NOTICE CONCERNING COPYRIGHT RESTRICTIONS

Transcription

1 NOTICE CONCERNING COPYRIGHT RESTRICTIONS This document may contain copyrighted materials. These materials have been made available for use in research, teaching, and private study, but may not be used for any commercial purpose. Users may not otherwise copy, reproduce, retransmit, distribute, publish, commercially exploit or otherwise transfer any material. The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material. Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specific conditions is that the photocopy or reproduction is not to be "used for any purpose other than private study, scholarship, or research." If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of "fair use," that user may be liable for copyright infringement. This institution reserves the right to refuse to accept a copying order if, in its judgment, fulfillment of the order would involve violation of copyright law.

2 + Electric Geothermal Resources Council Transactions, Vol22, September 2023, 1998 Microgravity Monitoring for the Oguni Geothermal Reservoir System, Japan A Preliminary Correction of Seasonal Gravity Changes Before Exploitation Shigetaka Nakanishi', Keiji lguchi +, Chitoshi Akasaka', and Nobuyuki lwai + + Power Development Co., Ltd. (EPDC), Ginza, Tokyo, Japan + Kaihatsu Computing Service Center, Ltd. (KCC), Fukagawa, Tokyo, Japan + ABSTRACT In the Oguni field in Japan, 20 MW geothermal power plant is going to be constructed. To iden* and study the reservoir behavior after exploitation, microgravity changes at 41 benchmarks have been monitored since May of In the background level measurements at the present stage, it was clear that observed gravity changes included seasonal variations with microgals due to water level changes of shallower groundwater aquifer as well as changes due to the another sources of noises. To remove the effect due to the shallow groundwater level changes, we tried at first to calibrate microgravity data observed at a benchmark near a shallow aquifer well by the changes in groundwater table depth observed in the well. To extend the calibration method to another data of microgravity changes at different stations, a simple empirical equation was applied to precipitation data to reproduce the water level changes that observed in the shallow aquifer well. Microgravity changes calculated fiom the inferred water level changes estimated fiom the precipitation data were also compared with the actual gravity measurement data, and the reasonable match was obtained. It suggests that the effect of water table changes of the shallow aquifer on the surface microgravity monitoring could be compensated within the accuracy of *lo microgals by the method using the data of microgravity measurements at each benchmark before exploitation coupled with precipitation data, even if there are no ground water level observation wells. The changes of 130 microgals over the production period of 30 years are estimated at the Oguni field by a calculation using the gravity postprocessor based on the existing best reservoir model of the field. It is concluded that the correction of the seasonal gravity changes by such a method is essential for the refinement of the reservoir model and the reservoir management after exploitation at the Oguni field. Introduction The Oguni geothermal field is located in the central part of Kyushu island, southwestem Japan, as indicated in Figure 1. 4 YAMAKAWA 3OM W Mom OONUMA 1 OMW Figure 1. Location of the Oguni field and operating geothermal power plants in Japan. Electric Power Development Company Ltd. (EPDC) has been developing Oguni geotheml power station with a capacity of 20Mw aiming start operation in the year of In order to evaluate the reservoir performance and to conduct proper management of the reservoir behavior after field exploitation, surface gravity monitoring was planned. The repeat measurements of surface microgravity changes at 11

3 41 benchmarks in the field were started in May of 1996 to get enough data of background level to appraise noises and unexpected changes of gravity prior to power station startup. Microgravity data measured so far indicate that seasonal variations with microgals probably due to water level changes of shallower groundwater aquifer are included as well as microgravity changes by another sources of error. In this paper, preliminary studies of the correction of microgravity changes caused by shallow groundwater level changes are discussed. The result of the numerical reservoir simulation study using the STAR geothermal simulator and the gravity postprocessor (Pritchett, 1995) based on the best present state model is also described, which gave us considerable ideas to make the gravity monitoring program to be successful. Geothermal System of the Oguni Field The Oguni field lies on the northwestern boundary of the Kuju Uplift Belt, where the basement drops steeply to the northeast towards the Shishimuta Subsidence Belt. It is infenred that many faults and fracture systems are developed along the steep slope of the basement. The Oguni geothermal system is shown in Figure 2. A representative fault with a strike of NWSE, called the Takenoyu fault, runs through the central part of the field. In addition, the drilling results suggest that a hidden fault lies to the north of the Takenoyu fault and that these two faults form a local horst structure. Fracture zones including the Takenoyu fault as a nearly vertical high permeable zone are encountered by exploratory wells in the horst structure and its periphery. (see Abe et al, 1995). The Hohi Volcanic Rocks, and the Kusu Group and the Kuju Volcanic Rocks are most important for the Oguni reservoir system. The Hohi Volcanic Rocks forms a fractured permeable layer for reservoir above the steep slope of the basement. The Kusu Group and the Kuju Volcanic Rocks have impermeable layers such as the Nogami Mudstone and the altered rocks, respectively. They play the role of the cap rock above the reservoir. The Oguni reservoirs are separated into two parts, i.e., the northem and the southern reservoirs. This separation is confmed by the pressure difference between two reservoirs and the results of interference tests. The northem reservoir covers a large area including the Takenoyu fault, whereas the southern high pressure one is restricted to a small area. The geothermal fluid originated from meteoric water penetrates into deep and is heated by the conduction from the residual magma of Mt. Waita. The fluid (NaCl type water) flows up in a northwest direction along the Takenoyu fault zone and flows laterally northwards passing though the high permeable zones of Shishimuta formation and Hohi Volcanic Rocks. The extent of the reservoir is approximately 4 km in NS direction and 2 lan WE direction. The cap rock exists from 300 to 750 m ASL. Below the cap rock, the geothermal reservoir exists in a temperature range from 200 to 240 OC with a small region of steam zone at a top of the reservoir. Above the cap rock, the shallow unconfined aquifer exists with a groundwater table of the depth of around 100 m from the ground surface in Sugawara area, northern part of the Oguni field. a I Kuju Volunic rock^ Waiwul lak~ (WI) Kuju Volcanic Racks: Yamakawa luff w y * m. ', Faults infd from well geology \ Faults inlap;tlcd from Seismickfkcrion sumy (d'iabk) Hohi Volcpnic Rocks (Ho) 1 ; ud lw KUSU Formuion: ~ ~gyni mudsionc (NS) Kusu Fomtion: MJchiL lam (Ma) I;.Shishimutr Fonrwtion (Sh) a Taio Formation (Ta), e ' din0 (las ldiablc) NdyimpnwMemne. Wcll ~nce (dashed lines arc projected) (Mid lost circuluion) Mainpmnubleoonc Ruidflow Figure 2. Geothermal system of the Oguni field. (Revised after Abe, et al. (1 995)) 12

4 Microgravity Monitoring Surface graviiy meas~emen~ have been conducted since May of 1996, in every month. Totals of 41 benchmarks are spread over the whole field including production and reinjection bore field. The measurements are conducted in the way of reciprocating manner on several survey lines based on the standard benchmark of Geographical Survey Institute (GSI) located far (11 lcm) away from the field. Therefor, the survey is actually the measurements of relative changes to the gravity value at the standard benchmark. The Scintrex CG3M gravity meter is used for the rn~as~~ents The gravity mater is a microprocessorbased, automated meter, which con~uously samples data and auto~tically corrects for earth tides and tilt errors. In the actual survey, gravity data were sampled maximum of 120 times at each benchmark, and the sampling was stopped at a time that the error mean square of the maximum likelihood value was reduced within 5 microgals. The nominal accuracy of the CG3M gravity meter is 1 microgal. The gravity changes that observed fiom May 1996 to December 1997 at the station of 201, OG2, MG23 and 205 are shown in Figure 4, for representative examples. Locations of these benchmarks are shown in Figure 3. Unusual shifls in gravity between October and November of 1996 were observed at all of the benchmarks. It is considered that the reason of this change is due to the change of the condition of the standard benchmark of GSI, probably by either the gravity value or ground level changes. Or it could be due to the another source of error relating to the measurements that we did not notice yet, and we should keep on examining these data. Apart from the unusual increase in gravity mentioned above, seasonal gravity ~uc~ations are also observed. It appears that the changes of gravity due to the changes of shallow groundwater table depth are also included among these fluctuations. These seasonal changes could be corrected by the data of groundwater level of a shallow well. And it is considered that at least the effect of shallow groundwater table changes on gravity changes should be removed, in order to iden@ the production induced gravity changes of the reservoir after onset of production. Correction of the Seasonal Gravity Changes A shallow groundwater aquifer is idenwied by several shallow wells of the total depth of 20Om in Sugawara area, above the Nogami Mud Stone, which acts cap rock of the geothermal reservoir. The changes of groundwater table depth have been observed since 1990 as shown in Figure 5, by the shallow well BW2. Precipitation data are also shown in Figure 5. This area is the one of the areas of much rain, and we have rainfall of 1,130 3,450 d y r in recent 9 years (2,300 d y r on the average), that resulted in the substantial changes of groundwater table depth as shown Figure 5. To remove the effect of groundwater table changes on the surface microgravity changes, we studied at fnst the relations between the groundwater level changes observed in the shallow well and microgravity measurements data obtained at the benchmark (station 201) near the well. ~~ ~~ ~ ~ ~ Figure 3. Location of the benchmarks for gravity measurements. The change in gravity associated with a change in water table depth will be given by (AUis and Hunt, 1986): Ag = 2 n G 4 pw Ah where Ag is the gravity change, G is the universal gravitation constant ( 6.67~10 m3kg1sec2), 4 represents the porosity at the depth of the water table, pw is the grouxidwater density, and Ah is the change in water table. Several calculations have done by the equation using several porosity values based on the groundwater changes data. Solid line in Figure 6 shows the calculated gravity based on the measured changes of water level with a porosity of It seems that the calculated gravity is reasonably comparable with the measured one, although ambiguous shift and noises are still recopized. (We made a assumption that the calculated values after November 1996 were shifted to start fiorn the measured value of November 1996.) The porosity value of 0.13 means that a change of groundwater table of 1 m causes gravity change of 5 microgals in the Sugawara shallow groundwater system. Although it appears that the data of groundwater table changes is extremely important for calibration the gravity data, it is also common case that enough data of groundwater levels are not available. Therefor, based on the above mentioned results, to extend the method for the gravity r n e ~ ~ data ~ ewithout n ~ the groundwater level data, we tried to estimate changes of groundwater table from the precipitation data by using a simple empirical equation after Yuhara and Sen0 (1969). At first, we assume that a maximum increase of a water table associated with a precipitation rate ean be expressed by : AH=aR 13

5 ,nata ~j a r moa0 d k r 2 4 W,_ M. 5 I a E E ubjy wsm (00.0 : 9?93!J9.5 #7W#Ul. e793sao v I Figure 4. Gravity changes at the representative benchmarks. Error bars mean the differences between reciprocating measurements. where AH is a maximum increase of a water table (in m), R is a precipitation ( in mm), and a is a constant. And we also assume that decline of the water table after that time can be expressed by : H = HI + AH exp(b t) where HI is an initial water level before rainfall, t is time, and b is constant. Combining these equations, water level at certain day can be expressed by : H =HI + a C R, exp(b (t b)) where I&, is a precipitation at hth day from beginning (in =) Reproducing of the groundwater level same as the measured one in the well were tried by using the equation, and finally we got a set of parameters by least squares fitting as follows: HI = 671 (m ASL), a = , and b = 9.99 x (day') Calculated groundwater level using these parameters is shown in Figure 7 by dashed line, and it seem that reasonable matching with the measured water level was attained. The residual differences in m between measured and calculated water level changes are also shown in Figure 7. The calculated gravity changes using the inferred changes of water table depth are also shown in Figure 6 by dashed line. These results suggest that the surface gravity changes induced by the groundwater table changes in shallow aquifer can be corrected within the reasonable accuracy (A10 microgals or so) by estimating of the set of parameters (i.e. HI, a, b, and 4) for each measurement points using the microgravity data measured for a period of time prior to exploitation, even in a case that there are no groundwater level measurement data in the field. Estimation of Microgravity Changes Induced by Exploitation Changes in surface microgravity caused by field operation can be calculated by resent numerical simulation techniques (see e.g. Pritchett, 1995; Ishido et. al., 1995). Numerical simulation studies were performed to evaluate the reservoir potential of the Oguni field in 1992, and the numerical model representing the natural state was developed, which provides a high degree of conformity with reservoir parameters (pressures, temperatures and so on) measured in the field (see Pritchett and Garg, 1995). The possible gravity changes after exploitation were estimated by using the STAR simulator and the gravity postprocessor (Pritchett, 1995) based on the natural state model, which we refined based on the model developed by Pritchett and Garg (1995). It could provide us a usefiil information to conduct the proper microgravitymonitoring program in the field. 14

6 Precipitation and Groundwater Level = In ylcnud W.L rdo, IbdO UO I 5 The STAR reservoir simulator was used to carry out a numerical simulation which forecasted changes in underground conditions caused by currentlyplanned reservoir exploitation of 20 MW Oguni plant (and also future possibly planned Sugawara 10 MW binary plant), and then the gravity postprocessor was used to predict the resulting changes that would occur in the surface distributions of microgravity. The spatial grid employed for the reservoir forecasting simulation and its relationship to the study area of gravity changes employed for the postprocessor are shown in Figure 8. The STAR grid extends vertically upward from 1900 m ASL to the local ground surface. Figure 5. Groundwater level changes measured in the shallow well BW2. *? Shift by unknown reason I L so! 5 I L a ea er aa Y W Figure 6. Measured and calculated gravity changes at the station 201. Figure 7. Measured and calculated water level changes at well BW $TAR grld orlontatlon In horizontal plme 2aOOO~.. '. I ""," ",' '.'.,....j L g z a z hters East Figure 8. STAR computional grid and study area of gravity changes. Figure 9 shows the calculated distribution of microgravity changes at the swface (relative to preproduction condition) for 30 years of plant operation. The calculated result of gravity changes at 30 years exhibits slight gravity increases (about +35 microgals) in the main reinjection wellfield at Sugawara area, and relatively large gravity decreases (about 130 microgals) centered on the middle of the production wellfield (neighborhood of production wells IH2, GH11, GH20, GH10, GH12, and GH 19), caused by boiling arising from productioninduced local pressure decline. Because of a relatively small scale in capacity of the power plant and relatively high permeability of the reservoir system, gravity disturbances induced by exploitation are not expected to be substantially intense in the Oguni field. It seems, however, that the gravity changes should be detectable by a properconducted gravity survey. The study 15

7 Nakanishi, et at. indicates that the correction of the gravity data affected by various sources of noise, in particular by ground surface subsidence and groundwater table fluctuations, is extremely important in the Oguni field, to detect the reservoir behavior in microgravity after exploitation. STAR computed gravity change betwen Contow tnterval IS S mlcrogalr 0 md d w It is concluded that the correction of the seasonal gravity changes by such a method is extremely important in the Oguni field, based on the study of prediction of the gravity changes after fullscale field exploitation. More microgravity measurements data will, fortunately, be available before exploitation, and it will provide us extremely useful information to appraise the background noises in the measured surface microgravity and to establish the correction of these noises, that promises to make success of the hture reservoir management. Acknowledgements The authors wish to express their deep gratitude to the management of EPDC for supporting this research project and the permission to publish this paper. The authors also would like to thank Mr. Kazuo Moribe and Mr. Hirotsugu Ooshima of KCC for their assistance in the arrangements of the gravity measurement data. The authors are grateful to New Energy and Industrial Technology Development Organization for providing the data of shallow aquifer well BW2. References btur East Figure 9. Computed changes in surface microgravity after 30 years of field operation. Contour interval is 5 micro gal. Shaded area denotes gravity increase. Summary Microgravity measurements at 41 benchmarks have been conducted prior to startup of the Oguni 20 MW power plant. The seasonal gravity fluctuations probably due to the changes of shallow groundwater table seemed to be included in the microgravity changes observed so far. The comparison between the measured microgravity changes and the calculated ones using the observed shallow groundwater table changes were studied. The preliminary correction of the seasonable gravity changes by using a simple empirical equation and precipitation data was tried. And it appeared that the method would be a promising way to correct the gravity changes affected by the changes of groundwater table for all of the benchmarks with no groundwater level data. Abe, M., Yamada, M., Kawano, Y., Todaka, N., and Tezuka, S. (1995). Development of the Oguni Geothermal Field, Japan, Proc. World Geothermal Congress, Florence, pp Allis, R.G. and Hunt, T.M. (1 986). Analysis of exploitationinduced gravity changes at Wairakei Geothermal Field, Geophysics, Vol. 51, pp Ishido,T., Sugihara, M., Pritchett,J.W. and Ariki,K. (1 995). Feasibility Study of Reservoir Monitoring Using Repeat Precision Gravity Measurements at the Sumikawa Geothermal Field. Proc. World Geothermal Congress, Florence, pp Pritchett, J.W. and Garg, S.K. (1995). A Modeling Study of the Oguni Geothermal Field, Kyushu, Japan, Proc. World Geothermal Congress, Florence, pp Pritchett, J.W. (1995). STAR a Geothermal Reservoir Simulation System, Proc. World Geothermal Congress, Florence, pp Pritchett, J.W. (1995). STAR Users Manual, SCubed Report No.SSS TR , Revision E. Yuhara, K. and Sen0 K. (1969). Onsengaku (Geophysics on Hot Springs), ChijinShokan, pp (in Japanese). 16