Geochemical evidence for the nonexistence of supercritical geothermal fluids at the Yangbajing geothermal field, southern Tibet

Three multiple geothermal a temperature of ∼ 320 at an ∼ 8-km Yangbajing Abstract Exploring and exploiting high-temperature (even supercritical) geothermal resources are significant to meet energy demands and reduce carbon emissions. The Yangbajing geothermal field is the most exploited in China, with the currently highest temperature (329.8 °C) measured in a geothermal well. However, whether there are supercritical geothermal fluids beneath the deep parts of this geothermal field is under controversy. In this paper, the water isotope, chemical compositions, and C–He isotopes of gas samples were collected and analyzed. The geothermal water originated from the mixing of meteoric water and magmatic water (25%). The sources of CO2 in the geothermal field were dominated by the thermogenic degassing of carbonates and metasediments in the crust while the radioactive decay of U and Th in granite is the dominated source of He. The temperatures of three different reservoirs are 150 ± 15 °C, 250 ± 10 °C, and ∼ 320 °C (with a depth of ∼ 8 km), respectively. These were obtained using dissolved gas, soil CO2 flux, and noble gas geothermometers. Unlike other supercritical geothermal fields worldwide with larger, shallower, basaltic magma chambers, the Yangbajing geothermal field has a deep-seated, small-scale, granitic magma chamber. Thus, its geological conditions are not conducive for gestating supercritical fluids. These results are of great significance for exploring and developing high-temperature (even ultra-high-temperature) geothermal resources in China.


Introduction
Clean and renewable energy sources are increasingly becoming important in response to global climate change.
Among these, geothermal energy contributes more and more to reducing environmental pollution and increasing the proportion of renewable energy (Karlsdottir et al., 2020). Earth's heat energy is immense-over 99% of the Earth has temperatures exceeding 1000 °C (Chambefort and Stefá nsson, 2020). Thus, the exploration and exploitation of geothermal energy, particularly high-temperature and supercritical (T > 374 °C, P > 22 MPa) geothermal resources (Reinsch et al., 2017, Feng et al., 2021, have become essential in strategically planning energy consumption for all countries around the world.
Most of the high-temperature geothermal systems in China are located in Tibet, Yunnan, and Sichuan provinces, forming part of the Himalayan geothermal belt (Liao, 2018). The Yangbajing geothermal field in southern Tibet, China, is the earliest explored and developed high-temperature geothermal field in China (Dor, 2003, Guo et al., 2007. It has been continuously studied for more than 40 years, and its current installed capacity has reached 25.18 MW (Zhang et al., 2019). A steady-state temperature measurement of its borehole ZK4002 was conducted in 1994 and the measured temperature was 329.8 °C at a depth of about 1500 m, which is currently the highest measured temperature in a geothermal well in China (Fan, 2002, Guo et al., 2007. After years of research, the characteristics and properties of the fluid sources (Guo et al., 2007, Guo, 2012, Yuan et al., 2014, heat source (Brown et al., 1996, Nelson et al., 1996, Wei et al., 2001, and fluid channels and reservoir (Zhao et al., 2011, Feng et al., 2012 of the Yangbajing geothermal field have been studied systematically on varying degrees from the perspectives of geology, fluid geochemistry, geophysics, and altered minerals (Makovsky and Klemperer, 1999, Dor, 2003, Wang et al., 2018. Thus, a relatively consistent knowledge of the geothermal field has been formed. The circulation of the thermal groundwater of the Yangbajing geothermal field is driven by the huge terrain elevation difference. The thermal groundwater is currently estimated to exceed a 5 km depth, where it is speculated to have reached the brittle-ductile transition zone (Su et al., 2020). As previously mentioned, the geothermal fluid temperature exceeds 300 °C at a depth of 1.5 km. If it reaches a depth of 5 km, the temperature would exceed 400 °C according to the geothermal gradient (40 °C/km) (Kang et al., 1985, Zhao et al., 2011 and form supercritical geothermal fluids. This raises the question of whether there are supercritical geothermal fluids in the deep part (>5 km) of the Yangbajing geothermal field. Analyzing the gas characteristics of the geothermal field and estimating the temperature of the reservoir are the most economical and convenient ways to answer this question.
Gases are the bridge between the deep structures and energy characteristics and the surface thermal manifestations of a geothermal field. The chemistry and the isotopes of gases play an important role in geothermal exploration and development (Arnórsson, 2000). For example, geothermal gas and isotopic characteristics can reveal the heat source properties (Karakuş, 2015, Tian et al., 2018, Pan et al., 2021 and reflect the geodynamic mechanisms (Ciotoli et al., 2016) of geothermal fields during the geothermal exploration stage. Changes in geothermal gas characteristics over time can reveal the responses of thermal reservoirs during geothermal energy development and utilization, providing meaningful guidance for the sustainable use of geothermal resources (Caracausi et al., 2005, Monterrosa and Montalvo López, 2010, Tassi et al., 2014. Through the research and analysis of geothermal gas and isotopic characteristics, identifying the source of the fluid composition of the geothermal field and the geochemical processes that the fluid experienced (such as magma degassing and mixing) is possible (Fan et al., 2019;Tian et al., 2019;Hao et al., 2020b). Moreover, some abnormal gas components such as hydrogen and helium are also important associated resources (Hao et al., 2020a).
Geothermal gases can also be used in estimating the temperature through the thermodynamic equilibrium process of geothermal gases in a reservoir. Sets of gas geothermometers based on the concentration of reactive gases have been used to investigate the subsurface temperatures of geothermal systems. These include some single gas thermometers such as CO2, H2S, or H2 (Arnórsson and Gunnlaugsson, 1985) and those with two or more gases such as CO2-H2, H2S-H2, and H2O-H2-CO2- CO-CH4 (Chiodini andMarini, 1998, Fiebig et al., 2004). These may also be based on gas ratios such as CO2/Ar, H2/Ar, and CO2/N2 (Arnórsson, 2000). Notably, the estimated temperatures based on gas-gas relationships and gas-mineral equilibria in reservoirs and the gas content and ratio in the geothermal fluid are quenched as these fluids ascend to the surface (Giggenbach, 1980, Arnórsson, 2000.
Thus, these reactive gases are sensitive to the exiting processes (e.g., steam condensation and boiling) of fluids as they ascending to the surface. Even so, they have been used in many cases and still give appropriate reservoir temperatures (Karolytė et al., 2019). However, these inherent processes limit the use of gas geothermometers and make it hard to determine reservoir conditions and secondary processes (Stefá nsson, 2017). Moreover, reservoir temperatures obtained by assuming a gas-gas or gas-mineral equilibrium are not always acceptable (Lowenstern et al., 2015, Karolytė et al., 2017. Fortunately, the development of some gas geothermometers in recent years (Harvey et al., 2017, Byrne et al., 2021 has provided new opportunities for reservoir temperature estimation with advanced experimental technologies, which may be useful in detecting supercritical geothermal systems. Most supercritical geothermal systems are related to magma chambers at depths of 2-6 km (Scott et al., 2015, Stimac et al., 2015, Jolie et al., 2021. Significant and focused heat transfer typically occurs between the upside of magma chambers and the bottom of production wells. This part is almost at the root of conventional geothermal systems (Scott et al., 2017), with the most favorable conditions for high-enthalpy and supercritical geothermal systems (Scott et al., 2016) and major types of ore deposits (Heinrich and Candela, 2014). Several supercritical geothermal systems have been observed around the world, including those at the Salton Sea, The Geysers, and Hawaii, USA (Garcia et al., 2016, Kaspereit et al., 2016; Kakkonda, Japan (Muraoka et al., 1998);Larderello, Italy (Minetto et al., 2020); Reykjanes, Krafla, andNesjavellir, Iceland (Friðleifsson et al., 2017, Reinsch et al., 2017); Los Humeros, Mexico (Peiffer et al., 2018);and Menengai, Kenya (O'Sullivan et al., 2015). In this regard, supercritical fluids are related to the circulation of infiltrated meteoric water around magma intrusions (Hayba andIngebritsen, 1997, Scott et al., 2016). Their primary components are CO2, H2S, and other acid gases due to the magma degassing (Henley andSeward, 2018, Heřmanská et al., 2020). They may result from chemical and physical processes, such as phase segregation, conductive heating of groundwater near an intrusion, magma degassing, or condensation of liquids (Heřmanská et al., 2019). The host rock permeability, the emplacement depth and geometry of an intrusion, the temperature-dependent permeability near an intrusion, and the depth and extent of boiling zones are the key geological controls affecting supercritical fluids (Scott et al., 2016). Boiling zones are located within 1 km above intrusions at depths of 3-4 km with permeabilities of >10 − 15 m 2 . These boiling areas get smaller with deeper intrusions and lower permeabilities (Scott et al., 2017). The permeability may decrease to 10 −20 m 2 at a depth of 10 km, making it impossible for supercritical geothermal fluids to form (Wanner et al., 2020).
In this paper, we collected geothermal fluids samples from hot springs and geothermal wells and compiled the historical data of the Yangbajing geothermal field. We wanted to determine whether there are supercritical geothermal fluids in Yangbajing using fluid chemistry and isotope evidence. The geochemical characteristics of the geothermal fluids were analyzed, and the reservoir temperatures of the geothermal field were re-evaluated using several new geothermometers. The results are significant for the exploration and development of high-temperature and ultra-high-temperature geothermal resources around China and for the in-depth understanding of the geodynamic processes of the Yadong-Gulu rift zone.

Geological setting of the Yangbajing geothermal field
The Yangbajing geothermal field is located in the hinterland of the Qinghai-Tibet Plateau, 90 km northwest of Lhasa ( Fig. 1). This geothermal field has the largest installed capacity for geothermal power and provided 60% of Lhasa's electricity supply in the 1990s (Dor, 2003). The geothermal wells and numerous thermal springs in this geothermal field cover an area of approximately 40 km 2 . The annual average temperature in Yangbajing is 2.5 °C, and the highest and lowest recorded air temperatures are 23.4 °C and −25.7 °C, respectively. The average annual rainfall in Yangbajing is 383 mm, and 65% of which is concentrated from July to August (Guo et al., 2007).
The Yangbajing geothermal field is located in the middle part of the famous Yadong-Gulu rift system. To the northwest are the Nyainqentanglha Mountains, and to the southeast are the Tang Mountains (Fig. 1b). The basement of the geothermal system comprises Carboniferous-Permian slate, Upper Cretaceous carbonate, and Jurassic metasediments in the southeast, with Paleozoic gneiss and migmatite in the northwest (Dor, 2003, Wang et al., 2014. The magmatic rock around the geothermal field includes the Late Yanshanian-Himalayan granite, with a K-Ar radiometric age of 8.1 Ma (Fig. 2a) (Zhao et al., 2003). However, no Quaternary volcanic activity has been found in the Yangbajing area so far, and partial melting at depths of 15-20 km is speculated as the heat source of the geothermal system (Teng et al., 2019) based on the observed "bright spots" from the INDEPTH (International Deep Profiling of Tibet and the Himalaya) project (Brown et al., 1996, Chen et al., 1996, Kind et al., 1996, Makovsky et al., 1996, Nelson et al., 1996. Two sets of normal faults striking NNE and NE are respectively distributed on the east and west sides of the faulted basin. These extensional faults are inferred as flow channels for the storage and migration of geothermal fluids as they continued to be active until the late Quaternary ( Fig. 3) (Guo et al., 2007). The Yangbajing area has strong surface thermal manifestations, including boiling springs, hot springs, hydrothermal explosions, steaming grounds, and sinters deposited before the geothermal power plant was constructed in 1981 (Tong et al., 2000). However, some of these hydrothermal activities have disappeared as the geothermal exploration continues (Liao, 2018). The current electric power generated at the geothermal field mainly depends on the geothermal fluids from the shallow reservoir, whereas less than one-fifth is from the deep reservoir (Guo et al., 2007). The precipitation and snowmelt from the Nyainqentanglha Mountains (with an elevation between 4500 and 5800 m) are the primary recharge source of the geothermal field in the northwest (Guo, 2012, Tan et al., 2014. Meteoric water infiltrates to a certain depth and is heated by the host rock. Thermal fluids rise into the deep reservoir along the fault belt and then flow into the shallow thermal reservoir, where they mix with cold groundwater and flow to the southeast part of the field together ( Fig. 3) (Guo, 2012).
There are two distinct reservoirs under the Yangbajing geothermal field: the deeper one is at a depth of 950-1850 m, whereas the shallower reservoir is at 300-500 m (Dor, 2003, Guo, 2012. The temperature of the deep reservoir is about 250 °C, according to the data from the borehole ZK4001. In contrast, the temperature of the shallow reservoir is between 130 °C and 170 °C (Guo, 2012). The petrography of the shallow reservoir is mainly weathered granite and Quaternary sandstones in the northwestern part of the geothermal field and Himalayan granite overlain by Quaternary sediments and underlain by boulder clay and silty clay in the southeastern part ( Fig.   3) (Guo et al., 2007). The deep reservoir comprises fissure granitic mylonite, biotite granite, and fissure granites covered with strongly weathered granites, which have low permeability as the initial feldspar has been altered to kaolinite. The plastic shear zone consists of granitic mylonite toward the southeast, with a dip of 30° cutting the granite bedrock. The high-angle normal faults in front of the Nyainqentanglha Mountains extend to a depth of more than 5 km, cutting the granitic mylonite and forming deep thermal reservoirs along with these fragments (Fig. 3) (Guo et al., 2007).    Searle et al., 1987, Dor et al., 1997, Zhao et al., 2001, Craw et al., 2005, Kapp et al., 2005, Lee and Whitehouse, 2007, Wu et al., 2007Hacker et al., 2014;Wang et al., 2014, Weller et al., 2016Gé belin et al., 2017;Wu et al., 2021.

Sampling and analysis
Several types of fluid samples were collected from the Yangbajing area in 2020 for the chemical and isotopic analyses: four geothermal water, two cold groundwater, four condensate water, three surface water, and four gas samples. Also, parameters such as the pH, temperature (°C), total dissolved solids (TDS) (ppm), oxidationreduction potential (ORP) (mV), and electrical conductivity (EC) (µS/cm) of the water samples were measured at the field using handheld meters (HQ40D, Hach). The sampling locations are shown in Fig. 2b. The borehole gas samples were collected using the water displacement method (Tian et al., 2019) after the geothermal fluids were separated using wellhead water-steam separators, following the method described by Arnórsson et al. (2006). A 50-mL narrow-mouth glass bottle was filled with geothermal water from the sample well and submerged upside down into the thermal water in a barrel. Then, a tube linked to the separator was inserted into the glass bottle, and the water in the bottle was displaced by the gas. The bottle was plugged with a silicone stopper and sealed with an aluminum cap after the water in the bottle was discharged to two-thirds of the bottle's volume. The sampling bottle was sealed upside down in a 500-mL HDPE bottle full of corresponding geothermal water with no headspace to prevent atmospheric contamination of the gas sample. Additionally, we collected four or five parallel samples from each well for the detection of their gas components, noble gas isotopic compositions, and carbon isotopic compositions.
Hydrogen and oxygen isotopes were measured by a laser water isotope analyzer (L1102-I, Picarro). The isotopic ratios were based on the standard of the Vienna Standard Mean Ocean Water (VSMOW). The analytical precisions of δD and δ 18 O were 0.5‰ and 0.1‰, respectively. The compositions of the major gas chemical species, including CH4, CO2, H2, N2, O2, Ar, He, and H2S, were determined using a MAT 271 mass spectrometer with relative standard deviations of less than 5% and expressed in percent by volume. δ 13 C-CO2 values (with respect to Vienna Pee Dee Belemnite [VPDB]) were obtained using a gas isotope mass spectrometer (Delta VTM, Thermo Finnigan) coupled with an online sample preprocessor (Tian et al., 2019). The measurement error for the ratio of carbon isotope values was ±0.2‰. 3 He/ 4 He and 4 He/ 20 Ne ratios were analyzed using a Noblesse noble gas mass spectrometer produced by Nu Instruments, UK. The noble gas isotopic ratio measurement results had errors of less than 7%. These measurements were completed within 1 month after the field works.

Chemical parameters and isotope characteristics of the water
The water types of all the samples were classified as river water, cold groundwater, thermal groundwater, and condensate water (Table 1). All samples in the study were alkaline (pH > 7), and most of the geothermal water had a pH of ∼9. The TDS of the geothermal water ranged from 439 to 1754 ppm with an average of 1323 ppm, whereas that of cold water was less than 700 ppm. Most of the geothermal water had a high EC of over 2000 μS/cm, whereas the cold water had an EC below 1000 μS/cm. Almost all of the water samples had negative ORP values except No. B-02, indicating that those waters are in a reducing environment.  (Table 3). c Mantle: δ 13 C = −6.5‰, CO2/ 3 He = 1.5 × 10 9 ; limestone: δ 13 C = +1.5‰, CO2/ 3 He = 1 × 10 13 ; sediments: δ 13 C = −30‰, CO2/ 3 He = 1 × 10 13 (Sano and Marty, 1995).

Dissolved gas geothermometer
Based on gas-gas and gas-rock reactions, the relationships between gas components could be used to evaluate the physical and chemical states of geothermal reservoirs. However, it is recognized that gas geothermometers have some characteristics that are difficult to control. A significant hurdle to using gas component geothermometers is the redox state of fluids, especially for reactive phases such as H2S, H2, CH4, or CO, which are sensitive to redox potential. Highly immature geothermal areas are closely related to active volcanoes (Giggenbach, 1987). They usually have acidic, sulfate chloride waters; and an assembly of highly oxidizing alteration minerals, such as gypsum and alunite. In contrast, mature geothermal areas have neutral-pH discharges; low sulfuric acid, chlorine-dominated hot waters; potassium feldspar and potassium mica (muscovite) symbiosis; and calcite alteration to produce Ca-Al-silicate rock. For more oxidized states, such as those with gypsum and alunite, CO2/CH4 geothermometers are less suitable. Moreover, Giggenbach (1993) showed that RH is a parameter that quantifies the redox state, with RH = −2.8 indicating a steady-state chemical equilibrium of fluid components.
The gas system H2O-CO2-H2-CO-CH4 is an effective geothermometer widely used for geothermal fluids (Chiodini and Marini, 1998). However, the H2O steam in a sample pool is affected by boiling as geothermal water flows to the surface. Also, CO undergoes oxidation after mixing with shallow aquifer during the ascent of geothermal fluids. Therefore, we used the CO2-CH4-H2 system in the geothermal system of Yangbajing. A redoxcondition independent diagram ( Fig. 4; H2/Ar* vs. CH4/CO2) was combined to determine the gas equilibrium temperature, and the degree of full redox equilibrium was attained. In typical situations, the temperature-controlled With the assumption that Ar in a geothermal fluid is consistent with air-saturated water (ASW) (Giggenbach, 1987), a geological thermometer based on the H2-Ar-CH4-CO2 gas-liquid dissolution equilibrium relationship was established. When water boils, the relationship between hydrogen and Ar is as follows:(4)logXH2XArv=RH+6.52in the liquid phase, the characteristic formula is:(5)logXH2XArL=RH-logBH2+6.52where BH2 is the equilibrium coefficient of H2 between the gas and liquid phases. Therefore, the log(XH2/XAr * ) vs. log(XCH4/XCO2) diagram (Giggenbach, 1993) shows that the RH value in the gas-liquid equilibrium is −3.6 to −2.8, which also represents a temperature increase from 125 °C to 374 °C ( Fig. 4 in Giggenbach (1993)). The four formulas above show BCO2, BCH4, and BH2 values as reported by Sepúlveda et al. (2007). The thermometer diagram of the CO2-CH4-H2 system (Fig. 4) shows that the solution equilibrium temperature of the Yangbajing geothermal field tends to about 140 °C-160 °C, and the RH value is between −3.2 and −3.6.

Soil CO2 flux geothermometer
Hot springs, especially those formed by steam-heated meteoric water, are usually of the acid sulfate variety. These hot water samples contain limited reservoir chemistry information and are not perfect for geochemical geothermometers (Arnórsson, 2000, Mukherjee and Singh, 2020). Thus, obtaining sufficient samples and reconstructing the composition of the parent geothermal fluid by correcting the mixing and degassing processes are necessary for these hydrochemical samples to be usable as geothermometers (Fournier, 1977, Palandri andReed, 2001). A common problem with geochemical thermometers is that geothermal water easily rebalances with surrounding rocks as it rises from the reservoir to the surface. They also easily precipitate and dissolve, making deviations common. Thus, the dissolved gas equilibrium geothermometer may have some deviations due to easy air mixing during sampling or the enrichment of non-condensable gas due to vapor condensation (Arnórsson and Gunnlaugsson, 1985). Then, the steam flux in the geothermal area can be as follows: (7)Fstm=Qtoths-hw-1where Fstm is the steam flux (kg m −2 s −1 ), Qtot is the inferred heat flux (Eq. (6)), hs is the enthalpy of steam at the local boiling point (kJ kg −1 ), and hw is the enthalpy of liquid water at ambient conditions (kJ kg −1 ). The CO2 concentration in the steam is then converted into temperature:(8)TCO2Flux=-44.1+269.25R-76.88R2+9.52R3where TCO2Flux is the reservoir temperature (°C), and R is the logarithm of the CO2 concentration in the steam supplying the thermal area (log mmol kg −1 ) from the CO2 flux measurements and Eq. (7). Eq. (8) applies to high-temperature geothermal reservoirs hosted in mafic to silicic rocks (Arnórsson and Gunnlaugsson, 1985). This method should also be effective in the Yangbajing area.
Soil CO2 flux and soil temperature data from Zhang, 2015 have been used in the Yangbajing geothermal field to estimate its reservoir temperature (Table S2). The data shows two different geothermal zones in the Yangbajing geothermal field (Fig. 3).

Noble gas geothermometer
Among geothermal fluid geothermometers, many empirical gas geothermometers have been developed earlier, such as those for CO2 and H2S (Arnórsson and Gunnlaugsson, 1985) or the CO2-H2S gas equilibrium system (Arnórsson andGunnlaugsson, 1985, Chiodini andMarini, 1998). There are also CO2/Ar and H2/Ar gas ratios, among others (Arnórsson, 2000), these gas geothermometers are useful to some degree. However, they are all based on the theoretical assumption that the gas compositions or gas ratios in a geothermal reservoir are controlled by gas-gas equilibrium or gas-mineral thermodynamic equilibrium and that this equilibrium is fixed when the local thermal fluid rises from the reservoir to the surface (Giggenbach, 1980). Some problems arise with this assumption, such as the chemical rebalancing of some of the reacting gases as they rise and the condensation and boiling of shallow vapors, which change the compositions of the gases. These conditions limit the use of empirical geothermal gas geothermometers. Further, one more critical problem is the difficulty of distinguishing between the reservoir state and secondary processes (Stefá nsson, 2017). Recently, Byrne et al. (2021) developed a noble gas geothermometer by adopting Icelandic geothermal gas data based on the thermodynamic equilibrium process of noble gas dissolution in geothermal reservoirs. The most typical characteristic of noble gases is that they almost do not react with surrounding rocks. Thus, their rise from the reservoir to the surface is basically only controlled by the thermodynamic equilibrium of dissolution. Therefore, they have a broader application prospect than those of empirical geothermometers.
This method assumes that the isotope fractionation of noble gases (Ne, Ar, Kr, and Xe) between gases and liquids is controlled by temperature. Furthermore, the distribution relationship between noble gas concentration and saturated groundwater can be deduced as follows: (9)[i]v=[i]aswKDi(T)KDi(T)-1Xv+1where the partitioning can be described at equilibrium by the vapor-liquid distribution coefficient. This coefficient is defined as the ratio between the mole fractions of a species (i) in the vapor (v) and liquid (aq) phases, which is the abundance in the initial ASW phase and the subsequent vapor and liquid phases, respectively. In an extreme case, when Xv = 0, the relation can be presented as: (10) We adopted the noble gas data of Zhao et al. (2001) for the Yangbajing geothermal field and estimated the temperatures of single noble gases, as shown in Fig. 6. The relation diagram of 20 Ne, 84 Kr, and 36 Ar shows that the gas temperature of the geothermal field is concentrated on the evolutionary equilibrium line at 220 °C-260 °C.
Considering that these geothermal gas samples could easily mix with atmospheric components, we also used a mixed noble gas geothermometer to estimate the reservoir temperature of the Yangbajing geothermal field, as shown in Fig. 7. As can be seen from the figure, the reservoir temperature is basically concentrated at 240 °C-260 °C, and the proportion of atmospheric mixing is 0.2 on average. This ratio is closely related to that of the samples from the high-altitude area of Yangbajing.  5. Discussion

Sources of geothermal water
The δD vs. δ 18 O relationship in the Yangbajing geothermal area was plotted based on the data from the collected samples and the compiled data (Table S1) (Fig. 8). The cold groundwater, thermal groundwater, and condensate water were near the global meteoric water line (GMWL) (Craig, 1961)   Isotope exchanges between water and rock often result in "oxygen shift" in many high-temperature geothermal systems (Giggenbach, 1992). However, the oxygen isotope shift in the Yangbajing geothermal field was not observable, which might be because of mixing with magmatic water, weak water-rock interactions, or both (Pichler, 2005). The "horizontal only" shift of δ 18 O to the right due to water-rock interactions might be reversed by isotope exchange between CO2 and H2O during CO2 degassing. Oxygen isotope shifts to the left have been observed in some CO2-rich geothermal systems, while most geothermal systems in the Himalaya geothermal belt are rich in CO2 (Girault et al., 2014, Wang et al., 2021. Fig. 8a shows that both δD and δ 18 O shifted right slightly, indicating that they were dominated by mixing with magmatic water. On the basis of the δD and δ 18 O relationship, we suggest that a small amount of the geothermal fluid beneath the Yangbajing geothermal area was affected by magmatic water at depth.

The origin of N2 and Ar
The probable nitrogen sources in the geothermal area are (a) the atmosphere, (b) the mantle, including the mantle wedge and the subducting oceanic crust, and (c) sediments (both subducted oceanic deposits and continental crust) (Inguaggiato et al., 2004). Nitrogen sources in the crust usually involve biological activities (NH4), deposition, and diagenesis (Bebout et al., 2013, Mysen, 2019, Mukherjee and Singh, 2021. As can be seen from the N2-He-Ar triangle diagram of the gas components (Fig. 9), the N2/Ar ratios (ranging from 82 to 128.1) of the newly collected samples were slightly higher than that of ASW (40) and generally similar to that of the air (83). These indicate that these ratios comprise atmospheric sources and excess nitrogen (Table 2). Almost all the samples were close to the He corner within a narrow area (shadow in Fig. 9), indicating mixing with deep-seated nitrogen sources (excess nitrogen), probably crustal metamorphic rocks or mantle volatiles (Evans et al., 2008, Zhang et al., 2017b.
Considering the abundant 4 He from crustal radioactive activities, we inferred that most of the deep-seated N2 of the hot spring gases likely has a crustal genesis from metasedimentary sources. Previous studies on helium in the Tibetan Plateau (Yokoyama et al., 1999, Hoke et al., 2000 showed that only a very small proportion of mantle-derived helium (<5%) was found between the Bangong-Nujiang suture zone and the Yarlung Zangbo suture zone (IYSZ). To the south side of IYSZ, helium is purely crust-derived. Mantle helium is mainly injected into the crust by the upwelling of the asthenosphere. In the IYSZ, the influence of mantle fluid gradually decreases southward until it is replaced by crust-derived helium at about 100 km north (Hoke et al., 2000, Newell et al., 2008. For the Yangbajing area, the 3 He/ 4 He value was very low (generally less than 0.2 Ra) ( Fig. 10). This is much lower than the R values of hot spots or plumes (15-30 Ra;Lupton, 1983), the depleted mantle (8 ± 1 Ra), and oceanic island arcs (7.4 ± 1.5 Ra; Sano and Fischer, 2013). This indicates a dominant crustal source with little mantle-derived helium.  Table 3 show that more than 97% of He is crust-derived. According to the tectonic background, as Yangbajing is located at the boundary between the crust-and mantle-derived helium (Hoke

The origin of CO2
The relationship between δ 13 C-CO2 and CO2-He in gases is an effective tracer to identify the source of CO2. The δ 13 C value in the Yangbajing area ranged from −12.29‰ to −7.72‰, which was between those of marine carbonates (0 ± 2‰) and crustal sediments (−30 ± 10‰) (Tardani et al., 2016), suggesting a possible mixture of the two. In addition, Fig. 11 shows that CO2 in the Yangbajing area may also contain a small amount of MORB components. The samples with high δ 13 C values and CO2/ 3 He ratios were largely associated with carbonate rocks (CO2/ 3 He = 10 13 , Sano and Marty, 1995). Sano and Marty (1995) established a three-terminal mixed model of C-He isotopes, which can identify different endmembers and estimate the proportion of each endmember, especially for those hydrothermal fluids with high CO2 content. The three endmembers in this study were as follows: mantle, carbonate rock, and sediments. We assumed that the mantle portion is a MORB-type mantle to simplify the calculation. The values of each endmember were δ 13 C = −6.5 ± 2.5‰ and CO2/ 3 He = 2 × 10 9 for the mantle portion, δ 13 C = 0 ± 2‰ and CO2/ 3 He = 10 13 for carbonate, and δ 13 C = −30 ± 10‰ and CO2/ 3 He = 10 13 for sediments. The formulas are as follows:C13C12sample=C13C12M×fM+C13C12C×fC+C13C12S×fS1C12/H3esample=fMC12/H3eM+fCC12/H3 eC+fSC12/H3eSfM+fC+fS=1the subscripts sample, M, C, and S, refer to the measurement values of the sample, mantle, carbonate rocks, and sediments, respectively. Fig. 12 shows that the samples in the Yangbajing area (except ZK4001) basically fall within the limits of the mantle, carbonates, and sediments. This indicates that these samples only experienced weak He-CO2 fractionation, which also indicates the effectiveness of calculating the mixing ratios of the three endmembers. The abnormal sample ZK4001 will be discussed later.  Table 3. The proportion of the carbonate components was 54%-68%, with an average of ∼60%. That of the sediment component ranged from 26% to 43%, with an average of ∼35%. The average ratio of the mantle endpoints was ∼5%. The high ratio of the carbonate and C/S (carbonate/sediments) value in the geothermal gases in the Yangbajing area indicated an obvious heterogeneity of the CO2 reservoir in the rift area.
Moreover, a C/S value>1 indicates that the inorganic carbon contribution is greater than that of organic carbon. In the Yangbajing area, the source of organic carbon might have been some metamorphic-sedimentary rocks, including gneiss and migmatite (Singh et al., 1998, Richards et al., 2005. The primary sources of inorganic carbon were marine carbonate rocks, some of which were exposed along the Yadong-Gulu rift near Yangbajing ( It should be noted that as geothermal fluids rise to the surface, they usually involve degassing, calcite precipitation, and other phase separation processes. These affect the He-CO2 characteristics (such as δ 13 C-CO2 and CO2/ 3 He) of hydrothermal fluids (Ray et al., 2009, Barry et al., 2014. The continuous degassing in the local thermal system without deep-source CO2 replenishment led to the decreased He concentration and the very high CO2/ 3 He ratio in the collected gas samples (Ozima and Podosek, 2002). Conversely, degassing of hydrothermal systems at low temperatures (<110 °C) also increases the δ 13 C value in the remaining liquid (Mook et al., 1974). This can be seen in the samples with low CO2/ 3 He ratios, such as sample ZK4001 (Fig. 12). Previous studies showed that the shallow reservoir of Yangbajing (<400 m) also had mixing of cold water and deep thermal fluid, and mixing with shallow oxygen-rich cold water may have stimulated a secondary oxidation process and thus consumed H2. This is supported by the oxidation state coefficient in the Yangbajing area is lower than that of the FeO/FeO1.5 buffer (RH = −2.8), which is a common pair in geothermal systems. This suggests that deep fluids were either oxidized during ascent or mixed with shallow oxidizing fluids, which is consistent with the predicted lower reservoir temperatures.
The resulting temperature range is significantly lower than the measured temperature value (329.8 °C) of borehole ZK4002. This may be because the CO2-CH4 system had obvious H2 rebalancing when the geothermal fluid rose to the surface. In addition, the oxidation state of the CO2-CH4-H2 equilibrium system was stronger than that of the original hydrothermal system (RH = −2.8) (Giggenbach, 1987, Sepúlveda et al., 2007, Cinti et al., 2014. This low RH value generally considers the gas-liquid interactions that occur at ∼140-160 °C. Therefore, it can be inferred that the CO2-CH4-H2 equilibrium system is very sensitive to the rebalancing of H2 due to the decreased temperature. After considering the CO2-CH4-H2 re-equilibrium system, the temperature calculated by the CH4-CO2 system Eq. (3) at RH = −2.8 is 250 ± 15 °C, which can be considered as an estimation of the deep reservoir (∼2 km, Fig. 3) temperature in the Yangbajing geothermal field and it is consistent with borehole temperature measurement (Guo et al., 2007).
There are kaolinite, muscovite, sulfate and natural sulfur, anhydrite, alunite, and other alteration minerals in the northern part of Yangbajing and calcite (travertine) in the southern part (Dor et al., 1997). The characteristics of the altered minerals show that the Yangbajing area experienced different redox states in history. However, the present geothermal water is slightly alkaline, with low SO4, Cl-dominated hot water, and calcite (travertine) as the main precipitate. These indicate that the geothermal system of Yangbajing gradually attained a stable chemical equilibrium state.
On the basis of the above analysis, we prefer that the existence of the weak oxidation state in the gas components may be related to the mixing of the geothermal fluid and shallow cold water during migration. Gas composition rebalancing has occurred, which is more reflective of the shallow geothermal reservoir (∼500 m, Fig. 3) temperature of 150 ± 15 °C (Fig. 4). This result also considered gas composition measurement errors. In addition, this result is consistent with previous research and borehole temperature measurement results (Guo et al., 2007).

Soil CO2 flux geothermometer
The soil CO2 flux geothermometer was developed based on the principle of an earlier CO2 geothermometer (Arnórsson and Gunnlaugsson, 1985) and a large number of CO2 flux measurement data sets (Harvey et al., 2017).
The thermodynamic principle of previous CO2 geothermometers was that the concentration of CO2 in hightemperature hydrothermal geothermal systems is controlled by basic water-rock reaction: plagioclase + CO2 = clay + calcite (Giggenbach, 1981). Empirical geothermometers assume that geothermal reservoir fluids go through adiabatical boiling from the reservoir to the atmosphere. Thus, the CO2/H2O ratio in geothermal reservoirs can be obtained by measuring those in surface fumaroles. At a certain temperature, deep geothermal water boiling at atmospheric pressure is accompanied by a certain mass fraction of steam separation, enabling the prediction of a reservoir's CO2/H2O ratio from fumaroles. However, nonadiabatic heat transfer to the surface, such as secondary boiling or steam condensation in fumaroles, makes geothermometer less suitable. In addition, pulsed degassing CO2 from magma may overwhelm the mineral buffering capacity of overlying geothermal reservoirs, resulting in unbalanced, physically controlled reservoir CO2 concentrations and higher soil CO2 flux. It may be inferred that there is a distinct characteristic display in the northwestern geothermal area of the Yangbajing geothermal field. As a geothermal fluid seepage zone, it has a high reservoir temperature (about 280 °C), whereas the reservoir temperature of the entire Yangbajing geothermal field is ∼257 °C. This result is very close to the measured temperature of 252 °C from borehole ZK4001.
The value of the reservoir temperature is consistent with the soil CO2 flux (Zhang, 2015), indicating that CO2 flux is controlled by H2O flux. According to the calculation process (Table S2), the CO2 flux greatly varies, indicating a very heterogeneous shallow CO2 channel or deep fractures or highly permeable soil cover in the concentrated area.
The H2O flux is converted from an empirical relationship of the soil temperature, and the accuracy of calorimetry measurements has been validated in the Taupo Volcanic Zone of New Zealand (Rissmann et al., 2012, Bloomberg et al., 2014. However, the regression equation for determining the boiling point depth was measured in the Taupo Volcanic Zone in summer with an air temperature of 20 °C. In this paper, the soil CO2 in Yangbajing was also measured in summer, but the average air temperature in the plateau area was slightly lower than that in the Taupo Volcanic Zone. From the estimation results of the reservoir temperature, such temperature difference may introduce a relatively large error, but it is impossible to draw a conclusion at present. If a more accurate regression equation is to be obtained, conducting soil temperature measurements at multiple plateau geothermal areas will be necessary for the future to correct the boiling point depth. Our reservoir temperature estimation results show that the reservoir temperature is consistent with some previous research results (Zhao et al., 1998, Guo et al., 2007, albeit in a broader range (Fig. 13). This range could be because of subsurface differences or various measurement errors. First, an important factor is the heterogeneity of the permeability of the cover layer in the shallow part of the geothermal area that prevents gas from escaping to the atmosphere. For example, huge variability in the CO2 flux occurs when the ground is altered (Chiodini et al., 1996), In addition, according to the genesis model of the Yangbajing geothermal field (Fig. 3), the geothermal system is located in a rift valley. Therefore, it has multiple layers of thermal reservoirs at different depths. Those with temperature estimates over 300 °C may represent deeper reservoir (may be >5 km) temperatures, consistent with the fact that most of the CO2 gas came from deep thermogenic decarbonization. Moreover, in their study on the factors influencing soil CO2 flux measurements, Fairley and Hinds (2004) reported that the deep fractures in the rift zone contributed to the direct arrival of deep CO2 to the surface (Lewicki et al., 2003, Hunt et al., 2017. Such deep-source CO2 likely represents the deep temperature information of the rift zone. Considering that the extremely high CO2 flux was affected by some extreme conditions, we selected the 90% quantile representing the deep reservoir (over 5 km) temperature (Fig. 13) of about 327 °C. This temperature value may represent the upper limit of the reservoir temperature in the Yangbajing area. Further, the geothermal gradient of 4 °C/100 m in this area (Kang et al., 1985) indicates that the circulation depth of the geothermal fluid has reached ∼8 km, which is consistent with that of Gonghe Basin, north of the Tibetan Plateau (Pan et al., 2021). However, this depth exceeds the 2-6-km depths of typical supercritical fluids, and ~320 °C is far lower than the supercritical temperature of pure water. Therefore, we prefer that there is no condition from reservoir temperature of the existence of supercritical geothermal fluids in the Yangbajing geothermal field.

Noble gas geothermometer
Noble gas geothermometer has attracted great attention because of its insensitivity to secondary processes and chemical re-reactions (Mazor and Truesdell, 1984). However, isotopic fractionation due to boiling and phase separation also results in defects (Mazor et al., 1990, Pinti et al., 2017. Thus, the water-gas equilibrium of atmosphere-derived noble gas (ANG) isotope ( 20 Ne, 36 Ar, 84 Kr) abundance in geothermal waters controlled by temperature is developed as a new geothermometer (Byrne et al., 2021).
According to the He and C isotope analysis, the gas in Yangbajing was mixed with a certain proportion of atmosphere-derived gas. This was mainly due to the influence of deep geothermal fluids that reach shallow depths and mix with cold water. However, the 4 He/ 20 Ne ratio of one sample showed obvious atmospheric pollution due to operational problems during the sampling process, which was not considered in the estimation of the reservoir temperature. In addition, the CO2/ 3 He relationship showed that the mantle-derived part of the gas samples comprised less than 3% of the total samples. Therefore, we can consider that the noble gas isotopes collected in the geothermal fluid have atmospheric isotope signatures, which entered the ground after the infiltrated meteoric water and air reached dissolution equilibrium. The noble gas isotope component of the ASW is an endmember. Fig. 6 shows the relationships of the 20 Ne, 84 Kr, and 36 Ar abundances, which fit well with the estimated fractionation lines at temperatures between 200 °C and 350 °C. Notably, even if the air endmember is also shown on this curve, if the noble gas is contaminated with some air, the data points plot close to the air side, resulting in an underestimation of geothermal reservoir temperatures, although the data points are still close to the mixing curve.
In addition, when a high proportion of vapor (Xv) is present, the abundance of ANG is significantly reduced, and the data points are shifted toward the vapor phase. This results in overestimated reservoir temperatures. Therefore, a temperature estimation method considering the air mixing ratio should be introduced. Fig. 7 shows that the extra amount of air mixed into the gas samples is caused by either the sampling process or the degassing process, which increases the ANG abundance and leads to underestimated reservoir temperatures.
According to the relationship of 84 Kr/ 36 Ar and 20 Ne/ 36 Ar vs. 1/ 36 Ar (Fig. 7), air mixing causes adverse effects on the estimation of reservoir temperatures. The samples show apparent deviations from the vapor phase separation line and are close to the air end element. However, this correction might be imperfect because a small fraction of the gas isotopes came from the mantle. Our correction mainly considered only the effects of atmospheric mixing.
Nonetheless, the gas source shows that the influence of mantle gas mixing of less than 3% on the estimation of reservoir temperatures can be ignored. Based on results of this new type of geothermometer, the temperature of the geothermal reservoir in Yangbajing is around 250 ± 10 °C (Fig. 7), which is determined through the comparative analysis of the single noble gases and multiple noble gas mixtures.
It is important to note that conventional gas geothermometers primarily consider temperatures at the last stage of gas-rock equilibria. However, noble gas geothermometer record temperatures at which gases and liquids coexist in reservoirs at phase separation equilibria. Therefore, we must not expect that the reservoir temperatures obtained using multiple methods will be identical. However, different geothermometers can reflect the evolution of reservoir temperatures at different stages. Generally, the reservoir temperatures obtained by noble gas geothermometers are consistent and acceptable. The differences in the results of various geothermometers indicate that each geothermometer has its applicable conditions and complexities. However, the obvious advantage of noble gas geothermometers is that the chemical reactions during the ascent of geothermal fluids can be ignored. heat (Nábělek et al., 2009, Searle et al., 2010, Tian et al., 2020. Yangbajing, we hope to find evidence of no supercritical geothermal fluids in the Yangbajing area. In this paper, water and gas chemistry data from Yangbajing were studied intermittently over 20 years. From the earlier analyses, we can infer that the Yangbajing geothermal field showed relatively stable CO2/ 3 He and R/Ra values during its evolution for more than 20 years. The helium isotope ratios were generally less than 0.2 Ra (Fig. 10). Although the degassing process may have affected and changed the δ 13 C value, CO2/ 3 He did not differ by an order of magnitude over 20 years (Fig. 12), indicating a stable deep heat source and a volatile source at Yangbajing.
We sorted out the relationships between the He and C isotopes of the typical supercritical geothermal systems worldwide in Fig. 15 respectively. Meanwhile, those for OIB were not particularly uniform around 2-20 × 10 9 and 9-30, respectively (Graham 2002). The CO2/ 3 He and R/Ra values for CLM varied from 10 9 to 10 11 (Day and Hilton, 2020) and 2-7.8, respectively (Dodson et al., 1998). Fig. 15 shows that the R/Ra vs. CO2/ 3 He plots for all the typical supercritical geothermal systems fall in the region directly related to magma. This is because, for instance, Iceland is primarily associated with OIB, Yellowstone with a mid-ocean ridge or the continental asthenosphere, Los Humeros with island arc magma, and Kenya rift with CLM. In contrast, Yangbajing has no relation with mantle-derived magma.
Further, the He-C isotope characteristics of the Yangbajing area indicate that there may be no supercritical geothermal fluids therein. Moreover, we can infer that the formation of supercritical geothermal fluids is closely related to magma chambers. Thus, the magma chamber characteristics of Yangbajing and other typical supercritical geothermal systems are further compared.

Comparison of the Yangbajing magma chamber with those of other supercritical geothermal fields
The characteristics, burial depth, and volume of magma are all crucial factors for the formation of supercritical geothermal systems. In this study, we collected the characteristics of the magma chambers of Yangbajing and the world's supercritical geothermal systems (Table 4). We then compared and analyzed the mechanisms of the nonexistence of a supercritical geothermal system at Yangbajing. The formation of supercritical geothermal systems is closely related to the tectonic background of geothermal fields. According to various geological and geophysical data, no Quaternary volcanic activity has been found in the Yangbajing area up to now (Zhao et al., 2001, Teng et al., 2019. Moreover, the type of heat source in the Yangbajing geothermal field was previously controversial (Tong andZhang, 1981, Hochstein andRegenauer-Lieb, 1998). However, as discussed in section 5.5, the heat energy generated from shearing deformation and radioactive (2015) specifically studied the geological controls for the formation of supercritical geothermal systems, and one critical parameter is permeability. This is because high permeability (>10 −15 m 2 ) is more conducive to geothermal fluid migration and rapid heat transfer. Such permeability occurs at depths less than that of the BDT zone (4-6 km) in many geological conditions (Watanabe et al., 2017, Jolie et al., 2021. Another important feature for supercritical geothermal fluid formation is that supercritical geothermal fluids occur almost within 1 km above the magma chamber. Such area results from the combined heat transfer efficiency and permeability (Scott et al., 2015), and this requires a shallow emplacement depth of the magma chamber. The depth of the BDT zone in the Yangbajing area is about 11 km (Weller et al., 2016). On the basis of the distance between the magma chamber and the BDT, the geological conditions in the Yangbajing geothermal field might be unfavorable for the formation of a supercritical geothermal system because of Yangbajing's deeply emplaced magma chamber.
The magma chamber in the Yangbajing geothermal field formed from the partial melting of sedimentary rocks in the crust, and its size might be too small for such a formation mechanism. According to the volume of the lowvelocity area beneath the Yangbajing geothermal field, the maximum single volume does not exceed 2000 km 3 (Hacker et al., 2014, Heté nyi et al., 2011. On the basis of the maximum melt ratio of 30% in southern Tibet, the current largest volume of the magma chamber is less than 600 km 3 (Chen et al., 2018). This scale is an order of magnitude different from those of other supercritical geothermal fields worldwide. By systematically comparing the volumes of the magma chambers beneath the discovered supercritical geothermal systems (Table 4), we can see that their volumes are generally greater than 1000 km 3 . Even so, those systems with volumes of only a few hundred cubic kilometers also have another characteristic, that is, their burial depths are very shallow (∼3 km) (e.g., Krafla). Therefore, the volume of the magma chamber beneath the Yangbajing geothermal field also negates the existence of supercritical geothermal fluids therein.

Conclusions
This paper presented a systematic analysis of the components and isotopes of the geothermal fluids in the Yangbajing geothermal field. Using multi-gas geothermometers, we estimated the reservoir temperatures of the Yangbajing geothermal field, and relatively acceptable results were obtained. Moreover, we determined that there are no supercritical geothermal fluids in the Yangbajing area after comparing Yangbajing's geothermal and geological characteristics with those of other supercritical geothermal systems worldwide. The specific conclusions are as follows: The main component of the geothermal gases in the Yangbajing geothermal field was CO2, with a volume percentage of more than 70%. N2 was mainly derived from mixing with gases from metamorphic degassing and the atmosphere. The multi-isotopic relationship of CO2 and He revealed that the high CO2 content of the samples is related to the thermogenic decarbonatization of marine carbonates and metamorphic sediments. He mainly originated from radioactive decay in the crust, whereas the magma-derived volatiles comprised less than 3% of the total samples.
Using the CO2 -CH4-H2 gas thermometer, soil CO2 flux geothermometer, and noble gas geothermometer, we constrained the reservoir temperatures of the Yangbajing geothermal field. We determined at least three reservoirs.
The third reservoir was the source of the parent geothermal fluid, with temperatures of ∼320 °C and a depth of ∼8 km. The second reservoir had a temperature of ∼250 ± 10 °C, whereas the first reservoir had a temperature of 150 ± 15 °C.
Comparing the geochemical (He-C isotope) and geological (magma chamber) characteristics of supercritical geothermal fields, we inferred that the Yangbajing geothermal field could not host supercritical geothermal fluids with less mantle-derived volatiles (CO2 < 5% and He < 3%) and its fairly deep-seated, small-volume granitic magma chamber.
CRediT authorship contribution statement

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.