Journal of Conference Abstracts

Session J11:3B

J11 : 3B/25 : F2

Origin and Evolution of Groundwater in the Eastern Batinah Coast of Oman ­ A Hydrogeochemical Approach Based on Isotopic Tracers

Constanze, E. Weyhenmeyer (weyhen@geo.unibe.ch)¹,

H. Niklaus Waber (nick@mpi.unibe.ch)²,

Stephen, J. Burns (burns@geo.unibe.ch)¹,

Jan Kramers (kramers@mpi.unibe.ch)² &

Albert Matter (amatter@geo.unibe.ch)¹

¹ Geological Institute, University of Bern, Baltzerstrasse 1, CH-3012 Bern, Switzerland

² Institute of Mineralogy and Petrology, University of Bern, Baltzerstrasse 1, CH-3012 Bern, Switzerland

The alluvial aquifer beneath the Batinah Coastal Plain of Oman is one of the most important aquifers in Oman since it supplies water for the most densely populated, cultivated and industrialized areas in the Sultanate. In recent years, however, overexploitation of this groundwater resource has resulted in seawater intrusion into the coastal alluvial aquifer. In an attempt to derive a water budget and to develop a sustainable water management strategy, this study investigates the sources of recharge and the subsequent groundwater evolution along various flowpaths using a hydrogeochemical approach including isotopic data of O, H, C, Sr and Ar. Oxygen isotope values of over 200 wells distributed over the 10 000 km² study area indicate that the main source of recharge to the coastal plain is high altitude rainfall from the neighboring Oman Mountains. This isotopically depleted water forms two relatively narrow 'plumes' that stretch across the coastal plain to the coast thereby maintaining their homogeneity vertically and horizontally. Strontium isotope ratios of water within these two plumes are as high as 0.7115 indicating that the groundwater has interacted with the pre-Cambrian rocks of the Oman Mountains. In contrast, water samples from areas adjacent to the two plumes are characterized by a limestone-dominated strontium isotope signal of around 0.7085. Surprisingly, very little influence of oceanic basement rocks, the Samail Ophiolite, is detectable in the strontium isotope ratios, even in catchments composed entirely of ophiolite. This suggests that groundwater is predominately flowing through calcite and magnesite lined fractures thereby developing the more typical carbonate strontium isotope signal of 0.7080 to 0.7085. The assumption of fracture flow is consistent with elevated tritium values found in all bedrock wells, even at depths > 100 m. Elevated tritium values were also found in wells and springs at the foothills of the Oman Mountains suggesting higher flowrates through the karstified, pre-Cretaceous, autochthonous limestones of the Oman Mountains than previously assumed. Tritium values fall below detection limits a few tens of km's further out onto the plain, suggesting much slower water movement through the thick alluvial gravels, possibly due to cementation and porosity reduction by rapid diagenesis of ophiolitic clasts.

J11 : 3B/26 : F2

Identification of Subsurface Fluxes into the North-Western Kalahari Basin Using Environmental Tracers

Christoph Kuells

(christoph.kuells@mail.uni-wuerzburg.de)¹,

Eilon Adar (eilon@bgumail.bgu.ac.il)²,

Mebus Geyh (geyh@bgr.de)³ &

Peter Udluft (udluft@mail.uni-wuerzburg.de)¹

¹ Institute of Geology, Pleicherwall 1, 97070 Wuerzburg, Germany

² Inst. for Desert Research, Ben Gurion University, 84990 Sede Boker Campus, Israel

³ Nieders. Landesamt f. Bodenforschung, 14-C Labor, Stilleweg 2, 30655 Hannover, Germany

The extent of recent recharge into the Kalahari basin has been a subject of lively debate since the discovery of tritium traces in its phreatic aquifers. While most of the discussion has centred on local rainfall recharge at an experimental scale, few contributions have addressed the regional distribution of recharge areas as well as the identification of lateral subsurface fluxes into and within the Kalahari aquifers itself. Within the context of water resources development in the north-eastern Namibian Kalahari potential sources of recharge and the sub-surface flow pattern are examined with environmental tracers such as hydrochemistry and isotopes.

This study combines hydrochemical process with a mixing cell modelling for elaborating on the flow system, end members (potential recharge sources), and their quantitative interaction in an otherwise poorly defined aquifer system. On the grounds of an extensive hydrochemical database including major ions, trace elements (Si, Sr, Br, F) and stable isotopes ( ¹⁸O,  ²H,  ¹³C) hydrochemical process and groundwater regions are defined. These include dominant dissolution/precipitation processes and modifying exchange, redox or mixing processes. Additional information on the recharge environment and groundwater evolution is taken into account using stable isotopes. The regional extent as well as the genetic internal relationship between each of the hydrochemical process groups are statistically defined by hierarchical cluster analysis. Using this information and the physical constraints of the flow system a 3-dimensional conceptual model of internal links has been developed. The host rocks of the main recharge areas and their recharge characteristics (calcitic, dolomitic, quarzitic) produce a well-defined hydrochemical and isotopic fingerprint, hence, the major flow paths and convergence zones can be derived. An important recharge mechanism was identified involving infiltration from ephemeral sheet flows into a re-dissolving calcrete cover along the Kalahari fringe. Results also indicate a general pattern of a stratified aquifer system: Inflows into the Kalahari derived from preferential recharge regions mix with deeper sources including possible gaseous components along the Waterberg thrust fault. Finally, for a quantitative assessment of the groundwater flow system, a set of the linear mass balance equations were derived based on the hydrochemical and isotopic database.

J11 : 3B/27 : F2

Evaluation of Groundwater Circulation and Volcanism Influences in the Astian Sandy Aquifer (Mediterranean Coast, South of France)

Florent Barbecot (barbecot@geol.u-psud.fr)¹,

Yves Travi (yves.travi@univ-avignon.fr)²,

Christelle Marlin (marlin@geol.u-psud.fr)¹ &

Laurent Dever (dever@geol.u-psud.fr)¹

¹ Laboratoire d'Hydrologie et de Géochimie Isotopique, Université Paris-Sud, Bât. 504, 91405 Orsay Cedex, France

² Laboratoire d'Hydrologie, Université d'Avignon, 83 rue Pasteur, 84000 Avignon Cedex, France

In South coastal France, the Astian confined aquifer, made of fine grained micaceous sands, is under severe pressure due to tourism. This aquifer is one of the main source for fresh water supply in the Agde area even if salty water have been recognized in its eastern part.

In order to develop the management of this aquifer, assessment of water circulation and aquifer recharge may be approached by geochemical studies. Moreover, the purpose of this study is to identify the origin of salty water which can be related to eather a modern sea intrusion due to pumping or a palaeo-circulation within the aquifer. Twenty two boreholes have thus been sampled for chemical (major elements, F, Br) and isotopic (water  ²H,  ¹⁸O and dissolved carbon A¹⁴C and  ¹³C) analyses.

From the first results, two main geochemical domains separated by the Herault river can be identified. On the Western part of the aquifer, groundwaters are of Ca-HCO₃ type and are characterized by low mineralizations (less than 0.7 g.L^-1) with ¹⁴C residence time increasing from modern ages on the recharge area (Northern part of the aquifer) to 6 ka. B.P. nearby the coastline. On the opposite, the Eastern part of the aquifer is characterized by highly mineralizated water (up to 10 g.L^-1) of Na-Cl type and are associated with anomalous dissolved carbonate contents (up to 1.9 g.L^-1 of HCO₃), low A¹⁴C and high  ¹³C. Moreover, all the area has been affected by a distensional tectonic stage during the middle Quaternary allowing the formation of a sub-meridian volcanoes train on the right bank of the Herault river. Consequently, the Astian aquifer is affected by faults and crossed by basaltic dykes that could constitute ways for deep water and/or CO₂ upward leakage. This is confirmed by CO₂-sparkling water founded in the most salty area of the aquifer on the South-East of the Thau pond. The geochemistry of C-enriched groundwater lead to define a salty and deep end member characterized by  ¹³C_CO2 and A¹⁴C of - 4.4 per mil vs PDB and 0 pMC respectively, in agreement with the endogenous carbon domain (-10 to -2 per mil vs PDB) known for the peri-Alpine arc. Consequently, after correction of the C-mixing, the evolution of ¹⁴C ages,  ¹³C contents and chemical composition allow us to propose a conceptual model for this complex system of water and carbon mixings. The C-mixing comes from carbonate dissolution due to deep CO₂ diffusion and water mixing comes from the intrusion of old salty groundwater from the underlying aquifer(s).

J11 : 3B/28 : F2

Boron and Strontium Isotopes in Groundwaters and Brines of the North German Basin (Gorleben Diapir)

Wolfram Kloppmann (w.kloppmann@brgm.fr)¹,

Philippe Negrel (p.negrel@brgm.fr)¹,

Joël Casanova (j.casanova@brgm.fr)¹,

Catherine Guerrot (c.guerrot@brgm.fr)¹ &

Hans Klinge (h.klinge@bgr.de)²

¹ BRGM 3, avenue C. Guillemin - BP 6009, F-45060 Orléans cedex2, France, France

² BGR Stilleweg 2, D-30655 Hannover, Germany, Germany

Sr- and B-isotope ratios of groundwaters from sediments overlying the Gorleben diapir in the North German Basin and from its marginal deeps have been measured together with those of potential mixing endmembers, Permian salts and surface water. The complex salinity distribution with depth reflects the geological structure in the surroundings of the diapir with highly heterogeneous Quaternary glacial deposits overlying Tertiary sediments disturbed by salt tectonics. Low Br/Cl and B/Cl ratios plead for secondary brines from salt weathering rather than for dilution of residual brines as origin of salinity. Groundwaters with salinities <10 g/l show the highest ⁸⁷Sr/⁸⁶Sr ratios (mean value of 0.7095). Isotopic composition of these waters gives evidence of water-rock interactions, in particular cation exchange with silicates. Deep waters with salinities up to 330 g/l show a clear tendency towards lower Sr- isotope ratios the lowest values being close to those of the Upper Permian ocean (0.7071 - 0.7076; Martin & McDougall, 1995) and to the ⁸⁷Sr/⁸⁶Sr ratios of salt samples from the Zechstein diapir. This trend reflects mixing with brines derived from the diapir.  ¹¹B values of near surface waters are situated in the range of 7 to 15‰. Deep NaCl groundwaters are enriched in ¹¹B up to values of 33‰. Combining the highly fractionating B-isotopes with the non fractionating Sr-isotopes gives additional information on the nature of brines. The potential high salinity mixing endmember has  ¹¹B values around 40‰. Salt dissolution therefore seems to be the predominating process, provided that the B-isotope ratios of the Zechstein ocean were similar to those of modern sea water. Primary brines evaporated beyond halite saturation would in fact be enriched in ¹¹B with respect to seawater whereas associated salts are relatively depleted (Vengosh et al., 1992). Saltwater lenses in the upper parts of the Quaternary deposits lie on the same isotopic mixing lines as deep brines suggesting a common origin.

Martin EE & McDougall JD, Chem. Geol, 125, 73-99, (1995).

Vengosh A, Strarinsky A, Kolodny Y, Chivas AR & Raab M, Geology, 20, 799-802, (1992).

J11 : 3B/29 : F2

Origin of the Elements and Control of the Chemical Composition of Geothermal Waters in Sedimentary Environment; 'The Bagnères-de-Bigorre' (Central Pyrenees) Example

Stéphanie Levet (levet@lucid.ups-tlse.fr)¹,

Jean-Paul Toutain (toutain@lucid.ups-tlse.fr)¹,

Philippe Negrel (negrel@exchange.brgm.fr)²,

Margot Munoz (munoz@lucid.ups-tlse.fr)¹ &

Franck Poitrasson (Franck.Poitrasson@cict.fr)¹

¹ OMP, UMR5563, 38 rue des 36 ponts, 31400 TOULOUSE, France

² BRGM, direction de la recherche, 45060 ORLEANS Cedex, France

Origin of the elements and control of the chemical composition of geothermal waters in sedimentary environment; 'The Bagnères-de-Bigorre' (central Pyrenees, France) example.

Few geochemical studies have been performed on geothermal waters located in sedimentary reservoirs relative to those found in granitic environments. In the west pyrenean station 'Bagnères-de-Bigorre' (France), located along a major active normal fault (the Bigorre-Adour fault), 11 springwaters rising up from a Mesozoic sedimentary cover in the North Pyrenean Zone at a temperature between 20 and 50°C were collected for major and trace elements, and strontium isotope analysis. As shown by stable isotopes  ¹⁸O and  D, the fluids have a meteoric origin. The chemistry of the waters evidences enrichments in (SO₄-Ca-Cl) and (CO₃-Mg) indicating interactions with Triassic evaporites and Mesozoic carbonates. Li-Cl and Na-Cl plots show the conservative behavior of these elements along the flowpath unlike that of an igneous environment. This indicates that they are not controlled by the composition of the sedimentary reservoir. In such sedimentary reservoir, the applicability of calibrated geothermometers commonly used in waters equilibrated with volcanic rocks is of limited use. However, the good agreement between the silica and Na-Li geothermometers suggests the occurrence of two groups of waters with equilibrium temperatures of 80°C and 50°C respectively. Moreover, solution speciation modelling performed from these temperatures down to the surface temperature indicates that these waters are saturated with gypsum and anhydrite, supersaturated with calcite and dolomite and undersaturated with K-feldspar and plagioclase. The latter indicates a lack of control of the potassium and sodium concentrations by granitic-rock dissolution. The Sr isotopes (⁸⁷Sr/⁸⁶Sr) and trace element ratios (Rb/K versus Cs/K) clearly confirm the absence of water-rock interaction with the granitic basement. Then, the use of geochemical tracers (Ca/Cl, SO₄/Cl ratios) suggests the existence of two distinct flowpaths, one deeper than the other. The respective fluids chemical characteristics are mainly dependent on host-rock composition and dissolution rate, which is relative to temperature, pH and water/rock ratios. These fluids circulate at a depth up to 2 km and return to the surface undergoing cooling and dilution as they mix with superficial waters equilibrated with karstic aquifer.

J11 : 3B/30 : F2

Chemical and Isotopic Investigation of Cold and Warm Water Inflows in the Gotthard Tunnels

Sabrina Pastorelle (Sabrina.Pastorelli@imp.unil.ch)¹,

Johannes Hunziker (Johannes.Hunziker@imp.unil.ch)¹ &

Marini Luigi²

¹ Institut de minéralogie et pétrographie, Université de Lausanne, BFSH-2 CH-1015, Switzerland

² Dipartimento di Scienze della Terra, Università di Genova, Corso Europa 26 I-16123, Italy

Groundwaters detected crossing mountain chains by deep tunnels provide valuable information about subsurface flow conditions and rock-water heat transfer. The warm water inflows in the Gotthard tunnels also represent important low-enthalpy geothermal resources, due to their remarkable flow rates. This study is aimed at elucidating the origin of these groundwaters and the processes governing their chemical evolution, within the geological-hydrogeological framework of the Gotthard area. To achieve these objectives, thirty-five water samples were collected and analysed chemically (major and some trace elements) and isotopically  ¹⁸O and  ³⁴S): 21 from the Gotthard Highway Tunnel (GHT) and 14 from the Gotthard Exploration Tunnel of the new Swiss railway project (Alp-Transit) through the Alps.

Geologically the GHT is located within the Aar and Gotthard basement massifs represented mainly of late Variscan granites and Permocarboniferous volcanoclastic sequences with tight synclines of Mesozoic cover composed by evaporates and by dolomitic marbles. Sampled waters belong to different hydrogeochemical types (Ca-HCO₃, Ca-SO₄, Na-HCO₃ and Na-SO₄) reflecting both the lithologies where the hydrogeological circuits are located and the time of interaction of these waters with the surrounding rocks.  ³⁴S values are consistent with the occurrence of two processes that control the origin of dissolved sulphate: the leaching of anhydrites of the Upper Triassic age or the leaching of sulphide minerals contained in crystalline rocks and oxidation to sulphate. Some samples have  ³⁴S values heavier than those consistent with the leaching of Triassic anhydrites and show a negative correlation with the sulphate contents. These samples are probably affected bacterial reduction of SO₄^-2 that generate H₂S and produce SO₄^-2 enriched in ³⁴S. Intermediate isotopic values are originated through different rates of mixing of the three end-members.

A simple equilibrium model serves to show the evolution of the chemical composition of an aqueous solution interacting with gneissic rocks: the generation of a Na-HCO₃ water sample is simulated by the interaction, at 25°C, of a Ca-HCO₃ water with a rock that has the average composition of Gotthard Massif granite. Eq3-6 computer code (Wolery, 1983) allow us to carry out this exercise. In this model the granite is dissolved in two steps with different P-CO₂: the first step with P-CO₂ values of 10^-2.76 bars (initial value for shallow groundwater and the second step with P-CO₂ values of 10^-4 bars which is the P-CO₂ of the Na-HCO₃ water. Temperature constraints are given by quartz equilibrium temperature. Calculated results fit quite well with experimental data and reaction-path from the Ca-HCO₃ sample to the Na-HCO₃ is shown in activity-activity diagrams considering stability relations between (a) kaolinite, muscovite, albite and K-feldspar and (b) calcite and dolomite.

J11 : 3B/33 : F2

Relation of Saline Springs to the Regional Tectonic Framework (Southern France)

Luc Aquilina (l.aquilina@brgm.fr)¹,

Bernard Ladouche (b.ladouche@brgm.fr)¹,

Michel Kakalowicz (baka@dstu.univ-montp2.fr)²,

Nathalie Doerfliger (n.doerfliger@brgm.fr)¹ &

Paul Le Strat (p.lestrat@brgm.fr)¹

¹ BRGM, Direction de la Recherche, 1039 rue de Pinville, 34000 Montpellier, France

² CNRRelation of saline springs to the regional tectonic framework (Southern France)S, Géofluides Bassin Eau, 4 place E. Bataillon, 34000 Montpellier, France

In the South of France, several karstic systems have been investigated from a hydrologic and hydrochemical point of view (Ladouche et al., 1999; Schoen et al., 1999; Pinault et al., 1999). These systems not only show common karstic springs with low salinity, but also saline springs which range from 1 g/L to almost 10 g/L. A thorough chemical and isotopic investigation of theese was carried out, which include: major and trace elements, ¹⁸O, ²H, ³H, ⁸⁶Sr, ¹³C, ¹⁴C and for some of them ³⁶Cl isotopic ratios. For the different systems, the results are as follow:

- Balaruc: Fluids are sampled in drilling used for thermal purposes. Cl-Br and SO₄, and ¹⁸O-D relation ships show that salinity originates from seawater. Ca, Sr and Sr isotopic ratios indicate intensive dissolution of carbonates. Dissolution is related to the influx of deep CO₂. Tritium and ³⁶Cl indicate long residence time and the presence of paleo-meteoric waters in the thermal reservoir.

- Robine de Vic: The chemical composition of this saline spring is relatively similar to the previous one

- Fonstestramar: Two saline springs are known in this area. Water stable isotopes indicate a mixing with seawater. Cl/Br ratio however differs from seawater, as does the Cl-¹⁸O correlation. These elements show that the salinity originates from a seawater end-member which has undergone dissolution processes. No indication of sulphated Triassic influence is observed from SO₄ concentrations.

The geochemistry of the saline springs indicate that the karstic system investigated have a deep part buried below the base level of the mediterranean sea. In the case of the Balaruc system, the outcopring limestones and the thermal reservoir are distinct and related by a fault. This behavior is related to the tectonic evolution of the area. Indeed, all these systems are related to deep faults. These faults belong to the same tectonic structure which is constituted of two branches which merge in the Mediterranean sea.

Ladouche et al., J. Conf. Abs., 4, (1999)

Pinault et al., J. Conf. Abs., 4, (1999)

Schoen et al., J. Conf. Abs., 4, (1999)

J11 : 3B/34 : F2

Evidence for Self-Organization During Reactive Fluid Flow in a Porous Medium

Francois Renard (frenard@obs.ujf-grenoble.fr)¹ &

Jean Pierre Gratier (gratier@obs.ujf-grenoble.fr)²

¹ University of Oslo, Institute of Geology, postboks 1047 Blindern, Norway

² LGIT, CNRS-Observatoire, BP 53, 38041 Grenoble, France

When a reactive fluid circulates inside a porous medium, dissolution can occur, resulting in porosity and permeability modifications of the rock (Ortoleva, 1994). The positive feedback between fluid flow and mineral dissolution may destabilise an initially planar dissolution front and lead to more complex morphologies such as karsts. This is due to the self-focusing of the fluid inside the regions where permeability is increased because of local dissolution. Our study is carried out with two objectives: 1) Evaluate experimentally this process at a decimetre scale, 2) Compare the experiment to a numerical model of water-rock interaction. The experiment consists of an analogous two-dimensional porous medium that allows for the dissolution of halite under an imposed fluid flow. The numerical code used solves the equations of reaction and transport in a porous medium, represented as a two-dimensional grid, and hence predicts the temporal evolution of the reaction front. Both experiment and numerical simulation indicate the development of a dissolution instability (Renard et al., 1998).

This reaction-transport instability shows that if a rock is homogeneous at a considered scale, self-organization patterns can emerge, and large initial non-uniformities (due to tectonic events or sedimentary features for example) are not required to create porosity variations and localised deformation in the upper crust.

Ortoleva, Geochemical Self-Organization, Oxford University Press, (1994).

Renard F, Gratier JP & Ortoleva P, Geophysical Research Letters, 25, 385-388, (1998).

J11 : 3B/35 : F2

Simulation of Permeability Changes in Aquifers Caused by Chemical Reactions

Michael Kuehn (m.kuehn@tu-harburg.de)¹,

Joern Bartels (joern.bartels@gga-hannover.de)²,

Hansgeorg Pape³,

Wilfried Schneider (w.schneider@tu-harburg.de)¹ &

Christoph Clauser (c.clauser@gga-hannover.de)²

¹ TU Hamburg-Harburg, AB 1-09, Dampfschiffsweg 11, D-21079 Hamburg, Germany

² Geowissenschaftliche Gemeinschaftsaufgaben (GGA), Postfach 51 01 53, D-30631 Hannover, Germany

³ Geodynamik, Rheinische Friedrich-Wilhelms-Universitaet, Nussallee 8, D-53115 Bonn, Germany

A new simulation model for coupled flow, heat transport, multi-component transport, and chemical reactions is used to quantify the effect on flow and transport of chemically induced changes in the pore space structure of deep sandstone aquifers. The complex heterogeneous chemistry of these water-rock interactions depends on temperature, salinitiy, and pressure.

Therefore, a chemical module was developed to describe precipitation and dissolution of minerals at high temperatures and high ionic strength. The numerical module is based on the geochemical simulation program PHRQPITZ (Plummer et al., 1988) incorporating Pitzer's equations for calculation of geochemical reactions in brines. The thermodynamic data base was updated and the code expanded. Now the model comprises the simulation of mineral reactions in brines for a temperature range from 0-90°C in the system Na-K-Ca-Mg-H-Cl-SO₄-OH-HCO₃-CO₃-H₂O. The calculations are based on temperature-dependent Pitzer coefficients and the modelling results were verified with laboratory data. Both, coefficients and experimental results are available in the literature.

A new, petrophysically well founded permeability-porosity-relationship was developed and implemented. It is based on the KOZENY-CARMAN equation and a fractal model of the pore space. Permeability is expressed as a power series of porosity. The exponents of this series are calculated from the fractal dimension of the fluid-matrix interface. This expression was calibrated and validated with large petrophysical data sets from sandy to shaly sedimentary rocks.

For studying the effects of coupling of all processes extensively, the chemical reaction module and the permeability-porosity relation were implemented in an existing numerical code, simulating fluid flow and heat and multi species transport.

It was applied to investigate the relationship between mineral reactions and permeability changes in porous aquifers. Results of three-dimensional coupled simulations of a geothermal doublet installation are presented, focussing on the complex water-rock-interactions at the propagating temperature front and the vicinity of the injection well. It is demonstrated how the precipitation-dissolution processes vary in time and space during an assumed period of 30-100 years of geothermal use. Consequences for a geothermal doublet installation and its production performance are discussed.

Plummer LN, Parkhurst DL, Fleming GW & Duke SA, U.S. Geological Survey, Report, Reston, 88-4153, 306, (1988).

J11 : 3B/36 : F2

Dynamic Experimental Study of Rock-Water Interactions on the Soultz Granite at 200°C

Valerie Plagnes (plagnes@nire.go.jp)¹,

Isao Matsunaga (matsunaga@nire.go.jp)¹,

Hiroaki Tao (hiroaki@nire.go.jp)¹,

Koichiro Fujimoto (fujimoto@gsj.go.jp)²,

Mohamed Azaroual³ &

Luc Aquilina⁴

¹ National Institut for Resources and Environment, 16-3 Onogawa, Tsukuba, Ibaraki 305-8569, Japan

² Geological Survey of Japan, Higashi 1-1-3, Tsukuba, Ibaraki 305, Japan

³ BRGM, Direction de la Recherche, 3 av. Cl. Guillemin, BP 6002, F-45060 Orleans, France

⁴ BRGM, Direction de la Recherche, 1039 rue Pinville, F-34000 Montpellier, France

Several experiments of rock-water interactions have been conducted on granite from Soultz (NE of France), candidate site of the future European Hot Dry Rock geothermal exchanger, in order to investigate chemical reactions over 3000 m of depth. Experiments were performed on both whole rock and separated minerals (Biotite, K-Feldspar, Plagioclase) at 200ºC during 11 days. Two different approaches were used to investigate the chemical reactions of dissolution and precipitation that occur inside the column. 1) First, chemical analysis of the outlet fluid has been made (electric conductivity, pH, major and trace elements). Even is Si is the main component of the outlet fluid, his content is clearly weak (30-50 ppm) in front of the content which should be reached during a stationary state at 200ºC (~100 ppm). After 11 days, the chemical reactions seem to be still far from the chemical equilibrium. 2) Second, rocks and minerals were observed with Scanning Electronic Microscope after experiment in order to evaluate the morphological changes. K-Feldspar show the greatest degree of reaction with many etchpits developed along cleavages planes of the mineral and many precipitates spread at the surface. The same dissolution and precipitation features exist also on the surfaces of plagioclase minerals, but the alteration seems to have been less active. The dissolution is more tenuous on the biotite and well cristallographically controled on the quartz surfaces. The morphologic study of the minerals from separated minerals experiments and whole rock experiments didn't show so big differences. Intensity of dissolution is quite comparable and the precipitates are chemically the same (mainly phyllosilicates). The mixing in the whole rock doesn't seem to allow different kinds of precipitation. To obtain the equilibrated state of the chemical evolution and to improve the mineralogical characterization, the next experiments will have a longer duration: 21 days instead of 11 days. As a matter of fact, the dissolution features will be more advanced and the secondary minerals precipitated will develop in larger sizes.

J11 : 3B/37 : F2

Geochemical Modelling of Water-Rock Interactions between Crystalline Basement and Bassin Fluids Applied to the Geothermal Hot Dry Rock (H.D.R.) Site of Soultz-sous-Forêts (Alsace, France)

Emmanuel Jacquot (jacquot@illite.u-strasbg.fr)¹,

Bertrand Fritz (bfritz@illite.u-strasbg.fr)¹ &

Jacques Leroy (leroy@g2r.u-nancy.fr)²

¹ Centre de Géochimie de la Surface, UMR 7517 ULP-CNRS, 1, Rue Blessig, France

² Laboratoire de Géologie et de Gestion des Ressources Minérales et Energétiques. UMR 7566, BP 239, 54506 Vandoeuvre-lès-Nancy Cedex, FRANCE

The present study deals with the geochemical modelling of the geothermal Hot Dry Rock site located at Soultz-sous-Forets (Alsace, France). On this site, numerous studies were performed over the past decade. Hence parameters required for modelling can be well constrained.

Two boreholes were drilled down to about 3500 meters depth, within the granitic basement at a temperature of about 150 to 170°C. At the production well, a hot intragranitic formation fluid is extracted. It is then cooled to a surface heat exchanger, that simulates the use of the thermal power carried by the fluid. The cold fluid is re-injected at the injection well and then stored or circulating within the granite stimulated fractures network.

The driving force for the geochemical evolution of such a fluid-rock system is the thermodynamic desequilibrium induced by the imposed fluid temperature variations. We are to understand how the cold re-injected fluid behave within the hot rock. This requires first to characterise the water-rock system, that is to say the rock mineralogy and the fluid composition.

The geological features of this site are related to the Rhine graben. Numerous N-S orientated fractures allowed fluids circulations over long time periods, far before any perturbation induced by the geothermal plant itself. Hence the granite mineralogy is already strongly modified at the neighbouring of the most intensively fractured zones, where present fluid circulations take place. Beside quartz and K-feldspar from the original granite, clay minerals are abundant at the contact of the circulating fluid. Their possible swelling due to the injection of a cold fluid, as well as the occurrence of secondary mineral phases have to be studied if we are to foresee the evolution of the deep reservoir transport properties.

The initial fluid composition was constrained from previous chemical analysis and computations, assuming initial thermodynamic equilibrium between the previously defined rock and the formation fluid at 165°C. The fluid cooling to the surface heat exchanger was then simulated down to 65°C, the re-injection temperature. If the cooling occurred under thermodynamic control in the thermal heat exchanger, a total precipitated mass of approximately 8 kg per hour should have been observed, which was not the case during a four month circulation test performed during the summer 1997. Actually, the fluid cooling is very rapid and it occurs under kinetic control. The precipitation rates are slow when compared to the rate of the fluid temperature decrease, hence oversaturations with respect to secondary mineral phases occur.

The problem is then to know if these oversaturations are large enough for minerals homogeneous nucleation from the cold fluid to occur, and if the rock mineral surfaces at the contact of which the fluid is circulating behave as secondary minerals growth sites (heterogeneous nucleation).

J11 : 3B/38 : F2

Mass Balance Versus Equilibrium Control of Groundwater Composition in Crystalline Basement

Kurt Bucher (bucher@uni-freiburg.de)¹ &

Ingrid Stober (stober@uni-freiburg.de)²

¹ Institute of Mineralogy, Petrology and Geochemistry, University of Freiburg, Albertstr. 23b, 79104 Freiburg

² Geologisches Landesamt, Baden-Württemberg, Albertstr. 5, 79104 Freiburg, Germany

Water stored in the interconnected fracture system of the upper continental crust beneath the near-surface zone is in general characterized by high mineralizations (TDS) and by relative abundance of the major species Na, Ca >> Mg, K and Cl >> SO₄ > HCO₃, concentrations of all other species are typically low and the waters are remarkably oxidized. High-TDS Na-Ca-Cl waters (brines) have been reported from the continental basement world-wide including deep research wells and mine waters from the Canadian, the Scandinavian, Ukrainian shields and from the younger crust of the British Isles and Germany and many other sites. Reported groundwater compositions from crystalline basement indicate that dissolution of the major components of basement rocks, feldspar and mica essentially dominant the chemical evolution of the waters and that the reaction with the granitoid rock matrix is the major source of cations. The deep waters are, in general, saturated with respect to quartz. The relationships of water compositions depicted on stability diagrams (activity-activity diagrams) for relevant low-temperature phases at quartz saturation are consistent with equilibrium with kaolinite. Activities of Na, Ca, K, and Mg of basement waters indicate that pH-normalized evolution paths are controlled by mass balance and bulk dissolution rather than by equilibrium with precipitating product phases. High-TDS basement brines remain normally in the kaolinite field because of low pH. High pH-waters, indicating silicate dissolution in the absence of CO₂ and sulphide-oxydation, are generally rare and very few are consistent with equilibrium with phases other than kaolinite. If product phases such as chlorite, smectite, montmorillonite and zeolites are formed, they grow from strongly oversaturated fluids by non-equilibrium processes. In general, the product-phases cannot buffer the overall composition of water in the basement. In addition, CaCl₂-rich deep brines act as efficient filters for CO₂. It seems unlikely that CO₂-rich waters common in the uppermost crust receives CO₂ from deep sources (lower crust, mantle) because the gas cannot move through high-density CaCl₂ brines trapped at depth. All deep waters in the Precambrian shields are extremely low in HCO₃.

Session J11:4A

J11 : 4A/01 : F2

Thermodynamic Modelling of the Partitioning of Bromine During the Evaporation of Seawater

Michael Schramm

(schramm@immr.tu-clausthal.de) &

Michael G Siemann

(siemann@immr.tu-clausthal.de)

Fachgebiet Geochemie,Technische Universität Clausthal, A.-Roemer-Str. 2A, 38678 Clausthal-Zellerfeld

Trace elements in brines and marine evaporites are often used as genetic indicators for metamorphic processes. To calculate complex mineral-fluid reactions like the evaporation of seawater, where the chemical composition of the brine is evolving rapidly, a large set of distribution coefficient data is necessary but currently not available. The alternative approach is the thermodynamic modelling of trace elements applying Pitzer's equation for brines. Trace elements in solids can be treated as an extreme form of solid solutions and can thus be calculated by a set of different mathematical models. To obtain data for trace elements (Xtrace < 0.05) two different methods are envisaged. The usual way is to use thermodynamic data for the complete solid solution system and to extrapolate using mathematical models to trace element ranges of the mole fraction Xtrace. Due to the fact that no satisfying fits for the complete system were obtained by conventional mathematical methods, a new concept was developed. It is assumed, that the activity coefficient of the trace component ltrace can be described correctly by the following formalism: where R denotes the gas constant, T the temperature, Xmain the mole fraction of the main component and Wi the model parameter. To obtain these model parameters, experimental data from simple ternary solid solution systems (e.g. NaCl-NaBr-H₂O or KMgC_l3 6H₂O- KMgBr₃ 6H₂O-H₂O) were fitted using the above equation and a new mathematical formalism which transforms the classical distribution coefficient into activity coefficients.

The new formalism as well as the model parameters were implemented in the program EQ3/6. The evaporation of seawater was calculated using a chemical composition with 0.84 mmol Br/kg and neglecting the precipitation of carbonate minerals. The calculations are in good agreement with observed mineral sequences from the Zechstein II (Late Permian) section in the "Berlepsch" mine which are regarded as primary layers. The same section was investigated by BOEKE (1908) considering Br in halite and carnallite. The thermodynamic calculation of Br in the Cl-minerals is in good agreement with the natural data. The results show that it is possible to calculate the incorporation of trace elements in complex systems with satisfying accuracy. For these calculations, only experimental data from ternary aqueous systems are necessary to obtain the model parameters.

Boeke, HE, Über das Krystallisationsschema der Chloride, Bromide, Jodide von Na, K, und Mg, sowie über das Vorkommen des Br und Fehlen von J in den Kalisalzlagerstätten -- Zeitschrift für Krystallographie, 45, 346 - 391, (1908).

J11 : 4A/02 : F2

On the Origin of Salinity of Deep Groundwater in Crystalline Rocks

Ingrid Stober (stober@uni-freiburg.de)¹ &

Kurt Bucher (bucher@uni-freiburg.de)²

¹ Geologisches Landesamt, Baden-Württemberg, Albertstr. 5, 79104 Freiburg, Germany

² Institute of Mineralogy, Petrology and Geochemistry, Albertstr. 23b, 79104 Freiburg, Germany

Deep groundwater in fractured crystalline basement has been reported from deep mines and from scientific deep wells. Highly saline brines have been described from several km depth in the continental basement of the Canadian, Fennoskandinavian and Ukrainian shields and elsewhere in the world. In the Black Forest area, which forms a part of the central European continental crust, TDS of typical Na-Cl-type groundwater is more than 5 g/kg at a depth below 2-3 km. Upwelling thermal waters are therefore usually highly saline.We compare the compositional evolution of deep waters in the Black Forest basement with deep crystalline waters from elsewhere and use the halogen systematics (e.g. Cl/Br ratios and other parameters) of the waters to deduce the origin of their salinity. In the Black Forest the composition of deep thermal waters results from chemical interaction of surface water with the rock matrix (mainly weathering of plagioclase and mica) and from mixing of the reacted water with stagnant saline deep water of marine origin. The Cl/Br systematics of about 50 water samples suggests a marine origin: the Cl/Br ratios on a ppm basis are very close to normal marine ratios (Cl/Br=288). In contrast, Cl/Br ratios of other possible sources of salinity show distinctly different Cl/Br ratios. Water derived from dissolved Tertiary halite deposits of the rift valley is on the order of Cl/Br=2400 and water from dissolved Muschelkalk halite deposits has values of about Cl/Br=9900. Therefore, the origin of the salinity in deep waters of the Black Forest is clearly not derived from evaporitic halite deposits. Leaching experiments on crystalline rocks and mineral composition data, on the other hand, show that the average Cl/Br ratio of crystalline rocks world-wide is about Cl/Br=80. Theses values are significantly lower than the observed Cl/Br-values of deep Black Forest waters. In contrast, deep groundwater from Precambrian shield areas has significantly smaller Cl/Br ratios than water from the Black Forest. These waters have a much longer reaction history and consequently contain a much higher water-rock-interaction component. Their Cl/Br values evolved from prolonged reaction with crystalline rocks.

J11 : 4A/03 : F2

Fluid Circulation in Crystalline Basement Rocks: Constraints from Fluid Inclusions

Sarah Gleeson (sarah@earth.leeds.ac.uk)¹,

Bruce Yardley (bruce@earth.leeds.ac.uk)¹,

Joe Cann (j.cann@earth.leeds.ac.uk)¹ &

Ingrid-Anne Munz (ingridm@ife.no)²

¹ School of Earth Sciences, University of Leeds, Leeds, LS2 9JT, United Kingdom

² Institutt for Energiteknikk, Instituttveien 18, P.O. Box 40, N-2007 Kjeller, Norway

Recent deep drilling programs have shown the presence of hydrostatically pressured fluids, driven by topographic head at depths of up to 9 km in the crust. However, fluid inclusion studies of palaeohydrological systems suggest that in vein systems fluid pressures may cycle from above hydrostatic to lithostatic and commonly quartz growth occurs when fluid pressures are close to lithostatic. Fluid inclusions, therefore, can offer constraints on the relative roles of pressure and temperature fluctuations on vein formation and give insights into processes such as mixing at depth.

The high grade basement rocks of the Modum Complex in southern Norway are host to late, prismatic, growth-zoned, quartz veins which contain both complex hydrocarbon and aqueous fluid inclusions (Munz et al., 1995). The aqueous fluids have homogenisation temperatures in the range 120 to 340°C, with a strong mode at 180°C, and have a striking range in salinity from 0-40wt% NaCl equiv. Individual quartz zones can show a wide range in salinity in coeval fluid inclusion populations, at relatively constant temperature. Calculated fluid densities at homogenisation have a wide spread, from 800 to greater than 1050 kg/m³ and densities of fluid inclusions within a single growth zone can range from 930 to 1050 kg/m^3. While fluid inclusion densities indicate a spectrum of fluid pressures from near-lithostatic to hydrostatic, most quartz growth occurred when there were strong overpressures.

The spread in salinity at relatively constant homogenisation temperature suggests that the salinity variations reflect the composition of primary source aquifers. This, combined with the large variations in densities, indicates multiple processes may have be important in these vein systems. Possible mechanisms for vein formation include: the influx of fluids to dilational sites in the basement during faulting; sealing of fluid-filled fractures by mineral precipitation; convective overturn of variable density fluids within fractures; segregation of quartz to the cooler parts of fluid-filled, sealed fractures; repeated fracturing and fluid migration. Some of these processes imply fluid pressures that are near hydrostatic, while those taking place in sealed cracks may occur at near-lithostatic pressure. Temperature and compositional variations of fluid inclusions within growth zoned quartz attest to sporadic rapid influxes of both hotter and colder waters, and given sufficient permeability, buoyancy driven convection within the fractures may have been an important process.

Munz IA, Yardley BW D, Banks D, Wayne D, Geochim. Cosmochim. Act, 59, 239-254, (1995).

J11 : 4A/04 : F2

Analysis of Individual Fluid Inclusions from Alpine Clefts by Laser Ablation - Optical Emission Spectroscopy (LA-OES)

Cécile Fabre (fabre@cregu.cnrs-nancy.fr),

Marie Christine Boiron

(boiron@cregu.cnrs-nancy.fr) &

Jean Dubessy (dubessy@cregu.cnrs-nancy.fr)

BP 23, CREGU, Vandoeuvre les Nancy, France

The quantitative chemical analyses of fluid inclusions are a prerequisite for understanding and modelling fluid rock interactions as only these inclusions contain direct evidence of the composition of fluids responsible for past geological processes. Recent developments using laser ablation optical emission spectroscopy (LA-OES) (Boiron et al.,1991, 1997; Moissette et al., 1997), have shown that this method is adapted for the analysis of the ion content in individual fluid inclusions. The principle is the following: a Nd-YAG laser is coupled with an optical emission spectrometer which analyses the emission lines of the elements present in the plasma resulting from the interaction laser - material.

Calibration and reproductibility: The work has been focused on the 580-780 nm spectral range where Na, Li, K and Ca emission lines can be obtained simultaneously for each shot. Calibrations of element ratios have been carried out on synthetic glasses, minerals and synthetic fluid inclusions in quartz. Reproductibility for net intensity of Na, Ca, Li emission lines in glasses is in the range 5 to 15%. In fluid inclusions, RSD element ratios are in between 10 to 25%.

LA-OES appear to be a powerful localised technique for the analysis of major and trace elements at the ppm level in glasses and minerals, and 10-300 ppm in fluid inclusions (respectively for Na, Ca, Li and for K).

For Alps fluid inclusions element emission lines of Na, Ca, K and Li were identified. The LA-OES technique allows a cartography of the sample, showing the element ratios, inclusion to inclusion. The significant variation which may occur on the final mean, calculated on several inclusions, can be explained by this kind of repartition. The element ratios obtained by LA-OES (Na/Li, Na/Ca and Na/K) are close to those from crush leach method. The possible variation can be explained by the differences between the two techniques.

Considering that the fluid is at equilibrium with the Na and K feldspars present in the rock, the estimation of fluid trapping temperature gives a value around 370±20°C and fluid pressure between 2,5 and 3 kbar.

Boiron M C et al., Geochim. Cosmochim. Acta, 55, 917-923, (1991).

Boiron MC et al, XIV ECROFI, 44-45, (1997).

Moissette Aet al, XIV ECROFI, 211-212, (1997).

J11 : 4A/05 : F2

Geochemistry of the Thermal Springs and Fumaroles of Basse-Terre Island, Guadeloupe, Lesser Antilles

Tatjana Brombach (tatjana.brombach@imp.unil.ch)¹,

Luigi Marini (luigi@ugo.dister.unige.it)² &

Johannes Hunziker (johannes.hunziker@imp.unl.ch)¹

¹ Institut de Minéralogie et de Pétrograhpie, Université de Lausanne, BFSH2, 1015 Lausanne, CH

² Dipartimento di Scienze della Terra, Università di Genova, Corso Europa26, 16132 Genova, I

Typical high-salinity, Na-Cl geothermal liquids circulate in the Bouillante geothermal reservoir, whose temperature is close to 250°C. These Na-Cl aqueous solutions rise towards the surface, undergo boiling and mixing with groundwater and/or seawater and feed most Na-Cl thermal springs located in the central Bouillante area. These Na-Cl thermal springs (BO₃, HS₂, Thomas, BO₄ and BO₁) are contoured by Na-HCO₃ thermal springs (BO₂, Bain du Curé and La Lise) and by the Na-Cl thermal spring of Anse à la Barque (a groundwater slightly mixed with sea water), which are all heated through conductive transfer.

The two main fumarolic fields of the Soufrière area discharge vapors formed through boiling of typical hydrothermal aqueous solutions at temperatures of 190-215°C below the Ty fault area and close to 260°C below the dome summit. The boiling liquid producing the vapors of the Ty fault area has  D and  ¹⁸O values relatively similar to those of the Na-Cl liquids of the Bouillante geothermal reservoir, whereas the liquid originating the vapors of the summit fumaroles is strongly enriched in ¹⁸O, due to input of magmatic fluids from below. This process is also responsible of the virtual absence of CH₄ in these fumaroles. The thermal features situated around the Soufrière dome include: (1) Ca-SO₄ springs, produced through absorption of hydrothermal vapors in shallow groundwaters, (b) conductively heated, Ca-Na-HCO₃ springs, and (c) two Ca-Na-Cl springs produced through mixing of shallow Ca-SO₄ waters and deep Na-Cl hydrothermal liquids. The geographical distribution of the different thermal features of the Soufrière area indicates the presence of: (1) a central zone dominated by the ascent of steam, which either discharges at the surface in the fumarolic fields or is absorbed in shallow groundwaters; (2) an outer zone, where the shallow groundwaters are heated through conduction or addition of Na-Cl liquids coming from hydrothermal aquifer(s).

J11 : 4A/06 : F2

Chlorine-Isotopic Evidence for Advection of Highly Modified Pore Water and Long-Range Diffusion in Blake Ridge Submarine Gas-Hydrate Zone

Reinhard Hesse

(Reinhard.Hesse@ruhr-uni-bochum.de)¹,

Per K. Egeberg² &

Shaun Frape³

¹ Institut für Geologie, Ruhr-Universität Bochum, D-44801 Bochum, Germany

² Department of Chemistry, Agder College, 4604 Kristiansand, Norway

³ Department of Earth Sciences, University of Waterloo, Waterloo, ON N2L 3G1, Canada

A steady downward depletion of the heavy chlorine isotope to nearly -4‰  ³⁷Cl at ~750 m below sea floor (mbsf) at Ocean Drilling Program (ODP) Site 997 in the Blake Ridge gas-hydrate field in West Atlantic (Hesse et al. in press), one of the strongest ³⁷Cl depletions hitherto reported for marine pore waters, is interpreted as the result of mixing of two end member fluids. A strongly ³⁷Cl depleted low-chlorinity (506 mM) water is advected into the drilled sequence from below, similar to some North Sea formation waters (Eggenkamp & Coleman, 1998). This water is in diffusive communication with a near-0‰  ³⁷Cl paleo-seawater reservoir at shallow subsurface depths with a chlorinity maximum (561 mM) in the upper 20 mbsf reflecting saltier connate water buried during the Pleistocene glaciations and a minor contribution from salt exclusion during gas-hydrate formation at the roof of the hydrate zone (postulated to occur at 24 mbsf). The source of the isotopically light, low-chlorinity water is not known, as are the mechanisms for generating the large ³⁷Cl depletion. Some unknown process must be operating during advection at depth that may be dominantly lateral, which enriches the light Cl isotope, perhaps preferential adsorption of ³⁷Cl (?on organic matter) during transport through the sediment matrix, or clay reactions at greater depth. This water may come from the flanks of the Blake Ridge hydrate field, whose hydrate cap forms a broad anticlinal structure that is cut by numerous deep-reaching subvertical faults, which would facilitate dewatering. The chloride concentration-gradient driving diffusion is primarily controlled by environmental factors above and below the hydrate zone; the contribution from hydrate melting at the base of the hydrate zone (at 452 mbsf) is subordinate. Chlorine isotope diffusion continues below the hydrate base at 452 mbsf as shown by the  ³⁷Cl values which continue to decrease, although no chloride concentration gradient exists below 450 mbsf. The reason is that the light and heavy chlorine isotopes themselves show vertical concentration gradients caused by the reactions that preferentially remove ³⁷Cl at greater depth. A combined diffusion-advection model (Egeberg & Dickens, in press) reproduces the  ³⁷Cl curve and is used, together with the in situ chlorinity measurements, to estimate hydrate concentrations, which are 3.8% of the pore space on average (up to 24.5% in hydrate-rich layers; near-100% in rare massive hydrate layers).

Hesse R, Frape SK, Egeberg PK & Matsumoto R, Proc. Ocean Drilling Progr. Scient. Res, 164, in press, (1999).

Eggenkamp HGM & Coleman ML, Water-Rock Interaction, 9, 309-312

Egeberg PK & Dickens G, Chem. Geol, in press, (1999).

J11 : 4A/09 : F2

Geochemical Mapping of Magmatic Gas-Water-Rock Interactions in the Aquifer of Mount Etna Volcano

Walter D'Alessandro (walter@igf.pa.cnr.it)¹,

Alessandro Aiuppa²,

Patrick Allard³,

Sergio Bellomo²,

Lorenzo Brusca,

Agnes Michel⁴ &

Francesco Parello²

¹ IGF - CNR, Via U. La Malfa, 153, Palermo, Italy

² CFTA, Università di Palermo, via Archirafi 36, Palermo, Italy

³ LSCE, CEA-CNRS, Orme des Merisiers, Gif/Yvette, France

⁴ LPS, CEA-CNRS, Saclay, Gif/Yvette, France

Systematic analysis of major and minor elements and dissolved gases in groundwaters from the slopes of Mt. Etna in 1995-1998 provides a detailed geochemical mapping of the aquifer of the volcano and of the interactions between magmatic gas, water bodies and their host rocks. Strong spatial correlations between the largest anomalies in pCO₂, K, Rb, Mg, Ca and Sr evidence a dominant control of magmatic gas (CO₂) and consequent basalt leaching by acid waters upon the chemistry of the shallow (meteoric) Etnean aquifer. Most groundwaters displaying this magmatic-type interaction, outflow within active faulted zones on the S-SW and E lower flanks of the pile, but also in a newly recognized area on the northern flank, possibly tracking a main N-S volcano-tectonic structure. In contrast the distribution of other parameters (T, TDS, Na, Li, Cl and B) in the aquifer does not fit with a simple gas-water-basalt interaction. The W and E hydrogeological basins strongly differ in all these parameters, the former being hotter and more saline than the latter. These differences are due to the hydrogeological conditions (stronger precipitations, greater steepnees of the volcano's slope and lower altitude of the recharge zone) that allow a considerably higher water flow through the E basin. If we roughly assume that both basins are interested by similar magmatic-gas fluxes, the above mentioned differences may be the result of the longer water residence time in the W sector. Moreover, these hydrogeological conditions allow the water chemistry of this sector to be affected by small contamination by thermal brines with high salinity, high content in B, Cl and reduced gas (H₂S, CH₄) and low K/Na ratio, likely hosted in the sedimentary basement and of "old" marine origin. The rise of these waters, probably driven by the oil gases through faults, is generally masked by the high flow rate of the meteoric groundwater through the shallow aquifer. Only where the aquifer is thinner or has a smaller flow, and where the volcanic pile tend to disappear, this mixing process can be observed. Apart from the Paternò area (S), where these brines spring out undiluted, the mixing is also detected to the W (Bronte, Maletto), SW (Adrano) and SE (Acireale), suggesting its possible widespread occurrence beneath Etna.

J11 : 4A/10 : F2

The Mechanism and Dissolution Rates of Basaltic Glass as a Function of Temperature, pH, and Solution Composition

Eric H. Oelkers (oelkers@cict.fr)¹,

Sigurdur R. Gislason (sigrg@raunvis.hi.is)² &

Jacques Schott (schott@lucid.ups-tlse.fr)¹

¹ Laboratoire de Geochimie, Universite Paul Sabatier, 38 rue des Trente Six Ponts; 31400 Toulouse, France

² Science Institute, University of Iceland, Dunhagi 3, 107 Reykjavik, Iceland

Because of its widespread occurrence on the ocean floor and volcanic terrains, its large scale emission during volcanic eruptions, and its relatively rapid dissolution rate, basaltic glass plays a major role in the global flux and cycling of numerous metals and nutrients. To allow quantification of these processes, this study has been dedicated towards deducing the mechanism and a predicative model to describe basaltic glass dissolution in natural systems.

The mechanism of basaltic glass dissolution must account for a plethora of observations including:

-Hydrogen depth profiling and XPS analyses show that dissolving basaltic glass surfaces are depleted in network modifying cations such as Na, Ca, and Mg, which are rapidly and reversibly exchanged with protons (Schott, 1990).

-The pH dependence of its dissolution rate mimics that of aluminum (hydr)oxide mineral solubility: it exhibits a sharp increase with decreasing pH at acidic conditions and a more gentle increase with increasing pH at neutral to basic conditions. The minimum rate is found at ~pH=6 at low temperatures, but this minimum moves to lower pH with increasing temperature (Guy and Schott, 1989).

-At far from equilibrium conditions and constant pH, basaltic glass dissolution rates are 1) independent of aqueous silica activity, 2) decrease substantially with increasing aqueous aluminum activity, and 3) in mildly acidic solutions increase dramatically with increasing oxalic acid concentration (Gislason et al., 1999).

Taking account of these observations, basaltic glass dissolution is described in terms of its control by a silica rich precursor complex, similar to that which controls the dissolution rates of the alkali-rich feldspars (Oelkers et al., 1994). This precursor complex is formed by several metal for hydrogen exchange reactions; those with network modifying cations go essentially to completion and do not effect the overall concentration of this precursor complex, whereas the exchange of glass network formers such as Al do not go to completion and therefore their concentration strongly effects precursor complex concentration and thus the rate.

Consideration of this mechanism leads to an equation describing the variation with solution composition of basaltic glass dissolution rates at far from equilibrium conditions (r₊ ) and constant temperature given by

r₊ = k₊ (a_H+³/a_Al3+)ⁿ (1)

where, k₊ designates a rate constant, a_i stands for the activity of the subscripted aqueous species, and n represents a stoichiometric coefficient. Use of Equation (1) together with an Arrhenius relationship yields an accurate description of basaltic glass dissolution rates as a function of temperature, pH, and the aqueous activities of Si, Al, and organic acids.

Gislason SR, Oelkers EH & Schott J, (In preparation), (1999).

Guy C & Schott J, Chem. Geol, 78, 181-204, (1989).

Oelkers EH, Schott J & Devidal J-L, Geochim. Cosmochim. Acta, 58, 2011-2024, (1994).

Schott J, in Aquatic Chemical Kinetics, Stumm Ed, 337-366, (1990).

J11 : 4A/11 : F2

An Experimental Study of Aqueous Cesium-Acetate and Strontium-Acetate Complexing at Temperatures from 25 to 180°C

K. Vala Ragnarsdottir (vala.ragnarsdottir@bris.ac.uk)¹,

Patrica Fournier (fournier@lucid.ups-tlse.fr)²,

Jean Claude Harrichoury² &

Eric H. Oelkers (oelkers@cict.fr)²

¹ Department of Earth Science, University of Bristol, Queens Road, Bristol BS8 1RJ, UK

² Laboratoire de Geochimie, Universite Paul Sabatier, 38 rue des Trente Six Ponts; 31400 Toulouse, France

Cs and Sr isotopes are among the most widespread radionuclides in the environment. Owing to their ~30 year half-lifes, the radioactivity stemming from ¹³⁷Cs and ⁹⁰Sr can persist over significant time frames. For example, the 'reference transuranic waste' generated at the Rocky Flats plant in Colorado USA obtains 93% of its initial radioactivity from ⁹⁰Sr; this value drops only to 90% after 100 years.

Due to the efforts of numerous investigators, we now understand reasonably well the retardation of these radionuclides resulting from their absorption and co-precipitation with minerals in natural environments. Little attention, however, has been placed on determining the effect of the presence of organic acids on the speciation and transport of Cs and Sr. To address this problem, dissociation constants of aqueous cesium acetate (CsCH₃COO⁰) and strontium acetate (Sr(CH₃COO)₂⁰) were determined potentiometrically in this study at temperatures from 25 to 180°C (see Fournier et al., 1998 for experimental methods). Aqueous acetate complexing was investigated in this study because it is representative of natural organic acids in general and is found in abundance in both surface and subsurface waters. Resulting dissociation constants are in reasonable agreement with those predicted by Shock and Koretsky (1993).

Species distribution calculations performed using dissociation constants obtained in this study indicate that Cs-acetate complexing is weak; CsCH₃COO⁰ accounts for less than 10% of dissolved Cs in aqueous solutions containing up to 1 molal acetate at all temperatures up to 180° C. In contrast aqueous Sr-acetate complexing is relatively strong; Sr-acetate complexes comprise over 50% of all aqueous strontium in neutral and basic solutions containing 1 molal acetate. In acidic conditions (pH<2), however, CH₃COOH⁰ becomes the dominate acetate species and little aqueous strontium acetate complexing is present. Taken together, the results of this study imply that the presence of natural organic acids will enhance considerably the aqueous stability and thus the transport of radioactive strontium at all but extremely acidic conditions.

Fournier P, Oelkers EH, Gout R & Pokrovski G, Chem. Geol., 151, 69-84, (1998).

Shock EL & Koretsky CM, Geochim. Cosmochim. Acta, 57, 4899-4922, (1993).

J11 : 4A/12 : F2

Stability of Polysilicic and Monosilicic Acid at the Mineral Surface of Hydroxides

Martin Dietzel (mdietze@gwdg.de)

Goldschmidtstr.1, D-37077 Göttingen, Germany

In natural waters silicic acid is generated by the dissolution of rocks. The composition and concentration depend on the geochemical environment. In sedimentary surroundings silicic acid may withdrawn from the solution by adsorption at the surface of solids, especially hydroxides, and by crystallisation of silica containing solids. These interactions of aqueous solutions and solids are related to the polymerization behaviour of silica in the solution and at the surface of solids.

Adsorption experiments with polysilicic and monosilicic silica were carried out by the addition of lepidocrocite, hematite, feroxyhyte, goethite, akaganeite, magnetite, ferrihydrite and gibbsite. In all cases the adsorption of monosilicic acid is followed by first order reactions. At equilibrium the adsorption may be described by complexation of silicic acid at the surface of the solids. Maximum adsorption is obtained at pH >> 9.8.

However, the processes are more complex if polysilicic acid is adsorbed [1]. One part of the polymer is adsorbed at the surface within a rather short time. The other part decomposes to monomer via the depolymerization kinetic in homogeneous solution, without a solid [2]. The stability of the initial adsorbed polymer depends strongly on the chemical composition of the solution. At pH > 6 polymer decom-poses to monomer at the surface until equilibrium of monosilicic acid is reached. At pH < 6 polymer is stabilized at the surface of the solids.

If these results are applied to natural systems, adsorption of monosilicic acid is favoured in slightly alkaline solutions and polymerization of silica at the hydroxide surface may be favoured in acidic solutions. The adsorption of silica onto aluminium and iron hydroxides may represent the initial step for the formation of silicates (e.g. [3], formation of kaolinite). In general, the disordered linked silica polymers are expected to inhibit the crystallisation (e.g. [4]). The present results show, that these effect has to be considered only in acidic solutions.

Dietzel M, Böhme G, Chem. Erde, 57, 189-203, (1997).

Dietzel M, Usdowski E, Colloid Polym. Sci., 273, 590-597, (1995).

Brady PV, House WA, In: Physics and Chemistry of Mineral Surfaces. CRC Press, New York, Chapter 4, 225-305, (1996).

Iler RK, The chemistry of silica. Wiley-Interscience, New York, 866pp, (1979).

J11 : 4A/13 : F2

Drying of Tuffeaux: Relation with Microporosity and Adsorption

C. Thomachot (celine@illite.u-strasbg.fr) &

D. Jeannette (dj@illite.u-strasbg.fr)

Centre de Géochimie de la Surface, ULP Strasbourg 1, rue Blessig 67084 Strasbourg Cedex

Tuffeaux of Touraine are siliceous chalks (SiO₂>26%), with elevated porosity (30-50%) and mercury stress-hole of around 1µm. Microscopic observations show homogeneous pore structures with localisation of microporous ranges in microcalcites and opal-chalcedony spheres.

Let a saturated tuffeau sample, isolated except for one face in a water-proof sheath, be put to dry in a tub where relative humidity (R.H.) is controlled by brine. The drying curve -constructed by writing out water loss per surface units according to the time- is generally divided in 3 phases (Pavone, 1982).

A first phase (I) during which the flow doesn't vary very much. The surface is wet because the capillarity water transfer toward the surface is enough to equilibrate the loss by drying (Hallaire, 1958, Tournier, 1998). At the end of this phase, the porosity still occupied by water corresponds to critical saturation (Sc).

The last phase (III) with a very low and nearly constant flow. The desaturation is obtained by the diffusion of vapour through the porous media (Gardner, 1958). A phase (II) ensures the transition between phases I and III. The flow varies because there is coexistence of the two last mechanisms.

During phase III we consider traditionally that the meniscus forming a wet head regresses into the sample where the flow is constant. We show here that:

The importance of Sc is in direct relation with the porosity inferior to the mercury-stress-hole. For example, in the case of the tuffeau here: Sc=13.56% with 13.95% of microporosity.

The experimental adsorptions at 95% R.H. sensibly indicate the beginning of phase III. The end of the curve merges with the adsorption value at the R.H. at which the drying is done. In phase III, we shouldn't have any wet head and drying would correspond to a desorption.

Drying curve at 55% R.H. of a 31% total Porosity Tuffeau

Pavone, Thèse Paris VI, (1982).

Hallaire M. et al, C.R. Acad. Sci., 46, 2151-53, (1958).

Tournier B. et al, EUG, (1998).

Gardner W.R., Soil Sci., 85, 228-33, (1958).

J11 : 4A/14 : F2

Oxfordian Carbonate Concretions Preserve Chlorine Isotope Profiles Since Their Formation

Hans Eggenkamp (h.g.m.eggenkamp@reading.ac.uk)¹,

Neville Hollingworth (nth@wpo.nerc.ac.uk)² &

Max Coleman (m.l.coleman@reading.ac.uk)¹

¹ PRIS, University of Reading, Whiteknights, Reading RG6 6AB, United Kingdom

² NERC, Polaris House, North Star Avenue, Swindon SN2 1EU, United Kingdom

We have collected clay samples and co-existing carbonate concretions from the Oxford Clay (Jurassic) at Rixon Farm Pit near Cirencester, England. These concretions have grown in situ after deposition of the clay (Hudson, 1978). We studied the chemical and chlorine isotopic compositions of the pore fluids we leached from them in order to determine pore-water chemical changes during the formation of these concretions. We took clay samples from 8 depths in a freshly dug pit, three concretions from this pit, four subsamples from a concretion found elsewhere in the quarry, two water samples extracted from concretions and a sample of rainwater from a puddle in the quarry.Salt was extracted from all samples and we determined the Li, Na, K, Mg, Ca, Cl and sulphate content using ICP and ion chromatography. Bicarbonate content is calculated as the residue to neutralise the ionic balance. We determined the Cl isotope composition of all the samples, except the water samples which were too small.

As the exact amount of fluid in the samples was not known, and sulphate was the most common ion in all samples, all chemical data are reported as the ion/sulphate ratio. Although the clay and the concretions have a completely different lithology, the chemical composition of the salt leachates show an overlap between the two sample types. However, it seems that (relative to sulphate) the concretions are enriched in K, Mg, Ca, bicarbonate, and Cl, while the clay is enriched in Li and Na. These variations however are not very systematic. It may be that generally in the clays cation exchange reactions keep Li and Na concentrations higher, while Ca, sulphate and bicarbonate contents are higher in the calcite environment of the concretions.

The chlorine isotope compositions of the clays and the concretions show very clear differences. All of the measured isotope values of the concretions are negative (their isotope ratio is lower than the standard seawater), while all clay samples show positive values. Even more interesting are the subsamples taken from one of the concretions: the centre sample was less negative than the two subsamples taken near the rim. If there had been diffusion of Cl from the concretion into the clay, then a positive chlorine isotope value at the centre would be expected, however, it is negative. Very few low temperature processes are known to fractionate Cl isotopes significantly from seawater, however in this case, there must be not only a large chlorine isotope fractionation during formation of these concretions, but also a large chlorine isotope variation can exist over a long time between concretions and clay, over a very short distance.

Hudson, JD, Sedimentology, 25, 339-370, (1978).

Session J11:4P

J11 : 4P/01 : PO

Influence of the Pore Structures and Evaporating Surfaces on Drying Kinetics of Stone Samples

B. Tournier (tournier@illite.u-strasbg.fr)¹,

J. -P. Sizun (jean-pierre.sizun@univ-fcomte.fr)² &

D. Jeannette (dj@illite.u-strasbg.fr)¹

¹ Centre de géochimie de la surface, EOST, UMR 7517 CNRS-ULP, 1 rue Blessig, 67084 Strasbourg Cedex

² Université de Franche-Comte, U.F.R Sciences & Techniques, Laboratoire de Géosciences, 16, route de Gray, 25030 Besancon Cedex, France

Drying experiments made on stone samples whith different petrophysical characteristics allow us to analyse the influence of pore structures and evaporating surface on the drying kinetics. The sandtone used here is characterised by a bulk vacuum porosity (Nt) of 12%, a partial porosity by capillary imbibition (Nc) of 6% and a mercury pore threshold (Ra) of 10µm, whereas the micrite characteristics are Nt = 27%, Nc = 25% and Ra = 0.35µm. Cylindrical cores of 8 cm in height and 4 cm in diameter are dried after Nc or Nt water-saturation. They are wrapped in an impermeable sheath with an open and regular surface at the top. The relative humidity is set at 33 or 75% in the two containers used for drying. Evaporation flows of the samples and of a free water surface are deduced from the curves obtained by reporting the loss of weight per surface unit in relation to time.

If we compare the rate of evaporation of the samples and of the free water surface, we can notice that:

variations in the evaporation rate of the samples define three stages of drying;

differences between the evaporation rates and those of the free-water surface are in relation to the estimation of the evaporating surface.

Drying Kinetics of a Free Water Surface Compared to Micrited Sandstone after Saturation under Vacuum

Considering that during the first stage of drying (1, on the figure), the surface of the samples remains wet because of capillary transfers, it appears that:

during this first stage the flow and consequently the evaporating surface, are not constant;

these flow variations are intimately linked with the initial water saturation and the pore structures.

The results allow to understand better the weathering mecanisms and the effect of the protection products of building stones.

J11 : 4P/02 : PO

Monitoring of the Karymskoe Lake Condition (Kamchatka) in Connection with a Sub Aqueous Volcanic Eruption

Sergey Fazlulin (plof2@hotmail.com) &

Sergey Ushakov

Institute of Volcanology, Russia Academy of Sciences, 9, Piip boulevard, 630090 Petropavlovsk-Kamchatsky, Russia

In a number of cases volcanic lakes are one of the elements of volcanic hydrothermal systems. Chemical composition and circulation of volcanic lakes are more various then those of ordinary fresh basins. This diversity is conditioned by the combination of endogenic and exogenetic flows of substance and energy.

The subaqueous eruption in the Karymskoe lake (water reserve ~ 460.6 million m³) on January 2-3, 1996 completely changed its regime. During 10-14 hours of the subaqueous eruption the lake which was a fresh oligotrophic mountain basin with various biota became one of the largest repositories of acid water (pH ~ 3.0-3.1) in the world. Mineralization decreased in 5 times (up to 1 g/l) and, in spite of great heat-loss, the layer of water with temperature 12-25^oC retained on the surface during several months. As a result of the subaqueous eruption a volcanic structure 600 m in diameter and up to 60 m deep was formed on the bottom of the lake.

A shallow (up to 3 m) strait connected it with the main part of the lake. Seasonal vertical circulation on the one hand and water exchange through the strait between the main and crater parts of the lake on the other hand determine recent hydrological regime of the lake. Depending on season two conditions of water exchange can be observed. From autumn till spring water exchange is more intensive. As summer begins (1997) water mass of the crater part over the range of depths 15-60 m is "isolated" and is getting warm owing to considerable inflow of heat through the bottom. Monitoring of the temperature of this water mass allowed to evaluate heat discharge through the bottom. In summer 1997 it was 526.7 mln.J. Great dilution of the lake with surface waters in spring period of 1997 led to vertical mixing up to 45 meters, formation of cold intermediate water mass (~2^oC) and isolation of relatively warm, acid and mineralized water in the benthic layer.

Peculiarities of vertical water exchange in the Karymskoe lake which was formed by summer 1997 determined its hydrochemical regime. It is dilution of lacustrian waters with surface fresh waters, on the other hand - it is active discharge of chemical elements from fresh erupted material of the volcanic structure into the lake by thermal waters. With all this one can notice considerable discharge of iron into the lacustrine system. During the process of oxidation iron passes into suspension and deposits on the bottom all round the water area. Water type formed after the eruption is Cl-SO₄ and Ca-Na.

J11 : 4P/03 : PO

Snow Composition of the Areas of Present Volcanic Activity of Kamchatka

Sergey Fazlulin (plof2@hotmail.com) &

Aleksey Kumarkov

Institute of Volcanology, Russia Academy of Sciences, 9, Piip boulevard, 630090 Petropavlovsk-Kamchatsky, Russia

It is well known that the snow layers can be used for the ecological and geochemical monitoring, particularly, in the field of the study of volcanic aerosols pollution into surrounding environment. Parallelly we can estimate the volume and composition of the volcanic products that have been polluted upon the surface, made the models of the migration of volcanogenic elements from the atmosphere to the snow layer and further - to the soil and water.

In report the data of many years research works in volcanic areas of Kamchatka are given. As a background, we consider the samples that were picked up from transkamchatka's submeridional profile, that is relatively far from the antropogenic and volcanogenic sources of aerosol inflow.

It was clarified that there exist several types of volcanic influence on the melted waters composition. These influences depend on the presence or absence of volcanic ash in volcanic pollution, quantity of combinations sorbed on the volcanic ash, phase of eruption, stage of fumarolic activity.

The process of volcanic gases condensation leads to the significant elimination of pH in atmospheric precipitation, It is noted that in atmospheric precipitation that is formed under the influence of volcanic activity, contains microelements, the proportions of sulfates and haloides are increased. Microelements (in a form of heavy metals) come to the soil in concentrations that highly exceed the concentrations in the background and can affect biota. Increasing of the proportion of sulfates, both under the influence of antropogenic and volcanic sources, causes the troubles with identification of prevailing source of substances in the zones of its mutual influence. Identification of the sources is possible with the help of diagrams pH-pSO₄. On such diagrams the points, that belong to different extracts in genesis, are grouped into the individual fields of distribution.

J11 : 4P/04 : PO

Hydrogeochemistry of the Clara Mine, Black Forest, Germany

Yinian Zhu (zhuyinia@ruf.uni-freiburg.de)¹,

Ingrid Stober (stober@uni-freiburg.de)² &

Kurt Bucher (bucher@uni-freiburg.de)¹

¹ Institute of Mineralogy, Petrology and Geochemistry, Albertstr. 23b, 79104 Freiburg, Germany

² Geological Survey of Baden-Wuerttemberg, Albertstr. 5, 79104 Freiburg, Germany

Water from > 20 different localities in and around the underground barite and fluorite mine Clara near Wolfach (Black Forest, Germany) were collected monthly during one year and analyzed chemically in order to elucidate the interaction of surface water with the various rock types and to better understand the underground flow systems. The barite and fluorite mineralization occurs in 3 major vein systems in gneisses of the Black Forest basement. It is associated with quartz, carbonate, various sulfides, natural alloys and 250 species of secondary minerals. Water transport is strongly focused within the mineral veins owing to their high fracture and cavity related permeability. In contrast, two major fault and shear zones in the mine are nearly impermeable. Temperature, pH, electrical conductance, Eh and O₂ concentration, 6 anions and 18 cations were determined in waters from exploration drillholes in the mine, mine waters, surface waters , rain and snow sampled in 97/98. Four types of water can be distinguished on the basis of the relative proportions of the major ions: (1) Ca-HCO₃ water, (2) Ca-HCO₃-SO₄ water, (3) Ca-SO₄ water and (4) Ca-SO₄ (Na-Cl) water. From the type-one to type-four water the electrical conductance increases from about 0.100 ms/cm to more than 1 ms/cm. On diagrams showing the pH-normalized activity of the major cations versus activity of dissolved silica, all waters fall into the stability field of kaolinite which is an observed major alteration product in vein rocks. All waters are very close to quartz-saturation and no major systematic differences between the 4 water types are evident on aa-diagrams. On an Eh-pH diagram for the system Fe-Cu-O-H₂O-S, the data plot into 3 stability fields: Fe₂O₃ + Cu²⁺ + SO₄, Fe₂O₃ + CuO + SO₄ and Fe₂O₃ + Cu₂O + SO₄. Sulfides are not stable in these waters.The source of the solutes is partly related to fluid-rock reaction and partly anthropogenic (e.g. mining, fly-ash deposition). The two basic sources can be distinguished on the basis of trace element patterns.

J11 : 4P/05 : PO

Experimental Water-Rock Reaction of Black Forest Gneiss and Granite

Richard Liegl¹,

Ingrid Stober (stober@uni-freiburg.de)² &

Kurt Bucher (bucher@uni-freiburg.de)¹

¹ Institute of Mineralogy, Petrology and Geochemistry, Albertstr. 23b, 79104 Freiburg, Germany

² Geological Survey of Baden-Wuerttemberg, Albertstr. 5, 79104 Freiburg, Germany

Two characteristic crystalline rocks of the Black Forest basement, a gneiss from the Pauli-Schänzle quarry (Kinzig Valley) and a granite from Triberg (Central Black Forest) have been experimentally reacted with water in a batch reactor under a series of different experimental conditions in order to better understand the composition and evolution of groundwater in the crystalline basement. Experiments with fine-grained powders (< 0.01 mm) showed a remarkably high mobility of chloride: gneiss contains up to 48 mg dissolvable chloride per kg rock, granite up to 140 mg Cl / kg rock. The salinity of leachates is proportional to the amount of preserved fluid inclusions which is higher in Qtz-rich granites than in feldspathic gneisses with little Qtz. Experiments with CO₂-saturated water did not appreciably increase Cl, although TDS increased by a factor of 6. In coarsely crushed gneiss and granite (grain size > 1 mm) 5 and 8 mg Cl was extractable per kg rock, respectively. Average measured Cl/Br ratio was about 100 (on a ppm basis), a value typical of primary high-T fluids trapped in fluid inclusions. The Cl/Br-ratios of leachates from granite and gneiss were very similar and independent on temperature indicating that fractionation was not important at high-T. The release of the saline fluids from the granite to the groundwater is clearly related to the rate of fracture formation and faulting. In both, granite and gneiss, more Mg than Ca is extractable. However, Ca + Mg is higher by an order of magnitude in gneiss leachates. Leachates from both rocks are strongly alkali dominated and K > Na under most experimental conditions. Sulfate and nitrate are low (5 mg/kg) compared with Cl which is typically higher by a factor of 30 (!) in granite. Experiments at 50°C resulted in higher TDS and bulk reaction rate. Na/Ca ratio increased from 4 at 25°C to 13 at 50°C. The origin of the solutes is dissolution of secondary calcite (Ca) alteration of biotite and seconday chlorite (K, Mg), alteration of plagioclase and opened fluid inclusions (Na). The dominate primary anion is chloride which resides in fluid inclusions and in halite on grain boundaries (in contrast to HCO₃ which is athmospheric). Weathering of typical basement rocks and reaction of groundwater with gneiss and granite produces waters of a Na-K-Cl type. With the exception of biotite alteration, the contribution of silicate weathering is insignificant. Ion exchange on the fine rock powder and reaction products (clays) is important and increases K+(Na) relative to Ca+(Mg) in the water.

J11 : 4P/06 : PO

Experimental Leaching on Basaltic Rocks to Recognise Geological Environment of Reunion Island Springs

Jean-Luc Hoareau (jlhoarea@univ-reunion.fr) &

Eric Nicolini (nicolini@univ-reunion.fr)

¹ Laboratoire des Sciences de la Terre, Université de la Réunion, BP 7151, 97715, Saint-Denis Messag, Cédex 9, 97400, Réunion Island, France

In our study, we have performed a chemical alteration on the main basaltic rocks observed on Reunion Island by ultrapure water percolation. The hydrochemical profile of the experimental water is determined by ionic chromatography (main cationic and anionic elements (Na+, K+, Ca²⁺, Mg²⁺, F^-, Cl^-, NO₃^-, SO₄^2-), titration (HCO₃- and CO₃^2-), spectrophotometry (SiO₂)).

On the basis of these experimental results, we have shown a differential alteration of the parts of the rocks according to the repartition of the element. This study highlights the preferential hydrolysis of sodium and potassium when they are included on crystalline phases (like feldspars or pyroxenes). Conversely, magnesium and, in lesser part, calcium are preferentially leached by the differentiated rocks. Indeed, in this type of rocks, those elements are not found in mineral form, and are thus more hydrolysable.

In fact, our leaching experiment shows a differential signature of the water according to the rocks chemistry. Based on these first results, we can propose a new tool for determining the geological environments of the natural springs (preferentially not thermal) on Reunion Island.

These properties will be used to build a new nomenclature of the natural springs. Thus, by comparing the hydrochemical profiles of natural waters to the ones of experimental water. we can define which kind of rocks was altered and, consequently, which kind of rock is constituting the aquifer of each natural spring .

J11 : 4P/07 : PO

Source Rocks of REE+Y and Pb Isotopes of the Geothermal Water of Kizildere, Turkey

Marc A. Conrad (conrad@gfz-potsdam.de) &

Peter Möller (pmoe@gfz-potsdam.de)

¹ GRZ Potsdam, Am Telegrafenberg, 11473, Potsdam, Germany

REE and Pb in geothermal waters of Kizildere, Turkey, and rocks of the surroundings have been analysed. Additionally, the rocks have been leached at about pH₃ by which the easily accessible fraction of trace elements (Möller and Giese, 1997). These leaching experiments do not simulate natural alteration processes, but their results indicate whether or not soluble components released during water-rock interaction have been carried off. The comparison of the trends and the anomalies REE patterns of waters and the various rocks indicate that the trends of REE patterns favour the Sazak Formation (with low REE+Y contents and a small positive Eu anomaly) as the most probable source due to dissolution of carbonates, whereas the negative Eu anomaly of the water indicate that other sources must have contributed REE considerably. The most probable source could be the Igdecik Formation. Including information derived from the leaching solutions of rocks, it is evident that the Eu and Y anomalies of the water cannot be derived from the Sazak Formation. The resulting patterns might be described as the result of leaching of the underlying Igdecik Formation (mica schists, marbles and gneisses) and ion exchange with the marbles of the Sazak Formation. The Pb isotope ratios of the water point to the Igdecik Formation as the determining source. Because of the low and high Pb contents of the Sazak marbles and Igdecik schists the mixing of Pb from both sources is of minor significance. In conclusion, the Igdecik Formation seems to be the dominant source for Pb and REE+Y, whereas the Sazak Formation only influences the trend of REE by ion exchange. This is only possible if the interaction of mantle CO₂ already occurred with the marbles lenses included in the mica schists of the Igdecik Formation.

Möller P, Giese U, Chem. Geol, 137, 41-55, (1997).

J11 : 4P/08 : PO

Hydrogeochemistry of the Thermomineral Springs in Marmara Region, Northwestern Turkey

Rustem Pehlivan (pehlivan@istanbul.edu.tr) &

Osman Yilmaz

Dept.of Geological Engineering, University of Istanbul, 34850, Avcilar-istanbul, Turkey

This investigation was performed to determine hydrogeochemical features of thermomineral springs in Marmara region. For this purpose, hydrogeological features and geology of Marmara region were evaluated within regional geological situation. Results of the chemical analysis of the 24 thermomineral springs (thermal springs and hot springs) in Marmara region which are done in the previous years by us and others are given. In the investigation, analysis results of thermomineral springs are evaluated with respect to hydrogeochemstry view.

The cation and anion trends obtained from the diagrams show that thermomineral springs derivede from volcanic and asitic rock with limestone. Furthermore, thermomineral springss are mixed type waters according to their ions distribution Na+K>Ca+Mg and (CO₃+HCO₃)>SO₄+CI. Consequently, supported to the rock type and ion distribution of thermomineral springs.

J11 : 4P/09 : PO

Geochemical Investigation and Geology of the Surrounding of the Ozancik hot Water Spring, Canakkale, Turkey

Rustem Pehlivan (pehlivan@istanbul.edu.tr)

Dept. of Geological Engineering, University of Istanbul 34850, Avcilar-l.stanbul, TURKEY

The study area is located at 15 km southwest of Qan town. This investigation was performed to determine the geochemical properties of the Ozancik hot water spring. During the investigation, geological map of the hot water spring and its surrounding was prepared, and hot water and rock samples were collected from the study site. Paleogene -Neogene aged andesite, trachyandesite, andesitic tuff, silicified tuff and tuffites form the basement rocks. At the top of all these units Quaternary aged alluviums is taken place.

The results of the chemical analysis on the samples from the hot water spring indicated that the hot water richs in SO₄^2-, Cl^-, Na⁺, Ca²⁺, SiO₂ and SiO₂ of the rock samples. The distrubition of the ions in hot water on the Schoeller diagram has an arrangement of r(Na⁺+K⁺)> rCa²⁺> rMg²⁺ and r(SO₄^2-)> rCl^-> r(HCO₃^-).

The water analyses carried out following the ICPMS-200 technique were evaluated according to the World Health Organization (WHO) and Turkish Standards (TS). The evaluations revealed that utilization and the effects of the hot water on human health were determined.

J11 : 4P/10 : PO

The Composition of Non-Soluble Residues of Evaporite Sequences

Hans-Joachim Engelhardt (engelhardt@dbe.de)

DBE mbH, Dept. V-BGG, Rottlebener Weg 1, D-29475 Gorleben, Germany

In most cases the elemental composition of fine clastic sediments are close to the respective abundances of the upper continental crust. With the exception of increasing K and decreasing Si contents with diagenetic grad, diagenesis in mudrocks is largely isochemical with respect to the major elements. These observations are not in accordance with findings of non-soluble residues of evaporites ('salt clays'). They are affected in the hypersaline depositional environment by unusual physicochemical conditions (salinity, pH, Eh). The temperature of Zechstein salt rocks, recently lying in depths up to 1000 m, was, with local exceptions, lower than 90 ± 10 degrees Centigrade.

Geochemical investigations of samples from the Zechstein basin (cf. Brammer, 1992) show the following results: In comparison to palaeozoic shales there is a significant increase in the Mg-content and the K/Rb-ratio, whereas the Fe²⁺/Al-, Fe²⁺/K-, Fe²⁺/Fe³⁺- and the Mn/Fe³⁺-ratios are decreased. A comparison with recent North Sea deposits support these findings. The salt clays have higher Mg-contents, lower <epsilon> Fe₂O₃/K₂O- and MnO/<epsilon> Fe₂O₃-ratios. K is enriched in comparison to Rb. Samples of the Kupferschiefer have on average similar <epsilon> Fe₂O₃/K₂O-ratios and lower Mg concentrations than salt clays.

It is possible to correlate these findings with mineralogical results: North Sea deposits consist of illite, smectite, kaolinite, and chlorite (rhipidolithic composition). Illite and Fe-chlorite are the dominant clay minerals in Kupferschiefer deposits and common shales. However, in the non-soluble residues of evaporites illite co-exist with chlorite (pennine, clinochlore, sheridanite) and Mg-rich minerals, like koenenite, talc, and mixed-layer minerals. The occurrence of diagenetic corrensite is consistent with an interpretation of a precursor smectite component. The illite/chlorite ratio is extraordinary low (< 1).

A comparison with data extracted from the literature show similar trends of clay mineral assemblages in evaporite rocks of North America, the alpine 'Haselgebirge' (Austria), the Alsace (France), and the german 'Gipskeuper'.

The reactions have significant effects on brine composition: Thus, the precipitation of 1 kg clinochlore consume about 200 g Mg (=803 g MgCl₂). Additionally, saline solutions have high amounts of Mn, Fe, and Al, which could not be explained by reaction with other constituents of the salt formations. Fe amounts are partly flocculated, as hydroxides (high Eh), or as a constituent of FeCl₂ containing salts (low Eh) like Rinneit, Rokühnit, and/or Zirklerit.

The investigations demonstrate that concentrated salt brines react with alumino silicates and effect dissolution, recrystallization and precipitation of crystalline phases. In contrast to common shales, with illitization as the dominating process, especially chloritization and 'Mg-metasomatism' are controlling the composition of the non-soluble mineral fraction.

Brammer, K-J, Thesis, TU Clausthal, (1992).

J11 : 4P/11 : PO

Concentration Gradients in Single Minerals Caused by Open System Meteoric Water Infiltration: A Laser Ablation ICPMS Study

Klaus Simon (ksimon@gwdg.de)

Geochemisches Institut, Goldschmidtstr. 1, 37077 Göttingen, Germany

Granites from the Schwarzwald (Black Forest, SW Germany) are strongly altered by meteoric water interaction (Hoefs and Emmermann, 1983). Oxygen and hydrogen isotope compositions of whole rock and constituent minerals are depleted to values as low as -4 per mil and -110 per mil, respectively. Water-rock ratios are as high as 0.5 atom ratio (Simon, 1990). Due to heterogeneities in oxygen isotope composition of single minerals in hand specimen it is concluded that the alteration was channelized and has affected preferentially the rims of single minerals. Primary magmatic minerals are almost always to a great degree altered to secondary minerals like chlorite and albite. Balancing the chemical composition of the minerals of rocks altered to a varying degree, it becomes evident that some elements are mobilised and leached (Simon and Hoefs, 1987). This has been shown for Ca, Sr, Ba and Fe by Möller et al. (in press).

To get an idea of the process and the chemical effects of alteration the elemental ratios and concentrations in different parts (rim to core) of single minerals have been analysed with LA-ICPMS (spot diameter of 30 to 100 µ m, spot depth about 100 µ m; Simon et al., 1997). The resulting gradients and/or heterogeneities in trace elements like REE show the complex multi-stage (in time and space?) alteration processes in an open system the rocks have suffered.

Hoefs J, Emmermann R, Contrib Mineral Petrol, 83, 320-329, (1983).

Simon K, Contrib Mineral Petrol, 105, 177-196, (1990).

Simon K, Hoefs J, Chem Geol, 61, 253-261, (1987).

Möller P, Simon K, Giese U, Chemie der Erde, in press, (1998).

Simon K, Wiechert U, Hoefs J, Grote B, Fresenius J Anal Chem, 359, 458-461, (1997).

J11 : 4P/12 : PO

X-Ray Computerized Tomography Analysis of Fluid Repartition in Rocks During Drying

Jean-Pierre Sizun

(jean-pierre.sizun@univ-fcomte.fr)¹,

Bénédicte Tournier (tournier@illite.u-strasbg.fr)²,

Daniel Jeannette (dj@illite.u-strasbg.fr)² &

Gérardine Schramm²

¹ Dpt Géosciences, EA 2642 Contraintes, déformations, écoulements, Université de Franche-Comté, 16, route de Gray, 25000 Besançon, France

² Centre de Géochimie de la Surface, EOST, UMR 7517 CNRS-ULP, 1, rue Blessig, 67084 Strasbourg, France

X-ray tomography imaging is used to visualize porosity distribution and fluid repartition at various levels of water saturation in rocks during a controlled drying. We present results of a study performed on two rock types which differ by their petrographical nature and their petrophysical properties (total water porosity N_t; imbibable porosity N_48h; threshold pore access radius R_a,): a moderately cemented Fontainebleau sandstone (N_t: 10-12%; N_48h: 6-7%; R_a: 10 µ m) and a micritic limestone from Lourdines (N_t: 25-28%; N_48h: 23-24%; R_a: 0.20-0.35 µ m).

Gas saturation maps have been obtained on 18 cm long cylindrical plugs submitted to a drying by their upper surface (diameter: 10 cm). From these experimental results, we conclude that (1) water drainage is strongly controlled by pore structures; it starts along the borders and reaches inside the samples by higher porosity areas characterised by wide intergranular pore spaces and bigger pore throats; as the drying progresses, residual water profiles coincide with porosity profiles; (2) porosity heterogeneity can produce local breakdown in saturation profiles: water drainage is retarded in the subjacent part of the plug, while it strongly increases above, closeby evaporation surface.

Water drainage evolution during drying poses the problem of air migration which controls water displacement in porous media. The irregularity of water distribution in shallower part of the rock shows that no well defined and regular drainage front exists during drying of natural rocks.

J11 : 4P/13 : PO

Chromatographic Exchange of two Independent Chemical Components between an Aqueous Fluid and a Three-Component Solid Solution; Application to Evolution of Skarn Fe-Mg-Mn Clinopyroxenes and Ca-Mn-Fe Garnets

Abderrahim Sedqui (sedqui@emse.fr) &

Bernard Guy

Centre SPIN, Dép. Géochimie, ENSM-SE, 158 Cours Fauriel, F-42023 St-Etenne Cedex 2, France

The theory of metasomatism, as exemplified by chromatographic exchange, was set out by Korzhinskii (Korzhinskii, 1970) in the scalar case where one single component is exchanged between a solid and an interstitial fluid. In the present work, we have discussed the case with two independent components and applied it to the chemical exchange between an aqueous solution and a three-component solid solution.

The first step is to derive the isotherm function ruling the equilibrium between the fluid and the solid. This may be obtained by writing the equality of the chemical potentials of each component in the fluid and in the solid; the thermodynamic properties of the solid and of the liquid solutions are needed. Because the equilibrium data are often known in terms of dissociation constants for end members of solid solutions in equilibrium with bulk solution, we prefered to use these data rather than those on the properties of each species in solution separately. The important is the ratios of the the dissociation constants. At high temperature, the isotherms read like Langmuir isotherms.

The second step is the writing of the mass balance of the chemical components between the solid and the fluid during its movement within the porosity of the rock. Taking the hypothesis of local equilibrium with an isovolumic exchange, and the isotherm data obtained in the first step, we obtain an hyperbolic system of chromatographic type. This was studied numerically. The influence of the solid solution parameters, of the respective magnitudes of the three dissociation constants were studied. Evolution paths in composition triangles were drawn depending on the initial conditions (starting rock composition and inlet fluid). Depending on the cases «détentes», i.e. smooth spatial composition variations, shocks (sharp fronts) or combination of both may obtain. The results of the numerical simulation were compared with data on natural spatial evolutions of compositions of skarn minerals.

Korzhinskii, Theory of Metasomatic Zoning, Clarendon Press, Oxford, 162 p, (1970).

J11 : 4P/14 : PO

Origin of Deep Thermal Water in Relation with Structural Features Along the Middle Atlas (Morocco)

Anne Winckel (winckel@geol.u-psud.fr)¹,

Jean-Luc Morel (jlmorel@geol.u-psud.fr)²,

Laurent Dever (dever@geol.u-psud.fr)¹ &

Christelle Marlin (marlin@geol.u-psud.fr)¹

¹ L.H.G.I.- Université Paris-Sud, Bâtiment 504, 91405 Orsay Cedex, France

² Laboratoire Tectonique et Bassins, Bâtiment 504-Université Paris-Sud, 91405 Orsay Cedex, France

The tectonic evolution of Rif Belt (Morocco) and its foreland results of both the effects of the convergence of the Africa toward Eurasia and the south-westward drift of the Alboran block. The drift causes a ENE-WSW shortening while the convergence induces a NS to NW-SE shortening. These movements have generated a network of faults; some of them are associated with the overlap and are schematically oriented NE-SW (Nekor direction) and the others, oriented NS, cut again the structures and so are more recent. In this geological context, numerous thermal springs and boreholes are found both in the Rif and the Middle Atlas (transverse Rabat-Fes-Nador). They are all located at the east of the Nekor, whatever is the structural zone. The emergence temperature of this groundwater largely varies between 19.5 and 50.1°C for a wide range of mineralisation (electrical conductivity from 0.8 and 41.5 mS.cm^-1). According to the mineralisation, the chemical type of groundwater evolves between two end-members: Ca-HCO₃ and Na-Cl. Moreover, some of the groundwaters are characterised by high concentrations of SO₄ and Ca. These variations of chemical types are highly dependent of the dissolution of Triassic evaporites made of halite, gypsum and locally sylvite. The chemical equilibrium between liquid and solid phases could be used as chemical geothermometer, this latter being validated by the isotopic geothermometer.

The groundwaters are characterised by oxygen-18 contents between -8.7 and -4.5 per mil and deuterium contents between -53 and -25 per mil vs SMOW. The depletion in heavy isotopes observed in groundwater with respect to local rainfall ( ¹⁸O=-3.5 per mil and  ²H=-20 per mil) is explained by both an altitude effect and palaeoclimatic effect. The position of analytical points along the World Meteoric Line ( ²H = 8* ¹⁸O+10) indicates that the recharge to the deep groundwater systems have occurred through fracturated zones and have removed rapidly water from the evaporation layers. The preliminary results of carbon-14 activity display the presence of groundwater recharge infiltrated during the Late Quaternary. The results of the chemical and isotopic compositions indicate that the springs may come from a single hydrothermal province without any direct relationship with the structural countries.

J11 : 4P/15 : PO

Hydrothermal Alteration and Kaolinization in the Northwestern Border of Ossa Morena Shear Zone

Celso Gomes (cgomes@geo.ua.pt) &

Iuliu Bobos

University of Aveiro, Department of Geosciences, Aveiro-3810, Portugal

The northwestern border of the Ossa Morena zone consists of metamorphic rocks (micaceous schist, gneiss, migmatite-gneiss and tonalite gneiss, porphyroblast schist, amphybolite etc) attributed to Middle and Late Proterozoic age. The metamorphic evolution of the region is represented by three or four phases of Hercynian deformation. Four lithostratigraphic units had been defined in the northwestern border of the Ossa Morena zone. In the present contribution an attention is paid to the Lourosa unit sub-divided in two members: lower member (migmatite, orthogneiss) and upper member (garnet micaceous schist). The shear zone is characterised by the development of an intense foliation in migmatite rocks, and the largest zone is coincident with the elongate alteration system hosting the advanced argillic alteration. Migmatite complex can be considered as an external marker of the shear zone. The migmatite complex shows evidence of the passage of large volumes of fluid. Introduction of great quantities of water into major structural fractures (shear zone) is a major cause of the widespread alteration at a regional scale. The source of metasomatic water in the shear zone may have been related to the migmatization process. Migmatite complex hosts a relict unmineralised greisen alteration (quartz-muscovite bearing F, Cl and quartz-tourmaline) and an advanced argillic alteration (well ordered kaolinite ± illite and halloysite-7Å assemblages). Late stage of hydrothermal alteration is characterised by quartz + tourmaline veins which coexist with the advanced argillic alteration. Advanced argillic alteration is characterised by two stages: a post-greisen stage, recognised in São Vicente de Pereira area, by mica to kaolinite transformation and a second stage superimposed on the first related to hydrothermal event which produced a discharge of metamorphic water on a large scale in the migmatite complex. Large accumulations in situ of kaolinite and halloysite-7Å are found in the area, typical of regions characterised by intense fracturing and shearing, where rapid discharge of metamorphic waters did take place. The alteration products were investigated through X-ray diffraction, X-ray fluorescence, by electron microscopy.  ¹⁸O and  Dcarried out on kaolinite attest the contribution of metamorphic waters for the advanced argillic alteration. Typology of the fluid inclusions in the quartz veins was determined. A moderate to low salinity (eq. NaCl) was identified.

J11 : 4P/16 : PO

Computational Simulation of Hydrothermal Dissolution and Precepitation of Quartz

Hans Herrmann (hans.herrmann@bgr.de)¹,

C. Clauser (c.clauser@bgr.de)¹,

H. J. Neugebauer (neugb@geo.uni-bonn.de)² &

U. Hein (ufhein@hotmail.com)³

¹ PO Box 51 01 53, 30631 Hannover, Germany

² Nussallee 8, 53115 Bonn, Germany

³ Jhering Str. 6, 37081 Goettingen, Germany

Understanding hydrothermal processes requires information on the transient species mass balance. Hydrothermal dissolution and precepitation of quartz is described physically and simulated numerically. Numerical code is verified by comparing the numerical results with an analytical solution of a reactive flow as well as with laboratory experiments. The analysis of several scenarios of the growth of quartz veins show that vein growth occurs mainly between 200°C to 330°C. Our results agree well with field observations (Yardley, 1983) which predict a fast sealing of cracks.

Yardley, BWD, J. Geol. Soc. London, 140, 657-663, (1983).

J11 : 4P/17 : PO

Permeability and Diffusivities of Fracture Skins and a Weathering Surface in a Quartz Arenite: The Keweenawan (Upper Precambrian) Hinckley Sandstone, Minnesota

Lorenzo Marchesini (lmarche@geomin.unibo.it)

Dip. Scienze della terra e geol. amb., V. zamboni 67, 40127 Bologna, Italy

Permeabilities and diffusion coefficients were measured on surfaces (fracture skins and a varnished surface) and the unaltered rock matrix of the Upper Keweenawan (Late Precambrian) Hinckley Sandstone. The study was undertaken in order to quantitatively assess heterogeneities between rock matrix and its fracture surfaces, which represents the interface between fracture and matrix flows. The bulk sandstone revealed an average composition of Q₉₅, F₄, R₁ by petrographic analysis that classifies it as a quartz arenite of great compositional maturity. Pore spaces were little affected by compaction and the silica cement was volumetrically unimportant. Two types of surfaces were observed. The first is characterized by superficial and intergranular precipitation of iron and manganese oxides; the second contains infiltered clay in addition to oxides. In both types, thin section observations showed decreasing presence of precipitates with increasing distance from the surface until no intergranular material is found at a distance of 1 cm from the fracture surface. Permeabilities were analyzed with a gas permeameter probe (EFP) for surface measurements. A high number of single measurements were collected by dividing each surface in squared grids of 1 cm spacing intervals. Results indicate that surficial permeabilities were reduced of a factor 4 with respect to unaltered rock as a result of the pore occlusion by the precipitates. Diffusion coefficients were measured experimentally and calculated from Fick's Law. Diffusivities of both fracure skin and the weathering surface were approximately one order of magnitude lesser than the unaltered rock matrix as a result of pore-filling precipitates. Decreasing porosity with increasing tortuosity (i.e. deviation of the flow paths from idealized flow lines) and a molecular sieving or membrane effect were the principal factors responsible for the changes in diffusivity. Fracture skins are an integral and ubiquitous part of fractured geologic systems with important consequences in processes such as mass exchange and solute diffusion between the fractures and the rock matrix and contaminant transport through fracture flows. Detection of different and definite hydraulic properties for high porosity-high permeability, compositionally mature sandstone confirms the importance of considering and quantifying fracture properties in numerical and stochastic modeling of dual-porosity hydrologic media.