Hydrochemical Characteristics and Formation Mechanism of Geothermal Fluids in Zuogong County, Southeastern Tibet

Abstract

Zuogong County is located in the southeast of Tibet, which is rich in hot spring geothermal resources, but its development and utilization degree are low, and the genetic mechanism of the geothermal system is not clear. Hydrogeochemical characteristics of geothermal water are of great significance in elucidating the genesis and evolution of geothermal systems, as well as the sustainable development and utilization of geothermal resources. The hydrogeochemical characteristics and genesis of the geothermal water in Zuogong County were investigated using hydrogeochemical analysis, a stable isotope (δD, δ¹⁸O) approach, and an inverse simulation model for water–rock reactions using the PHREEQC. The results indicated that the Zuogong geothermal system is a deep circulation heating type without a magmatic heat source. The chemical types present in the geothermal water from the Zuogong area are HCO₃ and HCO₃·SO₄, and the main cations are Na⁺ and Ca²⁺. The groundwater is replenished by atmospheric precipitation and glacier meltwater. The salt content of geothermal water mainly comes from the interaction between water and surrounding rocks during the deep circulation process. The reservoir temperature of geothermal water in Zuogong is 120–176 °C before mixing with non-geothermal water and drops to 62–98 °C after mixing with 58 to 79% of non-geothermal water. According to the proposed conceptual model, geothermal water mainly produces water–rock interaction with aluminosilicate minerals in the deep formation, while in shallow areas it interacts mainly with sulfate minerals. These findings contribute to a better understanding of the geothermal system in Zuogong County, Tibet.

1. Introduction

Geothermal resources play a crucial role in providing green and sustainable clean energy, particularly in the context of achieving carbon neutrality [1,2,3]. It is essential to study the hydrogeochemical characteristics and formation mechanisms of geothermal fluids to understand their chemical composition, groundwater recharge sources, and thermal circulation mechanisms. This understanding is vital for the evaluation and utilization of geothermal resources [4].

The Tibet geothermal zone is an important geothermal resource-rich area in China. In the 1970s, over 600 hydrothermal active areas were discovered in Tibet during the first comprehensive scientific expedition to the Qinghai–Tibet Plateau. These areas have great potential for the utilization of geothermal resources. Since then, many scholars have studied the geothermal resources in Tibet from various aspects, including their geological setting, hydrogeochemistry, and fluid isotopic characteristics [5,6,7,8,9,10]. Currently, there is a lack of systematic research on the medium-low temperature geothermal resources in the Tibet area. Furthermore, the understanding of the formation mechanism and heat source types of geothermal resources in this area is not deep enough. Zuogong County, located in southeastern Tibet, experiences intense geothermal activity, with over 10 medium-low temperature hot springs such as Wuya and Boko. Conducting a study on the hydrogeochemical characteristics of geothermal resources in Zuogong could provide us with a better understanding of the formation process of medium-low temperature geothermal resources in eastern Tibet.

Exploring the causal mechanisms of geothermal systems can help the rational development and utilization of geothermal resources. Hydrochemistry and isotope methods are important for studying the hydrogeochemistry of geothermal water. The characteristics of the hydrochemical components can be used to analyze the degree of water–rock interaction during the deep circulation of geothermal water and to infer the deep thermal storage temperature of geothermal water. Geothermometers obtained from the dissolution equilibrium characteristics of hydrochemical components can be widely used to evaluate the temperature of thermal reservoirs, providing a cost-effective alternative to drilling holes to measure deep thermal reservoir temperatures. The multi-mineral equilibrium method and the silica-enthalpy mixing equation compensate for the significant error of traditional geothermometers in calculating the resulting deep temperature after mixing cold water, enabling a more accurate evaluation of the thermal storage temperature of the geothermal system. Determining the source of recharge is an important part of the geothermal genesis mechanism, and the stable isotope signature of hydrogen and oxygen can be used to determine the source of recharge and the recharge height of the geothermal water to determine the recharge zone [11,12,13,14]. Therefore, it is significant to use the hydrochemical and isotopic characteristics of geothermal water samples to explain the circulation process and formation mechanism of geothermal fluids.

This study explores the recharge mechanism, water–rock interaction processes controlling the chemical characteristics of thermal waters, and thermal reservoir temperature of the geothermal system in Zuogong County. With this aim, multiple geothermal water samples were collected from the area to test their hydrochemical and isotopic (δD, δ¹⁸O) composition. Additionally, an inverse simulation model of the study area’s geothermal water was performed to quantitatively analyze the interaction between geothermal water and the surrounding rock minerals during the upwelling process. This paper delves into the genetic model of the geothermal system in Zuogong County and provides a scientific basis for evaluating, developing, and utilizing medium-low temperature geothermal resources in Tibet.

2. Geological Setting

Zuogong County is located in the northern section of the Hengduan Mountains in the eastern part of the Qinghai–Tibet Plateau and the upper reaches of the Nujiang River and Lancang River basin. Its geographical coordinates are 97°04′ to 98°42′ E and 28°26′ to 30°27′ N. The average temperature is 4.6 °C, with precipitation mostly concentrated in summer. Concentrated precipitation combined with hot seasons results in high evaporation. The average annual precipitation is 445.9 mm, while the yearly evaporation is 1680.4 mm, 3.9 times the rainfall. The landform in the area belongs to the Bershula Ridge–Taniantawong sub-region of extremely high mountains and river valleys in eastern Tibet. It describes an alpine canyon geomorphology with rivers and a mountain range NW–SE oriented. The lowest altitude in the area is 2300 m, and the highest is 5806 m. The relative height difference is about 2200 m, and the maximum is 3506 m. In a cross-sectional view, its topography describes several V-shaped troughs related to the inception of the Nujiang River, Yuqu River, and Lancang River from the southwest to the northeast. In the southwest and southeast of Zuogong County, there are glaciers on some high mountains. The regional setting is shown in Figure 1a.

The study area belongs to the Wuqi–Zuogong stratigraphic division, such as the south Qiangtang–Zuogong stratigraphic area, and mainly exposes Mesozoic strata. The exposed lithology corresponds to Upper Triassic Dongdacun Formation (T₃ddc) sandstone, Upper Triassic Jiapila Formation (T₃j), and Upper Triassic Adula Formation (T₃a) slate, with limited outcrops of bioclastic limestone, micrite, and sparite of the Upper Triassic Adula Formation (T₃a). A small number of Quaternary rocks are distributed in the gully mouths and mountain pass areas of Yuqu and its tributaries, comprising sand, gravel, and soil layers. The regional magmatic rocks are developed and roughly bounded by the Yuqu Basin. The east side is dominated by Late Triassic acidic intrusive rocks, and the west and north sides are dominated by Jurassic–Early Cretaceous acid intrusive rocks. The lithology is dominated by monzogranite and granodiorite. There are also a small number of intermediate to acidic rock dykes developed in the area, dominated by aplite, diorite porphyrite, and syenogranite.

The groundwater in Zuogong County can be categorized into three types: bedrock fissure water, karst fissure water, and loose rock pore water. Bedrock fissure water is found in fractures, fissures, and folds of layered or block rock masses. It is relatively abundant and forms springs when it encounters impermeable layers. Karst fissure water is mainly found in medium-thick layered limestone, marble, and dolomite formations. It is characterized by small karst caves developed along some cracks. Overall, cracks are the main water-bearing medium. The main water source is vertical infiltration of atmospheric precipitation. Karst fissure water is discharged as springs at the bottom of valleys and enters surface streams. Loose rock pore water is found in river valleys, plains, and tributary valleys, with aquifers ranging in thickness from several meters to tens of meters. Groundwater is plentiful in river valleys and plains, but it is relatively scarce in tributary valleys due to variations in geomorphic units and water-bearing media.

3. Materials and Methods

3.1. Hydrochemical Analysis

In 2022, we collected nine groups of geothermal water samples (four groups of thermal well water and five groups of thermal spring water) and ten groups of non-geothermal water samples (six groups of spring water and four groups of surface water) from Zuogong County (Figure 1b). The temperature, pH, and Total Dissolved Solids (TDS) values were measured in situ using a portable multi − parameter detection device. Water samples were filtered using a 0.45 μm micron membrane and stored in high-density polyethylene (HDPE) bottles, which were rinsed three times with the water samples. Then, 2 mL of dilute nitric acid was added to the samples that needed to be analyzed for metal ions to acidify the water sample (pH < 2). All bottles were sealed with sealing film to prevent leakage and evaporation. Water samples were sent to the central laboratory of the Tibet Autonomous Region Geology and Mineral Exploration and Development Bureau for complete sample analysis and testing. The cations (K⁺, Na⁺, Ca²⁺, and Mg²⁺) were analyzed by inductively coupled plasma emission spectrometry (ICP-AES). For SiO₂ analysis, the thermal water samples were diluted five times using deionized water to prevent SiO₂ in the precipitation water. Silicon and other trace elements were detected using ICP-MS. The analytical precision for major cations and trace elements is less than 0.5%. The anions (F⁻, Cl⁻, SO₄²⁻, and NO₃⁻) were measured using ion chromatography (ICS-100) with a precision of ±2%. Through the anion and cation equilibrium test, the relative error of the hydrochemistry data was within 5%, which could meet the requirement of this study. Moreover, eight groups of isotope samples of natural water (two groups of thermal well water, two groups of thermal spring water, two groups of spring water, and two groups of surface water) were collected and sent to the Ministry of Land and Resources Groundwater of the Key Laboratory of Science and Engineering in the Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences. Wavelength scanning-cavity ring-down spectroscopy was used for the analysis of hydrogen and oxygen stable isotopes. The results were reported using standard δ notation relative to Vienna Standard Mean Ocean Water (V-SMOW) [15]. The measurement accuracies were within ±0.1‰ for δD and ±1‰ for δ¹⁸O. The test results are shown in

Table 1. Hydrochemical and isotopic characteristics of geothermal sampling water in Zuogong County, Tibet.

Sample ID		Elevation	Temperature	pH	Na	K	Ca	Mg	HCO3	CO3	Cl	SO4
		m	°C		mg/L
BKJ	thermal well	3866	71	6.85	388.1	27.9	17.85	7.85	1105.00	0	29.97	BKJ
BKZK01	thermal well	3899	51	7.04	385.4	24.78	22.65	9.01	1080.00	0	26.87	BKZK01
WYDR4	thermal well	3826	71	7.41	365.8	16.92	32.68	4.13	1034.00	0	23.38	WYDR4
WYZK01	thermal well		70	7.06	357.7	16.4	33.38	5.24	1007.00	0	24.09	22.63
DRZG3	thermal spring	4091	55	7.34	596.8	37.32	74.56	12.48	1313	0	65.84	DRZG3
DRBX2	thermal spring	3359	34	7.95	20.09	1.4	63.05	18.06	277.10	0	1.99	DRBX2
DRCD1	thermal spring	3558	39	8.18	85.95	5.93	52.06	7.41	268.15	8.65	5.94	DRCD1
DBJB2	thermal spring		58	7.1	70.55	8.2	95.65	21.67	432.20	0	14.05	78.26
WSDR	thermal spring	3424	63	7.95	147.73	11.39	20.3	7.62	365.96	0	11.89	WSDR
ZGQ1	spring	3850	22	7.68	37.37	3.32	77.75	16.86	346.80	0	4.05	ZGQ1
KSQ2	spring	3013	8	8.29	3.73	0.28	22.63	1.65	68.83	5.05	0.53	KSQ2
QGQ3	spring	3094	7	8.02	1.26	0	23.55	2.6	71.11	0	0.01	QGQ3
GZQ4	spring	2697	8	8.3	0.55	0.8	23.58	0.8	71.02	5.77	0.01	GZQ4
GZQ5	spring	3818	7	8.2	0.76	1.26	26.3	0.71	75.43	5.05	0.01	GZQ5
KSQ6	spring	2932	11	8.24	2.12	0.15	44.35	10.5	112.88	2.88	0.39	KSQ6
BKQ7	spring	4058	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a	BKQ7
BKQ3	spring	3905	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a	BKQ3
QLQ	surface water	3890	8	7.8	2.75	0.84	19.36	9.97	108.40	0	0.41	QLQ
XQ	surface water	3880	8	8.14	3.19	0.55	20.96	13.18	123.99	3.59	0.16	XQ
YQH1	surface water	3818	2	8.55	5.04	0.73	38.18	12.19	144.29	10.09	0.66	YQH1
YQH2	surface water	3815	2	8.58	4.66	0.64	40.54	12.62	143.54	11.54	1.8	YQH2
Sample ID		NO3	SiO2	B	F	Li	Sr	As	TDS	δD	δ18O	Types of hydrochemistry
		mg/L						ug/L	mg/L	‰
BKJ	thermal well	1.12	66.79	7.49	7.42	1.32	0.73	n.d	1673	−160	−19.5	BKJ
BKZK01	thermal well	n.d	50.37	7.80	6.95	1.48	0.87	n.d	1598	n.a	n.a	BKZK01
WYDR4	thermal well	2.3	70.14	3.44	3.77	0.88	1.38	0.7	1439	−165	−21	WYDR4
WYZK01	thermal well	n.d	76.72	4.25	3.15	0.94	1.23	n.d	1519	n.a	n.a	WYZK01
DRZG3	thermal spring	2.42	46.89	8.41	7.37	2.96	2.15	1130	2627	−159	−19.2	DRZG3
DRBX2	thermal spring	1.45	24.15	0.17	0.99	0.06	0.4	26	415.6	−149	−19.9	DRBX2
DRCD1	thermal spring	1.09	45.39	0.93	4.78	0.29	0.39	2.7	542.3	n.a	n.a	DRCD1
DBJB2	thermal spring	n.d	62.63	n.d	4.37	0.23	0.85	n.d	758.1	n.a	n.a	DBJB2
WSDR	thermal spring	0.77	59.97	2.31	8.04	0.59	0.581	1.2	699	n.a	n.a	WSDR
ZGQ1	spring	2.04	12.81	0.2	0.35	0.10	0.59	1.6	511.9	n.a	n.a	ZGQ1
KSQ2	spring	0.74	9.88	0.0375	1.92	0.0074	0.12	12	108.4	n.a	n.a	KSQ2
QGQ3	spring	0.92	7.73	n.d	0.84	0.0065	0.21	13	116.1	n.a	n.a	QGQ3
GZQ4	spring	1.35	7.4	0.00375	n.d	0.0058	0.098	3.8	105.1	n.a	n.a	GZQ4
GZQ5	spring	1.77	8.7	0.00275	0.035	0.0064	0.1	2.5	115.2	n.a	n.a	GZQ5
KSQ6	spring	1.29	5.93	0.00725	0.48	0.016	1.13	6	223	n.a	n.a	KSQ6
BKQ7	spring	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a	−138	−17.2	BKQ7
BKQ3	spring	n.a	n.a	n.a	n.a	n.a	n.a	n.a	n.a	−147	−18.2	BKQ3
QLQ	surface water	4.08	6.45	0.0215	0.69	0.0038	0.075	n.d	142.7	−136	−17.6	QLQ
XQ	surface water	0.94	7.26	0.01625	0.095	0.0046	0.078	n.d	167.7	−134	−17.2	XQ
YQH1	surface water	1.74	7.75	0.016	0.18	0.015	0.12	1.2	223.6	n.a	n.a	YQH1
YQH2	surface water	2.52	6.5	0.0165	0.29	0.018	0.15	1.1	226	n.a	n.a	YQH2

Note: “n.d” indicates undetected, and “n.a” indicates not analyzed.

.

By analyzing the hydrogeochemical characteristics and hydrogen and oxygen isotopic characteristics in the Zuogong area, the formation mechanism of geothermal water in the Zuogong area has been summarized. The applicability of the cation thermometer, SiO₂ thermometer, mineral saturation index method, and silica-enthalpy mixing model method in calculating the thermal reservoir temperature was comprehensively analyzed in the study area. In addition, the thermal reservoir temperature and circulation depth were determined. Finally, the water–rock interaction inverse simulation model established by PHREEQC was used to quantitatively analyze the water–rock interaction of geothermal water during the deep cycle process.

3.2. Hydrogeochemical Geothermometer

The hydrogeochemical geothermometer is the most common traditional method used to evaluate the temperature utilization of thermal reservoirs. It estimates reservoir temperature based on the equilibrium reactions of multiple minerals and is calibrated empirically or experimentally. The method includes silica and cation geothermometers. The calculation formulas are shown in Table 2.

The multi-mineral equilibrium geothermometer is based on the multi-component chemical equilibrium of the geothermal system to simulate the deep reservoir temperature of geothermal fluid. It has been widely used in thermal reservoir temperature estimation [16]. The basic assumption of the multi-mineral equilibrium geothermometer is that there is a chemical equilibrium (or approximate equilibrium) between the deep geothermal fluid and the mineral combination in the reservoir. The equilibrium of multiple minerals may converge to a small temperature range, and the convergence temperature of this range is considered to be the thermal reservoir temperature.

Table 2. Calculation formulas of hydrogeochemical geothermometers [17,18,19,20,21,22].

Geothermometer	Calculation Formula	Reference
a. Chalcedony (no loss of steam)	$T={\displaystyle \frac{1032}{4.69-\mathit{log}{SiO}_2}}-273.15$	Fournier (1977)
b. Chalcedony (maximum steam loss)	$T={\displaystyle \frac{1112}{4.91-\mathit{log}{SiO}_2}}-273.15$	Arnórsson et al. (1983)
c. Quartz (no loss of steam)	$T={\displaystyle \frac{1309}{5.19-\mathit{log}{SiO}_2}}-273.15$	Fournier (1977)
d. Quartz (maximum steam loss)	$T={\displaystyle \frac{1522}{5.75-\mathit{log}{SiO}_2}}-273.15$	Fournier (1977)
e.Na-K	$T={\displaystyle \frac{1217}{log\frac{Na}{K}+1.438}}-273.15$	Fournier and Potter (1979)
f.K-Mg	$T={\displaystyle \frac{4410}{13.95-log\frac{K^2}{Mg}}}-273.15$	Giggenbach (1988)
g.Na-K-Ca	$T={\displaystyle \frac{1647}{log\frac{Na}{K}+\beta \left(log\frac{\sqrt{Ca}}{Na}+2.06\right)+2.47}}-273.15$ β = 4/3(when t < 100 °C) or β = 1/3(when t > 100 °C)	Fournier and Truesdell (1973)
h.Na-Li	$T={\displaystyle \frac{1590}{log\frac{Na}{Li}+0.779}}-273.15$	Kharaka et al. (1982)

Table 2. Calculation formulas of hydrogeochemical geothermometers [17,18,19,20,21,22].

Geothermometer	Calculation Formula	Reference
a. Chalcedony (no loss of steam)	$T={\displaystyle \frac{1032}{4.69-\mathit{log}{SiO}_2}}-273.15$	Fournier (1977)
b. Chalcedony (maximum steam loss)	$T={\displaystyle \frac{1112}{4.91-\mathit{log}{SiO}_2}}-273.15$	Arnórsson et al. (1983)
c. Quartz (no loss of steam)	$T={\displaystyle \frac{1309}{5.19-\mathit{log}{SiO}_2}}-273.15$	Fournier (1977)
d. Quartz (maximum steam loss)	$T={\displaystyle \frac{1522}{5.75-\mathit{log}{SiO}_2}}-273.15$	Fournier (1977)
e.Na-K	$T={\displaystyle \frac{1217}{log\frac{Na}{K}+1.438}}-273.15$	Fournier and Potter (1979)
f.K-Mg	$T={\displaystyle \frac{4410}{13.95-log\frac{K^2}{Mg}}}-273.15$	Giggenbach (1988)
g.Na-K-Ca	$T={\displaystyle \frac{1647}{log\frac{Na}{K}+\beta \left(log\frac{\sqrt{Ca}}{Na}+2.06\right)+2.47}}-273.15$ β = 4/3(when t < 100 °C) or β = 1/3(when t > 100 °C)	Fournier and Truesdell (1973)
h.Na-Li	$T={\displaystyle \frac{1590}{log\frac{Na}{Li}+0.779}}-273.15$	Kharaka et al. (1982)

3.3. Hydrogeochemical Modeling

According to the hydrochemical characteristics, the geothermal water is mixed with shallow non-geothermal water, and the silica-enthalpy mixing model can be used to calculate the cold water mixing ratio and the thermal reservoir temperature before cold water mixing [23]. The calculation formulas are as follows (1):

$$\left\{\begin{array}{c}{S}_{c}x+S_h\left(1-x\right)=S_s\\ Si{O}_{2c}+Si{O}_{2h}\left(1-x\right)=Si{O}_{2s}\end{array}\right.$$

(1)

where x denotes the mixing ratio of non-geothermal water; S_c and SiO_2c denote the enthalpy of surface water (J/g) and SiO₂ content (mg/L), respectively; S_h and SiO_2h denote the initial enthalpy (J/g) and initial SiO₂ content (mg/L) of geothermal water, respectively; and Ss and SiO₂s denote the enthalpy (J/g) and SiO₂ content (mg/L) of geothermal water after mixing. The surface water sample YQH1 selected for this calculation has an enthalpy of 2 J/g and a SiO₂ content of 7.75 mg/L. Above 100 °C, the relationship between temperature and saturated water enthalpy can be seen in Table 3. Below 100 °C, the saturated water enthalpy is equal to the temperature.

The temperature of geothermal water is positively correlated with the thermal circulation depth of the reservoir. Thus, the following formula can be used to calculate the circulation depth of geothermal water (2) [24]:

$$H={\displaystyle \frac{t_1-t_2}{I}}+h$$

(2)

where H denotes the thermal circulation depth, m; t₁ denotes the maximum temperature of thermal water in the deep part, °C; t₂ denotes the constant temperature zone temperature, °C; h denotes the thickness of the constant temperature zone, m; and I denote the geothermal gradient. Based on the results of previous data collection, t₂ = 5 °C, h = 30 m, and I = 3.5 °C/100 m were taken in this study [25].

Based on the elevation effect of stable isotopes of hydrogen and oxygen, the recharge altitude of groundwater is determined, and the recharge area of groundwater is identified [26,27,28]. The groundwater recharge altitude is calculated as follows (3):

$$H=(\delta s-\delta p)/k+h$$

(3)

where H denotes the altitude of the groundwater recharge area, m; h denotes the height of the sampling point, m; δs denotes the δD (δ¹⁸O) of the sampling point; δp denotes the δD (δ¹⁸O) of atmospheric precipitation near the sampling point; and k denotes the altitude gradient of atmospheric precipitation δD (δ¹⁸O). Since δ¹⁸O in water is affected by water–rock interaction, the recharge altitude was calculated using δD. δp was selected in this study to obtain the average δD of surface water (−135‰), and the k value is −2.6‰/100 m [29].

To quantitatively analyze the formation and evolution of geothermal water, the inverse simulation data block in the PHREEQC program can be used to create a water–rock interaction model. This model focuses on the deep circulation process of atmospheric precipitation in the recharge area, which forms geothermal water. By comparing predicted and modeled concentrations, this model can help in deducing and explaining the changes in water–rock interaction [30]. When building an inverse model using PHREEQC, the first step involves determining the hydrochemical components of the initial and final water samples. The second step is to consider the geological background of the study area and select minerals for the water–rock interaction. Finally, specifying uncertainty intervals is necessary, with the typical uncertainty of the inverse model being around 5%. However, since the components of the initial geothermal fluid were derived from model calculations rather than measured data, the uncertainty of the inverse simulation was increased to a maximum tolerance of 10%.

Table 3. Relations between temperature, enthalpy, and SiO₂ content.

Temperature (°C)	Enthalpy (J/g)	SiO2 (mg/L)	Temperature (°C)	Enthalpy (J/g)	SiO2 (mg/L)
50	50.0	13.5	200	203.6	265.0
75	75.0	26.6	225	230.9	365.0
100	100.1	48.0	250	259.2	486.0
125	125.1	80.0	275	289.0	614.0
150	151.0	125.0	300	321	692
175	177.0	185.0

Table 3. Relations between temperature, enthalpy, and SiO₂ content.

Temperature (°C)	Enthalpy (J/g)	SiO2 (mg/L)	Temperature (°C)	Enthalpy (J/g)	SiO2 (mg/L)
50	50.0	13.5	200	203.6	265.0
75	75.0	26.6	225	230.9	365.0
100	100.1	48.0	250	259.2	486.0
125	125.1	80.0	275	289.0	614.0
150	151.0	125.0	300	321	692
175	177.0	185.0

4. Results

4.1. Water Chemistry

The sampling temperatures obtained from geothermal and non-geothermal water ranged from 34 °C to 71 °C and 2 °C to 22 °C, respectively. In geothermal water collected from thermal wells, the temperature was higher than that of thermal spring water, with average temperatures of 66 °C and 50 °C, respectively. In non-geothermal water samples, the temperature of spring water was slightly higher than that of surface water, with average temperatures of 10 °C and 5 °C, respectively. The pH value of all underground thermal water was between 6.85 and 8.18, which is neutral to weakly alkaline water. In comparison, the pH of non-geothermal water was slightly higher than that of geothermal water, ranging from 7.68 to 8.58, which is weakly alkaline water. The TDS content of geothermal water ranged from 415.6 to 2627.0 mg/L. The TDS content of thermal well water was 1439.0 to 1673.0 mg/L (average: 1557.0) mg/L, while that of thermal spring water varied greatly, with an average lower than that of thermal well water, with a content between 415.6 and 2627.0 (1008.4) mg/L. The salinity of non-geothermal water was relatively low, with a TDS content of 105.1–511.9 mg/L, of which the spring water was between 105.1 and 511.9 (196.6) mg/L, and the surface water was between 142.7 and 226.0 (190.0) mg/L.

The concentrations of major anions in geothermal water were mainly higher than those in non-geothermal water. The HCO₃⁻ content of geothermal water ranged from 277.10 to 1313.00 mg/L (the values of thermal well water and thermal spring water ranged from 1007.00 to 1105.00 mg/L and from 259.50 to 1313.00 mg/L, respectively), and that of non-geothermal water ranged from 63.78 to 346.80 mg/L (the values of spring water and surface water ranged from 63.78 to 346.80 mg/L and from 108.40 to 134.20 mg/L, respectively). The Cl⁻ contents of thermal well water and thermal spring water ranged from 23.38 to 29.97 mg/L and from 1.99 to 65.84 mg/L, respectively. In contrast, the Cl⁻ contents of spring water and surface water ranged from 0.39 to 4.05 mg/L and from 0.16 to 1.80 mg/L, respectively. The SO₄²⁻ contents of thermal well water and thermal spring water ranged from 20.30 to 37.12 mg/L and from 35.43 to 454.30 mg/L, respectively. In comparison, the SO₄²⁻ content of non-geothermal water was between 3.55 and 52.48 mg/L. Geothermal water had a higher SO₄²⁻ content than thermal well water.

Regarding trace anionic constituents, the F⁻ contents of thermal well water and thermal spring water ranged from 3.15 to 7.42 mg/L and from 0.99 to 8.04 mg/L, respectively. The F⁻ contents of spring and surface water were lower than those of geothermal water, which ranged from 0.04 to 1.92 mg/L and from 0.10 to 0.69 mg/L, respectively. The NO₃⁻ contents of geothermal and non-geothermal water ranged from 0.77 to 2.42 mg/L and from 0.92 to 4.08 mg/L, respectively.

The cationic constituents in geothermal water were higher than those in non-geothermal water. The Na⁺ content of geothermal water was between 20.09 and 596.80 mg/L (the values of thermal well water and thermal spring water ranged from 365.80 to 388.10 mg/L and from 20.09 to 596.80 mg/L, respectively), and that of non-geothermal water was between 0.76 and 37.37 mg/L (the contents of spring water and surface water ranged from 0.55 to 37.37 mg/L and from 2.75 to 5.04 mg/L, respectively). The Mg²⁺ content of geothermal water was from 1.40 to 37.32 mg/L (average: 16.69 mg/L), higher than from 0.28 to 3.32 (0.95) mg/L of non-geothermal water. The Ca²⁺ contents of thermal well water and thermal spring water ranged from 17.85 to 33.38 (26.64) mg/L and from 20.3 to 95.65 (61.12) mg/L, respectively, and those of spring water and surface water ranged from 22.63 to 77.75 (36.36) mg/L and from 19.36 to 40.54 (29.76) mg/L, respectively. The Mg²⁺ contents of thermal well water and thermal spring water ranged from 4.13 to 9.01 (6.56) mg/L and from 7.41 to 21.67 (13.45) mg/L, respectively. The Ca²⁺ contents of spring and surface water ranged from 0.71 to 16.86 (5.52) mg/L and from 9.97 to 13.18 (11.99) mg/L, respectively.

The contents of dissolved SiO₂, Li, Sr, B, and As in geothermal water were higher than those in non-geothermal water. The contents of SiO₂, Li, Sr, B, and As in geothermal water ranged from 24.15 to 76.72 (55.89) mg/L, from 0.06 to 2.96 (0.97) mg/L, from 0.39 to 2.15 (0.95) mg/L, from 0.17 to 8.41 (4.35) mg/L, and from 0.65 to 1130.00 (230.00) µg/L, respectively. However, the contents of SiO₂, Li, Sr, B, and As in non-geothermal water were as low as from 5.93 to 12.81 (8.04) mg/L, from 0.0038 to 0.1000 (0.0184) mg/L, from 0.075 to 1.130 (0.267) mg/L, from 0.0028 to 0.2000 (0.3572) mg/L, and 1.10–1130.00 (515.00) µg/L, respectively.

4.2. Hydrogen and Oxygen Isotope Compositions

The δD and δ¹⁸O values in geothermal water were lower than those in non-geothermal water. The isotope values ranged from δD = −165 to −149‰ and from δ¹⁸O = −21.0 to −19.2‰ for geothermal water samples and from δD = −147 to −134‰ and from δ¹⁸O = −18.2 to −17.2‰ for non-geothermal water samples. Further differentiation showed that the δD and δ¹⁸O contents of thermal well water (from −165 to −160‰ and from −21 to −19.5‰) were poorer than those of thermal spring water (from −159 to −149‰ and from −19.9 to −19.2‰). The δD and δ¹⁸O contents of spring water (from −147 to −138‰ and from −18.2 to −17.2‰) were slightly lower than those of surface water (from −136 to −134‰ and from −17.6 to −17.2‰).

5. Discussion

5.1. Hydrochemical Characteristics of Thermal Groundwater

The geothermal water contained HCO₃ and HCO₃·SO₄ water, and the difference in hydrochemical types was primarily reflected in the cation content. The thermal well water had a hydrochemical type of HCO₃-Na, while thermal spring water differed slightly. Samples WSDR and DBJB2 belonged to HCO₃-Na and HCO₃-Ca·Na type water, respectively. As the amount of SO₄²⁻ in the water increased, the hydrochemical types of DRZG3, DRBX2, and DRCD3 were converted to HCO₃·SO₄-Na, HCO₃·SO₄-Ca·Mg, and HCO₃·SO₄-Na·Ca types (Figure 2). The primary cations in thermal springs change from Na⁺ to Ca²⁺, which is thought to be related to the mixing ratio of hot spring water and non-geothermal water. The greater the amount of non-geothermal water mixed in, the higher the concentration of Ca²⁺ in the thermal spring. Non-geothermal water was HCO₃ water, except for KSQ6. The hydrochemical type of spring water was mostly HCO₃-Ca type, while that of ZGQ1 and KSQ6 was HCO₃-Ca·Na and HCO₃·SO₄-Ca·Mg type, respectively. The surface water contained mostly HCO₃-Ca·Mg and HCO₃-Mg·Ca hydrochemical types.

Typically, water samples’ TDS content is mainly influenced by water–rock interactions during the water circulation. For geothermal and non-geothermal water, there was a good positive linear correlation between TDS content and sampling temperature and between HCO₃⁻, Cl⁻, Na⁺, and K⁺ contents (Figure 3). This indicates that the dissolution of siliciclastic minerals at high temperatures may be an important source of the chemical constitution of geothermal water (4,5). However, TDS had a weak positive correlation with SO₄²⁻, Ca²⁺, and Mg²⁺ contents. The SO₄²⁻, Ca²⁺, and Mg²⁺ contents in thermal well water were even lower than those in non-geothermal water. This may be caused by fluid decompression and cooling during the upwelling process of deep geothermal water. The content of thermal spring water was relatively high, which indicates that the increased contents of SO₄²⁻, Ca²⁺, and Mg²⁺ may be controlled by the dissolution of sulfate minerals such as gypsum in the shallow parts of the earth’s surface. From large to small, the TDS contents of geothermal and non-geothermal water were thermal well water, spring water, and non-geothermal water. This indicates that deep thermal water may have a mixed effect with shallow non-geothermal water during the upwelling process.

$$4NaAl{Si}_3{O}_8+4C{O}_2+6H_{2}O\to 4{Na}^{+}+{Al}_2{Si}_2{O}_5{\left(OH\right)}_4+4{HCO}_3^{-}+8{SiO}_2$$

(4)

$$2KAl{Si}_3{O}_8+2C{O}_2+3H_{2}O\to 2K^{+}+{Al}_2{Si}_2{O}_5{\left(OH\right)}_4+2{HCO}_3^{-}+4{SiO}_2$$

(5)

There was a positive correlation between the TDS content in non-geothermal and geothermal water and the typical geothermal components of SiO₂, F, Sr, Li, and B (Figure 4). This indicates that the contents of the characteristic components such as SiO₂, F, Sr, Li, and B in geothermal water are mainly controlled by water–rock interactions such as mineral dissolution. The R² between Li and B contents and TDS reached more than 0.9, indicating that Li and B have the same source between geothermal water and non-geothermal water. Water–rock interactions during deep circulation increase Li and B concentrations in geothermal water.

The geothermal system in the Tibet zone can be classified into three genesis types: (1) magma reservoir-type high-temperature geothermal system undergoing cooling in the crust; (2) locally molten geothermal system within the crust in collision zones; and (3) deep circulation heating geothermal system [31,32,33,34]. Geothermal fluids lacking magma heat sources are created by deep heating of groundwater through high heat flow circulation. Wang et al. (2023) relied on the CRUST1.0 model and a large amount of geological data to analyze the distribution characteristics of the crustal heat generation rate in China’s continental areas [35]. The Qinghai–Tibet Plateau was found to have the highest crustal heat flow value (crustal heat flow: 55–75 mW·m⁻²). This conclusion provides a basis for using a heat source for deep circulation heating geothermal systems. High-temperature geothermal systems such as Yangbajing, Yangyi, Gudui, and Gulu in central and southern Tibet have all been confirmed to have heat sources related to local melting in the crust. Moreover, the chlorine enthalpy model and silicon enthalpy diagram were used to determine the existence of parent geothermal fluid beneath the geothermal field.

To clarify whether the geothermal water in Zuogong is affected by molten magma, this study used the anion ternary diagram (Cl–SO₄–HCO₃) to compare the contents of four characteristic elements (Li, B, F, and As) in geothermal water possibly derived from magma with hydrochemical constituents of Tengchong Rehai (RH) in Yunnan, Yangbajing (YBJ) in Tibet, and Yellowstone National Park (YSNP). These locations are known to have clear magma heat sources [36,37,38]. As shown in Figure 5, the geothermal water in the Zuogong area is bicarbonate geothermal water, and water samples are all distributed in the peripheral cycle water area. In terms of the characteristics of the anionic components, Zuogong geothermal water is not affected by the molten magma.

For geothermal water with magmatic heat sources, the four characteristic elements, Li, B, F, and As, may come from volatile substances in the cooling process of deep magma [39,40]. Therefore, Li, B, F, and As elements were simultaneously enriched in geothermal fluid with magmatic heat sources. The Li, B, F, and As element contents of the geothermal water in Yangbajing and Rehai were compared with those of the Zuogong geothermal water (Figure 6). The results showed that the Li, B, F, and As concentrations of the Zuogong area’s geothermal water substantially differed from those in Yangbajing and Rehai with magmatic heat sources. The Li and As contents even reached an order of magnitude difference. The difference could not be attributed to the different degrees of water–rock interaction but was caused by differences in heat sources. The above research shows no magmatic heat source in the geothermal system of the Zuogong area.

The high levels of Na⁺, K⁺, HCO₃⁻, Cl⁻, SiO₂, and TDS in both geothermal and non-geothermal waters suggest that the interaction between water and silicate minerals deep within the earth is responsible for the salts found in geothermal water. Additionally, the strong correlation between TDS and B and Li fractions in the water samples suggests that the interaction between water and B- and Li-bearing minerals during deep circulation is the primary source of B and Li fractions in geothermal water. The presence of SO₄²⁻ in the hot well water and hot spring water suggests that sulfate minerals near the earth’s surface contribute to the water’s SO₄²⁻ content. Furthermore, the presence of characteristic ions (F, Li, B, and As) in the geothermal water, along with the Cl-SO₄-HCO₃ diagram, indicates that there was no mixing of magmatic fluids during the formation of the geothermal water.

5.2. Estimation of Circulation Temperature and Depth for Thermal Groundwater

The results of calculating the thermal reservoir temperature using a geochemical thermometer are shown in

Table 4. Calculated temperatures using selected chemical geothermometers and the estimated reservoir depth.

Sample	Measured Temperature	a	b	c	d	e	f	g	h	MEE	SEMM	Thermal Circulation Depth
	°C											After Mixing (m)	Before Mixing (m)
BKJ	71	87	90	116	115	190	96	179	216	98	158	2687	4401
BKZK01	51	72	77	102	103	182	91	171	225	90	160	2459	4459
WYDR4	71	90	92	118	117	159	91	150	195	88	165	2401	4601
WYZK01	70	95	96	123	121	158	87	150	200	92	163	2516	4544
DRZG3	55	69	74	99	100	180	97	167	242	89	146	2430	4059
DRBX2	34	39	49	71	75	188	23	121	208	63	120	1687	3316
DRCD1	39	67	73	97	99	187	59	143	216	62	176	1659	4916
DBJB2	58	84	87	113	112	230	55	160	214	84	171	2287	4773
WSDR	63	81	85	110	110	196	74	165	227	66	159	1773	4430

Note: MEE: multicomponent mineral equilibrium; SEMM: silica-enthalpy mixing model.

. The thermometer calculation results of chalcedony with no steam loss and maximum steam loss ranged from 39 to 95 °C and from 49 to 96 °C, respectively. The thermal reservoir temperatures calculated by quartz with no steam loss and quartz with the maximum steam loss thermometers ranged from 71 to 123 °C and from 75 to 121 °C, respectively.

Generally speaking, the higher the SiO₂ concentration in the water sample, the higher the thermal reservoir temperature. This is because the solubility of SiO₂ changes with temperature. Since the equilibrium concentration controlled by different silica minerals changes differently with temperature, it is necessary to select appropriate silica minerals when selecting a silica geothermometer [41]. The sampling temperatures of geothermal water were all lower than the local boiling point; thus, the maximum steam loss thermometer was unsuitable for this area. The calculated results of the chalcedony and quartz thermometers with no steam loss were both higher than the sampling temperature (Figure 7). It indicates that the results obtained by the silica thermometer may represent the thermal reservoir temperature in the Zuogong area, and the calculated results of the chalcedony thermometer with no steam loss were in good agreement with the temperature estimated by the multi-component mineral equilibrium simulated using the PHREEQC Interactive RC1 software (see below).

The calculation results of the cation thermometer are as follows: The thermal reservoir temperature obtained by the Na-K thermometer was between 158 and 230 °C, the calculated result by the K-Mg thermometer was between 23 and 97 °C, the temperature range obtained by the Na-Li thermometer was between 195 and 242 °C, and the thermal reservoir temperature calculated using the Na-K-Ca thermometer was between 121 and 179 °C.

The results of Na-K and Na-Li thermometers generally yield higher values, which inaccurately characterize the thermal reservoir temperature in the region. The Na-K thermometer is used for the equilibrium control between geothermal water and alkaline feldspar in high-temperature geothermal systems, and it is suitable for thermal water above >150 °C [42]. When the Na-K thermometer is used in medium- and low-temperature geothermal systems, the calculated thermal reservoir temperature will be higher due to the increase in Ca²⁺ and SO₄²⁻ contents [43,44]. The calculation results obtained by the Na-Li thermometer were similar to those of the Na-K thermometer. This is because the Na-Li thermometer is also suitable for characterizing the temperature of deep thermal reservoirs and controlling minerals in medium- and low-temperature geothermal systems not to reach equilibrium. The K-Mg thermometer is controlled by the mineral equilibrium between muscovite, K-feldspar, and chlorite, and it is easier to reach equilibrium again than the Na-K thermometer. Therefore, the K-Mg thermometer can represent the thermal reservoir characteristics after mixing with shallow, cold water. However, the calculated results of the geothermal water in the study area using the K-Mg thermometer were lower than the sampling temperature, indicating that the K- and Mg-controlled minerals in the water samples had not reached equilibrium. Therefore, the K-Mg thermometer is not suitable for this analysis. The Na-K-Ca thermometer is a correction to the Na-K thermometer and is suitable for geothermal systems with lower temperatures, especially in geothermal water rich in Ca²⁺. In this study, the thermal reservoir temperature calculated by the Na-K-Ca thermometer was lower than that of the Na-K and Na-Li thermometers and higher than that of the silica thermometer. It is inferred that the thermal reservoir temperature calculated by the Na-K-Ca thermometer is relatively reliable.

Multicomponent mineral equilibrium temperatures were inferred for the four thermal well samples and five thermal spring samples with the highest measured temperatures using the PHREEQC geochemical code and the lnll thermodynamic database. Based on this field survey and previous research results, minerals that may exist in the reservoir and be used in modeling were selected [45], including calcite, chrysotile, quartz, muscovite, kaolinite, chalcedony, sepiolite, and diopside. Al data in geothermal water are often missing, and the FixAl method needs to be used to force the solution to maintain equilibrium with Al-containing minerals. Microcline is commonly used in forced equilibrium systems with pH values ranging from 5.5 to greater than 9 [46].

The results of the geothermal thermometer simulation of multiple mineral equilibrium are shown in Figure 8. The predicted thermal reservoir temperature is between 62 and 98 °C. The predicted results are similar to the results of the chalcedony thermometer with no steam loss, proving that the results obtained by this method can represent the thermal reservoir temperature in the Zuogong area. The analysis showed that the temperature of the thermal well water reservoir was 88–98 °C, indicating that the deep thermal reservoir and the degree of water–rock interaction are similar. The simulation result of thermal spring water was 62–89 °C, inferring the degree to which the thermal spring is mixed with shallow non-geothermal water while rising to the shallow part.

According to the silica-enthalpy mixing model, the mixing ratios of thermal well water, thermal spring water, and non-geothermal water were calculated to be from 58 to 69% (62%) and from 61 to 79% (68%), respectively. The thermal reservoir temperatures before cold water mixing ranged from 159 to 165 °C and from 120 to 176 °C, respectively (Figure 9). The thermal reservoir temperature before the mixing of geothermal water and non-geothermal water was similar to the reservoir temperature obtained by the Na-K-Ca thermometer, indicating that the calculation results of the Na-K-Ca thermometer can be used to characterize the initial reservoir temperature of geothermal fluid in Zuogong County. It is generally believed that the initial temperature of geothermal water before non-geothermal water mixing is between 120 and 176 °C. The mixing ratio of thermal spring water was higher than that of thermal well water. This is because the mixing of cold water is a dynamic process, and the ratio of cold water gradually increases as the geothermal water rises to the shallow part.

According to the calculation results of thermal reservoir temperature, it can be seen that there are two thermal reservoir temperatures during the upwelling process of geothermal water. One is the thermal reservoir temperature after experiencing the mixing of non-geothermal water, which is obtained by the multimineral saturation index method (from 62 to 98 °C). The second is the thermal reservoir temperature before non-geothermal water mixing, obtained by the silica-enthalpy mixing model (from 120 to 176 °C). Since geothermal water inevitably accepts the mixing of non-geothermal water during the circulation process, this study explained the two circulation depths before and after the mixing of non-geothermal water when calculating the geothermal water circulation depth (Table 3). The thermal reservoir cycle depth before mixing is supposed to represent the initial depth of geothermal fluid [47,48]. According to the calculation results of cycle depth (Equation (2)), the thermal reservoir circulation depth of geothermal fluid in Zuogong County was between 1659 and 2687 m after experiencing non-geothermal water mixing from 58 to 79%, while that before non-geothermal water mixing was between 3316 and 4916 m.

5.3. Recharge Source of Geothermal Water

The stable isotope characteristics of hydrogen and oxygen can help determine the source of geothermal water and the scope of the recharge area. Figure 10 shows that both geothermal and non-geothermal waters in Zuogong County are located near the global meteoric water line (GMWL: δD = 8δ¹⁸O + 10) and below the local meteoric water line (i.e., the eastern Tibetan Plateau atmospheric precipitation line: LMWL: δD = 8.2δ¹⁸O + 19). Non-geothermal water samples were plotted in the straight line between snow-melted water and magmatic water, while geothermal waters reveal an oxygen drift phenomenon compared to snow-melted water. The results showed that the recharge sources of groundwater are atmospheric precipitation and mountain glacier meltwater. During the deep circulation of geothermal water, there is an oxygen isotope exchange between the geothermal water and the surrounding rock, leading to the enrichment of δ¹⁸O in the geothermal water.

The isotope elevation effect can be used to calculate the recharge elevation of geothermal water, thus determining the source of geothermal water recharge. Oxygen isotope values are not representative of geothermal water because they are affected by water–rock interaction. On the other hand, stable deuterium isotopes rarely undergo isotope exchange with surrounding aquifer materials, making them the ideal isotope for estimating water recharge elevation. Therefore, deuterium isotopes were used in this study to determine recharge elevations.

Table 5. The calculation results of the thermal circulation depth of Zuogong County, Tibet.

Sample	Type	Elevation (m)	δD (‰)	Recharge Altitude (m)
BKJ	thermal well	3866	−160	4828
WYDR4	thermal well	3826	−165	4980
DRZG3	thermal spring	4091	−159	5014
DRBX2	thermal spring	3359	−149	3897

shows the recharge altitude results calculated using the elevation effect of stable hydrogen and oxygen isotopes in the study area (Equation (3)). The recharge altitude of thermal well water and thermal spring water ranged from 3897 to 5014 m, and the average was 4680 m. Based on the recharge altitude of geothermal water and the geological condition of the study area, it is inferred that the groundwater recharge area in the study area is the high mountain area of Zuogong County.

This study analyzed the degree of water–rock interaction during geothermal water circulation. The δ¹⁸O and δD were used to determine the degree of water–rock interaction of geothermal water, which is explained according to the δD excessive parameter d (d = δD − 8δ¹⁸O) [49,50,51,52]. Generally speaking, the smaller the d value, the slower the groundwater runoff speed, the more closed the geological environment, the weaker the renewable capacity of geothermal water, and the deeper the degree of water–rock interaction. In this study, the d value of geothermal water was between −5.4 and 10.2 (0.95), while that of non-geothermal water was between −1.4 and 4.8 (1.65) (Figure 10). The average values of geothermal and non-geothermal water were greater than 0, indicating that the geothermal water in the study area has not experienced long-term water–rock interaction during the deep circulation process.

The Na-K-Mg diagram can be used to determine the equilibrium of geothermal water (Figure 11). It can be seen in Figure 11 that all thermal well water and thermal spring water in the study area are located in the immature water area, which indicates that it may be that the geothermal fluid did not reach the complete equilibrium with surrounding minerals during the circulation process due to a small degree of water–rock interaction and the mixing and dilution of non-geothermal water. This indicates that the geothermal fluid is in a relatively open environment. Combined with the excessive parameter d value characteristics of hydrogen and oxygen stable isotopes, it is comprehensively shown that the geothermal water in the study area is in a relatively open thermal reservoir environment during the deep circulation process. The geothermal water body has a short runoff path, resulting in a faster renewal rate and shallower water–rock interaction. However, due to the mixing effects, we are unable to use the Na-K-Mg diagram for calculating the thermal storage temperatures.

5.4. Water–Rock Interaction Modeling

When using PHREEQC software to establish an inverse simulation model of geothermal water–rock interaction, it is essential to select the flow path, possible reactant minerals, and the hydrochemical constituents of the initial and final water sample [53]. This study collected two types of geothermal water: thermal well water and thermal spring water. According to previous studies, thermal spring water will also produce water–rock interaction with surrounding rocks while rising to shallow depths. Therefore, the endpoint of this modeling was selected to be one thermal well water (BKZK01) and one thermal spring water (WSDR), and the starting point sample was the surface water (XQ) collected this time. Since both geothermal and thermal spring water have varying degrees of mixing of non-geothermal water, the binary mixing model was used to correct the impact of mixing. The calculation method is as follows (6):

$$Co{n}_{mix}=Co{n}_D(1-f_c)+Co{n}_C{f}_c$$

(6)

where Con_mix denotes the chemical constituents after thermal water mixing with cold water; Con_D and Con_C denote the chemical constituents of geothermal water and non-geothermal water before mixing, respectively; and fc denotes the mixing ratio of non-geothermal water (BKZK01:69%; WSDR:61%). The non-geothermal water constituent is the surface water sample (XQ), and the corrected geothermal hydrochemical constituent results are shown in

Table 6. The concentrations of the major constituents of the initial and final members (mg/L).

Flow Path	Sample	Na	K	Ca	Mg	HCO3 + CO3	Cl	SO4	Si	F	Sr
Initial	XQ	2.75	0.84	19.36	9.97	108.40	0.41	5.69	6.45	0.095	0.078
Final	BKZK01	1237.10	78.07	29.97	6.87	3242.59	85.76	52.82	61.49	22.21	2.633
Initial	XQ	2.75	0.84	19.36	9.97	108.40	0.41	5.69	6.45	0.095	0.078
Final	WSDR	374.49	27.89	21.77	3.95	768.81	29.85	173.13	61.66	20.47	1.368

.

According to the geological conditions and rock mineral constitution of Zuogong County, the reactant minerals of this inverse simulation model were determined to be albite, K-feldspar, clinochlore, kaolinite, muscovite, quartz, calcite, gypsum, halite, celestite, and fluorite. In addition, CO₂ (g) was added to balance the dissolution equilibrium of carbonate minerals in geothermal water. The uncertainty of the model was set to 0.06 and 0.05. Based on the rock mineral constitution characteristics and the reaction characteristics of aluminosilicate minerals, an appropriate reaction path was selected from the model’s simulation results. The quantitative simulation results of the interaction between geothermal water samples in the study area are shown in

Table 7. Water–rock interaction models along the flow path from surface fluids to deep geothermal fluids in the Zuogong area. Mole transfer in mmol/L; positive numbers indicate dissolution, and negative numbers indicate precipitation.

Flow Path	Mode Reactant	Mole Transfer	Flow Path	Mode Reactant	Mole Transfer
Thermal well (XQ-BKZK01)	Albite	4.73 × 10−2	Thermal spring (XQ-WSDR)	Albite	1.51 × 10−2
	Calcite	−3.88 × 10−3		Calcite	−2.18 × 10−3
	Celestite	2.22 × 10−5		Celestite	1.48 × 10−5
	Clinochlore	−6.78 × 10−4		Clinochlore	−4.94 × 10−5
	Fluorite	5.65 × 10−4		Fluorite	5.37 × 10−4
	Halite	2.47 × 10−3		Gypsum	1.73 × 10−3
	Kaolinite	−4.42 × 102		Halite	8.32 × 10−4
	K-Feldspar	−4.42 × 102		Kaolinite	5.90 × 10−2
	Muscovite	4.42 × 102		K-Feldspar	6.75 × 10−2
	Quartz	8.83 × 102		Muscovite	−6.68 × 10−2
	CO2(g)	4.43 × 10−2		Quartz	−1.64 × 10−1
				CO2(g)	1.30 × 10−2

.

The results of the inverse simulation model showed that during the runoff of thermal well water and thermal spring water, albite, fluorite, celestite, and halite were all dissolved, while calcite and clinochlore were precipitated. Compared with thermal spring water exposed on the surface, aluminosilicate minerals in geothermal water are primarily saturated during deep runoff.

The amount of calcite precipitation and albite dissolution in thermal well water was higher than in thermal spring water, showing that the Na⁺ concentration in thermal well water was much higher than the Ca²⁺ concentration. Thermal spring water mixed with non-geothermal water during the upwelling process caused some calcite to be redissolved into the water; thus, the Ca²⁺ concentration in thermal spring water was higher than that in thermal well water. In addition, the gypsum dissolution process occurs in thermal spring water, and the gypsum minerals in thermal well water do not participate in this reaction. This shows that the SO₄²⁻ content in thermal spring water originates from the dissolution of sulfate minerals in the shallow parts of the earth’s surface.

Quartz dissolved in thermal well water but precipitated in thermal spring water. It indicates that the cooling of thermal spring water during the upwelling process and the mixing of non-geothermal water lead to the re-precipitation of quartz. Celestite and fluorite were dissolved in thermal well water and spring water, indicating that the fluorine and strontium in geothermal water come from the dissolution of fluorite, celestite, and other minerals rich in fluorine and strontium.

5.5. Geothermal Conceptual Model

Based on the above analysis, atmospheric precipitation and glacial meltwater in the Zuogong area enter the deep stratum along the deep fault structure, receive heating from high-temperature surrounding rocks during the deep circulation process, and form thermal water of 120 to 176 °C in the deep layer. During the runoff process, deep thermal water experiences 58 to 79% non-geothermal water mixing during upwelling along different secondary structural fracture zones, forming a shallow thermal reservoir at 62 to 98 °C near or exposed to the surface. Thermal springs are formed in the valley stage area. The water–rock interaction with surrounding rocks during the runoff process is the main source of salt for geothermal water, creating deep geothermal water of HCO₃-Na. During the process of thermal springs exposed on the surface, it dissolves and leaches with shallow sulfate minerals to form HCO₃-Na., HCO₃·SO₄-Na·Ca, and HCO₃·SO₄-Ca·Mg geothermal water. The presumptive geothermal genetic model is shown in Figure 12.

6. Conclusions

By analyzing the chemical constituents of 19 groups of water samples in Zuogong County, the hydrochemistry of thermal well water in geothermal water is all HCO₃-Na type, and the thermal spring water is mainly HCO₃ and HCO₃·SO₄ water. Non-geothermal water is all HCO₃ water. According to the correlation between the TDS water sample and water chemical elements, the analysis of the anion ternary diagram and geothermal characteristic ion content (F, Li, B, and As) shows that the geothermal water in Zuogong County has no magmatic heat source, and the salt in the water sample comes from water–rock interaction. Near the atmospheric precipitation line of geothermal water hydrogen and oxygen stable isotopes, the calculated δD recharge altitude is 3897–5014 m. Analysis shows that geothermal water recharge sources are atmospheric precipitation and glacier meltwater from the surrounding mountainous areas.

Cation thermometer, silica thermometer, multicomponent equilibrium method, and silica-enthalpy mixing model have been used to comprehensively calculate the thermal reservoir temperature in the Zuogong area. Comprehensive analysis shows that geothermal resources have two sets of reservoir temperatures before and after cold water mixing: the thermal reservoir temperature after shallow cold water mixing (62–98 °C) and the thermal reservoir temperature before deep cold water mixing (120–176 °C). On this basis, the shallow circulation depth of the thermal reservoir is calculated to be between 1659 and 2687 m, and the deep circulation depth is between 3316 and 4916 m.

The geothermal water samples are all distributed in the immature water area in the Na-K-Mg diagram, indicating that the geothermal water is in a relatively open thermal reservoir environment with a shorter runoff path and a faster renewal rate. Inverse simulation models of water–rock interaction have been established for thermal well water and spring water, respectively. The results show that the water–rock interaction in deep geothermal water is dominated by mineral dissolution and precipitation of aluminosilicates. During the upwelling process of geothermal water, quartz precipitates from the water due to the mixing of non-geothermal water and the cooling effects, while carbonate minerals dissolve again into the water, causing the Ca²⁺ content in shallow water to increase. The water–rock interaction of sulfate minerals only exists in shallow areas. The dissolution of minerals such as fluorite and celestite is the source of characteristic ions in geothermal water.