Matsukawa was the first geothermal power plant established in Japan. It started power production in October 1966, and is the only vapor-dominated geothermal field developed to date. The power plant has been continuously producing full power, 22MWe, for almost 27 years. Matsukawa is located about 600km northeast of Tokyo and about 27km northwest of Morioka, Japan. It is in the Hachimant*i volcanic region, one of the most active volcanic regions in Japan. This paper reviews reservoir engineering studies at Matsukawa. This includes a study of the current state of the reservoir studied mainly by pressure buildup tests, a study of the initial state of the reservoir studied mainly by reconstruction of a reservoir pressure profile, and a numerical modeling study of the natural state of the reservoir. Continued pressure buildup tests since 1986 have revealed that there is a lateral steam flow from southwest to northeast in the Matsukawa vapor-dominated reservoir, and most of the steam is suplied from southwest of the development area. This result suggests that the vapor-dominated reservoir extends further southwest than the area now being exploited. These conclusions are supported by production records and chemical data of produced steam. The study on the natural state of the Matsukawa geothermal reservoir has revealed that there was a thin vapor-dominated zone at the shallow part of the reservoir (around 300m to 400m depth) and the current production zone (800m to 1300m depth) was filled with liquid before exploitation. Early production wells produced wet steam with some hot water at first, but they turned to produce only dry steam after a production period of 6 months to 1 year, because of the existence of low permeability aureole around the reservoir and high heat flow. Estimated conductive heat flux, 1.5W/m2 is as high as that of The Geysers. The natural state modeling study showed that the model of the initial state of the reservoir, described above, was feasible. The results also indicated that the low permeability aureole was very important for the evolution of its natural state and also for production of superheated steam from the liquid zone below the thin vapor-dominated zone in the shallow part of the reservoir. Initial temperature distribution and the results of the simulation study suggest that there is an extensive heat source in the southwestern part of the reservoir.
We carried out electrical resistivity, 30cm depth temperature and gamma-ray spectrum exploration at a fumarole zone in Kuju-Iwouyama, the central Kyushu island, southwest Japan, to clarify a subsurface structure beneath high temperature fumaroles. A survey line, 200m long, was set across fumaroles having a maximum temperature of approximately 260°C. Apparent resistivity of both 'Wenner' and 'Eltran' type four electrodes arrays were measured every 4m along the line. 30cm depth temperature and gamma-ray spectrum of 40K, 214Bi and 208Tl were measured at 8m intervals. High underground temperature zones (>90°C) were observed at the fumarole zone and up-flowing zone of thermal water. Relatively low gamma-ray intensities and high Bi/K, Bi/Tl values were recognized around the fumarole zone. A steam reservoir beneath the fumarole zone was detected as a high resistivity zone (>1000 ohm, m) which was 60m wide and 30m deep. We have discussed the physical properties of the steam zone in the volcano.
AE (Acoustic Emission) source location technique is essential to evaluate dynamic behavior of subsurface cracks in geothermal developments. The triaxial hodogram method has an advantage, especially in the evaluation of deeper geothermal reservoir, because it can determine AE source locations by single point obserbation. However, source locations estimated by the simple triaxial hodogram method are directly afected by subsurface velocity structure. This paper describes a modeling method of subsurface velocity structure to reduce the location error in the multi-point triaxial AE measurement. The model is optimized with a evaluating function in terms of the distances between the located sources at each detector. Feasibility of the proposed method has been confirmed in both simulation and field studies using an artificial seismic source. We also applied this technique to AE source locations in geothermal fields. The results have demonstrated the effectiveness of the technique for the precise evaluation of geothermal resevoir.
A knowledge of stress corrosion cracking of rocks under hydrothermal conditions is required to evaluate longterm geothermal reservoir performance. In this study, stress corrosion cracking tests were carried out on Granite in high temperature water environments from 150°C to 350°C using Pre-notched thick-walled cylindrical specimens. The inner bore of the specimen was hydraulically pressurized up to a predetermined level to give an initial stress intensity factor Ki to the notch tip. This caused radial flow of hot water through the specimen. The pressure was kept constant until the specimen failed, and the time to failure tf was measured. During the tests, the variation in permeability with time was measured by collecting water drained from the specimen. The following results were obtained. (1) The mechanism of fracture changed with test conditions. Specimen failure was generally caused by a straight crack extending from the notch tip to the outer wall. On the other hand, under the conditions of low Ki and high temperature (≥300°C), a crack initiated at the notch tip branched out during the extension and developed into multiple cracks. (2) Crack extension by stress corrosion occurred preferentially in the direction parallel to the lift plane where pre-existing microcracks are oriented preferentially, because the microcracks had higher sensitivity to stress corrosion. (3) Permeability measurements showed that pre-existing microcracks, which were paths of water, grew and widened markedly owing to stress corrosion and rock dissolution. The developing rate was enhanced in high temperature conditions, especially above 200°C. (4) Under the conditions of high temperature, stress corrosion cracking rate of pre-existing microcracks in the vicinity of a main crack tip might be enhanced up to about that of a main crack with low Ki. Development of multiple cracks was thought to be a result of marked extension of the microcracks and cupling of the extended microcracks with the main crack. (5) Stress corrosion cracking laws were deduced from relationships between Ki and tf. Results revealed that stress corrosion cracking rate was strongly enhanced by water permeation and high tem-perature, especially above 200°C.
Three dimensional axisymmetric analysis on heat transfer is performed to examine heat extraction rate and behavior of the solidified magma region during heat extraction from molten magma by the so-called open heat extraction system. In the analysis, it is assumed that heat transfer in the solidified magma is almost horizontal, because the heat exchange system is expected to be very slender in the vertical direction. The variation of temperature of the solidified magma with respect to depth is mainly due to heat transfer from the solidified magma to water which is flowing in the vertical direction. It is revealed that the optimum heat extraction rate can be achieved for the flow rate of injected water of 10kg/s∼20kg/s depending on the value of coefficient of heat transfer between the solidified magma and water. The solidified region grows up with time after the onset of heat extraction, but the growth rate decreases with time and is very small after 3000 hours. If heat extraction is stopped, the solidified region would melt down. This phenomena of melting down is also examined.
Takigami geothermal field, central Kyushu, Japan, is a hot water dominated geothermal system. A new 25MW geothermal power plant is expected to commence the operation in 1996 at this field. A short period production and reinjection test was conducted from November 1991 to February 1992. We conducted repeated gravity surveys at specified intervals (one to three months) for two years before and after the test period to detect gravity changes caused by production and reinjection. Gravity changes of up to 150 microgals have been observed during the two years. Most of such large changes are found to be due to seasonal variations of shallow water level. After correction of such a shallow water effect, we detected gravity increases of up to 30 microgals in the reinjection area. This change is reasonable if we assume that the reinjected water stays for a while beneath the reinjection area. Gravity decreases of up to 40 microgals were also detected in and eastward of the production area. The gravity decreases have not become apparent by the end of March, that is, one month and a half after the test finished.