VU Research Portal

Isotope analysis of fluid inclusions de Graaf, S.

2018

document version Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA) de Graaf, S. (2018). Isotope analysis of fluid inclusions: Into subsurface fluid flow and coral calcification.

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

E-mail address: [email protected]

Download date: 01. Oct. 2021 Chapter Fluid flow evolution in the Albanide fold- 3 and-thrust belt

ectonic forces generated during thrust emplacement along active margins may drive complex fluid flow patterns in fold-and-thrust belts and foreland basins. In the Albanide fold-and-thrust belt, fracture-controlled fluid flow Tled to the development of calcite vein networks in a sequence of naturally fractured Cretaceous to Eocene carbonate rocks. Fluid inclusion isotope data of these calcite veins demonstrate that fluids circulating in the carbonates were derived from an underlying reservoir that consisted of a mixture of meteoric water and evolved marine fluids, probably sourced from deep-seated evaporites. The meteoric fluids infiltrated in the hinterland before being driven outward into the foreland basin and ascended as soon as fracturing induced a sufficient increase in permeability. The contribution of fluids derived from evaporites increases towards the thrust front in association with elevated deformation, which is expected to be the main driver of expulsing connate waters from the evaporites. Structural and petrographic observations provide time constraints for the various phases of fracture infilling and reveal an increasing dominance of meteoric water in the system through time as migration pathways shortened and marine formation fluids were progressively flushed out. Similar fluid flow evolutions have previously been recorded in various fold-and-thrust belt settings elsewhere in the world.

Based on: De Graaf, S., Nooitgedacht, C. W., Le Goff, J., Van der Lubbe, H. J. L., Vonhof, H. B., and Reijmer, J. J. G. (accepted for publication in AAPG Bulletin) Fluid flow evolution in the Albanide fold-and-thrust belt: Insights from δ2H and δ18O isotope ratios of fluid inclusions Nooitgedacht, C. W., De Graaf, S., Van der Lubbe, H. J. L., Vonhof, H. B., Le Goff, J., and Reijmer, J. J. G. (in prep.) Tracking fluid migration in the external Albanides using hydrogen and oxygen isotopes of fluid inclusions in calcite veins 46 Chapter 3

3.1 Introduction Fold-and-thrust tectonics is commonplace along convergent plate margins and is associated with the development of intricate fracture and fault networks (Roure and Sassi, 1995). The structural complexity of fold-and-thrust belts opens up a wide array of conceivable fluid flow patterns (Travéet al., 2007; Morley et al., 2011; Lacroix et al., 2014). Tectonic loading due to thrust emplacement, for instance, may instigate basin-ward expulsion of fluids along large-scale migration pathways like faults or basement detachments (Oliver, 1986; Ge and Garven, 1992). The timing of fluid flow and circulating fluid types may differ considerably between foreland fold-and-thrust belts depending on their stratigraphic architecture and structural evolution (Ferket et al., 2003; Schneider et al., 2004; Barbier et al., 2012; Crognier et al., 2018). Predicting the evolution and interplay of deformation and fluid flow in foreland fold-and-thrust belts remains complicated, while it is of key importance when assessing the potential existence of aquifers or petroleum-bearing reservoirs (Agosta et al., 2010; Morley et al., 2011). Carbonate rock formations are of particular interest since the permeability of lithified limestone is generally governed by brittle deformation structures (i.e., fracture and faults) due to limited matrix porosities (Nelson, 2001). In this Chapter, fluid inclusion isotope data of calcite veins are used to delineate the fluid flow system of the Albanide foreland fold-and-thrust belt. Prominent calcite veining in the Albanide fold-and-thrust belt exists in fracture networks in Cretaceous to Eocene Ionian Basin carbonates, which are the main reservoir rocks in the area (Velaj et al., 1999) and an analogue to producing reservoirs in the Adriatic Sea and onshore Italy (Cazzola and Soudet, 1993; Zappaterra, 1994). The carbonate rocks are mainly composed of calciclastic gravity flows and mass transport deposits (Rubertet al., 2012; Le Goffet al., 2015). Events of fracturing provided discrete pathways facilitating the episodic migration of fluids and governing the emplacement of abundant hydrocarbon accumulations in the area (Graham Wall et al., 2006). Previous vein-based fluid flow reconstructions in the Ionian Zone focused on the areas of Kremenara (Van Geet et al., 2002; Swennen et al., 2003), Shpiragu (Graham Wall et al., 2006), Kelçyrë (Vilasi et al., 2009) and Saranda (Lacombe et al., 2009). These previous researches inferred multiple fluid flow episodes throughout the evolution of the foreland fold-and- thrust belt. Although the sequence of fluid flow and vein cementation events has quite elaborately been studied, the precise origin and migration pathways of fluids requires further constraints, especially for the early-stage fluid system. Excellent outcrops of Cretaceous to Eocene carbonate rocks are present in the Mali Gjere mountain ridge, which is a topographic expression of the Kurveleshi thrust sheet in the Ionian Zone of south (Figure 3.1). In addition to studying veins from the Mali Gjere, calcite veins from a set of outcrops from the two other main thrust units in the Ionian Zone – the Çika and the Berati belt – were analyzed Fluid flow evolution in the Albanide fold-and-thrust belt 47

A B Berati BeltB Montenegro Albanian alps Kosovo

Mirdita Zone Mashkullorë Korabi 15 N Macedonia

Pre-Adriatic Depression Pre-Adriatic Gjirokastër

Mali Gjere

Korça Basin Kurveleshi Belt

Kruja Zone 20 Krasta Zone Vanister 40˚ Ionian 00’ Berati belt 18 Çika belt Sazani Zone Zone Kurveleshi belt 20 Research area Muzinë Çika Belt 22 Greece 41 Legend Quaternary Bodrishtë A Sarandë 14 Neogene

Paleocene Ionian Sea Cretaceous 39˚ 50’ Jurassic 10 km Triassic 20˚ 00’ 20˚ 15’

C SW Çika belt Kurveleshi belt Berati belt NE A Mali Gjere B

Jura Flysch ssic-Cr and m etace olass ous c e cove arbon r Delvina oil field ates an Triassic evaporites d shales 1 km 4 km

Figure 3.1 (A) Map showing the main structural units of Albania (Moisiu and Gurabardhi, 2004) with the research area indicated by the black square. (B) Geological map of southern Albania after Moisiu and Gurabardhi (2004). Seven outcrop locations (Mashkullorë, Vanister, Terihat, Frashtan, Muzinë, Jorgucat and Bodrishtë) were studied along the Mali Gjere mountain ridge, which exposes Cretaceous to Eocene carbonate rocks within the Kurveleshi thrust sheet. Red lines depict faults. (C) Simplified geological cross-section through southern Albania after Velaj (2015), indicated by the line A-B in the geological map.

to assess spatial variations in the fluid flow evolution throughout the entire fold-and- thrust belt. Besides fluid inclusion isotope analysis, complementary petrographic and δ18O and δ13C isotope analyses of the calcite vein cements were carried out. An analysis of the arrangement and cross-cutting relations of distinct structural sets was performed to acquire a chronological framework for the various generations of fracture-infilling calcite. 48 Chapter 3

3.2 Background 3.2.1 Geological evolution of the Ionian Zone The Ionian Zone is one of the structural domains that makes up the Al- banides (Figure 3.1) and refers to an assembly of folds and thrusts of Ionian Basin deposits situated in the southwest of Albania (Meço and Aliaj, 2000; Robertson and Shallo, 2000). Deposition in the Ionian Basin was closely linked to the structural evolution of the Adriatic plate (Channell et al., 1979). In Triassic times, at least 2000 m of evaporites and shallow-water carbonates were deposited in an epicontinental marine environment (Vlahović et al., 2005; Karakitsios, 2013). Subsequent exten- sional tectonics within the Adriatic plate from the Late Triassic onward led to the development of rift basins in a horst and graben system. The Ionian Basin was such an intracontinental rift basin and formed between the Apulian and Kruja carbonate platforms (Robertson and Shallo, 2000; Heba and Prichonnet, 2009). The syn-rift sequence in the Ionian Basin from the Jurassic until the Oligocene is characterized by sediment input from the bordering carbonate platforms (Hairabian et al., 2015; Le Goffet al., 2015). The sequence of Cretaceous to Eocene carbonate gravity flows that is presently exposed along the Mali Gjere mountain ridge exhibits a thickness of 900-1350 m (Velaj, 2015) and is mainly composed of decimeter to meter-scale strata of fine-grained mud- to wackestones with minor amounts (< 2%) of planktic foraminifera. Uplift of the hinterland provoked by the Alpine Orogeny led to an increasing input of terrigenous material into the basin and the deposition of up to 2000 m of Oligocene flysch, Neogene molasse, and Quaternary sand- and siltstones on top of the Ionian Basin carbonates (Meço and Aliaj, 2000). The position of the Ionian Zone in the southern extremity of the Alpine orogenic belt was crucial for its structural development from the Oligocene onward (Robertson and Shallo, 2000; Nieuwland et al., 2001). Southwest-verging fold-and- thrust tectonics involving an intricate network of structural elements (e.g., ramp anticlines, back-thrusts, strike-slip faults and salt diapirs) caused a regional uplift and the formation of three main NNW-striking anticlinal thrust sheets: the Çika, Kurveleshi and Berati belt. Triassic evaporites and to lesser extent Early Jurassic Posidonia shales act as decollement levels facilitating thin-skinned thrusting (Velaj, 2002; Swennen et al., 2003). The Ionian Zone is bordered by the Vlora- lineament in the north and extends towards the south into the Hellenides (Neumann and Zacher, 2004; Karakitsios, 2013).

3.2.2 Fluid flow in the Cretaceous to Eocene carbonates Primary porosity of the Cretaceous to Eocene carbonate rocks is low (0-1%) due to compaction and calcite pore infill during early marine diagenesis (Van Geetet al., 2002; Vilasi et al., 2006; Dewever et al., 2007). Multiple events of brittle deforma- tion induced episodic increases in rock permeability and circulation of fluids in the Fluid flow evolution in the Albanide fold-and-thrust belt 49

Cretaceous to Eocene carbonates (Graham Wall et al., 2006). Episodes of fluid flow occurred before and during the main phase of folding and thrusting (Swennen et al., 2000). Fluids were overpressured and buffered by the host-rock during the early pre-folding fracture infilling events (Van Geet et al., 2002). Meteoric fluids played an important role in vein cementation events during the late stage, which is also associated with the development of karst networks (Vilasi et al., 2006). Evidence for leaching of evaporites indicates that fluids were supplied along deep-seated decolle- ment levels (Van Geet et al., 2002; Vilasi et al., 2009). Although burial histories may have varied throughout the entire Ionian Basin, peak burial of the carbonates is estimated to have been reached at depths of 2-4 km and temperatures of 80-100˚C in Early Miocene times (Rigakis and Karakitsios, 1998; Roure et al., 2004; Lacombe et al., 2009; Vilasi et al., 2009). Brittle deformation structures provided reservoir qualities to the Cretaceous to Eocene carbonates (Velaj et al., 1999). Hydrocarbon charge in the Late Miocene and Pliocene was derived from underlying Triassic to Jurassic shales and filled structural traps in strongly deformed parts in the north of the Ionian Zone (Zappaterra, 1994; Prifti and Muska, 2013). The Oligocene flysch deposits that overlie the carbonate reservoir rocks act as an effective seal (Velaj, 2015). Although the sealing cover has largely been removed in the central part of the Ionian Zone, hydrocarbon accumulations may still occur in structural traps associated with the impermeable Triassic evaporites (Vilasi et al., 2009). The Ionian Zone is renowned for hosting the most important petroleum accumulations in Albania (Curi, 1993; Velaj, 2015).

3.3 Methodology An evaluation of the structural development of the naturally fractured Cretaceous to Eocene carbonate rocks from the Mali Gjere may allow for establishing chronological relationships of distinct generations of calcite fracture infill and tracking the hydrogeological system through time. Orientations of fractures and stylolites were measured in seven distinct outcrops (Figure 3.1B; Table 3.1) using an iPad equipped with GeoID (Lee, 2014), a professional application for measuring geological structures. Fractures were subdivided in barren ones (joints) and infill-bearing ones (veins), with the extent of fracture infill being documented for the veins (i.e., infill width). The studied outcrops are all part of the upper part of the sequence of Cretaceous to Eocene calciturbidites (i.e., Upper Cretaceous to Eocene). An outcropping pavement or vertical section of typically 30-100 m2 (depending on structure density) was delimited and the structures within it were documented. Measurements were plotted in equal-area lower hemisphere projections using OSXStereonet (Cardozo and Allmendinger, 2013) to discern different structural sets. Abutment relationships and physical intersections between structural sets, referred to as cross-cutting relationships, were documented to determine relative time constraints for the structural sets. Thirty-four hand samples of veins were retrieved from the same outcrops studied in the structural analysis. Collected vein samples cover the various phases of 50 Chapter 3 fracture infilling recognized in the Mali Gjere. Isotope analyses of calcite vein cements and fluid inclusion water were performed following the analytical protocols as described in Chapter 1. Apart from the isotope analyses, thin sections were produced from 32 vein samples for petrographic analysis of vein textures, fluid inclusion assemblages and cross-cutting relationships. The petrographic study was performed using a polarizing microscope (Nikon Polarizing Microscope Eclipse 50i POL) and a confocal laser scanning microscope (Olympus Confocal Microscope - FV3000).

Table 3.1 Geographic coordinates and bedding orientations (dip direction/dip angle) per outcrop location (see Figure 3.1B)

Outcrop Coordinates Bedding Bodrishtë 39°53'57.93"N 20°18'24.56"E 013/14 Jorgucat 39°55'49.92"N 20°16'18.84"E 068/41 Muzinë 39°56'39.72"N 20°13'53.26"E 059/22 Frashtan 39°58'16.99"N 20°14'1.52"E 055/20 Terihat 39˚59'25.84"N 20˚12'58.35"E 064/18 Vanister 40°0'50.97"N 20°11'38.05"E 036/20 Mashkullorë 40°6'58.54"N 20°5'27.59"E 037/15

SSE NNW

SSE NNW

064/18

20 m

Figure 3.2 Interpretation of fracture patterns in a gently dipping pavement in the Mali Gjere (39˚59’19.76”N, 20˚13’4.84”E). Recent dissolution activity along fracture planes enhanced the visibility of their traces, which may be continuous over tens of meters. Two systematic sets predominate in this particular outcrop: a pervasive WNW-ESE pre-folding set (blue) and a set of NE-SW syn-folding fractures (black). Measured fracture orientations are presented in the lower hemisphere stereonet projection with thick lines representing averages for the pre-folding (blue) and syn-folding (black) set. Fluid flow evolution in the Albanide fold-and-thrust belt 51

3.4 Results 3.4.1 Structural elements Bedding in the Mali Gjere is consistently northeast dipping (Table 3.1) and devoid of major faulting and folding (Figure 3.1B). Across the seven studied outcrops, the orientations of in total 362 structural elements were measured. Orientations of fractures exhibit a high degree of systematics (Figure 3.2) and are organized in four recurring sets (Figure 3.3). An extensive network of WNW-ESE oriented opening- mode fractures is most prominent and recognized in every outcrop. Substantial calcite fracture infill of up to 2 cm thick occurs in 82% of the fractures that belong to this structural population. In the outcrops of Bodrishtë, Terihat and Vanister, another fracture set with similar infill characteristics is present at a slightly different angle (i.e., more E-W orientation). A third set of opening-mode fractures is oriented NNE-SSW, which is approximately perpendicular to the WNW-ESE fracture set. Although NNE-SSW fracturing is less pervasive, it is recognized in all outcrops except Terihat. Calcite infill occurs in 75% of the NNE-SSW fractures and is usually thinner (< 0.5 cm) compared to the WNW-ESE fractures. Fractures of the WNW-ESE, E-W and NNE- SSW sets display an orthogonal relationship to the bedding. A fourth deformational event recognized in the Mali Gjere is characterized by sub-vertical NE-SW oriented

Bodrishtë Jorgucat Muzinë Frashtan n = 54 n = 41 n = 10 n = 38

Terihat Vanister Mashkullorë Stylolite (plane) n = 46 n = 84 n = 75 Pole to fracture plane

Fracture sets

Syn-folding

Pre-folding 2

Pre-folding 1

Figure 3.3 Orientations of fracture planes (poles) and tectonic stylolitic surfaces (dashed great circles) presented in contoured lower hemisphere stereonet projections for the seven outcrops studied. Squares represent averages of opening-mode fracture sets; different colors correspond to the distinct structural sets. Interpretation of structural sets is based on structure orientations and cross-cutting relationships. Bedding-parallel burial stylolites are plentiful in all outcrops, but not included in the stereonet projections for clarity. Structures that are identified to be pre-folding were rotated back to their original position at the time of development (i.e., before inclination of bedding). 52 Chapter 3 fractures. Fractures related to this stress field were not identified in every outcrop and are predominantly barren; calcite infill occurs in 37% of the fractures and is typically thin (< 0.2 cm). Bedding-parallel stylolites are abundant in all outcrops. Besides, two sets of tectonic stylolites were recorded. One set strikes approximately NNW-SSE and displays a perpendicular relationship to the surface (Figure 3.4D). This set is well- defined and documented in all outcrops except Muzinë, being especially pervasive in the outcrops of Jorgucat and Mashkullorë. A second set of tectonic stylolites is oriented NE-SW and only observed in the outcrops of Mashkullorë and Bodrishtë. Across the outcrops, 33 cross-cutting relations between the various structural sets were recorded, in addition to countless relationships involving burial stylolites (Figure 3.4). WNW-ESE and E-W fractures are systematically cross-cut by the other two fracture sets. The major portion of these fractures (70%) post-date burial stylolites (Figure 3.4C), whereas fractures of the NNE-SSW set display a more ambiguous relationship with burial stylolites. NE-SW fractures systematically cut through the WNW-ESE, E-W and NNE-SSW fracture sets and post-date burial stylolites. Cross- cutting relations between fractures and tectonic stylolites are scarce. Only the outcrops of Mashkullorë and Terihat present clear evidence of veins of both the WNW-ESE and NNE-SSW set being dissolved by – and thus predate – NNW-SSE oriented tectonic stylolites.

A B WNW-ESE

228/67

273/69 NNE-SSW 147/86

C D

Figure 3.4 Compilation of small-scale structures observed in the field. (A) Cross-cutting relationships of three sets of pure-opening mode veins in Frashtan. (B) A WNW-ESE oriented vein that is cut by an NNE-SSW oriented vein in Bodrishtë. (C) A WNW-ESE oriented vein that is dissolved by a bedding- parallel burial stylolite in Vanister. (D) An NNW-SSE striking tectonic stylolite that is oriented vertically with respect to the present-day surface in Bodrishtë. Fluid flow evolution in the Albanide fold-and-thrust belt 53

A B C

1 mm 1 mm 1 mm

D E F

1 mm 0.5 mm 1 mm G H I

1 mm 1 mm 0.5 mm

Figure 3.5 Overview of photomicrographs. (A) Blocky calcite vein cement in a fracture from Muzinë (MU1). (B) Blocky calcite vein cement in a fracture from Muzinë (MU6). The vein is consumed on one side by a tectonic stylolite. (C) Disordered calcite cement in a fracture from Terihat (TE1), possibly due to recrystallization. (D) A rim of small crystals (<< 0.1 mm) at the walls of a vein from Vanister (VA2; XPL image). The vein displays opening along two distinct bands. (E) High-resolution banding in a vein from Vanister (VA3). (F) Elongated crystals growing inward perpendicular to the walls of a vein from Vanister (VA8). (G) Brecciated shards of host rock in a calcite vein from Muzinë (MU2). (H) Blackish aspect of a vein from Terihat (Te6 and Te7). This vein developed in a chert concretion. (I) Growth bands within a calcite crystal in a vein from Bodrishtë (BO6.3).

3.4.2 Vein petrography Mineral infill in fractures related to the three main deformational phases in the Mali Gjere is uniquely composed of calcite. Bitumen staining, although common in fractures in the north of the Kurveleshi belt (Graham Wall et al., 2006), is not observed in fractures from the Mali Gjere. Calcite fracture infill is typically blocky (Figure 3.5A-B) with well-defined growth zones (Figure 3.5I), while the infill of a minor number of samples exhibits a syntaxial texture (Figure 3.5F). A coarsening inward trend is common with rims of small crystals (<< 0.1 mm) at the vein walls (Figure 3.5D). More disorganized fine-grained infill characterized by poorly defined crystals makes up sample Te1 (Figure 3.5C) and is encountered in minor quantities 54 Chapter 3

(< 20%) in two other veins (Mu4 and Fr2). The vein from which Te6 and Te7 were sampled is hosted in a chert concretion and displays a blackish color with poorly defined crystals when studied under a polarizing microscope (Figure 3.5H). Deformational calcite twinning and banded textures are common in veins from all generations. Banding may be minor (Figure 3.5D) or highly frequent consisting of innumerable generations separated by thin bands of host rock (Figure 3.5E). Vein walls are in general matching (Figure 3.5A, 3.5C and 3.5D). Clear indications of shear were not observed in the infill of any of the fracture generations. Brecciated textures with shards of host rock dispersed in the calcite fracture infill occur in three WNW-ESE and one NNE-SSW vein (Figure 3.5G). Stylolitic surfaces developed along the walls of three E-W to WNW-ESE oriented veins (Figure 3.5B). Veins retrieved from the Mali Gjere are dominated by primary fluid inclusions. Primary fluid inclusions are typically small (< 2m m) and mostly occur as isolated inclusions in clusters within single crystals (Figure 3.6A-B) or organized along growth bands (Figure 3.5I). Although primary fluid inclusions are mostly monophase, larger inclusions may be two-phase with a low vapor-liquid ratio (< 0.1; Figure 3.6A). Primary fluid inclusions exhibit an equidimensional rounded or rectangular shape and do not display signs of post-trapping modifications (e.g., decrepitation or leakage). Secondary monophase fluid inclusions, which are arranged along trails that are unrelated to the crystal growth faces (Shelton and Orville, 1980; Smith and Evans, 1984; Bodnar, 2003), can reach a size of up to 10 mm and are less common than the primary ones.

A B

z = 5-6 μm

Gas phase

z = 6-11 μm 10 μm C

z = 10 μm

z = 7-9 μm z = 4-12 μm

50 μm 0.5 cm

Figure 3.6 Petrographic appearance of fluid inclusions hosted in calcite veins from the Mali Gjere. (A) Compiled image of a 3D confocal surface scanning analysis (up to 14 mm depth) of vein Va2 showing the distribution of primary fluid inclusions. Some of the larger inclusions display a minor gas phase (see inset). The scanning depth (z) is given for the different panels in the image. (B) Confocal image of all-liquid primary fluid inclusions in Bo6. (C) Petrographic view of secondary fluid inclusion trails (next to the red lines) in vein Te3. Fluid flow evolution in the Albanide fold-and-thrust belt 55

3.4.3 Isotope data Fluid inclusion isotope data were obtained for 26 calcite vein samples. The 18 2 δ Ow and δ Hw isotope signatures hold a significant positive correlation (Figure 3.7) 18 2 and range from -7.8 to 1.1‰ for δ Ow and -56.0 to -14.7‰ for δ Hw (Table 3.2). Isotope data plot either onto or to the right of the Global Meteoric Water Line (GMWL), which expresses the isotope fractionation of terrestrial meteoric waters as a global average (Craig, 1961). Veins belonging to the WNW-ESE, E-W and NNE-SSW fracture sets display higher isotope values than veins hosted by NE-SW fractures. Complementary carbon and oxygen isotope data were obtained for the calcite of 36 vein samples, as well as for five host rock samples. Average values of the measurements are presented in Table 3.2. Acquired isotope values of the veins 18 13 range from -6.5 to 0.9‰ for δ Oc and from -3.0 to 3.0‰ for δ Cc (Figure 3.8). Host rock samples display consistent isotope values throughout the various outcrops 18 13 (-2.3 to -1.5‰ for δ Oc and from 1.1 to 2.7‰ for δ Cc). Carbon isotope values of the veins closely resemble those of the host rock, apart from a number of samples from Frashtan, Bodrishtë and Mashkullorë, which display a depletion in 13C. Veins are mostly depleted in 18O with respect to the host rock (Figure 3.8). Especially veins 18 from the Frashtan outcrop display remarkably low δ Oc values.

3

3 3

Figure 3.7 Fluid inclusion isotope data of calcite veins from the Mali Gjere. 18 2 Fluid inclusion water displays a distinct correlation in the δ Ow-δ Hw space that is at a lower slope than the GMWL. See Figure 3.10 for interpretation of the data. 56 Chapter 3 Petrographic characteristics Blocky calcite infill with clear twin lamellae infill Blocky calcite One generation of blocky calcite of blocky One generation One generation of blocky calcite of blocky One generation Minor vein banding. Blocky calcite Minor vein One generation of blocky calcite of blocky One generation One generation of blocky calcite of blocky One generation Coarse blocky calcite crystals. Pervasive twin Pervasive crystals. blocky calcite Coarse walls vein on the lamellae. Stylolitic surfaces One generation of blocky calcite with a calcite of blocky One generation curved Intense trend inward coarsening twinning deformational One generation of blocky calcite with intense with intense calcite of blocky One generation twin development Banded vein with a syntaxial to blocky calcite blocky calcite to with a syntaxial Banded vein of twin lamellae Minor development infill. Syntaxial to blocky calcite infill blocky calcite to Syntaxial One generation of blocky calcite with clear calcite of blocky One generation twinning calcite Mostly blocky calcite with a coarsening with a coarsening blocky calcite Mostly and (± 20%) is fine-grained Part trend. inward (recrystallized) disorganized One generation of clean blocky calcite of clean blocky One generation Fine grained disorganized calcite calcite disorganized Fine grained (recrystallization) High frequency vein banding vein High frequency Brecciated shards of host rock in a blocky calcite in a blocky calcite rock of host shards Brecciated infill Vein banding. Blocky calcite infill with clear twin infill banding. Blocky calcite Vein lamellae Syntaxial calcite infill with minor development with minor development infill calcite Syntaxial of twin lamellae Blocky calcite with a coarsening inward trend, trend, inward with a coarsening Blocky calcite banded appearance One generation of calcite infill with a coarse coarse with a infill calcite of One generation rock host Brecciated texture. syntaxial blocky to twinning. calcite (curved) and intense fragments wall vein along one Stylolitic surface Blocky calcite with brecciated host rock rock host with brecciated Blocky calcite fragments Mostly blocky calcite. Part (± 10%) is fine- Part blocky calcite. Mostly (recrystallized) and disorganized grained Banded vein with syntaxial to blocky calcite to with syntaxial Banded vein Blocky and somewhat messy calcite infill calcite messy Blocky and somewhat Vein composed of blocky calcite with clear twin of blocky calcite composed Vein development One generation of blocky calcite with intense with intense calcite of blocky One generation twin lamellae curved Blackish and chaotic infill. Vein hosted in chert Vein hosted Blackish and chaotic infill. One generation of clean blocky calcite of clean blocky One generation Blocky calcite exhibiting curved deformational deformational curved exhibiting Blocky calcite twins Blocky calcite. A stylolitic surface developed on developed surface A stylolitic Blocky calcite. wall one vein Blackish and chaotic infill. Vein hosted in chert Vein hosted Blackish and chaotic infill. High-frequency vein banding with brecciated banding with brecciated vein High-frequency rock of host shards w H 2 16.8 δ - - 27.9 - 40.2 - 28.3 - 36.4 - 35.1 - 25.8 - 22.4 - 35.9 - 14.7 - 24.4 - 40.3 - 20.7 - 37.7 - 56.0 - 46.3 - 34.1 - 26.0 - 47.9 - 41.9 - 36.8 - 19.6 - 29.6 - 41.9 - 28.8 - 23.8 w O 18 0.3 0.7 1.1 - 0.1 - 0.1 - 5.5 - 0.8 - 2.3 - 2.6 - 3.3 - 1.5 - 0.4 - 5.4 - 4.2 - 5.7 - 6.9 - 6.4 - 3.2 - 0.4 - 7.8 - 5.7 - 2.0 - 1.0 - 0.7 - 3.7 - 1.2 δ (‰ vs. VSMOW) Fluid inclusions c O 18 1.0 0.1 - 4.3 - 1.5 - 1.8 - 1.7 - 1.9 - 2.3 - 2.5 - 2.3 - 2.6 - 1.1 - 2.1 - 0.6 - 3.4 - 2.1 - 1.3 - 3.3 - 3.0 - 2.2 - 2.8 - 0.3 - 0.8 - 1.7 - 5.3 - 5.2 - 4.8 - 1.2 - 1.5 - 1.5 - 6.5 - 2.2 - 0.8 - 4.0 - 1.6 - 5.1 - 4.5 - 0.7 - 3.7 - 3.2 δ c Calcite C 13 1.9 1.6 2.7 1.5 1.7 1.1 1.8 1.9 1.5 1.6 1.5 0.4 1.8 1.7 0.8 2.8 1.9 1.6 1.5 1.3 2.3 0.2 2.9 0.5 0.5 1.5 1.8 1.3 2.8 1.1 1.8 2.9 1.5 1.8 2.2 1.1 3.0 (‰ vs. VPDB) - 3.0 - 0.7 - 2.8 δ ? Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Syn-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 2 Pre-folding Syn-folding Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Pre-folding 2 Pre-folding Syn-folding Syn-folding Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Syn-folding Syn-folding Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Syn-folding Syn-folding Pre-folding 1 Pre-folding ? Pre-folding 1 Pre-folding Pre-folding 2 Pre-folding ? Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Fracture set 263/40 291/51 178/78 218/61 191/56 306/79 211/73 183/80 168/80 171/80 184/75 330/76 079/11 280/80 190/78 323/78 144/88 146/84 281/79 214/71 327/81 327/81 264/67 349/80 195/72 224/72 308/85 149/78 184/75 310/70 224/72 273/69 263/40 240/90 248/77 Orientation 221/66 Terihat Jorgucat Terihat Mashkullorë Jorgucat Terihat Mashkullorë Bodrishtë Terihat Vanister Bodrishtë Muzinë Terihat Vanister Bodrishtë Muzinë Terihat Vanister Bodrishtë Muzinë Frashtan Frashtan Frashtan Vanister Bodrishtë Muzinë Frashtan Vanister Bodrishtë Jorgucat Muzinë Frashtan Terihat Bodrishtë Bodrishtë Outcrop Muzinë TE6 Host rock Bodrishtë rock Host Host rock Muzinë rock Host Host rock Frashtan rock Host Host rock Terihat rock Host Host rock Vanister rock Host JG1b TE5 MA2 JG1a TE4 MA1 BO7 TE3 VA9 BO6.3 MU5 TE2 VA8 BO6.2b MU4 TE1 VA3 BO6.2a MU3 FR3.1 FR3.2 FR4 VA2 BO5 MU2 FR2 VA1 BO4 JG2 MU1 FR1 TE7 Overview of isotope data and petrographic observations of calcite vein samples and five host rock samples. The three phases of fracture fracture of phases three The samples. rock NNE-SSW host subsequent five to and 2’ samples vein ‘pre-folding calcite observationsof petrographic and data isotope of veins, Overview oriented 3.2 E-W Table and WNW-ESE burial early of set the to referring 1’ ‘pre-folding with covered, are infilling ‘syn-folding’ oriented to orientations veins the and are latest as Vein generation measured in of the NE-SW field oriented and veins related to LPS. angle’ convention. ‘dip direction/dip in the given Sample BO1 MU6 BO2 Fluid flow evolution in the Albanide fold-and-thrust belt 57 Petrographic characteristics Blocky calcite infill with clear twin lamellae infill Blocky calcite One generation of blocky calcite of blocky One generation One generation of blocky calcite of blocky One generation Minor vein banding. Blocky calcite Minor vein One generation of blocky calcite of blocky One generation One generation of blocky calcite of blocky One generation Coarse blocky calcite crystals. Pervasive twin Pervasive crystals. blocky calcite Coarse walls vein on the lamellae. Stylolitic surfaces One generation of blocky calcite with a calcite of blocky One generation curved Intense trend inward coarsening twinning deformational One generation of blocky calcite with intense with intense calcite of blocky One generation twin development Banded vein with a syntaxial to blocky calcite blocky calcite to with a syntaxial Banded vein of twin lamellae Minor development infill. Syntaxial to blocky calcite infill blocky calcite to Syntaxial One generation of blocky calcite with clear calcite of blocky One generation twinning calcite Mostly blocky calcite with a coarsening with a coarsening blocky calcite Mostly and (± 20%) is fine-grained Part trend. inward (recrystallized) disorganized One generation of clean blocky calcite of clean blocky One generation Fine grained disorganized calcite calcite disorganized Fine grained (recrystallization) High frequency vein banding vein High frequency Brecciated shards of host rock in a blocky calcite in a blocky calcite rock of host shards Brecciated infill Vein banding. Blocky calcite infill with clear twin infill banding. Blocky calcite Vein lamellae Syntaxial calcite infill with minor development with minor development infill calcite Syntaxial of twin lamellae Blocky calcite with a coarsening inward trend, trend, inward with a coarsening Blocky calcite banded appearance One generation of calcite infill with a coarse coarse with a infill calcite of One generation rock host Brecciated texture. syntaxial blocky to twinning. calcite (curved) and intense fragments wall vein along one Stylolitic surface Blocky calcite with brecciated host rock rock host with brecciated Blocky calcite fragments Mostly blocky calcite. Part (± 10%) is fine- Part blocky calcite. Mostly (recrystallized) and disorganized grained Banded vein with syntaxial to blocky calcite to with syntaxial Banded vein Blocky and somewhat messy calcite infill calcite messy Blocky and somewhat Vein composed of blocky calcite with clear twin of blocky calcite composed Vein development One generation of blocky calcite with intense with intense calcite of blocky One generation twin lamellae curved Blackish and chaotic infill. Vein hosted in chert Vein hosted Blackish and chaotic infill. One generation of clean blocky calcite of clean blocky One generation Blocky calcite exhibiting curved deformational deformational curved exhibiting Blocky calcite twins Blocky calcite. A stylolitic surface developed on developed surface A stylolitic Blocky calcite. wall one vein Blackish and chaotic infill. Vein hosted in chert Vein hosted Blackish and chaotic infill. High-frequency vein banding with brecciated banding with brecciated vein High-frequency rock of host shards w H 2 16.8 δ - - 27.9 - 40.2 - 28.3 - 36.4 - 35.1 - 25.8 - 22.4 - 35.9 - 14.7 - 24.4 - 40.3 - 20.7 - 37.7 - 56.0 - 46.3 - 34.1 - 26.0 - 47.9 - 41.9 - 36.8 - 19.6 - 29.6 - 41.9 - 28.8 - 23.8 w O 18 0.3 0.7 1.1 - 0.1 - 0.1 - 5.5 - 0.8 - 2.3 - 2.6 - 3.3 - 1.5 - 0.4 - 5.4 - 4.2 - 5.7 - 6.9 - 6.4 - 3.2 - 0.4 - 7.8 - 5.7 - 2.0 - 1.0 - 0.7 - 3.7 - 1.2 δ (‰ vs. VSMOW) Fluid inclusions c O 3.7 18 1.0 0.1 - 4.3 - 1.5 - 1.8 - 1.7 - 1.9 - 2.3 - - 2.5 - 2.3 - 2.6 - 1.1 - 2.1 - 0.6 - 3.4 - 2.1 - 1.3 - 3.3 - 3.0 - 2.2 - 2.8 - 0.3 - 0.8 - 1.7 - 5.3 - 5.2 - 4.8 - 1.2 - 1.5 - 1.5 - 6.5 - 2.2 - 0.8 - 4.0 - 1.6 - 5.1 - 4.5 - 0.7 - 3.2 δ c Calcite C 13 1.9 1.6 2.7 1.5 1.7 1.1 1.8 1.9 1.5 1.6 1.5 0.4 1.8 1.7 0.8 2.8 1.9 1.6 1.5 1.3 2.3 0.2 2.9 0.5 0.5 1.5 1.8 1.3 2.8 1.1 1.8 2.9 1.5 1.8 2.2 1.1 3.0 (‰ vs. VPDB) - 3.0 - 0.7 - 2.8 δ ? Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Syn-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 2 Pre-folding Syn-folding Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Pre-folding 2 Pre-folding Syn-folding Syn-folding Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Syn-folding Syn-folding Pre-folding 2 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Syn-folding Syn-folding Pre-folding 1 Pre-folding ? Pre-folding 1 Pre-folding Pre-folding 2 Pre-folding ? Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Pre-folding 1 Pre-folding Fracture set 263/40 291/51 178/78 218/61 191/56 306/79 211/73 183/80 168/80 171/80 184/75 330/76 079/11 280/80 190/78 323/78 144/88 146/84 281/79 214/71 327/81 327/81 264/67 349/80 195/72 224/72 308/85 149/78 184/75 310/70 224/72 273/69 263/40 240/90 248/77 Orientation 221/66 Overview of isotope data and petrographic observations of calcite vein samples and five host rock samples. The three phases of fracture fracture of phases three The samples. rock host five and samples vein calcite observationsof petrographic and data isotope of Overview Terihat Jorgucat Terihat Mashkullorë Jorgucat Terihat Mashkullorë Bodrishtë Terihat Vanister Bodrishtë Muzinë Terihat Vanister Bodrishtë Muzinë Terihat Vanister Bodrishtë Muzinë Frashtan Frashtan Frashtan Vanister Bodrishtë Muzinë Frashtan Vanister Bodrishtë Jorgucat Muzinë Frashtan Terihat Bodrishtë Bodrishtë Outcrop Muzinë TE6 Host rock Bodrishtë rock Host Host rock Muzinë rock Host Host rock Frashtan rock Host Host rock Terihat rock Host Host rock Vanister rock Host JG1b TE5 MA2 JG1a TE4 MA1 BO7 TE3 VA9 BO6.3 MU5 TE2 VA8 BO6.2b MU4 TE1 VA3 BO6.2a MU3 FR3.1 FR3.2 FR4 VA2 BO5 MU2 FR2 VA1 BO4 JG2 MU1 FR1 TE7 infilling are covered, with ‘pre-folding 1’ referring to the set of early burial WNW-ESE and E-W oriented veins, ‘pre-folding 2’ to subsequent NNE-SSW subsequent to 2’ ‘pre-folding veins, oriented E-W and WNW-ESE burial early of set the to referring 1’ ‘pre-folding with covered, are infilling ‘syn-folding’ oriented to orientations veins the and are latest as Vein generation measured in of the NE-SW field oriented and veins related to LPS. angle’ convention. ‘dip direction/dip in the given Table 3.2 Table Sample BO1 MU6 BO2 58 Chapter 3

3

Host rock δ 

3 3 3

Figure 3.8 Carbon and oxygen isotope data of vein and host rock samples from different outcrop locations along the Mali Gjere. Squares and crosses represent vein and host rock samples, respectively, and correspond in color per outcrop. Fluids acquired DIC through host rock dissolution as demonstrated by the similarity in 13 18 δ Cc between veins and host rock. Depletions in O with respect to host rock are related to variable degrees of fluid-DIC oxygen isotope equilibration under meteoric conditions and/or elevated temperatures. Water-rock interactions (WRI) or vein precipitation from isotopically enriched evaporation fluids could explain enrichments in 18O with respect to the host rock.

3.5 Discussion 3.5.1 Structural evolution of the Mali Gjere 3.5.1.1 Early-burial fracturing events Early burial diagenesis caused a loss of porosity and permeability of the Cretaceous to Eocene carbonate rocks deposited in the Ionian Basin (Van Geet et al., 2002; Vilasi et al., 2006). Multiple events of fracturing were subsequently key for causing periodic increases in rock permeability (Figure 3.9). Cross-cutting relationships between the different structural sets that are present in the Mali Gjere reveal a clear sequence of deformational events. Bedding-parallel burial stylolites are especially useful time markers due to their development before the onset of intense thrusting in the Oligocene (Swennen et al., 2000). Fractures that are oriented WNW- Fluid flow evolution in the Albanide fold-and-thrust belt 59

ESE and the minor set of E-W fractures are interpreted to have formed first because they systematically pre-date the other fracture sets and burial stylolites. Although the precise chronology of the WNW-ESE and E-W fractures is unclear, their similar orientations and infill characteristics support a similar time of formation. Because these fractures systematically pre-date burial stylolites, they are interpreted to have formed during early burial in a foreland basin setting, which is in line with their bedding-orthogonal position. NNE-SSW oriented tensile fractures typically post-date the WNW-ESE fractures and pre-date NE-SW fractures. Cross-cutting relationships between the NNE-SSW fracture generation and burial stylolites are ambiguous, suggesting that they formed synchronously under a vertically-oriented s1. Along with their layer- orthogonal orientation, this suggests that the NNE-SSW fractures developed at a later stage during burial compared to the first phase of brittle deformation, but still before inclination of bedding with the carbonates being situated in a foreland basin setting. In the structural analysis (Figure 3.3), orientations of fractures and stylolites of the pre-folding fracture sets were returned to their original position at the time of formation (i.e., before inclination of bedding) to account for slight variations in bedding orientation throughout the Mali Gjere.

3.5.1.2 Burial fracturing in foreland basins The orientations of the two main pre-folding fracture assemblages (WNW- ESE and NNE-SSW) are similar to orientations of pre-folding fractures that were previously recognized elsewhere in the Ionian Zone (Graham Wall et al., 2006; Lacombe et al., 2009; Vilasi et al., 2009), suggesting that a regionally uniform stress regime persisted in the foreland basin. Syn-subsidence fracturing is an important process in structurally shaping carbonate rocks in a wide array of settings, ranging from shallow-water carbonate platforms (Bertotti et al., 2017; Lavenu and Lamarche, 2017; Nooitgedacht et al., 2018) to basinal carbonate deposits (Vitale et al., 2012). Pervasive fracturing during early burial in carbonates is facilitated by the rapid embrittlement of carbonate rocks (Lamarche et al., 2012; McNeill and Eberli, 2012; Lavenu et al., 2014). Rock failure may occur even in the absence of tectonic stresses as soon as the overburden-induced stress is sufficiently high, which may occur already before complete lithification at shallow burial depths below 500 m (Lavenu and Lamarche, 2017). Pinpointing the nature of the controlling stresses during syn-burial fractur- ing of the Ionian Basin carbonates is difficult due to the complexity of fold-and-thrust belt systems. The architecture of the collisional system ranging from the foreland basin to the orogenic wedge may encompass lateral variations in stress regimes. Whereas compressional stresses prevail within the fold-and-thrust belt, the foreland basin is typically dominated by flexural stresses that are related to lithospheric bend- ing provoked by tectonic loading in the hinterland (Beaumont, 1981; DeCelles and Giles, 1996; Ferket et al., 2003). Strata in a foreland basin may experience switches in stress fields as the thrust front advances, and different organizations of structural 60 Chapter 3 elements may be expected depending on the position of the rock within the devel- oping fold-and-thrust system (Tavani et al., 2015). The main pre-folding fracture sets are roughly orthogonal (WNW-ESE and NNE-SSW). Orthogonal fracture assemblages are typical for the early orogenic deformation phase in foreland fold-and-thrust belts and thought to be associated with perpendicular stretching directions throughout the foreland basin (Tavani et al., 2015; Figure 3.11). In the forebulge and outermost part of the foredeep, flexural stresses related to bending of the lithosphere predominate (Bradley and Kidd, 1991), which typically leads to the development of longitudinal (i.e., parallel to the trend of the foreland basin) bed-perpendicular opening joints (Billi and Salvini, 2003; Lash and Engelder, 2007). As rocks migrate towards the thrust front, transverse (i.e., perpendicular to the trend of the foreland basin) joints become more common and develop as a second early orogenic assemblage (Tavani et al., 2015). Transverse joints develop in association with along-strike curving and stretching of the foredeep close to the thrust front, which may yield a s3 that is oriented parallel to the foredeep trend (Quintà and Tavani, 2012). Although transverse joints are most common close to the thrust front, they may also form as cross-joints in the flexuring regime where longitudinal joints predominate (Destro, 1995; Quintà and Tavani, 2012). Since the first syn-burial assemblage of WNW-ESE fractures is roughly parallel to the trend of the Mali Gjere, it could possibly represent a longitudinal fracture set. The subsequent orthogonal NNE-SSW fracture set would then represent a set of transverse joints. We do stress that his interpretation must be treated with caution due to the complexity of fracture networks in fold-and-thrust belt settings.

Early diagenesis Fracturing Fracture infilling (i.e., fluid flow) Burial stylolitization Deposition Tectonic stylolitization K Paleocene Eocene Oligocene Miocene < 5 Ma

Pre-folding Syn-folding depth

Burial curve ~3km

Figure 3.9 Paragenetic sequence showing the three events of fracturing and fluid flow that affected the Cretaceous to Eocene carbonate sequence. The time spans shown are approximations. Fracturing and fluid flow were most intense during the first deformational event, taking place in the early burial stage. Intensities of small- scale fracturing and fluid flow decrease through time. Fluid flow evolution in the Albanide fold-and-thrust belt 61

3.5.1.3 LPS fracturing and tectonic stylolitization NE-SW oriented fractures abut on and cross-cut the other fracture sets and are, therefore, interpreted to have formed last. They post-date burial stylolites, and their orientation is parallel to the direction of tectonic transport during thrusting in the Ionian Zone (Velaj et al., 1999). Hence, NE-SW fractures are most likely associated with layer-parallel shortening (LPS) from the Oligocene onward. When the rocks approach the thrust front, the tectonic stress component increases and the s1 shifts from a vertical position to a horizontal position perpendicular to the direction of tectonic transport, which leads to the development of sub-vertical extensional joints parallel to the s1 (Tavani et al., 2015). Syn-folding fractures related to NE-SW oriented LPS are recognized throughout the entire Ionian Zone (Graham Wall et al., 2006; Lacombe et al., 2009).The typically barren aspect and low density of the NE-SW fractures might indicate that no interconnected fracture network was established that could facilitate fluid flow. Tectonic stylolites encountered in the Mali Gjere are mostly perpendicular to the NE-SW fractures, pointing to a cogenetic relation (i.e., formed under the same stress regime with a sub-horizontal s1) (Fletcher and Pollard, 1981). The fact that tectonic stylolites are sub-vertical with respect to the surface (Figure 3.4D) supports a late genesis after tilting of bedding during thrust emplacement. Tectonic stylolitization may be an effective deformation mechanism for accommodating strain during LPS in carbonate rocks (Alvarez et al., 1978; Evans et al., 2003). Inherited veins and joints that are perpendicular to the shortening direction (longitudinal) may be reactivated as preferential sites for the development of tectonic stylolites and facilitate LPS (Figure 3.5B; Railsback and Andrews, 1995). In the outcrop of Mashkullorë, a clearly defined set of NE-SW oriented tectonic stylolites deviates from the predominant orientation of tectonic stylolites. This set is perpendicular to the WNW-ESE fracture set; however, a relation would be improbable considering the extensional regime in which these fractures probably formed. Because the bedding is horizontal at Mashkullorë, it cannot be established whether these stylolites are related to a pre- or post-folding tectonic event. Tectonic stylolites recorded by (Vilasi et al., 2009) in outcrops near Kremenara, Saranda and Kelçyra in the Ionian Zone fall apart in the same two directions as we observe. However, also in this study, there was no convincing evidence for the timing and origin of the NE-SW oriented stylolite set.

3.5.2 Fluid characteristics 3.5.2.1 Fluid inclusion isotope data Petrographic analyses indicate that vein cements are largely original (e.g., syntaxial structures, vein banding, inward coarsening of crystals) and dominated by primary fluid inclusions. The only exceptions include i) the disorganized infill of sample Te1, which is probably the result of recrystallization in relation to late- stage reactivation during active thrusting, and ii) the blackish infill of the vein of 62 Chapter 3 samples Te6 and Te7; also the nature and timing of this chert-hosted vein remained unclear. Even though fluid inclusion water in the remaining vein samples seems to be predominantly of primary origin, certain post-entrapment alteration processes may drive shifts in fluid inclusion isotope signatures. In-situ isotope alteration of fluid inclusion water in high-temperature settings may occur due to bulk diffusion of water (Bakker, 2009) or oxygen isotope exchange between entrapped fluids and host minerals (Rye and O’Neil, 1968). However, at the estimated precipitation temperatures of veins in the Albanides of below 70˚C (Swennen et al., 2000; Vilasi et al., 2009), the rates of these processes are thought to be too low to cause significant alteration (Rye and O’Neil, 1968; Bakker, 2009). Various studies indeed demonstrated that isotope signatures of fluid inclusion water may be preserved in ancient vein systems over geologic time scales (Vityk et al., 1993; Naden et al., 2003; De Graaf et al., 2017). Isotope data of vein-hosted fluid inclusions from the Mali Gjere exhibit a 2 18 positive correlation between δ Hw and δ Ow with lowest values matching a meteoric fluid endmember (Figure 3.10). A hypothesis featuring the mixing of meteoric water with an isotopically enriched end-member would most readily explain this isotope pattern. The meteoric recharge would have isotope values of approximately -7.5‰ 18 2 for δ Ow and -50‰ for δ Hw. Calcite isotope data could hold an inverted J-shape 18 13 towards low δ Oc and δ Cc values (Figure 3.8), which would support meteoric

3

3 3

Figure 3.10 Fluid inclusion isotope data of calcite veins from the Mali Gjere reveal the isotope evolution of circulating fluids in the Albanide fold-and-thrust belt. Early pre-folding veins probably precipitated from a mixture of meteoric water and evolved marine formation fluids, whereas later stage syn-folding veins precipitated predom- inantly from the meteoric fluid endmember. Sample TE1 deviates from the isotope evolution due to post-depositional alteration (Figure 3.5C). Fluid flow evolution in the Albanide fold-and-thrust belt 63 diagenetic conditions (Lohmann, 1988). Based on isotope data of early-stage calcite veins that are presently exposed in the adjacent Berati thrust sheet, (Vilasi et al., 2009) also inferred that meteoric-derived fluids could be circulating in the Cretaceous to Eocene carbonates already during the burial stage. Exact isotope values of the other fluid end-member cannot be constrained but would lie somewhere along the extension of the isotope evolution trend in Figure 3.10. This fluid end-member most probably corresponds to marine formation or connate water that experienced isotope enrichment due to either sub-aerial evaporation in a shallow-water environment or high-temperature water-rock interaction (WRI) at depth (Horita, 2005). Evaporated marine fluids could have been sourced from formation waters trapped in Triassic evaporites, which are abundantly present in the subsurface. 87Sr/86Sr isotope data of calcite veins in Cretaceous to Eocene carbonate rocks from the adjacent Berati thrust sheet indicate that interaction between evaporites and fluids occurred already during the earliest flow events (Vilasiet al., 2009). The 18 minor enrichments in δ Oc with respect to host rock that some vein calcites display could also be the result of interaction of paleo-fluids with evaporites (Swennen et al., 2000; Vilasi et al., 2009). Interaction with evaporites is still observed in the salinity signature of present-day groundwater in the Albanides (Eftimi and Frashëri, 2016). Positive correlations between δ18O and δ2H similar to those observed in Figure 3.10 are common for groundwater reservoirs in modern sedimentary basins and are often explained by mixing of meteoric water with saline formation waters (e.g., Knauth et al., 1980; Moldovanyi et al., 1993; Taylor, 1997; Sanjuan et al., 2016). Alternative explanations for the isotope data are less plausible. Whereas progressive isotope enrichments may result from boiling (steam loss) of subsurface fluids, this is only feasible in in high-temperature fluids (>> 100˚C) (Truesdellet al., 1977; Darling and Ármannsson, 1989). Precipitation temperatures of veins in the Kurveleshi belt are relatively low (< 70˚C; Swennen et al., 2000; Van Geet et al., 2002), suggesting that fluids in the Albanides did not reach sufficiently great depths during migration for a significant effect of such Rayleigh-type fractionation processes. Moreover, fluid inclusions do not reveal signs of boiling or effervescence (i.e., co-existing gas and liquid inclusions; Moncada et al., 2012). Membrane filtration in clay-rich sediments or exchange reactions with hydroxyl-bearing clay minerals could also force minor modifications of δ18O and δ2H in groundwater (Taylor, 1997; Horita, 2005), but are unlikely to account for the strong isotope variations observed. Hydrogen isotopes, in particular, are assumed to remain largely unaffected by WRI (Clayton et al., 1966; Ohba and Matsuo, 1988). For being a bulk fluid inclusion isotope study, the extent of infiltration of secondary fluids and their effect on the observed isotope signatures cannot be precisely determined. Hence, one hypothesis that cannot be completely excluded is that the trend (partly) reflects the admixture of secondary meteoric fluids to a primary fluid inclusion content with a marine signature. However, clear evidence for abundant secondary fluid inclusion trails is absent in thin sections, whereas the samples with lowest isotope values would have their fluid inclusion content almost completely replaced in this scenario. Altogether, a fluid mixing system is most likely to underlie the recorded fluid inclusion isotope variations. 64 Chapter 3

3.5.3 Oxygen isotope fractionation systematics From thermodynamic laws, it follows that fluid temperatures in an equili- 18 18 brated system are constrained by: 1) the δ Oc of precipitating calcite; and 2) the δ Ow of the mineral-forming fluid (Kim and O’Neil, 1997; Zheng, 2011). The combination 18 18 of δ Ow and δ Oc may thus provide meaningful precipitation temperatures, when assuming equilibrium precipitation. As for the veins studied here, oxygen isotope fractionation factors correspond to a temperature range of -12 to 39˚C with a mean of 16˚C (Kim and O’Neil, 1997). Whereas the upper temperature limit could theo- retically agree with shallow-crustal fluids, the lowest temperatures are unrealistically low for the paleo-fluid system. Experimental equations for mineral-water oxygen isotope fractionation have previously been identified to produce unrealistically low temperatures in other natural systems (Yan et al., 2012). Even in the slowly growing calcites of Devils Hole – considered a natural laboratory – temperatures calculated using the experimental equation of Kim and O’Neil (1997) are 8˚C lower than actual precipitation tem- peratures (Coplen, 2007). Non-equilibrium fractionation of oxygen isotopes may be commonplace for calcite precipitation in natural systems (Watkins et al., 2014). As for the veins in this study, the calculated precipitation temperatures below 0˚C could point to a system that was not in isotope equilibrium either. A possible driving force behind non-equilibrium kinetic effects on oxygen isotope fractionation in vein systems could be high precipitation rates, which may lead to considerable 18O enrichments in calcite cements (Scholz et al., 2009; Gabitov et al., 2012). In ancient vein systems, however, part of the apparent temperature discrep- ancy could also be due to the late infiltration of (meteoric) waters that are depleted in 18O with respect to the primary fluid inclusion content. Such waters may infiltrate through recrystallization or micro-fracturing under deformational stresses, as evi- denced by secondary fluid inclusion trails. Besides, certain chemical parameters may also put a control on oxygen isotope fractionation apart from temperature and cause an apparent out-of-equilibrium system, including pH (Beck et al., 2005; Dietzel et al., 2009) and fluid salinity (Truesdell, 1974; Hu and Clayton, 2003). Altogether, the precise reasons for the anomalous fractionation factors remain unclear. In any case, oxygen isotope fractionation factors between bulk fluid inclusion water and calcite from this study are not suitable for establishing reliable temperature and depth estimates for vein precipitation.

3.5.4 Fluid flow in the Albanide fold-and-thrust belt Vein-forming fluids in the Ionian Zone are composed of a mix of meteoric 13 and marine formation/connate waters. The resemblance in δ Cc between veins and host rock indicates that mineralizing fluids acquired dissolved inorganic carbon (DIC) species mainly through dissolution (e.g., stylolitization) of the hosting Cretaceous to Eocene carbonate rocks. Depletion in 13C with respect to host rock as demonstrated by some vein samples could be related to an input of carbon from Fluid flow evolution in the Albanide fold-and-thrust belt 65 external sources (e.g., hydrocarbons). Degradation of organic matter may also 13 introduce C-depleted methane and CO2 into circulating fluids and significantly 13 lower δ Cc signatures (Stahl, 1979; McManus and Hanor, 1988; Prikryl et al., 1988). Observed vein banding probably results from a crack-and-seal type of infilling, in which fractures repeatedly break open along the vein walls (Ramsay, 1980). Vein banding and syntaxial cements (Figure 3.5F), which were observed in fractures of all generations, suggests that fluid flow and fracture infilling occurred concurrently with fracture opening for each phase of deformation. Matching vein walls (i.e., devoid of significant dissolution; Figure 3.5A, 3.5C and 3.5D) point to a short residence time of fluids in the fracture networks and a transient character of fluid flow. Brecciated textures are common in early burial veins (Figure 3.5G) and indicate that high hydraulic pressures exerted by underlying fluid reservoirs may have facilitated fracturing (Roberts and Nunn, 1995; Cosgrove, 2001; Sibson, 2003).

-7.5‰ δO SW -50‰ δH NE Forebulge Foredeep Orogenic wedge

1 3 2

σ σ σ σ σ σ 1 2 3 σ σ σ

Figure 3.11 Schematic representation of large-scale fluid flow patterns in the Albanide fold-and-thrust belt. Meteoric water is charged in topographically high areas of the orogenic wedge and driven basin-ward due to compressional tectonic forces and the elevated position of the hydraulic head. Meteoric fluids ascend as soon as the fracture-controlled permeability of the Cretaceous to Eocene carbonate rocks was sufficiently high. The carbonates experienced various stress regimes from the moment of deposition until incorporation into the fold-and-thrust belt. Foreland basins are typically characterized by the development of opening-mode fractures parallel and perpendicular to the trend of the foreland (Tavani et al., 2015; blocks 1-2). Late-stage fractures develop parallel to the direction of tectonic transport and are associated with orthogonal tectonic stylolites (block 3). The vertical scale is exaggerated in the figure. 66 Chapter 3

Overpressured fluid reservoirs may be installed under an increasing overburden in the case of the pre-folding fluid system. Regional compressional stresses related to thrusting may also contribute to elevating hydraulic pressures in confined fluid reservoirs (Oliver, 1986; Ge and Garven, 1989). High fluid pressures during vein formation in the Cretaceous to Eocene carbonates of the Ionian Zone were also identified by(Van Geet et al., 2002; Lacombe et al., 2009; Vilasi et al., 2009). The evidence for hydraulic overpressures suggests that fluids were derived from a deeper- seated reservoir. Lateral flow of meteoric water through this reservoir from a terrestrial recharge zone is indispensable because the carbonates were being covered by Oligocene flysch deposits in an underfilled foreland basin during the syn-subsidence events of fluid flow. Pinpointing the rock formation(s) that accommodated lateral flow of fluids is complicated as multiple deep-seated porous formations exist in the Ionian Zone including Triassic breccias and Jurassic limestones (Zelilidis et al., 2015). We propose a model with gravity-driven basin-ward flow of rainwater that is charged in topographic highs shaped by active fold-and-thrust tectonics in the hinterland (Figure 3.11). Large-scale compression induced by advancing thrust tectonics may elevate pore pressures in the foreland to facilitate basin-ward flow of groundwater (Oliver, 1986; Ge and Garven, 1989). Basin-ward expulsion of fluids is commonly observed in fold-and-thrust belt settings worldwide (Machel and Cavell, 1999; Ferket et al., 2003; Lynch and Van der Pluijm, 2017). Meteoric fluids that are charged in topographic highs in the hinterland may be transported far into the foreland basin as demonstrated in vein-based studies in various fold-and-thrust belt settings, including the Appalachians (Evans and Battles, 1999; Kirkwood et al., 2000), the Rocky Mountains (Beaudoin et al., 2011, 2014) and the Mexican fold-and-thrust belt (Fitz-Diaz et al., 2011). Similar to what we infer for the Albanides, the charged meteoric fluids in these studies are identified to mix at depth with saline formation fluids and deposit vein cements upon fracturing in relation to lithospheric flexure. In the present-day Ionian Zone, gravity-driven flow and upwelling of meteoric waters is still recognized (Eftimi and Frashëri, 2016), indicating that the large-scale fluid flow system with outflowing meteoric water has remained temporally stable in the Albanide fold-and-thrust belt. Early stage pre-folding veins and late stage syn-folding veins plot in distinct fields along the observed fluid inclusion isotope trend (Figure 3.10). The meteoric water endmember becomes progressively more dominant through time, demonstrating that marine formation waters were gradually flushed out of the system due to the continued throughput of meteoric recharge. Progressive shortening of fluid migration pathways as the carbonate rocks advanced towards the thrust front (i.e., approached the zone of meteoric recharge) may have contributed to the increasing dominance of meteoric fluids through time. The inferred evolution of fluids is supported by overall lower fluid inclusion salinities of late stage veins in the Ionian Zone (Vilasi et al., 2009). A fluid evolution with meteoric waters gaining in dominance through time was also inferred for the Mexican fold-and-thrust belt by (Fitz-Diaz et al., 2011), who recorded a similar decrease in δ2H of circulating fluids from the earliest to the latest stages of deformation. Fluid flow evolution in the Albanide fold-and-thrust belt 67

Berati Podgoran (Pd)

Kelçyrë (Ke)

Sheper (Sh)

Skore (Sk) Çika Kurveleshi

Porto Palermo (PP) Poliçan (Po) Borsh (Br) Shën Vasil (SV)

Sarandë (Sa) N Metoq (MQ) Ksamil (Ks) Ionian Sea 20 km (Sp) Greece Shalës (SL)

Figure 3.12 In addition to the seven outcrops studied in the Kurveleshi belt, we studied veins in eight outcrops in the external-most Çika belt and six outcrops in the internal-most Berati belt. The abbreviations used for the outcrops are given in parentheses. The map view is slightly enlarged compared to Figure 3.1B.

3.6 Spatial fluid variations throughout the Ionian Zone Fluids circulating in fracture networks within the Kurveleshi belt display a clear temporal evolution with meteoric fluids gaining in dominance through time. To assess spatial variations in fluid flow pathways throughout the entire fold-and- thrust belt setting, the study area was expanded in the work of Nooigedacht et al. (in prep.) to the adjacent Çika and Berati belts, which are along with the Kurveleshi belt the main structural units of the Ionian Zone. In total, vein and fracture assemblages were studied in twelve outcrops of Ionian Basin sediments spread along the Çika and Berati belts (Figure 3.12). The sedimentological aspect of the rocks is highly consistent throughout the entire Ionian Zone (i.e., low-permeability fine-grained calciturbidites). Fracture patterns within the studied outcrops in the Çika and Berati 68 Chapter 3

belt reflect the same stress fields as inferred for the Kurveleshi belt, supporting the notion that regional forces controlled deformation of the Ionian Zone carbonates. A total of 39 vein samples were collected from the Çika and Berati belt for isotope 2 18 analyses; a summary of the δ Hw and δ Ow measurements is presented in Table 3.3. When fluid inclusion isotope data of early burial veins are plotted per thrust sheet, a remarkable grouping shows up (Figure 3.13). Fluids in the external Çika belt are dominated by the meteoric fluid end-member, while vein-forming fluids in the internal Berati belt are shifted considerably more towards the evaporite end-member. The isotope composition of fluids in the Kurveleshi belt in turn is intermediate between the bordering Çika and Berati belts. This trend probably reflects a gradual increase of the evaporite signal towards the thrust front, where tectonic forces and deformation are elevated, causing widespread expulsion of connate waters from the evaporites. An obvious implication is that the basin-ward flow of meteoric waters was not guided through the evaporites, but rather through a shallower overlying aquifer, which experienced episodic inputs of upwelling fluids from the evaporites.

3

Figure 3.13 Fluid inclusion isotope signatures of pre-folding veins from the three thrust belts in the Ionian Zone. Fluids in the external Çika belt are shifted towards the meteoric side of the mixing system, while the internal Berati belt is dominated by connate seawater evaporation fluids. Vein-forming fluids in the Kurveleshi belt are intermediate. The data reveal an increase of the evaporite signal towards the thrust front. Fluid flow evolution in the Albanide fold-and-thrust belt 69

Since syn-folding fractures with substantial calcite infill are scarce in the Çika and Berati belts, there is not sufficient fluid inclusion isotope data of the syn-folding stage to establish spatial variations in fluid flow during active thrusting. During this stage, however, variable fluid isotope signatures might be expected in relation to the compartmentalization of the fold-and-thrust belt. Each thrust unit may constitute a separate hydrogeological unit with distinct fluid flow characteristics. Although the thrust sheets are all positioned within the same fold-and- thrust complex and subject to the same regional tectonic forces, the composition of circulating fluids may vary considerably between them. Along with the clear temporal evolution in fluid flow patterns as revealed from the Kurveleshi belt, this demonstrates that the prediction of fluid flow in fold-and-thrust belt settings is highly challenging. Nevertheless, the general evolution from large-scale basin-ward migration pathways and pervasive fluid flow during early burial to more localized flow of meteoric fluids during later stage tectonic activity may be widely applicable for carbonate rock formations in fold-and-thrust belt settings.

Table 3.3 Fluid inclusion isotope signatures of veins in the Ionian Basin carbonates of the Berati and Çika thrust sheets. Vein orientations are given in the ‘dip direction/dip angle’ convention.

Berati Belt Çika Belt

18 2 18 2 Sample Orientation Fracture set δ Ow δ Hw Sample Orientation Fracture set δ Ow δ Hw Sk1 027/46 Pre-folding 1 1.1 -20.2 SL1 210/35 Pre-folding 1 -3.7 -30.2 Po2 176/51 Pre-folding 1 0.5 -15.4 SL2 210/35 Pre-folding 1 -3.5 -42.4 Po3a 188/75 Pre-folding 1 -0.2 -28.1 SL3 210/35 Pre-folding 1 -1.9 -26.5 Po3b 313/24 Pre-folding 2 -1.1 -26.2 SL4 210/35 Pre-folding 1 -3.2 -35.1 Po4 195/75 Pre-folding 1 -0.8 -14.7 Sp2 236/68 Pre-folding 1 -3.8 -36.5 Sh1 013/70 Pre-folding 1 -1.8 -27.7 Sp3 244/66 Pre-folding 1 -4.2 -23.5 Sh3 017/67 Pre-folding 1 1.4 -25.9 KS1 241/75 Pre-folding 2 -4.7 -29.2 Sh4 021/80 Pre-folding 1 2.2 -18.7 KS3 237/86 Pre-folding 2 -2.3 -29.7 Sh5 023/77 Pre-folding 1 0.6 -23.8 MQ1 297/66 Pre-folding 2 -4.8 -32.5 Sh6 339/49 Pre-folding 1 1.8 -24.4 MQ2 240/50 Pre-folding 1 -4.8 -34.4 Sh7 312/50 Pre-folding 1 -1.4 -28.3 MQ3 218/65 Pre-folding 1 -5.2 -35.7 Sh8 021/31 Pre-folding 1 -0.1 -20.2 Sa1 213/38 Pre-folding 1 -2.0 -22.5 Sh9 348/26 Pre-folding 1 -0.9 -28.2 Sa2a 270/48 Pre-folding 2 -3.0 -33.1 Ke1 265/65 Pre-folding 2 2.5 -32.4 Sa2b 185/48 Pre-folding 1 -1.6 -18.0 Ke3 216/65 Pre-folding 1 -0.2 -26.9 Sa3 213/38 Pre-folding 1 -3.1 -13.9 Ke4 281/66 Pre-folding 2 3.2 -27.2 Sa4 175/48 Pre-folding 1 -1.0 -14.8 Ke5 210/51 Pre-folding 1 -2.5 -35.2 Sa5 186/40 Pre-folding 1 -3.0 -30.0 Ke6 180/64 Pre-folding 1 1.7 -33.3 SV1 310/45 Pre-folding 1 -3.8 -39.0 Ke7a 226/49 Pre-folding 1 2.8 -27.4 SV2 333/54 Pre-folding 1 -2.3 -37.9 Ke7b 226/49 Pre-folding 1 0.3 -23.8 Br1 209/39 Pre-folding 1 -0.8 -19.3 Ke7c 180/48 Pre-folding 1 1.4 -24.7 Br2 207/57 Pre-folding 1 -2.1 -31.1 Pd1.1 261/53 Pre-folding 2 2.1 -22.1 Br3 028/67 Pre-folding 2 -3.6 -27.6 Pd1.2 261/53 Pre-folding 2 1.4 -13.4 PP1 206/69 Pre-folding 1 -3.0 -14.8 Pd1.3 261/? Pre-folding 2 -2.7 -34.2 Pd2 191/70 Pre-folding 1 0.8 -19.9 Pd4 265/36 Pre-folding 2 -4.2 -29.2 70 Chapter 3

3.7 Conclusions Fluid inclusion isotope data of calcite veins hosted in carbonate rocks from the Ionian Zone of Albania provide unique insight into the evolution of fluid migration pathways throughout the development of the Albanide fold-and-thrust belt. Fluid flow in the studied carbonate rocks is closely associated with events of brittle deformation, which establish discrete pathways (i.e., fractures) allowing for fluids to migrate. Fluids circulating within the carbonates were episodically supplied through upwelling from an underlying confined reservoir. This groundwater 18 reservoir probably comprised a mixture of meteoric water (-7.5‰ for δ Ow and 2 -50‰ for δ Hw) and evolved marine waters derived from deep-seated evaporite occurrences. The contribution of fluids derived from the evaporites is higher towards the thrust front in association with elevated deformation, which is expected to be the main driver for expulsing connate waters from the evaporites. The topographically high hinterland of the orogenic wedge most likely formed a zone of meteoric recharge that fed the deep-seated fluid reservoir. Charged rainwater flowed outward into the foreland basin driven by gravity and gradually flushed out the marine formation waters. As a consequence, fluid inclusions contained in late stage veins, which formed during active thrusting, are dominated by the meteoric water endmember. The model for fluid flow in the Albanide foreland fold-and-thrust belt characterized by the continued basinward expulsion of meteoric waters from an early stage onward applies to many fold-and-thrust belt settings worldwide. A pronounced water-DIC oxygen isotope disequilibrium 18 18 precludes the use of δ Ow/δ Oc pairs as a paleothermometer for vein precipitation. The disequilibrium could be related to kinetic isotope fractionation during vein precipitation and/or secondary infiltration of surface waters.