northern Apennines: the Tuscan Nappe megabuddins in the M. Amiata geothermal area and their influence on Neogene sedimentation. Bollettino della Società Geologica Italiana 123, 513–529. Bull, P.A., 1981. Some fine-grained sedimentation phenomena in caves. Earth Surface Processes and Landforms 6, 11–22. Burger, D., 1989. Dolomite weathering and micromorphology of paleosoils in the Franconian Jura. Catena Supplement 15, 261–267 Cremlingen. Campy, M., Chaline, J., 1993. Missing records and depositional breaks in French Late Pleistocene Cave sediments. Quaternary Research 40, 318–331. Capezzoli, E., Priori, S., Costantini, E.A.C., Sandrelli, F., 2009. Stratigraphic and paleopedological aspects from the Middle Pleistocene continental deposits of the southern Valdelsa Basin. Bollettino della Società Geologica Italiana 128, 395–402. Capezzuoli, E., Sandrelli, F., 2004. I sedimenti quaternari del settore meridionale della Valdelsa (Provincia di Siena). Italian Journal of Quaternary Sciences 17, 33–40. Carmignani, L., Kligfield, R., 1990. Crustal extension in the northern Apennines: the transition from compression to extension in the Alpi Apuane core complex. Tectonics 9, 1275–1303. Carmignani, L., Decandia, F.A., Disperati, L., Fantozzi, P.L., Lazzarotto, A., Liotta, D., Oggiano, G., 1995. Relationships between the Sardinia–Corsica–Provençal Domain and the Northern Apennines. Terranova 7, 128–137. Carmignani, L., Decandia, F.A., Disperati, L., Fantozzi, P.L., Kligfield, R., Lazzarotto, A., Liotta, D., Meccheri, M., 2001. Inner Northern Appennines. In: Vai, G.B., Martini, I.P. (Eds.), Anatomy of an Orogen: The Appennines and Adjacent Mediterranean Basins. Kluwer Acad., Publ., pp. 197–214. Collinson, J.D., Thompson, D.B., 1982. Sedimentary structures. London, U.K, Allen & Unwin. 194 pp. Costantini, A., Decandia, F.A., Lazzarotto, A., Liotta, D., Mazzei, R., Pascucci, V., Salvatorini, G., Sandrelli, F., 2009. Note illustrative della Carta Geologica d'Italia alla scala 1:50.000: Foglio 296-Siena. APAT-Roma. Tip. A.T.I., 129 p. De Waele, J., Plan, K., Audra, P., 2009. Recent developments in surface and subsurface karst geomorphology: an introduction. Geomorphology 106, 1–8. Elter, F.M., Sandrelli, F., 1994. La fase post-nappe nella Toscana meridionale: nuova interpretazione sull'evoluzione dell'Appennino settentrionale. Atti Ticinesi Scienze della Terra 37, 173–193. Fabrizi, F., 1962. Map of the Mugnano Cave (T/SI 258). Tuscan register of caves, Federazione Speleologica Toscana (F.S.T.)Available on line to www.speleotoscana.it. Fornós, J.J., Ginés, J., Gràcia, F., 2009. Present-day sedimentary facies in the coastal karst caves of Mallorca island (western Mediterranean). Journal of Cave and Karst Studies 71, 86–99. Gandin, A., Giamello, M., Guasparri, G., Mugnaini, S., Sabatini, G., 2000. The calcare cavernoso of the Montagnola Senese (Siena, Italy): mineralogical-petrographic and petrogenetic features. Mineralogica et petrographica Acta 43, 271–289. Ghinassi, M., Colonese, A.C., Di Giuseppe, Z., Covoni, L., Lo Vetro, D., Malavasi, G., Martini, F., Ricciardi, S., Sala, B., 2009. The Late Pleistocene clastic deposits in the Romito Cave, southern Italy: a proxy record of environmental changes and human presence. Journal of Quaternary Science 24, 383–398. Giannini, E., Lazzarotto, A., 1967. Sulla giacitura, origine ed età della coltre di “Cavernoso” alla Montagnola Senese. Atti della Società Toscana di Scienze Naturali, Memorie 74, 63–77. Giannini, E., Lazzarotto, A., 1970. Studio geologico della Montagnola Senese. Memorie della Società Geologica Italiana 9, 451–495.

Harms, J.C., Southard, J.B., Spearing, D.R., Walker, R.G., 1975. Depositional environments as interpreted from Primary Sedimentary Structures and Stratification Sequences. SEPM Short Course No. 2, Lecture Note. Society of Economic Paleontologists and Mineralogists, Tulsa, OK. Harms, J.C., Southard, J.B., Walker, R.G., 1982. Structures and sequences in clastic rocks. SEPM Short Course No. 9, Lecture Note. Society of Economic Paleontologists and Mineralogists, Tulsa, OK. Häuselmann, P., Tognini, P., 2005. Kaltbach cave (Siebenhengste, Switzerland): phantom of the sandstone? Acta Carsologica 34, 383–396. Hill, C.A., 1999. Sedimentology and paleomagnetism of sediments, Kartchner Caverns, Arizona. Journal of Cave and Karst Studies 61, 79–83. Hunt, C., Davison, J., Inglis, R., Farr, L., Reynolds, T., Simpson, D., el-Rishi, H., Barker, G., 2010. Site formation processes in caves: the Holocene sediments of the Haua Fteah, Cyrenaica, Lybia. Journal of Archaeological Science 37, 1600–1611. Jolivet, L., Daniel, J.M., Fournies, M., 1991. Geometry and kinematics of extension in Alpine Corsica. Earth and Planetary Letters 104, 278–291. Klimchouk, A., 2009. Morphogenesis of hypogenic caves. Geomorphology 106, 100–117. Lugli, S., 2001. Timing of post-depositional events in the Burano Formation of the Secchia Valley (Upper Triassic, northern Apennines) clues from gypsum-anhydrite transition and carbonate metasomatism. Sedimentary Geology 140, 107–122. Lugli, S., Morteani, G., Dominique, B., 2002. Petrographic, REE, fluid inclusion and stable isotope study of magnesite from the Upper Triassic Burano Evaporites (Secchia Valley, northern Apennines): contributions from sedimentary, hydrothermal and metasomatic sources. Mineralium Deposita 37, 480–494. Martini, I., 2007. Analisi di facies dei depositi terrigeni di grotta: nuove possibilità di studi geo-speleologici. TALP 35, 22–27. Merla, G., 1952. Geologia dell'Appennino settentrionale. Bollettino Società Geologica Italiana 70, 95–382. Nemec, W., Kazanci, N., 1999. Quaternary colluvium in west-central Anatolia: sedimentary facies and paleoclimate significance. Sedimentology 46, 139–170. Pascucci, V., 2004. Karst and palaeonvironment of the Siena area (Central Italy). In: Guerrieri, L., Rischia, I., Serva, L. (Eds.), 32nd International Geological Conference, Florence, Italy, 20–28 Agosto 2004: Field trip guide book, 3-DO2, pp. 3–16. Pascucci, V., Bianciardi, G., 2001. Geologia e morfologia: una serie di modesti rilievi caratterizzati da una struttura complessa e da un notevole sviluppo di fenomeni carsici. In: Manganelli, G., Favilli, L. (Eds.), La Montagnola Senese, una guida naturalistica. WWF, Italia, pp. 11–23. Passeri, L., 1975. L'ambiente deposizionale della formazione evaporitica nel quadro della paleogeografia del Norico Tosco-Umbro-Marchigiano. Bollettino della Società Geologica Italiana 94, 231–268. Passeri, L., 1979. The hypothesis of a northern Apenninic Peninsula during the upper Triassic. Memorie della Società Geologica Italiana 20, 151–161. Piccini, L., Mecchia, M., 2009. Solution weathering rate and origin of karst landforms and caves in the quartzite of Auyan-tepui (Gran Sabana, Venezuela). Geomorphology 106, 15–25. Pickering, R., Hancox, P.J., Lee-Thorp, J.A., Grün, R., Mortimer, G.E., McCulloch, M., Berger, L.R., 2007. Stratigraphy, U–Th chronology, and paleoenvironments at Gladysvale Cave: insights into the climatic control of South African homininbearing cave deposits. Journal of Human Evolution 35, 602–619. Quinif, Y., 2010. Fantômes de roche et fantômisation. Karstologia Mémories, 18. 196 pp. Quinif, Y., Meon, H., Yans, J., 2006. Nature and dating of karstic filling in the Hainaut Province (Belgium). Karstic, geodynamic and paleogeographic implications. Geodinamica Acta 19, 73–85. Sasowsky, I.D., Mylroie, J.E., 2004. Studies of Cave Sediments. Kluwer Academic/Plenum Publisher, New York. 329 pp. Signorini, R., 1946. Cenni preliminari su un rilevamento nella Val di Merse. Bollettino della Società Geologica Italiana 65, 29–31. Signorini, R., 1966. I terreni neogenici del Foglio “Siena”. Bollettino della Società Geologica Italiana 85, 639–654. Signorini, R., Pieruccini, U., 1968. La cava di gesso di Montisi e la serie ridotta nella Valle del Merse a sud di Siena. Bollettino della Società Geologica Italiana 87, 157–170. Springer, G.S., 2005. Clastic sediments in caves. In: Culver, D., White, W. (Eds.), Encyclopedia of Caves. Elsevier, pp. 102–108. Springer, G.S., Kite, J.S., 1997. River-derived slackwater sediments in caves along Cheat River, West Virginia. Geomorphology 18, 91–100. Springer, G.S., Kite, J.S., Schmidt, V.A., 1997. Cave sedimentation, genesis, and erosional history in the Cheat River Canyon, West Virginia. GSA Bulletin 109, 524–532. Terzaghi, K., Peck, R.B., Mesri, G., 1996. Soil Mechanics in Engineering Practice, 3rd Edition. Wiley-Interscience. 549 pp. Tognini, P., 1999. The Mt. Bisbino (Northern Italy) karst: a new speleogenetic processes. Etudes de géographie physique 28, 185–190. Trevisan, L., 1955. Il Trias della Toscana e il problema del Verrucano triassico. Memorie della Società Toscana Scienze Naturali 62, 1–30. Vergari, A., Quinif, Y., 1997. Les paléokarsts du Hainaut (Belgique). Geodinamica Acta 10, 175–187. Walker, R.G., 1992. Facies, Facies models and modern stratigraphic concepts. In: Walker, R.G., James, N.P. (Eds.), Facies Models: Response to Sea Level Change. Geological association of Canada, St John's, pp. 1–14. White, W.B., 2007. Cave sediments and paleoclimate. Journal of Cave and Karst Studies 69, 76–93. Zupan Hajna, N., 2003. Chemical weathering of limestone and dolomites in a cave environment. Speleogenesis and Evolution of Karst Aquifers 1, 1–6.