V24B-01 INVITED 16:00h
Overview and the achievement of the Unzen Scientific Drilling Project
Unzen volcano is an active volcano in SW Japan, and its 1990-95 eruption caused the frequent dome-collapse-type pyroclastic flows and associated debris flows. Detailed observations have enabled the constructions of magma ascent and eruption models. Unzen was born in 0.5 Ma and has grown inside the regional tectonic graben. Most of its eruption products have been thickly accumulated inside the active graben more than 1200 m beneath the current surface. Unzen Scientific Drilling Project (USDP) is a six-year term international project started in April 1999, co-sponsored by the Japanese Government and International Continental Scientific Drilling Program. The project includes not only scientific drillings but also related geological, geophysical and geochemical studies to totally understand the growth history, subsurface structure and magma ascending mechanism of Unzen Volcano. The highlight of USDP is to drill into the still-hot magmatic conduit of 1990-95 eruption (conduit drilling) to prove the magma ascent model based on detailed observations during the eruption, and is scheduled during the Phase II (2002-2005) of the project. Phase I of USDP (1999-2002) consists of drilling two boreholes (USDP-1: 752 m and USDP-2: 1462 m) into the flanks of Unzen Volcano and conducting associated researches mainly to reveal the time-integrated process of Unzen volcano; i.e. three-dimensional structure and the growth history of the volcano. Pilot drilling (USDP-3, 350m) associated with the scientific and drilling strategy for the conduit drilling was also conducted. Following the scientific results of Phase I, Phase II was approved by the funding agency and the conduit drilling started in February 2003. Repeated troubles necessitated some modifications of the drilling plan, but the drilling was successfully ended in July 2004 by penetrating into the 1990-95 magmatic path about 1500m beneath the summit of Unzen volcano. Construction of an integrated evolutional and magmatic model of Unzen will be the final goal combining all scientific results in the end of the project.
V24B-02 16:15h
Growth History Of Unzen Volcano, Kyushu, Japan _| Results of Two Flank Drillings of Unzen Scientific Drilling Project
Unzen volcano is an active volcano composed of lava domes, thick lava flows and pyroclastic deposits of hornblende andesite to dacite. Tectonically active Unzen graben dissects volcanic edifices of the volcano. During the phase I of the Unzen Scientific Drilling Project (USDP), two drillings were conducted at the northeastern (USDP-1: 752 m depth) and eastern (USDP-2: 1463 m depth) flanks of the volcano, respectively, to fully recover accumulated deposits of the volcano hidden beneath the younger eruptives. Extensive K-Ar and 40Ar/39Ar age determinations have also been conducted on both surface rocks and drilling cores. <BR>Unzen volcano starts to grow at 0.5 Ma above the Pre-Unzen pyroxene andesite (0.5-0.8 Ma). Unzen volcano has been divided into three volcanic stages, Early and Late stages of the Older Unzen and the Younger Unzen. <BR>The Early stage of the Older Unzen (0.3-0.5 Ma) products consist of pumice-rich pyroclastic flows, block and ash flows, associated volcaniclastic debris flows and thick lava flows. The north- and south-dipping fans spreading outside the graben are sharply cut by the faults. This suggests that Unzen volcano grew rapidly in the first 200,000 years of its history and formed a conical volcanic edifice. <BR>The Late stage of the Older Unzen (0.15-0.3 Ma) products mainly fill in the graben. In the western half of the deposits of this stage, thick lava flows cover widely inside the Unzen graben. On the other hand, thick alternated piles of pyroclastic deposits were recovered both from USDP-1 and -2 cores. In the USDP-2 core, phreatomagmatic deposits about 250 m thick with essentially abundant glass materials of ca. 0.3 Ma. These findings suggest that rapid subsidence of the Unzen graben at around 0.3-0.2 Ma led strong interaction between the magma and groundwater. <BR>Younger Unzen volcano (0-0.15Ma) is composed of four edifices, Nodake, Myokendake, Fugen-dake and Mayuyama volcanoes, all locate in the eastern half of Unzen volcano. Block-and-ash flow deposits or related debris flow deposits were continuously supplied to the eastern flank of the volcano.
V24B-03 INVITED 16:30h
Unzen Scientific Drilling Project: Challenging drilling operation into the magmatic conduit shortly after eruption
Drilling operation was aimed at penetration into the core of the volcano 8 years after eruption of Unzen, including directional drilling in high temperature and with high inclination. The project started with fixing drilling site. Scientists and drilling engineers agreed to settle it at the northern slope of Mt. Unzen at 840 m asl, and the drilling target was set at sea level. Drilling operation was started in Feb. 2003. In the shallow section, frequent lost circulation and accidental side-track occurred due to the unconsolidated zone, and caused_@many troubles. Although the drilling was delayed, we succeeded in drilling down to 396m with the inclination of 25 degree in 17-1/2 inch hole and 13-3/8 inch casing section. 12-1/4 inch hole was drilled using TDS, EM-MWD, and DHM. When the inclination was built up to 75 degree at 795 m, we changed the drilling mode of trajectory control to keep the angle. A large fracture of total loss was encountered at 807m, and serious cuttings bed occurred. The latter made the drilling impossible to continue. Then, we inserted 9-5/8 inch casing down to 796 m. Trajectory correction runs was completed in 8-5/8 inch hole, and 7 inch casing was set down to 1550m. In 6-1/4 inch hole, though EM-MWD and DHM were not used, drilling inclination and azimuth were stable. Spot coring was started at 1582 m, the levels of spot coring depth were chosen based on the data of temperature measurement and cuttings observation. Though the drilling exceeded 1800m, the original target depth, drilling was continued, because we could not encounter the high temperature conduit at that time. Finally, the well reached the 1995 m, and succeeded in taking cores highly probable of magmatic conduit in July 2004. We could carry out geophysical logging mostly throughout the whole sections. Spot coring were done at 16 times; its total length was 75m. Although the highest measured temperature was 155 deg. C, the formation temperature may reach at least 200 deg. C. The reasons of smooth drilling after 9-5/8 inch casing installation were easy hole cleaning due to smaller diameter, least lost circulation, and no gas invasion.
V24B-04 16:45h
Unzen Volcano Scientific Drilling: Well Logging Data of the USDP-4
The Unzen Volcano Scientific Drilling Project (USDP) has been conducted to target the magma conduit shortly after the 1990-1995 eruption. After two drillings of 752 m and 1463 m deep at the flank site, the conduit surveying well (USDP-4) was drilled to the depth of 1995.75 m in the mountainside to clarify the ascending and degassing mechanisms of magma. We have conducted physical logging in the USDP-4 well to elucidate the structure and material properties in and around the conduit. The logging items are as follows: Gamma Ray (167 to 1780 m), Resistivity (167 to 1795 m), Self-Potential (167 to 1775 m), Density (392 to 1782 m), Neutron Porosity (770 to 1777 m), Sonic velocity (392 to 1787 m), Full-bore Formation Micro Imager (FMI : 167 to 1540 m), Formation Micro Scanner (FMS : 1550 to 1791.5 m) and VSP (237 to 737m). Because of the high inclination of this well (Sakuma et al., this meeting), we used the Tough Logging Condition System (TLCS) below the depth of 800 m where the well inclination is up to 70 degrees. We had some concern because of a possible well collapse and high temperatures at the conduit zone before drilling. However, a good well condition and low temperature enabled us to obtain good logging data from this well. Comparing the logging data and lithology, determined mainly from drilled cores and cuttings (Nakada et al., this meeting), we can make clear the features of its formation and material properties found within the well. Gamma Ray varies between 40 API to 100 API, with the high (90 to 100 API) value coinciding with a lava dike. Resistivity structure can be classified into 5 layers. The value of resistivity above 240 m, 240 to 550 m, 550 to 1100 m, 1100 to 1760 m and below 1760 m are a few hundred ohm-m, 500 to 1000 ohm-m, about 100 ohm-m, about 10 ohm-m, about 100 ohm-m, respectively. The lava dike indicates a characteristic feature of about 100 ohm-m even though it is distributed in the 10 ohm-m layer. P-wave velocity varies 3 to 5 km/s through all depths. The velocity of Pyroclastic rocks, lava flow, volcanic breccia and lava dike is 3 km/s, about 4.5 km/s, 4 km/s and 5 km/s, respectively. The density of Pyroclastic rocks, lava flow, volcanic breccia and lava dike is 2.3 to 2.4 g/cm3, about 2.5 g/cm3, about 2.5 g/cm3 and 2.6 g/cm3 respectively. The porosity of the lava flow and lava dike is low (0.1), and high porosity (0.4) coincides with low resistivity (10 ohm-m), low velocity and low density zones. Furthermore, the high porosity zone around 1700 m coincides with the zone where a high NaCl concentration in the drilling mud was observed during the drilling. The anomaly of Self-Potential recognized around the lava dike seems to be related to the fluid flow. FMI and FMS images show not only the boundaries of lava dikes but also vertical pyroclastic veins in the conduit zone. Some conductive layers are recognized in the lava dikes. Drilling induced tensile fractures and borehole breakouts were also partly recognized. According to the FMS analysis, the main dips and strikes of the boundary of lava dikes and pyroclastic veins are a high dip and in an east-west direction. These measurements are indispensable to better understand the dynamics of magma emplacement, tectonic settings, and mechanical controls on the orientation of fractures.
V24B-05 17:00h
Conduit drilling at Unzen volcano, Japan: core description and interpretation
A 1996-m-long hole has been drilled on a slanted trajectory that passed beneath Unzen volcano, Japan. The hole penetrated central conduit region of the volcano at ~1600 m below the summit. Drill cores at 50 m intervals were recovered from the hole. The principal facies of the drill cores in the conduit region are polymictic breccia and coherent dacite. The polymictic breccia is non-stratified, poorly sorted and made up of heterolithologic, polyhedral dacite clasts 10-50 mm across, which are embedded in a cogenetic matrix. Larger clasts up to 50 cm across rarely occur. The clasts consist of non-vesicular, porphyritic dacite, which varies in color (greenish grey to grey), alteration, phenocryst size and phenocryst proportion. They show clast-rotated texture, and no jigsaw-fit texture. The edges of the clasts are commonly rounded. The matrix of the breccia is reddish grey, dense (low porosity) and composed of angular dacite fragments up to 5 mm across. The coherent dacite is grey, massive and porphyritic. It contains plagioclase and hornblende phenocrysts. The matrix of the dacite is mostly crystalline, but gradation from light grey, crystalline dacite to dark grey, glassy dacite occurs. These polymictic breccia and coherent dacite have been in places intruded by veins up to 10 mm wide. The veins vary in color (white, greenish grey and black) and components (fine-grain cray minerals, coarse rock fragments). Some veins show laminations along the vein wall. The morphological features of the polymictic breccia suggest that the breecia formed within the conduit region of the volcano in response to fragmentation of rocks due to explosions, but did not extrude from the vent and recycled within the conduit region. The coherent dacite facies is probably a dyke with chilled margins. The veins may have formed by injections of high-temperature gas and/or liquid into fractures around dykes during dyke intrusions. The descriptions of the drill core, together with composition of drill cuttings, electric conductivity of the borehole (FMS image) and seismological data during 1990-1995 eruption, suggest that the conduit region of the Unzen volcano is 300-350 m wide at ~1600 m below the summit, and consists of polymictic breccia and several, parallel, vertical dykes, intruding the breccia. Each dyke is tabular-shaped and 3-30 m thick. Thin dykes (3-8 m thick) have simple, uniformed interior and are inferred to have formed by single injection of magma. On the other hand, thick dykes (26-30 m) have several cooling units within, and probably formed by several magma injections.
V24B-06 INVITED 17:15h
Real Images of Magmatic Conduit: Progress of the Conduit Drilling in Unzen
The conduit of the 1990-1995 lava dome at Unzen was penetrated in the scientific drilling project (USDP). The drilling target was set in the hyprocenter zone of isolated tremors that occurred prior to magma extrusion in 1991. The drilling operation was sponsored by MEXT-Japan and ICDP and completed in the summer of 2004. Drilling was started vertically from the northern slope of Mt. Unzen at 840 m asl and directional to below the summit. The conduit zone was reached near the sea level (ca. 1.3 km below the crater). Rock sampling and logging in and around the conduit were succeeded. The conduit zone consists of multiple EW-oriented vertical dykes, up to about 4 m thick and of different formation ages, within volcanic breccias. Both the dykes and breccias are altered hydrothermally. Pyroclastic veins of the similar orientations are abundant in the conduit zone. The volcanic breccias of the conduit zone are physically homogeneous. Their porosity is less than 0.2 and density is close to 2.5. Mud loss seldom occurred during drilling in the volcanic breccias. These facts imply little cracks developed and very low permeability of the conduit zone. Relatively fresh dyke with the highest temperature was located in the 1975 to 1999 m drilling depth, beneath a little south of the previous Unzen peak. The temperature was as low as 200 deg C, much cooler than expected. It is likely that hydrothermal fluid circulation within the conduit zone accelerated cooling and alteration of the newest conduit. Identification of the conduit as the latest eruption was based on its chemical inconsistency with that of the new dome lava. It is important that the conduit is neither single nor bundled into older ones throughout the growth history of the volcano. Magma prefers intrusion isolatedly from older conduits in every eruption event as far as in this depth. It is likely that forming cracks through one of which magma ascended was recorded as pyroclastic veins and is responsible for isolated tremor events.
V24B-07 17:30h
Transition Of Degassing Processes From Syn-Eruptive To Post-Eruptive Activities At Unzen Volcano
Budget of volcanic gas components during the dome-forming eruption of Unzen volcano, 1991-1995, has been evaluated with the SO2 emission rates, magma effusion rates and volatile contents in melt inclusions (Hirabayashi et al., 1995; Satoh et al., 2003). The estimate concluded a closed-system degassing implying that most of the gas component in the ascending magma was discharged through the summit degassing; there is no need of the gas-loss to the volcanic edifice from the conduit wall nor the excess gas supply from non-erupting magma. As a consequence, the intense gas emission quickly decreased after the end of magma effusion. However, the relatively minor but high-temperature (ca. 400?_qC in 2001) gas emission still continues even after five years. The post-eruptive gas emission is characterized by Cl-rich and S-poor composition, indicating that the acid gases were derived from already degassed magmas. The low delta D of water in the high-T fumarolic gases might also indicate the water discharge from the already degassed magma, either from the hot lava dome or the shallow conduit. The results of the conduit drilling revealed that the shallow conduit was already cooled; therefore the Cl-rich components need to be derived from the lava dome. In contrast, the fumarolic gases also contain high amount of CO2 with the isotopic ratio (delta 13C = -6 per mil) of the typical arc magmatic gas. Emission rate of the magmatic CO2 was estimated to be in the order of 1 t/d, based on the steam flux estimates and composition of the fumarolic gases. If the CO2 was derived from a magma containing 1000 ppm of CO2, which is estimated based of the gas budget during the eruption, the CO2 emission rate of 1 t/d requires magma degassing rate of 1000 t/d, corresponding to 0.5 m3/d, or 1000 m3 for the six-years emission. Although the volume of the source magma is much less than volume of the conduit, the CO2 needs to be derived from much deeper sources, because of the low CO2 solubility in magmas. Therefore the already cooled magma conduit serves also as a gas passage from the deep sources.
V24B-08 17:45h
The Present and Future of Volcano Research Drilling
The success of the Unzen Scientific Drilling Project (USDP) in reaching and sampling the conduit of Unzen Volcano demonstrates that: (1) Directional drilling is applicable to volcanic systems; (2) Geophysical monitoring during eruption and high-resolution surveys afterward provide a reliable view of the magma pathway; (3) At least some of our ideas about syneruptive magmatic behavior appear to be confirmed through this glimpse of new magma emplaced under a load of more than 30 MPa, containing melt that had lost less than half its water; and (4) Temperatures are relatively low, indicating stronger hydrothermal dissipation of magmatic heat than anticipated. There remains, of course, a great deal more to be learned about the conduit of Unzen from detailed analysis of the core samples and interpretation of drilling and logging data. Further analysis will determine whether we have opened a new door to understanding volcanoes or engaged in a one-time trip to the "Moon". While the cost of USDP is unprecedented in volcanology, it is less than one percent of the cost of damage that Unzen did in the 1990-1995 eruption. The USDP conduit data point lies toward one end (excepting very large continental systems) of a continuum in terms of repose time. Unzen has a repose period of a hundred or more years. It is therefore not surprising that the conduit of Unzen is a warm ribbon in a cold edifice. Higher temperatures are to be expected in systems with shorter repose times. This is important because the thermal and mechanical state of the edifice will likely affect the rise of new magma; associated precursory seismicity, deformation, and thermal and gas emission; and subsequent eruptive behavior. Other parameters are of interest as well, such as how recharge from a crater glacier or lake affects the hydrologic regime of a conduit, with resultant phreatomagmatic and hydrothermal activity. And to what extent might forecasts be improved by placing geophysical and geochemical instruments near or even directly in the magma pathway during repose? The fact is that many characteristics of the conduit environment in active systems, characteristics that we have previously only guessed at, can now be measured directly. If we believe that advances in understanding volcanic systems will save lives, then we should drill on, rather than being content with repetitive surface studies of volcanoes. Continued research drilling should be targeted on volcanoes where the risk of eruption to populations is high and where results will span the parameter space of conduit environments. Some of this drilling may yield not just advances in basic science and in eruption forecasting, but geothermal energy as well.