Fluid migration and interaction in extensional basins: application to the Triasic and Neogene rift in the central part of the Catalan Coastal Ranges, NE Spain: field guide (F5)

´2000 Geofluids III (F5) Fluid migration and interaction in extensional basins: application to the Triasic and Neogene rift in the central part of the Catalan Coastal Ranges, NE Spain. Field Guide. F. Calvet(1,*), A. Canals(1), E. Cardellach(2), J.M. Carmona(3), D. GómezGras(2), D. Parcerisa(2), K. Bitzer(3), E. Roca(2) & A. Travé(2) 1Facultat de Geologia, Universitat de Barcelona, 08071 Barcelona, Spain. 2Institut de Ciències de la Terra (C.S.I.C.), 08071 Barcelona, Spain. 3Departament de Geologia, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain (*Corresponding author E-mail: [email protected]) . TABLE OF CONTENTS 1. OBJECTIVES ________________________________ 5 2. GEOLOGICAL SETTING _____________________ 7 2.1. CATALAN COASTAL RANGES __________________ 7 2.2. GEODYNAMIC EVOLUTION ____________________ 9 2.3. GEODYNAMIC EVOLUTION AND PALEOFLUIDS ____ 10 3. STOPS _____________________________________ 13 S1. COLLSEROLA TOWER ________________________ 13 S2. BERTA MINE ______________________________ 14 Introduction ________________________________________ 14 Objectives __________________________________________ 15 Berta Mine Vein Filling _______________________________ 16 Constraints on the age of the hydrothermal events __________ 18 S3. RUBÍ ____________________________________ 21 Outcrop description __________________________________ 21 Early Burdigalian host conglomerate petrology ____________ 21 Calcite veins ________________________________________ 22 Interpretation and discussion ___________________________ 24 S4. TRIASIC MARTORELL QUARRY ________________ 26 S5. MIOCENE MARTORELL QUARRY _______________ 27 Outcrop description __________________________________ 27 Calcite veins ________________________________________ 27 S6. TORRENT’S SPRING, GELIDA __________________ 31 Outcrop description __________________________________ 31 Dolomite petrology __________________________________ 32 1.6.3. Dolomite geochemistry __________________________ 34 Calcite cement petrology and geochemistry _______________ 35 Discussion and interpretation __________________________ 36 S7. FINI-MIOCENE FLUID FLOW MODELING AND ECONOMIC IMPORTANCE ____________________ 39 Fluid flow during dolomitization ________________________ 39 Late postrift paleofluids: General diagenetic sequence in the late neogene horst-graben system _____________________________ 41 S8. OLÈRDOLA CASTLE __________________________ 43 Outcrop description __________________________________ 43 S9. MONTJUÏC HILL ____________________________ 44 Introduction ________________________________________ 44 Objec tiv es __________________________________________ 46 Sandstone diagenesis _________________________________ 46 Geochemical characteristics ___________________________ 47 S10. CALDETES _______________________________ 49 Geological setting ___________________________________ 49 Hydrochestry of the present post-rift hydrothermal system ______________ 49 Quantitative modelling _______________________________ 52 Modeling results_____________________________________ 53 Conclusions ________________________________________ 54 4. REFERENCES _______________________________ 55 Geofluids III. F5 Field Guide. 15 - 16 July 2000 5 1. OBJECTIVES The aim of this field-trip is to recognize hydrothermal episodes and main diagenetic products related to the evolution of extensional basins. The history of fluid flow and fluid migration in these extensional basins will be discussed on the basis of field observations and analytical data. Finally, modelling of fluid migration in the neogene basin will be presented. The following outcrops will be visited: Mesozoic (F-Zn-PbBa) and Neogene (F-Ba) mineralizations (Mina Berta, Rubí). Syn-rift geofluids: diagenesis of proximal deposits filling neogene half-grabens and calcite veins (la Pedrera and la Gasolinera outcrops, Rubí), diagenesis of distal deposits filling neogene half-grabens and calcite veins (Martorell). Post-rift geofluids: Dolomitization related to normal fault system (La Font del Torrent, Gelida) and diagenesis in upper Miocene delta system (Montjuïc Mountain, Barcelona). Calvet, F. et al. 6 Geofluids III. F5 Field Guide. 15 - 16 July 2000 7 2. GEOLOGICAL SETTING 2.1. Catalan Coastal Ranges The Valencia Trough is a NE-SW oriented basin located between the Iberian Peninsula and the Balearic Promontory. The Valencia Trough is an extensional basin which comprises two domains: The Catalan-Valencia domain in the northwest and the Betic-Balearic domain in the southeast (Roca et al. 1990; Roca, 1994). The Catalan Coastal Ranges are situated in the Catalan-Valencia domain, and correspond to the northeasternmost part of the continental margin that separates the thinned crust of the Valencia Trough from the undeformed or thickened crust of the Iberian microplate. In this setting, the Catalan Coastal Ranges display a well-developed horst-and-graben structure, which was acquired during the late Oligocene-early Miocene opening of the Valencia Trough. This extensional structure is superimposed on the older, Eocene-early Oligocene thick-skinned thrust sheets which formed during the Paleogene collision between Iberia and Euroasia. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 0 25 km + + Jurassic & Cretaceous Carboniferous Triasic Cambrian Silurian & Devonian Ordovician Late Hercynian granites + + + + + + + + + + + + Neogene Quaternary Thrust Paleogene Normal fault Fault Geological Legend Mediterranean Sea N S0 S1 + + + + Sop S8 S9 S2 S3,5 S4 S7 S10 Alacant Catalan Coastal Ranges BARCELONA GIRONA Barcelona Figure 1. Geological map of the adjacent area of the field trip. In the central part of the Catalan Coastal Ranges, the main extensional structures are NE-SW and ENE-WSW trending extensional faults formed by the reactivation of an earlier major basement thrust. Calvet, F. et al. 8 These normal fault splits the area into a set of ENE-WSW blocks mainly tilted to the NW (Guimerà, 1984; Bartrina et al., 1992) which controlled the formation of three differently deformed domains (Prelittoral range, the Vallès-Penedès half-graben and MontnegreGarraf horst, and the offshore Barcelona graben) separated by two major normal faults: the Vallès-Penedès and Barcelona faults (Fig. 1). Both faults dip to the SE and have a vertical displacement which reaches 4 km in the Vallès-Penedès fault and 8,5 km in the Barcelona fault. i. Prelitoral range: Located NW of the Vallès-Penedès master fault, this is characterized by the presence of Paleogene structures only. The Prelittoral range consists of a Hercynian basement, Triassic cover and Ebro basin-fill gently tilted to the NW except in the proximity of to the Vallès-Penedès fault where these units are strongly deformed by north-west verging contractional structures. ii. The Vallès-Penedès half-graben and Montnegre-Garraf horst: Between the Vallès-Penedès and Barcelona faults the structure is relatively simple. It consists of a Hercynian basement overlain by a Mesozoic cover tilted north-westward, forming a major depocenter close to the Vallès-Penedès fault (the Vallès-Penedès half-graben) and a parallel system of highs to the SE (the Montnegre-Garraf horst). This structure is cut by ENE-WSW hectometric normal faults. The Vallès-Penedès half-graben, which is up to 100 km long and 10-14 km wide, is filled with a thick successions of terrigenous rocks which describe a tight syncline and are cut by normal faults. These successions have been divided into the following lithostratigraphic units (Cabrera et al., 1991), which from base to top are: 1) Lower continental syn-rift complexes Aquitanian?- early-middle Burdigalian in age; 2) Continental and transitional early post-rift units with reefal carbonate platforms Langhian in age; And, 3) Continental late post-rift complexes middle-late Serravallian-Tortonian in age, which consist of thick red bed sequences deposited on alluvial fan environments. Along the Llobregat river, these three deformed complexes are covered by pliocene continental to marine deposits which onlap a major Geofluids III. F5 Field Guide. 15 - 16 July 2000 9 Messinian regional erosive surface affecting the underlying deposits. The Montnegre-Garraf horst is up to 125 km long and up to 20 km wide. The NE part of this horst is called the Montnegre horst and it consists of metapellitic rocks from late Ordovician to Carboniferous and Hercynian granitoids (Gil Ibarguchi & Julivert, 1988; Julivert & Durán, 1990). Southwest of the Llobregat river, the horst is called the Garraf horst and consists of thick Mesozoic cover (Triassic, Jurassic and chiefly Cretaceous limestones) which overlie an Hercynian basement. Close to the offshore Barcelona half-graben, the MontnegreGarraf horst presents a relatively complex structure. This zone includes some minor normal faults, dividing the area into a number of north-west tilted blocks covered by Miocene to Quaternary deposits (i.e., Barcelona city depression and Montjuïc tilted block). iii. The offshore Barcelona half-graben: The Barcelona halfgraben is up to 60 km long and 16 km wide. It is bounded in the NW by the Barcelona SE dipping extensional fault and in the SE margin by ENE-WSW highs cut by several hectometric normal faults (Bartrina et al., 1992; Álvarez-de-Buergo & Meléndez, 1994). The Barcelona half-graben is filled with the following lithostratigraphic units that overlie a thick Mesozoic cover (Bartrina et al., 1992): 1) Paleogene pre-rifts units, which are constituted by red-bed sequences, evaporites and carbonate coalbearing beds; 2) Early Miocene syn-rift units (Aquitanian?- Burdigalian) made up of basically alluvial fan and marine shelf deposits; 3) Langhian early post-rift terrigenous shelf to slope and locally coralgal carbonate platforms; 4) Late SerravallianTortonian late post-rift units,which present marine shales and transitional sandstones; and 5) Pliocene-Pleistocene undeformed deposits, which overlie the Messinian angular unconformity. 2.2. Geodynamic evolution The geodynamic evolution of the Catalan Coastal Ranges has undergone two major phases since the early Tertiary (Fontboté et al., 1990; Roca & Guimerà, 1992; Bartrina et al., 1992; Roca, 1994): 1) a Paleogene compressional phase which extended from the Eocene to the Calvet, F. et al. 16 Berta Mine Vein Filling The Berta mine, located 20km W of Barcelona, consists of a subvertical N60 oriented fault system developed in a late-Hercynian granite. Mineralized portions of the fault occur sporadically (pinch and swell) within a vein system that is at least 600m long and up to 5m wide. The veins are close to a major fault which controlled the Tertiary Valles-Penedes graben (Fig. 1). The Tertiary sediments close to the outcrops are essentially conglomerates and breccias and are occasionally cemented by calcite. The presence of two different and temporally separated hydrothermal events can be recognized from: 1. Vein mineralogy: two different mineral associations are present. A first, quantitatively the most important, is constituted by fluorite, sulphides (sphalerite, galena, chalcopyrite and pyrite), barite (I), calcite (I) and quartz (I), and was mined in the past; the second is composed by minor amounts of green octahedral fluorite (II), quartz (II), calcite (II) and barite (II). The paragnetic succession is showed in figure 2. QUARTZ FLUORITE SPHALERITE GALENA CALCITE BARITE Time Th = 180-200°C NaCl-H2O Tmi=-4 to 0°C Th = 80 to 150°C NaCl-CaCl2-H2O Tmi=-12 to -22°C Th = 100 to 150°C NaCl-H2O Tmi=-3 to 0°C Tmi=-8 to -6°C Figure 2. Paragenetic succesion of Berta Mine 2. Fluid chemistry: from fluid inclusion data, three types of fluids have been recognized. Fluid I, found in quartz I belongs to the NaCl- Geofluids III. F5 Field Guide. 15 - 16 July 2000 17 H2O system, has a low salinity (Tmi from –4 to 0°C) and relatively high temperature (between 180 and 230°C); fluid II, found in fluorite I, is a polysaline brine with high salinity (Tmi from –12 to –22°C) and temperatures ranging from 80 to 150°C. A positive correlation between Th and Tmi in fluid II suggests a mixing process during mineral precipitation. Fluid III, is the only one present in the green octahedral fluorite (II) although it is also found in secondary fluid inclusions in quartz I and fluorite I. It belongs to the NaCl-H2O system and has a low salinity (Tmi between –3 and 0°C) and a low temperature (from 100 to 150°C) (see Fig. 3). AR-11 no-oct. 5 -5 -15-25 75 125 175 225 AR-11 oct. Tmi Th Q AR-27 AR-9 AR-8 AR-11 RB-2 MBG-1a MBG-4 90 110 130 150 170 190 210 230 0 2 4 6 8 10 12 Th nº nº 0 -5 -10 -15 -20 -25 0 10 20 Tmi Figure 3. Analitical data of fluid inclusions in Berta Mine Br/Cl and I/Cl ratios obtained by the crush-leach method (Tritlla et al. 1995) confirm the presence of at least two different fluids. Halogen ratios in the fluid trapped in quartz I and in fluorite II (octahedral) are very close to seawater values whereas the fluids in fluorite I are clearly different. 3. Isotopic (radiogenic and stable) signature of the fluids: 18O of early quartz (I) has a value around 10.4‰. Assuming a T of formation of 200°C, the 18O of water in equilibrium is –1.2‰ (Canals and Cardellach, 1996). This value together with the fluid chemistry data (salinity and halogen ratios) point to a surficial (probably seawater) Calvet, F. et al. 18 origin for fluid I. 18O of water in equilibrium with fluid III calculated from 18O of calcite II at 155°C has a value of –1‰. This composition, together with the fluid chemistry suggests an origin related to the circulation of seawater through the system. The isotopic composition of the fluids responsible for the formation of the main stage of mineralization (fluorite I, sulfides and barite I) are not possible to calculate. However, fluid chemistry points to an evolved waters of surficial origin in accordance with the 34S values of sulfides, Sr and Pb isotopes. The uranogenic lead of main stage galena is compatible with a source related to granites, whereas the 208Pb/204Pb ratios point to a mixture with a higher Th/Pb source than that of the granites (Canals and Cardellach, 1997) (metamorphic basement (?)). 87Sr/86Sr ratios of vein fluorite, calcite and barite show different values matching the paragenetic sequence with values of the early minerals less radiogenic than the late ones. All values are in accordance with a granitic source rock whose 87Sr/86Sr ratio evolved with time. The 87Sr/86Sr ratio of octahedral fluorite is distinctly higher and can be explained by a leaching event of strontium from granites during at least Tertiary times. 4. Host rock alteration: in the enclosing rocks around the veins, the following mineral assemblages have been recognized (Roig and Canals, 1994): a) sericite, calcite, pyrite, chlorite and epidote; b) kaolinite; c) adularia and quartz crosscutting the previous assemblages. Constraints on the age of the hydrothermal events Crustal evolution of the eastern part of the Iberian Peninsula, based on the evaluation of Mesozoic extensional tectonics and the available geophysical data, includes three main successive evolutionary stages: Mesozoic crustal thinning, Paleogene crustal thickening and Neogene crustal thinning (Salas and Casas, 1993). Sequence stratigraphy, subsidence analysis and the integration of the basin fill data allowed Salas and Casas (1993) to identify four successive evolutionary stages in the basins of the eastern Iberian margin during Mesozoic extension: 1) Triassic rift (Late PermianHettangian); 2) Early and Middle Jurassic postrift (SinemurianOxfordian); 3) Late Jurassic and Early Cretaceous rift (Kimmeridgianmiddle Albian) and 4) Late Cretaceous postrift (Late AlbianMaastrichtian). Constraints on the age of the hydrothermal events can be inferred from: Geofluids III. F5 Field Guide. 15 - 16 July 2000 19 a. Field relationships: some of the Ba-F (Pb-Zn) mineralized veins in the CCR locally cut across lower Triassic strata. Consequently the age of the related mineralizing event must be at least younger than Triassic. b. Presence of hydrothermal barite cementing the lower Triassic conglomerates: significative amounts of barite are found cementing the lower Triassic conglomerates and sandstones (up to 50% of the total cement) in areas close to Ba-bearing veins enclosed within the Palaezoic basement. In the diagenetic sequence this cement represents an early precipitate (see Stop 3). c. Indirect age constraints: from Sr isotopes in fluorites and barites and S isotopes in barites from several F-Ba veins of the CCR, Canals and Cardellach (1993) were able to constrain the age of the main hydrothermal event. 34S of barites (15 to 20‰) are similar to those of sulphates from the Triassic-Jurassic evaporites of the area (11 to 19‰). Strontium isotopic composition of barite and fluorite fall within the range 0.7093 to 0.7167, reflecting at least two different strontium sources: the granites of the Hercynian basement and seawater or leached evaporites of Triassic-lower Jurassic age. The genetic model based on a mixing process suggest that the mineralizing processes occurred from about 220 Ma to 200 Ma, (Triassic to lower Jurassic). On the other hand, Solé et al. (1998) found an Ar loss affecting the K-feldspars of granites from the CCR (the most important enclosing rock of the F-Ba veins). This fact was attributed by these authors to an inhomogeneous distributed thermal event associated to fluid circulation that took place after the exhumation of the plutons, probably during the Mesozoic times (between 187 and 200 Ma). In Berta Mine, 87Sr/86Sr ratios in late fluorite and barite are distinctly higher (0.7168-0.7187). Using a similar mixing model as for the earlier vein filling minerals, these ratios would indicate an age for this mineralizing event as young as Miocene. d. Recent hydrothermal activity: in the CCR, thermal springs are found associated to the Valles-Penedes graben and the ENE- Calvet, F. et al. 20 WSW and NW-SE faults, reactivated during the Neogene. According to Fernández and Banda, (1990) these waters have a meteoric origin and were heated during their flow to depths of 3 km, ascending then through fractures up to the surface. An example is the La Garriga-Samalús hot spring where Canals et al. (1990), found precipitates of fluorite in a core drilled in a strongly altered granite, at depths between 500 and 600 m. Homogenization temperature and salinity of fluid inclusions from this fluorite are similar to those found in fluorite II of Berta Mine. This observation is an evidence of the mineralizing potential of the recent hydrothermal systems in the area. The available data suggests that hydrothermal activity in Berta Mine took place in two separate events. The first was probably related to the extensional regime that took place during the Mesozoic (Upper Triassic-Lower Jurassic…). Age constraints from vein mineral geochemistry, age of K-feldspat resetting of enclosing granites and the presence of hydrothermal barite cementing the lower Triassic conglomerates supports this hypothesis. This hydrothermal episode has been recognized in most of the Hercynian massifs of Central and Southern Europe (e.g. Jébrak (1984); von Gehlen (1987); Galindo et al. (1994)). The second event is clearly later, has significative geochemical differences and produced a distinct mineralogical record. This hydrothermal episode could be related to the filling of veins in the Burdigalian-Serravalian rocks (see stop 2). The similitude of the Sr isotopes between the green octahedral fluorite (II) of Berta Mine and calcite from the stop 2 veins would support this hypothesis. Thus, the age of the second episode of mineralization in Berta Mine could be related to a syn-rift stage during the Miocene. Mineralogical differences between the veinlets enclosed in granites and those enclosed in the syn-rift sediments of Tertiary age could be explained as the result of a mineral zonation around the hydrothermal fluid source. Geofluids III. F5 Field Guide. 15 - 16 July 2000 21 S3. Rubí Location: The outcrop is located around a petrol station in Road B1413 from El Papiol to Sabadell (Km 6.5) in an abandoned quarry in the proximity. Objectives: Synrifts paleofluids in synrifts fractures affecting proximal filling half-graben deposits. Outcrop description The fractures affect the lower sediments (early Burdigalian) which consist of conglomerates arranged in alluvial fan sequences located in the marginal parts of the basin. The structural analyses show that the fractures originated in a single fracturation phase which was approx N-S-trending, with nearly vertical fractures, dipping from 66 to 90° either to the east or to the west, with a decimetric to metric spatial distribution. The direction of the extension within the fractures is perpendicular to the vein wall, and no displacement between both lips can be observed, indicating the extensive character of the fractures. These fractures formed perpendicularly to the direction of maximum incremental stretching as a result of brittle or semi-brittle failure. The calcite cement filling the fractures is affected by horizontal slickensides formed during the Serravalian NNW-SSE compresion, indicating that the fractures formed during the early syn-rift stage. Early Burdigalian host conglomerate petrology The conglomerates are made up of clasts of Jurassic and Cretaceous grey dolomite (70%), metamorphic pelitic shales (10%), Cretaceous limestones (6-10%), quartz (3%), granitoids (3-5%), and sandstones of Bundsandstein facies (3-5%). The matrix of the conglomerate is formed by reddish sandstone, very rich in quartz and metamorphic pelitic grains. The conglomerates are clast-supported and heterometric. The clasts are angular and mostly range from 5 to 15 cm in size, although some can reach up to 2 meters. The deposits are subhorizontally stratified (from 070/09 N to 110/08 N) and contain hectometric scale faults, microfaults and tension fractures. Some of these fractures are filled with calcite cements. Calvet, F. et al. 22 The early Burdigalian conglomerates are cemented by two generations of calcite pore-cement (Travé & Calvet, in press). The first generation consists of a microstalactitic pore-cement displaying a geopetal distribution below some of the clasts with a black-bright yellow-black zoned luminescence. The second generation of porecement consists of anhedral spar calcite, which is non luminescent, presenting either a rim distribution or a mosaic pattern filling the porosity between the clasts of the conglomerates. Calcite veins The calcite veins filling the fractures affecting the early Burdigalian conglomerates, are from 1 mm to 30 cm thick. Most fractures present two filling stages, although one single stage is also observed in some fractures (Travé & Calvet, in press). The calcite veins are affected by horizontal slickensides indicating that the calcite fillings occurred prior to the Serravallian strike-slip movement. Petrology: The first fracture-filling stage of calcite cement has a milky aspect in hand specimens. It consists of millimetric to centimetric irregular alternances of diffuse laminated "crystalline" bands and thin sedimentary laminae. The "crystalline" bands consist of different textures such as palisade calcite cement, anhedral spar calcite texture, rhombic calcite cement, and euhedral spar calcite filling vug cavities. The palisade calcite cement displays a laminar disposition and is formed by submillimetric to few millimetre thick laminae with a smooth wave morphology. In general, the base of the laminae consists of equant-to-bladed subhedral crystals grading rapidly to the palisade morphology crystals. The palisade calcite crystals are from 100 to 500 µm long and from 10 to 50 µm wide. They are curved with a pseudoradial disposition and look dirty in the optic microscope presenting undulating extinction. The palisade calcite crystals are non, or dull-orange, luminescent. The anhedral spar calcite forms individual submillimetricmillimetric laminae or laminae interlayered with the palisade calcite laminae. The crystals are equant or slightly elongated, anhedral to subeuhedral and normally range from 21 to 500 µm in size, though some can reach 2 mm. The anhedral spar calcite presents millimetric pseudo-fibres and have been interpreted as aragonite Geofluids III. F5 Field Guide. 15 - 16 July 2000 23 pseudomorphs.The anhedral spar calcite is dull-orange to non luminescent. The sedimentary laminae, from 100 µm to 5 mm thick, are made up of lime-mudstones, submillimetric calcite laminae, clasts of calcite crystals and, locally, micritic grains. The sediment layers present submillimetric diffuse lamination and, locally, a geopetal disposition. The calcite laminae and the detrital crystals are non-luminescent, whereas the lime-mudstone sediment is dull-orange luminescent. The second fracture-filling stage has a translucent aspect in hand specimens. It consists of euhedral spar-megaspar calcite cement. This cement also fills the host conglomerates and microfractures affecting the first filling stage. The euhedral spar-megaspar calcite is made up of crystals measuring from 50 µm to 1 cm, with abundant mechanical planar twin planes. It has a zoned luminescence with dull to bright orange to bright yellow bands. Elemental geochemistry: The first filling stage calcite cements are characterised by having a high content of Mg and variable contents of Mn, Fe and Sr. Most of the second filling stage calcite cements are characterised by having Mg, Fe and Sr contents that are lower than 2000 ppm, and variable Mn content Stable isotopes: The _18O and _13C values of the calcite cements from the two fracture-filling stages plot in two different well defined areas. The values of the first fracture-filling stage calcite cements (n=21) plot in a very narrow range, the _18O values range from –9.0 to -5.2 0/00 PDB and the _13C range from –9.3 to -6.7 0/00 PDB. No differences between the different cements are observed. The _18O values of the second fracture-filling stage calcite cements (n=15) range from –16.2 to –9.9 0/00 PDB and the _13C values range from –7.4 to -1.9 0/00 PDB. Strontium isotopes: The calcite cements from the two fracture-filling stages show different 87Sr/86Sr ratios, the values of the second filling stage being more radiogenic.The 87Sr/86Sr values of the first stage cements range from 0.71088 to 0.71361and the 87Sr/86Sr values of the second stage cements range from 0.71671 to 0.71800. Calvet, F. et al. 24 Interpretation and discussion The structures affecting the lowermost sediments (early Burdigalian conglomerates) deposited in the SE margin of the Vallès-Penedès halfgraben correspond to the early deformation of syn-rift sediments. Strike-slip movements which are Serravallian in age post-date the two stages of the calcite filling the fractures. The elemental geochemistry similarities of the two calcite pore-cements in the host conglomerate and the calcite filling the fractures (first and second fracture-filling stages) indicate that the fractures were formed in porous and water-rich host conglomerates. Cementation of the host conglomerates took place slightly before and synchronously with fracture cementation. The first pore-cement in the host conglomerates and the first fracture-filling stage of calcite cement illustrate the fluid behaviour characteristic of the vadose meteoric zone. The meteoric fluid was a superficial fluid which was influenced by soil-CO2, percolating through existing cavities (fractures), originating speleothems. Fluid circulation within conduits produced a great variability of the elemental geochemistry by fluctuation in supply of the different ions and in the water flux, producing short-lived microenvironments. These microenvironments were characterised by particular redox conditions changing through time from oxidising to reducing groundwaters as result of a progressively more confined aquifer. This fluctuation in the redox conditions in the aquifer correlates with the main calcite textures. The great variability in the Sr2+ and Mg2+ content of the calcite cements is interpreted as reflecting different original mineralogies (calcite and aragonite). The _18O_and _13C values of the first fracturefilling stage calcite, ranging from –9.0 to –5.2 and from –9.3 to –6.7 0/00 PDB, respectively, are typical values for calcite precipitated from meteoric waters. Assuming precipitation temperatures of 15-20°, calcites with _18O between –9.0 and –5.2 0/00 PDB would have precipitated from fluids with –9.3 and –4.4 0/00 SMOW, which are within the range of values or slightly lower than modern rainfall (-6.4 to –4.6 0/00 SMOW) in the same area. The 87Sr/86Sr ratios of the first-filling cements present values that are relatively similar to the first mineralazation event in the adjacent horst (Berta mine), which acted as a possible source flux area. A major change in fluid composition occurred prior to the precipitation of the second generation of pore-cement within the host Geofluids III. F5 Field Guide. 15 - 16 July 2000 25 conglomerate and the second fracture-filling stage of calcite cement, which illustrates fluid behaviour characteristic of the phreatic meteoric zone. During the Langhian these sediments were buried up to a few hundred meters and the porosity of the conglomerates was filled with water. The meteoric fluid was a more evolved fluid owing to significant interaction with sediments coupled with an increase in the residence time of the water in the aquifer and the reservoir effect. The high temperature (hydrothermal origin) during calcite formation is demonstrated by the low _18O_ values of the second fracture-filling stage calcite cement, ranging from –16.2 to –9.9 0/00 PDB. Thermal waters due to high temperature leaching of limestones along flowpaths are also suggested by the relatively enriched _13C values of the second fracture-filling stage, ranging from –7.4 to –1.9 0/00 PDB (Bakalowicz et al. 1987). A hydrothermal event characterised by temperatures around 120°C, low salinities and high radiogenic 87Sr/86Sr values (between 0.7168 and 0.7194) has been reported in the Miocene close to the studied area (Canals and Cardellach, 1996). The 87Sr/86Sr ratios of this horst mineralization and the second fracture-filling stage of calcite cements are similar and relatively radiogenetic, suggesting a similar origin produced during this hydrothermal event. The change from a vadose to a phreatic environment took place during the evolution of the rifting in a well constrained period of time ranging from early Burdigalian to late Langhian-early Serravallian. Calvet, F. et al. 32 second generation of spar calcite cement. Locally, the intensity of the fracturation in the dolomite body forms centimetric patches of breccia. Dolomite petrology The dolomite occurs as detrital dolomite, replacive dolomite and dolomite cement. Detrital dolomite: The detrital dolomite consists of dolomite clasts and grains, sand-size detrital dolomite crystals, and a small proportion of sand quartz grains. The dolomite clasts and grains are well rounded ranging from 100 µm to several mm in diameter, and are basically from Mesozoic dolomites (Triassic and Cretaceous), either non or dull orange luminescent. The detrital dolomite crystals have a cloudy nucleus, non luminescent to dull orange luminescent, and a translucent overgrowth rim 7-21 µm in thickness, bright yellow luminescent. The matrix of the detrital dolomite consists of dolmicrospar, clay and Fe-oxids. The dolmicrospar is made up of dolomite idiotopic rhombs from 7 to 35 µm in size and silt-size detrital dolomite crystals. The larger idiotopic rhombs present a thin outer rim. The centre of the rhombs is non luminescent or dull orange luminescent and the outer rim is bright yellow luminescent. Replacive dolomite: The replacive dolomite shows three textures: pink idiotopic dolsparite, white idiotopic dolsparite and white xenotopic dolsparite. The pink idiotopic dolsparite texture consists of crystals with the intercrystalline porosity filled with pink-to-brown matrix. The dolomite crystals have an euhedral (rhombic) to subeudral (curved rhombic) habit varying from 27 to 400 µm in size. The crystals have a black nucleus which may account for 50% of the crystal. The dolomite crystals are dull orange luminescent, with a bright yellow luminescent irregular band within the black nucleus. The white idiotopic dolsparite texture presents euhedral to subedral crystals, ranging from 25 to 250 µm in size, and exhibit undulose extinction. Less than 10% of the dolomite crystals have a black nucleus. The dolomite crystals display a non to dull orange luminescent inner zone and a thin dull orange luminescent outer zone. The inner part of the dolomite crystals is dissolved and partially to completely filled with spar calcite cement. Geofluids III. F5 Field Guide. 15 - 16 July 2000 33 The white xenotopic dolsparite, locally present, consists of crystals exhibiting an anhedral habit. The crystals are from 50 to 500 µm in size, show undulose extinction and subgrain formation, and are non luminescent. Dolomite cement: Dolomite cement forms the first generation of fracture filling and displays a rim disposition. The dolomite rim in the submillimetric fractures consists of one-crystal wide fringes, whereas in the millimetric or centimetric fractures the rim, up to 5 mm thick, includes several crystals. The dolomite cement fills the breccia porosity. The dolomite cement is constituted by euhedral dolomitic crystals, from 50 µm up to 1mm in thickness, displaying c-axes perpendicular to the fracture walls with a normal to slight undulose extinction. The dolomite cement presents microvug porosity filled with spar calcite cement. The dolomite cement is non to dull orange luminescent. Locally, it exhibits a zonation with non-luminescent zones and dull orange luminescent zones. Dolomite textures distribution: The distribution of the dolomite textures in the dolomite body indicates a gradient of the dolomitizing process. The white idiotopic and xenotopic dolsparites are preferentially located close to the N-S trending normal fault and close to the fractures, grading to the pink idiotopic dolsparite. These two textures are interpreted as the most evolved stages of the dolomitization process where all the textural signs of the original detrital deposits have been erased. The pink idiotopic dolsparite is preferentially located few centimetres from the N-S trending normal fault and few millimeters from the fractures. This texture preserves ghosts of the original detrital deposits (clay matrix, some quartz grains and dolomite clasts) and is interpreted as the initial stage of the dolomitization process. This dolomite texture grades to the detrital dolomite. The detrital dolomite presents abundant remains and ghosts of clasts and detrital dolomite crystals and is interpreted as the result of a lower degree of dolomitization. The overgrowth rims and the dolmicrospar dolomite idiotopic rhombs are related to this early stage of the dolomitization process. Finally, the detrital dolomite grades to the nondolomitized detrital deposits. Calvet, F. et al. 34 1.6.3. Dolomite geochemistry Elemental geochemistry: The dolomite textures show two different elemental compositions: one formed by the detrital dolomites (clasts and the nucleus of the detrital dolomite crystals) and the other formed by the crystals originated by the dolomitizing process (overgrowth rims, replacive dolomites and dolomite cement). The dolomite clasts and the nucleus of the detrital dolomite crystals have relatively rich-magnesium dolomites (averaging 44.9 and 45.1 % mol CO3Mg content respectively) and are low in iron (averaging 580 and 310 ppm respectively). The dolomite overgrowth rims, the replacive dolomite and the dolomite cement have relatively poor-magnesian dolomites (a mean of 40.1, 41.3 and 41.6 % mol CO3Mg content respectively) and are relatively high in iron (a mean of 2.660, 1.720 and 1.680 ppm respectively). These dolomites are practically ferroan dolomites that have been defined by Taylor and Sibley when there is more than 2 mol % FeCO3. The dolmicrospar matrix of the detrital dolomites formed by a mixture of detrital dolomite crystals (rich-magnesian and low iron) and dolomitized textures (low-magnesian and high iron) show intermediate values. Stable isotopes: The oxygen and carbon isotopic compositions of the different dolomites (detrital dolomite, replacive dolomite and dolomite cement) have similar values. The oxygen isotopic composition of detrital dolomites vary from -4.5 to -3.4‰ PDB and carbon isotopic composition varies from -2.8 to -2.1‰ PDB. The different replacive dolomite textures (pink idiotopic dolsparite, white idiotopic dolsparite and white xenotopic dolsparite) have very similar oxygen and carbon isotopic compositions. The d18O varies from -4.8 to -3.2‰ PDB and the d13C ranges from -2.9 to -2.0‰ PDB. The dolomite cement has d18O varying from -4.6 to -3.2‰ PDB and d13C varying from -3.1 to - 2.4‰ PDB. Strontium isotopes: The 87Sr/86Sr ratio of replacive dolomite varies from 0.70830 to 0.70896 while that of dolomite cement is 0.70868. The dolomite strontium values (except one sample) coincide with the range of the of the strontium seawater from the end of early miocene, -20 my, to mid miocene, -11 my. This range, in the literature varies from approx 0.7085 at -20 my to approx 0.7089 at -11 my. But Geofluids III. F5 Field Guide. 15 - 16 July 2000 35 the dolomite strontium isotope values present a small dispersion, which could be attributed to the influence of the clastic content (scattered metamorphic clasta, feldspar grains and clay matrix) present in the continental terrigenous deposits. Calcite cement petrology and geochemistry Spar calcite cement forms the second generation of fracture filling and fills locally intercrystalline porosity. The most abundant crystal habit is euhedral but subhedral and anhedral morphologies are also present. The intercrystalline limits are straight and present numerous triple junctions. The crystals range from 100 µm to 10 mm in size and have a translucent aspect and uniform extinction The calcite cement is non-luminescent, although locally it exhibits a complex zonation. In the fractures the boundary between dolomite cement and calcite cement is defined by a microcorrosion surface. The spar calcite cement has a high magnesium content varying from 940 ppm to 6.300 ppm (average 2.390 ppm), a low iron content (from below the detection limit to up to 6.670 ppm), and strontium values are from below the detection limit to 750 ppm. The spar calcite cement presents an oxygen isotopic composition ranging from -7.0 to -5.1‰ PDB and a carbon isotopic composition varying from -8.2 to -6.0‰ PDB. The 87Sr/86Sr ratio of the calcite cement ranges from 0.70794 to 0.70818. The calcite cement strontium isotope values present a very narrow range of variations. The provenance of the meteoric water that originated the calcite cement was the Garraf horst (with Triassic and Cretaceous materials). The Triassic consists of detrital units and marine carbonates (mainly dolomites) units which currently present an important karstic aquifer reservoir. The Cretaceous is made up of marine carbonates. The meteoric waters have in general a strontium isotope signature similar to the materials through which the waters pass (Vahrenkemp, 1998). The 87Sr/86Sr ratio of the calcite cement shows a possible triassic influence which presents similar values of the triassic dolomites (Calvet et al., in progress). By contrast, the 87Sr/86Sr ratio of the calcite cement are not cin agreement with the strontium values of the Cretaceous limestones (Travé et al., 1998). In consequence, the fluids originating the calcite cement were meteoric in origin and passed through the triassic aquifers. Calvet, F. et al. 36 Discussion and interpretation Fluid composition: The dolomite crystals, which were formed during the dolomitizing process (overgrowth rims, replacive dolomites and dolomite cement) have a relatively high iron content, ranging from 2.030 to 3.390 ppm, from 1.305 to 2.175 ppm and from 1.180 to 2.980 ppm, respectively. Applying the distribution coefficient equation the Ca/Fe molar ratio of the dolomitizing fluid ranges from 120 to 1.735, which are consistent with formation water. Dolomitizing fluid pathway: The geometry of the dolomite body and the distribution of dolomite textures within the outcrop suggest that the dolomitization process started along the fault plane moving in the direction of flow. The N-S fault acted as vertical conduit for the dolomitizating fluids. The rocks of the footwall of the N-S normal trending fault (Muschelkalk dolomites and Keuper mudstones) have a very low porosity and permeability whereas the rocks of the hangingwall (mainly conglomerates and lithoarenites) have a relatively high porosity and permeability. These differences in the petrophysical parameters controlled the lateral migration of the dolomitizing fluids and produced an asymmetric dolomite body. The presence of 30 metres of clays of the uppermost clayed unit probably acted as a seal for the dolomitizing fluids (Fig. 4). Figure 4. Fluid modelling in Vallès-Penedès half-graben. Geofluids III. F5 Field Guide. 15 - 16 July 2000 37 Origin of dolomitizing fluids. Magnesium and Iron: The origin of the magnesium of the dolomites studied was probably the miocene seawater. The marine shales, which are late Burdigalian to early Serravallian in age, are located in the center of the basin and contain a large volume of marine seawater, which could have been expelled during the shale compactation. The 87Sr/86Sr ratio of the miocene seawater varies from 0.7088 to 0.7090 (Burke et al., 1982, DePaolo, 1986, Hodell et al., 1991, Miller et al., 1991 Hodell and Woodruff, 1994) which are similar to the replacive dolomite ratios (from 0.70830 to 0.70896) and to the dolomite cement ratio (0.70868). The only possible source of iron in these dolomites is the marine shales filling the basin center. Mass-balance calculations: If the magnesium source was the miocene seawater expelled from the shales of the basin center, it is possible to calculate the volume of seawater necessary to produce the dolomite body. The volume of the dolomite body can be calculated assuming that it is 50 m long, 25 m wide and has a thickness of 50 m. The volume of the dolomite body is 137.500 m3. Taylor and Sibley (1986) calculated in the Ordovician of the Michigan basin that 1.700 volumes of seawater was necessary to dolomitize one volume of host rock. The seawater volume necessary to dolomitize the example under study, and in accordance with the data of Taylor and Sibley (1986), is 23.275.000 m3 ( = approx 2 x 107 m3). The volume of the seawater presumably ttrapped in the marine shales porosity can also be calculated. The approx in the size of the marine shale body is 200 m thick and has a distribution area of 10 km for 10 km. The volume of the shale body is 2 x 1010 m3. If the shales had an original porosity of 50% the seawater ttrapped in the shales was 1010 m3. The differences in volume between the available seawater (1010 m3) and the volume of seawater (2 x 107 m3) necessary to dolomitize the dolomite body accounts for the marine source of magnesium. Timing: One important objective is to determine the timing of the dolomitization process. The relationships between the dolomite body and some structural characters allow us to deduce the following: i) The slickenside of the N-S trending normal fault is vertical, implying that the fault movement occurred after the tilting (mid SerravallianTortonian) of the host rocks (the continental terrigenous deposits dip up to 30º). ii) The number of fractures associated with the N-S trending Calvet, F. et al. 38 normal fault is more abundant close to the fault plane. The dolomite cement filling the fractures present the same geochemical characteristics as the replacive dolomite, indicating a unique dolomitizing fluid. Consequently, fracturation and dolomitization occurred at the same time. Geofluids III. F5 Field Guide. 15 - 16 July 2000 39 S7. FINI-MIOCENE FLUID FLOW MODELING AND ECONOMIC IMPORTANCE Objectives: Late postrift paleofluid events and their economic importance. Fluid flow during dolomitization The late Miocene paleotopography was probably similar to the present one, which is characterized by an elevated NW rift shoulder represented by the Prelittoral Range with elevations reaching up to 700 m and a less pronounced flank which is formed by the Garraf horst. The graben itself forms a topographic depression. This configuration implies a topography-driven flow from the NW part, where mesozoic and paleozoic rocks are exposed, and to a lower degree a topography driven flow from meteoric infiltration on the Garraf horst where crops basically calcareous mesozoic rocks. These systems depend strongly on the hydraulic conductivity within the sediment fill of the basin. The sediment fill is characterized by a strong asymmetry with thick permeable conglomerates at the NW margin and more finegrained, less permeable rocks with less thickness at the SE margin. The basin center is dominated by finegrained low permeable rocks. Basal conglomerates equivalent to the Barcelona formation offshore in the Barcelona-half graben (Bartrina et al., 1992), are unconformably deposited above mesozoic and paleozoic rocks at the base of the graben. These basal conglomerates are unsorted, subangular, occasionally reaching block size, and may play a major role as hydraulic conduit for fluids infiltrating on the NW rift shoulder as they constitute the only continuous hydrostratigraphic unit within the graben. Additionally, the bounding faults probably serve locally as preferential pathways, as can be seen from numerous thermal springs in the Vallès-graben in the NE prolongation of the Penedès graben. The conceptual model used in this study focusses on the possibility of late Miocene fluid transfer from the NW rift shoulder by infiltration through high permeable pathways such as the coarse grained alluvial units and the bounding fault, assuming an elevated conductivity of the fault. Such fluids could pass through the basal conglomerates towards the SE margin of the graben, where they mix with the topography Calvet, F. et al. 40 driven system from the Garraf horst and move upward along the bounding fault. Finally, they infiltrate laterally into the continental terrigenous deposits and produce dolomitization. In order to test this conceptual model of paleo-fluid flow, a twodimensional finite element model was applied, which simulates fluid flow along a vertical cross section through the graben, incorporating the paleozoic and mesozoic base of the graben, the uplifted shoulders and the asymmetric basin fill. A simplified geological NW-SE section from the Prelittoral Range to the Garraf horst was taken. The bounding faults are represented in the model as zones with hydraulic conductivities elevated by a factor of 10 with respect to the surrounding rocks. Low permeabilities in the center of the basin correspond to the fine-grained sediment fill, whereas elevated conductivities are related to coarse grained sediments close to the NW bounding fault. The simulated flow field shows, that the hydraulic gradient created from the topographic relief may be sufficient to drive fluids from NW rift shoulder along the bounding fault into the basal conglomerates towards the SE margin of the basin, where these fluids ascend along the SE bounding fault and infiltrate laterally into the local continental terrigenous. For the assumed hydraulic conductivities, flow velocities are calculated in the order of 100 m/y in the surficial sediments with elevated topographic relief, whereas they decrease rapidly towards depth and reach values in the order of meters per 1000 years in the basal conglomerate unit. Instead of a continuous flow system as suggested by this model we rather think of a episodic flow system with periods of limited duration, with hydraulic conductivities controlled by tectonic activity. The SE bounding fault may locally drain fluids due to opening of fractures due to relaxation of tectonic stress. Such episodic flow systems would better account for the temperatures under which dolomites have been formed. The temperature distribution (calculated without advective transport) indicates maximum temperatures of the sedimentary basin fill of about 90º at 2.400 m with a basal heat flow of 82 mW/m2. The values however, may change if advection is taken into account. Geofluids III. F5 Field Guide. 15 - 16 July 2000 41 Late postrift paleofluids: General diagenetic sequence in the late neogene horst-graben system The general diagenetic sequence during the late neogene times in the neogene horst-graben system of the NE of the eastern of the Valencia Through presents a general pattern. This diagenetic sequence can been divided into the following stages: 1) reservoir porosity formation; 2) hydrocarbon migration; and 3) late diagenetic stage. 1. Reservoir porosity formation: The reservoir porosity formation stage is related to the late neogene dolomitization process. This dolomitization process affects a variety of tertiary statigraphic units presenting different lithologies and perhaps mesozoic limestones. The dolomitization process presents the following characteristics: i) The end dolomite textures are similar (from idiotopic to xenotopic textures) and these textures are independent of the parent rock (early neogene continental terrigenous deposits, early to middle neogene detrital limestones, middle neogene shallow marine limestones and reefal limestones). The dolomite bodies present abundant porosiy, especially vug, breccia, intercrystalline and fracture porosities. The origin of these porosities are related to the dolomitizing fluids. ii) The dolomitization fluids are from warm (below 50º C) to relatively hot (approx 100ºC). In general the fluid had a high Fe/Ca ratio indicating a formation water. The dolomite 87Sr/86Sr ratios suggest a miocene seawater and locally some original modification due to the fluid-rock interaction or other influences. iii) The dolomites are fault-related in the onshore outcrops and probably also in the offshore examples. The dolomite bodies in onshore and in offshore are located along the horst borders where the deformation is more intense. And these bodies present a certain linear trend parallhronous pocesses. iv) These fault-related dolomite bodies form reservoir compartments. The non continuous distribution of oil reservoirs offshore is perhaps the consequence of this compartmentalization. The compartmentazation of oil reservoirs in fault-related dolomites is characteristic of the Michigan Basin Calvet, F. et al. 48 composition has changed through time, so we have assumed a phreatic system with high flow rates (Thiry, 1999). In the case of the Montjuïc sandstones, meteoric flow systems and convective fluid circulation in the onshore Barcelona half-graben may have interacted (Fig. 7). Phreatic waters flux Palaeozoic wall rock Tertiary deposits Flexural Subsidence Barcelona Graben Collserola Horst Montjuïc Topographical Flux Convctive Flux Compaction Flux Figure 7.Fluid flow in the Barcelona Graben. The presence of alunite indicates the existence of saline and acidic fluids. As feasible process which may have contributed to the acidification is the oxidation of pyrite (Goldbery, 1978, 1980). Oxidation of Fe2+ may release additional H+ (Mann, 1983 & McArthur et al., 1991). Fe2+ may be derived from microbial activity on iron oxide surfaces (McArthur et al., 1991). 2FeS2 + 2H2O + 7O2  2FeSO4 + 4H+ + 2SO42- (Pyrite) 4Fe2+ + 6H2O + O2  4FeOOH + 8H+ (Goethite) Oxidation of organic-rich and pyritic material of the marly strata in the Miocene section of Montjuïc may have caused acidification in the fluids passing through the sandstone units. Hydrolysis reactions of aluminosilicates at acidic conditions may release enough silica to supply oversaturated solutions capable of inducing Montjuïc sandstone silicifications in the more permeables units of de section. Geofluids III. F5 Field Guide. 15 - 16 July 2000 49 S10. Caldetes Geological setting The Caldetes thermal water spring is located 30 km NE of Barcelona (Fig. 1) at about 300 m distance from the coastline on the foothills of the Corredor-Montnegre massif, which is formed in its southwestern section by granites belonging to the Maresme batholith and metamorphic paleozoic rocks in its northeastern section. The Corredor-Montnegre massif extends 75 km in length and 20 km width with altitudes up to 773 m. It is part of the Catalan Coastal Ranges, which comprise a horst-block system of Neogene grabens which belong to the north-western edge of the Valencia Trough (Roca & Guimerà, 1992). The Corredor-Montnegre massif represents a horst and separates the offshore Barcelona halfgraben from the onshore Vallès-Penedès graben. Both grabens are neogene basins, bounded on their northwestern margin by SE dipping listric faults and with a sedimentary fill reaching 2000 m thickness in the Vallès-Penedès graben and 5000 m in the Barcelona halfgraben. The Catalan Prelitoral Range with mostly paleozoic metamorphic rocks and altitudes up to 1712 m in the Montseny massif limits the northwestern boundary of the Vallès-Penedès graben. Hydrochestry of the present post-rift hydrothermal system The thermal waters in the Catalan Coastal Ranges show varying chemical compositions. Two principal hydrochemical facies can be distinguished (Albert, 1976): 1. Sodium bicarbonate waters, circulating principally in granites, 2. Sodium chloride waters with elevated total dissolved solutes, circulating principally in sedimentary units. There is, however, no strict coincidence between the chemistry of thermal waters and the structural position of the springs. The Caldes de Montbui spring and the La Garriga spring, for example, are both located at the bounding fault between the Catalan Prelitoral Range and the Vallès-Penedès graben (Fernández & Banda, 1988), but they exhibit different chemical compositions. The Caldes de Montbui thermal waters are intensely mineralised chloride sodic waters, whereas the La Calvet, F. et al. 50 Garriga thermal waters exhibit a composition which is typical for a circulation through granite rocks. The Caldetes thermal water spring is located about 25 km SE from La Garriga and shows a chemical composition similar to Caldes de Montbui, however with slightly lower chloride and sodium contents. The temperature is constant at 39ºC. Application of hydrochemical geothermometers shows that the last water-rock equilibrium is at 93ºC, indicating a deep circulation of these waters (Albert, 1976). Such a temperature explains the +0.5 shift in d18O, with respect to local meteoric compositions, due to fluid interaction with silicates (fig. 8). LNWL-CM LNWL-M GMWL δ 18 O δD δ 18 O thermal water shift Non thermal springs transec along the Corredor-Montnegre-Vallès Caldetes thermal waters Non thermal springs along the Montseny -30 -35 -40 -45 -50 -55-9.0 -8.5 -8.0 -7.5 -7.0 -6.5 -6.0 -5.5 -5.0 Figure 8.Caldetes isotopic composition as a result of fluid-rock interaction. LMWLCM: Local Meteoric Water Line of the Corredor-Montnegre massif. LMWL-M: Local Meteoric Water Line of the Montseny massif. In order to determine the origin of the thermal waters we plotted the dD and d18O values versus elevation in the Corredor-Montnegre and Montseny massifs and compared the data to the isotopic compositions of the Caldetes thermal waters. In order to avoid seasonal effects in precipitation, we sampled springs with limited recharge areas at welldefined altitudes. Geofluids III. F5 Field Guide. 15 - 16 July 2000 51 Taking the Corredor-Montnegre and Montseny dD-altitude and d18O-altitude and taking into account the d18O shift observed in figure 8, the recharge altitude is determined between 550 m and 650 m (see the case of dD-altitude in fig. 9). Corredor-Montnegre springs at well defined recharge altitudes Montseny springs at well defined recharge altitude Caldetes thermal waters (deuterium al emergence altitude) δD-recharge altitude regression lines δD Recharge altitude (m) 1400 1200 1000 800 600 400 200 0-55 -50 -45 -40 -35 -30 Figure 9.dD-recharge altitude regression line for Corredor-Montnegre and Montseny massif. Several lines of evidence indicate that the origin of the Caldetes thermal waters is not by infiltration of meteoric waters in the Corredor massif, but in the Montseny massif, at elevations of about 750 m: 1. The Corredor-Montnegre massif reaches only locally elevations of more than 700 m. 2. Additionally, the hydrochemistry of the waters points towards a circulation in a sedimentary aquifer, whereas the Corredor-Montnegre massif is constituted by granite rocks and paleozoic metamorphic rocks in its northeastern section. 3. Finally, the spring discharge at Caldetes is constantly in the order of 150 L/min, indicating, that the aquifer has sufficient extension to equilibrate seasonal changes in recharge. Assuming a mean precipitation of 680 L/m2/y and a 1% infiltration rate, the recharge area required for this spring discharge is in the order of 12 km2. This area is Calvet, F. et al. 52 clearly above what can be expected in the Caldetes area because a recharge surface of this size at the indicated altitude is not available. Quantitative modelling In order to test this hypothesis, we applied a finite element model to a simplified geologic cross section (Fig. 10), incorporating the principal hydrostratigraphic units, starting at the Montseny area, crossing the Vallès-Penedès graben at Sant Celoni and the Corredor massif at Caldetes. The cross section ends within the offshore Barcelona halfgraben. The cross section has a horizontal extension of 39 km incorporating the principal extensional faults of the Vallès-Penedès graben and the Barcelona halfgraben, as determined from seismic sections. These faults present the principal hydrostratigraphic units which might provide a circulation from the Montseny massif to the thermal water spring at Caldetes. They are incorporated in the hydrogeologic model as zones of elevated hydraulic conductivity. Using this model we identified fault conductivities and topographic gradients necessary to cause a regional flow along these faults. We also calculated advective transport of a hypothetic tracer introduced in the Montseny massif and advective heatflow is calculated, assuming a basal heatflow of 80 mW/m2 and a fixed surface temperature of 10º C. The 2D modeling code "basin" (Bitzer, 1999) was used. Figure 10. Finite Element-mesh and structural units of the simulated cross section. Geofluids III. F5 Field Guide. 15 - 16 July 2000 53 Modeling results The distribution of hydraulic conductivity assumes permeabilities in the order of 10-9 to 10-11 m/s for the granitic rock in most areas of the model. The faults are introduced as zones of elevated permeability with 10-8 m/s, and 10-7 m/s in its most surficial areas. The sedimentary rocks of the Vallès-Penedès graben and Barcelona halfgraben are incorporated as permeable zones with values ranging between 10-6 to 10-8 m/s, depending on sediment type and burial depth. The simulated flow field shows that fluids infiltrating on the Montseny massif are partially drained along the hydraulic conducting fault and move towards the Barcelona halfgraben, where they are finally drained along the bounding faults of the Barcelona halfgraben. In order to test the hypothesis that the isotopic composition of the Caldetes thermal waters corresponds to meteoric infiltration on the Montseny massif, we performed a solute transport calculation by injecting a hypothetic "conservative" tracer (in fact deuterium acts as a conservative tracer in a granitic context) on those parts of the Montseny massif with elevations above 700 m. Solutes are transported along the fault zones and finally reach the Caldetes site. The heatflow model assumes thermal conductivities in granite in the order of 0.9 cal/(m s C) and 0.6 - 0.3 cal/(m s C) for sediment, depending on its porosity. Specific heat for granite was set to 190 cal/(kg C) and 220 - 500 cal/(kg C) for sediment depending on its porosity. The distribution of vertical heatflow (Fig. 11) shows the effect of advective transport. The coastal areas with mainly vertical fluid flow exhibit elevated vertical heatflow rates. Calvet, F. et al. 54 Figure 11.Vertical heatflow (given in mW/m2). Conclusions 1. Hydrochemical and isotopical data from Caldetes thermal waters indicate a large scale fluid flow system with recharge in the Montseny massif. 2. Fluid circulation along the principal fault zones as preferential pathways is capable of explaining the hydrochemical and thermal characteristics of the Caldetes thermal springs. 3. Modeling shows, that such a flow system for the given permeabilities is physically possible, and indicates that residence times of the Caldetes waters should be very long, in the order of tens of thousands of years. This hypothesis, however, still needs to be verified. Geofluids III. F5 Field Guide. 15 - 16 July 2000 55 4. REFERENCES Albert, J. F., 1976. Estudio geotérmico preliminar de Cataluña. Ph.D. Thesis, Barcelona Univ. Álvarez-de-Buergo, E., Meléndez, F., 1994. Características generales de las subcuencas del margen peninsular mediterráneo (“Rift” del Surco de Valencia). Acta Geol. Hisp., 29, 67-79.Agustí 1985. Bakalowicz, M.J., Ford, D.C., Miller, T.E., Palmer, A.N., Palmer, M.V., 1987. Thermal genesis of dissolutional caves in the Black Hills, South Dakota. Geol. Soc. Am. Bull., 99: 729-738. Bartrina, M.T., Cabrera, L., Jurado, M.J., Guimerà, J., Roca, E., 1992. Evolution of the central Catalan margin of the Valencia trough (Western Mediterranean). Tectonophysics, 203: 219-247. Bitzer, K., 1999. Two-dimensional simulation of clastic and carbonate sedimentation, consolidation, subsidence, fluid flow, heat flow and solute transport during the formation of sedimentary basins. Computers & Geosciences, 25:431-447. Cabrera, L., & Calvet, F., 1996. Onshore Neogene record in NE Spain: Vallès-Penedès and El Camp half-grabens (NW Mediterranean). In: P.T. Friend and C. Dabrio (Editors), Tertiary Basins of Spain pp. 97-105. Calvet, F., Travé, A., Roca, E., Soler, A. & Labaume, P., 1996. Fracturación y migración de fluidos durante la evolución tectónica neógena en el Sector Central de las Cadenas Costero Catalanas. Geogaceta 20: 1715-1718. Calvet, F., Travé, A., Bitzer, K., Roca, E. & Baker, J., in progress, Dolomitization of Detrital Deposits Related to Neogene Extensional Faults, Catalan Coastal Ranges (Spain). Canals, A. Albert, J. & Ayora, C., 1990. El sistema geotérmico de La Garriga-Samalús: comparaciones con sistemas hidrotermales fósiles. Geogaceta, 7: 88-90. Canals, A.; Cardellach, E.; Rye, D.M. & Ayora, C., 1992. Origin of the Atrevida vein (Catalonian Coastal Ranges, Spain): Calvet, F. et al. 56 mineralogic, fluid inclusion and stable isotope study. Econ. Geol., 87:142-153. Canals, A. & Cardellach, E., 1993. Strontium and sulphur isotope geochemistry of low-temperature barite-fluorite veins of the Catalonian Coastal Ranges (NE Spain): a fluid mixing model and age constraints. Chem. Geol., 104; 269-280. Canals, A. & Cardellach, E., 1996. Caracterización de los fluidos en sucesivas fases de fracturación en las cadenas Costero Catalanas y su registro en los filones de Ba-F de baja temperatura. Geogaceta, 20: 696-699. Canals, A. & Cardellach, E., 1997. Ore lead and sulphur isotope pattern from the low temperature veins of the Catalonian Coastal Ranges (NE Spain). Min. Dep., 32: 243-249. Cardellach, E.; Canals, A. & Tritlla, J., 1990. Late and postHercynian low temperature veins in the Catalonian Coastal Ranges. Acta Geológica Hispánica, 25: 75-81. Fernández, M. & Banda, E., 1988. Aproximación a la anomalia geotérmica de La Garriga - Samalús (Vallès-Penedès). Acta Geológica Hispánica, 23: 1-20. Fontboté, J.M.; Guimerá, J.; Roca, E.; Sàbat, F. & Santanach, P., 1989. Para una interpretación cinemática de la génesis de la cuenca catalano-balear: datos estructurales de sus márgenes emergidos. Libro Homenaje a R. Soler, AEGGP, Madrid, pp.37-51. Fontboté, J.M., Guimerà, J., Roca, E., Sàbat, F., Santanach, P. & Fernández-Ortigosa, F., 1990. The Cenozoic Geodynamic Evolution of the Valencia Trough (Western Mediterranean). Revista de la Sociedad Geologica de España, 3: 249-259. Galindo, C.; Tornos, F.; Darbyshire, D.P.F. & Casquet, C., 1994. The age and origin of the barite-fluorite (Pb-Zn) veins of the Sierra del Guadarrama (Spanish Central System, Spain): a radiogenic (Nd, Sr) and stable isotope study. Chem. Geol., 112: 351-364. Gil Ibarguchi, J.I., Julivert, M., 1988. Petrología de la aureola metamórfica de la granodiorita de Barcelona en la Sierra de Collcerola (Tibidabo). Estudios Geol., 44, 353-374.Guimerà, J., Geofluids III. F5 Field Guide. 15 - 16 July 2000 57 1984. Palaeogene evolution of deformation in the northeastern Iberian Peninsula. Geological Magazine, 121: 413-420. Jébrak, M., 1984. Contribution à l'histoire naturelle des filos (F-Ba) du domaine varisque français-marriquain. Essai de caractérisation structurale et géochimie des filons en extension et en décrochement. Doc. Bur. Rech. Géol. Min., no. 99; 510pp. Julivert, M., Durán, H., 1990. Paleozoic stratigraphy of the Central and Northern part of the Catalonian Coastal Ranges (NE Spain). Acta Geol. Hisp., 25, 3-12.Middlenton, K., Coniglio, M., Sherlock, R. & Frape, S.K., 1993. Dolomitization of Middle Ordovician carbonate reservoirs, southwestern Ontario. Bulletin of Canadian Petroleum Geology, 41: 150-163. Roca, E., 1994. La evolución geodinámica de la Cuenca CatalanoBalear y áreas adyacentes desde el Mesozoico hasta la actualidad. Acta Geológica Hispánica, 29: 3-25. Roca, E. & Guimerà, J., 1992. The Neogene structure of the eastern Iberian margin: structural constraints on the crustal evolution of the Valencia trough (western Mediterranean). Tectonophysics, 203: 203-218. Roig, A. & Canals, A., 1994. Caracterización de las condiciones de formación de Mina Berta (F-Pb-Zn, Cadenas Costeras catalanas): datos mineralógicos y de inclusiones fluidas. Bol. Soc. Esp. Min., 17: 194-195. Salas, R. & Casas, A., 1993. Mesozoic extensional tectonics, stratigraphy and crustal evolution during the Alpine cycle of the eastern Iberian basin. Tectonophysics, 228: 33-55. Solé, J; Delaloye, M. & Enrique, P., 1998. K-Ar ages in biotites and K-feldspars from the Catalan Coastal Batholith: evidence of a postHercynian overprinting. Eclogae Geol. Hel., 91: 139-148. Taylor, T.R. & Sibley, D.F., 1986.Petrographic and geochemical characteristics of dolomite types and the origin of ferroan dolomite in the Trenton Formation, Ordovician, Michigan Basin, U.S.A. Sedimentology, 33: 61-86.