Hydrogeochemical Characteristics and Genetic Mechanism of the Shiqian Hot Spring Group in Southwestern China: A Study Based on Water–Rock Interaction

Abstract

Shiqian County, located within a key geothermal fluids belt in Guizhou Province, China, has abundant underground hot water resources. Therefore, elucidating the hydrogeochemical characteristics and formation mechanisms of thermal mineral water in this area is essential for evaluating and sustainably utilizing regional geothermal fluids. This study focuses on the Shiqian Hot Spring Group and employs integrated analytical techniques, including rock geochemistry, hydrogeochemistry, isotope hydrology, digital elevation model (DEM) data analysis, remote sensing interpretation, geological surveys, mineral saturation index calculations, and PHREEQC-based inverse hydrogeochemical modeling, to elucidate its hydrogeochemical characteristics and formation mechanisms. The results show that strontium concentrations range from 0.06 to 7.17 mg/L (average 1.65 mg/L) and metasilicic acid concentrations range from 19.46 to 65.51 mg/L (average 33.64 mg/L). Most samples meet the national standards for natural mineral water and are classified as Sr-metasilicic acid type. Isotope analysis indicates that the geothermal water is recharged by meteoric precipitation at elevations between 911 m and 1833 m, mainly from carbonate outcrops and fracture zones on the southwestern slope of Fanjingshan, and discharges south of Shiqian County. The dominant hydrochemical types are HCO₃·SO₄-Ca·Mg and HCO₃-Ca·Mg. Strontium is primarily derived from carbonate rocks and celestite-bearing evaporites, whereas metasilicic acid mainly originates from quartz dissolution along the upstream groundwater flow path. PHREEQC-based inverse modeling indicates that, during localized thermal mineral water runoff in the middle-lower reaches or discharge areas, calcite dissolves while dolomite and quartz tend to precipitate, reflecting calcite dissolution-dominated water–rock interactions and near-saturation conditions for some minerals at late runoff stages.

1. Introduction

Geothermal resources, including geothermal energy, geothermal fluids, and the extractable chemical constituents they contain, represent an important category of renewable and clean energy that can be effectively utilized under appropriate geological and techno-economic conditions [1,2,3]. Globally, geothermal energy has been extensively applied in electricity generation, district heating, medical therapy, and balneotherapy. Countries such as Iceland, New Zealand, Turkey, Italy, and Japan have gradually formed mature and systematic geothermal development frameworks through decades of technological advancement and management optimization [4,5,6]. In comparison, China began its geothermal development relatively late, yet the sector has expanded rapidly in recent years, driven by national energy-transition policies and increasing demand for low-carbon heating solutions [7,8]. Meanwhile, international research increasingly emphasizes the roles of geological structures, hydrothermal circulation, and water–rock interactions in shaping geothermal systems, thereby providing a scientific foundation for resource evaluation and sustainable utilization. In addition to energy and industrial applications, geothermal resources, particularly thermal mineral waters, possess significant wellness and health-promoting potential. The unique chemical composition, temperature, and flow characteristics of these waters have been linked to therapeutic effects, rehabilitation, and preventive health practices, forming the basis of spa tourism and balneotherapy industries worldwide [9,10]. Recognizing and leveraging the health and wellness value of geothermal systems not only expands their socio-economic benefits but also provides opportunities for sustainable regional development, public health improvement, and integration with cultural and ecological tourism initiatives. Within this global context, strengthening geothermal research in China, particularly investigations of thermal mineral water systems, is essential for improving understanding of regional geothermal processes and will further promote China’s deeper engagement in international geothermal research and collaborative practices.

Thermal mineral water, as the surface manifestation of geothermal resources [11], is a natural resource integrating heat, minerals, and water, formed through prolonged geological processes [12,13]. Its primary utilization forms include natural hot springs and artificially drilled geothermal wells [14]. The genesis of thermal mineral water is mainly governed by recharge sources, geological structures, and water–rock interactions [15,16,17], with meteoric precipitation serving as the principal recharge source for most geothermal fluids [18,19]. Fault structures exert a decisive influence on the occurrence and formation of thermal mineral water: on one hand, they act as conduits for transferring heat from deep to shallow zones, creating localized high-temperature anomalies [20,21,22]; on the other hand, they provide the main ascending channels for thermal fluids, where the convergence of heat and water drives the circulation of the thermal system [23,24]. Water–rock interaction is a fundamental process regulating the chemical composition of groundwater, as hydrochemical characteristics are directly influenced by the mineral assemblages of the geological formations through which the water flows. The hydrochemistry of geothermal fluids is thus decisively shaped by water–rock interactions, which are jointly controlled by tectonic activity, lithological composition, groundwater dynamics, climatic conditions, and pressure [25,26]. The relative significance of these factors varies across regions. For example, in the Fenton Hill geothermal system (New Mexico, USA), elevated temperature and pressure markedly enhance water–rock reaction rates, leading to increased concentrations of dissolved constituents in geothermal fluids [27]. In northwestern Sichuan, China, lithological variations dictate the patterns of water–rock interaction, resulting in differences in the dominant ionic composition of thermal waters [26]. Experimental studies further illustrate these processes: Satake et al. [28] demonstrated that, under high-temperature (~250 °C) and elevated pressure conditions, water–rock interaction involving CO₂-rich fluids fosters intense secondary mineral precipitation. At near-neutral pH (~7.3), the authors observed rapid formation of smectite and other clays, illustrating how pH modulates mineral precipitation and thereby influences the continuous release (or sequestration) of dissolved geochemical species. Recent reviews indicate that CO₂ dissolution into formation waters produces acidic conditions that, under elevated temperature and pressure, substantially accelerate water–rock reactions and the release of divalent cations (e.g., Ca²⁺, Mg²⁺), whereas subsequent evolution toward higher alkalinity promotes carbonate and secondary silicate mineral precipitation, thereby modulating the continued release—or sequestration—of dissolved species [29]. These observations indicate that the controlling factors of water–rock interactions differ significantly among regions, with their effects shaped by local geological structures, lithology, and geothermal circulation patterns. Consequently, evaluations of thermal mineral water systems must account for site-specific geological, hydrogeological, and hydrochemical conditions to accurately reveal their formation and evolution mechanisms.

Under hydrothermal conditions, the leaching and dissolution of surrounding rocks by underground thermal water are significantly enhanced. This process facilitates the release and migration of trace elements from host rocks and promotes their subsequent enrichment in thermal spring waters, thereby conferring potential for development as natural mineral water or therapeutic thermal mineral water resources [30,31]. For instance, investigations of thermal and mineral springs in different countries have revealed substantial variations in trace-element concentrations, reflecting differences in lithology and geotectonic settings. In Greece, analysis of 276 springs documented wide-ranging concentrations of Mn, Fe, Co, Ni, Cu, Zn, Se, and I, highlighting the strong influence of local geological and tectonic conditions on mineral composition [32]. In a case study of the Aab-E-Shifa hot spring in Pakistan, the zinc concentration in the spring water reaches a relatively high level (6.73 mg/L), and the hot spring water is considered to possess a certain therapeutic potential [33]. Owing to their rich ionic compositions, hot spring waters have long been used for health and wellness purposes and are regarded as important natural resources for the rehabilitation and health-care industry. Therefore, investigations of the hydrochemical compositions of hot spring waters in different regions provide valuable guidance for the development of the spa and wellness industry and for optimizing the sustainable utilization of these resources.

Understanding the sources of dissolved ions and trace elements and their flow pathways is of great significance for elucidating the hydrogeochemical evolution of geothermal fluids and their resource characteristics [34]. Globally, a variety of methods have been widely applied to trace ion sources and to analyze water–rock interaction mechanisms in geothermal resource studies [35]. Hydrogeochemical analysis is commonly used to identify major water types, characterize ionic features, and infer water–rock interactions, and has been extensively applied in hot spring studies in other countries [36]. Stable isotope tracing, including δ²H, δ¹⁸O, δ³H, carbon isotopes, and noble gases, can provide valuable information on groundwater recharge sources, fluid mixing, and circulation processes [37,38]. Mineral saturation indices and multicomponent geothermometers are effective for identifying ongoing mineral dissolution and precipitation trends, as well as for estimating reservoir temperatures. These methods have been broadly validated in European geothermal areas [39]. However, each of these individual approaches has certain limitations. Hydrogeochemical analysis can reveal ionic distributions and correlations, but it is challenging to quantify ongoing water–rock reactions along complex flow paths [40]. Isotope tracing may fail to clarify sources and mixing ratios when signals overlap or representative end-members are lacking. Although mineral saturation indices can indicate chemical equilibrium, they cannot independently quantify the contributions and dynamic changes in multiple minerals [39,41]. To address these limitations, the integration of PHREEQC inverse hydrogeochemical modeling with hydrochemical analysis and isotope tracing enables the quantitative interpretation of ion sources and mineral dissolution processes [40,42]. This approach can systematically elucidate water–rock interaction mechanisms, clarify the enrichment processes of minerals and trace elements in thermal waters, and provide a scientific and quantitative basis for understanding the formation mechanisms of thermal mineral water and the controlling factors of its characteristic ionic composition, thereby supporting resource assessment and sustainable development.

Building on the global and regional understanding of thermal mineral waters and their characteristic ionic compositions, this study focuses on Shiqian County, located in the northeastern part of Guizhou Province and the central region of Tongren City, China. Shiqian represents one of the principal distribution areas of low-to medium-temperature geothermal fluids in Guizhou Province, characterized by abundant thermal mineral water resources, high water temperatures, and large discharge rates. In addition, the waters contain a variety of trace elements whose concentrations reach or exceed the threshold values specified in relevant natural mineral water standards [43]. The thermal waters are primarily concentrated at the intersection of the NNE-trending Shiqian Fault and the NE-trending Hongshi strike-slip fault system, and are mainly hosted in the Cambrian Middle-Upper and Ordovician Lower carbonate strata [44,45]. As water flows through these formations, it undergoes leaching and dissolution processes that enrich the fluids with minerals, forming thermal mineral waters with specific chemical compositions. Previous studies in Shiqian have primarily focused on the geological background, hydrochemical compositions, thermal water types, or the formation mechanisms of individual springs. However, systematic and comprehensive studies on the hydrogeochemical characteristics and formation mechanisms of the entire Shiqian hot spring group are still limited. To address these gaps, this study employs a multi-dimensional approach integrating rock geochemistry, hydrogeochemistry, isotope hydrology, DEM analysis, remote sensing interpretation, geological surveying, mineral saturation indices, and inverse hydrogeochemical modeling. The objectives are to elucidate the hydrogeochemical characteristics and formation mechanisms of the Shiqian hot spring group, and to quantify water–rock interactions and the source contributions of dissolved constituents. The findings provide scientific guidance for exploration, sustainable utilization, and management of geothermal resources in the region and offer a foundation for promoting local tourism development and the rational exploitation of thermal mineral waters.

2. Overview of the Study Area

Shiqian County, located in the southwest of Tongren City, Guizhou Province, is characterized by a warm climate year-round and abundant rainfall, with an average annual precipitation of 1113.9 mm. Thermal mineral water constitutes the primary mineral resource in the area. The terrain exhibits a “high in the northeast, low in the southwest” pattern, with a maximum elevation of 1836 m, a minimum elevation of 370 m, and a relative relief of 1466 m. It lies within a stepped, large-scale slope zone transitioning from the Xiangxi Hills to the Yunnan-Guizhou Plateau. Regionally, the area belongs to the Zunyi fault-uplift structural zone of the northern Guizhou bulge on the Yangtze paraplatform, within the NNE-trending Fenggang structural deformation belt and the complex structural deformation zone of Guiyang, bounded by the NE-trending Hongshi-Shijiachang (Shigu)-Shiqian Fault [46]. Tectonic architecture and plate-tectonic setting govern the formation and evolution of thermal mineral waters. During the Yanshanian, subduction of the Pacific Plate beneath Eurasia drove Circum-Pacific orogenesis, imposing a SE-NW compressive stress field across eastern Eurasia and establishing a regional NNE-trending fold-fault system [47]. As tectonism progressed, the relative southwestward motion of Eurasia reoriented the stress field to NNE-SSW compression, providing the driving force for the East Asian strike-slip zone and fostering NE-trending strike-slip faults within it. Because these two stress regimes were temporally close, their superposed effects imparted an S-shaped curvature to the strike-slip faults and adjacent NNE-trending folds [48]. This tectonic framework underpins the region’s present-day structure and geomorphology and overprints earlier Wulingian and Caledonian systems [49]. Since the late Pliocene, indentation of the Indian Plate into Eurasia has uplifted the Qinghai–Tibet Plateau and produced widespread thrust-nappe systems [47]. Influenced by western Tethyan evolution, plateau uplift, and far-field extrusion, neotectonics in Guizhou under the Himalayan regime is characterized by active faulting and block uplift, forming horst-graben assemblages, subparallel strike-slip systems, and shallow detachments. These young structures inherit, superimpose on, and rework older tectonic fabrics [47]. Multi-stage, reactivated NE-trending faults are most pervasive; they cut NNE, near-EW, and NW-trending active faults, and their oblique intersections create highly permeable conduits where numerous thermal mineral springs emerge. This pattern shows that active faults are the principal pathways that (i) transmit deep geothermal energy upward and (ii) drive downward circulation of shallow meteoric water for heating, thereby controlling the spatial distribution of geothermal waters in Guizhou [50]. Bedrock exposures strongly influence groundwater recharge and reservoir connectivity. In the study area, thermal mineral waters are hosted in water-bearing limestones, dolomites, and limestone-shale contacts that exhibit high permeability and transmissivity. Within storage structures such as the Shiqian Anticline, the Shiqian Fault, and the Hongshi strike-slip zone, thermal reservoirs along the anticline axis locally crop out, whereas the flanks are bounded by lower-permeability strata [51,52]. The hinge-parallel Shiqian Fault cuts Cambrian–Silurian strata and intersects obliquely with the Hongshi strike-slip zone near Shiqian County, focusing flow where highly conductive NE-trending channels meet [49]. Lithologic contrasts produce strong permeability anisotropy and lower thermal resistance along the anticline axis, which together focus recharge, throughflow, and reservoir connectivity. Consequently, most thermal springs are concentrated within limestones and dolomites of the Tongzi–Honghuayuan formations and at their contacts with Dawang Formation shales [53].

The Shiqian Anticline stretches in a NNE direction, presenting a narrow and banded distribution. Its core is frequently disrupted by the Shiqian compressive-torsional fault with the same strike, resulting in an “S”-shaped curvature. The two limbs of the anticline are asymmetric, with the western limb being gentler than the eastern one; the dip angle of the western limb ranges from 20° to 40°. Two sets of crisscross extensional fractures (trending NNE and NWW respectively) are well-developed in both the core and limbs [13]. The Shiqian Fault, striking NNE (Figure 1), is a secondary structural unit in the eastern part of the Zunyi Fault-Arch. It develops on a large scale and is closely associated with the Shiqian Anticline, sharing the consistent “S”-shaped curvature [14]. The Hongshi Fault, formed during the Xuefeng Period [15,16], strikes NE (Figure 1) and traverses the entire study area obliquely. With a great cutting depth, it facilitates the upward conduction of deep geothermal energy and intersects with the Shiqian Fault at the southern end of Shiqian County. According to the interpretation profile [54] (Figure 2) of controlled-source audio magnetotelluric (CSAMT) geophysical exploration for the Shiqian and Hongshi Faults, the Shiqian Fault has a dip angle of 70°, while the Hongshi Fault has a dip angle of 65°. Both are high-angle steeply inclined faults, which are conducive to the rapid ascent of deep heat flow to the shallow layer, thereby promoting the rapid evolution of thermal fluids. Geological surveys show that the fractured zone and its affected zone of the Shiqian Fault are 10–150 m wide, often accompanied by secondary small faults, flexures, and visible breccias. The Hongshi Fault is a transpressional fault with a fractured zone width of 5–10 m and an affected zone of 100–300 m, and four sets of fractures are developed on the fault plane. Accordingly, the rocks on both sides of the faults are fragmented, with well-developed tension fractures, shear fractures, and branch faults. This not only provides favorable channels for the infiltration of surface water and shallow groundwater but also facilitates the acquisition of deep heat sources [13]. The Shiqian Hot Spring Group is distributed along the NE-trending Hongshi Fault, NNE-trending Shiqian Fault, and their secondary fault bundles in the area.

Specifically, the synergistic effect of geological structures, bedrock outcrops, and lithological differences collectively regulates groundwater recharge, runoff, and geothermal reservoir connectivity. Taking the Shiqian Anticline, the core structure of the study area, as an example, the axial part of the anticline is the outcrop area of geothermal reservoir strata, serving as a natural recharge window for groundwater. In contrast, the two flanks of the anticline are sealed by impermeable strata, which can effectively converge recharge water and reduce heat loss. The Shiqian hinge fault associated with the anticline axis cuts deeply through the Cambrian to Silurian strata and intersects obliquely with the Hongshi strike-slip fault bundle near the county seat. This structural composite zone forms an efficient water-conducting channel, enabling the convergence of thermal mineral water migrating along the NE-trending faults [55]. Meanwhile, the permeability differentiation caused by lithological differences further optimizes the migration paths of heat flow and groundwater. Due to the significant difference in rock permeability, heat flow tends to concentrate toward the anticline axis with low thermal resistance during conduction. Most thermal mineral springs in the study area are exposed at the contact zone between the limestone and dolomite of the Tongzi–Honghuayuan Formation and the shale of the Dawang Formation, eventually forming a thermal mineral water enrichment zone with strong water-bearing capacity and water abundance.

In the study area, except for the absence of Devonian, Carboniferous, Jurassic, Cretaceous, and Tertiary strata, the exposed stratigraphic sequence from oldest to youngest comprises the Sinian, Cambrian, Ordovician, Silurian, Permian, Triassic, and Quaternary systems [56,57]. The hosting horizon of thermal mineral water is the second carbonate thermal reservoir, occurring from the Cambrian Qingxudong Formation to the Ordovician Honghuayuan Formation (Figure 3), with lithologies dominated by dolomite and limestone. Overlying strata consist of the Meitan Formation or Dayawan Formation up to the Silurian Hanjiadian Group, characterized primarily by shale and sandstone lithologies. The impermeable basement is composed of the Upper Proterozoic Fanjingshan Group and Banxi Group, as well as the Sinian to lower Cambrian Niutitang, Mingxinsi, and Jindingshan formations, exhibiting lithologies of metamorphic rock, shale, and sandstone. Since the Xuefeng movement, Guizhou Province has been situated within the interior of the Eurasian Plate. Being far from plate margins, the region has experienced no significant magmatic intrusion or volcanic activity [43]. Apart from minor granite intrusions in the Mesoproterozoic Fanjingshan Group, strata from various geological periods are predominantly composed of sedimentary carbonate rocks and terrigenous clastic formations [58]. Therefore, the thermal source of Shiqian thermal mineral water is primarily derived from conductive heat from the deep Earth, with temperature increases sustained by the normal geothermal gradient [43].

3. Data Sources and Processing Methods

To investigate the hydrogeochemical characteristics and formation mechanisms of the Shiqian hot spring group, we collected thermal mineral water and reservoir rock samples in the study area on 12 October 2024. The data sources and analytical procedures are summarized below. Because characterization of the thermal reservoir is essential for interpreting the hydrochemical composition of thermal waters, assessing water–rock interactions, and simulating water–rock reaction processes, seven fresh rock samples were collected from the reservoir profile, spanning the Cambrian Qingxudong Formation to the Ordovician Honghuayuan Formation. Major element concentrations were determined by X-ray fluorescence spectrometry (ARL PERFORM’X 4200, Thermo Fisher Scientific, Waltham, MA, USA) at the State Key Laboratory of Mineral Deposit Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences. Scanning electron microscopy (SEM) was performed at the Key Laboratory of Karst Geological Resources and Environment, Guizhou University, using a high-resolution environmental SEM with fully automated mineral analysis (Quattro S + MapsMineralogy, Thermo Fisher Scientific, Waltham, MA, USA).

The hydrochemical data of thermal mineral water were acquired through field surveys and laboratory analyses. All water sample collections were conducted in strict accordance with the “Specification for Geothermal Resources Geological Exploration” (GB/T11615-2010 [59]). A total of 13 thermal water samples were collected within the study area (sampling locations are shown in Figure 1). Before sampling, 500 mL polyethylene bottles were soaked in 10% nitric acid for 24 h, rinsed 3–4 times with deionized water, and air dried. During sampling, a portable water quality analyzer (Multi 3620 IDS, Xylem Analytics, Weilheim, BAV, Germany) was used on site to measure parameters such as water temperature, pH, and total dissolved solids (TDS). Before filling, each sample bottle was rinsed three times with the to-be-collected water sample. The collected water was then filtered through a 0.45 μm microporous membrane, placed in the pre-rinsed polyethylene bottle, and sealed and refrigerated for transport to the laboratory. Cation concentrations (Na⁺, K⁺, Ca²⁺, Mg²⁺, Sr²⁺) and H₂SiO₃ were determined at the Guizhou Provincial Key Laboratory of Karst Mountain Ecology and Environment (Guizhou Normal University), using a Shimadzu AA-7000 atomic absorption spectrophotometer (Shimadzu Corporation, Kyoto, Japan) and a TU-1901 UV spectrophotometer (Beijing Puxi General Instrument Co., Ltd., Beijing, China), respectively. HCO₃⁻ concentrations were measured by acid-base titration at the training base of the State key laboratory of Karst Mountain Ecology and Environment. SO₄²⁻ and Cl⁻ were analyzed at the State Key Laboratory of Environmental Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences, using an ICS-90 ion chromatograph (Dionex Corporation, Sunnyvale, CA, USA). δD and δ¹⁸O stable isotopes were determined at the Institute of Karst Research, Guizhou Normal University, by a high-precision liquid water stable isotope analyzer (L2130-i, PICARRO, Santa Clara, CA, USA).

Remote sensing data consisted of Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) (NASA, Washington, DC, USA) imagery provided by NASA/USGS. Relative to ETM+, Landsat 8 offers an optimized spectral configuration, two additional bands (coastal aerosol and cirrus), and 12-bit radiometric resolution, improving feature discrimination and quantitative analysis [60]. OLI supports lithologic and structural mapping, whereas TIRS provides dual thermal infrared bands that enable high-precision land surface temperature (LST) retrieval for monitoring geothermal anomalies, with typical clear-sky errors on the order of 1 K [60]. For long-term comparisons, ETM+ data (7 reflective bands plus a 15 m panchromatic band) remain useful for tracking vegetation and surface changes [61]. Landsat 8 OLI multispectral imagery was downloaded from the Geospatial Data Cloud (http://www.gscloud.cn). The visible-to-shortwave infrared bands have 30 m spatial resolution. All selected scenes were cloud-free over the study area at overpass. After download, images were preprocessed in ENVI (Version 5.6), including radiometric calibration, atmospheric correction, geometric checks, and cloud/shadow masking, followed by cropping, contrast stretching, principal component analysis (PCA), optimal band selection, and filter-based enhancement

4. Results and Analysis

4.1. Geochemical Characteristics of Thermal Reservoir Rocks

4.1.1. Mineralogical Characteristics

In the Shiqian area, thermal mineral waters are primarily distributed in a zonal pattern along the Hongshi Fault and Shiqian Fault, where extensive faulting has facilitated the migration and accumulation of geothermal fluids. These waters are hosted within carbonate rock strata ranging from the Cambrian Qingxudong Formation to the Ordovician Honghuayuan Formation, which are predominantly composed of dolomite and limestone (Figure 4). Detailed scanning electron microscope (SEM) observations (Figure 5) reveal that the thermal reservoir rocks consist mainly of dolomite, calcite, quartz, gypsum, and celestite. Dolomite crystals present rhombic or rhombohedral forms with smooth surfaces and exhibit a range of crystallization sizes. Calcite is typically observed as rhombohedral or prismatic crystals with triangular or rhombic cross-sections. Quartz occurs as hexagonal prismatic crystals, often terminated by hexagonal pyramids at either end, which together create a regular crystal morphology. Gypsum is predominantly found in tabular or fibrous forms, while celestite displays smooth surfaces and manifests as prismatic or tabular crystals. The mineral composition and crystal morphology of these reservoir rocks reflect the geological evolutionary history of the study area and play a key role in controlling reservoir properties, as well as the hydrochemical characteristics and formation mechanisms of the thermal mineral waters.

4.1.2. Characteristics of Major Elements

The thermal reservoir in the study area is mainly composed of carbonate rocks, and its mineral assemblages are dominated by dolomite, calcite, and quartz. Major element analysis of the thermal reservoir rock samples (Figure 6;

Table 1. Major element analysis results of thermal reservoir rock samples from the study area.

Sample Number	Geothermal Reservoir	Lithology	Analytical Parameters(%)
			CaO	MgO	SiO2	Al2O3	Fe2O3	K2O	Na2O	MnO	P2O5	TiO2	SO3	LOI	SUM
YK01	∈3–4O1l	Dolomite	37.02	10.88	9.31	0.72	0.27	0.16	0.01	0.02	0.00	0.02	0.30	40.89	99.60
YK02	∈3–4O1l	Dolomite	33.21	12.81	10.9	1.28	0.54	0.26	0.04	0.02	0.01	0.05	0.05	40.27	99.44
YK03	∈3–4O1l	Dolomite	36.92	13.67	5.26	0.31	0.16	0.10	0.00	0.01	0.01	0.02	0.07	43.49	100.02
YK04	O1h	Limestone	17.26	8.76	30.91	10.50	3.15	3.65	0.03	0.05	0.07	0.32	0.03	25.18	99.91
YK05	O1h	Limestone	27.45	17.67	12.16	0.60	0.49	0.18	0.02	0.02	0.01	0.02	0.15	40.57	99.34
YK06	O1t	Dolomite	29.66	17.34	11.53	0.25	0.15	0.06	0.04	0.02	0.00	0.01	0.01	41.40	100.47
YK07	O1th	Siliceous Dolomite	30.68	17.23	7.86	0.35	0.24	0.10	0.04	0.02	0.00	0.01	0.01	42.99	99.54

) reveals that CaO, MgO, and SiO₂ are the principal components, while Fe₂O₃, Al₂O₃, Na₂O, K₂O, MnO, P₂O₅, TiO₂, and SO₃ are present at relatively low concentrations. Specifically, CaO content ranges from 17.26% to 37.02%, with an average of 30.31%; MgO ranges from 8.76% to 17.67% (average 14.05%); and SiO₂ ranges from 5.26% to 30.91% (average 12.56%). Other major oxides, including Al₂O₃ (0.25%–10.50%, average 2.00%), Fe₂O₃ (0.15%–3.15%, average 0.71%), K₂O (0.10%–3.65%, average 0.64%), Na₂O (0.00%–0.04%, average 0.03%), MnO (0.01%–0.05%, average 0.02%), P₂O₅ (0.00%–0.07%, average 0.01%), TiO₂ (0.01%–0.32%, average 0.06%), and SO₃ (0.01%–0.30%, average 0.09%), are significantly less abundant. In addition, the high loss on ignition (LOI), which ranges from 25.18% to 42.99% (average 39.26%), further reflects the substantial volatile content in these rocks.

These geochemical characteristics indicate that the abundance of CaO, MgO, and SiO₂ in the thermal reservoir rocks provides a substantial material foundation for the ionic composition of thermal mineral waters. During water–rock interaction, these major components are continuously released into the thermal water via dissolution and ion exchange. This ongoing process not only shapes the chemical profile of the thermal mineral waters but also offers important insights into the mechanisms of water–rock interaction and the dynamics of material migration and transformation within the reservoir. Moreover, these processes contribute to the health benefits associated with thermal mineral waters.

4.2. Hydrochemical Characteristics

4.2.1. Major Cation and Anion Characteristics and Hydrochemical Types of the Shiqian Hot Spring Group

Thermal mineral waters were sampled from geothermal wells and natural spring outlets across the study area. As summarized in

Table 2. Hydrochemical composition analysis results of thermal mineral water from the study area.

Sample Number	Name of Hot Spring/Geothermal Well	Hydrochemical Components (mg/L)
		T(°C)	pH	TDS	K+	Na+	Ca2+	Mg2+	HCO3−	SO42−	Cl–	H2SiO3	Sr2+
SQ01	Guanyuliang Hot Spring	28.7	7.24	756.0	5.89	2.42	104.01	23.94	242.01	246.14	1.01	41.98	7.17
SQ02	Shichang No.1 Hot Spring	46.5	7.13	354.0	1.45	1.48	46.01	15.37	226.40	30.53	0.60	65.51	0.38
SQ03	Shichang No.2 Hot Spring	23.9	8.13	407.0	1.08	0.89	53.93	17.80	320.08	11.96	2.24	19.46	0.06
SQ04	Kaixiahe Karst Cave Hot Spring	38.9	7.99	434.0	3.28	4.31	73.23	16.10	242.01	97.40	6.28	35.97	1.34
SQ05	Shiqian Beita Geothermal Well	40.4	7.56	353.0	1.17	1.29	41.82	16.42	195.17	46.94	0.42	30.01	1.47
SQ06	ChengbeiQuandu Geothermal Well	30.1	7.64	371.0	2.08	4.41	48.34	12.67	484.02	37.54	4.99	20.63	0.81
SQ07	Chengbei Hot Spring	26.1	7.97	363.0	1.19	2.23	43.77	16.41	210.78	37.54	2.36	23.63	1.23
SQ08	Old Public Security Bureau Geothermal Well	41.7	7.43	404.0	1.65	3.59	48.14	17.72	226.40	71.58	0.37	31.69	1.38
SQ09	Guochang Geothermal Well	41.3	7.58	493.0	2.66	11.99	55.46	18.74	257.62	103.33	1.97	32.53	1.85
SQ10	Chengnan Hot Spring	41.1	7.42	661.0	4.71	11.81	86.21	21.74	226.40	201.39	1.16	41.14	3.12
SQ11	Wujiawan Geothermal Well	37.5	7.36	691.0	4.41	82.63	31.73	14.00	429.37	65.62	2.94	39.68	2.04
SQ12	Zhongba Jiangpo Geothermal Well	43.2	8.03	678.0	1.72	127.39	18.86	10.61	452.79	1.49	1.97	26.14	0.21
SQ13	Qiaobian Geothermal Well	38.9	8.06	677.0	1.45	127.08	19.59	10.80	507.44	1.54	1.76	28.92	0.27

T in °C; pH is dimensionless; all other hydrochemical parameters are in mg/L.

, temperatures range from 23.9 to 46.5 °C (mean 36.8 °C), classifying the resource as low- to medium-temperature thermal water (per GB/T 11615-2010 [59]). Measured pH values are 7.13–8.13 (mean 7.66), indicating overall weak alkalinity. Total dissolved solids (TDS) are 353–756 mg/L (mean 519.92 mg/L), consistent with moderate mineralization. As shown in Figure 7 and Table 2, the dominant cations are Ca²⁺ and Mg²⁺, followed by Na⁺ and K⁺. Concentrations (range; mean) are: Ca²⁺ 18.86–104.01 mg/L; 51.62 mg/L; Mg²⁺ 10.61–23.94 mg/L; 16.33 mg/L; Na⁺ 0.89–127.39 mg/L; 29.35 mg/L; and K⁺ 1.17–5.89 mg/L; 2.52 mg/L. Elevated Ca²⁺ and Mg²⁺ are consistent with dissolution of carbonate aquifers. By contrast, samples SQ11–SQ13 show anomalously high Na⁺, plausibly linked to flow along a compressive–torsional fault that intersects silicate-rich Banxi Group strata in the Zhongba area, enhancing silicate weathering and/or cation exchange. Despite these local enrichments, Ca²⁺ and Mg²⁺ control the bulk cation chemistry of the thermal waters. Among anions, HCO₃⁻ and SO₄²⁻ predominate: HCO₃⁻ 195.17–507.44 mg/L (mean 309.27 mg/L); SO₄²⁻ 1.49–246.14 mg/L (mean 73.31 mg/L); and Cl⁻ 0.37–6.28 mg/L (mean 2.16 mg/L). To assess seasonal stability, we compared dry-season results with wet-season data from prior work [53]. As shown in

Table 3. Comparison of hydrochemical analysis results of thermal mineral water in dry and wet seasons in the study area.

Sample Number	Translating Sampling Time	Hydrochemical Components (mg/L)
		Ca2+	Mg2+	H2Si03	Sr2+	HCO3−
SQ01	2024.10	104.01	23.94	41.98	7.17	242.01
	2018.05 *	121.65	26.72	39.37	7.82	191.49
SQ02	2024.10	46.01	15.37	65.51	0.38	226.39
	2018.05 *	49.71	14.83	61.01	0.06	185.86
SQ03	2024.10	53.93	17.80	19.46	0.56	320.07
	2018.05 *	56.64	16.47	24.63	0.56	242.39
SQ04	2024.10	73.23	16.10	35.97	1.34	242.00
	2018.05 *	84.67	18.14	37.32	1.67	215.43
SQ07	2024.10	43.77	16.41	23.63	1.23	210.78
	2018.05 *	47.49	15.93	23.15	1.43	176.0
SQ10	2024.10	86.21	21.74	41.14	3.12	226.39
	2018.05 *	78.00	21.08	36.35	2.91	208.39
SQ11	2025.10	31.73	14.00	39.68	2.04	429.37
	2018.05 *	43.65	16.42	28.59	1.67	170.37
SQ12	2024.10	18.85	10.61	26.14	0.21	452.79
	2018.05 *	6.47	1.96	34.36	0.28	580.10

* denotes data sourced from Reference [53].

and Figure 8a, HCO₃⁻ exhibits the same spatial variation trend in both seasons, indicating stable hydrochemical patterns; absolute concentrations are generally lower in the wet season, consistent with precipitation-driven dilution and increased recharge.

The ionic combination ratio coefficient method is widely applied to investigate the origin of groundwater solutes and to characterize hydrochemical processes [55,62]. In this study, the milliequivalent concentration ratio of Ca²⁺/Mg²⁺ (meq/L) was used to assess the relative contributions of calcite and dolomite dissolution to the ionic composition of thermal mineral water. As shown in Figure 8b, Ca²⁺/Mg²⁺ ratios in both dry and wet season samples are predominantly distributed between the 1:1 and 1:2 reference lines, indicating that the majority of divalent cations originate from calcite dissolution [63]. This pattern suggests that the hydrochemistry of the thermal waters is primarily governed by water–rock interactions within carbonate aquifers. Furthermore, the absence of a marked seasonal shift in the Ca²⁺/Mg²⁺ ratio implies that carbonate dissolution processes operate with notable stability throughout the year.

The hydrochemical properties of natural waters are relatively complex, making hydrochemical type classification an essential approach for investigating these characteristics [64]. In this study, the Shklomanov classification method [65] was applied to classify the thermal mineral water types in the study area based on the distribution patterns of seven major cations and anions (Cl⁻, HCO₃⁻, SO₄²⁻, K⁺, Na⁺, Ca²⁺, and Mg²⁺). As a result, three principal hydrochemical types were identified: HCO₃·SO₄-Ca·Mg, HCO₃-Ca·Mg, and HCO₃-Na. Specifically, samples SQ01, SQ04, SQ09, and SQ10 are characterized by the HCO₃·SO₄-Ca·Mg type, while SQ02, SQ03, SQ05, SQ06, SQ07, and SQ08 are classified as the HCO₃-Ca·Mg type. The HCO₃-Na type, represented by samples SQ11, SQ12, and SQ13, has been attributed to the compressive-torsional faulting in the Zhongba area. This faulting has disrupted the Banxi Group strata, resulting in the incorporation of silicate rock fragments into the aquifer and thus transforming the hydrochemical composition of those samples towards Na⁺ enrichment [66].

The Piper trilinear diagram provides a clear visualization of hydrochemical types and the evolutionary trends of water chemistry [62,67]. As shown in Figure 9, cations in thermal mineral water samples SQ01–SQ10 are dominated by Ca²⁺ and Mg²⁺, with Ca²⁺ accounting for more than 50% of the milliequivalent percentage in all cases. In contrast, Na⁺ is the dominant cation in samples SQ11, SQ12, and SQ13, where its milliequivalent percentage also exceeds 50%. The anion plots for all samples cluster along the left boundary of the diagram, within the HCO₃⁻ end-member field, indicating that the hydrochemical components of the thermal mineral water originate from the weathering and dissolution of carbonate and silicate rocks [68].

4.2.2. Characteristics of Special Components

Geothermal water under hydrothermal conditions intensifies water–rock interactions within the aquifer, thereby promoting the enrichment of both conventional constituents and trace elements in the water [69]. The principal sources of trace elements in thermal mineral water are the water–rock interactions occurring during migration through country rocks, as well as the possible addition of deep mantle-derived magmatic water [70]. As indicated by Table 2 and Figure 10, strontium (Sr²⁺) concentrations in thermal mineral water from the study area range from 0.06 to 7.17 mg/L, with an average of 1.65 mg/L. The highest Sr²⁺ content (7.17 mg/L) is recorded in sample SQ01 (Guanyuliang Hot Spring), while the lowest (0.06 mg/L) is found in sample SQ03 (Shichang No. 2 Hot Spring). Metasilicic acid (H₂SiO₃) concentrations range from 19.46 to 65.51 mg/L, with an average of 33.64 mg/L. The highest value, 65.51 mg/L, occurs in sample SQ02 (Shichang No. 1 Hot Spring), and the lowest, 19.46 mg/L, is observed in SQ03 (Shichang No. 2 Hot Spring). As shown in Table 3 and Figure 11, the concentrations of metasilicic acid and strontium in the thermal mineral water exhibit similar variation trends during the dry and wet seasons, suggesting that the overall circulation regime of the thermal mineral water remains stable throughout the year. According to the Chinese national standard GB 8537-2018 [71] for Drinking Natural Mineral Water, groundwater with Sr²⁺ concentrations ≥ 0.2 mg/L and temperatures above 25 °C may be classified as strontium–rich thermal mineral water. All hot spring and geothermal well samples in the study area meet this standard except SQ03, which has Sr²⁺ below 0.2 mg/L. Similarly, for groundwater at temperatures above 25 °C, a H₂SiO₃ concentration ≥ 25.0 mg/L is required to classify the water as metasilicic acid–rich thermal mineral water. As shown in Table 2, all hot spring and geothermal well samples satisfy this criterion except SQ03 (Shichang No. 2 Hot Spring), SQ06 (Chengbei Quandu Geothermal Well), and SQ07 (Chengbei Hot Spring), each with H₂SiO₃ below 25 mg/L.

According to GB/T 11615-2010 Specification for Exploration of Geothermal Resources (“Appendix E Hydrotherapy Thermal Mineral Water Quality Standards”), geothermal water with a temperature of at least 34 °C and a strontium (Sr²⁺) concentration of 10 mg/L has health benefits. Table 2 shows that Sr²⁺ levels at most sampling sites in the study area are close to this threshold. Similarly, when the concentration of metasilicic acid (H₂SiO₃) reaches 25 mg/L, the geothermal water also possesses health benefits, and Table 2 indicates that most sampling sites have H₂SiO₃ concentrations above 25 mg/L. Furthermore, when the H₂SiO₃ concentration is at least 50 mg/L, the water can be designated as metasilicic acid–rich hydrotherapy thermal mineral water. In the study area, sample SQ02 (Shichang No. 1 Hot Spring) has an H₂SiO₃ concentration of 65.51 mg/L, meeting the requirements for both drinking natural mineral water and hydrotherapy thermal mineral water.

Overall, most hot springs and geothermal wells in the study area contain metasilicic acid (H₂SiO₃) and strontium (Sr²⁺) at concentrations that meet the standards for strontium-rich and metasilicic acid-rich thermal mineral drinking water. Consequently, these waters can be classified as Sr–H₂SiO₃ composite-type natural drinking mineral water. Notably, the SQ02 Hot Spring satisfies both the drinking water and hydrotherapy classification criteria. The presence of these high-quality thermal mineral water resources presents opportunities for their development as natural mineral drinking waters and for hot spring hydrotherapy. Such development has the potential to stimulate the local mineral water industry and tourism sector, thereby enhancing the region’s economic value and competitiveness.

4.3. Hydrogen and Oxygen Isotope Characteristics

Stable isotope analysis plays an important role in hydrogeological studies of both thermal and non-thermal waters, as these isotopes record information about fluid sources and related processes [72], making them highly significant for clarifying watershed water cycles and groundwater recharge relationships. As shown in

Table 4. Stable hydrogen and oxygen isotope composition of water samples.

Sample Number	Name of Hot Spring/Geothermal Well	δD/‰	δ18O/‰
SQ01	Guanyuliang Hot Spring	–56.54	–8.63
SQ02	Shichang No.1 Hot Spring	–57.12	–8.76
SQ03	Shichang No.2 Hot Spring	–49.37	–7.82
SQ04	Kaixiahe Karst Cave Hot Spring	–48.64	–7.64
SQ05	Shiqian Beita Geothermal Well	–53.94	–8.43
SQ06	ChengbeiQuandu Geothermal Well	–41.04	–6.48
SQ07	Chengbei Hot Spring	–43.49	–6.76
SQ08	Old Public Security Bureau Geothermal Well	–55.62	–8.67
SQ09	Guochang Geothermal Well	–56.02	–8.68
SQ10	Chengnan Hot Spring	–56.64	–8.73
SQ11	Wujiawan Geothermal Well	–58.31	–9.06
SQ12	Zhongba Jiangpo Geothermal Well	–56.40	–8.42
SQ13	Qiaobian Geothermal Well	–56.00	–8.45

, the δ¹⁸O values of thermal mineral water in the study area range from –9.06‰ to –6.48‰, with an average of –8.19‰, while the δD values range from –58.31‰ to –41.04‰, with an average of –53.01‰. Hydrogen and oxygen stable isotopes are commonly used to investigate relationships between recharge sources of thermal mineral water and atmospheric precipitation, to characterize hydrogeochemical processes within geothermal systems, and to serve as effective indicators for evaluating the influence of evaporation on water chemistry [73,74,75].

The hydrogen and oxygen isotope data of thermal mineral water in the study area were plotted on the δD-δ¹⁸O relationship diagram of the Global Meteoric Water Line [(GMWL: δD = 8δ¹⁸O + 10) [76] and the Local Meteoric Water Line of Southwest China (LMWL: δD = 7.44δ¹⁸O + 6.36) [77](Figure 12). The fitted linear equation is δD = 6.93δ¹⁸O + 3.75 (R² = 0.977982). Its slope and intercept are lower than those of the local meteoric water line, indicating that atmospheric precipitation is affected by evaporation during groundwater recharge [78]. The hydrogen and oxygen isotope values of water bodies in the study area all fall near the Global Meteoric Water Line and the local meteoric water line, suggesting that the thermal mineral water in this area is formed by atmospheric precipitation infiltration and recharge, followed by deep circulation and heating under the effect of geothermal temperature. It can be seen from the distribution of thermal mineral water in the diagram that all thermal mineral water in the area has undergone varying degrees of negative drift, indicating low degree of isotope exchange of thermal mineral water and an unclosed water–rock interaction environment [79].

To evaluate the extent of oxygen isotope exchange between regional water and rocks, W. Dansgaard [80] introduced the concept of deuterium excess (d-excess), defined as d = δD − 8 δ¹⁸O. This parameter reflects the degree of oxygen isotope exchange between water and rock within a region and serves as a key indicator for quantifying variations in this exchange [81]. Once the Local Meteoric Water Line (LMWL) of a region is determined, its deuterium excess parameter (d-excess) of atmospheric precipitation is fixed accordingly. Theoretically, it remains constant and is not affected by factors such as season and altitude [73]. However, after atmospheric precipitation infiltrates and recharges the aquifer, water–rock interaction becomes the core factor modifying the isotopic composition. Thermal mineral water undergoes isotopic exchange with oxygen-bearing rocks, resulting in a significant increase in δ¹⁸O. In contrast, rocks contain minimal hydrogen-bearing components, exerting a negligible effect on δD, thus δD remains basically stable. According to the definition formula of d-excess (d = δD − 8δ¹⁸O), a constant δD coupled with an elevated δ¹⁸O will inevitably lead to a decrease in d-excess. The magnitude of δ¹⁸O enrichment in thermal mineral water is determined by three factors: the composition of oxygen-bearing rock components, aquifer temperature, and groundwater residence time. Within the same aquifer, δ¹⁸O exhibits a significant positive correlation with residence time, meaning a longer residence time corresponds to a higher δ¹⁸O value [50,82]. As can be seen from the definition of d-excess, when δD remains unchanged and δ¹⁸O increases, the d-excess value decreases accordingly. A lower d-excess value indicates a greater circulation depth of thermal mineral water, which leads to a higher aquifer temperature under the influence of the geothermal gradient. A greater circulation depth of thermal mineral water also implies a longer residence time in the aquifer and a more sufficient duration of water–rock interaction. Long-term accumulation results in a more thorough modification of groundwater isotopes by rocks, ultimately leading to a stronger water–rock interaction, and finally presenting the characteristics of a significant increase in δ¹⁸O and a synchronous decrease in d-excess [50,82]. As shown in Figure 13, the d-excess values in the study area range from 10.61‰ to 14.18‰, with an average of 12.55‰, and plot above the global meteoric water d-excess line (10‰). This suggests that the water–rock interaction environment in the study area is open, which means that the reservoir temperature is low, and water–rock interactions are weak, resulting in limited isotope exchange.

In atmospheric precipitation, both δD and δ¹⁸O values decrease progressively with increasing elevation, reflecting the altitude effect of stable isotopes in precipitation [83,84]. Based on the altitude effect formula proposed by Li Weijie [77], which was derived from studies on the stable isotope characteristics of precipitation across different terrains in Southwest China (δ¹⁸O = −0.0028 H − 3.93). This formula has incorporated the effects of local factors such as evaporation conditions and water vapor sources by adopting the meteoric water lines of the study area and its corresponding sub-regions. the estimated recharge elevation in the study area ranges from 911 m to 1833 m (

Table 5. Recharge elevation of geothermal water in the study area.

Sample Number	SQ01	SQ02	SQ03	SQ04	SQ05	SQ06	SQ07
Recharge elevation(m)	1679	1723	1389	1325	1607	911	1011
Sample number	SQ08	SQ09	SQ10	SQ11	SQ12	SQ13
Recharge elevation(m)	1692	1697	1714	1833	1602	1614

), with an average of 1523 m.

Based on the results of age determination of mineral water in Shiqian area by Chen Lü’an [85] and Zhang Shicong [86] using the isotopic methods of δ¹⁴C, δ¹³C and δ³H, the age of mineral water from Guanyuliang (SQ01) in Shiqian is 1700–2400 a, that from Shichang (SQ02) is 6400–6700 a, that from Chengnan (SQ10) is 10,000–10,400 a, and that from Wujiawan (SQ11) is 9511–9900 a. It can be seen that the thermal mineral water in Shiqian area has an age of approximately 10,000 years, belonging to the late Pleistocene paleowater, and is characterized by long circulation cycle and long recharge path.

5. Discussion

5.1. Recharge, Runoff, and Discharge Processes of the Shiqian Hot Spring Group and Their Controlling Factors

The recharge-runoff-discharge relationships of thermal mineral water are important factors influencing the formation of its hydrochemical composition and are primarily controlled by topography, lithology, and tectonic conditions [87]. Analysis of δD and δ¹⁸O characteristics shows that the thermal mineral water in the study area is recharged by atmospheric precipitation, with recharge elevations ranging from 911 m to 1833 m and an average of 1523 m. Combined with the elevation map of the area (Figure 14a), regions with an average altitude of 1523 m are found along the southwestern foot of Fanjingshan, indicating that the recharge area for the thermal mineral water is located in the southwestern foothills of Fanjingshan. The study area’s overall terrain presents a northeast-high to southwest-low trend. This topography drives groundwater recharge under gravity, resulting in a runoff pattern from higher to lower elevations, specifically moving from the southwestern foothills of Fanjingshan toward the southwest. Tectonic movements have led to faulting, displacement, uplift, and other structural changes, forming numerous fault zones and fracture zones of varying orientations and types [88], which provide favorable channels and spaces for heat transfer and enrichment of thermal mineral water. Accurately determining fault locations is crucial for understanding the recharge-runoff-discharge relationships of thermal mineral water [20]. Remote sensing imagery interpretation enables identification of fault structures at both macro and micro scales and assists in analyzing and predicting geothermal and mineralization conditions [89]. Faults in remote sensing images can be recognized using indicators such as tone, geomorphology, texture, and linear features; comprehensive analysis of these indicators confirms the presence of faults [90]. In this study, Landsat 8 remote sensing imagery was used to identify fault structures in the area through visual interpretation of these indicators. As shown in Figure 14b, multiple faults and anticline structures were interpreted, with the faults displaying distinct linear traces in the imagery. The Shiqian Fault exhibits large-scale extension, stable morphology, and prominent spiral structures; its geometry runs parallel to mountain ranges and river valleys, appearing in both parallel and conjugate forms and displaying characteristics of compressional and torsional faults [91,92]. The Hongshi Fault is characterized by wavy undulations, linear folds, feather-like fractures, drag structures, and small spiral forms, and also exhibits both compressional and torsional characteristics [91,92]. Both faults show combined torsional and compressional features, possess strong water-blocking properties, and control the spatial distribution of thermal mineral water in the area. The distribution of hydrothermal spring outflows indicates that springs are mainly located between the Shiqian and Hongshi faults, with fewer found outside these faults and almost none south of the Hongshi Fault, confirming the strong water-blocking nature of these faults and their role as primary boundary faults of the geothermal field in the study area.

In summary, the thermal mineral water in the study area is primarily recharged by atmospheric precipitation originating at the southwestern foothills of Fanjingshan. Under the influence of the Shiqian and Hongshi faults, the water flows from high-altitude areas to lower elevations in a southwestward direction, emerging south of Shiqian City as hot springs. Thus, the northeast-high to southwest-low geomorphological pattern and the water-blocking boundary faults of Shiqian–Hongshi are the principal factors governing the recharge-runoff-discharge relationships of thermal mineral water in the study area.

The thermal reservoir area refers to the effective distribution range of the thermal reservoir. Accurately delineating this range is a prerequisite for ensuring the accuracy and reliability of resource quantity calculation results. Based on the results of geothermal geological survey, geophysical exploration and recharge elevation calculation in the study area, combined with the characteristics of boundary faults, this study delineated the thermal reservoir boundary (Figure 14a) and determined that the thermal reservoir area in the study area is approximately 2.8 × 10³ km². According to the thermal reservoir histogram of the study area (Figure 3), the calculated thermal reservoir thickness ranges from 1060 m to 1895 m, with an average thickness of 1447.5 m. In addition, referring to the analysis results of Chen Jin et al. [93] on the distribution status and genesis of geothermal resources in Shiqian County, the total flow rate of geothermal water in this area is 58.45–62.4 L/s, indicating that the thermal reservoir in the study area is large in scale and the thermal mineral water resources have good development potential. The temperature attenuation rate (t) of thermal mineral water, calculated based on the thermal reservoir temperature (T1) derived from the thermal SiO₂ geothermometer and the measured temperature (T), ranges from 32.11% to 65.20%, with an average of 48.28%. The detailed calculation results are shown in

Table 6. Geothermometer calculation results and estimated temperature attenuation rates of thermal mineral waters in the study area.

Sample Number	T1	T2	T3	$\mathbf{t}=\frac{{\mathbf{T}}_1-\mathbf{T}}{{\mathbf{T}}_1}\times 100\%$
SQ01	82.46	7.60	51.27	65.20%
SQ02	102.16	18	72.26	54.48%
SQ03	52.88	−11.77	20.45	54.80%
SQ04	76.09	−3.88	44.57	48.88%
SQ05	68.91	−0.56	37.06	41.37%
SQ06	54.96	−10.20	22.59	45.23%
SQ07	59.88	−6.63	27.67	56.41%
SQ08	71.04	0.79	39.28	41.30%
SQ09	72.07	1.43	40.36	42.69%
SQ10	81.61	7.11	50.38	49.64%
SQ11	80.11	6.25	48.80	53.19%
SQ12	63.63	−4.04	31.57	32.11%
SQ13	67.48	−1.49	35.57	42.35%

Note: “t” denotes the temperature attenuation rate, and “T” denotes the measured temperature.

.

5.2. Sources of Major Ions and Special Components in the Shiqian Hot Spring Group

Gibbs [94], based on the relationship between TDS and ionic ratios in precipitation, river water, lake water, and seawater samples worldwide, summarized three principal mechanisms that control the hydrochemical characteristics of natural waters: precipitation dominance, rock weathering, and evaporation–concentration. Analysis of Gibbs diagrams for thermal mineral water provides valuable insights into the origins of hydrochemical components and the water–rock interaction processes occurring during groundwater circulation [94,95]. Using hydrochemical test data from the Shiqian thermal mineral water, all sampling points from the Shiqian Hot Spring Group were plotted on the Gibbs diagram (Figure 15). As shown in the figure, both cations and anions from the thermal mineral water samples are distributed on the left side of the diagram, which corresponds to the rock-weathering dominance type. This indicates that water–rock interactions are the primary factor controlling the hydrochemical composition of the thermal mineral water. In the Na⁺/(Na⁺ + Ca²⁺) ratio diagram (Figure 15a), samples SQ11, SQ12, and SQ13 plot on the right side with moderate TDS values and Na⁺/(Na⁺ + Ca²⁺) > 0.5, suggesting that cation-exchange reactions influence the composition of the thermal mineral water. In the Cl⁻/(Cl⁻ + HCO₃⁻) ratio diagram (Figure 15b), the thermal mineral water samples are distributed in the left to middle portion of the diagram with low Cl⁻ concentrations, indicating limited leaching of halite during circulation. Overall, all samples fall within the rock-weathering domain of the Gibbs diagram, confirming that rock weathering predominantly controls the hydrochemical composition of the thermal mineral water in the study area.

Differences in the concentration ratios of Mg²⁺/Na²⁺, Ca²⁺/Na⁺, and HCO₃⁻/Na⁺ resulting from the weathering of carbonate rocks, silicate rocks, and evaporites can further elucidate the sources of ions in thermal mineral water [96]. As shown in Figure 16, most thermal mineral water samples are distributed near the carbonate-weathering zone, while a few are located near the silicate-weathering zone. This distribution indicates that the ionic composition of the thermal mineral water is primarily controlled by the dissolution of carbonate minerals, with silicate weathering as a secondary influence. Overall, these results demonstrate that carbonate rocks are the main factor influencing the hydrogeochemical composition of the thermal mineral water in the study area, with silicate rocks playing a subordinate role.

Correlation analysis of cation and anion concentrations in thermal mineral water can be used to determine the degree of association between ions and to effectively reveal their possible sources [97]. Based on hydrochemical data, ion correlation diagrams were plotted (Figure 17). As shown in Figure 17, the TDS content of thermal mineral water in the study area exhibits positive correlations with the concentrations of K⁺, Na⁺, Ca²⁺, Mg²⁺, HCO₃⁻, SO₄²⁻, H₂SiO₃, and Sr²⁺, indicating that these ions all contribute to TDS. K⁺ displays positive correlations with Ca²⁺, Mg²⁺, SO₄²⁻, and Sr²⁺. HCO₃⁻ is strongly correlated only with Na⁺, while its correlations with other cations, anions, and TDS are weak. This suggests that the source of HCO₃⁻ is relatively complex and may be influenced by rock weathering, mineral leaching, mixing processes, and degassing of deep-seated CO₂ [6]. Cl⁻ exhibits a positive correlation with HCO₃⁻, but weak or even negative correlations with other ions. H₂SiO₃ shows no correlation with any of the other ions.

Cluster analysis dendrogram of the thermal mineral water in the study area was performed, as shown in Figure 18. When the truncation distance is set to D = 0.6, the ionic components can be divided into three groups: Group 1 includes Na⁺, HCO₃⁻, and Cl⁻; Group 2 comprises K⁺, SO₄²⁻, Sr²⁺, Ca²⁺, and Mg²⁺; and Group 3 consists solely of H₂SiO₃. It can be concluded that Na⁺, HCO₃⁻, and Cl⁻ in the first cluster share similar material sources, while K⁺, Ca²⁺, Mg²⁺, SO₄²⁻, and Sr²⁺ in the second cluster also originate from analogous sources. This geochemical association indicates that the origin and enrichment of strontium are closely related to carbonate rocks and gypsum-salt layers [47]. In natural systems, strontium is primarily hosted in celestine (SrSO₄) and strontianite (SrCO₃), which commonly occur paragenetically with gypsum. The thermal reservoirs in the study area are rich in gypsum and associated celestine, as confirmed by scanning electron microscopy (SEM) observations of reservoir rocks, and this mineral assemblage is consistent with the characteristics of carbonate-hosted thermal reservoirs in Guizhou Province. Furthermore, strontium and calcium both belong to Group II elements of the periodic table and share similar ionic radii and identical charges. As a result, strontium can readily substitute for calcium in the crystal lattices of minerals such as gypsum (CaSO₄) and calcite (CaCO₃) through isomorphic replacement, leading to the enrichment of celestine and minor strontianite within carbonate strata [47]. Overall, these results indicate that strontium in the thermal mineral waters is mainly derived from the dissolution of celestine, with a subordinate contribution from strontianite. Metasilicic acid is classified into the third cluster, suggesting an independent material source. In thermal mineral waters, metasilicic acid is typically generated through the hydrolysis of silicate minerals and the direct dissolution of SiO₂-bearing phases. During recharge by atmospheric precipitation in the recharge area and the upper reaches of the runoff zone, low-temperature and low-mineralization infiltrating waters remain in prolonged contact with surrounding rocks, enabling the dissolution of quartz and the initial acquisition of metasilicic acid. Combined SEM observations (Figure 5) and X-ray fluorescence (XRF) data (Table 1) demonstrate that quartz is the dominant SiO₂-bearing mineral in the thermal reservoirs. In addition, major-element analyses of reservoir rocks show an average SiO₂ content of 12.56%. These results indicate that quartz dissolution overwhelmingly dominates the formation of metasilicic acid and represents its primary source in the thermal mineral waters.

By performing a Gibbs diagram analysis on the Shiqian thermal mineral water, one can preliminarily infer that its hydrochemical composition is dominantly controlled by the dissolution of soluble minerals during water–rock interaction. Further comparison of the molar ratios Mg²⁺/Na⁺, Ca²⁺/Na⁺, and HCO₃⁻/Na⁺—which reflect the weathering of carbonate rock, silicate rock, and evaporite lithologies, respectively—allows for a more detailed quantification of the contributions from these different lithologies. In addition, Correlation analysis of major anions and cations, combined with evaluations of water-sample similarity and grouping characteristics based on cluster dendrograms, indicates that strontium (Sr) and metasilicic acid (expressed as H₂SiO₃ or dissolved SiO₂) are characteristic components of the Shiqian thermal mineral waters. Their concentrations and material sources are significantly controlled by water–rock interactions, among which carbonate rocks play a dominant role in regulating the contents of Sr and metasilicic acid, followed by silicate rocks. In addition, geochemical processes such as rock weathering, mineral leaching, fluid mixing, and deep-seated CO₂ degassing also influence their abundances. According to Table 2, Sr concentrations in the Shiqian thermal mineral waters range from 0.06 to 7.17 mg/L, which falls within the typical interval for carbonate-type and sedimentary geothermal systems. The Sr is mainly derived from the dissolution of celestine (SrSO₄) associated with gypsum and, to a lesser extent, from strontianite (SrCO₃) within carbonate rocks [47]. By contrast, continental rift–type geothermal systems (e.g., the East African Rift) often have Sr concentrations in the range of 0.1–5 mg/L, indicating moderate-temperature, mixed carbonate–silicate water–rock interactions; while high-temperature magmatic geothermal systems (e.g., Yellowstone or some Icelandic fields) typically exhibit much lower Sr concentrations (<0.05 mg/L), due to interactions primarily with Sr-poor volcanic silicate minerals. According to Table 2, the metasilicic acid content in the Shiqian thermal mineral waters ranges from 19.46 to 65.51 mg/L, indicating a low to mid-temperature geothermal regime. This range is comparable to that of mid-temperature systems such as the East African Rift (20–150 mg/L) or Denizli/Pamukkale in Turkey (30–110 mg/L), reflecting the influence of silicate mineral weathering in moderate-temperature reservoirs. In contrast, high-temperature volcanic systems (e.g., Yellowstone or Iceland) often show much higher dissolved silica (150–450 mg/L or more), consistent with rapid silicate dissolution and long circulation paths under high-temperature conditions. Therefore, the relatively low-to-moderate Si content in Shiqian thermal waters suggests a moderate intensity of water–rock interaction, with a reservoir dominated by carbonate lithology and subordinate silicate contribution—a conclusion that is consistent with mineralogical evidence that quartz (SiO₂) is the primary Si-bearing phase in the reservoir [98].

5.3. Estimation of Geothermal Reservoir Temperature by Equilibrium Mineral Method and Judgment of Mineral-Fluid Chemical Equilibrium

The saturation index (SI) of minerals can reflect the reaction state between minerals and aqueous solutions, and it is one of the important parameters for exploring water–rock interactions [39]. When SI = 0, the mineral is in an equilibrium state; when SI > 0, the mineral is in a supersaturated state and tends to precipitate; when SI < 0, the mineral is in an undersaturated state and tends to dissolve, as shown in Figure 19. This study calculated the saturation indices of key minerals including quartz and gypsum. The results show that quartz is in a saturated state while gypsum is in a dissolved state in the thermal mineral water of the study area.

The multi-mineral equilibrium diagram can be used to judge the chemical reaction equilibrium state between hot water and minerals [99]. The method is to plot the function of mineral saturation index (SI) against temperature as curves. When multiple minerals converge simultaneously within a narrow temperature range and approach equilibrium, it can be considered that these minerals have reached dissolution equilibrium with the hot water. The corresponding mineral dissolution equilibrium temperature can then be regarded as the geothermal reservoir temperature. The PHREEQC software was used to calculate the mineral saturation indices of calcite, chalcedony, gypsum, quartz and amorphous silica at different temperatures. Curves depicting the relationship between saturation index (SI) and water temperature (T) were plotted for each mineral in the thermal mineral water. As shown in Figure 20, the saturation curves of some minerals intersect at a certain point, which indicates that these minerals coexist at this temperature, and this temperature is also the geothermal reservoir temperature here. The geothermal reservoir temperatures are approximately as follows: Guanyuliang Hot Spring (SQ01) at 80 °C, Shichang No.1 Hot Spring (SQ02) at 105 °C, Shichang No.2 Hot Spring (SQ03) at 81 °C, Kaixia River Cave Hot Spring (SQ04) at 98 °C, Shiqian Beita Geothermal Well (SQ05) at 100 °C, and Guochang Geothermal Well (SQ09) at 110 °C. The geothermal reservoir temperatures derived by the equilibrium mineral method show a relatively wide convergence range, which requires further estimation using geothermometers.

In geothermal systems, a fundamental prerequisite for the application of geothermometers is that certain solutes, gases used as geothermometers, and minerals in the geothermal reservoir achieve an equilibrium state [100]. In 1988, Giggenbach [101] proposed the Na-K-Mg ternary diagram method based on the contents of Na, K, Mg and their temperature-dependent reaction processes. This method uses relative content parameters, namely Na/1000, K/100 and Mg1/2, to determine the applicability of chemical geothermometers, calculate the temperature of deep reservoirs, and identify geothermal fluids that have reached equilibrium with surrounding rocks [102]. The Na-K-Mg ternary diagram is divided into three zones: the “nonequilibrium zone”, the “partial equilibrium zone” and the “complete reaction equilibrium zone”. As shown in Figure 21, the hydrochemical components of thermal mineral water in the study area all fall in the Mg corner and the “immature water” region of the ternary diagram, indicating that the overall degree of water–rock interaction in the study area is relatively low, the equilibrium state has not been achieved yet, and the dissolution process is still ongoing. Therefore, cation geothermometers are not suitable for estimating the geothermal reservoir temperature in the study area. Silica is a mineral widely distributed in the lithosphere and hydrosphere. When the temperature is below 300 °C, the dissolved silica in water is generally not affected by other ions or complexes, and thus is commonly used for geothermal temperature calculation [103]. Silica geothermometers include the quartz geothermometer and the chalcedony geothermometer, which are applicable to reservoir temperature calculation under high-temperature and low-temperature conditions, respectively [104]. Their calculation formulas are as follows:

(1). Quartz geothermometer-no steam loss, applicable range: 0–250 °C.

$$\mathrm{T}1={\displaystyle \frac{1309}{5.19-{\mathrm{l}\mathrm{g}\mathrm{S}\mathrm{i}\mathrm{O}}_2}}-273.15$$

(2). Amorphous silica geothermometer.

$$\mathrm{T}2={\displaystyle \frac{731}{4.52-{\mathrm{l}\mathrm{g}\mathrm{S}\mathrm{i}\mathrm{O}}_2}}-273.15$$

(3). Chalcedony geothermometer-no steam loss, applicable range: 0–250 °C.

$$\mathrm{T}3={\displaystyle \frac{1032}{4.69-{\mathrm{l}\mathrm{g}\mathrm{S}\mathrm{i}\mathrm{O}}_2}}-273.15$$

The calculation results are presented in Table 6. It can be seen from the table that the estimated values obtained by the amorphous silica geothermometer (T2) are negative, which are inconsistent with the actual conditions and thus discarded. The calculated values of the chalcedony geothermometer (T3) are only slightly higher than the measured wellhead temperatures, and even lower at some sampling points, so they are also excluded. Based on the mineral saturation index of quartz, quartz is in a saturated state. Therefore, the calculated values from the quartz geothermometer (T1) are adopted as the geothermal reservoir temperatures of the study area.

5.4. Numerical Simulation of Water–Rock Interaction

Numerical simulation of water–rock interaction is an important tool for deepening our understanding of these processes; it can identify complex reactions between groundwater and various minerals and enables quantitative analysis under reasonable conditions [105]. In this study, the hydrogeochemical software PHREEQC (Version 2.8) was used for inverse modeling to analyze mineral dissolution-precipitation processes and water–rock interactions of thermal waters along the flow path. The model is based on mass and charge conservation and couples the thermodynamic equilibrium of mineral dissolution–precipitation during fluid transport. An improved Newton–Raphson iterative method is employed to solve a system of equations involving water activity, mineral dissolution equilibria, and elemental mass balance, thereby quantifying the mass transfer of each mineral phase [106,107]. To effectively elucidate the water–rock reaction processes in the study area, the initial and terminal water sample data along the flow path, together with uncertainty coefficients for potentially reactive mineral phases, were input into the inverse modeling module of PHREEQC. Given that carbonate dissolution is closely related to CO₂, CO₂(g) and H₂O were included as potential reactants to simulate the effect of CO₂ partial pressure on equilibrium in the carbonate system [50]; the reactions considered are listed in

Table 7. Reaction phase used in phreeqc.

Phase	Formula	Reaction
H2O(g)	H2O(g)	H2O(g) = H2O(a)
CO2(g)	CO2(g)	CO2(g) = CO2(a)
Calcite	CaCO3	CaCO3 = CO32− + Ca2+
Dolomite	CaMg(CO3)2	CaMg(CO3)2 = Ca2+ + Mg2+ + 2CO32−

. To ensure the specificity and rationality of the simulation, and based on the geological background of the study area, mineral saturation indices, mineral–fluid chemical assessments, and hydrochemical data, the following boundary and prior conditions were set: the simulation domain follows the natural flow path of the thermal waters, which is controlled by regional tectonics. The Hongshi Fault exhibits a pronounced water-blocking property, forming a natural hydraulic barrier that constrains the movement of thermal waters; therefore, the starting boundary was set at the higher-elevation Guanyuliang Hot Spring (SQ01) and the terminal boundary at the lower-elevation Chengbei Hot Spring (SQ07) (Figure 22). Measured hydrochemical data from the start and end samples were used to constrain the initial and final states of the reaction system. Based on XRD and SEM analyses of reservoir rocks, the reactive mineral phases were limited to calcite, dolomite, and quartz. The uncertainty coefficient for mineral reactions was set to 0.02.

The path selected for this PHREEQC inverse simulation is located in the middle and lower reaches of the runoff system or near the discharge area. The simulation results are shown in

Table 8. Results of inverse hydrogeochemical modeling for the major hydrochemical components of geothermal water in the study area.

Sample Number	Confidence Level	Model	Calcite (mmol/L)	Dolomite (mmol/L)	Quartz (mmol/L)	CO2(g)	H2O(g)
SQ01(initial sample) SQ07(terminal sample)	98%	1	6.834 × 10−4	−9.003 × 10−5	−1.200 × 10−4	-	1.235 × 101
		2	8.888 × 10−4	−3.105 × 10−4	−2.403 × 10−4	−7.795 × 10−4	-
		3	6.997 × 10−4	−9.003 × 10−5	-	1.4733 × 10−4	1.235 × 101

Note: Positive values represent mineral dissolution, and negative values (“−”) represent mineral precipitation.

. As listed in Table 8, all three models can well reflect the hydrochemical processes along the flow path, with calcite dissolution and dolomite precipitation occurring in each model, and quartz precipitation occurring in Models 1 and 2. In Model 1, the calcite dissolution amount is 6.834 × 10⁻⁴ mmol/L, and the dolomite and quartz precipitation amounts are 9.003 × 10⁻⁵ mmol/L and 1.200 × 10⁻⁴ mmol/L, respectively. In Model 2, the calcite dissolution amount is 8.888 × 10⁻⁴ mmol/L, and the dolomite and quartz precipitation amounts are 3.105 × 10⁻⁴ mmol/L and 2.403 × 10⁻⁴ mmol/L, respectively. In Model 3, the calcite dissolution amount is 6.997 × 10⁻⁴ mmol/L, and the dolomite precipitation amount is 9.003 × 10⁻⁵ mmol/L. The thermal mineral water in the study area is mainly recharged by atmospheric precipitation infiltration from the high-altitude areas on the southwestern foot of Fanjing Mountain. In the initial infiltration stage, the low-temperature, low-ion-concentration infiltrating water contacts the surrounding rocks and has a strong dissolution capacity. At this stage, dolomite may undergo preliminary release, and minerals such as quartz dissolve slowly under water–rock interaction, providing initial solutes for the water body. During the migration of thermal mineral water along the runoff path, the water body gradually warms up and undergoes a long-term water–rock interaction process, leading to a relative increase in ion concentrations in the solution, which approach or reach the saturation state of minerals. When the water body enters the middle-lower reaches of the runoff area or the discharge area (the model simulation interval), due to changes in thermodynamic conditions, dolomite and quartz tend to be supersaturated, thus exhibiting precipitation phenomena. Combined with SEM and major element analysis of the reservoir rock, this study reveals that dolomite is abundant and prone to reach saturation and precipitate, while quartz, despite its low solubility, also readily approaches saturation under water–rock interaction conditions. Calcite, being less abundant in the reservoir than dolomite, tends to remain in a net dissolving state along the flow path. This result is consistent with previous inverse-modeling studies in carbonate geothermal systems. For instance, in the Wumishan Formation dolomite reservoir, PHREEQC-based inverse modeling has quantified calcite and dolomite dissolution/precipitation behaviors, revealing characteristic reaction pathways under reinjection-type geothermal conditions [108]. In another study in a German sandstone/natural fluid geothermal system, inverse modeling highlighted the suitability of calcite as a reactive phase, emphasizing that selection of mineral reaction phases should be guided by mineralogical data (XRD / SEM) [109]. Furthermore, reactive-transport studies in carbonate reservoirs using models like PHREEQC or IPHREEQC have shown that surface complexation on calcite surfaces can significantly influence dissolution and precipitation dynamics [110]. Therefore, our inverse-modeling results for the SW-directed flow path in the Shiqian system namely alternating calcite dissolution and dolomite/quartz precipitation offer strong evidence that the dominant source of geochemical constituents in the thermal water is carbonate rock dissolution, emphasizing the primary control of a carbonate-dominated reservoir on the hydrochemical evolution of the system.

5.5. Genesis and Evolutionary Characteristics of the Shiqian Hot Spring Group

Integrating thermal-reservoir characteristics, hydrogeochemical signatures, H–O stable isotope evidence, fault interpretation, and water–rock reaction modeling within the regional geological context, we constructed a genetic model for the Shiqian hot spring system (Figure 23). Based on DEM data and hydrogen–oxygen isotope analyses, the thermal mineral water is recharged by atmospheric precipitation, with recharge areas located on the southwestern flank of Mount Fanjing to the northeast, at elevations of 911–1833 m. Under gravity, precipitation infiltrates from carbonate bedrock outcrops through fractures, pores, karst conduits, and fault damage zones, and then flows from high to low elevations under hydraulic head. Upon encountering the Hongshi and Shiqian faults—large transpressional structures with closely spaced linear folds, well-developed fault planes, and subsidiary structures—the lateral permeability is effectively impeded, forming a boundary system that confines thermal water to flow along fault strike. Consequently, thermal waters migrate within carbonate reservoirs along the NNE-trending Shiqian Fault and NE-trending Hongshi Fault, with an overall flow from northeast to southwest, discharging as springs where the two faults intersect in the southern Shiqian and Zhongba areas. Groundwater is heated by terrestrial heat flow during its deep circulation along structural fractures and undergoes water–rock interactions with surrounding rocks. Based on the analysis of regional thermal reservoir mineral composition and hydrochemical characteristics, Ca²⁺ and Mg²⁺ in the thermal mineral waters of the study area are mainly derived from the dissolution of carbonate rocks, Sr²⁺ is primarily sourced from the dissolution of celestine associated with gypsum and strontianite, and H₂SiO₃ mainly originated from the dissolution of quartz. As also indicated in Table 6 and Figure 20, the thermal reservoir temperatures of hot springs in the area are mostly below 100 °C. At such temperatures, the saturation index of celestine is less than 0, indicating a dissolved state, which further confirms that strontium is mainly derived from the dissolution of undersaturated celestine minerals. PHREEQC inverse simulation (corresponding to the segment from the middle-lower reaches of the runoff to the discharge area) revealed the net mineral reactions occurring along this path segment. The results show that within the simulation interval, calcite exhibits dissolution, while dolomite and quartz show precipitation. This result characterizes the comprehensive water–rock interactions along the path segment. Therefore, it is inferred that the enrichment of H₂SiO₃, a characteristic component in the thermal mineral waters, mainly occurs during the upstream and deep circulation processes prior to the simulation segment, achieved through the dissolution of minerals such as quartz. The observed net precipitation of dolomite and quartz in the simulation segment reflects the supersaturated state of these minerals under specific geochemical conditions in the late stage of runoff. The thick regional sedimentary cover and the distribution of thermal waters indicate that the primary heat supply is transferred upward along deep NNE–NE faults acting as thermal conduits. Meanwhile, meteoric and shallow cold waters percolate downward through fault zones and paleo-karst surfaces within aquifers, acquiring heat during deep circulation. The heated water becomes buoyant and ascends. In addition, the results of water–rock reaction modeling show that there are differences in the mineral reaction rates of thermal mineral water along the simulated runoff path. The calcite dissolution amounts in Models 1, 2, and 3 reach 6.834 × 10⁻⁴ mmol/L, 8.888 × 10⁻⁴ mmol/L, and 6.997 × 10⁻⁴ mmol/L, respectively. The precipitation amounts of dolomite and quartz in Model 1 are 9.003 × 10⁻⁵ mmol/L and 1.200 × 10⁻⁴ mmol/L, respectively; in Model 2, the precipitation amounts are 3.105 × 10⁻⁴ mmol/L and 2.403 × 10⁻⁴ mmol/L, respectively; and in Model 3, the dolomite precipitation amount is 9.003 × 10⁻⁵ mmol/L. This indicates that the intensity of water–rock interaction of the fluid in this segment is controlled by local geochemical and dynamic conditions, respectively—indicative of strong dynamic forcing of fluid transport. Upflow is governed by the combined effects of buoyancy-driven free convection and tectonically forced convection, which exchange heat with continuously infiltrating cold water to establish a deep convective system. As the thermal waters rise, they enter carbonate-hosted reservoirs and are effectively sealed by overlying impermeable caprocks composed of the Meitan and Dawan formations up to the Silurian Hanjiaodian Group (shales, sandstones), allowing accumulation in structural highs such as anticlinal cores. Ultimately, the deep thermal waters ascend along fault damage zones and discharge naturally at topographic lows with reduced hydrostatic pressure (e.g., river valleys) as hot springs, or are released via drilled geothermal wells.

5.6. Analysis of the Water Quality Characteristics and Resource Attributes of the Shiqian Hot Spring Group

Hot spring water possesses unique physicochemical properties due to its content of various minerals and trace elements, and suitable water quality conditions are one of the foundations for maintaining the specific uses of the water body [49]. In China, hot springs have a long history of being used for bathing and leisure [52]. In recent years, with the advancement of exploration and development of hot spring resources, their water quality characteristics and resource potential have received increasing attention. Strontium, metasilicic acid, calcium, magnesium, potassium, sodium, bicarbonate, and sulfate are common ions and components in hot spring water. The hydrochemical characteristics of these components provide a scientific basis for the resource classification and utilization direction of hot springs. According to relevant water quality standards, when the content of these components reaches a certain threshold, the water body can be classified as a specific type of natural mineral water or therapeutic thermal mineral water. For example, the contents of strontium and metasilicic acid are important indicators in Chinese national standard GB 8537-2018 for Drinking Natural Mineral Water. The Shiqian area is rich in geothermal resources with numerous and widely distributed hot spring outcrops, and is known as the “City of Springs” [67]. Data from this study show that the thermal mineral water in Shiqian has high contents of trace elements such as strontium and metasilicic acid. Referring to Chinese national standard GB 8537-2018 for Drinking Natural Mineral Water, except for SQ03, the strontium content of most other hot springs (geothermal wells) has met the naming standard for strontium-rich mineral water; except for SQ03, SQ06, and SQ07, the metasilicic acid content of most other sampling points has also met the naming standard for metasilicic acid-rich mineral water. In addition, the metasilicic acid content of SQ02 hot spring meets both the above-mentioned drinking mineral water standard and the relevant content reference values for GB/T 11615-2010 Specification for Exploration of Geothermal Resources (“Appendix E Hydrotherapy Thermal Mineral Water Quality Standards”).

In terms of resource endowment, the thermal mineral water in the Shiqian area is characterized by suitable water temperature, abundant resources, and high content of trace elements. Among them, the contents of strontium and metasilicic acid in many places meet the standards for drinking natural mineral water, and some points simultaneously meet the relevant content reference values for therapeutic thermal mineral water, which provides a water quality basis for its development as drinking natural mineral water and therapeutic hot water resources. The rational development and utilization of thermal mineral water in this area can provide a resource foundation for the development of local mineral water industry and tourism industry, and has certain social and economic value.

6. Conclusions

In this study, the Shiqian Hot Spring Group was selected as the research object. A total of seven fresh rock samples and thirteen thermal mineral water samples were collected within the study area. Multiple technical approaches were employed, including petrogeochemistry, hydrogeochemistry, isotope hydrology, digital elevation model (DEM) data analysis, remote sensing interpretation, geological survey, mineral saturation index calculation, and inverse hydrogeochemical modeling, to systematically investigate the hydrogeochemical characteristics and genetic mechanisms of the hot spring group in this area. The main conclusions are as follows:

• The thermal mineral waters of Shiqian are hosted within the carbonate strata from the Qingxudong Formation to the Honghuayuan Formation, which constitute the second thermal reservoir in Guizhou Province. The lithology is dominated by dolomite, limestone, and siliceous dolomite, with mineral components mainly comprising dolomite, calcite, and quartz. The principal chemical constituents of the thermal reservoir rocks are CaO, MgO, and SiO₂.
• The dominant hydrochemical types of the thermal mineral waters are HCO₃·SO₄-Ca·Mg and HCO₃-Ca·Mg, followed by HCO₃-Na type. The hydrochemical composition is mainly controlled by meteoric precipitation and rock weathering, with carbonate mineral dissolution as the primary process, followed by silicate weathering. Based on the results of mineral saturation index calculation and PHREEQC inverse hydrogeochemical numerical modeling, the net mineral reaction processes within the selected path segment were quantified. The simulation results show that calcite dissolution and dolomite and quartz precipitation are the dominant mineral reactions in the simulation interval. This finding further reveals the specific characteristics of water–rock interaction during the local stage of thermal mineral water migration and supports the analysis of the hydrogeochemical evolution path in this area.
• Integrated analysis of hydrogen–oxygen stable isotope compositions, DEM data, remote sensing interpretation, and geological background reveals that the thermal mineral waters are recharged by meteoric precipitation, with recharge elevations ranging from 911 m to 1833 m. The DEM shows that recharge zones are located on the southwestern foot of Fanjingshan. Combined with remote sensing interpretation of geological structures and regional geological background analysis, the thermal mineral waters flow generally from northeast to southwest along the Shiqian Fault and the Hongshi Fault, ultimately discharging as springs in low-lying terrain, river valleys, or gullies under the control of regional geomorphology and geological structures.
• Integrating evidence from ion correlation analysis, cluster analysis, and petrographic and mineralogical investigations, the chemical composition of the thermal mineral waters in the study area is shown to be primarily controlled by water–rock interactions. Hydrochemical characteristics indicate that the enrichment of Sr is closely associated with the dissolution of strontium-bearing minerals, such as celestine, within carbonate rocks and gypsum-salt layers, and that Sr shares a common material source with ions in the second group (K⁺, Ca²⁺, Mg²⁺, SO₄²⁻, and Sr²⁺). In contrast, metasilicic acid exhibits an independent material source and is mainly derived from the dissolution of quartz in the upstream segment of the groundwater flow path.
• The recharge source of the thermal mineral waters is meteoric precipitation from the southwestern foot of Fanjingshan. After infiltrating through local fractures, pores, and karst conduits, the water is stored in the primary thermal reservoir composed of Cambrian Qingxudong Formation to Ordovician Honghuayuan Formation carbonates. During deep circulation, the groundwater is heated by terrestrial heat flow and continuously extracts mineral components from the surrounding reservoir rocks, ultimately forming mineral-rich thermal waters. Water quality analyses indicate that the concentrations of characteristic components such as strontium and metasilicic acid in the thermal mineral waters meet the relevant reference values of water-quality standards, demonstrating their potential for development as regional mineral water and therapeutic thermal water resources.