BRGM, Orléans, France 2 CFG Services, Orléans, France.

Similar documents
Main Results of a Long-Term Monitoring of the Bouillante Geothermal Reservoir During Its Exploitation

Procedia Earth and Planetary Science 13 ( 2015 ) 25 29

GEOTHERMAL EXPLORATION IN MARTINIQUE BASED ON MAGNETO-TELLURIC AND CONTROLLED-SOURCE EM SURVEY

Assessment of the Bouillante Geothermal Field (Guadeloupe, French West Indies): Toward a Conceptual Model of the High Temperature Geothermal System

Resistivity Imaging for Geothermal Exploration, using Controlled-Source EM where Magneto-Telluric is Not Applicable: Model and Field Study

Geothermal exploration in the Mount Pelée volcano-morne Rouge and Diamant areas (Martinique, West French Indies) : Geochemical data

In other active geothermal systems

Fluid Geochemistry at the Nir Geothermal Field, Nw-Iran

Geochemical Characteristics of Reservoir Fluid from NW-Sabalan Geothermal Field, Iran

Bouillante geothermal field: mixing and water/rock interaction processes at 250C.

Geophysical Methods Applied to the Assessment of the Bouillante Geothermal Field (Guadeloupe, French West Indies)

Lithium isotopes in island arc geothermal systems: Guadeloupe, Martinique (French West Indies) and experimental approach

GEOTHERMAL POTENTIAL OF ST. KITTS AND NEVIS ISLANDS

GEOTHERMAL DEVELOPMENT IN THE COMOROS AND RESULTS OF GEOTHERMAL SURFACE EXPLORATION

THE FLUID CHARACTERISTICS OF THREE EXPLORATION WELLS DRILLED AT OLKARIA-DOMES FIELD, KENYA

GEOTHERMAL ENERGY POTENTIAL FOR LONGONOT PROSPECT, KENYA. By Mariita N. O. Kenya Electricity Generating Company

The Initial-State Geochemistry as a Baseline for Geochemical Monitoring at Ulubelu Geothermal Field, Indonesia

Geochemical Modelling of Low-Temperature Geothermal Fields from Bihor County, Romania

SUBSURFACE HYDROTHERMAL ALTERATION AT THE ULUBELU GEOTHERMAL FIELD, LAMPUNG, SOUTHERN SUMATRA, INDONESIA. Suharno 1, 2 and PRL Browne 2

SUB-SURFACE GEOLOGY AND HYDROTHERMAL ALTERATION OF WELLS LA-9D AND LA-10D OF ALUTO LANGANO GEOTHERMAL FIELD, ETHIOPIA

GEOCHEMISTRY OF RWENZORI HOT SPRINGS. Vincent Kato Department of Geological Survey and Mines, Entebbe, Uganda

Structural Controls on the Chemistry and Output of the Wells in the Olkaria Geothermal Field, Kenya

WAMUNYU EDWARD MUREITHI I13/2358/2007

Visualizing Earth Science. Chapter Overview. Volcanoes and Eruption Types. By Z. Merali and B. F. Skinner. Chapter 9 Volcanism and Other

Structural Geology tectonics, volcanology and geothermal activity. Kristján Saemundsson ÍSOR Iceland GeoSurvey

Clay minerals : paleo-conditions and dynamic evolution of hydrothermal systems. P. Patrier Mas

Heat (& Mass) Transfer. conceptual models of heat transfer. large scale controls on fluid movement. distribution of vapor-saturated conditions

HIGH TEMPERATURE HYDROTHERMAL ALTERATION IN ACTIVE GEOTHERMAL SYSTEMS A CASE STUDY OF OLKARIA DOMES

Hydrothermal Chemistry/ Reverse Weathering. Marine Chemistry Seminar

Geochemical Modeling of Acidic Geothermal Fluids using SOLVEQ and CHIM-xpt

predictive mineral discovery*cooperative Research Centre A legacy for mineral exploration science Mineral Systems Q3 Fluid reservoirs

The Petite Anse-Diamant Geothermal System (Martinique Island, Lesser Antilles): Results of the Exploration

Geochemical monitoring of the response ofgeothermal reservoirs to production load examples from Krafla, Iceland

David I. Norman and Joseph N. Moore

Geothermometer, Geoindicator and Isotope Monitoring in Lahendong Wells during

3D Fault Model of the Bouillante Geothermal Province Interpreted from Onshore and Offshore Structural Knowledge (French West Indies)

Geophysical surveys Anomaly maps 2D modeling Discussion Conclusion

Geophysical Surveys of The Geothermal System of The Lakes District Rift, Ethiopia

RECENT RESULTS FROM THE SAN JACINTO - TIZATE GEOTHERMAL FIELD, NICARAGUA

A bowl shaped depression formed by the collapse of a volcano is called a. Magma that has left the vent of a volcano is known as. Lava.

Magneto-telluric (MT) surveys in a challenging environment at Lihir Island, PNG

Geothermal Model of the Lahendong Geothermal Field, Indonesia

BIBLIOGRAPHIC REFERENCE

Geochemical Study of Ampallas Geothermal Area, Mamuju District, West Sulawesi Province

EVALUATING HEAT FLOW AS A TOOL FOR ASSESSING GEOTHERMAL RESOURCES

Kizito Maloba Opondo. Kenya Electricity Generating Company

ASSESSMENT OF GEOTHERMAL RESOURCES AT THE AL-LITH AREA, KINGDOM OF SAUDI ARABIA

Itisanexcellentexampleofanactivehotspringssystem. Most pictures shown were taken on a PACRIM field trip in 1995(unless otherwise referenced).

Key Words: geothermal, exploration, extraordinary low water level, pressure imbalance, hidden vapor dominated reservoir ABSTRACT

Mount Spurr geothermal workshop August 27 28, 2007

Japan Engineering Consultants, Inc., Energy and Industrial Technology Development Organization,Tokyo, Japan

Graduate School of Science and Engineering, Kagoshima University

Interpretation of Subsurface Geological Structure of Massepe Geothermal Area Using Resistivity Data

Overview of Indonesian Geothermal System

The Role of Magnetotellurics in Geothermal Exploration

GEOCHEMICAL CHARACTERISTICS OF ACID FLUIDS IN MT PINATUBO, PHILIPPINES

THE GEOTHERMAL POTENTIAL OF BILIRAN ISLAND, PHILIPPINES

Effective geochemical methods for identifying geothermal systems in the western branch of the EARS

Department of Civil & Geological Engineering GEOE Engineering Geology ASSIGNMENT #2

Quantification of the 3D thermal anomaly in the orogenic geothermal system at Grimsel Pass

CHEMICAL GEOTHERMOMETERS FOR GEOTHERMAL EXPLORATION

GEOLOGY MEDIA SUITE Chapter 12

FLUID-MINERAL EQUILIBRIA AND SUBSURFACE TEMPERATURE EVALUATION OF SELECTED HOT SPRINGS, JIANGXI PROVINCE, SE-CHINA

Anna Yushantarti Center For Mineral, Coal, and Geothermal Resources, MEMR Jl Soekarno Hatta 444, Bandung, Indonesia

Geothermal Exploration on the Island of Montserrat, Caribbean

Evaluation of Subsurface Structures Using Hydrothermal Alteration Mineralogy A Case Study of Olkaria South East Field

Geology and Hydrothermal Alteration of the Hverahlid HT-System, SW-Iceland

Geo-scientific Data Integration to Evaluate Geothermal Potential Using GIS (A Case for Korosi-Chepchuk Geothermal Prospects, Kenya)

Exploration of Geothermal High Enthalpy Resources using Magnetotellurics an Example from Chile

TORFAJÖKULL, ICELAND A RHYOLITE VOLCANO AND ITS GEOTHERMAL RESOURCE

Current status of the high enthalpy conventional geothermal fields in Europe and the potential perspectives for their exploitation in terms of EGS

INVESTIGATION OF MINERAL ALTERATION IN ANDESITE AND DACITE FROM THREE DIFFERENT VOLCANO HYDROTHERMAL SYSTEMS ON DOMINICA, LESSER ANTILLES

Mt St Helens was know to have entered into active periods that lasted from years once every years over the last 500 years, (Figure 5).

CAMBRIAN INTRUSION-RELATED COPPER MINERALISATION AT THE THOMAS CREEK PROSPECT, SOUTHWESTERN TASMANIA

GEOTHERMAL STRUCTURE AND FEATURE OF SULFIDE MINERALS OF THE MATALOKO GEOTHERMAL FIELD, FLORES ISLAND, INDONESIA

Volcanic Eruptions and Hydrovolcanism

AN OVERVIEW OF THE GEOTHERMAL POTENTIAL OF PAPUA NEW GUINEA

From Punchbowl to Panum: Long Valley Volcanism and the Mono-Inyo Crater Chain

Imagine the first rock and the cycles that it has been through.

Foundations of Earth Science, 6e Lutgens, Tarbuck, & Tasa

** *Poitiers Univ.,HYDRASA Lab., 40 Av. recteur Pineau, Poitiers cedex, France

Summary of geothermal setting in Geo-Caraïbes target islands

Preliminary results of deep geothermal drilling and testing on the Island of Montserrat

Geothermal Potential Assessment in Northern Rwanda

A - Piton de la Fournaise activity

Geochemical and Hydrological Assessment of Thermal Features in the Vicinity of the Champagne Pool in the Waiotapu Geothermal System

Engineering Geology ECIV 2204

Geothermal Exploration in Eritrea

Monitoring of the Bouillante Geothermal Exploitation (Guadeloupe, French West Indies) and the Impact on Its Immediate Environment

THE CHEMICAL CHARACTERISTICS OF GEOTHERMAL FLUIDS IN JIAODONG PENINSULA, SHANDONG, CHINA

NOTICE CONCERNING COPYRIGHT RESTRICTIONS

The importance of understanding coupled processes in geothermal reservoirs. Thomas Driesner October 19, 2016

Magmatism in Western Cascades Arc. Early Tertiary Magmatism Part II. Washington Magmatism. Western Oregon. Southern Oregon

UGRC 144 Science and Technology in Our Lives/Geohazards

NUMERICAL MODELING STUDY OF SIBAYAK GEOTHERMAL RESERVOIR, NORTH SUMATRA, INDONESIA

Characterization of Subsurface Permeability of the Olkaria East Geothermal Field

TECHNICAL REVIEW OF GEOTHERMAL POTENTIAL OF NGOZI AND SONGWE GEOTHERMAL PROSPECTS, TANZANIA.

3D Numerical model of the Obama hydrothermal geothermal system, Southwestern Japan

HYDROTHERMAL ALTERATION IN THE SUNAGOHARA FORMATION, OKUAIZU GEOTHERMAL SYSTEM, JAPAN

Insights to High Temperature Geothermal Systems (New Zealand) from Trace Metal Aqueous Chemistry

Transcription:

GRC Transactions, Vol. 38, 2014 Toward a Continuum Geothermal Model to Explain Coexistence of Medium to High (100 to 250 C) Temperature Geothermal Systems in Martinique and Guadeloupe, French West Indies V. Bouchot 1, A. Gadalia 1, H. Traineau 2, and S. Caritg 1 1 BRGM, Orléans, France 2 CFG Services, Orléans, France v.bouchot@brgm.fr Keywords Martinique, Guadeloupe, Lesser Antilles, Bouillante, Montagne Pelée, high to medium temperature geothermal systems, continuum geothermal model, evolution in space and time, magmatic heat source, exploration guidelines, fluid equilibrium Abstract The recent surface exploration data acquired in Martinique led to a revisit of the conceptual models of three geothermal systems in active volcanic arc context: Petite Anse in the Southwest of the island, Lamentin in the center, Montagne Pelée in the North. These three explored systems are compared with each other and with the developed Bouillante geothermal field in Guadeloupe (15 MWe). From parameters, such as reservoir temperature, duration of the geothermal system using fossils events, age and duration of magmatism, degree of equilibrium of geothermal fluid, we propose a continuum model of geothermal system including three successive stages: prograde, peak and retrograde. (Bouchot et al., 2010). So while Bouillante is a classical system of high temperature (250 C), according to geothermometers it appears that the geothermal systems in Martinique show lower temperature reservoirs between 100 C and 220 C. These different reservoir temperatures coupled with their variations in time pose several questions. Do these temperature differences indicate different stages of the same type of geothermal development (with successive prograde, peak or retrograde stages)? What impacts can the age, position, volume and duration of the magmatic activity as source of heat, have on the evolution of the geothermal system (duration, extension) and the temperature of its reservoir? The comparison of different systems, based on selected parameters such as reservoir temperature, duration of the geothermal system using fossils events, age and duration of magmatism, degree of equilibrium of geothermal fluid, should provide relevant answers to these questions. This comparative synthesis should thus allow a better understanding of the temporal and spatial evolution of these systems and contribute to better exploration of them in the Lesser Antilles island volcanic arc (Table 1). A geothermal system associated with an active volcanic arc is, a priori, controlled by a magmatic source of heat. It is Introduction Recent surface exploration data acquired in Martinique, French West Indies, between 2012 and 2014, led to revisiting the conceptual models of three geothermal systems in this active volcanic arc setting: Petite Anse in the southwest of the island, Lamentin (close to Pitons du Carbet) in the centre, and Montagne Pelée in the north (Gadalia et al., 2014) (Figure 1). The three systems being explored in Martinique are compared with each other and with the Bouillante geothermal field (Guadeloupe, France) (Figure 1), the only exploited geothermal field in the Caribbean, with an installed capacity of 15 MWe Figure 1. Location of the four studied geothermal fields (red star), in Guadeloupe and in Martinique (French Lesser Antilles). 363

characterized by i) a geothermal reservoir resulting from hydrothermal convective circulation, ii) leaks from the reservoir to the surface and associated surface expressions (thermo-mineral springs, fumarole) and iii) recharge of the system involving, as on islands in the Lesser Antilles, meteoric and marine waters. Geothermal Fluids and Reservoir Temperatures In Martinique and Guadeloupe, geothermal reservoirs show little variations in terms of chemical composition of geothermal waters. They are usually close to the pole chlorinated sodium mainly because of the strong influence of seawater in the reservoir recharge (Figure 2). In fact, the involvement of seawater with a salinity around 35 g/l, complicates the interpretation of geothermal fluids in terms of water maturity. To do this, it favors the use of isotopic analysis of fluids (O, C), binary diagrams (e.g. Cl/Br), and the ternary diagram [Na-K-Mg] rather than the Table 1. Comparison between the geothermal systems of Martinique and Guadeloupe French islands. Geothermal Field Montagne Pelée Bouillante Petite Anse Lamentin-Carbet Stage of the geothermal evolution Duration of the magmatic activity in the region Latest volcanic activity recorded in the field 3 He/ 4 He signature (from hot spring) Temperature of the reservoir (geothermometer or measured) Fossil geothermal stage Duration of the geothermal system Current fluid composition with ph and TDS Degree of fluidrock interaction Recharge of the geothermal system Spatial extension of the geothermal activity Prograde stage 125 ky (550 ky if Mont- Conil included) Active volcanism (1929, last event) Peak stage (with plateau) 1 Ma (between 1.2 My and 230 ky ago) Retrograde stage (trend) ~1.1 Ma (between 1,5 My and 350 ky ago) 364 diagram [Cl-HCO 3 -SO 4 ] (Giggenbach, 1988) whose concept of mature waters is inappropriate because of sea water influence (Figure 2). In Lamentin, current geothermal water Advanced retrograde stage 700 ky (between 1 Ma and 337 ky ago) ~500 ky ago ~350 ky ago > 600 ky ago 7.2 Ra 4-4.7 Ra 7.75-7.93 Ra 2.2 Ra 180-200 C (R2) 200-220 C (R1) (estimated) No identify ~ 100 ky (from 125 ky ago to today) Na-HCO 3 to Na-Cl 5.6 <ph< 6.2 0.4 to 1.4 g/l Immature waters only 2% of sea water Supposed expansion in the future 250 C (measured) Adularia and quartz breccia (250 ± 50 ky by 40 Ar- 39 Ar) 190-210 C (estimated) Palaeo-caprock and palaeo-acidic alteration (fumarole) with evidence of boiling (undated) 100-140 C (estimated) 90 C (measured in well) Illitic alteration in depth and palaeosilica sinters (300 to 250 ky by U/Pb) > 350 ky ~200 ky >> 500 ky Na-Cl ph~5.3 20 g/l Na-Cl 6.3 <ph< 7 20g/l Na-Cl 5.6 <ph< 6.2 10-11 g/l Full equilibrium Partial equilibrium Partial equilibrium sea water (58%) fresh water (42%) Spatial stabilization of the geothermal since 300 ky Figure 2. Position of the different geothermal waters (from Martinique and Guadeloupe geothermal fields) in the Na-K-Mg diagram (left) and Cl-HCO 3 -S0 4 diagram (right). sea water (54%) fresh water (46%) Contraction of the system (in progress) sea water (30-35%) fresh water (65-70%) Contraction of the system since 250 ky (from 10 to 1 km²) from hot springs and wells are of sodiumchloride type, at ph <5.6 to 6.2 with a salinity of 10-11 g/l, resulting from a mixture of 30-35% sea water and 65-70% fresh water. According to the ternary diagram [Na-K-Mg] (Figure 2), theses sodiumchloride waters are in the field of partial equilibrium close to immature waters. In Montagne Pelée, two reservoirs are distinguished: a first reservoir, located on the SW flank of Pelée stratovolcano, is of sodium-chloride bicarbonate-type with an outflow feeding the springs of Rivière Picodo (0.4 g/l, ph 7.7-8.4), while a second reservoir, located in the apex area is composed of bicarbonate fluids with an outlet at the springs of Rivière Chaude (1.4 g/l, ph 6.4-6.8), and which appeared after the 1902 eruption. The proportion of seawater in the Picodo reservoir is estimated around 2% (Gadalia et al. 2014). Unlike conventional magmatic-hydrothermal model associated with an active stratovolcano, the summit reservoir (Rivière chaude), whose waters probably borrow the eruptive conduit, is not acid. Bicarbonate character reflects a neutralization of the water. According to the ternary diagram [Na-K-Mg] (Figure 2), theses geothermal waters (Rivière Picodo and Rivière Chaude) are clearly in the field of immature waters. In Petite Anse, Eaux Ferrées springs are of sodium-chloride water type, marked

by a salinity of 20 g/l at ph 7 to 6.3. The reservoir waters could result from a mixture of sea water (54%) and freshwater (46%), as in Bouillante. According to the Br/C diagram, a large excess of chloride (34%) could have a magmatic origin (Gadalia et al., 2014). According to the ternary diagram Na-K-Mg, the geothermal water is partially equilibrated (Figure 2) but close to immature waters field. In addition, high concentrations of certain trace elements (As, B, Cs, Li, Rb,...) and the isotopic ratios of 18 O water indicates a water-rock interaction at high temperature. In Bouillante, the geothermal fluid, taken from the reservoir or reconstituted from hot springs, is homogeneous in its composition. It corresponds to sodium chloride brine with a salinity approaching 20 g/l and a ph of 5.3 ± 0.3 at 250 C. Its chemical and isotopic compositions indicate that the reservoir is supplied both by sea water (58%) and fresh surface water (42%), the mixture of which is at chemical equilibrium with the host rock at a temperature of 250 C (Sanjuan et al., 2001). According to the ternary diagram [Na-K-Mg] (Figure 2), theses sodium-chloride waters reach the full equilibrium field. Geothermometers calculated from hot spring and wells waters of the four studied fields show strong variations in temperature of the geothermal reservoirs, ranging from 100 to 250 C. In Lamentin, the current reservoir is a low temperature type (100-140 C) according to geothermometers from Lamentin hot water (90 C) (Figure 3), while fossil reservoir known in Lamentin 250 ky is of high temperature type (>200 C). In Montagne Pelée, where at least two high temperature reservoir have been characterized (Gadalia et al., 2014), the chloride-bicarbonate reservoir (related to Rivière Picodo) would have an estimated temperature of 155-180 C while the bicarbonate reservoir (in relationship with Rivière Chaude spring) would be a little warmer around 180-200 C. In Petite Anse, the reservoir is a high temperature type, between 190-210 C (Sanjuan et al., 2003), according to the geothermometers obtained from the Eaux Ferrées springs (close to Petite Anse village). In Bouillante, the measured temperature in the developed reservoir, up to 250 C, is the highest of the four study geothermal fields. In terms of fluid equilibrium, it appears that the dominantly chloride-sodium type reservoir waters, characterized by a strong recharge of sea water (except for Montagne Pelée) show: i) a full equilibrium for Bouillante reservoir fluid (250 C), ii) only partial equilibrium for Petite Anse and Lamentin hot spring (close to immature field), and, ii) clearly immature waters for Montagne Pelée (Rivière Picodo and Rivière Chaude). Early Hydrothermal Stages as Fossil Witnesses of the Geothermal Systems Identification of early fossil stages of the geothermal systems can make a relevant contribution to explain their evolution in time and space (Figure 3): i) determining how long they are active, ii) estimating their thermal temporal evolution (cooling, heating or thermal peak?) and iii) assessing if their spatial extent varies with time. In Lamentin (Figure 3), prominent fossil silica sinter as high temperature occurrences, emplaced 250-300 ky ago (dated by U/ Th, in Chovelon et al, 1984) and chlorite-epidote and illite-i/s assemblages identified in wells (> 200 C) in total imbalance with the temperatures (80-90 C) of the current fluid, testify to a high temperature palaeo-system active 300 ky ago in the plain of Lamentin (Mas et al. 2003). In addition, in the Morne Rouge area, a large siliceous hydrothermal breccia (explosion breccia type), associated with silica sinter, suggests that the HT palaeo-system came into sharp boiling 300 ky ago. Later, temperatures inferior to 200 C, expressing the beginning collapse of the palaeo-system, has been suspected due to the presence of laumontite and corrensite in the wells (Mas et al. 2003). Today, kaolinite and smectite assemblage is stable with geothermal water (80-90 C). From a spatial point of view, the witnesses of the high temperature palaeo-system (silica sinters, explosion breccia, hydrothermal alteration and related arsenic anomaly) are distributed over a wide area of approximately 10 km² (at the surface), while the current water flow at 90 C and related kaolinite-smectite alteration (in wells) and conductive signature (CSEM method, Gadalia et al., 2014) are very limited in space (~ 1 km² at the surface). Therefore, assuming that both geothermal stages (fossil and current) are connected to the same geothermal system and comparing the current low temperature reservoir to the fossil high temperature reservoir, it can be seen that, between 250 ky and today, the Lamentin geothermal system gradually cooled (from ~ 250 C to 100 C in the reservoir) while contracting in space (from 10 km² to 1 km ² according to surficial expressions). In Petite Anse (Figure 3), the warm springs of Eaux Ferrées, located on the coast, are hosted in an intense argillic alteration with pyrite-montmorillonite smectite dominant (indicating ~ 100 C), which forms the caprock of a fossil geothermal system (Traineau et al., 2013). The scarce presence of interstratified illite-smectite and illite (> 200 C) in this palaeo-caprock is interpreted as the result of warmer fluid upflow coming from the underlying geothermal fossil reservoir. Above this caprock (Figure 3), a typical kaolinite-alunite palaeo-alteration was identified, on the relief, in several sites distributed along a 2 km long NNW-SSE corridor between Petite Anse and Anse d Arlet. This acid alteration is symptomatic of a palaeo-fumarolic activity, likely related to an early boiling event of a larger and / or shallower reservoir than the current one. These undated palaeoalterations (neutral caprock on the sea level and acidic alteration on the relief) are the witnesses of an intense high temperature geothermal palaeo-system, whose the caprock is today at the surface, because of the flank collapse of the Morne Jacqueline volcano (Traineau et al., 2013) (Figure 3). Therefore in Petite Anse, we suggest that the geothermal system could begin several hundreds of thousands of years ago (300-500 ky?) with an intense geothermal activity recognized on several kilometers (under boiling conditions), whereas the current activity seems to have decreased in intensity (and perhaps slightly in temperature 190-210 C) and is contracted around Morne Jacqueline volcano. We interpreted this evolution as a retrograde trend of the system. However, strong magmatic signature of the 3 He/ 4 He ratio (7.3-7.9 Ra according to Pedroni et al., 1999 and Gadalia et al., 2014) suggests a continuation (or renewal) of magmatic activity in the region despite the absence of volcano less than 350 ky, which could eventually maintain the current geothermal system at high temperature (~ 200 C). 365

Figure 3. Surficial expressions as witness of fossil and active geothermal (hot springs) stages of the Lamentin (left) and Petite Anse (right) fields (Martinique). In Bouillante, the high temperature geothermal system (250 C) is existing for at least 300 ky as indicated by an adularia-silica bearing breccia, partially eroded, and visible in outcrop above the current reservoir (Patrier et al., 2013). Recently dated at 250 ± 50 ky by 40 Ar/ 39 Ar on adularia (Vérati et al., 2014), this explosion breccia contains mainly adularia emplaced at >200 C, which is compatible with the current reservoir temperature. Thus, since 250-300 ky, the Bouillante geothermal system has maintained at high temperature (~250 C) without modification of its spatial extension, centered on the Bouillante Bay. This system appears particularly stable during a long period. In Montagne Pelée, geothermal activity appears at most as old as the building of Pelée stratovolcano, which began 127 ky ago (Germa et al., 2011; Boudon et al., 2013). Thus, the stratovolcano is the main host rock of the geothermal reservoirs, located between 1000 and 2000 m in depth. Comparing the spatial and temporal evolution of the four studied geothermal systems, it appears that the system of Bouillante remains stable in temperature (250 C) and in space during 300 ky, while the geothermal system of Petite Anse is marked by a spatial contraction of its reservoir and a reduction of its hydrothermal activity, which could be combined with a slight decrease of the temperature around 200 C. More advanced is the retrograde thermal evolution of the system Lamentin which gradually cools since 250 ky to reach 100 C today. The system of Montagne Pelée, as a still immature geothermal system, is likely to show a prograde evolution in the future (next 100 ky?). Magmatism as a Heat Source for Geothermal Systems Recent magmatic activity has been recorded in Martinique (Westercamp et al., 1990; Germa et al., 2011) and is thought to be providing the required heat sources for the development of geothermal systems. Relationship between magmatic activity and hydrothermal systems can be tentatively analyzed even if ages are not well-constrained. Petite Anse. The area of active and fossil hydrothermal activity is located within a volcano-tectonic corridor extending from Pointe 366 Burgos to Diamant and characterized with a 1.1 My-long magmatic activity ending 350 ky ago with the eruption of Morne Larcher volcano (Figure 3). It frequently exhibits evidences of mixing between basaltic magmas and differentiated, quartz-bearing dacitic magmas. The latter is thought to be evolving within shallow magmatic chambers at around 6 km depth according to Gourgaud and Gerbe (1991). A MT survey evidenced a 5-km long, resistive and high-density body striking north-south which could be viewed as a large dyke-like magmatic intrusion (Gadalia et al., 2014). Its top is at a depth of 500 m. A dyke of dioritic composition outcropping close to the Eaux Ferrées hot springs is probably related. Such a large and shallow intrusive body could be considered to be the heat source of the hydrothermal system. Gases collected with thermal fluids (CO 2, He) have clear magmatic origin as shown by the 3 He/ 4 He ratio (7.75-7.93 Ra; Pedroni et al., 1999) and are consistent with the occurrence of a magma cooling and degassing. Lamentin. The main magmatic activity in this area is related to the Pitons du Carbet volcano. Germa et al. (2011) evidenced two main stages of activity. During a first stage (1 to 0.6 My), dacitic lava domes were emplaced in the summit area of the volcano. Around 337 ± 5 ky, a large sector collapse occurred yielding a large horse-shoe caldera open to the west (Boudon et al., 2013). It was immediately followed by the eruption of several large dacitic lava domes which are well observed today. The low 3 He/ 4 He ratios (2.2-2.3 Ra; Pedroni et al., 1999) of gases collected with thermal fluids of the Lamentin hydrothermal system seem to be indicative of a long-lived degassing magma which is rather consistent with the first stage of magmatic activity (1-0,6 My). Moreover, the occurrence of hydrothermalized formations (acidic alteration) within the deposits of the sector collapse dated 337 ± 5 ky is indicative of the existence of an early hydrothermal system which has been active before this collapse event. Therefore it is suggested that the development of the Carbet-Lamentin hydrothermal system could be old and related to a heat source associated with the first stage of magmatic activity of the volcano dated 1-0,6 My. Another hypothesis could be a more local magmatic activity, expressed about 0.6 My ago in the southern part of the Lamentin depression by the volcanic stage of Rivière Salée. Montagne Pelée. The hydrothermal system seems to be mainly developed within, or close to, the central eruptive conduit. The volcano started its activity around 127 ky and is still active with its latest eruption recorded in 1929. During a previous stage, it seems that the location of the volcano was slightly northward. Gas collected with the thermal fluids of Sources Chaudes hot springs show 3 He/ 4 He ratios (7.2 Ra; Ph. Jean Baptiste, personal communication, 2014) which are similar to the ratios measured in the fumaroles of the Guadeloupe Soufrière volcano (6.5-8.2 Ra; Ruzié et al., 2014). Bouillante. The volcanic activity observed in the area of Bouillante belongs to the so-called Bouillante volcanic Chain which

has been active between 1.2-0.2 My (Gadalia et al., 1988). The most recent magmatic activity recorded at Bouillante is dated at 0.5My (500 ky): it is considered has the probable heat source of the high temperature geothermal system. The 3 He/ 4 He ratios of gas collected in the thermal fluids (4-4.7 Ra) is consistent with a rather long-lived degassing magma. Finally, comparison of the four hydrothermal systems shows: the age of the magmatic activity considered as the heat source varies: 600 ky (Lamentin), 500 ky (Bouillante), 350 ky (Petite Anse), actual (Montagne Pelée), the duration of the magmatic activity varies: 1,1 My (Petite Anse), 0.7-0.8 My (Lamentin, Bouillante), 130 ky (Montagne Pelée), a time lag exists between the latest magmatic activity and the development of the (mature) hydrothermal system. It is estimated at 200 ky (Bouillante), 100-200 ky (Petite Anse, Lamentin), and more than 100 ky because the mature stage is not yet reached (Montagne Pelée). Toward a Continuum Geothermal Model From comparative parameters, summarized in Table 1, we propose a continuum model of geothermal system with three successive stages, prograde, peak and retrograde stages (Figure 4): The early hydrothermal stage is characterized by a rise in temperature of the system (prograde conditions) that follows an intense magmatic activity. During this prograde stage, the reservoir temperature reaches 220 C (at Montagne Pelée) but the partial equilibrium of fluids with the rock is not yet reached ( immature waters ). At this stage, we can assume that the extension of system is limited and it will tend to be expanding later (if permeability and recharge are effective). This is the case of the system of Pelée stratovolcano; it is difficult to estimate the duration of this prograde stage but could be around 100 to 200 ky. The thermal peak stage reach 250 C in Bouillante system. This high temperature stage should be sustainable during Figure 4. Continuum geothermal system based on a diagram temperature of the reservoir versus duration of the geothermal system. at least 300 ky (plateau) and occupying the same volume of geothermal reservoir. This is the case of the system of Bouillante. A retrograde trend of the system could be characterized by a slight cooling of the resource (250 C to 200 C) coupled with a decrease of hydrothermal activity, partial equilibrium of the fluid and a contraction of the reservoir. This is the case of Petite Anse geothermal system. Advanced retrograde stage is marked by a significant cooling of the reservoir to reach 140-100 C (Lamentin), coupled with an important contraction of the system. In this case, we can assume that the magmatic heat source is at the end of cooling because of its old age (> 600 ky) and that the geothermal system is progressively dying. This is the case of geothermal system of the Lamentin. Summary Such a continuum model should contribute to exploration in active volcanic arc setting such as the Lesser Antilles volcanic arc. It appears that a magmatic environment, both sustainable (700 ky to 1 Ma) and wedged in the time around 500 ky (magmatic peak), appears more favorable to the development of active high temperature geothermal system (peak stage). To conclude, this model allows characterization of different geothermal systems, explored or exploited, according to their degree of evolution. Each stage of this development (prograde to retrograde) is associated with a specific use of the geothermal resource: production of cold or heat (around 100 C), plus/or electricity production by binary cycle for temperature <180 C, plus/ or power generation by flash turbine as in Bouillante geothermal plant (> 200 C). References Bouchot V., Sanjuan B., Traineau H., Guillou-Frottier L., Thinon I., Baltassat J.M., Fabriol H., Bourgeois B., Lasne E. (2010) - Assessment of the Bouillante geothermal field (Guadeloupe, French West Indies): Toward a conceptual model of the high temperature geothermal system. Proceedings World Geothermal Congress 2010 Bali, Indonesia, 25-29 April 2010. Boudon, G., Villemant B., Le Friant, A., Paterne M., Cortijo E. (2013) - Role of large flank-collapse events on magma evolution of volcanoes. Insights from the Lesser Antilles Arc. Journal of Volcanology and Geothermal Research. vol. 263, pp. 224-237. Chovelon P. (1984) - Cadre structural des minéralisations hydrothermales des prospects du Lamentin et du Morne Rouge (Martinique). Report BRGM 84 SGN 326 GTH, 51p. Gadalia A., Gstalter N., Westercamp D. (1988) - La Chaîne volcanique de Bouillante, Basse-Terre de Guadeloupe (Petites Antilles) : Identité pétrographique, volcanologique et géodynamique, Géologie de la France, n 2-3, 11, pp. 101-130. Gadalia A., Baltassat J.M., Bouchot V., Caritg S., Coppo N., Gal F., Girard J.F., Gutierrez A., Jacob T., Martelet G., Rad S., Tailame A.L, Traineau. H., Vittecoq B., Wawrzyniak P., Zammit C. (2014) - Compléments d exploration géothermique en Martinique : conclusions et recommandations pour les zones des Anses d Arlet et de la Montagne Pelée. Report BRGM/ RP-63019-FR, 230 p. 367

Germa A., Quidelleur X., Lahitte P., Labanieh S., Chauvel C. (2011) - The K-Ar Cassignol-Gillot technique applied to western Martinique lavas: A record of Lesser Antilles arc activity from 2 Ma to Mt Pelée volcanism. Quaternary Geochronology, 6, pp. 341-355. Giggenbach W.F. (1988) - Geothermal solute equilibria. Derivation of Na K Ca Mg geoindicators. Geochimica et Cosmochimica Acta, 52, pp. 2749 2765. Gourgaud A., Gerbe M.C. (1991) - Mise en évidence de réalimentations magmatiques dans le système volcanique linéaire Burgos-Diamant (Martinique). Bulletin Société Géologique France, t. 162, n 6, pp. 1017-1027. Mas A., Patrier P., Beaufort D, Genter A. (2003) - Active hydrothermal circulations in the geothermal system of the Lamentin Plain, Martinique. Journal of Volcanology and Geothermal Research, vol.124, pp.195-218. Patrier P., Bruzac S., Pays R., Beaufort D., Bouchot V., Verati C. (2013) - Occurrence of K-feldspar bearing hydrothermal breccias in the Bouillante geothermal field (Basse Terre Guadeloupe). Bulletin Société Géologique France, t.184, pp. 119-128. Pedroni, A., Hammerschmidt, K., and Friedrichsen, H. (1999) - He, Ne, Ar, and isotope systematics of geothermal emanations in the lesser Antilles Islands Arc. Geochimica & Cosmochimica Acta, 63, ¾, pp. 515-532. Ruzié L., Aubaud C, Moreira M., Agrinier P., Dessert C., Gréau C., Crispi O. (2014) - Carbon and Helium isotopes in thermal springs of La Soufrière volcano (Guadeloupe, Lesser Antilles): Implications for volcanological monitoring, Chemical Geology, accepted. Sanjuan B., Genter A., Baltassat J.-M., Roig J.-Y (2003) - Réévaluation du potentiel géothermique dans les régions de Morne Rouge - Montagne Pelée et du Diamant (Martinique). Conclusions et recommandations. Report BRGM/RP-52549-FR, 38 p. Sanjuan B., Lasne E., Brach M. (2001) - Bouillante geothermal fluid: mixing and water / rock interaction processes at 250 C. Proceedings, 10th Water-Rock Interaction (WRI-10), Cagliari, Italy, June 10-15. Traineau H., Bouchot V., Caritg S., Gadalia A. (2013) Compléments d exploration géothermique en Martinique : volet géologie. Report BRGM/ RP-62349-FR, 153 p. Verati C., Patrier-Mas P., Lardeaux J.M., Bouchot V. (2014) - Timing of geothermal activity in an active island-arc volcanic setting: First 40 Ar/ 39 Ar dating from Bouillante geothermal field (Guadeloupe, French West Indies). Geological Society, London, Special Publications, in: 40 Ar/ 39 Ar Dating: from Geochronology to Thermochronology, from Archaeology to Planetary Sciences, eds. F. Jourdan, D. Mark & C. Verati, n 378. Westercamp D., Pelletier B., Thibaut P.M., Traineau H., Andreieff P. (1990) - Carte géologique de la France (1/50 000), feuille Martinique, BRGM edition, Orléans. 368