Proceedings World Geothermal Congress 2015 Melbourne, Australia, 19-25 April 2015

Effective Field Geological Mapping Techniques for Sumatran Geothermal Fields, Indonesia Lucas D. Setijadji Department of Geological Engineering, Gadjah Mada University, 2 Grafika Bulaksumur, Yogyakarta 55281 Indonesia E-mail address: [email protected]; [email protected]

Keywords: geothermal, Sumatra, geology, field mapping ABSTRACT Sumatra island is considered to be a potential location to host high-enthalpy geothermal fields in Indonesia. However, despite recent efforts to accelerate the development of new geothermal fields in this area, the results are not so good. There are some success stories, such as Sarulla. However, some other projects are now struggling to discover economically feasible reservoirs. It is possible that the complex geological setting of the Sumatran geothermal fields, due to the influence of the Sumatran Great Fault, make these fields significantly different from the Java island ones, where a majority of existing producing fields have been developed in Indonesia. While many geothermal companies rely heavily on geophysical surveys, especially MT, this paper will discuss the importance of a good geological mapping program at the initial stage of geothermal exploration. Particularly, the focus will be on the important aspects of mapping techniques that are considered critical, based on the experience in several projects in Sumatra. Before the real field mapping program, the prefield stage is important to generate tentative lithological, structural, alteration, and thermal anomaly maps that can be used for an efficient field work campaign. Public-domain data such as topography (DEM) and geological maps are not adequate for this purpose. High-resolution DEM data at several meters ground resolution (typically radar) are needed to be used as base maps for appropriate desk studies on lithological units and structural aspects of the study area. The identification of monogenic volcanic centres (many are only several hundreds of meters in diameter) is needed as volcanism in the Sumatra island is present not only as a stratovolcano but also as many monogenic volcanic centres controlled by structures. Multispectral thematic remote sensing such as Landsat TM have been used in some cases to detect the distribution of clay and oxide minerals. Surface thermal anomalies can be detected mainly by the night thermal IR ASTER image. At the field survey stage, efficient field observations should focus on the different facies of volcanic deposits (coherent versus fragmental), volcanic stratigraphy of different volcanic units, structural geology, and alteration aspects of different units. As there are many Holocene ash fall deposits in Sumatra that covers the majority of Pleistocene volcanic units, one must observe the underlying rocks. The youngest ash and other volcanic products need to be mapped to evaluate the current activity of the volcano. Hydrothermal alteration zones in Sumatra are not always related with currently active geothermal systems, so the identification and study of the nature of alteration zones, including the observation of quartz vein-float, is required to elucidate which alteration zones are directly related with the active geothermal system. Structural geology data is typically difficult to interpret at the initial exploration program, but observations on the distribution of thermal manifestations, dykes, and epithermal veins may serve as a good indication of the existence of extensional structures. Reliable geological information should then be incorporated in the integration of other data, especially geochemical and geophysical data, in order to improve the targeting of exploration drilling sites. 1. INTRODUCTION Indonesia has the highest geothermal potential in the world, with geothermal fields that are scattered in many islands especially in Java, Sumatra, Sulawesi, Maluku islands and Nusa Tenggara islands. Although Java is the center of geothermal development in the country today, the highest potential for high-enthalpy geothermal fields is considered to be located in the Sumatra island. However, despite recent efforts to accelerate the development of new geothermal fields here, the results have been poor. There are some success stories, such as Sarulla, but other projects are now struggling to discover economical reservoirs. It seems that the complex geological setting of Sumatran geothermal fields, due to the influence of the Sumatran Great Fault, make them significantly different from the Java island cases. Many geothermal companies working in this island, as well as in other places in the country, rely much on geophysical survey, especially MT. It also seems to be influenced by the regulation of Indonesia government that each geothermal project must be evaluated of its resource potency even at the preliminary stage of project. In this case, MT-derived anomalies are the main standard for such evaluations. As a result, almost all the projects make detailed geophysical surveys from the beginning of the project. However, this marginalizes the contribution of geological surveys. In this contribution, the author will discuss the importance of geological mapping programs at the initial stage of geothermal exploration. In particular, the focus is on the important aspects of mapping techniques that are considered critical, based on experience in several projects in Sumatra. 2. GEOLOGICAL SETTING OF SUMATRA ISLAND The Sumatra island is the westernmost of five major islands in the Indonesian archipelago, measuring about 1,730 km long in a NW-SE orientation. This island was built as an active continental margin subduction zone as part of the Cenozoic Sunda arc that extends from NW tip of Sumatra, Java, Bali, Lombok and Sumbawa islands. Sumatra was built up by the amalgamation of several continental fragments derived from Gondwanaland since the Late Paleozoic or Early Mesozoic periods that formed the Sundaland which is the SE margin of Eurasia continent (e.g. Metcalfe, 2006). Since the Late Mesozoic era, the Sundaland was already a stable craton, which was bound by a subduction zone at the west of Sumatra. 1

Setijadji Subduction formed a convergent tectonic margin between the Indian-Australian plates and the Sundaland (Hamilton, 1979). Subduction takes place along the Java/Sunda trench that reaches the maximum depth of 6.75 km. The oceanic crust is being subducted northward, more or less perpendicular, to the Sunda volcanic arc at a rate of 6 to 7 cm/yr (Hamilton, 1979; Simandjuntak and Barber, 1996). The Benioff seismic zone, representation of the subducted slab, extends to the depth of more than 600 km in Java but only 200 km in the case of Sumatra (Figure 1). However, tomographic imaging studies suggest that the lithospheric slab penetrates to a depth of at least 1500 kilometers in all sections (Widiyantoro and Van der Hilst, 1996). Sumatra is now a part of the Sunda-Banda volcanic arc that extends approximately 3,700 km long, from the northern tip of Sumatra island through Java to east of Damar island (Hamilton, 1979). This long arc is divided into the Sunda arc (Sumatra-Java-BaliLombok-Sumbawa) and Banda arc for the islands east of Sumbawa. (Fig.1). The basement crust thins eastward, from approximately 30 km beneath Sumatra to 15 km beneath the Flores Sea (Ben Avraham and Emery, 1973).

Figure 1: Tectonic setting of Sunda Arc, Indonesia (modified after Setijadji, 2005).

3. SUMATRA VS JAVA ARC SEGMENTS Although the Sumatra and Java islands are part of the same Sunda arc, it is important to highlight the significant differences in terms of volcanic geology between these islands that affect the different styles of geothermal fields. There are several substantial differences. Sumatra has a continental basement and its subduction style is an active continental margin. Meanwhile, Java is underlain by different basement compositions, from a continental basement in West and Central Java, to an island arc crust in East Java (Fig. 1). In addition, different from other parts of the Sunda Arc, the subduction beneath Sumatra is oblique, at the rate of 50 to 70 mm/yr (Natawidjaja and Triyoso, 2007; Figure 2). As the direct result of this oblique subduction, there is a presence of a great Sumatran Fault Zone (SFZ) that crosscut the middle of the island entirely. It is an active, northwest-trending, right-lateral strike-slip fault that accommodates the arc-parallel component of oblique subduction along the Sunda Trench (Figure 2). The Sumatran fault zone (SFZ) traverses the back-bone of Sumatra, within or near the active volcanic arc (Sieh and Natawidjaja, 2000). Older belts of subduction-related plutonic rocks on Sumatra all have a roughly northwest trend subparallel to the present magmatic arc, suggesting that the configuration of this part of the subduction system has remained almost the same throughout its history. The Sumatra Fault Zone is clearly identified by the cluster of shallow earthquakes, and it separates older rocks (mainly Tertiary) of volcanics and plutonics in the southwest, from the younger ones (Quaternary) in the northeast. In southern and central part of Sumatera, all of Quartenary volcanoes are located within 50 km of the fault zone (Barber et al., 2005). The fault zone is highly segmented (Figure 2), in which most of the principal segment boundaries are dilational in nature (Sieh and Natawidjaja, 1998, 2000). A series of grabens (e.g., Katili, 1970; Simandjuntak and Barber, 1996) have formed at these extensional segment boundaries, and volcanic edifices are situated within some of them (Bellier and Sebrier, 1994; Sieh and Natawidjaja, 2000). The slip rates along the fault segments are different in the south and north of Equator, as determined from the observation of river channels and offset along the Sumatran fault which had incised many thick pyroclastic deposits. It is interpreted that dextral slip rates increase northward but not uniformly. In the south of Equator, slip rates are approximately 11mm/yr. In contrast, north of Equator, these increase to 27 mm/yr at 2oN (Sieh and Natawidjaja, 2000). 2

Setijadji

Figure 2: Regional geological-structural setting of Sumatra island, including the great Sumatran Fault Zone (SFZ) and the rate of plate movement along its subduction boundary (Natawidjaja and Triyoso, 2007).

4. SUMATRAN GEOTHERMAL FIELDS Sumatran geothermal fields are scattered from the Aceh (the northern tip) to Lampung (the southern tip of the island), with a total of 84 prospects are already identified in 2004 by the Ministry of Energy and Mineral Resources (Figure 3). The majority of geothermal prospects, especially those with high-enthalpy potentials have common characteristics as follows. Thermal manifestations are associated with active or inactive, young stratovolcanoes, including ones that are partly and deeply eroded. At least one solfatara field, either small or large, can usually be found with traces of magmatic gases. Significant discharges of acid sulfate-chloride or acid chloride-sulfate waters (hot acid springs) can be found on the flanks of the stratovolcano. Partially neutralized thermal waters with chemical affinity to the acid flank discharges occur in the foothill region. At still lower levels, hot springs with a neutral or slightly alkaline NaCl type of thermal water can sometimes be found. Warm springs discharging bicarbonate-type waters in the foothill region can also be recognized. Finally, acid surface alteration typically occurs. The SFZ is close to the axis of the Quaternary magmatic arc on Sumatra, and the concentration of thermal manifestations such as hot and warm springs, fumaroles, and altered rock are commonly found in the vicinity of the fault zone suggests its potential as an important source of geothermal energy (Hochstein and Sudarman, 1993). 5. CHALLENGES IN GEOLOGICAL SURVEY Sumatran geothermal fields are relatively more challenging to be geologically studied for several following reasons. 5.1 Accessibility Most of the geothermal prospects are located in the protected forests. As a result, they are located at very remote areas with very limited access. Working in protected forests also requires a special license from authorities and visiting national forests is typically forbidden for foreigners. Remote sensing is expected to be effective to help the survey at the initial stage, but the dense vegetation coverage influences the effectiveness of remote sensing techniques. 5.2 Complexity of Volcanic Geology Based on the experience of working in several projects, the author considers that the volcanic geology of many Sumatran geothermal projects to be more complex than the Java case. Many Indonesian geoscientists are not well trained or educated in volcanic geology. Moreover, those that have training are typically trained in Java island, which is the center of geoscientific education in Indonesia. In the field, many volcanic manifestations in Sumatra are not commonly found in Java, so that geologists may face many difficulties in recognizing important features that are critical to the determination of the locations of potential heat sources, permeability zones, and active versus fossil hydrothermal alterations. The difficulties in volcanic geology and related hydrothermal system studies are manifested by the phenomenon that many of such works are contracted to other parties including universities. 3

Setijadji

Figure 3: Location of geothermal prospects in Sumatra, based on the map by Ministry of Energy and Mineral Resources 2004.

Working in Sumatra geothermal fields must include the anticipation of the presence of several volcanic eruption centers, rather than a single center, as the influence of faults is very intense. Preliminary information on such different volcanic centers are difficult to be found from published geological maps since such maps typically have a scale at 1:250,000 in which all Quaternary volcanics are mapped as single unit. In several cases, the intensive geological mapping program using a volcanic stratigraphy mapping approach revealed many young volcanic edifices. Different styles of volcanoes are present, such as, monogenic, stratovolcano, caldera, and composite volcanoes. In general, there are two main styles of Sumatran volcanoes based on their setting relative to the Sumatra Fault Zone: stratovolcanoes (similar to the Java case) and complex of monogenic volcanoes. Stratovolcanoes are typically located at a distance of around 10 km to 20 km from the Sumatra Fault Zone, while the monogenic volcanic complexes are typically found at pull-apart tectonic-volcanic depression settings, located at the proximity of the Sumatra Fault Zone. Magmatic eruption centers can be present in many ways: from central eruption crater, caldera, to lava dome and dike. For Sumatran lava domes and dikes, many of them represent themselves as glassy obsidian and its derivative products rather than porphyritic textures, probably due to the rapid movement of magma extrusion related to the active Sumatra Fault Zone. Acid volcanic rocks produced by explosive eruptions, especially welded ignimbrite, are also more common in the Sumatran compared to the Java case. These rocks can cover the majority of survey areas and restrict the observation of underlying rock units. Hydrothermal systems are believed to be composed of multiple stages rather than a single event. Recognizing how many hydrothermal systems are present in the survey area is truly challenging. It is the regulation by the Indonesian government that each geothermal concession area or WKP must be composed of a single geothermal field only. This regulation has forced companies, as operators of certain geothermal prospect areas, to determine the nature of geothermal system, such as the boundary of reservoir, from the early stages of exploration. This is very difficult to determine. Companies typically apply MT survey to determine the boundary of geothermal fields, compared with information on the geochemistry of hot springs. However, they may not be properly matched with geological information on styles of alteration and thermal manifestations. Several geothermal fields in Sumatra are located at the proximity with active epithermal gold mines that indicate the presence of overlapping fossil geothermal systems that are now producing epithermal gold deposits and an active system that is being targeted for geothermal exploration. Even in the active thermal manifestations, especially those of neutral pH hot springs, it is common to find quartz veins already formed. This phenomenon should be carefully examined, either active hot spring discharge and quartz veining are coexistent or they come from 4

Setijadji two overlapping events (i.e. active versus fossil system). Detailed mapping of hydrothermal alteration zones can help in elucidating the problem. Fossil systems are typically already eroded that may be well represented by the presence of high temperature secondary minerals. However, since the Sumatra Fault Zone is a very active fault, it is also anticipated that high temperature secondary minerals can be exposed locally due to fault’s movement, while the whole alteration system is still an active and largely not eroded. Detailed and careful field observations, combined with proper laboratory analysis, are required to make solid conclusion. Structural geology data, such as joints and faults, are more commonly found in Sumatra geothermal fields compared with the case of Java. The logical reason is that the Sumatra Fault System is very active such that most of volcanic units, even the historical ones, may be affected by such structures. The geological mapping program at the initial stage of geothermal exploration surveys should consider structural geology as important geological data to make preliminary interpretations on the zones of permeability related to extensional joints and faults. In this case, it is not enough to make structural geology interpretations based on the regional setting of the Sumatra Fault Zone, as this regional strike-slip fault is in fact highly segmented with at least 20 segments. Most of the principal segment boundaries are dilational in nature (Sieh and Natawidjaja, 2000) and become the sites of a series of grabens with volcanic edifices situated within some of them. The slip rates along the fault segments are also different in the south and north. In general it is interpreted that dextral slip rates increase northward, but not uniformly. 5.3 Complexity in Geohazards Risk Because of the complexity of volcanic geology, geothermal fields in Sumatra also hold more complex geohazards potential for future development, compared with Java’s case. For example, recent eruptions of the Sinabung volcano, after its dormancy for more than 400 years, affected the development plan of the nearby Sibayak geothermal field. Ideal geothermal fields to be developed in Indonesia are considered to be the ones that are located within young volcanoes but do not show historical eruptions. Information of historical eruptions of Sumatran volcanoes are far from adequate, each geothermal project should collect their own evidence to evaluate the history of volcanic eruptions in the proximity of the survey area. Some major hydrothermal eruptions are recorded in the literature, such as the eruption in the Suoh valley, south Sumatra, in the 1930s and 1990s. However, much more are to be found by direct field surveys. Tephra deposits must be carefully examined, especially if they buried human artifacts, which is an indication of a historical event. Deposits surrounding the lakes must be examined to clarify the nature of lakes to determine whether they are sites of recent volcanic eruptions or not. 6. PLAN FOR EFFECTIVE GEOLOGICAL MAPPING Considering the complexity of the volcanic geology of Sumatran geothermal fields and the requirement to collect precise geological data to enable proper interpretation from the early stage of geothermal surveys, here are some suggestions that may help improve the planning and execution of geological field mapping programs. Effective field geological mapping programs, especially at the early stage of a geothermal survey program, should focus on several main targets. These targets include the distribution of surface rocks, understanding volcanism history, location of young volcanic eruptive centers, geological structures (faults and fractures), and the distribution and nature of hydrothermal alterations and thermal manifestations. These data are critical in making proper geological contributions for early assessments of the potential of the studied geothermal prospect. Well-designed geological mapping typically consists of three stages of activities: prefield, field work, and postfield work stage that consists of laboratory analysis and reporting (Figure 4). Remote Sensing

Pre-Work Geological Map, Reports

Landsat/ASTER

DEM/DTM

VNIR

Tentative Lineament map

Tentative Geological Map

Tentative Volcanic Units Map

SWIR

Tentative Alteration (Mineralogical) Map

TIR

Tentative Thermal Anomaly Map

Fieldworks Structural mapping

Geological mapping

Volcanic Facies mapping

Alteration mapping

Thermal Manifestation mapping Rock samples

Lab. Analysis

Petrography

XRD

Radiometric Dating: K-Ar, Ar-Ar, Carbon dating

Geological Structure Map

Data Analysis and Reporting

Geological Map

Volcanic Facies Map

Evolution of Geology

Alteration Map

Geochemical, Geophysical data

GIS Database

Conceptual Model of Geothermal System

Figure 4: Typical workflow of geological mapping program and its contribution for geothermal exploration. 5

Setijadji 6.1 Prefield Stage The prefield stage is aimed to generate tentative lithological, structural, alteration and thermal anomaly maps that can be used for efficient field work campaigns. This stage consists of the collection and integration of previous works in the survey area, combined with the interpretation of remote sensing images. Moreover, this stage must generate more detailed tentative geological maps, typically at a scale of 1:40,000 to 1:50,000, based on the reinterpretation of the published regional 1:250,000 geological maps. Additionally, the prefield stage must generate a plan map for the planned field survey tracks and observation points. Since many of Sumatran geothermal fields are located in remote areas and protected forests, effective field data collection can only be achieved from good planning on proposed survey tracks and locations of critical observation sites. Recently, high-resolution digital elevation model (DEM) data have become one of the best sources for prefield studies because they can be digitally and manually processed and interpreted with regard to lithology and structural geology information. Unfortunately, in the Indonesian case, for public-domain DEM data, such as topography maps on the scale of 1:50,000, raster ASTER DEM, and SRTM DEM are not adequate for this purpose. High-resolution DEM data at several meters ground resolution (typically radar) are needed to construct a base map for appropriate desk studies on lithological units and structural aspects of the study area. For example, the identification of monogenic volcanic centres (many of them are only several hundreds of meters in diameter) is needed, as volcanism in Sumatra is not only present as stratovolcanoes but also many monogenic volcanic centres controlled by structures. An example of the benefit of using high resolution DEM image is shown in Figure 5. Recently, Lidar has been introduced in Indonesia’s geothermal projects and this technology seems to have a high potential of being used in geothermal applications.

Crater lake

Central crater

Figure 5: Comparison of accuracy on volcanic features recognition between high resolution DEM data (left, 5m resolution) and public-domain SRTM data (right, 30m resolution) at one Sumatran active volcano.

Typically, one can apply several methods of digital processing on original DEM data using GIS software, such as ArcGIS spatial analysis tools. The purposes of DEM analysis are to identify two main targets: lithological unit interpretation (delineation) and structural geology analysis through lineament interpretation. Lineaments are an important aspect in interpreting geological structures using remote sensing digital images. The lineament is the aligned and elongated geometry of an object which is oriented systematically in a specific direction. Usually, the lineament is interpreted by following river or valley alignments which are controlled by strong erosion. For this purpose, several derivative thematic images can be generated from the original DEM map through some spatial analysis, such as: 1. Hillshade analysis, from this process one generates 3D-like images that enhance morphology, especially around steep slopes. 2. Lithological units interpretation, done manually (free hand), which considers the texture, shape and spatial association of features observable from the hillshade images. 3. Manual lineaments interpretation by free hand delineation following valleys or rivers which have relatively straight patterns at specific lengths. Tectonic lineaments typically have distinguished features, including a systematic interval and repetitiveness. 4. If available, a digital lineaments extraction can also be done using some specific software programs. This method of extraction is good for statistical analysis but it cannot be used for distinguishing tectonic versus nontectonic structures. Hence, a careful interpretation of the results is required. 5. On the large lineaments dataset generated by the DEM image, statistical analyses can be applied to generate conclusions on the dominant force (maxima) in the structural data cluster and its spatial direction. It is achieved by making a rose diagram for strike or bearing data. The use of multispectral thematic remote sensing images such as Landsat TM, SPOT and ASTER should be treated carefully, as there are several fundamental restrictions attached. Firstly, there is the fact that the majority of geothermal fields in Sumatra are located within dense tropical forest that limit the spectral information collected directly from the ground. Secondly, sensors from several of the most commonly available images, especially Landsat and ASTER, are currently not working properly, so that one 6

Setijadji needs to carefully select the scenes that are free of sensor errors. With several precautions, thematic mapping sensors can still provide important information on the studied geothermal fields. It is already proven in some cases, that some remote sensing data can be utilized to detect the distribution of clay and oxide minerals by a combination of the VNIR and SWIR bands of Landsat TM, SPOT, and ASTER. Surface thermal anomalies are mainly detected using the night thermal IR ASTER image (Figure 6). On processing remote sensing images, one typically starts with the initial processing that includes geometric and radiometric corrections, radiance calibration, dark pixel correction, cloud masking, and land cover classification. Then, mineral groups (i.e. iron oxide and clay groups) mapping can be done based on suitable band ratios. The resulting mineral group maps, such as kaolin, silicaillite and hematite groups can then displayed as RGB map. After this, surface temperatures are calculated using the digital number (DN) of TIR bands analyzed with a certain algorithm. Such analysis can be applied on Landsat TIR bands, but in the author’s experience, it is better to derive it from the night TIR ASTER bands.

Figure 6: Hydrothermal alteration map (left, derived from Landsat TM) and surface thermal anomaly map (right, derived from night TIR band ASTER) at one Sumatran active volcano.

In summary, on the prefield stage, the analysis methods consist of following major steps: 1. Rectification of many images, such as remote sensing images and geological map; 2. Digitizing of geological maps that consist of lithological and structural features; 3. Image processing of satellite images (Landsat, SPOT, and/or ASTER) for thermal and alteration manifestations utilizing VNIR, SWIR, and TIR bands; 4. DEM/DTM image processing for lithological and geological structures interpretation; 5. Integration of data in GIS for a final interpretation on the lithology, structural geology, hydrothermal alterations, and thermal manifestations; 6. Generating tentative maps on lithological units, geological structures, hydrothermal alteration zones, thermal anomalies, and proposed tracts for the field investigation. 6.2 Field Work Stage During the field work stage, efficient field work should follow the planned tracks and observation points derived from the prefield stage. On all the observation sites, observations should focus on the different facies of volcanic deposits (coherent versus fragmental), volcanic stratigraphy of different volcanic units, structural geology, and alteration aspects of different units. Volcanic rock facies classification can follow the schema by McPhie et al., 1993 (Figure 7). In each observation point, the appropriate data are collected, which consist of locational information, lithology, structural geology, and alteration-thermal manifestation aspects. If necessary, certain checklist forms can be prepared to make a standardized input data list from all observation points. In selected observation points, rock samples of enough quantities should be collected for further analysis with different purposes. Fresh rock samples are typically collected for petrography analysis. For lithological unit classification, XRD for analysis on secondary minerals detection, and selected samples for radiometric dating are used. Meanwhile, hydrothermally altered rocks are sampled for petrography and XRD analysis for defining hydrothermal alteration styles. Travertine, silica sinter, and silica veins are also collected for petrography and XRD analysis for the interpretation of the environment of deposition. Charcoal fragments can also be collected for carbon dating. As there are many Holocene ash fall deposits in Sumatra that covers the majority of Pleistocene volcanic units, one must observe the underlying rocks to check if the surface is covered by tephra. In addition, the youngest ash and other volcanic products need to be mapped to evaluate the current activity of the volcano. 7

Setijadji Hydrothermal alteration zones in Sumatra are not always related with currently active geothermal systems. Thus, the identification and study of the nature of alteration zones, including the observation of quartz vein-floats, is required to elucidate which alteration zones are directly related with an active geothermal system. The results of field checking on altered grounds suggest that, in several geothermal prospect areas, one can differentiate zones of altered grounds despite the fact that most areas are covered by rain forests. Furthermore, one can differentiate acid alterations (kaolin-rich), neutral (quartz-illite), and intermediate (iron-oxide) using the Landsat TM images. ASTER images also has a potential of achieving this, but it is difficult to find free images. Surface temperature values extracted from the TIR band of Landsat have different distributions from that of ASTER, especially from the night data. It is evaluated, based on the conceptual model and ground checking of results that, night TIR image of ASTER is more accurate. The results suggest that remote sensing interpretation can contribute significantly in the early stage of geothermal exploration, despite the presence of large vegetation coverages in Indonesia. Structural geology data is typically difficult to interpret at the initial exploration program, but observations on the distribution of thermal manifestations, dykes and epithermal veins may serve as a good indication of the existence of extensional structures. Field geological structure data typically consist of field measurements on shear joints, extensional joints, faults planes, and striations. These data can then be analyzed to determine the styles of geological structures, direction of the compressive field, and the geological structure history in combination with secondary data derived from DTM interpretation.

Figure 7: Classification of volcanic rocks by McPhie et al. (1993).

6.3 PostField Work Stage The postfield work stage typically consist of following steps: 1. Laboratory activities that typically consist of petrography and XRD analysis. Petrography is used to clarify the different lithological units, mineralogical compositions of volcanic rocks, and alteration degrees of rocks. Meanwhile, XRD analysis is done to clarify the secondary minerals present. XRD analysis is typically conducted on air dried bulk rocks, clay minerals separate, and ethylene glycol-treated clay samples. Selected rocks are analyzed for radiometric dating, such as fresh lavas for K-Ar or Ar-Ar dating, and charcoal deposits for carbon dating. 2. Structural geology analysis is not often performed in preliminary surveys, but it is an important part of the geological mapping program. In this stage, a local-scale structural analysis is done for determining structural patterns that play important roles in the development of the geothermal system by providing permeable zones in the working area. Local structural analysis uses mainly field measurements data that include joints, bedding orientation, and offset lithology. Statistical analysis can be used in the large structure dataset, such as field joints data. The purpose of this analysis is to get the dominant force of the geological structure data cluster by making a rose diagram for the strike or bearing data. Then, kinematic analysis can be performed for the fault data to determine the relative offset in the lithology and also to interpret the tectonic regime which controls the development geological structures. The results of all analyses are then synthesized in order to derive the best conclusions for the type of deformation processes occurring in the studied area. In addition, the finalization of structural geology map through classification of structures into faults, lineaments and other structural features is likewise included. 8

Setijadji 3. Integration of all geological data to generate lithological units, structural geology features, interpretation on volcanic history, and alteration-thermal manifestation styles 4. Integration of geological, geochemical, and geophysical data to develop early an conceptual model of the potential geothermal system 5. Generating digital GIS database from all data 7. CONCLUDING REMARKS Sumatran geothermal fields are considered to be more difficult to be explored compared with the Java cases. Thus, good planning and effective execution of field geological mapping are important for generating a good initial evaluation on the geothermal development potential at the early stage of exploration. With such complexity, it is considered that the integration of geological, geochemical, and geophysical data are required to fully understand Sumatran geothermal fields. Effective field geological mapping program can only be achieved after intensive prefield work to provide tentative lithological, structural, alteration, thermal anomaly, and planned survey tracks and observation points maps. Field work observation should focus on the different facies of volcanic deposits, volcanic stratigraphy of different volcanic units, structural geology, alteration and thermal manifestation aspects of the different units. Hydrothermal alteration zones in Sumatra are not always related with currently active geothermal system, so that the field identification and laboratory study of the nature of alteration zones is required to determine which alteration zones are directly related to the active geothermal system. Observations on the distribution of thermal manifestations, dykes and epithermal veins may serve as a good indication of the existence of extensional and permeable structures. Ideally, reliable geological information should then be incorporated in the integration with geochemical and geophysical data in order to better define the system and target the sites of exploration drilling. REFERENCES Gasparon, M.: Quaternary Volcanicity, in A. J. Barber, M. J. Crow, and J. M. (eds) Sumatra: Geology, Resources and Tectonic Evolution, Geological Society London, Memoirs 31, (2005), 120-130. Gozzard, J.R.: Image Processing of ASTER Multispectral Data, Department of Industry and Resources, Geological Survey of Western Australia, (2006). Hamilton, W.B.: Tectonics of the Indonesian Region, Professional Paper 1078, U.S. Geol. Surv., Washington, DC, (1979), 345 p. Hochstein, M. P. and Sudarman, S.: Geothermal Resources of Sumatra. Geothermics, 22, (1993), 181-200. Ibrahim, R. F., Fauzi, A., and Suryadarma: The Progress of Geothermal Energy Resources Activities in Indonesia. Proceedings, World Geothermal Congress 2005, Antalya, Turkey, (2005). Katili, J. A.: Volcanism and plate tectonics in the Indonesian island arcs, Tectonophysics, 26, (1975), 165-188. McPhie, J., Doyle, M., and Allen, R.: Volcanic textures : a guide to the interpretation of textures in volcanic rocks. University of Tasmania, Centre for Ore Deposit and Exploration Studies, (1993), 198 p. Natawidjaja, D. H. and Triyoso, W.: The Sumatran Fault Zone- From Source to Hazard. Journal of Earthquake and Tsunami, 1, (2007), 21–47. Neumann van Padang, M.: Indonesia. Catalog of Active Volcanoes of the World and Solfatara Fields, IAVCEI, Rome, 1, (1951), 1271. Puspito, N. T. and Shimazaki, K.: Mantle Structure and Seismotectonics of the Sunda and Banda Arcs, Tectonophysics, 251, (1995), 215-228. Siebert, L. and Simkin, T.: Volcanoes of the World: an Illustrated Catalog of Holocene Volcanoes and their Eruptions. Smithsonian Institution, Global Volcanism Program Digital Information Series, GVP-3, (http://www.volcano.si.edu/world/), (2002). Simandjuntak, T.O. and Barber, A.J.: Contrasting Tectonic Styles in the Neogene Orogenic Belts of Indonesia. In Hall, R. and Blundell, D.J. (Eds.), Tectonic Evolution of Southeast Asia, Geological Society Special Publication, 106, (1996), 185-201. Tatsumi, Y. and Eggins, S.: Subduction zone magmatism. Blackwell Science, Ann Arbor, (1995), 211 p. Van Bemellen, R. W.: The Geology of Indonesia, vol. 1A. Martinus Nijhoff, the Hague, (1949), 732 p. Widiyantoro, S. and van der Hilst, R.: Structure and Evolution of Lithospheric Slab Beneath the Sunda Arc, Indonesia, Science, 271, (1996), 1566-1570.

9