Mineral Exploration in the Central Xicheng Ore Field, China, Using the Tectono-Geochemistry, Staged Factor Analysis, and Fractal Model

Abstract

As China’s third-largest lead–zinc ore field, the Xicheng Ore Field has significant potential for discovering concealed deposits. In this study, a tectono-geochemical survey was conducted, and 1329 composite samples (comprising 5614 subsamples) were collected from the central part of the field. The dataset was analyzed using staged factor analysis (SFA) and concentration–area (C–A) fractal model. Four geochemical factors were extracted from centered log-ratio (CLR)-transformed data: F_2-1 (Ag–Pb–Sb–Hg), F_2-2 (Mo–Sb–(Zn)), F_2-3 (Au–Bi), and F_2-4 (W–Sn). Known Pb–Zn deposits coincide with positive F_2-1 and negative F_2-2 anomalies, as identified by the C–A fractal model, suggesting these factors are reliable indicators of Pb–Zn mineralization. Five Pb–Zn exploration targets were delineated. Statistical analysis and anomaly maps for F_2-3 and F_2-4 also indicate the potential for Au and W mineralization. Notably, some anomalies from different factors spatially overlap, indicating the possibility of epithermal Pb–Zn mineralization at shallow depths and mesothermal to hyperthermal Au and W mineralization at great depths. Overall, the integration of tectono-geochemistry, targeted and composite sampling, SFA, and C–A fractal modeling proves to be an effective and economical approach for identifying and enhancing ore-related geochemical anomalies.

1. Introduction

The Xicheng Ore Field has a long history of exploration and development. Between the 1960s and 1980s, major discoveries included the Changba–Lijiagou giant lead–zinc deposit, the Bijiashan large lead–zinc deposit, and many small- to medium-sized deposits [1]. However, the widespread distribution of deposits and mining activities has reduced the effectiveness of soil geochemical surveys for exploring buried deposits. In contrast, lithogeochemical methods, based on primary halos, are less affected by anthropogenic contamination and are the primary approach for geochemical exploration in bedrock-exposed areas [2,3,4]. Traditional lithogeochemistry typically involves collecting one sample per grid via regular spacing or along exploration lines. The approach is effective for detecting shallow buried or outcropping ores with strong, large-scale surface anomalies [5,6,7]. However, its limited sample representativity reduces its effectiveness in identifying anomalies related to deeper deposits [8,9,10,11]. Surface anomalies from deeply buried deposits are often subtle and localized, commonly associated with permeable structures such as faults, fractures, and veins. To address this, the tectono-geochemistry method, a regional geochemical approach designed for exploring buried magmatic and hydrothermal deposits in bedrock areas [11,12,13,14], targets materials like fissure fillings, fault gouges, mineralized or altered rocks, and veins to better capture deep-sourced anomalies [8,9,13]. Unlike traditional rock geochemical surveys, which typically collect one sample per grid, the tectono-geochemistry method gathers 3–6 samples per grid to create composite samples. This improves representativeness while reducing analytical costs. Comparative studies conducted as early as the 1960s demonstrated that tectono-geochemistry is significantly more effective and enhancing anomalies related to deeply buried deposits [8,9].

Over 30 ore deposits have been discovered at shallow depths in the central Xicheng Ore Field, and current efforts focus on exploring concealed deposits. Intense tectonic and magmatic activity associated with mineralization has produced well-developed leakage halos, making the area well-suited for applying the tectono-geochemistry method. In this study, a 1:50,000 tectono-geochemistry survey was conducted to investigate buried deposits. Staged factor analysis (SFA) on centered log-ratio (CLR)-transformed data was used to extract ore-related elemental associations, whose significance was interpreted based on classical hydrothermal zoning patterns. Thresholds of these factors were determined using the concentration–area (C–A) fractal model, and exploration targets were delineated based on identified anomalies.

2. Regional Geology

The Xicheng Ore Field, with proven Pb + Zn reserves exceeding 17 million tons, is the third-largest lead–zinc ore field in China. It is located in the western part of the West Qinling metallogenic belt (Figure 1) [15,16,17], bounded by the northern EW-trending Huangzhuguan Fault, the southern NEE-trending Rentushan–Luoba Fault, and the western NNE-trending Shanjiahe Fault, forming a wedge-shaped structural block [15,17]. This study focuses on the central part of the Xicheng Ore Field. The exposed strata in the study area mainly include the Upper Silurian Wujiashan (S₃w), Middle Devonian Anjiancha (D₁a), Middle Devonian Huangjiagou (D₂h), Middle–Upper Devonian Honglingshan (D_2-3hl), Upper Devonian Shuanglanggou (D₃sl), Triassic Longwuhe (TL), Middle Jurassic Longjiagou (J₂l), Lower Cretaceous Jishan (K₁js), and Neogene Gansu Group (NG) (Figure 2). The TL, J₂l, and K₁js formations are distributed in the southern part of the study area. Devonian strata, consisting of clastic and carbonate rocks, are dominant. The NG is sporadically exposed in the southeastern corner. Magmatic rocks belong to the Caoguan–Yushuba magmatic–tectonic belt, and they include Late Triassic granite and granodiorite, such as the Huangzhuguan granodiorite and the Changba and Shalipo monzogranites. Numerous granite veins are also present. Multiple tectonic events formed the Xihanshui fault–fold belt. The Wujiashan complex anticline is the dominant fold, accompanied by several secondary folds developed on its flanks [17]. Faults are primarily categorized into NE-trending, NW-trending, and nearly EW-trending shear fracture zones, with major structures including the Rentushan–Luoba, Wujiashan, and Huangzhuguan faults.

Pb and Zn mineralization dominates the area, with large-to-super-large lead–zinc deposits such as the Changba, Lijiagou, and Bijiashan. Smaller-scale Fe, Cu, Au, W, and Mo mineralizations are also present. The lead–zinc deposits in the Xicheng Ore Field likely formed through Devonian SEDEX-style processes, later modified by Triassic metamorphism and magmatic–hydrothermal overprinting, and tectonic uplift [1,18,19,20,21]. Late-stage geological and geochemical processes have transformed the ore bodies in several ways. (1) Structural alteration: Originally, horizontal stratabound ore bodies were modified into steeply inclined forms, with some exposed at the surface due to tectonic denudation (Figure 3a). (2) Metal remobilization and enrichment: Ore metals were remobilized and reprecipitated, thickening and upgrading ore bodies, particularly in fold-hinge zones (Figure 3b,c). (3) Hydrothermal overprinting: Intense tectonic–hydrothermal activity disrupted ore body contiguity and remobilized ore-related elements, forming new ore bodies and leakage halos along permeable structures such as faults or fractures (Figure 3c).

3. Samples and Methods

3.1. Sample Collection and Analysis

As a regional lithogeochemical method, tectono-geochemistry aims to identify and extract potential ore-related geochemical signatures in rock outcrop areas using targeted sampling media and a composite sampling approach. Before conducting field sampling, structural data of the study area must be collected, and then sample collection must be carried out in combination with the actual field investigation. Sampling layout and numbering scheme can be found in [11]. The method prioritizes the collection of altered and mineralized rocks and veins; when these are unavailable, unaltered rocks are collected to represent background conditions. To improve representativeness and reduce costs, 3~6 subsamples are collected per sampling grid and combined into a single composite sample for analysis. In this survey, the study area covers approximately 412.6 km², with sampling conducted on a 500 m × 500 m grid. A total of 5614 subsamples were collected and combined into 1329 composite samples. Sample locations are shown in Figure 4. The subsamples include 2454 altered rocks, 2143 fissure fillings (or coatings), 641 unaltered rocks, 319 rocks from structurally fractured zones, and 57 mineralized rocks. Each composite sample, weighing over 150 g, was crushed and ground to below 200 mesh. 19 elements (Au, Ag, As, Bi, Cu, Pb, Zn, Cd, Hg, B, Sb, W, Sn, Mo, Ba, Co, Cr, Mn, and Ni) were analyzed at the Experimental Testing Center of No.1 Institute of Geology and Mineral Exploration, Gansu Bureau of Geology and Mineral Exploration and Development, Tianshui, China. Analysis methods and detection limits are provided in

Table 1. Analytical methods and detection limits for elements analyzed in the tectono-geochemistry survey of the central Xicheng Ore Field.

Elements	Analysis Methods	Units	Detection Limits
Au	Inductively coupled plasma–mass spectrometry	ng/g	0.1
Ag	Emission spectrometry	μg/g	0.02
As	Atomic fluorescence spectroscopy	μg/g	0.2
Bi	Inductively coupled plasma–mass spectrometry	μg/g	0.05
Cu	X-ray fluorescence spectroscopy	μg/g	0.9
Pb	X-ray fluorescence spectroscopy	μg/g	2.0
Zn	X-ray fluorescence spectroscopy	μg/g	1.0
Cd	Inductively coupled plasma–mass spectrometry	μg/g	0.05
Hg	Cold vapor atomic fluorescence spectroscopy	ng/g	0.3
B	Emission Spectrometry	μg/g	1
Sb	Atomic fluorescence spectroscopy	μg/g	0.05
W	Inductively coupled plasma–mass spectrometry	μg/g	0.2
Sn	Emission Spectrometry	μg/g	0.6
Mo	Emission Spectrometry	μg/g	0.2
Ba	X-ray fluorescence spectroscopy	μg/g	18
Co	X-ray fluorescence spectroscopy	μg/g	0.8
Cr	X-ray fluorescence spectroscopy	μg/g	2.5
Mn	X-ray fluorescence spectroscopy	μg/g	10
Ni	X-ray fluorescence spectroscopy	μg/g	1.0

.

3.2. Staged Factor Analysis

Factor analysis is a widely used multivariate technique for interpreting geochemical data by revealing underlying patterns in large, complex compositional datasets through linear factors [24,25,26,27]. Staged factor analysis, introduced by Yousefi et al. [28,29], can progressively remove noisy or insignificant elements to isolate meaningful multielement signatures [29,30]. It begins with an initial factor analysis on the full dataset to extract meaningful factors, and then it eliminates low-significance elements and repeats the analysis on the reduced dataset, resulting in clearer and more interpretable factors. This method has been effectively applied in geochemical data processing [26,27,31,32,33,34]. In this study, lithogeochemical data were CLR-transformed using the CoDaPack 2.02.21 software [35,36,37], and staged factor analysis was conducted using SPSS 23 software. Factors with eigenvalues greater than one were extracted using the principal component extraction method, followed by varimax rotation to improve interpretability. A loading threshold of |0.55| was used to identify significant elements, with those values shown in bold for each factor.

3.3. Concentration–Area (C–A) Fractal Model

Cheng et al. [38] proposed the concentration–area (C–A) fractal model, which relates the element concentration to the area enclosed by concentration contours through a power-law relation. As an early advancement in the fractal analysis of geochemical data, the C–A fractal model has been widely applied to identify geochemical anomalies in complex geological settings [39,40,41,42,43,44,45,46,47]. The model is defined as follows [38]:

A(ρ ≤ ν)∝ρ^−D1; A(ρ > ν)∝ρ^−D2

where A(ρ) represents the area with concentration values greater than the contour value ρ; υ is the threshold; and −D₁ and −D₂ are fractal dimensions derived from the slopes of the straight line segments in the log–log plot of A (ρ) versus ρ. Breaks between segments and their corresponding ρ values are used to define anomaly thresholds.

4. Results

4.1. Statistical Analysis

Statistical parameters for 12 lithogeochemical elements are present in

Table 2. Statistical parameters of lithogeochemical data from the tectono-geochemistry survey in the central Xicheng Ore Field.

Element	Min	Mean	Max	STD	Skewness	Kurtosis	UCC	CV	EC
W (μg/g)	0.05	6.22	5222	146	34.9	1246	2.00	23.4	3.11
Pb (μg/g)	1.90	64.9	49,089	1353	35.9	1302	17.0	20.9	3.82
Bi (μg/g)	0.09	0.53	227	6.58	31.7	1066	0.13	12.4	4.18
Zn (μg/g)	2.70	138	41,654	1403	22.9	614	71.0	10.1	1.95
Cd (μg/g)	0.01	0.44	99.5	4.01	21.1	482	0.10	9.07	4.51
Ag (μg/g)	0.03	0.19	40.2	1.39	23.9	621	0.05	7.43	3.73
Au (ng/g)	0.10	8.90	1624	60.6	18.6	428	1.80	6.81	4.94
Hg (ng/g)	6.00	48.6	6019	289	15.0	253	56.0	5.94	0.87
Cu (μg/g)	1.90	21.0	3208	102	26.1	762	25.0	4.86	0.84
Sb (μg/g)	0.44	1.17	56.24	1.91	21.1	556	0.20	1.63	5.86
Mo (μg/g)	0.25	3.39	41.1	3.05	2.96	21.2	1.50	0.90	2.26
Sn (μg/g)	0.40	2.23	18.2	1.39	3.34	26.9	5.50	0.62	0.41

Notes: STD—standard deviation; CV—coefficient of variation (CV = SD/mean); UCC—upper crustal concentration; EC—enrichment coefficient (EC = mean/UCC). The UCC value for Hg is from [48]; all other element values are from [49].

, and their boxplots are shown in Figure 5. Elements with coefficient of variation (CV) and enrichment coefficient (EC) values > 3, namely W, Pb, Bi, Au, Ag, and Cd, are likely dominant ore or associated elements (Figure 6). The boxplots also reveal numerous extreme upper outliers for these elements. Notably, the high CV and EC values for Pb and Zn align with their status as the primary ore metals in the study area. Zinc, in particular, shows an exceptionally high CV (10) but a relatively low EC (1.95). Its maximum content is 4.17%, significantly exceeding its cutoff grade. Au and W also warrant attention, with maximum contents of 0.52% and 1.62 g/t, respectively. Frequency distribution histograms for the potential ore and associated elements (Figure 7) show that the raw data deviate from normality, with extreme values leading to a highly skewed distribution. In contrast, the CLR-transformed data are closer to a normal distribution.

4.2. Single-Element Geochemical Signatures

To better illustrate the spatial distribution of ore-related elements, anomaly maps were generated for Pb, Zn, Ag, Au, W, and Bi elements with high CV and EC values. Thresholds were determined using the boxplot method (threshold = upper quartile + 1.5 × [upper quartile − lower quartile]) and are listed in

Table 3. Thresholds of potential ore-related elements determined by boxplots.

Element	Lower Quartile	Upper Quartile	Threshold	Number of Anomalous Samples
Pb (μg/g)	9.0	21.9	41.3	80
Zn (μg/g)	16.0	61.7	130	51
Ag (μg/g)	0.05	0.13	0.26	93
Au (ng/g)	1.3	3.5	6.8	151
W (μg/g)	0.47	1.29	2.52	65
Bi (μg/g)	0.15	0.26	0.43	106

. The resulting anomaly maps are shown in Figure 8.

Pb, Zn, and Ag anomalies are primarily associated with Lower Devonian formations (D₁a) and adjacent units, such as the Silurian Wujiashan (S₃w), Upper Devonian Shuanglanggou (D₃sl), and Middle Devonian Huangjiagou (D₂h). The spatial distributions of Pb and Ag anomalies largely overlap, though they differ in intensity and scale. While Pb and Zn (Cd) are typically considered closely associated, their anomaly maps reveal notable differences—for instance, a strong Pb anomaly in the southwest is not accompanied by a Zn anomaly, and a southeastern Zn anomaly corresponds to only a small-scale Pb anomaly. Au anomalies occur across multiple formations, with three large-scale, high-intensity anomalies located in the D₁a, S₃w, and D₃sl units. The spatial pattern of Au anomalies differs from that of Ag; notably, strong Au anomalies in the western and southeastern areas are not accompanied by Ag anomalies. Zones where Au and Ag coexist may indicate the presence of electrum. Bi anomalies show a mixed pattern. Some coincide with Au anomalies, while others align with W anomalies—for example, in the southwestern region and near the Changba–Lijiagou area. W anomalies can be grouped into two types: (i) western and eastern anomalies associated with Triassic granodiorite porphyry (γδμT), monzogranites (nγT), and granodiorite (γδT); and (ii) northern and southern anomalies likely linked to different geological processes. The northern W anomaly, lacking volatile element associations, may be due to concealed granitic cupolas beneath the Devonian cover. The southern W anomaly, accompanied by Hg, Sb, and B anomalies, likely reflects epithermal fluid activity. The Au-Bi-W association reflects a geochemical signature linked to magmatic–hydrothermal processes and granite-related gold mineralization, distinct from SEDEX or orogenic systems. High-temperature zones near intrusions are marked by W-Bi-Au anomalies, while distal zones often show As-Sb enrichment. Deeper regions may exhibit Cu-Au-Ni-W-Bi associations, with shallower areas showing Hg-Pb-Zn-As-Sb signatures [50,51,52,53]. Finally, the Changba–Lijiagou mining area is marked by weak, small-scale Pb, Zn, and Ag anomalies, but strong, large-scale Au, Bi, and W anomalies.

4.3. Multi-Element Geochemical Signatures and Target Delineation

Two stages of factor analysis were conducted to extract clean, interpretable factors. In the first stage, Cd and Cu had absolute loadings below 0.55 in all factors and were excluded. The second stage was then performed on the remaining 10 elements. Varimax orthogonal rotation produced a four-factor model explaining 67% of the total variance (

Table 4. Rotated component matrix from staged factor analysis: Loadings in bold indicate selected factors with absolute values ≥ 0.55.

First Stage					Second Stage (Cd and Cu Are Excluded)
Element	F1-1	F1-2	F1-3	F1-4	Element	F2-1	F2-2	F2-3	F2-4
Pb	0.66	0.35	0.17	−0.01	Pb	0.73	−0.31	0.04	0.04
Zn	0.08	0.73	−0.46	−0.27	Zn	0.08	−0.72	−0.42	−0.31
Cd	0.47	0.40	−0.33	−0.19	Sb	0.63	0.58	0.03	0.14
Sb	0.64	−0.52	0.16	0.05	Hg	0.72	0.02	−0.19	−0.09
Hg	0.75	0.01	−0.14	−0.08	Ag	0.67	−0.13	0.48	0.17
Ag	0.57	0.21	0.59	0.02	Au	−0.14	0.07	0.77	−0.33
Mo	−0.09	−0.79	−0.10	−0.22	Bi	0.11	0.16	0.64	0.40
Au	−0.20	−0.10	0.65	−0.31	W	−0.39	−0.07	−0.06	0.67
Bi	0.10	−0.11	0.66	0.28	Sn	0.26	0.03	0.04	0.71
W	−0.29	0.03	−0.10	0.73	Mo	−0.19	0.79	−0.03	−0.23
Sn	0.27	0.003	0.19	0.66	Eigenvalue	2.19	1.64	1.45	1.43
Cu	−0.548	0.01	0.02	−0.37	% of Variance	21.94	16.39	14.48	14.30
Eigenvalue	2.47	1.78	1.65	1.45	Cumulative %	21.94	38.34	52.82	67.12
% of variance	20.54	14.84	13.76	12.07
Cumulative %	20.54	35.38	49.14	61.21

; Figure 9).

• F_2-1 (22%) shows strong positive loadings for Pb, Ag, Sb, and Hg.
• F_2-2 (16%) shows strong positive loadings for Mo and Sb, and a negative loading for Zn.
• F_2-3 (14%) has positive loadings for Au and Bi.
• F_2-4 (14%) shows strong positive loadings for W and Sn.

Factor loadings and scores are visualized in Figure 9b–d. Among elements with high loadings, the smallest angle is between Pb and Sb (Figure 9c), followed by Hg, Ag, and Pb (Figure 9b); W and Sn; and Au and Bi (Figure 9d), indicating strong correlations among these element pairs.

Thresholds for each factor were determined using the C–A fractal model. Four straight lines were fitted to the ln–ln C–A plot using the least squares method, resulting in five geochemical populations: low background, high background, weak anomaly, moderate anomaly, and strong anomaly (Figure 10). The corresponding thresholds are listed in

Table 5. Anomaly thresholds of factors derived using the C–A fractal model in the central Xicheng Ore Field.

Factor	Background		Anomaly
	Low	High	Weak	Moderate	Strong
F2-1 (positive)	0–0.05	0.05–0.13	0.13–0.35	0.35–1.1	1.1–7.6
F2-2 (negative)	0–−0.03	−0.03–−0.14	−0.14–−0.38	−0.38–−0.78	−0.78–−4.4
F2-3 (positive)	0–0.01	0.01–0.08	0.08–0.27	0.27–0.84	0.84–6.9
F2-4 (positive)	0–0.05	0.05–0.18	0.18–0.45	0.45–1.1	1.1–7.3

. Anomaly maps of factor scores illustrate the spatial relationships between elemental associations and known deposits (Figure 11).

F_2-1 (Pb–Ag–Sb–Hg) is primarily associated with Lower Devonian formations (D₁a) and adjacent formations. The Pb (galena)–Ag (silver sulfides)–Sb (stibnite)–Hg (cinnabar) association likely reflects low-to-moderate-temperature hydrothermal mineralization [54], typical of epithermal or mesothermal vein systems, influenced by tectonic or magmatic activity during the Devonian [1]. These elements commonly precipitate from fluids circulating through fault zones or volcanic–sedimentary sequences, particularly in extensional or volcanic arc settings. In marine-origin (SEDEX) or water-saturated sediments, stratiform sulfide layers enriched in these metals could also form through seafloor venting.

F_2-2 (Zn) largely represents the negative of F_2-1 within the Devonian formations and is associated with known Pb-Zn mineralization located in the eastern Devonian units and the Silurian Haijiushan Formation (S₃h). A secondary Zn anomaly occurs along fault crossings in the south, within the Triassic Longwuhe group (T_1-2L), possibly indicating a concealed mineralized zone beneath the Triassic cover.

F_2-3 (Au–Bi) mostly coincides with known deposits tied to acidic granitic intrusions (ilmenite series) and likely reflects magmatic–hydrothermal mineralization processes linked to reduced intrusion-related gold systems. The association with Au-Bi reflects the possibility of Bi to form complexes (e.g., bismuth tellurides) that facilitate gold mobilization [55]. Cooling or chemical changes destabilize these complexes, leading to Au-Bi co-precipitation in quartz veins or shear zones. Bi enrichment may signal proximity to intrusive centers or distal fluid pathways. The Au-Bi association supports a model where granitic cupolas provided heat and metals during tectonic stabilization or crustal thickening events.

F_2-4 (W–Sn) mineralization is typically associated with S-type (peraluminous) and I-type (metaluminous) granites, such as Triassic monzogranites (nγT) in the central part of the map and Triassic granodiorite porphyry (γδμT) in the west. This suggests the presence of concealed granitic cupolas beneath the Devonian cover. During late magmatic stages, hydrothermal fluids exsolved from cooling granitic melts released volatile-rich fluids that transported W and Sn through fracture networks. These were carried as chloride complexes (for Sn) and carbonate complexes (for W), leading to mineral deposition in quartz veins or shear zones. In regions like the Iberian Variscan Massif, W-Sn deposits correlate with specific granite suites (e.g., S₁ and S₂ types) [56,57], suggesting similar processes may apply in Xicheng if granitic cupolas are present. This interpretation aligns with the reported Late Triassic mineralization age (222 ± 2.2 Ma) of the Changba–Lijiagou giant lead–zinc deposit, based on Rb-Sr isotopic dating of sulfide samples [58]. This age corresponds to tectono-magmatic activity related to the Indosinian orogeny. The deposit is associated with Late Triassic granitic intrusions and metamorphosed sedimentary host rocks, reflecting hydrothermal circulation during post-collisional crustal extension [58].

Notably, positive F_2-1 (Pb–Ag–Sb–Hg) and negative F_2-2 (Zn) anomalies are key indicators of Pb–Zn mineralization, with all known Pb–Zn deposits located within these. The Changba–Lijiagou super-large deposit is characterized by strong negative F_2-2 (Zn) anomalies rather than the positive F_2-1 anomalies. Additionally, several Pb–Zn deposits are associated with strong positive F_2-3 (Au–Bi) and F_2-4 (W–Sn) anomalies, suggesting that well-developed F_2-3 and F_2-4 anomalies may signal the exploration potential for Au and W mineralization.

This study focuses on the strong anomalies to delineate mineral exploration targets, identifying eight targets (Figure 12). Target I is defined by strong positive F_2-1 (Pb–Ag–Sb–Hg) and F_2-4 (W–Sn) anomalies, along with moderate F_2-3 (Au–Bi), indicating potential for Pb–Zn and W mineralization. Target II shows strong positive F_2-1 (Pb–Ag–Sb–Hg) and negative F_2-2 (Zn) anomalies, suggesting Pb–Zn potential. Target III is characterized by strong positive F_2-1, negative F_2-2, and positive F_2-4 anomalies, pointing to Pb–Zn and W mineralization. Target IV is characterized by a strong negative F_2-2 (Zn) anomaly, indicating the exploration potential for Pb–Zn mineralization. Target V exhibits strong negative F_2-2 and positive F_2-3 anomalies, suggesting potential for Pb–Zn and Au mineralization. Targets VI and VII are marked by strong positive F_2-3 anomalies, indicating Au potential. Target VIII combines strong positive F_2-3 and F_2-4 anomalies, suggesting the exploration potential for Au and W mineralization.

5. Discussion

The tectono-geochemistry method is based on the concept of leakage halos and is well-suited for the geochemical exploration of buried magmatic and hydrothermal deposits in the bedrock exposed areas. Unlike diffusion haloes, leakage haloes form when ore-forming fluids migrate from the depth to the surface along permeable pathways such as faults and fractures [59,60,61,62]. As early as the 20th century, exploration geochemists compared the tectono-geochemistry method with traditional lithogeochemical approaches in bedrock exposed areas (e.g., the Guanmenshan lead–zinc and Baoshantao copper deposits). The results were compelling: the tectono-geochemistry method effectively identified and enhanced anomalies associated with buried deposits that conventional lithogeochemical methods failed to detect [8,9,62]. Traditional lithogeochemical surveys typically collect point samples along prospecting lines without targeting mineralization-related features such as altered rocks, fault gouges, or fracture zones. This often results in poor sample representativeness and the overlooking of anomalies from deeply buried sources. To address this, geochemists introduced a grid-based composite sampling approach, where multiple subsamples are collected per grid and combined into a single, more representative sample [9]. In recent years, the tectono-geochemistry method has proven successful in detecting and enhancing geochemical anomalies related to buried deposits [10,11,12,14,63,64,65,66,67].

Previous studies have shown that both structures and hydrothermal fluids played key roles in deposit formation in the study area, resulting in well-developed alterations and veins [1,20,68,69,70]. However, regional geochemical surveys often struggle to determine which of these features are related to concealed ore deposits, as the process is time-consuming and uncertain. The tectono-geochemistry method addresses these challenges through its use of targeted sampling media and a composite sampling scheme. Unlike traditional lithogeochemical surveys, which often overlook subtle anomalies, the use of targeted sampling media (e.g., altered rocks, veins, and fault gouges) is more effective in detecting leakage halos formed by ore-forming fluids migrating to the surface along permeable structures [59,60,61,62]. The composite sampling approach enhances anomaly detection by combining multiple subsamples per grid, maximizing the likelihood of capturing mineralization-related features. This not only improves sample representativeness but also reduces analytical costs [21,38,71,72,73]. Together, the targeted sampling media and composite sampling scheme make the tectono-geochemistry method an economical and efficient approach for identifying geochemical anomalies associated with concealed ore deposits in regional geochemical exploration.

The results of the tectono-geochemistry survey in the study area are promising. Nearly all lead–zinc deposits are located within moderate-to-high positive F_2-1 or negative F_2-2 anomalies. In addition, several large-scale and strong virgin anomalies, positive F_2-1 (Pb–Ag–Sb–Hg) and negative F_2-2 (Zn), were identified, suggesting significant exploration potential for Pb ± Zn ± Ag mineralization. Based on the empirical elemental zoning patterns of hydrothermal ore deposits [9], the Pb–Ag–Sb–Hg association is typically linked to epithermal Pb–Zn mineralization (Figure 13). Therefore, targets I to V should be prioritized for the exploration of concealed Pb–Zn mineralization at shallow depths. Another key insight from the tectono-geochemistry survey is the potential for concealed Au and W mineralizations. Although only two gold occurrences have been identified in the central Xicheng Ore Field [17], this survey reported a maximum Au value of 1.62 g/t. Statistical analysis shows that Au exhibits high variability (CV = 6.81) and significant enrichment (EC = 4.94). Additionally, Au mineralization is well-developed in the western part of the Xicheng Ore Field [17,74]. According to the empirical elemental zoning pattern of hydrothermal deposits [9], the Au–Bi association is a strong indicator of concealed mesothermal to hyperthermal Au mineralization at greater depths. Similarly, W shows a high maximum value (0.52%), CV (23.4), and EC (3.11), suggesting good potential for W mineralization. Based on zoning patterns [9], the W–Sn association is a useful exploration indicator for hyperthermal W systems. Notably, target I also shows promise for concealed W exploration at depths.

The superposition of anomalies from the four factors suggests that the geochemical patterns in the study area, including anomaly intensity, distribution, scale, and elemental associations, are jointly controlled by the original Devonian SEDEX-style mineralization [19,21] and subsequent tectonic–magmatic–hydrothermal overprinting [17,75]. Late-stage Au- and W-rich magmatic–hydrothermal activity likely led to deep Au and W mineralization, producing corresponding surface Au–Bi and W–Sn anomalies along permeable structures. Additionally, these later events may have remobilized the original Pb–Zn mineralization, forming new ore bodies in structurally favorable locations and generating leakage halos (Figure 3b,c). Tectonic denudation may also have exposed some ore bodies at the surface, creating strong anomalies (Figure 3a). For example, the Changba–Lijiagou mining area, a typical SEDEX-style mineralizations [1,18,19,21,23], is characterized by the absence of F_2-1 (Pb–Ag–Sb–Hg anomalies) but shows strong negative F_2-2 (Zn), and positive F_2-3 (Au–Bi), and F_2-4 (W–Sn) anomalies. Initial Devonian metallogenesis likely produced SEDEX-style Pb–Zn mineralization with surface anomalies in Pb, Zn, Cd, Hg, and Sb. However, late Triassic magmatic–hydrothermal activity played a key role in shaping the final geochemical patterns. Previous studies suggest this system was responsible for Au, Cu, Mo, and W mineralization in the Xicheng Ore Field [17,75]. Thus, surface Au–Bi and W–Sn anomalies likely reflect deep Au and W mineralization. More importantly, Triassic magmatic–hydrothermal processes may have introduced Ag and remobilized Pb–Zn mineralization, leading to the formation of the Changba–Lijiagou giant lead–zinc deposit and strong surface anomalies in Pb, Zn, Ag, Sb, and Hg. The absence of positive F_2-1 (Pb–Ag–Sb–Hg) anomalies may be partly attributed to mining-related disturbances.

6. Conclusions

(1)

As a regional lithogeochemical method, the tectono-geochemistry method—using tailored sampling media and a composite sampling scheme, combined with staged factor analysis (SFA) and C–A fractal model—can efficiently and economically identify ore-related geochemical anomalies.

(2)

With support from SFA and C–A fractal model, positive F_2-1 (Pb–Ag–Sb–Hg) and negative F_2-2 (Zn) anomalies are reliable indicators of Pb–Zn mineralization, while positive F_2-3 (Au–Bi) and F_2-4 (W–Sn) are vectors towards Au and W mineralization. Eight targets have been delineated based on strong anomaly signatures.

(3)

Based on empirical zoning patterns of hydrothermal deposits, Pb–Zn mineralization should be prioritized due to its likely shallow depths; Au and W also warrant focused exploration at greater depths, given the strong geochemical anomalies and limited existing discoveries.