Carbon and oxygen stable isotope record of upper Kimmeridgian shallow-marine ramp carbonates (Iberian Basin, NE Spain): the imprint of different burial and tectonic histories

DOI: 10.1344/GeologicaActa2021.19.14  C. Sequero, G. Della Porta, B. Bádenas, M. Aurell, 2021 CC BY-SA Carbonate ramp facies; Carbon and oxygen isotopes; Diagenetic resetting; Post-depositional history; Tectonic uplift.

Bulk carbon and oxygen stable isotopes of ancient shallow-marine carbonates can record the effects of multiple palaeoenvironmental factors, but also the imprint of several post-depositional processes, which may alter the original marine isotopic composition. In this study, carbon and oxygen stable isotope analyses were performed on bulk carbonate, bivalve calcitic-shell (Trichites) and calcite vein samples from two stratigraphic sections (Tosos and Fuendetodos, present-day distance 15km), representing proximal inner-and distal mid-ramp environments, respectively, of the uppermost Kimmeridgian ramp facies deposited in the northern Iberian Basin (NE Spain). These successions underwent different diagenetic pathways that altered the primary marine isotopic composition in each section in different ways. Different burial histories, tectonic uplift and a variable exposure to meteoric diagenesis from the end of the Kimmeridgian to the Cenozoic (following Alpine tectonic uplift) are reflected in the different alteration patterns of the carbon and oxygen stable isotope signatures. A significant deviation to lower values in both δ 13 C and δ 18 O is recorded in those carbonates mostly exposed to meteoric diagenesis (distal mid-ramp Fuendetodos section), because of post-depositional tectonic uplift (telogenesis). On the other hand, the deposits mainly affected by burial diagenesis (proximal inner-ramp Tosos section) only record low δ 18 O with respect to expected values for pristine Kimmeridgian marine carbonates. The different burial and tectonic uplift histories of these deposits in each sector, due to their different tectonic evolution in this part of the basin, resulted in a variable degree of diagenetic resetting. However, in spite of the different diagenetic resetting reported of the carbon and oxygen stable isotope signatures in each section, these carbonates show similar cement types in terms of fabrics and cathodoluminescence properties. The diagenetic resetting reported for these carbonates prevents the use of the δ 13 C and δ 18 O records for addressing palaeoenvironmental interpretations, but instead highlights useful features regarding the variable diagenetic overprint of the studied shallow-marine carbonate successions concerning their specific post-depositional history.

INTRODUCTION
Using the carbon and oxygen stable isotope composition of shallow-marine carbonates as proxies for palaeo-seawater, palaeoclimatic and palaeoceanographic reconstructions is generally challenging, due to several factors. i) Multiple local and global processes control the chemical variability and isotopic signature of shallow seawaters and DIC (Dissolved Inorganic Carbon), including: evaporation, local stagnation of marine waters, input of freshwater through rainfall and rivers, latitude, climate, and perturbations of the carbon cycle (Castro et al., 2019;Colombié et al., 2011;Immenhauser et al., 2003;Patterson and Walter, 1994;Swart, 2015;Zuo et al., 2018). ii) In bulk carbonate samples, different components record potential distinctive isotopic compositions due to the precipitation processes and the vital effect on carbon isotope fractionation (e.g. skeletal grains, coated grains, micrite matrix) (Anderson and Arthur, 1983;Hudson, 1977;Jenkyns and Clayton, 1986;McConnaughey, 1989a, b;Nelson and Smith, 1996;Schobben et al., 2015). iii) Diagenetic processes, from early meteoric eogenesis due to subaerial exposure to burial mesogenesis and meteoric telogenesis, can alter the primary marine isotopic signature, thus complicating the use of these isotopic records as palaeoenvironmental proxies (Brand and Veizer, 1981;Coimbra et al., 2014;Glumac and Walker, 1998;Huck et al., 2017;Immenhauser et al., 2002;Marshall, 1992;Moore, 1989;Oehlert and Swart, 2019).
The magnitude of the diagenetic alteration can be variable for specific carbonate grains, depending on the original mineralogy and its degree of stability, increasing from aragonite, high-Mg calcite to low-Mg calcite. On the other hand, the magnitude of the diagenetic alteration also depends on the microstructure of the grains and on the degree of the primary porosity and permeability of the sediment. Coarse-grained grain-supported fabrics, in the absence of mud between grains, can be more favourable for infiltration of diagenetic fluids, and may result in a higher degree of diagenetic alteration and resetting of stable isotope signatures (Al-Mojel et al., 2018;Coimbra et al., 2014;Huck et al., 2017;Immenhauser et al., 2002;Marshall, 1992;Vincent et al., 2004). Different diagenetic processes and products result from the interaction between carbonate sediments and rocks with diagenetic fluids of a different nature (e.g. seawater, meteoric waters, high-temperature burial or hydrothermal fluids). Which processes take place and the way they affect more intensively the sediment depend on multiple factors, such as the mineralogy, porosity and permeability of the carbonates, chemistry and saturation of the diagenetic fluid, temperature or pressure conditions, and the burial and tectonic history, among others. Diagenetic processes can involve the dissolution and re-precipitation of variable mineralogies (e.g. neomorphism of aragonite into low-Mg calcite), the formation of more stable mineralogical phases (e.g. high-Mg calcite to low-Mg calcite) or dolomite replacement (Allan and Matthews, 1982;Al-Mojel et al., 2018;Brand et al., 2012;Lohmann, 1989, 1997;Coimbra et al., 2014;Given and Lohmann, 1985;Jenkyns and Clayton, 1986;Joachimski, 1994;van der Kooij et al., 2009;Moore, 1989;O'Neil, 1977;Sharp, 2007;Tucker, 1993). Therefore, diagenetic processes usually are associated with chemical, mineralogical and/or textural changes of carbonates. These must be taken into account when interpreting the carbon and oxygen isotope signature of bulk rocks and specific diagenetic phases Veizer, 1980, 1981;Coimbra et al., 2014;Colombié et al., 2011;Huck et al., 2017;Lavoie and Bourque, 1993).
Diagenesis of marine carbonates commonly implies a decrease in δ 18 O values by the influence of 18 O-depleted meteoric waters and/or relatively high-temperature burial fluids (due to the temperature dependence of oxygen isotopic fractionation). The magnitude of such alteration depends on the degree of the fluid-rock interaction (Allan and Matthews, 1982;Al-Mojel et al., 2018;Brand and Veizer, 1980;Coimbra et al., 2014;Given and Lohmann, 1985;Grotzinger et al., 2011;Hudson, 1977;van der Kooij et al., 2009;Madden and Wilson, 2013;Plunkett, 1997). Post-depositional alteration also generally leads to a decrease in δ 13 C values, which is normally attributed to the interaction with meteoric fluids enriched in organic 12 C by soil weathering during penecontemporaneous subaerial exposure or during uplift after burial (telogenesis) (Allan and Matthews, 1982;Bahamonde et al., 2017;Derry, 2010;Huck et al., 2017;Hudson, 1977;Immenhauser et al.,2002;Lavastre et al., 2011;Lohmann, 1988;Moore, 2001;O'Neil, 1987;Patterson and Walter, 1994;Swart, 2015). In this sense, the degradation of buried organic matter or the presence of hydrocarbon and methane seeps can also act as source of 12 C during the post-depositional burial stages (Chung et al., 1992;Peckmann and Thiel, 2004). In ancient shallow-marine carbonate rock successions, the effect of meteoric and/or burial diagenesis on the carbon and oxygen stable isotope record has been highlighted in several studies. These works report local negative δ 13 C and δ 18 O excursions in bulk carbonates with coarser fabrics related to meteoric dissolution and cementation, neomorphism and recrystallization during subaerial exposure (vadose and phreatic zone), or decreasing δ 18 O values from burial cements (Al-Mojel et al., 2018;Coimbra et al., 2014;Huck et al., 2017;Immenhauser et al., 2003;Strasser, 2015). However, some authors propose that even in the presence of diagenetic alteration, carbon and oxygen stable isotope compositions can provide some valuable information about the palaeoenvironmental setting (Al-Mojel et al., 2018;Bartolini et al., 2003;Colombié et al., 2011;Zuo et al., 2018), even using the carbon stable isotope stratigraphy as a correlation tool between widely spaced separated logs (Colombié et al., 2011;Eltom et al., 2018;Glumac and Walker, 1998;Grötsch et al., 1998;Zuo et al., 2018).
In this study, bulk carbon and oxygen stable isotope analyses were performed on samples from two stratigraphic sections of the Higueruelas Formation cropping out in the north-eastern area of the Iberian Chain (NE Spain). These successions represent the proximal and relatively distal areas of the uppermost Kimmeridgian (Upper Jurassic) carbonate ramp that developed in the northcentral part of the Iberian Basin. These sections were selected to evaluate the influence of depositional facies and palaeoenvironmental conditions on the recorded stable carbon and oxygen isotope signatures. However, the obtained results show that post-depositional processes had a strong influence on the recorded isotopic compositions in each sector linked to their specific burial evolution and tectonic uplift history, mostly related to the Early Cretaceous and Cenozoic tectono-sedimentary evolution in this part of the Iberian Basin. The major objectives of this study are: i) to evaluate the diagenetic features and their influence on the carbon and oxygen stable isotope signatures of these shallow-marine carbonate ramp deposits in two selected sections affected by different post-depositional histories, and ii) to discuss the potential of possible primary signals from the stable isotope data with a local and/or regional significance.

GEOLOGICAL SETTING
The studied uppermost Kimmeridgian Higueruelas Formation (Fm.) represents inner-to mid-ramp deposits of a carbonate ramp that developed in the north-central part of the Iberian Basin (Aurell et al., , 2010 (Fig. 1A, B). At basin scale, these shallow-marine deposits grade eastwards into outer-ramp marlstone and lime mudstone of the Loriguilla Fm. (Aurell et al., , 2010 (Fig. 1C). The regional relative sea-level fall documented at the end of the Kimmeridgian, which was partly controlled by the rifting-related tectonic uplift of the Iberian Massif (Fig. 1A), resulted in a gradual shallowing-upward trend and the final subaerial exposure of the carbonate ramp in this part of the basin due to tectonic uplift Bádenas and Aurell, 2001). The shallowing-upward trend at the end of the Kimmeridgian culminated with deposition of a mixed carbonate-siliciclastic unit overlying the Higueruelas Fm.
The tectonic uplift of the north-central Iberian Basin margin, leading to the subaerial exposure of the carbonate ramp at the end of the Kimmeridgian, was related to the Late Jurassic-Early Cretaceous rifting stage of the Iberian Basin ( Fig. 3), followed by the subsequent post-rifting stage until the end of the Cretaceous Cortés et al., 1999). This resulted in the formation of several subbasins controlled by extensional faults, which were inverted during the Cenozoic Alpine compression Guimerà and Álvaro, 1990;Liesa et al., 2019;Salas and Guimerà, 1996) (Figs. 2B;3). The two studied sections belong to the Aguilón sub-basin, controlled by NW-SE oriented extensional faults   (Fig. 2B). In this sub-basin, after a long-lasting episode of nondeposition (latest Kimmeridgian to middle Valanginian), sedimentation resumed with the deposition of up to 300m of a Lower Cretaceous (mid-Valanginian to lower Barremian) syn-rift continental succession (Villanueva the Huerva and Aguilón formations; Cortés et al., 1999;Soria et al., 1995) (Figs. 2C;3). Subsiding areas were located SW of the Aguilón fault system and Villanueva de Huerva monocline (Cortés et al., 1999) (Fig. 2C). The Tosos sector belonged to the subsiding area southwest of the Aguilón fault system, and consequently a thick (180m) succession of mid-Valanginian to lower Barremian deposits accumulated over the Higueruelas Fm. By contrast, the Fuendetodos sector was located on the non-subsiding area, NE of the Villanueva de Huerva monocline, therefore lacking the mid-Valanginian to lower Barremian sedimentary record (Figs. 2C;3). After deposition of the Lower Cretaceous units, there was a major erosive and non-depositional gap until the deposition of syn-tectonic (Alpine) continental Cenozoic units (Paleocene to middle Miocene) (compressional stage in Fig. 3), with an interval of non-deposition and/or erosion from the latest Paleocene to early Oligocene (Cortés et al., 1999). Around Tosos, this set of Cenozoic units is up to 600m thick (Pérez et al., 1985;Pérez, 1989). However, the Fuendetodos sector was close to an Alpine structure Diagenesis in shallow-marine carbonates 4 (Mezalocha anticline) that represented a palaeo-high that remained exposed until the onset of the Miocene (Fig.  2C), and thus was mainly subjected to erosion and/or nondeposition during this stage (Fig. 3). In particular, around Fuendetodos, the Higueruelas Fm. is only overlain by a c. 200m-thick succession of middle Miocene lacustrine carbonates (Pérez et al., 1985;Pérez, 1989).
In summary, the Higueruelas Fm. underwent a different burial and uplift history in the two studied sectors. In Tosos, these carbonate deposits were buried by at least 780m of lower Cretaceous (180m thick) and Cenozoic (600m thick) continental deposits, and affected by three intervals of nondeposition and/or erosion from the latest Kimmeridgian to middle Valanginian, from the middle Barremian to Paleocene, and from the latest Paleocene to early Oligocene (Fig. 3). In contrast, Fuendetodos remained mostly exposed from the latest Kimmeridgian to the middle Miocene, and only a 200m-thick continental succession is known to have been deposited over this section during the Cenozoic (Alpine) compressional stage. Even though other episodes of burial may have occurred in Fuendetodos from the latest Kimmeridgian to the middle Miocene, erosion and/or non-deposition associated with tectonic uplift dominated. In addition, on the basis of the previous stratigraphic framework established in this region , the upper part of the Higueruelas Fm. in the Fuendetodos sector is eroded (Fig. 3), lacking the last two high-frequency sequences (8 and 9) recognized in nearby distal sections.

STUDIED SUCCESSIONS AND SEDIMENTOLOGICAL FRAMEWORK
The uppermost Kimmeridgian Higueruelas Fm. studied in Tosos and Fuendetodos sections is 71m and 52m thick,  Dercourt et al., 1993). B) Main facies belts in the Iberian Basin during the late Kimmeridgian (square in A) and location of the studied sections (south of the city of Zaragoza, NE Spain; compiled from Aurell et al., 2003 andIpas et al., 2004). C) Kimmeridgian stratigraphy in the northern Iberian Basin (modified from Aurell et al., 2019). The Higueruelas Fm., which is represented by the long-term third-order sequence Ki3, overlies open-marine marlstone and lime mudstone of the Loriguilla Fm., which is in turn laterally equivalent to the shallow-marine carbonates of the Higueruelas Fm. in eastern areas of the northern Iberian Basin (i.e. deep-marine areas in B A) Extent of the Jurassic outcrops in the study area with the position of the studied logs (stars) and the cross-section a-a' shown in (C). The position of the Early Cretaceous extensional faults is included (grey colour) (modified from Cortés Gracia and Casas Sainz, 1996). B) Main sedimentation areas and active faults during the Late Jurassic-Early Cretaceous rifting episode in the northern Iberian Basin. The studied sections (stars) are situated in the newly formed Aguilón sub-basin (Aguilón Sb) (modified from Liesa et al., 2019). C) Balanced cross-section (present-day arrangement) and suggested section restored to the top of the syn-rift sequence (Early Cretaceous) for the transect a-a' indicated in (A). The position of Tosos (TO) and Fuendetodos (FU) sections are included (modified from Cortés et al., 1999). respectively. These successions encompass high-frequency sequences 1 to 7 (in Tosos) and 2 to 7 (in Fuendetodos), and include a total of 14 facies representing deposition from inner-ramp domains (intertidal, lagoon, backshoal and shoal-sand blanket), mainly recorded in the Tosos section, to mid-ramp foreshoal and offshore areas represented in the Fuendetodos section , 20205). Distance between the two sections, without restoring tectonic shortening, is around 15km (Fig. 5). Some decimetre-to metre-thick beds of totally or partially dolomitized facies are also recorded in both sections (Fig.  5), which post-date the deposition of the Higueruelas Fm. The origin of this replacive dolomitization has been related to circulation of hydrothermal fluids through faults during the Cenozoic Alpine compression, which acted as extensional faults during the Late Jurassic-Early Cretaceous rifting stage (Aurell, 1990) (Fig. 2C). However, it is also probable that these dolomites formed through fault-related hydrothermal circulation during the rifting episode, as similarly described in later studies about the Late Jurassic-Early Cretaceous dolomites in eastern areas of the Iberian Basin (i.e. Maestrat Basin; Travé et al., 2019).
The detailed sedimentological framework of the studied successions is described in Sequero et al. (2019). Log correlation was based on the recognition of a number of well-marked bedding surfaces, which allowed the recognition of the high-frequency sequences 1 to 7  identified in these sections (Figs. 4;5), and represent sedimentary discontinuities related to short-term sealevel fluctuations (i.e. short-term eccentricity cycles). These bedding planes are planar surfaces, some of which show bioturbation or Fe-encrustations, and locally represent hardgrounds. The high-frequency sequences 1 to 7 identified in the proximal Tosos section represent a complete record of the Higueruelas Fm. in this sector, with the boundaries with the underlying Loriguilla Fm. and the overlying carbonate-siliciclastic unit being distinguished in the field (Fig. 1C). However, the record of the upper part of the Higueruelas Fm. (i.e. sequences 8 and 9 recognized in nearby distal sections) is not preserved in the southeast, i.e. the distal Fuendetodos sector, due to the erosive gap related to the tectonic uplift in this part of the basin (Figs. 2C; 3). In both sections, there is no evidence of subaerial exposure at sequence boundaries that could be associated with the high-frequency sea-level fluctuations (e.g. Strasser, 2015), and at the top there is no record of latest Kimmeridgian karst or palaeosol formation.
A best-fit correlation between the discontinuity surfaces in the Tosos and Fuendetodos areas was performed, as no physical tracing between isolated outcrops was possible. This correlation is supported by the physical tracing of these sharp bedding planes along continuous outcrops of the Higueruelas Fm. in the same area (see Sequero et al., 2018). This best-fit correlation was also constrained by the vertical facies evolution observed in the successive highfrequency sequences defined in these and other individual logs analysed in this region. Variable vertical facies trends are described for these sequences (shallowing, deepening, aggradational or deepening-shallowing) (Fig. 5), also varying from one section to another, since the local factors operating on the carbonate ramp strongly controlled the distribution and lateral migration of the facies types .

METHODOLOGY Petrographic characterization
A total of 72 samples in thin sections, already used for differentiating depositional facies by Sequero et al. (2019), was analysed for the petrographic descriptions of facies under the binocular microscope, in particular differentiating between primary sedimentary carbonate components (i.e. micrite matrix and carbonate grains) from diagenetic calcite cements. Dolomitized facies (Fig. 5) have not been included in this work, since only a few totally or partially dolomitized beds occur in both sections. Nomenclature for calcite cements follows the terminology based on crystal fabric summarised in Flügel (2004).
Petrographic analysis was complemented by CathodoLuminescence (CL) microscopy on 21 thin sections (7 in Tosos and 14 in Fuendetodos), with a Technosyn Cold Cathodo Luminoscope, operating under 10-14kV beam potential, 0.5µA beam current and a 0.05-0.1 Torr pressure. CL analyses were used to characterize the primary carbonate components and calcite cements, on the basis of the absence or type of luminescence, as traditionally followed in the literature (Hiatt and Pufahl, 2014;Machel, 1985;Richter et al., 2003;Savard et al., 1995).

Carbon and oxygen stable isotope analyses
The carbon and oxygen stable isotope analyses were performed on the 72 bulk-rock samples collected in Tosos (n= 35) and Fuendetodos (n= 37) sections, and on calcitic shells from three specimens of Trichites (large-sized bivalve with very thick outer shell wall, proposed as reliable proxy for Jurassic pristine stable carbon and oxygen isotopic signature; cf. Zuo et al., 2018) found in the offshore-proximal peloidal-bioclastic wackestone facies within sequence 2 of the Fuendetodos section (Fig. 5). In addition, two calcite veins found in the offshore-proximal peloidal-bioclastic packstone facies within sequence 5 of the Fuendetodos section were also analysed. Only two samples of calcite vein were analysed due to the difficulties in finding those veins with a suitable width to sample without contamination from the surrounding material. In order to ensure the reliability of the carbon and oxygen stable isotope compositions of each sample, a double measure per sample was carried out, giving a total of 173 stable isotope analyses (n= 77 in Tosos section, and n= 96 in Fuendetodos section) (see Table 1).
For the carbon and oxygen stable isotope analyses of bulk-rock samples, carbonate powders were extracted using a handheld micro-drill, avoiding calcite cements, calcite filled fractures and large bioclastic fragments of corals or stromatoporoids, due to the disequilibrium-isotopic precipitation by organisms (McConnaughey, 1989a, b). Diagenesis in shallow-marine carbonates    However, the incorporation of a certain amount of calcite cements filling the intraparticle or mouldic space of skeletal debris, interparticle spaces, and some fenestral pores, was in some cases inevitable. Isotope analyses of the interparticle or fenestral cement could not be performed in view of the difficulties in sampling without contamination from surrounding micrite or carbonate grains.
The carbon and oxygen stable isotope analyses were performed using an automated carbonate preparation device (GasBench II) connected to a Delta V Advantage (Thermo Fisher Scientific Inc.) isotope ratio mass spectrometer at the Earth Sciences Department, University of Milan (Italy). Carbonate powders were reacted with >99% orthophosphoric acid at 70ºC. The carbon and oxygen stable isotope compositions are expressed in the conventional delta notation calibrated to the Vienna Pee-Dee Belemnite (V-PDB) scale by the international standards IAEA-603 and NBS-18 and an internal laboratory standard. Analytical reproducibility for these analyses, being checked by repeated analyses of the certified carbonate standards, was better than ±0.1‰ for both δ 13 C and δ 18 O values.

Petrographic and cathodoluminescence analyses
The bulk-carbonate samples analysed from both the Tosos and Fuendetodos sections include a wide variety of primary carbonate components in variable proportions, mainly micrite matrix, peloids, ooids, oncoids and skeletal debris, as well as calcite cements filling mouldic vugs (generated by dissolution of skeletal debris), intraparticle and interparticle pores, fenestral pores and fissures (Figs. 6; 7). Local stylolites also occur in both sections.

Chaetetid-stromatoporoid-coral buildup
Inter-buildup stromatoporoid-chaetetid-coral and oncolitic W to P  Three types of calcite cement fabric have been recognized in all the investigated facies: granular, drusy and blocky. Granular to drusy calcite cements tend to fill mouldic, intraparticle and interparticle porosity. They generally constitute less than 10% of the thin-section area and are formed by anhedral to subhedral equant (granular, <10µm in size) and non-equigranular (drusy, 10 to 30µm in size) crystals (Figs. 6A-C; 7A-D). In the drusy calcite cements, crystal size increases towards the centre of the void (Fig. 6B). In some grainstone and packstone facies, granular and drusy interparticle cement follow the formation of elongated grain contact and mechanical compaction (Figs. 6C; 7C). Blocky calcite cements precipitated in fenestral pores, in rectilinear fissures (i.e. calcite veins; Fig. 6D) and locally in mouldic vugs (Fig. 7B). They constitute <10% of the thin-section area, and are formed by anhedral to subhedral crystals from hundreds of µm to several mm in size.

Carbon and oxygen stable isotope records
Carbon and oxygen stable isotope records for the Tosos and Fuendetodos sections are represented in Figure 5, and the values are listed in Table 1. The double measure per sample (labelled A and B following sample name in the table) confirms the consistency of the carbon and oxygen stable isotope composition obtained for each analytical point, with a mean reproducibility of ±0.2‰ for both δ 13 C and δ 18 O values.
The bulk δ 13 C and δ 18 O values of the inner-ramp Tosos succession range between -3.5 to 2.3‰ for δ 13 C and -7.1 to -2.4‰ for δ 18 O (Fig. 8), with a mean value of 0.9‰ and -4.4‰, respectively (standard deviation of 1‰ for both δ 13 C and δ 18 O). In general, this section shows low-amplitude fluctuations for both δ 13 C and δ 18 O records, in the order of ≤1‰ with respect to the mean values (Fig. 5). In some cases, there are negative shifts in the order of 2‰ for the δ 13 C record in the backshoal oncolitic-stromatoporoid wackestone-packstone facies within sequence 2 (at 31.5m), and in the order of 3-4‰ in both δ 13 C and δ 18 O values for backshoal peloidal-oolitic packstone facies at the onset of sequence 4 (at 44.5m).
The bulk δ 13 C and δ 18 O values along the mid-ramp Fuendetodos succession vary between -5.6 to 1.0‰ and -6.6 to -3.5‰, respectively (Fig. 8), with a mean value of -1.4‰ for δ 13 C and -5.3‰ for δ 18 O (standard deviation of 1.2‰ for δ 13 C and 0.7‰ for δ 18 O), thus reflecting a general 13 C and 18 O depletion compared to the Tosos section. In addition, both δ 13 C and δ 18 O values show slightly higheramplitude oscillations than in Tosos, in the order of 1-2‰ (Fig. 5). Two pronounced excursions with respect to the mean values are reported for the δ 13 C record, accompanied by a coupled and less pronounced shift in the δ 18 O values: a negative excursion at the onset of sequence 2 (at 0.5m) within offshore-proximal peloidal-bioclastic wackestone facies, with a deviation in the order of 3‰ for δ 13 C and 1‰ for δ 18 O, and a positive shift in sequence 6 (at 44.5m), coinciding with deposition of foreshoal peloidal grainstone facies, being the δ 13 C and δ 18 O values deviated in the order of 3 and 2‰, respectively. The isotopic signals of the three specimens of Trichites collected in the lower part of this section show higher δ 13 C values but a wider range for the δ 18 O signature (Fig. 8), ranging between 1.3 to 2.5‰ for δ 13 C and -4.8 to -2.2‰ for δ 18 O. The two analysed calcite veins show instead the lowest carbon and oxygen isotope values, ranging between -7.7 to -6.3‰ for δ 13 C and -7.7 to -7.3‰ for δ 18 O (Fig. 8).
The bulk carbon and oxygen stable isotope stratigraphy in both sections reveals no distinctive long-term δ 13 C and δ 18 O patterns, and the δ 13 C and δ 18 O trends observed within time-equivalent high-frequency sequences vary significantly from one section to the other (Fig. 5). Moreover, the bulk isotopic stratigraphy shows that the δ 13 C and δ 18 O values fluctuate in covariance, the degree of such δ 18 O-δ 13 C covariance being relatively lower in G e o l o g i c a A c t a , 1 9 . 1 4 , 1 -2 3 ( 2 0 2 1 l o g i c a A c t a , 1 9 . 1 4 , 1 -2 3 ( 2 0 2 1 l o g i c a A c t a , 1 9 . 1 4 , 1 -2 3 ( 2 0 2 1  Tosos (r 2 = 0.29) but moderate in Fuendetodos (r 2 = 0.52) (Fig. 9). In addition, the cross-plots of the bulk δ 13 C and δ 18 O values per facies types in each section do not show differentiated isotopic composition patterns regarding the facies type (Fig. 9), with the exception of the restricted lagoon peloidal-bioclastic-oolitic packstone facies in the Tosos section, which shows higher δ 18 O values compared to the rest of the facies (Fig. 9A).

INTERPRETATION AND DISCUSSION
Diagenetic features from petrographic and cathodoluminescence analyses The information obtained from the petrographic and cathodoluminescence analyses provides relevant data to evaluate the variable diagenetic features in the Tosos and Fuendetodos sections, and contributes to the interpretation of the stable isotope data. The analysed facies are predominantly wackestone to packstone accumulated in protected or below effective wave base depositional settings (Fig. 4). In view of the general low to moderate energy depositional domains, none of the investigated facies shows the presence of early marine cement in the form of isopachous rims of fibrous or radial fibrous cement (Figs. 6; 7).
The variable luminescence observed in the carbonates provides information about the redox conditions during precipitation of calcite (Barnaby and Rimstidt, 1989;Carpenter and Oglesby, 1976;Frank et al., 1982;Machel, 1985;Oglesby, 1976). Luminescence is a common indicator of the incorporation of Mn 2+ within the crystal lattice during water-rock interactions under suboxic to anoxic (generally burial) conditions (Flügel, 2004;Hiatt and Pufahl, 2014;Machel, 1985;Marshall, 1992;Richter et al., 2003;Savard et al., 1995). The absence of luminescence can in turn be attributed to both the absence of Mn 2+ in the diagenetic fluids, precipitation under oxidizing conditions, or the incorporation of Fe 2+ under reducing conditions, since Fe 2+ is a common quencher resulting in dull or no luminescence Oncolitic-stromatoporoid W to P Stromatoporoid P

Shoal -sand blanket
Oolitic G Oncolitic G Oncolitic P

MID-RAMP
Chaetetid-stromatoporoid-coral buildup Inter-buildup stromatoporoid-chaetetid-coral and oncolitic W to P r = 0.29 2 2 r = 0.52 A B FIGURE 9. Cross-plots of δ 13 C and δ 18 O values of the analysed bulk carbonates in A) Tosos and B) Fuendetodos sections according to the facies type.

Diagenetic alteration of carbon and oxygen stable isotope record
The scatter diagram for the measured bulk δ 13 C and δ 18 O values of the Tosos and Fuendetodos sections shows that diagenesis significantly modified the original isotopic signature of these uppermost Kimmeridgian shallowmarine carbonate deposits in both localities, although with distinctive differences in each section (Fig. 8).
Oxygen isotope values for bulk carbonates from Tosos (-7.1 to -2.4‰, with most of the values ranging between -5.9 to -2.4‰) and Fuendetodos (-6.6 to -3.5‰) are lower than those reported in the literature for pristine Kimmeridgian marine carbonates (between -2.8 to 1.4‰; e.g. Jenkyns et al., 2002;Nunn and Price, 2010;Riboulleau et al., 1998;Wierzbowski, 2004) (Fig. 8). If these published pristine Kimmeridgian marine values are assumed valid for the Kimmeridgian shallow-water domains in the Iberian Basin, the different magnitudes of deviation in the Fuendetodos and Tosos sections can be used to discuss both the imprint of the diagenetic alteration and the origin of the diagenetic fluids. In particular, the deviation of the δ 18 O values with respect to the pristine Kimmeridgian marine carbonates is higher in Fuendetodos than in Tosos, in the order of 1 to 3‰ for Tosos and of 2 to 3‰ for Fuendetodos (Fig. 8).
The variation for carbon isotope values also shows remarkable differences between both sections. In the Tosos section, the measured δ 13 C values (-3.5 to 2.3‰, with most of the values ranging between -1.4 to 2.3‰) are similar to those published for pristine Kimmeridgian marine carbonates (between -2.3 to 4.7‰; Jenkyns et al., 2002;Nunn and Price, 2010;Riboulleau et al., 1998;Wierzbowski, 2004), whereas in Fuendetodos, measured δ 13 C values (-5.6 to 1.0‰, with most of the values ranging between -3.8 to 0.3‰) are lower than those described in the literature, in the order of 3 to 4‰, suggesting a diagenetic overprint. Although during powder sampling for stable isotope analyses there was likely contamination of sedimentary components and diagenetic cements, samples from the two sections display different distinctive average stable isotope signatures and trends, which is indicative of resetting by different diagenetic processes and fluids.
The diagenetic alteration affecting the studied carbonate deposits overprinted a possible palaeoenvironmental signal from the isotopic record. The covarying δ 13 C and δ 18 O vertical trends in the Tosos and Fuendetodos sections can be interpreted as an indicator of this diagenetic resetting, as traditionally reported in the literature (Allan and Matthews, 1982;Fike et al., 2006;Grotzinger et al., 2011;Metzger and Fike, 2013). In recent studies (Swart and Oehlert, 2018), it has been suggested that such covariance is not a definitive marker of diagenetic alteration. However, in this case, this assumption can be considered since there is no other evidence in either the Tosos or Fuendetodos sections to support the preservation of a palaeoenvironmental signal from the isotopic record, such as: i) the absence of a faciesdependent pattern on the isotopic record which could be associated with changes in the depositional conditions (Colombié et al., 2011;Immenhauser et al., 2002Immenhauser et al., , 2003Patterson and Walter, 1994;Zuo et al., 2018) (Fig. 9); ii) the absence of distinctive long-term variation patterns in the δ 13 C and δ 18 O stratigraphy (Fig. 5), showing only low-amplitude oscillations in both sections with local excursions, and iii) the highly varying vertical δ 13 C and δ 18 O trends observed within the time-equivalent high-frequency sequences from one section to another (Fig. 5), being not possible to establish a significance for these isotopic variations. The significance for the local excursions in the δ 13 C and δ 18 O values observed in both sections is also uncertain, as no other differences from the petrographic features are found compared to the rest of the samples.
The three specimens of Trichites collected from the Fuendetodos section have the potential to record the original latest Kimmeridgian seawater stable isotope composition, as reported by previous authors (Zuo et al., 2018). Therefore, they could be used for quantifying the magnitude of the diagenetic resetting. However, the diagenetic overprint reported on the bulk carbonates in the studied sections can also be observed in the δ 13 C and δ 18 O values of the Trichites calcitic shells (Fig. 8). Sample TR-1 shows the closest δ 18 C and δ 18 O values to the pristine Kimmeridgian marine carbonates (a mean value of 2.4‰ for δ 13 C and of -2.4‰ for δ 18 O; see Table 1), but they also fall within the range of those reported for the diagenetically altered bulk carbonates in the Tosos section. In addition, the two other specimens (samples TR-2 and TR-3) display a higher deviation in the δ 18 O signature, in the order of 1-2‰ with respect to the pristine Kimmeridgian marine values. Therefore, the δ 13 C and δ 18 O values of these specimens of Trichites seem to have also been diagenetically altered.
In the Fuendetodos section, the isotope signatures obtained from calcite veins (samples CV-1 and CV-2 in Fig.  8) confirm that the isotope data in this section gradually approach the low values of the late diagenetic calcite in fractures.

Comparison between Tosos-Fuendetodos: burial vs meteoric diagenetic alteration
The δ 18 O record in the Tosos section, with most of the data points deviated towards very low δ 18 O values (Fig.  8), is characteristic of closed high-temperature burial diagenetic systems (Banner and Hanson, 1990;van der Kooij et al., 2009;Madden and Wilson, 2013;Marshall, 1992;Plunkett, 1997;Swart, 2015). Hot burial fluids in contact with carbonate rocks give rise to a depletion of 18 O in the δ 18 O values, because of the high temperature-related fractionation of the oxygen stable isotope (Brand and Veizer, 1981;Immenhauser et al., 2002;Lohmann, 1988;Magaritz, 1983;Marshall, 1992;Morse and MacKenzie, 1990). In contrast, the carbon stable isotope fractionation is less dependent on temperature, and in a closed system the dissolved inorganic carbon is provided by the host carbonate rock (Brand and Veizer, 1981;Lohmann, 1988;Marshall, 1992;Morse and MacKenzie, 1990). This fits with the δ 13 C values in the Tosos section, being that these values are close to those of the pristine Kimmeridgian marine carbonates (Fig. 8). In contrast, the dispersion observed for the data in the Fuendetodos section, with significant lower δ 13 C and δ 18 O values, suggests a higher interaction with 18 O-depleted meteoric fluids incorporating organic 12 C probably derived from soil weathering (Allan and Matthews, 1982;Bahamonde et al., 2017;Hudson, 1977;Immenhauser et al., 2002;Lavastre et al., 2011;Lohmann, 1988;Moore, 2001;O'Neil, 1987;Patterson and Walter, 1994;Vogel, 1993). The two calcite veins analysed in the Fuendetodos section (Fig. 8), showing the lowest δ 13 C and δ 18 O values compared to those reported for both the Tosos and Fuendetodos sections, reflect the isotopic signature of calcite likely precipitated from meteoric fluids (e.g. Hudson, 1977;Nelson and Smith, 1996) during telogenesis after tectonic uplift. In this regard, the proximity of the δ 13 C and δ 18 O values obtained in the Fuendetodos section to those reported for the calcite veins, confirms the pervasive diagenetic alteration of these limestones by meteoric fluids. This is also consistent with the higher and relatively significant δ 18 O-δ 13 C covariance observed in Fuendetodos (r 2 = 0.52) compared to Tosos (r 2 = 0.29) (Fig.  9), highlighting a major resetting of the primary isotopic composition in Fuendetodos by meteoric fluids (Marshall, 1992;Rosales et al., 2001;Swart, 2015;Swart and Oehlert, 2018;Veizer et al., 1997).
The variable diagenetic overprint recorded on the isotopic composition in the Tosos and Fuendetodos sections is consistent with the different post-depositional processes occurring in this part of the Iberian Basin from the end of the Kimmeridgian. In Tosos, the uppermost Kimmeridgian carbonates of the Higueruelas Fm. studied here are overlain by a thick (up to 780m) continental succession deposited during the middle Valanginian to middle Miocene (Pérez et al., 1985;Pérez, 1989;Soria et al., 1995), with three long-lasting periods of subaerial exposure during the latest Kimmeridgian to middle Valanginian, from the middle Barremian to Paleocene and from the latest Paleocene to early Oligocene (Fig. 3). In contrast, the Fuendetodos sector remained mostly exposed due to tectonic uplift until the middle Miocene, and the Higueruelas Fm. is overlain by only a c. 200m-thick succession of middle Miocene lacustrine carbonates (Pérez et al., 1985;Pérez, 1989). These specific post-depositional histories controlled different late-diagenetic processes affecting the uppermost Kimmeridgian deposits in both localities, as reflected on the isotopic record (Fig. 8). In the Tosos section, the Higueruelas Fm. was affected by a long period (from mid-Valanginian to mid-Miocene) of burial (negative deviation of δ 18 O, but not of δ 13 C values, due to hot burial diagenetic fluids). However, these deposits in the Fuendetodos section were mainly subjected to uplift-related meteoric alteration (negative deviation of both δ 18 O and δ 13 C values).

Is diagenetic alteration dependent on facies type?
Palaeoenvironmental factors controlling sedimentation in the latest Kimmeridgian shallow carbonate ramp of the Iberian Basin are not reflected in the isotopic composition obtained from these carbonates in the Tosos and Fuendetodos sections, due to diagenetic resetting. Different diagenetic processes governing the post-depositional evolution of the Higueruelas Fm. have instead overprinted the primary marine δ 13 C and δ 18 O signature in each section. In this regard, considering the variability in the textural characteristics observed for the analysed deposits, the effect of the diagenetic alteration on the carbon and oxygen stable isotope records can be variable as a function of the depositional texture type. Accordingly, the coarser textures, with higher primary porosity and permeability, are more favourable for fluid circulation, which may result in a higher diagenetic overprint (Al-Mojel et al., 2018;Coimbra et al., 2014;Huck et al., 2017;Immenhauser et al., 2002;Marshall, 1992;Vincent et al., 2004). Therefore, for the Higueruelas Fm., an approximation of the possible link between textural features and the degree of alteration of the primary isotopic signature has been attempted with the available data in Figure 10. o l o g i c a A c t a , 1 9 . 1 4 , 1 -2 3 ( 2 0 2 1  In Tosos, a variable response to the diagenetic alteration regarding the textural types can be appreciated (Fig. 10A). There is a particular difference in the isotopic composition scatter for the grainstone textures, showing the lowest δ 13 C and δ 18 O values compared to the other textures. In the Fuendetodos section, however, the diagenetic alteration indistinctly affected the carbonate deposits, not reporting differentiated δ 13 C and δ 18 O values regarding the textural type (Fig. 10B). This distinct pattern in the texture-diagenetic alteration relationship in the Tosos and Fuendetodos sections could be related to the different post-depositional processes affecting the Higueruelas Fm. at each locality. For the Tosos section, which underwent a long period of relatively shallow burial, it is probable that the diagenetic alteration predominantly reflects variable fluid-rock interaction depending on the porosity and permeability of the sediment, but also a variable degree of contamination by sampling burial cement. Most of the calcite cements incorporated during sampling correspond to granular (GCC) and drusy (DCC) calcite cements, which seem to have been precipitated as burial diagenetic phases, because they followed mechanical compaction and show dull luminescence (DCC; Fig. 6B'). In this sense, there is also a relationship between the amount of calcite cement precipitated and the alteration of the isotopic signature, the grainstone textures having a larger amount of burial calcite cement (around 15% of the total volume), compared to the rest of the textures (less than 10%). In contrast to this scenario, the Higueruelas Fm. in the Fuendetodos section was mostly uplifted (Figs. 2C; 3) and exposed to meteoric diagenesis during the long-term uplift from the latest Kimmeridgian to middle Miocene, and therefore to a probably more complex succession of diagenetic events (i.e. meteoric waters with variable chemical composition probably related to different climate and vegetation, or meteoric waters that interacted with different rocks in the phreatic zone), thus obliterating a texture-diagenetic alteration pattern. This is also supported by the higher diagenetic alteration reported for the Fuendetodos section (i.e. the lowest δ 13 C and δ 18 O values, close to the isotopic composition of the meteoric calcite veins) compared to the Tosos section.
A noteworthy aspect that arises from this study is that from the Fuendetodos section, which represents a deeper depositional environment, one would expect isotopic data less influenced by meteoric diagenesis. Instead, this section appears more altered by meteoric fluids than the Tosos section, which represents shallower-water domains. On the other hand, it is also remarkable that both sections, despite being governed by different diagenetic processes, show similar calcite cements in terms of fabrics and cathodoluminescence properties (Figs. 6A'-D'; 7A'-D'). These similarities, along with the diagenetic resetting revealed in each section, provide the background for further analyses in order to understand the imprint of specific burial and tectonic histories on the carbon and oxygen stable isotope data from these shallow-marine carbonates. On the other hand, the results obtained in this study can also be applied to similar cases in order to understand how the carbon and oxygen stable isotope signatures are altered in specific post-depositional contexts.

CONCLUSIONS
Bulk carbon and oxygen stable isotope compositions recorded in the uppermost Kimmeridgian shallow carbonate ramp deposits in the northern Iberian Basin (Higueruelas Fm., NE Spain), reveal the variable effect of burial and meteoric diagenesis (following tectonic uplift) in two selected sections (nearly 15km apart) representing the proximal (Tosos section) and relatively distal (Fuendetodos section) areas of the carbonate ramp.
This study reveals that the deeper-water facies, represented in the Fuendetodos section, are those more affected by meteoric alteration compared to the Tosos section, despite representing shallower-water domains. This particularity is a consequence of different burial and tectonic histories affecting the Higueruelas Fm. at each locality, which is reflected on the variable diagenetic resetting of the isotope data. In the Tosos section, the cross-plot of the bulk δ 13 C and δ 18 O values shows a significant decrease in the δ 18 O signature compared with the pristine Kimmeridgian marine carbonates, but instead a similar carbon isotope composition. This type of diagenetic alteration results from a post-depositional evolution in a closed high-temperature burial system. In this area, the Higueruelas Fm. was affected by long-term (from Cretaceous to Miocene) burial, as these deposits are overlain by a c. 780m-thick continental succession deposited during the middle Valanginian to middle Miocene. On the contrary, the Higueruelas Fm. in the Fuendetodos section remained mostly exposed due to tectonic uplift, and was affected by meteoric diagenesis from the latest Kimmeridgian until the middle Miocene, recording a significant depletion of both 13 C and 18 O. In both sections, the strong influence exerted by the diagenetic resetting makes the δ 13 C and δ 18 O records unsuitable for addressing palaeoenvironmental reconstructions.
A relationship between the magnitude of the diagenetic alteration on the isotopic signature and the facies type is observed. In the Tosos section, the grainstone textures are those showing the lowest δ 13 C and δ 18 O values compared to the other textures. For the Fuendetodos section, however, no differentiated isotopic composition patterns regarding the textural types are observed. In contrast to the isotopic record, similar cement types in terms of fabrics and cathodoluminescence properties are reported in both sections.
This study highlights the importance of integrated studies (sedimentology, regional tectonic evolution and burial history, diagenesis) in order to better understand and interpret the carbon and oxygen stable isotope signature in ancient shallow-marine carbonates. In addition, the Tosos section shows a relationship between the textural types and the diagenetic alteration, but this does not occur in the Fuendetodos section, affected by a long-term uplift and exposure to meteoric fluids rather than burial. These results can be applied to similar cases in order to understand how the carbon and oxygen stable isotope signatures are altered in specific post-depositional contexts, and so provide the background for further investigations focused on the relationship between the fabric type and the magnitude G e o l o g i c a A c t a , 1 9 . 1 4 , 1 -2 3 ( 2 0 2 1  Diagenesis in shallow-marine carbonates 20 of the stable isotope resetting in those carbonates mainly affected by burial diagenesis.