GRC Transactions, Vol. 43, 2019

Geochemical Characteristics of the Geothermal Reservoir, : An Overview Jeremiah Kipngok1, Leakey Auko, Sylvia Malimo, George Igunza, Shilla Kangogo Lawrence Ranka, Evans Bett, Janet Suwai and Koji Matsuda2

1Geothermal Development Company [email protected] 2West Japan Engineering Consultants, Inc. [email protected]

Key words: Menengai, chemical characteristics, geothermal reservoir, isotope data, geothermal wells

ABSTRACT

This paper presents an overview of the chemical characteristics of Menengai geothermal reservoir from wells discharge fluid. The fluid discharge from Menengai generally exhibits widely varying chemical properties from well to well. This observed variability points to heterogeneity of the reservoirs feeding Menengai wells. The fluid discharge belongs to the NaHCO3 facies although some of the wells located in the northern part of the central sector, in the vicinity of well MW-19, display a Na-HCO3-Cl type. Most wells in Menengai have excess enthalpy which is largely attributed to phase segregation. Three models were used to assess gas-mineral equilibria assuming a closed, isolated and an open system and unit activity. The models show that H2S in the reservoir(s) in Menengai are largely controlled by hydrothermal minerals but H2 is higher suggesting the presence of significant vapor fraction in the reservoir(s) feeding Menengai wells, albeit in varying proportions. CO2 may have contributions from the magmatic heat source as well as organic sources besides being fixed by hydrothermal minerals. The water recharging the Menengai reservoir is meteoric in origin while a mantle source is inferred for the gases. The main upflow is located within or in the locality of well MW-13 situated in the central part of the Menengai geothermal field. The main process affecting the chemistry of Menengai wells discharge fluid is reservoir boiling and phase segregation. Conductive heat transfer/cooling and to some extent addition of magmatic fluids are other factors that significantly contribute.

1. Introduction The goal of this paper is to provide an overview of the chemical characteristics of discharge fluid from tested geothermal wells in Menengai and thus the geothermal reservoir beneath the Menengai caldera. The Menengai geothermal field is hosted in the Menengai volcano located in the central GRC Transactions, Vol. 43, 2019 part of the Kenya Rift Valley, an area with promising geothermal potential. Surface geoscientific studies culminated in the drilling of geothermal exploration wells in 2011. The wells were successful in proving existence of commercial steam. Production drilling is currently ongoing with plans to generate 105 MW of electricity from the already realized steam (of over 160 MW equivalent) are still underway. Figure 1 shows the distribution of wells in the Menengai caldera. The geothermal field is being developed by geothermal development company (GDC), an institution fully owned by the government of Kenya. Several studies have been undertaken so far geared towards understanding the nature of Menengai geothermal reservoir from chemistry of fluids discharged during well flow tests (Kipng’ok, 2011; Sekento, 2012; Malimo, 2013; Kipngok, 2014, Auko, 2014). The present contribution builds on these previous works.

Figure 1: Map showing the Menengai caldera and location of geothermal wells

2. Geological Background Menengai is an elliptical piecemeal caldera of the Krakatau-style formed through different episodes of collapse that are associated with two major eruptions thus leading to the partly superposed lava flows of different ages. The lava flows cover virtually the entire caldera floor. Menengai surface geology is dominated by trachyte lavas (which exhibit variation in texture and flow), pyroclastics and ignimbrites as extensively described by previous researchers (McCall, 1967; Griffith, 1977; Jones & Lippard, 1979, Jones, 1985, Griffith, 1980; Griffith and Gibson, 1980; Leat, 1983, 1984, 1985; Leat et al., 1984; Geotermica Italiana Srl, 1987; MacDonald et al., GRC Transactions, Vol. 43, 2019

1994; Mungania et al., 2004; Lagat et al., 2010). The major structures associated with Menengai geothermal activity are the caldera ring faults, Molo TVA, Solai TVA and the Makalia fault to the south of the caldera. The general trend of many structures within the Menengai field is N-S, NNE- SSW and NNW-SSE in the western zone. Numerous studies on drill cuttings from the wells have been undertaken and published (Omondi, 2011; Mibei, 2012; Kipchumba, 2013; Lopetok, 2013; Kahiga, 2014 and Mbia, 2014). The findings indicate that the wells lithology show close similarity, with trachyte being the dominant rock. Chemical analyses reveal that the subsurface lithostratigraphy includes trachytes, tuff, pyroclastics, basalt and syenitic intrusives although trachyte constitutes the most dominant rock. Fractional crystallization, possibly in combination with magma mixing and crustal contamination is the most important processes that control the geochemical evolution of the Menengai rocks (Mbia, 2014). Fractional crystallization from rock chemistry indicate existence of high level magma chamber, hence a heat source for Menengai geothermal system. Five hydrothermal zones are recognized in Menengai geothermal system; the unaltered zone, smectite-zeolite zone, transition zone, quartz-illite-epidote zone and wollastonite-actinolite zone. The major reservoir regions occur at the quartz-illite-epidote zone marked by the appearance of quartz, illite, smectite, albite, chlorite, calcite, epidote, actinolite and wollastonite. 3. Fluid Sampling and Analysis Fluid sampling and analysis was done as described by Kipngok (2011) and Auko (2014). The liquid samples were collected into polyethylene bottles while the gas samples were collected into 325-340 ml evacuated gas sampling flasks containing 50 ml of 40% w/v NaOH solution. Additional liquid samples were obtained from the weirbox but the chemistry of these samples was not utilized in this report except where phase separation could not be achieved using the Webre separator either due to a high steam fraction or very low wellhead pressures. Chemical analysis of all fluid samples was done at GDC geochemical laboratory while analysis of isotope samples was mainly conducted in Japan with a few water samples done in Iceland.

3.1. Geochemical Data Handling The quality of the chemical data was checked prior to interpretation based on the charge balance error (CBE). This was done by means of the equation below based on electro-neutrality conditions:

∑ 푧 .푚 ∑ 푧 .푚 퐶퐵퐸(%) = 푐푎푡 푐푎푡 − 푎푛 푎푛 .100% (∑ 푧푐푎푡.푚푐푎푡 + ∑ 푧푎푛.푚푎푛)/2

Where zi is the charge of an ion, i, and mi is the molal concentration of i (mol/kg). A CBE of the order of magnitude of 10% is invariably regarded satisfactory (Arnórsson, 2000). Therefore, for the potentially complete data set of the aqueous components of the wells, a CBE within the permissible threshold was selected. For the water samples with a pH above 8.5, speciation of total - - inorganic carbon as HCO3 and silica as H3SiO4 was taken into account in the CBE calculations. Atmospheric contamination of gaseous components was used as a criterion for selecting suitable data for gas interpretation. In most cases, the chemical results were obtained at widely varying separation pressures and thus necessitating recalculation to a common condition for comparison. GRC Transactions, Vol. 43, 2019

4. Chemical Composition of Reservoir Liquids Calculation of chemical composition of reservoir liquids is usually affected by boiling and potential causes for observed discharge enthalpy. Figure 2 (left) shows the relationship between total discharge enthalpy (as measured) and aquifer temperature. The aquifer temperature used in the Figure was obtained from analysis of temperature logs (under flowing and static conditions) and fluid geothermometry. It is noted in Figure 2 (left) that most wells in Menengai are characterized by excess enthalpy, in some cases very pronounced, owing to reservoir boiling and preferential steam inflow into the wells yielding varying in-hole steam fractions ranging from 0.1 to over 0.9. The corresponding percentages of deep steam computed through analysis of the silica enthalpy plot of Figure 6 are within the same range. In some wells dilution and conductive cooling is inferred. In this regard, several assumptions must be made in order to translate chemical compositions of the well discharge fluids of excess enthalpy wells into geothermal reservoir fluid compositions. Reservoir pH values computed with the aid of WATCH computer code (Arnórsson et al. 1982, version 2.4 of Bjarnason 2010) range from 7.5 to 8.8. Notably, the WATCH program may slightly overestimate reservoir values given that pH values calculated by WATCH are controlled by partitioning of gases (chiefly CO2) between coexisting vapor and liquid phases.

The relative concentrations of major anions (HCO3+CO3, SO4, and Cl) adapted from Giggenbach (1988) show that Menengai reservoir liquids belong to the Na-HCO3 facies (Figure 2, right). A few of the wells though (MW-19, -19A, 19B and 20A may be classified as Na-HCO3-Cl waters. Notably, sulphate content in the reservoir liquids is relatively high (50 to about 300 mg/kg) suggesting varying degrees of mixing of high and low enthalpy fluids. The predominantly Na- HCO3 nature of Menengai reservoir contrasts with the Na-Cl composition of most geothermal “mature” liquids in many geothermal fields in the world. This could be attributed to the tectonic setting of Menengai, being situated along a continental-rift zone. Available data for volcanic gases, though comparatively limited in number, indicate that subduction-zones volcanic gases are enriched in Cl relative to hotspot and divergent-plate volcanic gases (Symonds et al., 1994; Sawyer et al., 2008).

Figure 2: Correlation plot of measured enthalpy with aquifer fluid temperature (left) and Cl-HCO3-SO4 ternary (right) GRC Transactions, Vol. 43, 2019

Sodium is the dominant cation in the reservoir liquids and inversely correlates with discharge enthalpy (Figure 3, left). In principle, these differences may be ascribed to the different temperatures present in the zones of provenance of these liquids. It is evident that the concentration of an aqueous solute in the total discharge will decrease as discharge enthalpy increases if excess enthalpy is caused by phase segregation, which seems to be mostly the case in Menengai. The concentrations of K also appear to generally have an inverse correlation with temperature except for wells MW-20 and MW-18A (Figure 3, right). Accordingly, the Na/K ratios appear to mostly decrease with increasing enthalpy (Figure 5, left), consistent with what is expected. Low calcium is observed from samples analysed (Figure 4) with no clear relationship with enthalpy. The low concentrations could be the consequence of uptake of Ca from the fluid as a result of formation of CaCO3 upon flashing of the geothermal water. On the other hand, fluoride concentrations are significantly high in Menengai well discharges relative to fluoride content of geothermal fluids which is usually typically less than 10 mg/kg according to Nicholson (2003). The high fluoride content in the Menengai reservoir (Figure 5, left) could largely be due to two possibilities; precipitation of carbonate minerals leading to excess fluoride or the addition of fluoride to the fluid by condensation of volcanic gases (HF). Additionally, studies show that fluoride is generally anomalously high in the East African Rift Valley mainly attributed to the development of hyper-alkaline volcanic rocks in the rift zone (Edmunds and Smedley, 2013). Figure 5 (right) shows that sulphate content is equally much higher than typical concentrations in in deep geothermal fluids in most geothermal fields of less than 50 mg/kg according to Nicholson (1993). The high sulphate concentrations are presumed to be largely due to oxidation of hydrogen sulphide. This suggests mixing of high enthalpy geothermal fluids from deeper parts of the reservoir(s) with lower enthalpy fluids from the shallower permeable horizons.

Figure 3: Correlation plots of enthalpy with Na (left) and enthalpy with K (right) GRC Transactions, Vol. 43, 2019

Figure 4: Correlation plots of enthalpy with Na/K ratio (left) and enthalpy with Ca (right)

Figure 5: Correlation plot of enthalpy with F (left) and with sulphate (right) GRC Transactions, Vol. 43, 2019

5. Gas chemistry and Processes Controlling Gas Composition Before considering gas equilibria, it is instructive to contrast the concentration of gas species with discharge enthalpy (Figures 6 and 7) to investigate the gas chemistry of reservoir vapors and reservoir liquids. The enthalpy has been preferred as a reference variable to the fraction of reservoir steam because the enthalpy data is available for all the gas samples whereas the fraction of reservoir steam is not. CO2 is the dominant gas in the Menengai geothermal reservoir accounting for over 80 percent of the gases present. It is observed that there is an inverse correlation of CO2 content in the steam with discharge enthalpy as measured. This could be attributed to boiling in the reservoir and subsequent gas loss. H2S concentrations are largely controlled by reservoir temperature although other processes like mixing of higher enthalpy with lower enthalpy fluids may be responsible to a reasonable extent. The concentration of H2 is also dependent on temperature and suggests significant but varying vapor/steam fractions in the reservoir(s) tapped by Menengai wells. Methane appears to vary with the locality of the wells, although notably present in relative abundance in lower enthalpy wells. This suggests that the main source of CH4 in the Menengai wells discharge fluid is not the high temperature reservoir at depth despite the possible existence of equilibrium between CO2 and methane inferred from the redox conditions approximated by the Fayalite-Hematite-Quartz (FHQ) buffer (Figures 14 and 15).

Figure 6: Correlation plot of enthalpy with CO2 (left) and with H2S (right) in the vapor phase GRC Transactions, Vol. 43, 2019

Figure 7: Correlation plot of enthalpy with H2 (left) and with CH4 (right) in the vapor phase

5.1. Gas-mineral equilibria The procedure adopted to assess fluid-mineral equilibrium is that described by Scott et al. (2014). As observed in section 4, the main cause of elevated discharge enthalpies in Menengai is liquid- vapor phase segregation i.e. the retention of liquid in the aquifer rock due to its adhesion onto mineral surfaces. Selected wells that represent the characteristics of different parts of the Menengai geothermal reservoir were used to assess equilibria between gases and the mineral assemblages that control these gases. Three of the models described by Arnórsson et al. (2007) and thermodynamic data presented in Arnórsson et al. (2010) were used to evaluate gas-mineral equilibria in these wells. The calculations were done using the WATCH program (Arnórsson et al., (1982) version 2.4 (Bjarnason (2010). The reference temperature is that used in the WATCH program (Fournier and Potter, 1982) which in most cases compared well with downhole temperature profiles. Model 1 represents an isolated system while model 2 and 3 correspond to a closed and an open system respectively. The equilibrium curves of mineral assemblages that could potentially control the activities of the main reactive gases CO2, H2S, H2 are compared with that of the calculated dissolved gas concentrations assuming unit activity. The results for the models are shown in Figure 8 to 13. From the graphs in Figures 8 to 13, it is concluded that H2 concentrations are higher than would be in equilibrium with respective mineral assemblages in all wells selected. This agrees with earlier observations of significant vapor fraction(s) in the reservoir(s) in Menengai, at least at depth for most wells. CO2 appears to be largely controlled by hydrothermal mineral assemblages except for wells represented by MW-21A which may be have a source controlled contribution. H2S is also in equilibrium with hydrothermal mineral GRC Transactions, Vol. 43, 2019

assemblages with the exception of well MW-18A. The relatively low H2S in well MW-18A and other related wells is still a subject of investigation. The graphs further show the most appropriate models that can be used for the different wells based on their enthalpies.

Figure 8: State of equilibrium between calculated concentrations of CO2, H2S and H2 in the aquifer water in well MW-21A using models 1 and 2. Both models 1 and 2 are appropriate for this liquid dominated well.

Figure 9: State of equilibrium between calculated concentrations of CO2, H2S and H2 in the aquifer water in well MW-01 using models 1, 2 and 3. Models 2 appears to be more appropriate for this well.

Figure 10: State of equilibrium between calculated concentrations of CO2, H2S and H2 in the aquifer water in well MW-18A using models 1, 2 and 3. Models 1 and 3 are appropriate for this well. GRC Transactions, Vol. 43, 2019

Figure 11: State of equilibrium between calculated concentrations of CO2, H2S and H2 in the aquifer water in well MW-19 using models 1, 2 and 3. Model 3 is more appropriate for this excess enthalpy well.

Figure 12: State of equilibrium between calculated concentrations of CO2, H2S and H2 in the aquifer water in well MW-01A using models 1, 2 and 3. Models 3 is the most appropriate for this excess enthalpy well.

Figure 13: State of equilibrium between calculated concentrations of CO2, H2S and H2 in the aquifer water in well MW-20 using models 1, 2 and 3. Models 2 is the most appropriate for this excess enthalpy well.

Kipngok et al. (2014) assessed mineral-gas equilibria in the Menengai geothermal reservoir assuming that Fayalite-Hematite-Quartz (FHQ) buffer approximates the redox conditions GRC Transactions, Vol. 43, 2019 supposedly present in the zones where the gases attained chemical equilibrium. The findings showed that gas equilibrium is attained in vapor-dominated environments not only for the single phase vapor wells but for a majority of two-phase wells also (Figures 14 and 15). Downhole pressure profiles in the single phase steam wells and in the deeper portions of a number of two- phase wells show the presence of vapor under shut-in conditions further strengthening these findings.

Figure 14: Graph depicting the reservoir fluids tapped by Menengai wells compared with the theoretical compositions expected for gas equilibration in a single vapor phase, in a single liquid phase (i.e. pure water), and in two-phase systems with different values of y = mass ratio of vapor/(vapor + liquid), under redox conditions controlled by the FHQ redox buffer (modified from Kipngok et al., 2014). GRC Transactions, Vol. 43, 2019

Figure 15: Graph of log KC as a function of temperature and y, the fraction of steam showing the effects of vapor gain or loss with respect to the composition expected for the discharge of a pure equilibrium liquid phase (from Giggenbach, 1980). Also shown are the reservoir fluids discharged by Menengai wells (modified from Kipngok et al., 2014).

6. Origin of Fluids Available water and gas isotope data is presented in Figure 16. Stable isotopes of water indicate a meteoric origin of the recharge water. The effect of water-rock interactions is evident in some of the wells, being enriched in O-18 (oxygen 18 shift). Steam condensate samples are typically isotopically lighter due to phase separation. Recent data of 3He/4He isotope ratio and 13C value of CO2 obtained by West Jec in 2018 through a capacity strengthening collaboration project 13 between JICA and GDC suggests a deep (mantle) origin for the gases. The  C value of CO2 in Menengai geothermal wells fluids however is generally lighter (ranging from -5.9 to -16 ‰) and could possibly be the result of mixing between deep CO2 (from the mantle) with CO2 produced by thermogenic decomposition of organic matter. Calcite precipitation at temperatures lower than 13 193°C could also be invoked to explain the comparatively low δ C values of CO2 (Kipngok et al., 2018). In order to ascertain the contribution of thermogenic decomposition of organic matter to 13  C value of CO2, isotopic composition of CO2 in CH4 is proposed and recommended. GRC Transactions, Vol. 43, 2019

12.0 Central Menengai Wells Eastern Menengai Well (MW-18A) Winsor BH wells Eburru fumarole fumaroles fumaroles Silali fumaroles 10.0 organic 13C mantle 13C

8.0

6.0

3He/4He (R/Ra) 3He/4He 4.0

2.0

0.0 -20 -15 -10 -5 0 13 delta- C(CO2) (permil)

18 3 4 13 Figure 16: Correlation plot of D with  O (left) and He/ He (R/RA) with  C of CO2 (right)

7. Geochemical Models

7.1. SiO2-Enthalpy Plot

The correlation plot of SiO2-enthalpy (measured by the Russel James method) as shown in Figure 15 was used to obtain the apparent silica equilibrium temperature of the geothermal liquids entering the wells as well as evaluate the mixing scenario between the shallow end-member feed zone and a deep end-member steam dominated feed zone. The SiO2 vs. enthalpy plot shows: (i) the solubility of quartz in pure water (liquid and steam), according to Fournier and Potter (1982), (ii) the silica function of Giggenbach et al. (1994), comprising the solubility of both quartz (at high temperatures) and chalcedony (micro- to crypto-crystalline quartz) at low temperatures. The solubility of quartz in pure water, according to Fournier and Potter (1982), is represented by a bell- shaped symmetrical curve, which reaches a maximum of 770 mg/kg at 1594 J/g (corresponding to 340°C) and decreases with further increase in enthalpy. The silica function of Giggenbach et al. (1994) is very close to it below 320°C (that is for enthalpy lower than 1462 J/g). Notably, the measured temperature of the deeper permeable horizons in Menengai wells is generally higher than that of quartz equilibrium for majority of wells as depicted in Figure 16. This is attributed to the significant variation in temperature for the shallower and deeper feed zones in these wells. Three categories of reservoir fluids are inferred from Figures 17 and 18: (i) fluids with a relatively high vapor fraction (MW-20A, -01A, -09C, -19, -09B, -10B, 1-0A, -19A), (ii) liquid dominated fluids (MW-01, -21A, -18A, -04, -13B, -12, -17A, -03,) and (iii) fluids that have undergone conductive cooling and/or mixing with lower enthalpy fluids (MW-19B, MW07 and MW-09A). GRC Transactions, Vol. 43, 2019

Figure 17: Correlation plot of total discharge concentration of SiO2 vs. enthalpy for the reservoir fluids tapped by Menengai wells, also showing quartz solubility in pure water (Fournier and Potter, 1982) and solubility of quartz/chalcedony (Giggenbach et al., 1994).

450 MW-01 MW-01A MW-03 MW-04 MW-07 MW-09A MW-09B MW-09C MW-10A MW-10B MW-12 MW-13B MW-17A MW-19 MW-19B MW-20 MW-21A MW-18A 400

350 )

℃ 300

contribution of cooler water from 250 shallower feed

200 T SiO2 SiO2 (Qz: T F&P) (

150

100

50 50 100 150 200 250 300 350 400 450 T measured at deep feed zone (℃) Figure 18: Correlation plot of quartz equilibrium temperature with measured temperature of deeper feeders in Menengai

GRC Transactions, Vol. 43, 2019

7.2. Cl-enthalpy Plot

The Cl-enthalpy plot of Figure 19 depicts the different characteristics of Menengai wells and the possible processes that are responsible for the observed variations in the chemistry of wells discharge fluid. It is inferred from Figure 19 that the main upflow is located in the central part of the caldera in the vicinity of well MW-13 that is characterized by high enthalpy fluids. In the eastern sector, it is possible that another upflow exists around well MW-18A (Lucy and Kipngok, 2018). More deep wells are being drilled in this sector and the results will augment the current findings. Several processes influence the fluid composition leading to the observed variation in the chemistry of the discharge fluid, which include; boiling and phase segregation (the dominant process), mixing/dilution and conductive cooling. Addition of magmatic fluids is another possible factor. The cooler fluids appear to be dominant in wells located within the margins of the upflow area in the central dome.

3000 MW-01 MW-01A MW-03 MW-04 MW-07 MW-09A High enthalpy fluids MW-09B MW-09C MW-10A MW-10B MW-12 MW-13 pure steam near heat source/up- MW-13A MW-13B MW-17A MW-19 MW-19A MW-19B flow MW-20 MW-21A MW-18A Boreholes 2500

Two-phase fluids sorrounding heat source/up- flow 2000 eastern part

condensation steam loss 1500 steam loss 300oC

dilution 250oC

1000 Discharge Enthalpy Discharge Enthalpy (kJ/kg) conductive cooling 200oC steam-heated 150oC 500 100oC Cooler liquid water at periphery

shallow aquifer 0 0 500 1000 Cl in total discharge (mg/kg)

Figure 16: Cl-enthalpy plot of Menengai wells fluids 7.3. Geochemical Conceptual Model Summing up, the chemical and isotopic characteristics of the Menengai wells discharge fluid can be explained using the conceptual model shown in Figure 20. The main recharge is from the east of the field with a possible outflow to the west. The major upflow is located within the central GRC Transactions, Vol. 43, 2019 dome area around well MW-13 although the existence of a second one on the eastern sector of the caldera in the vicinity of well MW-18A is possible. Three categories of fluids are deduced; high enthalpy fluids (dominantly steam), two phase fluids and fluids that have experienced conductive cooling and/or mixing.

W E

outside caldera MW-09 MW-07 boreholes MW-13 MW-03 MW-18A fumarole fumarole

~60oC dilution impermeable dilution dilution cap outflow High enthalpy fluids o <200 C <200oC impermeable Cooler liquid Cooler liquid water cap water conductive conductive cooling cooling ? Two-phase Two-phase fluids fluids ~330oC boiling ~340oC

meteoric meteoric magmatic gases recharge recharge (CO2, H2S, He, etc.) + organic carbon (?) magmatic gases & conductive heat (CO2, H2S, He, etc.) & conductive heat magmatic heat source Figure 17: Geochemical conceptual model of the Menengai geothermal reservoir

8. Conclusion The following deductions are made from the chemistry of well discharge fluids in Menengai;

 Menengai reservoir water belongs to the Na-HCO3 facies although fluids of wells located in the northern sector of the central part of the field may be classified as Na-HCO3-Cl type.  H2S in the reservoir(s) in Menengai is largely controlled by hydrothermal minerals but H2 is higher suggesting the presence of significant vapor fraction in the reservoir, albeit in varying proportions/degrees between wells. Besides being largely controlled by hydrothermal minerals, CO2 may also be source controlled with an organic contribution as well. Different models work for different wells.  The widely varying chemical properties from well to well point to heterogeneity of the reservoirs feeding Menengai wells. Phase segregation is the major cause of excess enthalpy.  The source of recharge water in Menengai is meteoric while a mantle origin of the gases is inferred although a thermogenic contribution cannot be ruled as deduced by carbon isotopes. Studies on isotopic composition of CO2 in CH4 are strongly recommended to confirm the contribution of organic CO2. GRC Transactions, Vol. 43, 2019

 The upflow is located in the locality of well MW-13 within the central part of the Menengai geothermal field although existence of another upflow is possible in the eastern sector of the field, in the vicinity of well MW-18A.  The dominant processes occurring in the Menengai reservoir are reservoir boiling and phase segregation with mixing and conductive heat transfer/cooling playing a notable role.

REFERENCES

Arnórsson, S., Angcoy, E., Bjarnarson, J.Ö., Giroud, N., Gunnarsson, I., Kaasalinen, H., Karingithi, C. and Stefánsson, A.: Gas chemistry of volcanic geothermal systems, Proceedings World Geothermal Congress, Bali, Indonesia, (2010), 6 pp.

Arnórsson, S., Angcoy, E., Bjarnarson, J.Ö., Giroud, N., Gunnarsson, I., Kaasalinen, H., Karingithi, C., Stefánsson, A. “Gas chemistry of volcanic geothermal systems.” Proceedings of World Geothermal Congress (2010), Bali, Indonesia, 6 pp.

Arnórsson, S., Sigurdsson, S., and Svavarsson, H. “The chemistry of geothermal waters in Iceland: I. Calculation of aqueous speciation from 0° to 370 °C.” Geochim. Cosmochim. Acta 46 (1982), pp. 1513–1532.

Arnórsson, S., Stefánsson, A., Bjarnarson, J.Ö. “Fluid-fluid interaction in geothermal systems.” Reviews in Mineralogy & Geochemistry (2007), 65, 259-312.

Auko, L.: Evaluation of Fluid Mineral Interaction in the Menengai Geothermal System, Central Rift-Kenya, UNU-GTP publications (2014), Report 8.

Bjarnason J.Ö. “The speciation program WATCH, version 2.4, user‟s guide.” Iceland water chemistry group (2010), Reykjavik, 9 pp.

Edmunds W. M. and and Smedley, L. P. “Fluoride in natural waters.” In: Essentials of Medical Geology, Second Edition (2013). Eds: Selinus, O., Alloway, B., Centeno, J.A., Finkelman, R.B., Fuge, R., Lindh, U. and Smedley, P.L. Springer pp 311-336.

Fournier R.O. and Potter R.W. II. “A revised and expanded silica geothermometer.” Geothermal Resources Council Bulletin, (1982): 3-12

Geotermica Italiana Srl.: Geothermal reconnaissance survey in the Menengai- Bogoria area of the Kenya Rift Valley, UN (DTCD)/ GoK, (1987). Giggenbach, W. F., Sheppard, D. S., Robinson, B. W., Stewart, M. K., & Lyon, G. L. “Geochemical structure and position of the Waiotapu geothermal field, New Zealand.” Geothermics (1994), 23(5-6), 599-644. GRC Transactions, Vol. 43, 2019

Griffith P. S., and Gibson, I.L.: The geology and the petrology of the Hannington Trachyphonolite formation, Kenya Rift Valley, Lithos, 13, (1980), 43-53pp. Griffith P. S.: Box fault systems and ramps: Atypical association of structures form the eastern shoulder of the Kenya rift, Geol. Mag., (1980), 579-586 pp. Griffith P. S.: The geology of the area around Lake Hannington and the Perkerra River , Kenya, PhD thesis, University of London, (1977). Jones, W.B., and Lippard, S.J.: New age determination and geology of Kenya rift-Nyanzian rift junction, west Kenya.” Joun. of Geol. Soc.Lon, (1979), vol 136 pg 63. Jones, W.B.: Discussion on geological evolution of trachytic caldera and volcanology of Menengai volcano, Rift Valley, Kenya, Journ. Geop\l. Soc. Lon, (1985), vol 142, 711pp. Kahiga, E.W. “Borehole geology and hydrothermal alteration mineralogy of well MW-13, Menengai geothermal field, Kenya.” UNU-GTP Publications (2014), Report 16, 34 pp. Kipchumba, L.J. “Borehole geology and hydrothermal alteration of wells MW-08 and MW-11, Menengai geothermal field, Kenya.” UNU-GTP Publications (2013), Report 10, 143-176.

Kipng’ok, J.: Fluid chemistry, feed zones and boiling in the first geothermal exploration well at Menengai, Kenya, UNU-GTP publications (2011), Report 15, 281-302.

Kipngok, J., Auko, L, Kanda, I and Suwai,J.: Characterization of the Menengai High Temperature Geothermal Reservoir using Gas Chemistry, Geothermal Resources Council Transactions, (2014). Vol. 38.

Kipngok, J., Suwai, J., Auko, L., Malimo, S., Mulusa, I., Chepkemoi S. and Marini L. “Assessment of Carbon Dioxide (CO2) Partial Pressures in the Menengai Geothermal Area, Kenya.” Proceedings, ARGeo C7 (2018), Kigali, Rwanda.

Leat, P.T., Macdonald, R. and Smith, R.L.: Geochemical evolution of Menengai, Kenya Rift Valley, J Geophys Res, 89, (1984), 8571-8592. Leat, P.T.: Discussion on the geological evolution of the trachytic caldera volcano Menengai, Kenya Rift Valley, Journal of the Geological Society; 142 (4), (August 1, 1985), 711 - 712. Leat, P.T.: Geological evolution of the trachytic volcano Menengai, Kenya Rift Valley, J Geol Soc London, 141, (1984), 1057-1069. Leat, P.T.: The structural and Geochemical Evolution of Menengai caldera volcano, Kenya Rift Valley, PhD Thesis, University of Lancaster, U.K, (1983). Lopeyok, T.P. “Borehole geology and hydrothermal mineralisation of wells MW-09 and MW-12, Menengai geothermal field, Kenya.” UNU-GTP Publications (2013), Report 15, 289-324 Malimo, S.J.: Fluid Chemistry of Menengai Geothermal Wells, Kenya, Geothermal Resources Council Transactions (2013). Vol. 37, GRC Transactions, Vol. 43, 2019

Mbia P.K. “Sub-surface geology, petrology and hydrothermal alteration of Menengai geothermal field, Kenya.” Msc. Thesis (2014).

McCall, G.J.H.: Geology of the -Thomson’s- Lake Hanningtons Falls area, Geological Survey of Kenya Report No. 78, (1967). Mibei, G., Mutua, J., Kahiga E. and Lopeyok, T. “The Use of Leapfrog Software in Geothermal Conceptual Modelling; Case Study of Menengai Geothermal Field.” GRC Transactions (2017), Vol. 41.

Nicholson, K. “Geothermal fluids: Chemistry and exploration techniques.” Springer-Verlag Berlin Heidelberg New York ISBN (1993) 3-540-56017-3 pp.

Njue, L and Kipngok J., “Menengai Geothermal Field - Eastern Upflow.” Proceedings, ARGeo C7 (2018), Kigali, Rwanda.

Sawyer G.M., Carn S.A., Tsanev V.I., Oppenheimer C., Burton M. “Investigation into magma degassing at Nyiragongo volcano, Democratic Republic of the Congo.” Geochem. Geophys. Geosyst., 9, Q02017, doi (2008):10.1029/2007GC001829.

Scott, W, S., Gunnarsson, I., Arnorsson, S., Stefansson, A. “Gas chemistry, boiling and phase segregation in geothermal systems, Hellisheidi, Iceland.” Geochim. Cosmochim. Acta, (2014) 124, 170-189

Sekento, L.R.: Geochemical and isotopic study of the Menengai geothermal field, Kenya, UNU- GTP Publications (2012). Report 31, 769-792.

Symonds R.B., Rose W.I., Bluth G.J.S., Gerlach T.M. - Volcanic-gas studies: methods, results, and applications. In: “Volatiles in Magmas”, M.R. Carroll and J.R. Holloway (Eds.), Mineralogical Society of America, Reviews in Mineralogy, 30, 1-66 (1994).