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 Geothermal Resources Council TRANSACTIONS, VOL 9 PART I, August 1985 GEOTHERMOMETRY IN THE RECENT EXPLORATION OF MOKAI ROTOKAWA GEOTHERMAL FIELDS, NEW ZEALAND AND R.W. Henley and K.I. Middendorf Chemistry Division, D.S.I.R., Wairakei, Private Bag, Taupo. ABS TRACT Chemical data for natural features and exploration wells at Mokai and Rotokawa, New Zealand are reviewed in the context oe (a) the use of chemical geothermometers during early exploration (prior to drilling) and (b) the interpretation of exploration well discharges. These data confirm the validity of the NaCaK geothermometer which is particularly precise above about 29OoC, while available alkali ratio geothermometers (Na/K, Na/Li) substantially overestimate source temperatures. Lithium concentrations and chlorideboron ratios appear to be useful in guiding field exploration during drilling. Gas data in general conform to simple gas phase equilibria but used in conjunqtion with quartz or alkali geothermometry are most valuable in monitoring the behaviour of discharging wells and interpreting feed zone response to production. INTRODUCTION This paper presents geochemical data from recent exploration wells at Mokai and Rotokawa, New Zealand. These fields have proven to be the hottest yet drilled in New Zealand and to have the highest undeveloped resource potential of the New Zealand fields. In this paper the contrasting roles of chemical geothermometry are reviewed with respect to the early exploration phase and to the interpretation of exploration well discharges. Mokai EXPLORATION The Mokai field, 16 km northwest of Wairakei is one of the New Zealand fields most easily related to a recognisable caldera margin, such relations in other fields being obscured by younger ash flows and sediments. The Mokai system outlined by the resistivity boundary zone in Figure 1 occupies intracaldera volcaniclastics which have been drilled to a depth of over 1600 m. Figure 1 shows the distribution of natural thermal features and exploration wells in the field. Warm chloride springs (6OOC) occur in a faultaligned incised gorge, 5 km north of the field, and vaporrelated features (fumaroles, mud pools and sulphatebicarbonate rich springs) occur in the south central portion of the field. Some representative data from these features are given in Tables 1 and 2. Solute geothermometry applied to the northern springs gives evidence of source temperatures up to 166OC using the magnesium corrected NaKCa geothermometer (Fournier and Truesdell 1983, Fournier and Potter, 1979) and up to 18OOC based on quartz geothermometry (e.g. Fournier, The potassiummagnesium geothermometer proposed by Giggenbach et a1 (1983) suggests source temperatures of about 1 OO C, while the Na/Li geothermometer proposed by Fouillac and Michard (1981) suggests around 280OC. As evident from Figure 1 the chloride springs lie outside the apparent resistivity boundary of the field, and despite the conflicting geothermometry, show that a moderate temperature chloride water reservoir lies to the south. The presence of vaporderived features within the resistivity boundary is sufficient in itself to suggest high temperatures at depth. use of gas geothermometers depends either on an assumption of the vapor fraction resulting from subsurface boiling to produce the fumarole, the carbon dioxide content of the deep fluid (D'Amore and Panichi, 1980) or effective mineral equilibria in the reservoir (Giggenbach, 1980). For the Mokai fumaroles these data provide temperatures in excess of 2200C.. Exploration drilling commenced with a shallow (606 m) well, MK~, which confirmed the presence of the large intermediate temperature reservoir indicated by hot spring chemistry. At 220 m a maximum temperature of 194OC occurred above a major temperature inversion and drilling fluid loss zone (300 to 350 m). Analytical and geothermometer data for the discharge is given in Tables 1 through 3. The NaKCa geothermometer correlates well with the measured well temperatures as does the temperature obtained from the silica concentration assuming equilibrium with chalcedony (Chalcedony was identified in drillcore). Gas analytical quotients such as log QCH~ = (Xco2 x4h..+)/xch4,for the discharge are however signlfican ly higher than expected at the discharge temperature (17OOC). 31 7

3 200( ENTHALPY kj/kg 1 ooc \ Figure 2: Mixing relations for exploration well fluids and natural features in the Mokai geothermal field. Downhole samples for MK4 are included. Solid symbols = chloride at quartz temperature, open symbol = estimated convective upflow for field. Figure 1 : Mokai Geothermal Field. Fumaroles, steaming ground and seepages of s teamheated water are shown by the symbol and ++++ chloride water springs by the black ornament. The resistivity boundary zone is shown by the crosshatching. In MK2 (1654 m) and MK3 (1678 m), drilled subsequently within the resistivity anomaly, much higher temperatures occur. In MK2 temperatures measured at the principal drilling fluid loss zone ranged from 250 to 278OC prior to and during discharge, but NaKCa geothermometry indicated source temperatures from 277 to 287OC. The NaK calibration of Fournier ( 1979) gives higher temperatures from 288 to 303Oc. Homogenisation temperatures for inclusion fluids in quartz and adularia ejected during the initial vertical discharge range from 295 to 308OC and 291OC respectively (Hedenquist, pers. comm. 1. These temperatures were confirmed by postdischarge downhole measurements. Similarly in MK3 (1678 m), NaKCa (and NaK) geothermometry was confirmed by downhole measurements. at MK5 (2592 rn) a temperature of 298OC was measured downhole; the tnakca geothermometer gave 31OOC but again quartz temperatures were significantly lower (27OOC). 31 8

4 These wells are of high enthalpy type as defined (Henley et al., 1984, Chapter 11) by the relation Ht < HM > HNa~ca HQ,,,~~ where H refers to the enthalpy either measured, M1 or at the respective geothermometer or measured downhole (t) temperatures. The relatively high enthalpy relates to underground boiling in response to well discharge and conductive extraction of heat during flow to the well. As a result quartz geothermometry provides indications of temperatures in the vicinity of the discharging well rather than that of the undisturbed reservoir (see Table 3). Gas analytical quotients relative to the calculated quotients at reservoir temperatures (Fig. 6) indicate some apparent vapour loss from the reservoir fluid, suggesting that the excess discharge enthalpy is largely due to heat gained conductively from host rocks during discharge. The effect is most marked in the discharge of MK5 although this may relate to the presence of multiple feed zones in the well. Figure 2 provides mixing relations for the Mokai well fluids and natural discharges. A simple mixing pattern is observed confirming both the relation of the northern springs to the high temperature field and a strong component of lateral flow toward the north (Figure 3). The principal upflow to the field appears to be at about 325OC with a chloride content of 2100 mg/kg and low gas content similar to that at Wairakei. ssw meteoric voter \ I, noko8 Ignimbrite rontommg sleom heoled Rhyollte Erosion Deposits * Major well inflow Figure 3: Longitudinal crosssection of the Mokai field showing principal geological features, the strong lateral flow and dispersion of fluid to the north and subsurface temperatures measured or estimated through chemical geothermometry. NNE a ROTOKAWA The Rotokawa field is situated 9 km east of Wairakei and has similar geology except for the presence of extensive hydrothermal eruption breccias apparently centred on Lake Rotokawa (Figure 4). The field margin has recently been redefined by electrical resistivity measurements, extending the previously inferred field boundary well to the north of the Waikato River. Another resistivity anomaly occurs at Ngatarnariki just a few kilometres further north. Although some hot seepages occur along the Waikato River the most intense thermal activity occurs to the north of Lake Rotokawa (ph = 2.3). All are acid sulfatechloride type waters associated with fumarolic activity, collapse pits, advanced argillic rock alteration and intense steaming ground. Native sulfur precipitation accompanies spring discharge presumably through atmospheric oxidation of H2S. Native sulfur is also abundant within the lake sediments near and beneath Lake Rotokawa and these have attracted much attention as a possible sulfur resource. Due to their acidity and shallow interaction with volcanics, none of the surface discharges provide useful information on subsurface temperatures. High magnesium contents lead to large (loo C) corrections to the NaKCa geo thermome ter and the K/Mg geo thermome ter of Giggenbach et a1 (1983) gives temperatures of OC. The intensity of activity alone is however indicative of high subsurface temperatures. Table 2 includes a fumarole gas analysis from Rotokawa. As at Mokai, extraction of subsurface temperatures from such data is dependent on arbitrary assumption but high temperatures are indicated by the D' AmorePanichi and gas ratio geothermometers (e.g. XH~/XH~S, Giggenbach, 1980). Three exploration wells were drilled at Rotokawa between 1966 and Selected data from these exploration wells are given in Tables I and 2. Although relatively shallow, RK1 (1198 m) at that time located the highest bottom hole temperature (316OC) recorded for a New Zealand system. The solute geothermometers indicate no excess steam in the initial discharge of RK1 and feedzone temperatures of 257 to 264OC corresponding to feed depths of m. RK2 by contrast had a high enthalpy discharge with tsi02 tnakca at 22OOC suggesting heat gains from the feed zone wallrocks. In RK3 solute geothermometers gave temperatures of 23OOC (quartz) to 24OOC (NaKCa) corresponding to feed zone depths of 400 to 900 m. The excess discharge enthalpies appear through the relation tquartz < tnakca, to be due to phase separation and heat gain from host rocks. 31 9

5 HENLEY ANI) MIDDENDORF The recent drilling of two deeper exploration wells has provided more valuable information on geological structure and has located the highest subsurface temperatures yet encountered in New Zealand (RK4, 320OC; ~ 5 340OC)., R K ~ and 5 penetrated a thick andesite sequence previous1 y unknown at Rotokawa and RK4 located greywacke at about 1850 m. A more comprehensive chemical data set is available from RK4 compared to earlier wells (RK5 had not been discharged at the time of writing). Figure 4: Rotokawa Geothermal Field (symbols as for Figure 1). In the initial discharge of RK4, tquart = tnakca = 322OC and the measured discharge enthalpy corresponded well with that of water at this temperature. Subsequently, during its three month discharge, silica temperatures fluctuated and fell to 29OOC while tnakca remained relatively constant at C. Gas data provide a similar picture but the movement of the methane and ammonia gas quotients into the field of apparent vapor loss (see Giggenbach, 1980) may indicate that more than one feed zone affects the discharge. The gas content of the discharge fell from an initial 2% to 1.2%, the higher figure representing that of the deep aquifer and this is similar to that of the Broadlands (Ohaaki) system 18 km to the northeast. 1 0 Y > 3 Q 1000 i & z " I RK4 /,,ROTOKAWA 7 MOKAI, \ *@ ACID SULFATE CHLORIE,E SPRINGS a \ io0 1 so CHLOHi DE rng./kg Figure 5: Mixing relations for exploration well discharges in the Rotokawa yeothermal field and natural surface discharges. The estimated convective upflow fluids in the Wairakei and Mokai fields are shown for reference. X Y 20 0, TEMPERATURE OC Figure 6: Gas analytical quotient (log QCH4) versus temperature as a function of underground gain or loss of steam at about 28OOC (isosteam fraction contours), after Giggenbach (1980). 320

6 A number of contrasts occur between the chemistry of the early shallower exploration wells and the deep well, m4. Although similar to RK1, the chloride content of the deep system reached by RK4 is less than half that of RK2 and RK3, as shown in Figure 5. It appears that the discharge of RK2 and RK3 represent a boiled derivative of a parent fluid like that in RK4. RK1 is dilute compared to RK2 and 3 and appears related to R K ~ by both lilution and boiling. The chlorideboron ratio (4.75.4) of the deep system is half that of the shallow wells and surface discharges, the high boron correlating with the proximity to greywacke (c.f. Mahon, 1970). Although often an indicator of homogeneity, chlorideboron ratio, in many fields (e.g. Rroadlands (Ohaaki), Waiotapu, Wairakei, New Zealand and "ongonan, Philippines ) Cl/B shows iistinct zonation. Absorption of boron by illite is the likely cause and it is possible that the zonation may find some use in targeting exploration #ells. The asymetric distribution of exploration wells dth respect to the field boundary precludes at this stage any realistic modelling of the flow structure of the field. COMMENTARY ON GEOTHERMOMETEKS Chemical geothermome ters have two contrasting roles in exploration. Prior to drilring they provide the only clue concerning subsurface temperatures in the vicinity of natural discharges. It should be remembered however, as exemplified by Mokai, that this technique provides information about the immediate subsurface aquifers in the vicinity and not convective upflow temperatures in the field. All the solute geothermometers are based on assumptions regarding mineral equilibria and rates of reequilibration, and these need continued reassessment as exploration proceeds. At Mokai, for example, MK1 provided data supporting the use of the chalcedony geothermometer at temperatures around 17OOC whereas at more than 2OO0C, silica concentrations in these and other geothermal fields uniformly indicate control by quartz solubility. The chalcedony and NaKCa temperatures for the Mokai springs were coincident and indicated the proximity of a large intermediate temperature reservoir, subsequently proved by lrilling. Fortunately in this case, gas data from fumaroles, and the resistivity mapping, indicated a higher temperature system further south. It is possible that in many regions useful geothermal resources may have been overlooked through an uncritical reliance on solute geothermometers. More important in assessment is the growing role of the difference between the solute geothermometers in indicating aquifer behaviour in the vicinity the discharging well. The discharges of MK2, 3 and 5 and RK4 suggest that additional heat is gained from the formation during two phase flow to the well and that permeabilities are moderate relative to those of normal enthalpy wells (Hmeas L H N ~ K C = ~ Hquartz). Similarly, as shown by Giggenbach (1980) gas quotients based on CH4 and NH3 equilibria in relation to solute geothermometers provide good evidence on underground flow processes, particularly the gain or loss of steam, and the effects of multiple feeds (Fig. 6). The H2/H2S relation of Giggenbach (1980) appears useful as a geothermometer in the initial discharge of the exploration wells. The newly available chemical data for the Mokai and Rotokawa wells provide the opportunity to test the empirical calibration of the NaKCa geothermometer (Fournier and Truesdell, 1973) at high temperatures where previously few reliable field data were available. As shown in Figure 6a, the f15oc precision claimed by the original authors is validated by these data as well as data from other recent exploration wells in New Zealand and the Philippines. Excellent correlation is achieved above 3OOOC although in this temperature range the calibration equation shown in Figure 7a may be more precise. Figure 7b however shows that the NalK geothermometers of Fournier ( 1979) and ArnOrSSOn ( 1983) may significantly overestimate temperatures where log Na/K (mg/kg) (0.65. Similarly the NaLi geothermometer of Fouillac and Michard ( 1981 provides excessive temperature estimates although, as pointed out by those authors and Mahon (19761, the presence of high lithium contents itself indicates extensive high temperature fluidrock reaction. Gas geothermometry based on homogeneous gas reactions provides excellent temperatures estimates for the first discharge of most wells. Gas data however appear to have their most important role in interpretation of well or reservoir response to production as suggested by D'Amore and Celati (1983). Used in association with the solute geothermometers, gas data analysis will eventually become a standard technique in well or reservoir performance analysis. Chemical geothermometry plays a contrasting role in the assessment of exploration well jischarges. In many cases (e.g. MK~) drilling Eluid retention obscures aquifer temperatures until dell after sustained discharge. In such cases the WaKCa geothermometer reliably provides indications 3f feed zone temperatures. Fluid inclusion ~eotherrnometry is also a very useful indicator if sui table material is available. 321

7 Nthmkzd sp* W1 15/ 6/ M(2 13/ 5/ xi Hc3 30/11/l Mo 3/12/ m 19/1y <5 mtckawa, Lake 16/ 6/ , WiWJ RKl 19/ 4/ * RK2 20/10/ @ RK3 10/ 3/70 1 1XO Mo Fac4 5/12/ FK4 10/1y *highso =lues my be due to 3s anddatim in sanple. TA&E 1: Selected analytical data for liquid le fran ell discharges ard hot spm at mkai ard RO~~M.. Mokai Date xg C02 H2S "3 H2 N2 CH4 C02/H2S H2/H2S log log QCH~ 109 Q" mmoles/mole mmo le s/mo ledr gas TD Fumarole (6.8) MK MK2 18/ 5/82 a a MK3 30/11/ MK3 3/12/ *8 MK5 19/12/ Ro tokawa Fumarole 26/ 2/79 (22.7) RK 1 16/ 5/ RK2 20/ 10/ RK3 10/ 3/ RK4 5/12/ RK4 10/12/ TABLE 2: Gas analyses for selected well discharges and fumaroles at Mokai and Rotokawa. The gas fractic xg is given as millimoles (total gases) per mole (total discharge). Discharge enthalpies are given for equivalent liquid samples in Table 1. QCH~ and Q" are analytical quotients based c the homogeneous methane and ammonia forming reactions (Giggengach, 1980). 322

8 X 0) 1.0 E" New Zealand wells b 5 h m f 0.5 m x Y c. 2 m Y non New Zealand wells & t C IT x 1000 Figure 7a: Comparison of NaKCa relations in recent high temperature geothermal discharges (mostly in New Zealand) with the NaKCa geothermometer Fournier and Truesdell (1973) of line A. (The broken lines A +O15 and A O15 indicate the precision limits indicated by those authors). Line B provides an alternative calibration valid for most New Zealand wells above about 29OOC with slope 1944 and intercept 3.63 (N = Ngawha, TH = Tauhara, BR = Broadlands, MK = Mokai, RK = Rotokawa, T = Tongonan (Philippines), CPM = Cerro. Prieto (Mexico)). The NaKCa function has been calculated from total discharge composition and plotted against silica temperature (Fournier and Potter, 1982) or measured downhole temperature (MK5) O C /T x 10GO Figure 7b: Comparison of NaK ratios for recent exploration well discharges in New Zealand. Line A is the calibration of Fournier (1979) and line B that of White and Ellis ( Truesdell, 1976). tspr ing c1 c1 OC Total at silica Discharge temp. w/kg tnkc tsio2 OC ~ N K. tlina (Fournier) C1/B N. Mokai springs MK1 MK2 MK3 MK3 MK5 Ro tokawa, Lake Rotokawa, Spring RK 1 RK2 RK3 RK4 RK (160)= ( O TABLE 3: Geothermometer and other derived data for selected well discharges and hot springs at Mokai and Rotokawa. [Silica temperatures: 179 = quartz temperature (160) = chalcedony, 55 = amorphous siica. NaKCa temperatures corrected for magnesium and see text]. 323

9 HENLEY AND MrDDENDOR? CONCLUSIONS Used with discretion chemical geothermometry is a most effective and inexpensive exploration technique particularly when used in conjunction with electrical resistivity mapping to obtain a preliminary outline of field structure, The data from the Mokai and Rotokawa fields validates the various silica (depending on temperature range) geothermometers and the empirical NaKCa relation, Alkali ratio geothermometry (Na/K, Na/LI) appears unreliable at temperatures above 3OOOC although high Li contents are useful in indicating the proximity of high temperature reservoirs. Gas geothermometry at the exploration stage provides useful indications of high temperature despite the range of assumptions involved. Trends in gas analytical quotients are highly sensitive to well behaviour and, used in conjunction with solute geothermometers (quartz, NaKCa), provide a valuable, developing tool for the monitoring of well performance. ACKNOWLEDGEMENTS The authors which to thank R.B. Glover for comments on the manuscript and L.E. Klyen and Paul van Boheemen for maintaining the chemical sampling programme in these fields. REFERENCES Arnorsson, S., 1983, Chemical equilibria in Icelandic geothermal sys terns implications for chemical geothermometry investigations, Geothermics, V. 12, p D'Amore, F. and Panichi, C., 1980, Evaluation of deep temperatures of hydrothermal systems by a new gas geothermometer. Geochim. et cosmochim. Acta, v. 44, p Fournier, R.O. and Potter, R.W., 1982, A revised and expanded silica (quartz) geothermometer, Geothermal Resources Council Bulletin, November 1982, p Fournier, R.O. and Truesdell, A.H., 1973, An empirical NaKCa geothermometer for natural waters, Geochim et Cosmochim Acta 37, p Giggenbach, W.F., 1980, Geothermal gas equi1ibria:geochi.m. Cosmochim. Acta, v. 44, p Giggenbach, W,F., GOnfiantini., R., Jangi, B.L, and Truesdell, A.H., 1983, Isotopic and chemical composition of Parbati Valley geothermal discharges, NorthWest Himalaya, India. Geothermics, v. 12, p Henley, R.W., Truesdell, A.H. and Bilrton, P.B. Jr, 1984, FluidMineral Equilibria in Hydrothermal Systems: Reviews in Economic Geology, v. 1, p Mahon, W.A.J., 1976, Review of hydrogeochemistry of geothermal sys tems prospecting, development and use. Proc. 2nd U.N. Symposium on the Development and Use of Geothermal Resources, v. 1, p Mahon, W.A.J., 1970, chemistry in the exploration and exploitation of hydrothermal systems. U.N. Symposium on the Development and Utilization of Geothermal Resources, Proc. (Geothermics, Spec. Issue 21, v. 2, p Truesdell, A.H., 1976, Geochemical Techniques in Exploration. 2nd U.N. symposium on the Development and Use of Geothermal Resources, V. 1, liiilxxxi. D'Amore, F. and Celati, R,, 1983, Methodology for calculating steam quality in geothermal reservoirs. Geothermics, v. 12, p, Fouillac, C. and Michard, G., 1981, Sodium/lithium ratio in water applied to geothermometry of geothermal waters. Geothermics, V. 10, p Fournier, R.O., 1979, A revised equation for the Na/K Geothermometer. Geothermal Resources Council Trans., v. 3, p, Fournier, R.o., 1981, Application of water geochemistry to geothermal exploration and reservoir engineering, in Rybach, L. and Muffler, L. P. J., Geothermal Sys tems : Principles and Case Histories, John Wiley and Sons, p