Three-Dimensional Attribute Modeling and Deep Mineralization Prediction of Vein 171 in Linglong Gold Field, Jiaodong Peninsula, Eastern China

Abstract

As shallow mineral resources become increasingly depleted, the search for deep-seated orebodies has emerged as a crucial focus in modern gold exploration. This study investigates Vein 171 in the Linglong gold field, Jiaodong Peninsula, using 3D attribute modeling for deep mineralization prediction and precise orebody delineation. The research integrates surface and block models through Vulcan 2021.5 3D mining software to reconstruct the spatial morphology and internal attribute distribution of the orebody. Geostatistical methods were applied to identify and process high-grade anomalies, with grade interpolation conducted using the inverse distance weighting (IDW) method. The results reveal that Vein 171 is predominantly controlled by NE-trending extensional structures, and grade enrichment occurs in zones where fault dips transition from steep to gentle. The grade distribution of the 171₁ and 171_sub-1 orebodies demonstrates heterogeneity, with high-grade clusters exhibiting periodic and discrete distributions along the dip and plunge directions. Key enrichment zones were identified at elevations of –1800 m to –800 m near the bifurcation of the Zhaoping Fault, where stress concentration and rock fracturing have created complex fracture networks conducive to hydrothermal fluid migration and gold precipitation. Nine verification drillholes in key target areas revealed 21 new mineralized bodies, resulting in an estimated additional 2.308 t of gold resources and validating the predictive accuracy of the 3D model. This study not only provides a reliable framework for deep prospecting and mineral resource expansion in the Linglong Goldfield but also serves as a reference for exploration in similar structurally controlled gold deposits globally.

1. Introduction

Deep exploration is pivotal for uncovering new mineral resources, expanding reserves, and optimizing mine development. As shallow mineral deposits are progressively exhausted, the localization and prediction of deep-seated orebodies have become critical challenges in modern gold exploration [1,2,3,4,5]. The Jiaodong Gold Metallogenic Province, ranked third globally, is characterized by complex geological structures and diverse gold deposit types, with cumulative proven gold resources exceeding 5500 t [6,7,8]. In recent years, exploration efforts have strategically shifted from shallow deposits to deep and concealed orebodies. Newly identified deep resources now surpass shallow reserves, highlighting the significant prospecting potential of the Jiaodong subsurface.

The Linglong gold field in Jiaodong is well known for its quartz-vein-type gold deposits. However, recent deep exploration has revealed large altered rock-type gold deposits, including the Taishang deposit, Vein 171 in Dongfeng Mine, and the Shuiwangzhuang deposit, with cumulative gold resources estimated at approximately 600 t [9,10,11]. Extensive research has been conducted on the Linglong gold field, yielding a series of significant findings. Regarding the geological characteristics of the deposits, Song et al. [12] classified Jiaodong gold deposits through regional geological mapping and proposed that both quartz-vein-type and altered rock-type deposits are controlled by a unified tectonic system. Geochemical studies by Feng et al. [13] highlighted sulfidation and fluid immiscibility as the primary mechanisms for gold precipitation, based on analyses of gold mineral occurrence states in the Linglong Field. Specifically, sulfidation destabilizes (Au-S) complexes, resulting in gold precipitation, while (Ag-Cl) complexes remain dissolved in the fluid phase. Geochronological studies by Li et al. [14], utilizing Ar–Ar and Re–Os isotopic dating, constrained the main mineralization stage to the Cretaceous period (130–120 Ma), coinciding with regional craton destruction and Pacific Plate subduction. However, a unified consensus on genetic mechanisms remains elusive. For instance, Song et al. [15] attributed the mineralization process to fluid exsolution from anatectic granites, with metals derived from crust–mantle mixing. By contrast, Zhu et al. [16] emphasized the role of extensional tectonics during the destruction of the North China Craton, which facilitated mantle upwelling and ore–fluid mobilization. Yang et al. [17] argued that Pacific Plate subduction triggered large-scale fluid release, establishing a regional metallogenic system. Recent advances in deep drilling and 3D geological modeling [18,19,20,21] have revealed vertical zoning between deep altered rock-type deposits and shallow quartz-vein-type deposits, reflecting staged fluid evolution. In northwestern Jiaodong, gold orebodies cluster at steep-to-gentle transitions in an ore-controlling fault dip, validating a “stepwise” mineralization model [22,23,24]. Nevertheless, fragmented mineral rights distributed among multiple enterprises have restricted exploration to isolated license areas, limiting macroscopic studies on grade distribution and 3D ore-controlling structures. As a result, mining operators face an urgent need for new exploration targets, yet the absence of systematic deep prospecting methodologies remains a critical obstacle to further development.

Building on previous research, this study systematically compiles and organizes geological exploration data from Vein 171 in the Linglong gold field to construct a 3D attribute model of the main orebody. The model examines the geological characteristics, spatial grade distribution patterns, and spatial coupling relationships with ore-controlling fault structures. By employing advanced 3D model integration techniques, grade anomalies within mineralized bodies were identified, enabling accurate deep prospecting predictions within the 3D framework. The analysis of the 3D attribute model for Vein 171 highlights the effectiveness of 3D modeling in enhancing deep prospecting forecasts and orebody delineation, offering scientific guidance for identifying new exploration targets to expand reserves. These findings significantly improve exploration efficiency, reduce costs, extend mine lifespans, and contribute to the sustainable development of gold resources in the Jiaodong region.

2. Regional Geology

2.1. Regional Geological Characteristics

The Linglong gold field is located within the Ludong Uplift of northwestern Jiaodong (Figure 1a), bordering the Sulu Orogenic Belt to the north and situated near the Yishu Fault. To the south, the region is overlain by the Jiaolai Basin [25,26], which comprises three primary stratigraphic units: the Lower Laiyang Group (Early Cretaceous), dominated by green and variegated fluvial-lacustrine clastic rocks; the Middle Qingshan Group (Late Early Cretaceous), composed of basic, intermediate, and acidic volcanic rocks; and the Upper Wangshi Group (Late Cretaceous–Paleocene), characterized by red fluvial-lacustrine clastic deposits. In contrast, northern exposures are dominated by Precambrian metamorphic terranes [27,28], including Archean Jiaodong Group and Paleoproterozoic Fenzishan Group lithologies [29,30].

As a key basement unit, the Jiaodong Group comprises Neoarchean tonalite–trondhjemite–granodiorite (TTG) gneisses, metagraywackes, and plagioclase amphibolites, while the Fenzishan Group is lithologically diverse and subdivided into upper and lower subgroups dominated by granulitic, schistose, and marble units. Following the Triassic Sulu orogenesis, Late Jurassic to Early Cretaceous magmatic pulses became prominent, with Early Cretaceous intrusions marking the peak of Mesozoic gold mineralization events [31,32,33,34]. The Ludong Uplift is dominated by Late Jurassic granitoids [35,36], including notable intrusions in northwestern Jiaodong, such as the Guojialiang (127–130 Ma), Linglong (150–160 Ma), and Luanjiahe (150–158 Ma) intrusions [37,38,39]. East of Linglong, the Biguo pluton (166 Ma) consists of peraluminous melts formed via lower crustal anatexis (Figure 1b).

Fault structures are extensively developed in the area, with the Zhaoping Fault serving as the primary mineralization-controlling structure. It exhibits a characteristic “Y”-type tectonic framework and spans approximately 120 km longitudinally, maintaining widths of 50–200 m, which locally expand to 800 m. Structural segmentation along its strike defines three distinct sectors: northern, central, and southern. At the interface between the Linglong and Luanjiahe plutons, the Zhaoping Fault branches into the NNE-oriented Jiuqujiangjia Fault and the NE-striking Potouqing Fault [12,40,41]. The Zhaoping Fault exhibits a normal fault regime with notable extensional signatures, constituting a listric system developed in a magma-driven, thermally uplifted extensional tectonic environment. NNE-striking structures significantly influence and modify the NE-trending faults, creating a listric architecture dominated by the Jiuqujiangjia–Potouqing fault pair. The Jiuqujiangjia Fault, striking at approximately 33° with a southeast dip, controls the spatial distribution of key gold deposits such as Shuiwangzhuang, Jiuqu, Fushan, and Damoqujia. In contrast, the Potouqing Fault, striking at ~60° with a southeast dip, governs the distribution of major deposits such as Taishang and Dongfeng Vein 171. The Zhaoping Fault formed during the Mesozoic era, from the late Early to Late Cretaceous, influenced by Sulu orogenic processes and Jiaolai Basin evolution, with evidence suggesting potential tectonic rejuvenation during the Cenozoic. The structural framework of the fault system exerts significant control on the morphology and distribution of footwall veins, leading to heterogeneous orebody distribution. Enrichment “hotspot” zones typically occur at abrupt changes in the fault dip, particularly at steep-to-gentle transitions. Under a transtensional regime, modifications to fault attitudes generate continuous dilation spaces, while the principal fault surface displays a sinuous arcuate geometry with gentle undulations. The two branching faults converge at depth during down-dip propagation. Scholarly interpretations diverge regarding the tectonic significance of the Zhaoping Fault within Jiaodong’s structural framework. Some attribute it to transpressive deformation, while others emphasize the Cretaceous extensional regime [42,43]. Nonetheless, normal fault-dominated brittle extensional systems under tectonic extension provide highly favorable architectures for mineral concentration and orebody localization.

Figure 1. Geotectonic location map and regional geological sketch of northwestern Jiaodong (modified from [34]). (a) Geotectonic location map of Shandong region; (b) Regional geological map.

Figure 1. Geotectonic location map and regional geological sketch of northwestern Jiaodong (modified from [34]). (a) Geotectonic location map of Shandong region; (b) Regional geological map.

2.2. Geological Characteristics of the Deposit

The Vein 171 in the Linglong gold field is located in the Jiuqu-Lijiazhuang area, Zhaoyuan City. Large-area strata exposed in the eastern part of the area are the Archean Jiaodong Group, while the Cenozoic Quaternary is relatively dispersed. Mesozoic granites genetically associated with gold mineralization are mainly distributed in the western part of the study area. The main pluton include the Guojialing pluton, Luanjiahe pluton, and Linglong pluton. Lithologies include porphyritic medium-grained hornblende quartz monzonite, medium-coarse-grained monzogranite, weakly gneissic fine-medium-grained garnet-bearing monzogranite, and porphyraceous coarse-medium-grained monzogranite. Among them, the Linglong body is mainly distributed in the footwall of the tectonic fault zone. The Archean Jiaodong Group gneissic fine-grained trondhjemite and Luanjiahe body are mainly distributed in the hanging wall of the fault zone (Figure 2).

Wall rock alteration is well-developed around ore bodies in the study area, mainly including potassic alteration, sericitization, silicification, pyritization, and pyritic-sericitization. Among these, potassic alteration and sericitization are early-stage wall rock alterations, providing the foundation for later mineralization. Silicification, pyritization, and pyritic-sericitization are syn-ore alterations, closely related to mineralization. Wall rock alteration exhibits obvious zoning phenomena, which is closely related to the spatial distance from the orebody. Alteration intensity increases near the orebody and weakens away from it. Vein 171 is located on the southwestern side of the Shuiwangzhuang mining area. Its main 171₁ orebody, is actually the same ore body as orebody II-1 in the Shuiwangzhuang mining area and the Lijiazhuang mining area to the north. Due to mineral rights belonging to different companies, this study only discusses Vein 171.

3. Orebody Hosting Characteristics and Model Construction

3.1. Characteristics of the Orebody

As an altered rock-hosted gold deposit, Vein 171 occurs at the junction of the Zhaoping and Potouqing faults, primarily hosted in pyritic-sericitic cataclasites. Seven orebodies (171₁ to 171₅, 171_sub-1, and 208Ⅱ₁) have been identified, with the 171₁ and 171_sub-1 orebodies collectively accounting for 98% of the reserves. The Potouqing Fault predominantly controls the morphology and distribution of Vein 171, with orebody orientations concordant with the southeast-dipping principal fault surface. The 171₁ orebody spans exploration lines 160 to 68, occurring between elevations of −1550 m and +80 m. It has a strike length of approximately 2500 m and an average thickness of 4.23 m. Its general strike is ~60° with a southeast dip, and the inclination varies between 36.5° and 43.5°. Gold grades range from 1.00 to 26.34 g/t, with an average grade of 2.71 g/t. The 171_sub-1 orebody, located in the northeastern mining area, occurs parallel to and beneath the 171₁ orebody, spanning exploration lines 96 to 68 at vertical elevations between −1118 m and −218 m. It extends 850 m in strike length, has an average thickness of 8.31 m, and exhibits gold grades ranging from 1.00 to 17.35 g/t, with a mean grade of 2.97 g/t. Geometrically, the 171_sub-1 orebody shares a similar orientation to the 171₁ orebody, with a southeast dip of 35–40° and relatively gentle inclination (Figure 3a).

Drill data reveal that the orebodies exhibit vein-like and stratoid distributions in cross-section, with an overall undulating geometry. Down-dip observations show distinct branching-merging configurations as well as localized swelling of the orebody, complicating delineation efforts due to dispersed mineralization (Figure 3b). The branching-merging phenomenon is primarily attributed to multiple mineralization episodes and the unique structural settings of the study area [6,7,15]. Stress concentration induces rock fracturing, facilitating the formation of interconnected fracture networks. Hydrothermal fluids ascending through primary faults preferentially infiltrate secondary fractures and lithological weaknesses, leading to the formation of branch orebodies.

3.2. Orebody 3D Modelling Process

The 3D orebody modeling in this study adopts an integrated surface-block model framework, in which the surface model defines spatial boundaries, while the block model interprets internal attribute distributions [44,45]. Vulcan 2021.5 3D mining modeling software (Maptek, Adelaide, Australia) was utilized to construct 3D models of the Vein 171 gold system, with optimized attribute model parameters to analyze spatial grade distribution patterns. The modeling process began with geological data collection, which involved extracting drilling parameters—such as coordinates, deviation surveys, lithostratigraphy, and assay results—from borehole logs using MapGIS67 software [46]. The data were subsequently imported into the Vulcan 2021.5 software to establish a geological database, forming the foundation for orebody reconstruction. Wireframing techniques [47,48] were then employed to delineate orebodies along exploration profiles. This technique involves sequentially connecting mineralized intercepts to achieve accurate morphological reconstruction. For vein-type gold deposits characterized by features such as bifurcations, barren gaps, xenoliths, and fault truncations, a block-seam integration method was developed. This approach decomposes complex geometries into discrete modular blocks, which are individually modeled. The models are subsequently integrated using topological stitching techniques to create a cohesive and accurate representation of the orebody [49].

The 3D orebody models play a crucial role in geological characterization, mineral prospectivity modeling, mine design optimization, production management, and digital mine visualization systems. In this study, block modeling techniques were employed to develop mineralization property models by discretizing the 3D orebody into uniform cubic cells [50,51,52]. Subsurface grade data derived from boreholes were interpolated using the IDW method, which applies power function parameters to estimate grades. The estimation process assigns weights inversely proportional to distance, raised to a specified power exponent, implementing distance-decay weighted averaging while constrained by block geometry. Mathematical and geological analyses of the constructed grade models revealed spatial distribution patterns that provide a scientific foundation for selecting initial mining zones and identifying deep-level exploration targets. These models significantly enhance mineral exploration efficiency by improving target precision and reducing exploration uncertainties (Figure 4).

4. Parameter Selection for Orebody Property Model Construction

4.1. Combined Sample Length Selection

In the field of geostatistics, standardizing drillhole sample lengths is critical for ensuring an unbiased estimation of resource parameters during quantity evaluation [53]. Variations in sample lengths introduce bias in the calculation of weighted average grades, which can significantly affect the accuracy and reliability of resource evaluations. To address this issue, length standardization is applied to original drillhole samples of unequal lengths, ensuring uniformity across all samples. Geostatistical methodologies are typically employed to determine the appropriate standardized sample size. This involves assessing whether the selected sample length achieves statistical significance, analyzing the variance of different sample length combinations, and evaluating their reasonableness. The effectiveness of the standardization process is determined by examining changes in the dispersion of the data distribution before and after the combination of drillhole sample lengths. The evaluation of sample length combinations is guided by two primary criteria:

Sample variance is a statistical measure used to quantify the degree of dispersion or spread within a dataset relative to its mean. A higher variance indicates that the data points are widely spread around the mean, signifying greater fluctuations in the dataset. Conversely, a lower variance implies that the data points are more tightly clustered around the mean, reflecting reduced fluctuation. This metric is critical for assessing the stability of the dataset and determining how uniform or variable the sample measurements are.

The coefficient of variation (CV) is calculated as the ratio of the standard deviation (SD) to the mean, serving as a normalized measure of data dispersion. It expresses the extent of variability in relation to the mean, providing a dimensionless metric that allows for the comparison of dispersion across datasets with differing units or scales. Larger CV values indicate greater relative variability and more dispersed data distributions, while smaller CV values reflect reduced variability and concentrated distributions.

Using the analytical capabilities of modeling software, sample length distribution frequencies were derived from the borehole database. The histogram for orebody 171₁ reveals a right-skewed unimodal distribution with an elongated tail, deviating from the normal distribution (Figure 5a). Over 75% of the samples cluster at a length of 1 m, with the mean sample length (1.009 m) slightly exceeding the median, indicating mild right skewness. The distribution’s tail extends to a maximum of 1.76 m and a minimum of 0.35 m. A SD of 0.159 and a CV of 0.158 demonstrate moderate dispersion in the sample lengths. Statistically supporting the selection of 1 m as the composite length, which accounts for over 75% of the samples. The sampling interval data for orebody 171_sub-1 exhibits a highly concentrated distribution (Figure 5b). The length histogram shows a single-bar peak at 1 m with no significant tails on either side. Over 96% of sampling intervals are 1 m in length. Both SD and CV approach zero, further justifying the exclusive use of a 1 m composite length for modeling.

4.2. Identification and Treatment of Abnormally High-Grade

4.2.1. Recognition of Abnormally High-Grade

Abnormally high-grade values represent anomalously elevated grades within non-uniform mineralization systems, typically several times higher than the mean grade. These high-grade anomalies can jeopardize the stability of resource estimation, leading to biased economic assessments and distorted mine planning outcomes. Common industry methods for detecting such outliers include geostatistical analyses and the empirical “6–8 times the mean grade” rule applied within resource estimation workflows [54,55]. While no universal validation criteria exist for defining cutoffs for abnormally high grades, practical approaches can adhere to established principles and methodological frameworks. Key principles guiding the treatment of high-grade anomalies include: (1) Reduced coefficient of variation (CV): A lower CV after treatment reflects successful mitigation of high-grade anomaly effects, resulting in a more stable and rational grade distribution. (2) Post-treatment arithmetic mean < Sichel’s T-Estimator: Sichel’s T-estimator, derived from logarithmic transformation and the geometric mean, serves as a theoretical reference. A lower processed arithmetic mean compared to the T-estimator indicates effective adjustment of high-grade anomalies. (3) Production reconciliation: Comparative analysis of exploration data before and after anomaly treatment with post-mining production data evaluates the impact of anomaly processing on resource estimation accuracy and reliability.

Numerous studies have shown that gold grade data in gold deposits typically follow log-normal distribution patterns [56,57]. Statistical methods based on normal distribution theory have proven effective in identifying ultra-high-grade values. In this investigation, geostatistical methods are employed to determine high-grade thresholds, using orebody 171₁ as a case study. The methodology involves importing orebody 171₁ assay data into statistical software (IBM SPSS 27) to generate normal-fit histograms, Q–Q plots, and statistical descriptors for grade distribution analysis (Figure 6a). The analyzed data exhibit substantial skewness, with a significant concentration of low-grade samples observed in the frequency distribution. To satisfy normality assumptions required for geostatistical variogram modeling, a distribution transformation was applied. Logarithmic transformation was selected based on a comparative analysis of tail characteristics in cumulative probability curves, as it effectively mitigated skewness and improved the suitability of the dataset for further geostatistical analysis.

A comparative analysis of histograms and Q–Q plots (Figure 6b) reveals that the log-transformed grade data from orebody 171₁ closely approximates a normal distribution. Q–Q plots demonstrate improved alignment with the theoretical normal line, thereby corroborating the log-normal distribution theory commonly observed in gold deposit grade structures. In accordance with stationarity requirements in geostatistics, the transformed data parameters (mean and standard deviation) were selected as the basis for variogram modeling. The log-normal distribution, widely applied in modeling gold deposit grades, is mathematically expressed as:

$$\ln x_i=\mathrm{\mu } + K\mathrm{\alpha }\sigma$$

(1)

x_i—Threshold of abnormally high-grade; μ—Mean of the log-normal distribution; K_α—Corresponding coefficient of standard normal distribution; σ—Standard deviation.

As shown in Figure 6, the log-transformed data follows a log-normal distribution (μ = 0.653, $\mathrm{\sigma } = 0$.906). In mathematical statistics, a confidence level of 5% is commonly adopted; events below 5% are considered rare events [58]. The probability that the natural logarithm lnx falls within the interval (0.653 − 0.906, 0.653 + 0.906) is 95%. The probability of appearing to the right of this interval is smaller, only 5%/2 = 2.5%. Using the standard normal distribution table, the corresponding coefficient K_0.025 is found to be 1.96, expressed as $\ln x_i$ = a + 1.96σ. Based on the properties of the normal distribution function in probability theory, when $\ln x_i$ = 0.653 + 1.96 × 0.906, the corresponding value is $x_i$ = 11.34 g/t.

4.2.2. Handling of Abnormally High-Grade

Following the identification of the abnormally high-grade threshold (11.34 g/t) for orebody 171₁ using geostatistical methods, the spatial distribution characteristics of high-grade values within the orebody were analyzed. If high grades occur continuously and form a distinct pattern, they should not be treated as anomalies but instead linked to define a high-grade ore belt. However, in this case, the high-grade values were found to be dispersed and clustered in a scattered manner, preventing the isolation of a coherent high-grade orebody. A review of the original dataset identified 21 anomalous data points with grades exceeding the determined high-grade threshold of 11.34 g/t. To address the local disturbances caused by these abnormally high-grade values in orebody modeling, the upper limit of 11.34 g/t was applied for truncation and replacement. This ensures that the influence of extreme values is mitigated, stabilizing the grade distribution and improving the reliability of the orebody model.

Analysis of mathematical statistical processing results (

Table 1. Validation results of Sichel’s T-estimator processed by mathematical statistical methods [58].

Test Method	Sichel’s T-Estimator Test
Calculation basis	GB/T 33444-2016 Specification for exploration of solid mineral resources. [59]
	DZ/T 0338.3-2020 Regulation of mineral resources estimation—Part 3: The geostatistical methods. [60]
Orebody grade	y = ln(x)	follows natural log-normal distribution
Number of samples	420
①	Geometric mean of samples:	1.88
②	Arithmetic mean of samples:	2.68
③	Sample variance of natural logarithm (σ):	0.74
④	Taylor series expansion (order 3 or 4):	3
⑤	Sichel’s T-coefficient = ③ × ④:	1.44
⑥	Sichel’s T-estimator = ① × ⑤:	2.71
Inspect	If ② ≤ ⑥, abnormally high-grade treatment is valid; otherwise, reprocess until satisfied	No abnormally high-grade

) using Sichel’s T-estimator revealed that, following the abnormal treatment of orebody 171₁ grade data, the statistical characteristics became substantially more aligned with geological reality. Key improvements are summarized as follows: (1) Transformation to log–normal distribution: The application of the natural logarithmic transformation y = ln(x) effectively converted the initially right-skewed data into a log-normal distribution. This transformation successfully removed the anomalous tailing effect caused by abnormally high-grade values, improving the overall distribution’s statistical integrity. (2) Improved mean alignment: The post-treatment arithmetic mean value of 2.68 closely approximates yet is slightly lower than Sichel’s T-estimator value of 2.71. This alignment indicates data convergence and reflects the success of high-grade substitution in mitigating the influence of localized anomalies, thereby preventing inflated overall grade assessments. The resulting estimates more accurately represent the true orebody grade distribution. (3) Enhanced data dispersion and continuity: The concurrent reduction in both the coefficient of variation (CV) and Sichel’s T-estimator coefficient signifies a strengthened spatial continuity of mineralization. This improvement in data dispersion patterns reflects more uniform grade variability, contributing to a more reliable orebody modeling process.

4.3. Establishment of Ore Block Size

The determination of block dimensions in 3D geological modeling is a synergistic optimization process, integrating geological understanding with mathematical representation. Block sizing must be grounded in ore-forming mechanisms, systematically considering orebody topology, spatial variability in mineralization, structural controls, and exploration data density. For parent blocks, their long-axis orientation should maintain spatial consistency with the orebody strike direction, with 3D dimensions constrained within 1/5 of the exploration grid density. Child blocks should be partitioned according to orebody morphology, with maximum dimensions not exceeding the minimum minable thickness while maintaining appropriate spatial buffer distances from wall-rock contacts.

Taking orebody 171₁ as an example, geological interpretation reveals this hydrothermal deposit exhibits typical vein-type and stratiform-like structures, with significant disparities between strike extension and thickness variation. Accordingly, 30 m × 30 m × 30 m cubic parent blocks were constructed to characterize the macroscopic orebody framework, with dimensional parameters rigorously constrained by the geometric relationship between median exposed thickness (2 m to 5 m) from development workings and exploration line spacing (150 m) within the mining lease. Sub-blocks of 3 m × 3 m × 1 m resolution were implemented to accurately capture internal lithological variations, enabling precise representation of intra-orebody heterogeneity (Figure 7). Hierarchical modeling preserves natural boundaries through parent-controlled structural frameworks and child-resolved grade distribution patterns, maintaining orebody morphological integrity. Field implementation demonstrates 5.7% volumetric error with controlled boundary deviations, achieving optimal balance between model fidelity and computational efficiency.

4.4. Interpolation Method Choice

Spatial interpolation algorithms for ore blocks predominantly utilize Ordinary Kriging (OK) and Inverse Distance Weighting (IDW). OK requires solving the Best Linear Unbiased Estimate (BLUE) before resource estimation through block modeling, exhibiting computational complexity and demanding rigorous data quality/quantity. IDW offers computational efficiency and simplicity for rapid interpolation but demonstrates sensitivity to outliers (e.g., abnormally high-grade). Post high-grade anomaly treatment via Sichel’s T-estimator validation confirmed the absence of outliers, justifying IDW implementation for grade interpolation. The weighting scheme assigns decreasing influence proportional to distance raised to a power, reflecting spatial correlation patterns (Figure 8), formalized as:

$$x_{\mathrm{b}=}{\displaystyle \frac{\sum _{i=1}^n\frac{x_i}{d_{ i}^m}}{\sum _{i=1}^n\frac{1}{d_{ i}^m}}}$$

(2)

where $x_i$—assay value of the i-th sample falling within the zone of influence; $d_i$—ith sample-to-block geometric distance; $n$—sampling population size; $m$—weighting power parameter; $x_{\mathrm{b}}$—target block interpolation value.

4.5. Ellipsoid Search Parameter Configuration

During grade estimation in block modeling, appropriate configuration of ellipsoid parameters (comprising principal, secondary, and minor axes) is critical to match orebody geometry and enhance estimation accuracy. Fixed ellipsoid parameters suffice for mineralized zones with homogeneous and stable geometry. In contrast, dynamic search ellipsoids prove superior for structurally complex orebodies with folding and stratigraphic controls.

Orebody 171₁ exhibits stable vein-type stratiform geometry with minimal structural disruption from dykes or faults, warranting a fixed ellipsoid search configuration. The Vulcan 2021.5 3D interactive estimation approach necessitates defining an origin point at the block center, followed by triaxial adjustment of the ellipsoid to achieve spatial congruence between its planar configuration and mineralized blocks in the interpolation range (Figure 9). The orientation of the principal axis azimuth should match the orebody’s strike plunge, with the secondary axis plunge angle synchronized to the deposit’s dip inclination. The search distance parameterization occurs in three stages: Primary estimation employs principal axis ranges of 1–1.2× the drilling interval (150 m × 150 m grid), secondary axis ranges matching 1–1.2× adjacent borehole spacing along exploration lines, and minor axis distances set to the mean orebody thickness. Primary estimation parameters are configured with 150 m (principal), 150 m (secondary), and 3 m (minor axes). Secondary estimation expanded search ranges to 1.5× initial values, followed by tertiary estimation using 2× secondary-stage distances. To reduce interpolation bias in high-density sampling areas, clustered samples should be dispersed, and minimum/maximum sample points within the search ellipsoid should be constrained to ensure unbiased estimators.

5. Enrichment Variation Patterns of Orebodies

5.1. Relationship Between Orebody Metal Accumulation and Alteration Zone Thickness Variation

The product of ore grade and thickness (hereafter referred to as gold mineralization value) serves as a key indicator reflecting mineralization enrichment. Based on attribute models of orebody 171₁ and 171_sub-1, gold mineralization values and alteration zone thickness along exploration sections (Figure 10a,b) were extracted to investigate their spatial correlation along strike. The alteration zone thickness of Vein 171 shows an increasing trend from the shallow Line 160 to the deeper Line 68 (Figure 10). Spatial distribution mapping (Figure 3) indicates bifurcation of Zhaoping Fault Zone at Line 108, coinciding with down-plunge extension of main Vein 171 ore shoots. Multiphase tectonic stress superposition enhanced wall-rock fracturing, modifying water-rock interaction conditions and fluid composition, thereby promoting increased mineral dissolution and reprecipitation. Consequently, intensified alteration processes caused progressive thickening of alteration zones. West-to-east and north-to-south transects reveal increasing structural alteration intensity and mineralization strength, accompanied by corresponding enlargement of alteration zones. Lines 160 and 80 exhibit peak mineralization coinciding with local ore shoot swellings along strike, followed by gradual contraction patterns of the ore-controlling structures.

5.2. Relationship Between Orebody Metal Accumulation and Fault Dip

The ore-controlling fracture (Potouqing Fault) in the study area exhibits listric fault characteristics, with dip angles showing a steep-to-gentle alternating pattern along its strike. Due to the limited controlled strike length of 171_sub-1 orebody and insufficient deep-level data, analysis of the relationship between fracture dip angles and ore metal accumulation variations remains constrained. Based on the 171₁ Au orebody model and fault zone architecture, 376 datasets of fracture dip angles and corresponding Au mineralization values were collected at 30 m vertical intervals along the dip direction from shallow to deep levels. These data were used to construct a scatter density plot of fracture dip angles versus mineralization distribution (Figure 11) and a dip angle-Au mineralization correlation diagram (Figure 12), facilitating investigation of dip angle controls on gold enrichment.

Figure 11 illustrates that the overall mineralization enrichment pattern can be divided into a strong mineralization enrichment zone and a weak mineralization enrichment zone. Fault dips in the strong mineralization enrichment zone are primarily concentrated between 30° and 41°, with nine gold-enriched areas (>20 m·g/t) mostly clustered in gentle-dip regions. The weak mineralization enrichment zone, marked near fracture dips of ~37°, displays scattered data points mostly distributed in gentle-dip areas. This indicates that regions with gentler fracture dip angles predominantly govern the enrichment of gold mineralization.

As shown in Figure 12, fracture dip angles generally range from 30.2° to 42.6°, exhibiting a “steep-gentle alternation” pattern, with the two polylines showing distinct opposing trends. Areas with significant variations in fracture dip angles correspond to zones of intense mineralization enrichment, while areas with minor dip angle variations control weak mineralization enrichment. The “stepwise” mineralization model can be divided into two distinct levels: the first mineralization stage, located between –405 m and –345 m, has an average gold mineralization value of 39.67 (m·g/t) and fracture dip angles ranging from 37.21° to 40.43°. This upper step occurs in the shallow part of the orebody, where steeper fracture dips initially host localized gold enrichment, followed by a cliff-like rapid decline in Au grades as mineralization extends downward along the dip direction. The second mineralization stage, situated between –1125 m and –975 m, has an average Au mineralization value of 22.93 (m·g/t) with fracture dip angles of 30.19–34.16°. Here, the opposing trends of the two polylines become more pronounced. As fracture dip angles decrease, the spatial extent of gold mineralization gradually expands. This indicates that variations in fracture dip angles exert a fundamental control on gold enrichment. During ore-forming fluid migration along fractures, steeply dipping zones inhibit mineral precipitation, whereas abrupt deceleration of fluid flow occurs at the transition from steep to gentle dips. Combined with sudden dilation of fracture space in gently dipping zones, these conditions favor fluid precipitation and the formation of mineralization enrichment domains. The gentle-dip segments within the “steep–gentle alternation” fracture system exhibit higher mineralization enrichment levels, demonstrating a clear negative correlation between fracture dip angles and gold mineralization. The observed relationship between fracture dip angles and gold enrichment provides critical guidance for mineral exploration in deep and peripheral zones of the orebody.

6. Spatial Distribution Laws of Ore Grade and Grade Data Analysis

6.1. Spatial Distribution Characteristics of Ore Grade

A 3D grade attribute model was constructed using IDW interpolation based on assay data from known ore zones. The 3D model visually characterizes spatial grade distribution of main orebody 171₁ and branch orebody 171_sub-1 (Figure 13a–f). Figure 13 reveals a heterogeneous grade distribution with bead-like and banded patterns, showing high-grade zones concentrated down-dip and down-plunge.

Spatial distribution analysis reveals that the main mineralization zone of orebody 171_sub-1 concentrated between Lines 80–68, while high-grade segments of orebody 171₁ migrate from Lines 88–80 towards deeper orebody 171_sub-1 at Line 68, exhibiting convergence trends (Figure 13d,e). This indicates vertical superposition of high-grade zones between the two orebodies, with near-equidistant parallel distribution patterns. Orebody 171_sub-1 displays multiple parallel vein branches (Figure 3b) with vertical zoning characteristics. At the Line 68 cross-section, orebody 171_sub-1 exhibits greater thickness and higher grades than orebody 171₁. These features suggest multistage magmatic intrusions, where later-stage magma transported new metallogenic materials, resulting in superimposed enrichment of pre-existing orebodies. The deeper position of orebody 171_sub-1 near magmatic conduits allows enhanced supply of a magma-derived metals, increasing its thickness and grade. The Lines 108–68 area lies within a stress superposition zone of bifurcated faults, providing favorable conditions for fluid migration and mineral deposition. Therefore, similar parallel orebodies likely to exist vertically along the extent of orebody 171_sub-1.

6.2. Ore Grade Data Analysis

The “spatial distribution” function in modeling software extracts grade data along E–W, N–S, and vertical directions. Figure 13 illustrates sinusoidal Au grade variation along E-W (x-axis) in orebody 171₁, with a wavelength of 2200 m and an amplitude of 1.68 g/t–4.55 g/t (Figure 14a). Au grade peaks at x = 547,650 (shallow) and x = 549,950 (deep), showing alternating high-low grade patterns between them. Orebody 171_sub-1 exhibits a systematic increase in Au grade with depth (Figure 14b), accompanied by thickening of the alteration zone. At the demarcation point x = 549,825 (Figure 3), which corresponds to the bifurcation of the Zhaoping Fault, the orebody orientation changes abruptly. Deep drilling reveals intensified structural deformation with enhanced wall-rock fracturing. Analysis of dip–mineralization relationships (Figure 11 and Figure 12) indicates the presence of dilatant zones at dip transition areas, controlled by extensional structures. These extensional structures enhance the permeability of fault zones, creating favorable pathways for fluid migration. The superimposition of multi-stage structural events further expands the mineralization space, resulting in imbricated orebody distributions with multiple parallel branch veins. This area exhibits significant increasing trends in both average Au grade and alteration zone thickness, highlighting the governing influence of superimposed tectonic stresses on orebody spatial distribution.

As shown in Figure 15, the average Au grade of the 171₁ orebody along the y-axis primarily ranges from 1.68 g/t to 4.68 g/t, with a stable variation trend and a peak value at y = 4,146,350 (Figure 15a). Integration with the grade distribution model of the main orebody in Vein 171 (Figure 13) reveals an upward trend in average grade when the orebody traverses barren intervals. Three-dimensional orebody modeling identifies barren intervals as regions with significant dip variations along the inclination direction, predominantly occurring in steep-dipping fracture zones. Further analysis reveals pronounced mineralization intensity variations along the deep extension direction, forming weakly mineralized (barren) separation belts between orebodies. It is proposed that steep-dipping structural configurations restrict effective mineral precipitation during ore-forming fluid migration through fractures. Post-barren interval fluid migration occurs within progressively gentler fracture dips, leading to reduced fluid transport velocities. Concurrent dilation of fracture-hosted ore accommodation spaces creates favorable conditions for mineralizing fluid precipitation. A sharp increase in Au grade at y = 4,147,500 in orebody 171_sub-1 indicates enhanced structural control by Zhaoping Fault bifurcations on ore distribution and grades at depth (Figure 15b). Tectonic superposition likely creates more favorable mineralization spaces at depth, triggering a sudden elevation of gold grades.

Previous studies demonstrate a near-equidistant parallel distribution of gold enrichment zones along the northern Zhaoping Fault [24,41,61,62,63]. Analysis of vertical (z-axis) grade data in orebody 171₁ reveals similar near-equidistant parallel distribution patterns (Figure 16a). Vertical grade profiles display distinct peaks at elevations of −375 m and −975 m. Field investigations confirm a NE-plunging orebody geometry, with underground exposures revealing open-ended mineralized structures along both strike and dip. As shown in Figure 16, peak gold grades occur at elevations of –1005 m to –1065 m in orebody 171_sub-1 and –975 m to –1025 m in orebody 171₁ (Figure 16a,b). This confirms that structural superposition creates favorable hosts for parallel veins in the vertical dimension.

7. Deep-Level Exploration Targeting and Preliminary Validation of Prediction Results

7.1. Deep-Level Exploration Targeting

Building upon previous studies on the metallogenic mechanisms of Jiaodong-type gold deposits [23,64,65] and the enrichment patterns of Vein 171, the tectonic-fluid coupling ore-controlling theory is synthesized as follows: (1) The tectonic-fluid coupling mineralization mechanism: Attitude variations, such as the “steep-flat alternation” structure within NE–NNE-trending extensional faults, create localized ore-hosting sites by providing favorable physical spaces (low-pressure zones) for gold precipitation. Variations in fault dip angles control orebody plunging directions and 3D geometries, exemplifying “structural control of mineralization positioning”. As deep-seated high-pressure ore-forming fluids ascend along faults, abrupt pressure drops occur at structural attitude mutation zones (e.g., fault bends), due to gouge layer barriers, triggering fluid decompression. Permeability contrasts within fault zones drive differentiation in fluid migration pathways, which in turn facilitate the progressive mineralization evolution from disseminated → veinlet → stockwork textures. (2) Stepwise mineralization model and alteration zoning: The interplay between complex fault architectures and ore–fluid migration result in a “stepped” mineralization distribution. Each “step” corresponds to a tectonic-fluid activity event, forming imbricate or bead-like orebody distributions in 3D space. From the main fault plane outward, alteration zoning develops sequentially from principal mineralized pyritic sericitization → sericitization → potassic alteration → weak sericitization, showing progressively decreasing mineralization intensity (Figure 17). (3) Critical exploration indicators: favorable prospecting targets include: Sudden “steep-flat alternation” transitions in fault attitudes, secondary fracture concentration zones, and cataclasite development belts.

In the constructed 3D coupling model of the Zhaoping Fault and mineralized domains (including Vein 171, the Lijiazhuang orebody, and the Shuiwangzhuang orebody, Figure 18), it was observed that the spacing between the two bifurcated faults gradually increases as they extend northeastward. In the northern segment of the Zhaoping Fault, the two bifurcated faults exhibit distinct convergence characteristics with increasing exploration depth. The industrial-grade intervals of the mineralized domains show enrichment trends, with the primary enrichment zone located between elevations of −1800 m and −800 m. Proximity to the center of the bifurcated faults is associated with increased wall rocks fragmentation near ore zones and a corresponding expansion of alteration zones. Based on these findings, the deep segments of bifurcated faults are identified as priority targets for future exploration: (1) Stress concentration and release in these areas induces rock fracturing, creating complex fracture networks that facilitate hydrothermal fluid migration and ore precipitation. (2) Listric faults (Potouqing Fault) formed under thermal doming-extension regimes transform local ore-bearing structures from compressional-shear to tensional-shear mechanics. This stress state transition generates dilational spaces for mineralizing fluids, facilitating multi-phase structural-fluid coupling during critical transformation stages. These changes in physicochemical conditions ultimately trigger the precipitation of metal-rich fluid. (3) The unique tectonic setting of fault bifurcation promotes gradual amplification and stabilization of fault scales while controlling the spatial distribution of mineralization. Guided by these structure-controlled orebody plunging patterns, geological exploration teams discovered the Shuiwangzhuang II-1 orebody (containing 110 t gold reserves) through deep drilling northeastward along the plunge direction of Vein 171. However, the near-equidistant parallel distribution of orebodies and their vertical zoning patterns remain insufficiently validated, providing crucial research directions for future studies.

7.2. Preliminary Validation of Predictive Results

Based on the grade distribution model of the Vein 171 main orebody (Figure 13), drilling holes were strategically deployed within the vertical axial space of the deep extension of the 171_sub-1 orebody to track parallel veins. These were aligned with the high-grade zone extension direction (Figure 19a) within the intersection area of two branching faults. Guided by the exploration rationale of “gently dipping veins exhibiting near-equidistant parallel distribution” and “prospecting near known mineralization while expanding boundaries”, deep drilling verification was conducted on the 171_sub-1 orebody. This was further supported by artisanal mining occurrences observed around the mining area. Nine drillholes were positioned along Lines 86–74 within the footwall of the Potouqing-Jiuqujiangzhu master fault bifurcation at the −660 m level (Figure 19a). Of these, three holes intersected mineralization. The drilling uncovered the 207 and 208 veins, delineating 21 mineralized bodies with proven ore reserves of 671,031 t averaging 3.44 g/t Au, containing 2.308 t gold metal.

Drillholes 74KZK01, 80KZK01, and 86KZK01 all intersected Vein 207 (Figure 19b) and Vein 208, revealing gold mineralized bodies within alteration zones, with the orebodies remaining open at shallow depths. Among these, Drillhole 74KZK01 displayed favorable mineralization intersections, distributed between survey lines 74–86, within the elevation range of −1092 m to −951 m. The exposed orebodies exhibit stratiform-like geometry with heterogeneous mineralization, striking approximately 60° and dipping northeast with an average dip angle of 30° (locally reaching 40°). The orebodies extend 300 m along strike, with a maximum depth continuity of 204 m, and a total of 13 mineralized zones were intersected. Vein 207 was encountered at depths of 382.60−389.60 m, showing massive pyrite aggregates on core surfaces, locally banded pyrite, and powdery pyrite distributed along fracture planes (Figure 19c). Comparative analysis of adjacent boreholes revealed that both 74KZK03 and 86KZK03 expose thick and stable potassic-altered granite rock masses in their deep sections.

Previous studies have indicated that potassic alteration associated with the early mineralization stage [8,17,67]. This alteration typically occurs in the outer zones of alteration belts, where gold content is low, mineralization is poorly developed, and it generally appears distal to orebodies. The formation of gold deposits in this district is strongly correlated with fault activities, occurring exclusively along NE-striking major faults and their footwall transtensional fracture systems. Gold potential decreases rapidly with increasing distance from these main fault structures. Drilling data reveals that proximity to NE-trending faults correlates positively with increased metallic sulfide content, more diversified metal mineral assemblages, and improved orebody thickness and grade within alteration zones. Conversely, both alteration intensity and mineralization quality decline as the distance from these faults increases. This suggests that ore-forming hydrothermal fluids migrated upward along NE-trending fault systems and precipitated gold under favorable physicochemical conditions. The listric fault system in this region, characterized by combined tensional-compressional stresses, provides integrated ore-conducting and hosting conditions, serving as a primary pathway for deep-seated metallogenic fluid migration. Integrated 3D modeling analysis further highlights the enhanced prospecting potential near the intersection zones of bifurcated fault systems.

8. Conclusions

Integrated studies utilizing 3D attribute modeling of Vein 171, deep exploration forecasting, geological characteristic analysis, mineralization pattern summarization, and drilling verification yield the following conclusions:

(1)

Using Vulcan 2021.5 3D mining software with surface-block model integration and geostatistical approaches, a refined 3D attribute model of the main orebody in Vein 171 was established through dynamic hierarchical modeling (parent blocks 30 m × 30 m × 30 m, child blocks 3 m × 3 m × 1 m) and IDW grade interpolation. The model reveals that high-grade zones concentrated along the dip and plunge directions of the orebody, displaying near-equidistant parallel distribution patterns.

(2)

The 3D coupled model further confirms a strong correlation between ore enrichment zones and the steep-to-gentle transitional positions of ore-controlling faults. Orebody 171₁ exhibits intense mineralization enrichment (Au > 20 m·g/t) within gentle-dip fault segments (30–41°). This is primarily attributed to the rapid velocity reduction of ore-forming fluids at steep-to-gentle dip transition zones, which significantly enhanced gold precipitation efficiency.

(3)

Based on 3D orebody modeling, spatial grade analysis, and the identified vertical stacking trend of high-grade intervals, combined with the “prospecting near known ores” strategy, the intersection zone of bifurcated faults within the northern Zhaoping Fault system (elevations −1800 m to −800 m) is predicted as a highly favorable prospecting target. This prediction has been effectively validated by nine drilled boreholes that show spatial consistency between high-grade intercepts and model projections. This demonstrates the high reliability of 3D attribute modeling for deep mineral exploration and provides critical guidance for future exploration activities.