FIRE and MUD Contents

Geothermal Exploration of the pre-1991 Mount Pinatubo Hydrothermal System

By F.G. Delfin,¹ Jr., H.G. Villarosa,¹ D.B. Layugan,¹ V.C. Clemente,¹ M.R. Candelaria,¹ and J.R. Ruaya¹

¹Geothermal Division, PNOC-Energy Development Corporation Ft. Bonifacio, Metro Manila, Philippines 1201.

Abstract

Geothermal exploration in Mount Pinatubo was conducted in two stages, surface investigations from 1982 to 1986 and deep exploratory drilling and well testing from 1988 to 1990. Discouraging results from the wells forced the abandonment of the prospect 13 months before the April 2, 1991 explosions.

Surface geological, hydrogeochemical, and geoelectrical studies indicated that Mount Pinatubo hosted a geothermal system with temperature of at least 200°C. That this brine system was most likely centered on the volcano's immediate northwestern flank was suggested by the better correlation of thermal features, faults, and resistivity anomaly in that sector. A possible magmatic input to this system was inferred from the high chloride contents of the high-altitude springs and solfatara pool. The hydrothermal features on the volcano's eastern slope were interpreted to be either leakages from the system on Pinatubo's northwestern flank or constituted a separate brine system.

Three exploratory wells drilled to depths ranging from 2,100 to 2,700 m encountered temperatures of 261 to 336°C. These boreholes were drilled through poorly permeable andesitic and dacitic volcanic rocks that had been altered to neutral-pH hydrothermal assemblages. Acid alteration minerals were not widespread, despite the bore fluids being acidic (pH = 2.3-4.3) Na-Cl waters. Low injectivity indices, low wellhead pressures, and the generally conductive temperature profiles of the wells confirmed the poor permeability of the hydrothermal system. This, together with the acid nature of the brine, made the wells noncommercial.

On the basis largely of the chemical and isotopic compositions of the deep reservoir fluids, significant magmatic input to the hydrothermal system is indicated. The deeper central part of the system was modeled to be a hot (>300°C) and generally impermeable two-phase zone produced by the condensation of magmatic volatiles. Two apparently separate brine systems with temperature of ~260°C convected above and on the flanks of the deep acid zone.

If our model of the pre-1991 Mount Pinatubo hydrothermal system applies to other systems associated with active or recently active stratovolcanoes, high-temperature acidic fluids are to be expected in the central parts of such reservoirs. Exploration efforts should be expended on the flanks of these systems, where neutral-pH brines are expected to form. Such efforts, however, must always include an assessment of the suitability for long-term development and exploitation of these "magmatic-hydrothermal systems."

Note to readers: Figures open in separate windows. To return to the text, close the figure's window or bring the text window to the front.

Introduction

As part of an effort to locate and develop new electric power sources in Luzon island, the Philippine National Oil Company-Energy Development Corporation (PNOC-EDC) implemented in 1982 a geothermal exploration program at Mount Pinatubo, approximately 80 km northwest of Manila (fig. 1). The presence of numerous hydrothermal features in the volcanic complex and its proximity to industrial centers made it then a worthy geothermal prospect.

Prior to PNOC-EDC's studies, little was known about Mount Pinatubo. Reconnaissance geologic mapping, hydrogeochemical surveys, and geoelectrical investigations were carried out from 1982 to 1986. Different interpretations and assessments of the surface exploration data were made by the staff of PNOC-EDC, Kingston Reynolds Thom and Allardice (KRTA) consultants, and the New Zealand Department of Scientific and Industrial Research. Subsequently, three deep exploratory wells were drilled in 1988-89. Discouraging results from these wells led to the abandonment of the prospect by early 1990.

The principal objective of this paper is to present a summary of the results of the various stages of the exploration program. On the basis of these and other studies during and after the 1991 eruption, we propose a model of the pre-1991 Mount Pinatubo hydrothermal system. This model has important practical implications for geothermal exploration and development of recently active stratovolcanoes.

Surface Exploration

Geology

Mount Pinatubo lies along the Luzon arc, a north-south-trending belt of late Tertiary to Quaternary volcanoes that extends northward from southern Luzon to Taiwan (Defant and others, 1989). The Luzon arc is associated with eastward subduction of the South China Sea oceanic crust along the Manila trench (fig. 1). In the Mount Pinatubo to Mount Mariveles segment of the arc, the volcanoes overlie the contact of the Zambales Ophiolite Complex (ZOC) and Tertiary sediments of the central Luzon basin.

The ZOC in the prospect area comprises variably serpentinized peridotites and pyroxenites, gabbros and related felsic intrusives, diabase dikes, and basaltic lavas. These rocks are exposed north, west, and southwest of Mount Pinatubo. North of the volcano (fig. 2), the ZOC unconformably underlies upper Miocene to lower Pliocene sandstones, siltstones, and conglomerate lenses (Delfin, 1984) and is intruded by a large body of Neogene diorite and quartz diorite (NI, fig. 1) (Philippine Bureau of Mines, 1963).

Mount Pinatubo is composed largely of hornblende andesite and dacite plus lesser amounts of pyroxene andesite and basalt emplaced as lava flows, pyroclastics, dikes, sills, and domes. The eruptive history of the volcano has been divided into two major periods (Delfin, 1984) termed here as ancestral and modern Mount Pinatubo (Newhall and others, this volume). Ancestral Mount Pinatubo was a Pleistocene andesite-dacite stratovolcano whose remnants are preserved in topographically elevated terrain surrounding the pre-1991 Pinatubo dome (fig. 2). Contemporaneous with this ancestral Mount Pinatubo stratocone are the domes of Mounts Negron, Cuadrado, and Mataba and the Bituin plug (Delfin, 1984). Modern Mount Pinatubo is a dacite-andesite dome complex surrounded by sheets of dacitic pyroclastic flows and related lahars. New ¹⁴C dating and stratigraphic analyses (Newhall and others, this volume) reveal at least six distinct eruptive episodes of modern Pinatubo beginning with a caldera-forming event (Inararo episode) >35,000 ¹⁴C yr B.P. The pre-1991 summit of Mount Pinatubo was a steep-sided, 2-km-wide hornblende andesite dome (Delfin, 1984) believed to have been emplaced during the Buag eruptive event about 800-500 cal yr B.P. (Newhall and others, this volume).

The Pinatubo volcanics are cut by predominantly northwest- and north-trending dip-slip faults. de Boer and others (1980) proposed the existence of a major northwest-striking structure, the Iba fracture zone, passing beneath Mount Pinatubo. Manifestations of this tectonic feature include the Marunot (or Maraunot) fault, the Dagsa fault, and the northwest-southeast alignment of surface hydrothermal discharges on Mount Pinatubo (Delfin, 1984). Shorter and less dominant northeast- and east-trending faults cut the major structures without any apparent offset. These shorter and easterly trending faults are particularly well defined near the summit region of the volcano (fig. 2).

Figure 1. Location map of the Mount Pinatubo area. Abbreviations for major geologic units: R, Recent sediments; Qv, Quaternary volcanics; TS, Tertiary sediments; NI, Neogene intrusive; ZOC, Zambales Ophiolite Complex. Inset shows major tectonic features of Luzon. Triangles, volcanoes.

Figure 2. Simplified geologic map of the Mount Pinatubo area. ZOC, Zambales Ophiolite Complex; TS, Tertiary sediments; AP, ancestral Mount Pinatubo; MP, modern Mount Pinatubo; PD, Pinatubo dome; ND, Negron dome; CD, Cuadrado dome; MD, Mataba dome; BP, Bituin plug. Heavy lines are faults. MF, Maraunot fault, TF, Tayawan fault, DS F, Dagsa fault, DF, Dangey fault. Outline of 1991 caldera from Newhall and others (this volume). Abbreviations for hot springs () are keyed to table 1. Altitude in meters.

Distribution and Chemistry of Thermal Springs

Prior to the 1991 eruptions, twelve groups of hydrothermal discharges were found in Mount Pinatubo. Most of the springs were concentrated along a narrow, ~25-km-long belt extending from Nacolcol in the northwest to Cuyucut in the southeast (fig. 2).

Representative chemical analyses of the spring waters are listed in table 1. The thermal springs have been classified on the basis of their relative abundance of Cl, HCO₃, and SO₄ (fig. 3). The chemistry of the springs varies systematically with altitude and with distance from the volcano. Farthest from Mount Pinatubo is the Nacolcol chloride spring, 15 km northwest at an altitude of 220 m (fig. 2). The Cl-HCO₃ springs of Dagsa, Cuyucut, and Asin lie between ~300 and 400 m in altitude at distances of 6 to 10 km from Mount Pinatubo. The Kalawangan Cl-HCO₃ spring, however, is situated 4 km east of Mount Pinatubo at a higher altitude of 675 m. At still higher altitudes (>500-1,100 m) and closer to the vent are the HCO₃-SO₄ springs of Upper Maraunot, Dangey, Mamot, Pula, and Pajo. The highest feature is the Pinatubo solfatara and the Cl-SO₄ solfatara pool located 1.5 km northwest of Mount Pinatubo at an altitude of 1,180 m. The Lower Maraunot springs, immediately south of the solfatara, are likewise Cl-SO₄ in composition. The Pinatubo solfatara and the Upper Maraunot springs reached 95°C; the rest of the springs have discharge temperatures of <60°C. Except for the acid solfatara pool, all the Pinatubo thermal springs discharge neutral-pH waters. Cation and silica mixing-model geothermometers for the high-chlorine springs yielded subsurface temperatures of at least 200°C (Clemente, 1984; Villaseñor, 1984).

The composition of gases collected from the Pinatubo solfatara is shown in table 2. In terms of mole percent, the solfatara gases are composed of 86-90 percent CO₂, 8-12 percent H₂S, 0.92-1.3 percent N₂, and 0.005 percent H₂. A source temperature of 260°C for the Pinatubo solfatara was obtained by using the D'Amore and Panichi geothermometer (Villaseñor, 1984). The possibility of magmatic input to the solfatara gases (Bogie, 1984), however, makes this temperature estimate dubious.

Figure 3. Cl-HCO₃-SO₄ ternary diagram of spring compositions, Mount Pinatubo. P1, P2, and P3 refer to water samples from PIN-1, PIN-2D, and PIN-3D, respectively. For spring names, see table 1.

Table 1. Selected chemical analyses of Mount Pinatubo thermal waters.

[Abbreviations for sources appear on fig. 2. Concentrations are given in milligrams per kilogram.]

Table 2. Representative gas compositions from the Pinatubo solfatara and wells.

[WHP = wellhead pressure; H = enthalpy; SP = sampling pressure; all concentrations in mmol/100 mol steam.]

Resistivity Surveys

Schlumberger resistivity traversing (SRT) and vertical electrical soundings (VES) were employed at Mount Pinatubo to delineate the most likely location and depth of the geothermal resource (Esperidion, 1984; Apuada, 1988). SRT measurements were conducted at 522 stations by use of half-current electrode separation (AB/2) of 250 and 500 m. The isoresistivity of the area at AB/2=500 m is shown in figure 4. Three low-resistivity anomalies were delineated, the Maraunot, Mount McDonald, and Dagsa anomalies. The Maraunot anomaly is a northwest-trending resistivity low defined by the 50 m contour and enclosing the Maraunot, Pula, and Pajo thermal springs. The Dagsa anomaly is a broad low-resistivity feature traversing the entire eastern flank of Mount Pinatubo. Defined by the 50 m value, it includes pockets of 10-20m lows that do not show any well-defined pattern. Between these two large anomalies is the Mount McDonald anomaly, a small northeast-trending 50 m low that is located southeast of the Pinatubo dome. It parallels the southeast caldera rim but is not associated with any surface hydrothermal feature.

Thirty-six VES measurements were undertaken to test the vertical persistence of shallow anomalies delineated by SRT. Maximum half-current electrode spread for the VES was 1,200 m. The contours of the interpreted resistivities of the VES bottom layers, derived from one-dimensional modeling, are broadly similar to what is shown in figure 4.

All three SRT anomalies persist at greater sounding depths. Both the Dagsa and Mount McDonald anomalies are characterized by almost uniform bottom resistivities of <=50m. On the other hand, the Maraunot anomaly shows further decrease in resistivity with depth attaining values of <=30m. The immediate vicinity of the Pinatubo dome, however, remained characterized by high resistivities of >100 to >1,000 m. These values suggest poor subsurface fluid circulation or minimal rock alteration around the Pinatubo dome, at least to depths penetrated by the VES.

Figure 4. Isoresistivity map of the Pinatubo area obtained at half-current electrode separation (AB/2) of 500 m. The anomalies are delineated by 50-m contours. Abbreviations for hot springs () are keyed to table 1. Altitude in meters.

Predrilling Model

At the conclusion of the surface exploratory studies in 1986, two conceptual models of the Mount Pinatubo hydrothermal systems were proposed. In the first model, Delfin (1984) and Villaseñor (1984) invoked two hot-water circulation systems. One of these systems, associated with Mount Pinatubo, was manifested by the Maraunot anomaly and by springs located on the western flank of the volcano. The Pinatubo solfatara and the numerous SO₄-HCO₃ springs at higher altitude were interpreted to indicate extensive boiling and steam condensation of upwelling geothermal fluids. The Maraunot fault and its subsidiary fractures were believed to be the main conduit of the hot water to the surface. A separate convective system, associated with Mount Negron, was invoked for the springs, altered rocks, and resistivity lows located on the eastern flank of the volcano. Of these two systems, the one associated with Pinatubo was deemed more promising but appeared small compared to other Philippine geothermal fields then explored.

The second model (Bogie, 1984) postulated a single large hydrothermal system in the complex. In this model, fluids from this huge system well up beneath Mount Pinatubo and flow out laterally in numerous directions to emerge as springs on both flanks of Mount Pinatubo. On the basis of the high chloride concentrations of the Maraunot springs and the solfatara pool, Bogie (1984) proposed a magmatic input to the hydrothermal system.

Exploration Drilling

In order to test the two exploration models and confirm the existence of a geothermal resource viable for electric power generation, three deep wells (fig. 5) were drilled between August 1988 and November 1989. Well PIN-1, southeast of Mount Pinatubo, reached a total vertical depth (TVD) of 2,733 m. PIN-2D (TVD of 2216 m) was located northeast of the dome and drilled directionally to the west. PIN-3D (TVD of 2190 m) was spudded from a pad northwest of the dome and deviated to the southwest. A summary of selected well data and bore outputs is shown in table 3.

Figure 5. Well location map of Mount Pinatubo exploratory wells showing pre-1991 caldera (hachures), major faults (heavy lines, ball on downthrown side), and fumaroles formed after April, 1991 explosions. Dashed lines indicate well tracks; BH refers to well bottom. Fault names are the same as in figure 2. A-A' is section line shown in figures 6 and 7. Contour interval is 200 m. Well PIN-1 had a total vertical depth of 2,733 m; well PIN-2D, 2,216 m; well PIN-3D, 2,190 m.

Table 3. Selected Pinatubo well data and bore outputs.

[MD/VD = measured and vertical depths, in meters; BHT = maximum bottomhole temperature; WHP = wellhead pressure; NCG = non-condensible gases]

Discharge Tests and Bore Outputs

PIN-1 was discharged by boiler stimulation after two attempts to discharge it by air compression failed (PNOC-EDC, 1990). This well had the longest flow test (138 days), and sampled brines were largely free of drilling fluids. The maximum measured temperature of the well was 261°C at the bottom. Discharge enthalpy ranged from 1,050 to 1,700 kJ/kg. Low wellhead pressure (<0.40 MPag) did not allow the well to sustain discharge beyond 138 days.

After 15 days of heat-up, PIN-2D discharged by natural buildup of its wellhead pressure. This initial discharge had to be cut off after 3 days because of the low pH (2.3) of the brine. Stable output thus could not be obtained, but the  initial enthalpy of 1,442 kJ/kg later increased to 2,145 kJ/kg. A second 3-day flow test yielded the same trend in fluid pH and discharge enthalpy. PIN-2D reached a maximum temperature of 336°C at the bottom, the highest measured in all the wells. Its mass flow of 24-69 kg/s was equivalent to a power output of 7.7 MWe.

PIN-3D discharged for 40 days after an 8-day heat-up period. The well reached 330°C at the bottom; its temperature profile was dominantly conductive from 500 to 1,400 m in depth, followed by a section of low gradient to the well bottom. Low wellhead pressure and low pH of the discharge brine made this well commercially nonproductive.

All the wells had poor permeability, as indicated by their generally conductive temperature profile, low injectivity indices, and low wellhead pressures (table 3). The resulting low mass outputs (except for PIN-2D) combined with the acid nature of the brine made the wells noncommercial.

Subsurface Geology

The subsurface stratigraphy of the volcano, based on deep drilling, is shown in figure 6. A pile of andesitic to dacitic volcanic rocks, 1,300 to more than 2,000 m thick, was intersected by the wells. The upper few hundred meters of this unit are largely dacite and andesite breccias and tuffs believed to be erupted from the modern Pinatubo. This sequence is underlain by dacitic lava flows and by a thicker sequence of waterlaid(?) pyroclastics presumed to be deposits of the ancestral Pinatubo. Dioritic dikes (D in fig. 6) of uncertain affinity crosscut the volcanics generally below 200 m in altitude. Only PIN-1 drilled through these volcanics to intersect the ZOC at mean sea level. In this borehole, the ZOC is a chaotic mixture of hornfels and moderately to completely altered microdiorite, micro-quartz monzodiorite, diabase, and gabbroic dikes.

Subsurface alteration is dominated by neutral-pH hydrothermal minerals. The distribution of hydrothermal clays at Mount Pinatubo (fig. 7) does not follow the sequential zonation from smectite, illite-smectite, illite, to biotite typically found in neutral-pH Philippine geothermal systems (Reyes, 1990). First, overlapping of some clay zones suggests that several hydrothermal regimes have existed within the volcanic field. Second, the smectite and illite-smectite zones persist to depths where temperatures are over 300°C. This occurrence is in marked contrast with other Philippine hydrothermal fields, where both smectite and illite-smectite zones are stable only up to temperatures of 180° and 230°C, respectively (Reyes, 1990). Such persistence of these two low-temperature alteration zones to high temperature at Pinatubo may be attributed to low permeability of the host rocks. Third, a well-developed biotite subzone is present in the last 380 m of PIN-2D but is not found in PIN-1 and PIN-3D.

Acid alteration assemblages are restricted to major permeable zones associated with fault intersections (fig. 7). Acid minerals in the drillholes include alunite, diaspore, pyrophyllite, dickite, anhydrite, and pyrite. These minerals form discrete acid horizons 2 to 65 m thick. They occur at altitudes as shallow as 165 m to as deep as -514 m, where temperatures range from 164° to 302°C. The limited distribution of acid minerals, despite the widespread occurrence of acid fluids in the boreholes, implies restricted permeability within the hydrothermal system. The limited structural flowpaths of acid fluids in the reservoir is further supported by the persistence of calcite to more than 1,040 m below sea level (PNOC-EDC, 1990).

Fluid-inclusion heating and freezing measurements were conducted in vein minerals, principally anhydrite, collected from most of the permeable horizons. Homogenization temperatures (Th) are plotted on well temperature-depth curves in figure 8. There is a good correlation between modal Th values and measured stable temperatures in PIN-1 and, to some extent, in PIN-2D. At a depth of 2,000 m in PIN-3D, there is a bimodal distribution of Th values at 290° and 320°C. The latter value is consistent with the present-day measured temperature, implying that in the past, fluids cooler by at least 30°C have circulated in this part of the reservoir.

Measured salinities of fluid inclusions from freezing measurements range from 2,000 to 109,000 ppm Cl equivalent. About 90 percent of the inclusions yielded values of 10,000 to 40,000 ppm, although values of 40,000 to 109,000 ppm were obtained at discrete intervals in PIN-2D and PIN-3D. These latter values are generally higher than the actual reservoir chloride found in the wells. High gas (CO₂) concentrations in the parent fluid of the inclusions, which lower the freezing point depression (Hedenquist and Henley, 1985), may be the cause of their higher salinities compared to actual reservoir chloride. Another possible cause is direct contribution from magmatic HCl gas. In both cases, high gas content of the inclusions's parent fluid is implied.

In general, the salinities increase with depth except in PIN-3D, where salinities have a bimodal peak at 2,000 and 23,000-26,000 ppm at -912 m. These salinities are consistent with the bimodal Th values measured at the same depth in the well, confirming that at least two distinct hydrothermal events have occurred in this part of the reservoir.

Figure 6. Generalized subsurface stratigraphy of Mount Pinatubo volcano. Symbols are the same as in figure 2. D=diorite dike.

Figure 7. Hydrothermal clay zones in Mount Pinatubo wells. Isotherms are based on downhole surveys.

Figure 8. Temperature-depth plot of Mount Pinatubo wells. The horizontal range on histograms is the range of temperatures (scale at base of figure) at various depths, inferred from fluid inclusion homogenization. The height of each histogram is the frequency of each temperature reading, in increments of 1 to 10 readings (the latter is the highest value at 1,250 m depth in well PIN-1). KT-14 (PIN-1) and KT-8 (PIN-3D) temperature profiles were measured temperatures 84 and 5 days after completion, respectively. KT-8 (PIN-2D) profile was measured 15 days after flow test.

Chemistry of Reservoir Fluids

Selected water chemistry of the wells is shown in table 4. Discharge and downhole fluids are moderately to highly acidic Na-Cl waters. Calculated reservoir chloride concentrations range from 6,000 to 10,000 ppm in PIN-1, from 2,000 to 3,000 ppm in PIN-2D and from 5,000 to 20,000 ppm in PIN-3D. The lower values for PIN-2D, despite its high temperature, may be due to the contamination of the brine by low-chloride drilling fluids brought by the limited discharge period. Sulfate concentrations are high particularly in PIN-2D, where they reach 2,000 to 3,000 ppm. Mg and Fe are present in anomalously high concentrations.

Silica geothermometry for the well waters yields temperatures that are close to, but generally lower than, measured. Calculated temperatures are 224-281°C for PIN-1, 314°C for PIN-2D, and 310°C for PIN-3D.

On a Na-K-Mg ternary diagram (fig. 9), PIN-3D and PIN-1 waters fall on the partially equilibrated field, whereas those of PIN-2D, and nearly all the thermal springs, can be classified as immature waters. Two distinct trends are apparent in figure 9. PIN-1 falls on a tie line with the Dagsa, Asin, and Cuyucut springs and yields a Na-K temperature of 240°C, 20°C lower than measured. A second trend is defined by the waters of PIN-3D, PIN-2D, and the Pinatubo solfatara pool. Thus, the fluid supplying the Pinatubo solfatara is probably the same fluid tapped by PIN-2D and PIN-3D. The Na-K temperature for PIN-3D and PIN-2D ranges from 300 to 320°C, close to actual measured temperatures in both wells.

The geochemical trends above suggest two possibilities: one, that two separate convective systems exist in Mount Pinatubo; or two, that one large hydrothermal system is present but fluids are differentially cooled and diluted prior to their discharge (Ruaya and others, 1992). The immature character of the well waters may be due to the limited discharge period. While this may be true for PIN-2D, and to some extent to PIN-3D, it probably cannot apply to PIN-1, which was discharged for 138 days. Thus, nonattainment of equilibrium with the reservoir rocks appears to be an inherent characteristic of the Mount Pinatubo hydrothermal fluids. This characteristic most likely reflects the poor permeability of the hydrothermal system.

Selected gas chemistry data are listed in table 2. CO₂ is the principal gas specie in the steam, followed by H₂S. H₂, N₂, and Ar are present in varying amounts. On a CO₂-N₂-Ar ternary diagram (fig. 10), gases from the solfatara and the wells, PIN-2D in particular, have compositions similar to volcanic gases from White Island volcano (Giggenbach, 1987) and selected Japanese volcanic areas (Kiyosu, 1985; Kiyosu and Yoshida, 1988). The CO₂/N₂ ratio of PIN-2D (192) is comparable to those of White Island (180), while the N/Ar ratios of PIN-2D (1,437) and PIN-3D (874-2,760) fall between those of White Island (800) and selected Japanese volcanoes (4,250). Magmatic contribution to the Mount Pinatubo gases is thus implied.

Figure 9. Na-K-Mg ternary diagram (Giggenbach, 1988) of Pinatubo spring and well waters. P1, P2, and P3 refer to water samples from PIN-1, PIN-2D, and PIN-3D, respectively. Spring abbreviations are given in table 1. Dashed lines represent trends discussed in text.

Figure 10. CO₂-N₂-Ar ternary diagram (Giggenbach, 1987) of Pinatubo solfatara and well gases.

Table 4. Selected chemical analyses of discharge and downhole waters from Pinatubo wells.

[WHP = wellhead pressure; H = enthalpy. Concentrations in milligrams per kilogram]

Stable Isotope Data

Figure 11 shows a plot of ¹⁸O versus D for the thermal springs and wells PIN-2D and PIN-3D. Most of the thermal springs fall along or near the meteoric water line, affirming their meteoric or diluted steam-condensate origin. Although the 13 ¹⁸O shift in PIN-2D may be attributed to rock-water interaction, the large shift in D suggests otherwise. In fact, the isotopic composition of PIN-2D is close to the average values for high-temperature fumarolic steam from volcanic areas in Japan (Matsuo and others, 1975). PIN-3D fluids show a much smaller shift in D compared to PIN-2D but still fall within the "primary magmatic water" area (Taylor, 1979).

The chloride-bearing springs of Asin and Cuyucut show small positive ¹⁸O and D shifts and fall either on a mixing or an evaporation line connecting PIN-2D to the thermal springs. The relationship, if any, between PIN-2D and PIN-3D is less clear. Vaporization of PIN-2D waters at 310°C can shift the isotopic values of the residual liquid in the direction of PIN-3D in figure 11. The magnitude of such shift, however, is much too small to approximate PIN-3D's actual composition (fig. 11). A possible explanation for the isotopic composition of the PIN-3D steam is evaporation of fluids previously modified by rock-water interaction. This is consistent with the partially equilibrated character of PIN-3D waters (fig. 9).

In summary, the ¹⁸O and D signatures of the well waters are comparable to those of high-temperature volcanic gases. This supports the contention that the hydrothermal system has a significant magmatic input.

Figure 11. ¹⁸O versus 9.0 D plot of Mount Pinatubo surface and well waters. Isotopic composition of high-temperature volcanic gases (HTVG) taken from Matsuo and others (1975). Field of primary magmatic water taken from Taylor (1979). Arrow indicates isotopic shift expected for evaporation of PIN-2D waters at 310°C. See text for discussion.

A Model of the Pre-1991 Mount Pinatubo Hydrothermal System

The geological, chemical, and isotopic data presented above show that the Mount Pinatubo hydrothermal fluids are not in chemical equilibrium with reservoir rocks dominated by neutral-pH alteration assemblages. Temperatures in excess of 300°C, moreover, suggest that a more efficient heat-transfer mechanism is operative than those associated with typical hot-water convective systems (Ruaya and others, 1992). A conceptual model of the pre-1991 Mount Pinatubo hydrothermal system is shown in figure 12. The temperature contours and the approximate location of the two-phase region are based on downhole observations.

We propose that the hydrothermal system was heated by a partially cooled or cooling magma chamber related to the emplacement of the Pinatubo dome ~500 years ago (Delfin and others, 1992). Hot vapors released from this cooling magma rise through fractures, heating the volcanic rocks, and condense in deep recharge waters. This results in the production of a hot two-phase zone beneath the dome where PIN-2D and PIN-3D bottomed. Further ascent of these waters, and interaction with the rocks and shallow ground waters, results in neutral SO₄-HCO₃-Cl springs such as Mamot, Dangey, Pajo, Pula, and Kalawangan. The chemical and isotopic signatures of the original magmatic vapors, however, are retained in the gas ratios and isotopic composition of discharges from PIN-2D and PIN-3D.

The thermal energy of the convecting magmatic vapor induces the formation of convective cells of largely meteoric water on the flanks of Mount Pinatubo. At least two separate brine systems have evolved. One is located northwest of the pre-1991 dome, and part of this--probably near the intersection of the brine and two-phase region--was intersected by PIN-3D. The other brine system lies to the southeast. The northern and deeper portion of this brine system is represented by PIN-1 fluids, while its major lateral plume is manifested by the chloride-bearing springs at Dagsa, Asin, and Cuyucut.

The poor permeability in the wells is consistent with Giggenbach and others' (1990) model of "magmatic-hydrothermal systems," where an extensive, hydrothermally sealed carapace is likely to form. The formation of this carapace is attributed to such processes as evaporation of locally formed brines, deposition of solids from rising waters, and densification of volcanic rocks by alteration. The carapace of the Pinatubo system most likely lies along the boundary of the two-phase zone and the brine systems, or approximately near the 300°C contour in figure 12. At PIN-2D and PIN-3D, the depth corresponding to the 300°C isotherm is characterized by intense argillization and silicification (PNOC-EDC, 1990).

Figure 12. Conceptual model of the pre-1991 Mount Pinatubo hydrothermal system (modified from Ruaya and others, 1992).

The Hydrothermal System and the 1991 Eruptions

Mount Pinatubo came abruptly to life on April 2, 1991, with a series of phreatic to phreatomagmatic(?) explosions that formed three closely spaced craters on the northeast flank of the volcano (Punongbayan and others, 1991). At least six high-discharge fumaroles formed during or soon after these eruptions along an east-northeast-trending belt parallel to the Tayawan fault (fig. 5). On June 12-15, the volcano erupted violently, discharging 3.7-5.3 km³ of dacitic magma (W.E. Scott and others, this volume).

The geothermal exploration conducted from 1982 to 1990 revealed that the Mount Pinatubo hydrothermal system had a significant magmatic input. This magmatic signature, unclear during the surface exploration stage, became distinct from the geochemical and isotopic signature of the deep reservoir fluids. Prior to April 2, 1991, however, there was little or no evidence of instability in the Mount Pinatubo hydrothermal system (Delfin and others, 1992). There were no significant changes in the hot spring and gas chemistry in the ~10 preceding years, nor were any historic phreatic explosion craters known in the area. The April 2 explosions and the ensuing intense seismic swarm were the first indications of a sudden magma intrusion event below the high-temperature hydrothermal system. Thus, prior to April 2, the accumulated information from geothermal studies did not indicate an imminent eruption.

Exploration Lessons

There is evidence from deep drilling at Mount Pinatubo that hot neutral-pH chloride waters, suitable for electric power generation, existed in the area but at some distance from those tested by deep drilling. If the model of the Mount Pinatubo hydrothermal system (fig. 12) is considered representative of most "magmatic-hydrothermal systems" in andesitic-dacitic volcanoes, high temperatures (>=300°C) may be expected in the central parts of such reservoirs. The model, however, clearly implies that acidic magmatic fluids, instead of neutral-pH chloride waters, are to be expected in the central region of such systems. Such acidic magmatic fluids, while attractive for their heat content, are not economically exploitable at present, owing to their high acidity and high gas content. Moreover, the limited permeability of the central part of such systems, as demonstrated in Mount Pinatubo, makes for poor fluid production.

A far more important lesson from Mount Pinatubo, as well as from recent eruptions of Nevado del Ruiz and El Chichón, concerns the need to assess the suitability for long-term development of geothermal systems associated with young volcanoes (Casadevall, 1992). Although volcanic hazard studies should be an important part of this assessment, other studies focusing on the degree of magmatic involvement in the hydrothermal system must be included. In this regard, the chemical and isotopic composition of waters and gases from wells and thermal features probably provide the best clues in determining magmatic input. Unfortunately, in most cases, such magmatic input becomes firmly established only after the expensive steps of drilling and testing deep wells have been completed.

Acknowledgments

We would like to thank the management of PNOC-EDC for permission to publish this paper. Willie Carmen and Manny Teocson are thanked for drafting the figures. Reviews of this manuscript by Tom Casadevall, Marianne Guffanti, David Sussman, and Chris Newhall are gratefully acknowledged.

References Cited

Apuada, N.A., 1988, Interpretation of the geo-electrical survey in Mt. Pinatubo geothermal prospect, Philippines: Pisa, International Institute for Geothermal Research, Diploma report, 29 p.

Bogie, I., 1984, Critique of Pinatubo geoscientific reports: Kingston Reynolds Thom and Allardice (KRTA) internal memorandum, 16 p.

Casadevall, T.J., 1992, Pre-eruption hydrothermal systems at Pinatubo, Philippines and El Chichón, Mexico: Evidence for degassing magmas beneath dormant volcanoes: Reports of the Geological Survey of Japan, v. 279, p. 35-38.

Clemente, V.C., 1984, A re-evaluation of the Mt. Pinatubo geochemistry: Philippine National Oil Company-Energy Development Corporation (PNOC-EDC) internal report, 16 p.

de Boer, J.Z., Odom, L.A., Ragland, P.C., Snider, F.G., and Tilford, N.R., 1980, The Bataan orogene: Eastward subduction, tectonic rotations, and volcanism in the western Pacific (Philippines): Tectonophysics, v. 67, p. 305-317.

Defant, M.J., Jacques, D., Maury, R.C., and de Boer, J.Z., 1989, Geochemistry and tectonic setting of the Luzon arc, Philippines: Geological Society of America Bulletin, v. 101, p.-663-672.

Delfin, F.G., 1984, Geology and geothermal potential of Mt. Pinatubo: Philippine National Oil Company-Energy Development Corporation (PNOC-EDC) internal report, 36 p.

Delfin, F.G., Sussman, D., Ruaya, J.R., and Reyes, A.G., 1992, Hazard assessment of the Pinatubo volcanic-geothermal system: clues prior to the June 15, 1991 eruption: Geothermal Research Council Transactions, v. 16, p. 519-527.

Esperidion, J.A., 1984, Resistivity traverse survey of Mt. Pinatubo: Philippine National Oil Company-Energy Development Corporation (PNOC-EDC) internal report, 20 p.

Giggenbach, W.F., 1987, Redox processes governing the chemistry of fumarolic gas discharges from White Island, New Zealand: Applied Geochemistry, 1987, p. 143-161.

------1988, Geothermal solute equilibria: derivation of Na-K-Mg-Ca geoindicators: Geochimica et Cosmochimica Acta, v. 52, p. 2749-2765.

Giggenbach, W.F., Garcia, P., N., Londoño C., A., Rodriguez V., L., Rojas G., N., and Calvache V., M.L., 1990, The chemistry of fumarolic vapor and thermal-spring discharges from the Nevado del Ruiz volcanic-magmatic-hydrothermal system, Colombia: Journal of Volcanology and Geothermal Research, v. 42, p. 13-39.

Hedenquist, J.W., and Henley, R.W., 1985, Importance of CO₂ on freezing point measurements of fluid inclusions: Evidence from active geothermal systems and implications for epithermal more deposition: Economic Geology, v. 80, p. 1379-1406.

Kiyosu, Y., 1985, Variations in N2/Ar and He/Ar ratios of gases from some volcanic areas in northeastern Japan: Geochemical Journal, v. 19, p. 275-281.

Kiyosu, Y. and Yoshida, Y., 1988, Origin of some gases from the Takinboue geothermal area in Japan: Geochemical Journal, v. 22, p. 183-193.

Matsuo, S., Suzuki, T., Kusakabe, M. Wada, H. and Suzuki, M., 1975, Isotopic and chemical compositions of volcanic gases from Satsuma-Iwojima, Japan: Geochemical Journal, v. 8, p. 165-173.

Newhall, C.G., Daag, A.S., Delfin, F.G., Jr., Hoblitt, R.P., McGeehin, J., Pallister, J.S., Regalado, M.T.M., Rubin, M., Tamayo, R.A., Jr., Tubianosa, B., and Umbal, J.V., this volume, Eruptive history of Mount Pinatubo.

Philippine Bureau of Mines, 1963, Geologic map of the Philippines: Manila, Philippine Bureau of Mines, scale 1:1,000,000, 9 sheets.

PNOC-EDC, 1990, Mt. Pinatubo resource assessment report: Philippine National Oil Company-Energy Development Corporation (PNOC-EDC) internal report, 46 p.

Punongbayan, R.S., Newhall, C.G., Ewert, J., Sussman, D., and Arevalo, E., 1991, Pinatubo-April 2, 1991 phreatic event and fumarolic activity summarized: Global Volcanism Network, v. 16, p. 12-13.

Reyes, A.G., 1990, Petrology of Philippine geothermal systems and the application of alteration mineralogy to their assessment: Journal of Volcanology and Geothermal Research, v. 43, p. 279-309.

Ruaya, J.R., Ramos, M.N., and Gonfiantini, R., 1992, Assessment of magmatic components of the fluids at Mt. Pinatubo volcanic-geothermal system, Philippines from chemical and isotopic data: Reports of the Geological Survey of Japan, v. 279, p. 141-151.

Scott, W.E., Hoblitt, R.P., Torres, R.C., Self, S, Martinez, M.L., and Nillos, T., Jr., this volume, Pyroclastic flows of the June 15, 1991, climactic eruption of Mount Pinatubo.

Taylor, H.P. Jr., 1979, Oxygen and hydrogen isotope relationships in hydrothermal mineral deposits, in Barnes, H.L., ed., Geochemistry of hydrothermal ore deposits (2d ed.): New York, Wiley-Interscience, p. 236-277.

Villaseñor, L.B., 1984, Summary of geochemical results--Pinatubo geothermal prospect: Philippine National Oil Company-Energy Development Corporation (PNOC-EDC) internal report, 9 p.

FIRE and MUD Contents

PHIVOLCS | University of Washington Press | U.S.Geological Survey
This page is <https://pubs.usgs.gov/pinatubo/delfin/>
Contact: Chris Newhall
Last updated 06.10.99