1 Introduction

In a world where the demand for fossil and renewable energy continues to rise, reservoir characterization and modeling is an essential step for the better management of resources. The reservoir properties of carbonates have been well studied in the past but still represent a challenge as limestones are strongly affected by diagenetic alteration. The heterogeneity of reservoir properties in carbonate rocks (e.g., Westphal et al. 2004; Davis et al. 2006; Dou et al. 2011) can be explained by (1) the fracturation leading to non-matrix flow network and (2) the complex diagenetic path of the reservoir (Walls and Burrowes 1985; Machel 1987; Mountjoy and Marquez 1997; Wilson and Evans 2002; Ehrenberg and Nadeau 2005; Rong et al. 2012). It is therefore difficult to understand and to predict the distribution of petrophysical properties in carbonate reservoirs. The GEothermy 2020 (GEo2020) program aims at exploring and eventually exploiting the deep geothermal resources of the Geneva Basin (GB). Based on a 3-D geological model of the Greater Geneva Basin combined with petrophysical data (Clerc et al. 2015), five units were targeted as potential reservoirs in the Lower Triassic (Bundsandstein and Muschelkalk), the Middle Jurassic (Bajocian), the Upper Jurassic (Kimmeridgian), and the Early Cretaceous (Barremian). This study focuses on the Upper Jurassic deposits well characterized in the past by numerous sedimentological and stratigraphic studies (e.g. Deville 1988, 1990; Fookes 1995; Charollais et al. 1996, 2013; Mouchet 1998; Gygi 2013; Rameil 2008; Strasser et al. 2015). The Upper Jurassic and, specifically, the Kimmeridgian limestones are known to be affected by dolomitization (Fondeur et al. 1954). Dolomitization is commonly associated with important modifications of the reservoir properties by poro-genesis or poro-necrosis mechanisms (Schmoker et al. 1985; Braithwaite 1991; Giorgioni et al. 2016). However, the paragenesis of the Kimmeridgian is poorly constrained in the GB, and dolomitization processes remain not well understood. The dolomitization model currently available is the one by Fookes (1995), following the interpretation made in Bernier (1984), and which proposed a mixing-zone (i.e., the mixing of phreatic seawater with meteoric waters) model to explain the dolomitization in this area. The mixing-zone dolomitization scenario has been the subject of many discussions among the scientific community in the last decades, mainly due to the fact that dolomitization depends on the stoichiometry of the reaction, the temperature, and the fluid composition. It is now accepted that the requirement in terms of volume and Mg concentration of fluid necessary for a mixing-zone dolomitization of large sedimentary bodies cannot be met (Warren 2000; Machel 2004).

Based on field analogs and subsurface data, the present study aims to provide a paragenetic framework for the Kimmeridgian in the GB and to discuss the potential dolomitization scenarios. Dolomitic fabrics and dolomitization processes from time-equivalent deposits reported in the literature are reviewed and compared with results from the present study. Furthermore, the impact of dolomitization on reservoir quality in the Upper Jurassic of the GB is evaluated with respect to possible future geothermal exploration of the GB.

2 Geological context and study sites

2.1 Geological setting

The Greater Geneva Basin is located at the Swiss-French transnational zone in the south-west of Lake Geneva. The basin is limited to the north-east by the internal chain of the Jura Mountain and to the south-east by the front of the Alpine units (Fig. 1). The basin has a Variscan crystalline basement covered by 3000–5000 m thick Mesozoic to Cenozoic successions. The Mesozoic series are mainly composed of carbonates and marls, along with evaporites. The Triassic is marked by a marine transgression during which the basin was connected to the Tethys Ocean, leading to the deposition of thick evaporitic series (Disler 1914). While the Lower Jurassic sediments are still influenced by the marine transgression, two successive regressive trends affect the Middle and Upper Jurassic series. Open-marine conditions prevailed from the Hettangian until the Toarcian, shifting to shallower conditions during the Bajocian and Bathonian. A shallow carbonate platform developed during the Upper Jurassic, extending towards the north-west until the early Cretaceous. During the Kimmeridgian patch reefs developed on top of pre-existing structural highs (Meyer 2000). The sealing of inter-reef depressions by prograding tidal deposits followed during the Tithonian, with the local occurrence of immersive facies (Strasser 1994).

Fig. 1
figure 1

Modified from Charollais et al. (2013)

Simplified geological map of the study area, with the location of the study sites, and a synthetic log of the Upper Jurassic units as exposed in Mount Vuache

Full size image

The Kimmeridgian and Tithonian sequences of the GB can be subdivided into four major units: the Couches à Céphalopodes, the Calcaires de Tabalcon, the Complexe Récifal, and the Tidalites de Vouglans (Fig. 1). The Complexe Récifal is subdivided in three subunits: the Calcaires Récifaux, Calcaires Plaquetés, and Calcaires de Landaize subunits. The Couche à Céphalopodes unit was deposited in a deep open marine environment and is mainly represented by marly limestones with a thickness of 100 m. The unit consists of meter-thick interbeds of fine grey to beige limestones and marly beige limestones. The fauna consists of ammonites, belemnites, and thin bivalves. This unit is dated as early to late Kimmeridgian (Enay 1969; Bernier and Enay 1972; Bernier 1984). The thin-bedded limestones consist of mudstone rich in organic matter and pyrite, along with rare oncoids and glauconite. Bioclasts are composed of echinoderms (echinoids, ophiuroids, holothurians), sponge spicules, bivalves, and serpulids. The Calcaires de Tabalcon unit is characterized by several decimeter-thick beds, about 20-m thickness, exhibiting discrete bedding. The facies is coarsening up with fine micritic limestones including sparse bioclasts at the base to a tight bioclastic limestone with debris originating from the dismantlement of a carbonate platform with corals, diceratids, pectinids, rare bivalves, gastropods, and crinoid ossicles. Locally, silica can be observed, likely remains of sponges. The upper part of the Calcaires de Tabalcon is affected by dolomitization. The Complexe Récifal forms a discontinuous white unit of variable thickness (Deville 1988). In the Vuache Mountains, Charollais et al. (2013) estimate a maximum thickness of about 200 m. The Calcaire Récifaux unit consists of a compact white limestone with facies associations typical for coral dominated reef environment: bindstones (including intervals rich in microbialites), mudstones, and rudstones. The Calcaire Plaquetés unit consists of thin-laminated beds rich in bitumen, gypsum, halite, and dolomite. Laterally, this unit varies significantly in thickness reaching up to 200 m to compensate depressions between reef bodies of the Calcaire Récifaux unit. The Calcaire Plaquetés unit was deposited in inter-reefal settings and could fill up all available space between the Calcaire de Tabalcon and the Tidalites de Vouglans formations. The Calcaire de Landaize unit is composed of bioclastic grainstone including coral fragments, stromatopores, echinoderms, and gastropods corresponding to a shallow lagoon with high-energy conditions. This unit consists of 0.2–2.5 m-thick beds with a maximum thickness up to 15 m. The Tidalites de Vouglans unit is composed of less than a meter thick limestone beds and marly to dolomitic interbeds. The dominant facies is a white to grey mudstone rich in pyrite, locally including microbial mats and mud cracks indicative of a foreshore tidal environment (Strasser 1994).

During the Early Cretaceous, overall shallow and warm-water conditions prevailed (Debelmas and Michel 1961; Sommaruga 1997; Charollais et al. 2013) leading to the deposition of bioclastic limestones, bioturbated limestones, and organic-rich marls. Upper Cretaceous deposits are not recorded in the GB, probably due to the late Cretaceous emersion and later erosion during the Early Cenozoic. This erosion led to the development of an important karstic system that was filled during the Eocene by the so-called “sidérolithique”, a red sandstone formation (Debelmas and Michel 1961). The Alpine orogeny and the associated development of the foreland basin induced the deposition of a thick Oligocene–Miocene detritic Molasse unit (Favre et al. 1880; Heim 1922; Charollais et al. 2007).

2.2 Study sites

The Kimmeridgian units in the GB were studied in one borehole and several outcrops across the basin (Fig. 1). Sedimentologic logs of the Saleve, Reculet, and Prapont sections as well as the Humilly-2 well are summarized in Fig. 2. In the south-eastern part of the basin, the Calcaire de Tabalcon unit was studied along the Etiollets section at the Salève Mountain (Fig. 2a). This outcrop was described by Deville (1988, 1990). Field data and sedimentologic interpretations were published in previous works (Joukowsky and Favre 1913; Carozzi 1950, 1954, 1955). Deville (1988) described four different facies: one micritic, two bioclastic and one dolomitized facies. The northern part of the basin is studied in several outcrops of the Folded Jura: Le Reculet Nord, Morillon, Valefin, Roche Blanche, and Le Col de la Faucille. The Reculet Nord section starts with the top of the Calcaires de Tabalcon unit (Fig. 2b) overlain by the Calcaires Récifaux unit (Fig. 2b). This section was previously described by Meyer (2000). The Rocher du Morillon is a succession of about 70 m of tectonically verticalized to sub-verticalized limestone beds, marly limestones, and marl beds in an anticlinal structure called Les Planches. Approximately 20 km south of Morillon, along the road D437, following the Foulasse and Bienne rivers, the Calcaires Récifaux, Calcaires de Landaize, and Tidalites de Vouglans units outcrop in Valefin Lès Saint-Claude and Roche Blanche (Fig. 2c). The Calcaires de Landaize outcrop in the Col de la Faucille section, a mountain pass above the city of Gex accessible by the N5 road in the direction of Les Rousses.

Fig. 2
figure 2

Sedimentologic logs for: a the Calcaires de Tabalcon unit the Salève Mountains, modified from Deville (1990). b The top of the Calcaires de Tabalcon and the Calcaires Récifaux in the Reculet Nord section, modified after Meyer (2000). c the Calcaires Récifaux, Calcaires de Landaize and Tidalites de Vouglans (TdV) unit in the Roche Blanche and Valefin sections. d The reef-front and reef-core sections of the Calcaires Récifaux in the Prapont section, modified from Fookes (1995). e Description of the 5 meters core available in the Humilly-2 well, representing the top of the Calcaires Récifaux unit. Stars indicate collected samples with sample labels

Full size image

The Calcaires Récifaux unit outcrops in the western part of the basin in Prapont and Champfromier. The Prapont section was well described by several authors (Pelletier 1953; Enay 1965; André 1962) and more recently in Fookes (1995). The outcrops are characterized from its base to the top by subtidal deposits followed by reef front deposits, a first reef sequence, inter-reef sands making the transition towards a second reef sequence on which lagoonal and subtidal storm deposits are found (Fookes 1995). Sampling focused on the reef front and first reef sequence, which is affected by dolomitization, and include numerous vuggy pores (Fig. 2d). Ten kilometers east from the Prapont section, the Champfromier section can be studied along the road D14 towards the village of Forens. The outcrop, described by Bernier (1984), consists of thick limestone beds with a regular dip of about 20° towards the east.

Subsurface data originates from the Humilly-2 well (Fig. 1) located in the center of the GB in the French department of Haute-Savoie. This well currently serves as a reference for the subsurface characterization of the GB (Clerc et al. 2015; Moscariello 2016; Rusillon et al. 2016). Stratigraphical and lithological control on the sedimentary units of HU-2 is provided by the outcrops in the Jura Mountains, Mount Vuache, and Mount Salève (from the Upper Triassic to the Quaternary). Only one core section including the Kimmeridgian was retrieved from 1021 to 1015 m, and is described in Fig. 2e. No sedimentological logs are drawn for the Morillon, Col de la Faucille, and Champfromier sections as they display virtually no variations in terms of facies.

3 Sampling strategy and analytical methods

At Mount Salève, 14 samples were taken along the path exposing the Calcaires de Tabalcon unit. Sample selection was tied to data from the stratigraphic section measured by Deville (1988) (Fig. 2a). In Champfromier, seven samples were taken along the outcrop. In the Folded Jura, 16 samples were studied: two in Morillon, three in Valefin, three in Roche Blanche, two in the Col-de-la-Faucille area, and seven in the Reculet Nord (Fig. 2b, c). In the Prapont section, seven samples were taken: three in the dolomitized reef front deposits and four in the first reef sequence following the scheme of Fookes (1995; Fig. 2d). Ten samples were taken from the HU-2 well, every 50 cm (Fig. 2e), in addition to those originating from the study by Rusillon et al. (2016). Core samples were taken directly from the core as plugs with a diameter of 25 mm. Samples from outcrops and cores were prepared for thin section analysis. Thin sections, impregnated with epoxy resin stained by Methylene blue, were used to define the texture, grain type, bioclast content, grain size, mineralogy, cement type, and pore-type distribution. Cathodoluminescence analysis was completed using an ERI-MRTech-optical cathodoluminescence microscope with a cold cathode mounted on an Olympus BX41 petrological microscope (Department of Earth Sciences, University of Geneva, Switzerland). The beam conditions were 15–18 kV at 120–200 mA with an unfocused beam of approximately 1 cm. Carbon coating (ca. 15 nm) by carbon thread evaporation was used prior to imaging with the Jeol JSM 7001F Scanning Electron Microscope (SEM, Department of Earth Sciences, University of Geneva, Switzerland). Semi-quantitative analyses and mapping were obtained with an EDS detector coupled with the JED 2300 software. Calcite staining (Dickson 1966) was performed to constrain the cement mineralogy and the paragenesis.

4 Petrography

4.1 Calcaires de Tabalcon unit

The Calcaires de Tabalcon displays four distinct sedimentological facies in the Etiollets section. The first facies is micritic with the accumulation of micropeloids and a faunal association typical for open-marine, low-energy, outer shelf environments. This facies is intensively dolomitized. Blocky calcite displays a dull luminescence with zonations (Fig. 3a, b). Dolomite is composed of fine to medium, euhedral planar-porphyrotopic, replacive rhombs. Crystals are generally displaying a cloudy core and a clear rim. The dolomite has a non-luminescent dark center surrounded by dull luminescent red zones toward the borders (Fig. 3a, b). The dolomite rhombs exhibit selective dissolution and calcitization along their core/rim interface (Fig. 3d), associated with intracrystalline microvugs. Dolomite rhombs are partly dissolved along younger stylolites (Fig. 3b). The second facies is bioclastic with wackestone and packstone to grainstone textures. The third one is lithoclastic. In both facies, two phases of calcite cements are observed: (1) syntaxial overgrowth around echinoderm fragments and (2) blocky calcite cementation filling the available inter-particular pore space. Under CL, the blocky calcite shows a dull luminescence with orange to brown zonations. Both facies are often affected by fractures sealed by sparitic calcite exhibiting a bright yellow to brown luminescence. Dolomite rarely occurs in these facies. The bindstone facies is composed of siliceous sponges including pyrite-limonite and is characterized by two stages of blocky calcite cementation (Fig. 3e–f): the first displaying a non-luminescent core with yellow bright luminescent zonations and the second marked by a bright yellow luminescence with zonations under CL. Two stages of dolomitization can be distinguished (Fig. 3e–f): (1) fine idiotopic, euhedral clear rhombs displaying a red bright luminescence and intersecting the blocky calcite cementations (dolomite 1) and (2) fine to medium euhedral to subhedral rhombs with cloudy cores and displaying a mauve bright luminescence, intersecting the first stage (dolomite 2). Both exhibit thin zonations under CL. Both stages of dolomites exhibit dedolomitization by calcitization. The dolomite is intersected by micro-fractures filled by calcite that is yellow bright luminescent under CL.

Fig. 3
figure 3

Photomicrographs (optical microscope and cathodoluminescence) of the Calcaires de Tabalcon unit. a, b Dolomitized and dedolomitized mudstone and late stylolithization in natural light and cathodoluminescence. Blocky calcite (BC) exhibits dull luminescence under CL (sample YM32). c Bioclast affected by compaction exhibiting two stages of blocky calcite cementation (BC1 and BC2, sample YM34). d SEM photograph of dedolomite composed of remaining dolomite and microvugs filling calcite (sample YM77). e, f Planar euhedral dolomite overprinting blocky calcite cementation (BC), in natural light and cathodoluminescence (sample YM40). g, f Sucrosic dolomite with cloudy cores and limpid rims, natural light and cathodoluminescence (sample YM61)

Full size image

In the Reculet Nord section, the Calcaires de Tabalcon is characterized by a dolomitized micritic limestone consisting of medium to coarse, euhedral, highly coalescent replacive rhombs that obliterated the initial fabric. Most of the rhombs display a dark cloudy core and clear outer rim (Fig. 3g). Under CL, the rhombs display a thick dull luminescent orange to brown-zoned core evidencing dissolution and mauve to pink dull luminescent zoned rims (Fig. 3h). The zoned rims form syntaxial overgrowth on crystal facing pores.

4.2 The Reef Complex unit—Calcaires Récifaux

The Upper Jurassic is about 800 m thick (non-published final drilling report of the HU-2 well) and can be subdivided in two units: a lower marly limestone unit (from 1855.6 to 1644 m) and an upper limestone unit (from 1644 to 812 m). The upper limestone unit includes the interval dated as Kimmeridgian (from 1271 to 846 m) with an unidentified lower limit. Several lithologies were observed from the base to the top: 153 m of beige limestone with evidences of sucrosic dolomite; 73 m of crystalline to micro-crystalline white limestone; 36 m of white chalky limestone; 60 m of beige oolitic grainstone and dolomitic limestone and 103 m of white limestone, often chalky, coarse, and oolitic. The Calcaires Récifaux unit in HU-2 consists of a white bioclastic limestone composed of gastropods, brachiopods, corals, echinoderms, and foraminifers. The majority of the core is composed of bindstones, rudstone, and grainstone in the first meter at the top of the core. Centimetric vugs are mostly associated with the presence of corals. Using CL, several stages of calcite cementation were identified in thin sections: (1) an isopachous cement rim around allochems, (2) syntaxial overgrowth around echinoderms, (2) three different stages of blocky calcite cementation filling the available inter-particular pore space (Fig. 4a, b). The first stage of blocky calcite cementation exhibits a dull brown luminescence under CL followed by a second mainly non-luminescent stage with bright yellow zonations. The third stage of blocky calcite cementation shows a dull to bright luminescence with orange to brown colors. Locally, this late stage of cementation is incomplete with intra-crystalline pores preserved (Fig. 4b). Only one sample (1020.5 m) displays dolomite after calcite staining.

Fig. 4
figure 4

Photomicrographs (optical microscope and cathodoluminescence) of the Calcaires Récifaux unit. a, b Multi-phased incomplete blocky calcite cementation leaving intra-crystalline porosity in the HU-2 well (sample YM49). c, d Dedolomitization as observed in the Prapont section (sample YM55). e, f Replacement planar dolomite overprinting pre-existing blocky calcite cementation in the Roche Blanche section (sample YM8). g Intensively fractured micritic limestone in the Champfromier section. The calcite filling the fracture exhibits thin twinning (sample YM14). f The same sample in natural light (top left) and under cathodoluminescence displaying a dissolution enhanced fracture filled by non-luminescent calcite

Full size image

In the Prapont section, the reef front deposits exhibit beige wackestone with microsolenoid corals, stromatoporoids, bryozoans, and rare foraminifers, along with oncoids and micropeloids associated with microbial crusts. Lateral variations at small scale are important in these deposits with locally dolomitized patches. Some of these patches are intensively bioturbated. Moldic and intergranular porosity as well as cm- to dm-size vuggy pores are common. Samples from the reef front consist of micritic limestone with large sparitic cements and rhombohedral pores (Fig. 4c, d). Under CL, the sparitic calcite displays a dull orange to brown luminescence with zonations. The rhombohedral pores are partially filled by de-dolomite with a dull to bright red luminescence (Fig. 4c, d). The Valefin and Roche Blanche outcrops consist of a set of bioherms at the base that are more resistant to erosion. These bioherms are composed of white limestone units rich in rudists and corals, separated by an oolitic grainstone with occurrences of nerinids and rudists. This grainstone is cemented at various degree and can exhibit partial interparticular moldic porosity. Samples exhibit rare vugs filled by blocky calcite cements (Fig. 4e) displaying a dull luminescent core followed by bright luminescent orange zones. Two stages of dolomitization can be distinguished (Fig. 4e, f): (1) very fine euhedral, pervasive replacive clear rhombs and (2) medium euhedral, pore ceiling, clear rhombs with cloudy cores and displaying a mauve bright luminescence, intersecting and surrounding vug-filling blocky calcite cements. Both exhibit a bright to dull luminescent yellow to orange-zoned core transitioning to thin red luminescent-zoned rimes (Fig. 4f).

The Calcaires Récifaux unit in the Champfromier section is characterized by micritic limestone including rare fragments of brachiopods, echinoderms, bivalves, and green algae. The limestone is intensively affected by fractures and micro-fractures filled by sparitic calcite displaying twining. Under CL, the calcite infilling of fractures is characterized by an orange to brown dull luminescence with zonations. This stage is crosscut by a second generation of dissolution-enhanced fractures filled by orange luminescent sparitic calcite (Fig. 4h). Non-planar anhedral dolomites are observed (Fig. 4g, h) and appear non-luminescent under CL (Fig. 4h).

In the Reculet Nord, the Calcaires Récifaux unit consists of a first reef sequence of about 10 m thickness, which is dolomitized throughout and characterized by an emersion surface at its top. This first sequence is followed by about 15 m of grainstone rich in gastropods, bivalves, and green algae on top of which a second reef sequence was deposited. The oolitic grainstone on top of the first reef sequence (Fig. 5a) is characterized by an important, partially replacive dolomitization affecting the interparticular space. These dolomite crystals are characterized by medium to coarse euhedral rhombs partly replacing the grains. Under CL (Fig. 5b), the dolomite displays a thick bright luminescent yellow to brown zoned core characterized by dissolution and followed by a bright luminescent red zoned core. Thick bladed isopachous cement rims are present around allochems (Fig. 5c) and a sparitic to micro-sparitic blocky calcite cement fills the interparticular space. Under cathodoluminescence, two stages of blocky calcite cementation can be distinguished: (1) a bright luminescent, orange zoned one and (2) a dull luminescent little to non-zoned one (Fig. 5d). Echinoderms fragments are surrounded by inclusion-rich, syntaxial overgrowth (Fig. 5e, f).

Fig. 5
figure 5

Photomicrographs (optical microscope and cathodoluminescence) of the Calcaire Récifaux and Calcaires de Landaize units. a, b Grainstone facies, as observed in the Reculet Nord section, showing replacement dolomite affected by calcitization with calcitized cores appearing orange and dolomite rims appearing red under CL (sample YM63). c, d grainstone facies in the Valefin section showing isopachous cement rim (IC) of allochems. The residual space is filled by blocky calcite (BC) cementation (sample YM2). e, f the same facies in the Roche Blanche section showing isopachous cementation and large syntaxial overgrowth around an echinoderm bioclast (sample YM6). g, h Calcaires de Landaize oolitic to oncoidic grainstone exhibiting important pressure-dissolution contact due to compaction. Under CL, the interparticular pore space is filled by blocky calcite cementation cross-cut by rhomboedral moldic pores due to dedolomitization (sample YM5)

Full size image

4.3 The Reef Complex units—Calcaires de Landaize

In all studied sections, this unit is characterized by limestone beds composed of a beige oncoid-rich grainstone with sparitic calcite cementation. The samples from Col de la Faucille, Valefin, and Roche Blanche sections share the same diagenetic features: (1) isopachous cement rims around allochems, syntaxial overgrowth around echinoderms fragments, and two successive stages of blocky calcite cementation. No dolomite was observed in these sections. The Morillon section differs from the other sections in that isopachous cements are absent, grain interpenetration with pressure-dissolution sutures is common, and rhombohedral interparticular moldic pores are widespread (Fig. 5g, h).

Overall, the Calcaires de Tabalcon unit displays three dolomite phases. In the Etiollet section, most of the unit is firstly characterized by fine to medium replacive, euhedral dolomite followed by a fine to medium, euhedral to subhedral dolomite. In the Reculet Nord section, a third stage of medium to coarse, planar-euhedral, fabric-destructive dolomite is observed. Both the first and third stages display dedolomitization by calcitization. The Calcaires Récifaux unit is also affected by three distinct stages of dolomite. The two first phases are similar to those previously described in the Etiollet section. The third stage, present only in the grainstone facies in the Reculet Nord section is characterized by medium to coarse, euhedral dolomite partially replacing the margins of carbonate grains. Dedolomitization, either as calcitization or as dissolution of dolomite crystals, led to porogenesis in the Calcaires Récifaux unit. The Calcaires de Landaize unit is affected only by one stage of dolomitization in form of fine euhedral replacive dolomite associated to microstylolites located at the grain contact. Dedolomitized grains are common, resulting in rhomb-shaped moldic pores.

5 Paragenesis

5.1 Paragenesis of the Calcaires de Tabalcon unit

The paragenesis of this unit is the following (Fig. 6a): Micritization affecting skeletal fragments and particles as the earliest stage recorded. It is interpreted to occur mainly in shallow-marine environments and directly on, or just below, the seafloor (Bathurst 1966; Alexandersson 1972; Land and Moore 1980). Moldic porosity created by organic matter degradation and aragonite dissolution is filled by drusy calcite crystals. Syntaxial overgrowth, rarely observed, is limited to the bioclastic facies of the Salève site. The idiotopic texture of the dolomites with well-visible rhombs is indicative of dolomitization occurring during the “early” stages of diagenesis, at very shallow burial, with temperatures below 50 °C (Gregg and Sibley 1984). Overall, this replacement dolomite exhibits cloudy cores surrounded by limpid rims indicating that the cores developed from fluids evolving from a state of near saturation with low-Mg calcite in the early stages to a state of undersaturation (Nader et al. 2007; Sibley 1980; Warren 2000). This early dolomitization is followed by a first stage of blocky calcite cementation (BC1) which, under CL, displays zonation indicating chemical modifications of the parent fluids during crystallization (Meyers 1974; Machel 1985). A first stage of compaction affected this unit as evidenced by the compaction features affecting skeletal fragments (Fig. 3c), indicating that the moldic porosity was not entirely filled during compaction. Then the second stage of blocky calcite cementation (BC2) filled the remaining pore space when available. BC2 shows zonation under CL, which could indicate minor modification of the chemical composition of the fluid during precipitation or be the results of changing redox conditions. Therefore, the blocky calcite cementation took place in the open porosity during mesogenesis and burial. A stage of calcite-filled micro-fractures with bright yellow luminescence under CL affects all stages described so far. The sucrosic euhedral dolomite observed in the Reculet Nord represents an advanced stage of replacement which obliterated the original texture. Most of the dolomite is affected by calcitization associated with intercrystalline microvugs (Fig. 3d). This process could be driven by the migration of Ca-rich water through the unit, which led to a beginning of dedolomitization by partial dissolution and later infilling of created porosity by calcite cement in a two-step process (Ayora et al. 1998). A late stage of compaction associated with calcite-filled fractures crosscuts all previous stages (Fig. 6a).

Fig. 6
figure 6

Paragenesis of the Calcaires de Tabalcon unit (a), Calcaires Récifaux unit (b) and Calcaires de Landaize unit (c). Of the three units, the Calcaires Récifaux is the only one displaying a third stage of incomplete blocky calcite cementation. While all units were affected by dedolomitization, the degree of this process varies with only the Calcaires Récifaux and Calcaires de Landaize units exhibiting complete dedolomitization leading to the creation of secondary pore space. Asterisk indicate in the Calcaires de Landaize unit, dolomite and dedolomite occur only in the Morillon section

Full size image

5.2 Paragenesis of the Reef Complex units

The Calcaires Récifaux unit shares almost the same early diagenetic history with the Calcaires de Tabalcon, namely micritization, moldic dissolution, and syntaxial overgrowth (see paragenesis in Fig. 6b). The most important differences are the blocky calcite cementation and dolomitization. The Calcaires Récifaux in the HU2 core underwent an early stage of isopachous cementation followed by a first stage of dolomitization, three successive stages of blocky calcite cementation (see Fig. 4b), a second stage of dolomitization and dedolomitization. Dolomitization affected most of units, as observed both in outcrops (Reculet Nord, Prapont) and subsurface (as described in the non-cored interval in HU2). The occurrence of replacive planar euhedral dolomites with well visible rhombs, preferentially in the fore reef facies and more precisely replacing the micritic matrix, indicates that the dolomitization occurred during the “early” stages of diagenesis. All of the blocky cement stages show zonation indicative of important changes in the chemical composition of the parent fluid during precipitation. The mostly non-luminescent, strongly zoned second stage of blocky calcite cementation is interpreted as a result of meteoric diagenesis, involving water typically poor in Fe and Mn and continually fluctuating chemical conditions (Tucker and Wright 1990; Holail 1992). The latest stage of blocky calcite cementation (BC3) is incomplete and the initial porosity is preserved with intercrystalline macropores, lacking dissolution (Fig. 4b–e). In all other study sites, two stages of blocky calcite cements were identified, with a predominance of the second stage, BC2 (dull luminescent, brown to orange zonation under CL). This blocky calcite cementation most likely took place during mesogenesis under reducing conditions. This stage is followed by dedolomitization, almost similar to that previously described in the Calcaires de Tabalcon unit. Complete dedolomitization, enhanced by the important secondary, rhombohedral moldic porosity, is observed only in Prapont (Fig. 4c, d). Remnants of dolomite tend to be localized on the edge of the pores and exhibit large dissolution features. Dedolomitization may indicate the flow of Mg-depleted water leading to the dissolution of dolomite with minor or no precipitation of calcite (Ayora et al. 1998; Reinhold 1998; Kyser et al. 2002).

The paragenesis in Champfromier differs from that of other study sites. First, the depositional context in Champfromier is particular: the micritic limestone was deposited in a calm open marine environment probably due to a lower topography (Bernier 1984). Second, the limestone contains large, sparitic, calcite-filled fractures that are later affected by dolomitization. The presence of thin calcite twinning could indicate that the calcite infillings were affected by tectonic deformation, under a temperature below 140 °C (Burkhard 1993; Ferrill et al. 2004). Moreover, the fact that dolomitization is characterized by a non-planar anhedral texture indicates that the temperature during its precipitation was greater than 50 °C (Gregg and Sibley 1984; see discussion below). Therefore, it appears that the Calcaires Récifaux unit in Champfromier was affected by very late diagenesis through greater burial. Moreover, dissolution enhanced fractures, filled by non-luminescent blocky sparite, indicates very late meteoric dissolution/precipitation process after exhumation.

The paragenesis of the Calcaires de Landaize unit is comparable to that from the other studied sites (Fig. 6c). Early diagenesis is marked by micritization and isopachous cementation followed by moldic dissolution. The isopachous cementation forms a thick rim around allochems. This first stage of cementation occurred prior to or during early compaction as it can be observed trapped between particles or as rims around particles already in contact. Syntaxial overgrowth started at the same time because the isopachous cement precipitation around echinoderm fragments is not rimmed. Both the presence of isopachous cements and well developed syntaxial overgrowth most likely indicate a marine phreatic environment during the early stages of diagenesis (Land 1970; Longman 1980). The syntaxial overgrowth is followed immediately by the first stage of blocky calcite cementation (BC1) displaying a bright luminescence with orange zonation under CL and then a second stage (BC2). Only the Morillon section exhibits rhombohedral moldic porosity pointing to dedolomitization.

6 Discussion

6.1 Models of dolomitization

The processes and diagenetic environments involved in the precipitation of dolomite have been and are still debated due to the fact that dolomite is rarely observed in modern marine depositional environments, despite its abundance in ancient sedimentary rocks (Arvidson and Mackenzie 1999). The “dolomite problem” results from the poor understanding of the chemical and/or hydrological conditions of formation and the difficulty to propose a single genetic origin derived from petrographic and geochemical data (Machel 2004). The following discussion on dolomitization processes affecting the Upper Jurassic limestones of the Geneva Basin is based mostly on petrographic data taking into account the paragenetic interpretation of this study. The discussion on dolomitization models for the three types of dolomite described in this study is based on previous work and results from local and regional studies summarized in Tables 1, 2 and 3. The dolomitization occurring in the Champfromier section is clearly different from that observed in the other sites and will be discussed separately.

Table 1 Examples of dolomitization reported throughout the Phanerozoic
Full size table
Table 2 Examples of dolomitization reported throughout the Phanerozoic
Full size table
Table 3 Examples of dolomitization reported throughout the Phanerozoic
Full size table

The first type of dolomite, exhibiting replacive euhedral rhombs mostly in a micritic matrix, fits well with a model of early reflux-type dolomitization (Adams and Rhodes 1960; Warren 2000; Machel 2004). The unimodal size distribution of the first dolomite indicates a single nucleation event while the idiotopic euhedral texture is indicative of a formation temperature below the critical roughening a temperature of 50 °C as defined by Gregg and Sibley (1984). The reflux-type model was proposed by Adams and Rhodes (1960) to explain extensive lagoonal and reefal dolomite in the Permian Basin of West Texas. In this model, the dolomitizing fluids are mesosaline brines with salinities controlled by surface evaporation in near-surface and shallow burial diagenetic settings. These brines originate from seawater evaporated beyond gypsum saturation in lagoonal and shallow-marine settings on a carbonate platform behind a reef acting as a barrier. The hypersaline brines, being denser than seawater, initiate downward fluid migration through the platform sediments, inducing dolomitization. This model is favored in the present case as the Kimmeridgian reef complex in the Geneva Basin developed in a marine, possibly protected, lagoonal zone (Meyer 2000). Moreover, the Prapont section includes, on top of the reef sequences, a succession of lagoonal and subtidal storm deposits, capped by supratidal mudflats and beach deposits (Fookes 1995). However two problems arise from this model in our study: (1) evidence for evaporite precipitation is lacking in the studied sections and (2) the dolomitization is limited to decametric to metric thick beds instead of a massive body of dolomite. Rameil (2008) pointed out this issue in his study of Upper Jurassic and Lower Cretaceous limestones of north-west Switzerland, a few kilometers north-east of our study area. The lack of evaporite deposits could be explained by the fact that the dolomitizing fluids were mesosaline (below gypsum saturation) instead of hypersaline, as documented by Simms (1984), Kaufman (1994), and Machel (2004). Moreover, during periods of small and high-frequency sea-level changes, prograding platforms tends to develop into platform-top areas with large supratidal domains regularly inundated by storm surges (Montañez and Read 1992; Sun 1994; Rameil 2008; see also Gabellone and Whitaker 2016). In this case, mesosaline fluids could become the source of reflux-dolomitization, a scenario known as “brine reflux dolomitization” (Warren 2000) or “penesaline dolomitization” (Qing et al. 2001). The scenario proposed by Rameil (2008) is a two-step process. During high-sea level, hypersaline brines are limited to the tidal flat environment with recharges coming from storms and reflux limited to this zone. In the lagoon, while evaporation occurs, free water exchanges with seawater are possible as the platform rim is below sea level. Therefore, brines produced in the lagoon are stenonohaline to mesosaline only, without reflux in the intertidal domain. When the sea-level drops, the platform rim may be emerged, limiting the water exchange between the lagoon and open marine seawater. Evaporation in the lagoon will lead to the production of mesosaline to hypersaline fluids of higher density that will initiate a reflux-type dolomitization as they sink towards the platform rim. The hypothesis of Rameil (2008), based on the work of Mutti and Simo (1994) and Qing et al. (2001), proposes that pulses of dolomitizing fluids are responsible for the observed distribution of dolomite bodies. These pulses can be induced by repeated high-frequency sea-level falls where the next sea-level rise induces a latent reflux. This alternating reflux, dependent on the sea-level changes, would then result in repeated infiltration horizons creating repeated dolomitization in the peritidal limestone succession (Rameil 2008).

As pointed out by Fookes (1995) for the Prapont section, the lagoonal deposits were possibly affected by an increase in water salinity evidenced by a gradual decrease in abundance of green algae and foraminifera. This may have resulted in forming mesosaline fluids. In addition, high-frequency sea-level fluctuations were observed and described both in the Prapont (Fookes 1995) and Etiollet (Deville 1990) sections. The first type of dolomite exhibits replacive euhedral rhombs, mostly in a micritic matrix matching the criteria of matrix-selective dolomitization which is commonly the earliest stage of dolomitization (Machel 2004). While it is not clear if a critical roughening temperature exists for replacive dolomite (Braithwaite 1991; Machel 2004), Baldermann et al. (2015) reported a moderate temperature of about 26 °C for their first type of replacive dolomite in the Upper Jurassic limestone of the North German Basin. Therefore, data from the present study agree with the scenario of an early dolomitization in near-surface to shallow conditions, following a reflux-type model with mesosaline to hypersaline waters.

The second stage of dolomitization exhibits fine to medium, euhedral to subhedral dolomite rhombs. This second dolomite is thought to have originated from the same dolomitization event that led to the first stage of dolomitization but resulting in two dolomite populations (Machel 2004). This texture is characterized by two distinct types of dolomite with different sizes, shapes, pore types, and connectivity. The crystals of the smaller sized population has a cloudy core with or without clear rims. Domains with the larger sized dolomite exhibit higher intercrystalline porosity (Machel 2004). This is the case for the Calcaires de Tabalcon (Fig. 3e–f) and Calcaires Récifaux (Fig. 4e–f) units. Machel (2004) suggests two hypotheses for this texture, namely a single event of dolomitization or recrystallization of precursor dolomite. Textural differences in the host limestone or heterogeneous lithification prior to dolomitization can explain the difference in dolomite texture. The second hypothesis can be linked to near-surface or evaporitic dolomite originating from reflux-type dolomitization. Metastable protodolomite prone to recrystallization during burial results in some rock domains recrystallizing into a coarser texture. In our study, the recrystallization scenario is likely because the second dolomite is characterized by a coarser texture, crosscutting the first dolomite. However the timing of this second event is difficult to constrain. While Machel (2004) stated that recrystallization occurred during burial, no information is given on the depth of this burial. Given the petrographic data provided in the present study and in agreement with Reinhold (1998), the second generation of dolomite could originate from recrystallization in a shallow burial environment closely related to the first reflux-type dolomite.

The third stage of dolomitization was observed only in the Reculet Nord section. In the Calcaires de Tabalcon unit the sucrosic dolomite represents an advanced stage of replacement that obliterated the original fabric creating high intercrystalline porosity. The rhombs exhibit cloudy cores surrounded by rims. In the Calcaires Récifaux unit, the partially replacive, medium to coarse euhedral dolomite crosscuts pressure-dissolution features. These observations tend to give this third stage a more complex origin than the previous ones, likely due to burial dolomitization.

Several models exist for the generation and migration of dolomitizing fluid during burial. Heydari (1997) discriminated three categories or burial diagenesis realms in function of hydro-tectonic processes: passive margin, collision margin, and post-orogenetic. Machel (2004) proposes four different scenarios also based on the processes driving fluid flow: compaction, thermal convection, topography, and tectonically driven. For the present study, the data at hand can help to discard some of these scenarios for the Geneva Basin.

In the compaction (or passive margin) driven model, seawater or modified trapped seawater is buried with the sediment during burial and pumped due to compaction dewatering (Illing 1959; Jodry 1969). This process can produce only limited extent of dolomitization due to the limited amount of compaction water but is viable when the fluid is funneled towards small volumes of limestones. Choquette and Hiatt (2008) discussed the importance of shallow burial dolomite cement as a component of ancient sucrosic dolomite, describing this as “cements that were never deeply buried, are limpid, have planar faces (non-saddle forms), often distinct zonation in cathodoluminescence and form syntaxial overgrowth on crystals facing pores”. The model proposed by Choquette and Hiatt (2008) is a four-step process common in lime mud (mudstone to wackestone textures), compatible with evaporitic dolomite. The first step represents the nucleus stage where dolomite crystals grow authigenically or are emplaced as detrital particles in the limestone, providing a very fine (1–10 µm) nucleus for the following step. During the second step, the cortex will grow around the nucleus, initiating the first stage of textural coarsening, replacing the initial limestone. This coarsening might lead to crystal interpenetration due to competition and/or compaction. Crystals may also appear cloudy due to residual inclusions. The third step starts after the depletion of sediment-sourced Mg and/or complete dissolution of remaining CaCO3. In this step, limpid planar-euhedral cement precipitated by overgrowing over the cortex formed in step two. The initial fabric is completely obliterated, leaving important intercrystalline water-filled pore while the rigid framework formed by the dolomite cement tends to slow compaction. The last step consists of further overgrowth of dolomite cement filling the available pore space, leading to coarsening of the initial dolomite texture as long as pore space is available. The third stage of dolomitization seems to follow to a certain extent this scenario as the Calcaires de Tabalcon unit still displays important intercrystalline porosity, meaning that step three was the last step occurring. Following this reasoning, the third stage in the Calcaires Récifaux unit could also originate from shallow burial dolomitization without complete obliteration of the original fabric. This is supported by the fact that dolomite crystals crosscut the interpenetration and resulting microstylolitization of grains. These pressure-solution features are known to occur very early during burial at depth ranging from 60 to 90 m (Schlanger 1963; Dunnington 1967). Moreover, several studies assumed that cement dolomitization occurred in a relatively closed system with a local source of Mg-rich fluids (Murray 1960; Weyl 1960; Warren 2000). This scenario agrees with that of Choquette and Hiatt (2008), as these processes “may go to completion after relatively early, shallow burial (tens to hundreds of meters burial depths), at relatively low diagenetic temperatures”. The thermal convection models are driven by spatial variations in temperature, due to elevated heat flux in the vicinity of igneous intrusions, lateral contrast between warm platforms waters and cold ocean waters, or lithology-controlled variations in thermal conductivity. This elevated heat flux will results in modification of the pore-water density (Kohout et al. 1977; Wilson et al. 1990). Two types of convection may form: open convection when the carbonate platform is open so seawater recharge laterally and discharge at the top; close convection if the temperature gradient is high enough with regards to the permeability of the limestone and in the absence of interbedded aquitards. While this model could be considered, there is no evidence that the requirement for an open or closed convection was met in the Kimmeridgian platform. Moreover, Machel (2004) reported that thermal convection can be “overpowered” by the presence of a reflux flow due to evaporitic waters, based on numerical modeling (Jones et al. 2002, 2003, 2004).

The tectonic or collision margin model (Oliver 1986; Heydari 1997; Warren 2000; Machel 2004) requires hot (100 °C to above 250 °C), metamorphic or hypersaline and highly pressured fluids, driven towards the basin margin and vertically through faults and fractures. Such hot fluid are responsible for non-planar and saddle dolomite precipitation. The dolomitization in the Champfromier section is the only one that contrasts the most with that observed in other sections studied. In this section, dolomitization seems to be associated with faults and fracturing. In addition the thin calcite twining is indicative of the calcite infillings affected by tectonic deformation (see above). The fact that dolomitization is characterized by a non-planar texture indicates that the temperature during its precipitation was greater than 50 °C (Gregg and Sibley 1984). Moreover, Radke and Mathis (1980) and Gregg (1983) postulated formation temperatures greater than 60 °C, which is in line with fluid inclusion data on saddle dolomites from the Upper Jurassic at the Franconian Alb, southeast Germany, which document precipitation temperatures between 60 and 90 °C (Liedmann and Koch 1990; Liedmann 1992). All of these observations support the assumption of deeper burial, where dolomitization would be the results of tectonic or hydrothermal driven fluids. The important overprint of tectonics in this section makes the paragenesis reconstruction more difficult to assess. Until further work is done on this section, no predominant scenario for dolomitization can be unequivocally privileged.

In the topography or post-orogenetic model (Tóth 1988; Garven 1995; Heydari 1997; Warren 2000; Machel 2004), dolomitization is considered as the result of an important amount of meteoric water migrating in uplifted sedimentary basins through recharge zones. During migration, the meteoric fluid will dissolve Mg-rich material prior to reaching the limestone where dolomitization will occur. The topography model does not appear to be common and only few examples were reported from the Cambrian carbonates in Missouri (Gregg 1985), the Cambrian-Ordovician carbonates in southern Canadian Rocky Mountains (Yao and Demicco 1995), and more recently from the Upper Cretaceous carbonates in the Dead Sea Transform, central Negev desert (Matthews et al. 2006). The active tectonics responsible for thrusting and folding of the Jura took place recently, in a time period ranging from the Miocene to the Pliocene (Laubscher 1986, 1992; Maurer et al. 1997; Sommaruga 1997; Burkhard and Sommaruga 1998; Mosar 1999; Becker 2000) while the Jura and Salève Mountains exhumation could have started as early as the Eocene (Schroeder 1958). A dolomitization following this scenario would occur later than the one resulting from compaction as discussed previously. Therefore, this model is not considered to explain neither early nor burial dolomitization observed in the present study. However, this process might have induced further dolomitization of buried rocks in the basin after exhumation. This issue is discussed further in Sect. 6.3.

6.2 Comparison with existing models

The bibliography on dolomitization is extensive, with various models available since the very first description of dolomite (de Dolomieu 1791). The objective of this section is not to review all previous studies. We selectively summarize some examples of dolomitization in limestones from the Lower Ordovician to the Upper Cretaceous, mostly in the Tethys realm (Tables 1, 2). Examples are presented along with a summary of the different stages of dolomitization, the respective pathways and scenarios invoked. Below, we focus on case studies from the Upper Jurassic and/or from areas close to the Geneva Basin. Reinhold (1998) published a thorough study dealing with the dolomitization of the Kimmeridgian formations in the Swabian platform (South Germany). The stratigraphic levels are equivalent to the Lower and Upper Kimmeridgian deposits in the Geneva Basin. Reinhold (1998) described six types of diagenetic dolomites. The first dolomite type displays rhombs that are fine to coarse, idiotopic euhedral to xenotopic anhedral, porphyrotopic, and fabric selective to pervasive. He interpreted this first stage as the result of an early dolomitization during shallow burial induced by modified seawater. The next two stages are quite similar and exhibit medium to (very) coarse rhombs that can be euhederal, subhedral to anhedral, and pervasive. This second and third stages are linked to two recrystallization phases of dolomite by an interaction with modified seawater or mixed meteoric/marine water during burial followed by downward migration of meteoric waters. The fourth and fifth stages are dolomite cements characterized by a first fine to coarse, euhedral to subhedral void filling cement directly followed by medium to coarse anhedral void filling syntaxial dolomite. These stages were interpreted as being the result of shallow burial dolomitization. The last stage is represented by fine to coarse, euhedral to anhedral void filling cements in the form of saddle dolomite being the results of late shallow burial related to deep-burial hydrothermal fluids transported along reactivated fractures. While Reinhold (1998) does not consider a reflux-type model, the timing of dolomitization is similar to that of the present study, with a first early stage followed by shallow burial dolomitization. The fourth and fifth stages exhibit the same characteristics as those of this study’s third stage in the form of a void filling and syntaxial dolomitization that could be the result of a scenario similar to that described in Choquette and Hiatt (2008). Moreover, the last stage of deep-burial dolomitization could be similar to that described in the Champfromier section with dolomitization associated with fluid circulation along fractures. Rameil (2008) studied the Upper Jurassic–Lower Cretaceous Twannbach formation in northwest Switzerland, a few kilometers to 50 km away from sites presented in the present study, and reported three diagenetic dolomites. The first stage displays medium euhedral to subhedral replacive rhombs exhibiting cloudy cores and clear rims. The second stage is characterized by a fine euhedral replacive dolomite. The third is composed of fine to medium subhedral rhombs observed only in burrows. The processes linked to the generation of the first two generations of our study are similar to those of Rameil (2008). The third stage of dolomitization is interpreted as microbial mediation where burrowing organisms induces biochemical modifications by concentrating organic matter that in turn will serve as a substrate for bacterial colonization. Under reducing conditions, sulphate-reducing bacteria would consume SO42−, which acts as an inhibitor to dolomite precipitation as it may bind with Mg2+ resulting in lower Mg/Ca ratio. In such a context, dolomitization by microbial mediation can occur. Furthermore, Rameil (2008) reported two types of dedolomite which he interpreted as the result of long-term emersion and interaction with meteoric waters. This dedolomitization scenario is similar to that recorded and presented in this study. Baldermann et al. (2015) focused on dolomitization of the Upper Jurassic Langenberg section in the North German Basin (Germany). The authors distinguished three stages of dolomitization: a first stages represented by fine to medium, euhedral to subhedral, dolomite; a second, displaying fine to medium, euhedral to subhedral, hypidiotopic to idiotopic fabric destructive dolomite exhibiting cloudy cores and clear rims; and a third stage exhibiting coarse euhedral void filling cements which are fabric retentive. Baldermann et al. (2015) interpreted these dolomitization events as the results of shallow seepage reflux and/or evaporitic tidal pumping at moderate temperatures (26° to 37 °C) by pristine marine to slightly evaporitic and reducing seawater derived from interstitial solutions. As for the Swabian platform (Reinhold 1998), dolomitization is thought to have been facilitated by bacterial sulfate reduction. In addition, the last stage of dolomitization is affected by dedolomitization. Data and interpretations of Baldermann et al. (2015) are very similar to those presented in this study, indicating that the processes leading to dolomitization and dedolomitization of the Upper Jurassic in the Geneva Basin and in the North German Basin might be related. The dolomitization processes reported in Reinhold (1998), Rameil (2008), and Baldermann et al. (2015) are very similar to those described in this paper. Textural characteristics of the different stages of dolomite in all studies described above are analogs to those recorded in the Upper Jurassic limestones in the Geneva Basin. Specifically the second stage of fabric destructive dolomite in the North German Basin (Baldermann et al. 2015) which is almost identical to the sucrosic dolomite reported in the third stage of our study. Therefore, a reflux-type model of dolomitization followed by syntaxial overgrowth during shallow burial, as suggested for the Geneva Basin, appears to apply to most cases reported so far for the Upper Jurassic from northern Switzerland up to the North German Basin (Table 1). Dolomitization is scarce in basins close to the Geneva Basin. In their study of the Upper and Middle Jurassic limestones in the Paris Basin, Vincent (2001) and Brigaud et al. (2009b) reported dolomite associated with stylolites and fractures. Both studies proposed that dolomitization was either the consequence of dewatering during compaction of clay-rich formation, providing Mg-rich fluids (most probably from the Callovian-Oxfordian clays) and/or a result upward migrating hydrothermal fluids from Triassic. Such a scenario is also reported from Middle Jurassic to Lower Cretaceous limestones in Italy where most of the dolomitization is thought to be associated to hydrothermal fluids and fault-related processes (Cervato 1990; Barale et al. 2013; Rustichelli et al. 2017). Other examples of such models of dolomitization were reported from Spain, from the Upper Carboniferous limestones of the Bodon unit (Variscan Cantabrian) (Gasparrini et al. 2006) and the Lower Cretaceous limestones of the Benicassim formation (Masetrat Basin) (Gomez-Rivas et al. 2014). Iannace et al. (2014) showed that the Lower–Upper Cretaceous limestones in the South Apennines of Italy (Mt. Faito and Mt. Chianello sections) were affected by two stages of dolomitization. These stages resulted from several pulses of “slightly” concentrated marine brines, similar to the scenarios of Rameil (2008). Farther from the Geneva Basin, a significant number of studies dealing with dolomitization in limestones with stages, textures, and characteristics comparable to those described in the present work were conducted. Some of these are discussed below. Readers can refer to Tables 1, 2 and 3 for references and detailed characteristics of dolomitization and their associated models in the Upper Jurassic and other time intervals.

Other studies in the Upper Jurassic exhibit data and interpretations fairly similar to those of this study. Goldberg (1967) showed that the Upper Jurassic limestones in the Negev of Southern Israel were affected by supratidal dolomitization due to a reflux of hypersaline water. This dolomite is then affected by a stage of dedolomitization probably related to subaerial exposure. In the North-West Golf of Mexico, Moore et al. (1988) showed that the Upper Jurassic limestones are affected by three stages of dolomitization (early pervasive then coarse replacive and pore filling dolomites) which are the results of an evaporative reflux-water scenario. In North East Iran, the Upper Jurassic limestones display five stages of dolomitization (coarsening from 40 to 500 µm) that are the result of a supratidal to upper intertidal seawater dolomitization followed by shallow burial dolomitization (Adabi 2009). Similar examples were reported from other time intervals, e.g., in the Lower Permian limestones of Central Oman where Beckert et al. (2015) described three stages of dolomitization resulting from seepage reflux of hypersaline fluids and shallow burial dolomitization. These dolomites were also affected by dedolomitization, probably due to meteoric fluids.

6.3 Implications for future subsurface prospection and exploitation

Assessing the potential of carbonate reservoirs is not easy because of their inherent heterogeneities. Limestones of the Geneva Basin are no exception as the diagenetic history proved to be complex with processes partly overriding precedent diagenetic phases. Nevertheless, the studied units revealed promising reservoir qualities, especially in the sucrosic dolomite where porosity values can range from 10 to 15% and permeability can reach 35 mD (Moscariello 2016; Rusillon et al. 2016). Dolomitization has an important impact on reservoir properties as it is often associated with increase in porosity and permeability, especially in sucrosic dolomites (e.g., Schmoker et al. 1985; Warren 2000; Wang et al. 2015; Giorgioni et al. 2016). In such sucrosic facies, the high intercrystalline porosity can form a highly connected porous network ensuring good fluid flow, storage and drainage properties. However, over-dolomitization in shallow burial settings, as discussed in Choquette and Hiatt (2008), can lead to an important decrease in porosity and, thus, possibly contribute to reducing reservoir properties. In the same way, dedolomitization can produce important amount of secondary porosity, also having important impacts on reservoir properties. Therefore the estimation of the volume of dolomitized limestone in the subsurface is very important in the actual plan to produce geothermal energy in the Geneva Basin. For such an estimate, it is necessary to constrain the volume of dolomitizing fluid migrating through the basin and quantify the extent of dedolomitization.

The scarcity of subsurface data is a major pitfall to estimate the volume of dolomitizing fluids in the basin. In the Humilly-2 (Hu-2) well, no cores or cuttings of dolomitized limestone are available, therefore only the non-published final drilling report can provide some insights. According to the report, the Upper Jurassic limestone exhibits two dolomitized intervals: one from 1181 to 1271 m (TVD) displaying a “beige limestone locally dolomitized with layers of darkened sucrosic dolomite” and one from 949 to 1009 m (TVD) displaying a “beige, fine to coarse limestone with presence of coarse, gray, compact dolomitic limestone”. Therefore we can roughly estimate that about 150 m of limestones are affected by dolomitization in Hu-2. The hônex-1 well (Th-1) is located 17.5 km east of Hu-2. Only two cores were recovered, one in the Lower Cretaceous and one in the Tidalites de Vouglans unit (Fig. 1). Here also, the non-published final drilling report is the only source of information. In this well, the final drilling report mentioned that the reef complex is about 216.6 m thick but dolomite was only observed at the base of the Calcaires de Tabalcon unit (2038 m depth). However, fractures are abundant and calcite filled according to the report. The unit underlying the Calcaires de Tabalcon seems to have fractures filled by dolomitic cements. The absence of dolomitization in the Th-1 well is rather interesting and unexpected. Indeed it is not unusual to find replacive texture present in a single formation and locally even in a single outcrop or thin section (Warren 2000). Without available cores or cuttings it is difficult to provide further discussion about the absence of dolomite in this well.

While the present study proposed that the dolomitization was an early process, a scenario of late dolomitization that may have affected the subsurface of the Geneva Basin cannot be excluded. First of all, the Jura Mountains are subject to active tectonics since the Miocene. Therefore, some questions arise: could topographic driven dolomitization affect the subsurface? Most of the dolomite is affected by calcitization which can be driven by meteoric fluid migrating through the limestone during telogenesis after exhumation (Ayora et al. 1998; Kyser et al. 2002). This could lead to Mg-enriched fluids possibly migrating toward the basin. As the Upper Jurassic is currently acting as an aquifer in the subsurface, such fluids could act as recharge water and possibly induce further dolomitization. Moreover, the Molasse Basin was affected by a decollement horizon in the thick Triassic evaporites (Philippe 1994). Compactional fluids originating from tectonic squeezing, as expected in a tectonic-driven scenario, are a viable alternative. The same applies for Mg-rich fluids originating from dewatering of clay-rich layers during compaction (Vincent 2001; Vincent et al. 2007; Brigaud et al. 2009a, b). Moreover, the northern part of the Geneva Basin is affected by four major wrench faults representing potential high-intensity fracture zones with associated enhanced permeability conditions (Clerc et al. 2015). In this context, hydrothermal driven fluids migrating upward could also be considered, which would fit the occurrence of dolomitic cements filling fractures in the Th-1 well.

Although some conclusions can be made from the present study, some issues remain open, particularly regarding diagenetic processes in the basin center. To better understand and constrain the volume and extension of the dolomitization, additional cores need to be acquired. Furthermore, the acquisition of geochemical data on the dedolomitization by sampling the remaining dedolomite will provide fundamental data to constrain and evaluate the type and timing of fluids that led to the calicization of the dolomitic Upper Jurassic limestones.

7 Conclusion

The Upper Jurassic carbonate rocks form a complex carbonate reservoir strongly affected by dolomitization and dedolomitization. This study provides a relative chronology of diagenetic stages for the different units present in the Kimmeridgian of the Geneva Basin. Based on the petrographic data acquired from sub-surface cores and outcrops, the following conclusions can be made:

  1. 1.

    Most of the initial porosity in the different units studied was filled by blocky calcite cement. This cementation occurred during early burial diagenesis. Most of units exhibit a two-stage blocky calcite cementation and only the top of the Calcaires Récifaux evidences a third incomplete stage of blocky calcite cementation with preservation of intracrystalline macroporosity.

  2. 2.

    Dolomitization occurred during early diagenesis and overprinted all precedent stages. The most affected units are the Calcaires de Tabalcon and the Calcaires Récifaux. To some extent, the Calcaires de Landaize unit is affected but only in the Morillon section. In most cases, dolomite has a planar, non-mimicking replacive texture.

  3. 3.

    Petrographic data was used to assess the dolomitization scenarios. The first stages of dolomitization are interpreted to be induced by a reflux-type model involving mesosaline to hypersaline fluid originating from evaporitic conditions in a lagoonal environment. High-frequency sea-level fluctuations acted as a mechanism for pulse migration of the brines through the limestone resulting in partial dolomitization. The second stage of dolomitization is thought to result from recrystallization during shallow burial. The third stage of replacive, fabric-destructive dolomite is explained by shallow burial dolomitization generating syntaxial overgrowth dolomite over pre-existing nuclei. This process is responsible for the highly porous sucrosic dolomite occurring in the Reculet section.

  4. 4.

    Dedolomitization is identified at different order of magnitude by either: (1) almost complete dissolution leading to the creation of secondary pore space or (2) two-step calcitization driven by the infiltration of Ca-rich water leading to dissolution, formation of micro-vugs, and secondary precipitation of calcite.

  5. 5.

    The creation of secondary pore space could provide good connectivity between the intraparticular or matrix related microporous network and the interparticular moldic macroporous network. This enhanced connectivity could therefore provide good reservoir properties suitable for geothermal energy exploitation.

As for most studies, carbonate heterogeneity remains to be a major issue when assessing the exploitation potential. Understanding the paragenesis affecting such reservoirs is an important step towards the better exploitation of resources currently available. This study provides further insights on a possible reflux driven dolomitization occurring in shallow carbonate platforms.