Scholarlink

<p>Structural fault frameworks impose major constraints on hydrothermal uranium metallogenic systems, calling for precise characterization of structural ore-controlling mechanisms to explore deep and concealed mineral resources. Conventional assessment methods are generally limited to qualitative or semi-quantitative analyses of surface traces and do not adequately quantify the deep-penetrating features of structural architectures. Furthermore, combining multi-source information often introduces biases due to subjective weighting methods. To surmount these limitations, a quantitative evaluation framework is proposed that integrates empirical geometric scaling with multivariate statistical analysis. The granite-hosted uranium field in the southern Zhuguang Mountain region of Northern Guangdong, South China, is utilized as a case study. The empirical displacement-length (D-L) scaling law is first applied to estimate fault vertical extensions, serving as an a priori architectural filter to verify the deep connectivity of surface fault traces. To address spatial multicollinearity, fault linear density, fault intersection kernel density, and Euclidean distances to faults and intersections are selected as key spatial parameters. Principal Component Analysis (PCA) elucidates the intrinsic correlations among these multidimensional variables and objectively determines factor weights, thereby facilitating the construction of a comprehensive Tectonic Control Index (TCI) model. The results show that favorable metallogenic zones, delineated using a statistical threshold based on the Central Limit Theorem, successfully capture the right-skewed spatial anomalies of known deposits. The principal innovation is the development of a quantitative methodological framework that combines empirical depth calibration with data-driven objective weighting, thereby enhancing predictive capacity to identify deep, concealed mineralization networks. These findings deliver a strong, data-driven spatial-statistical approach for the integrated exploration of concealed deposits in complex geological environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Data-driven tectonic prospectivity modeling for granite-hosted uranium systems: integrating empirical scaling and spatial statistics

  • 期刊论文

  • 发表日期：

    2026年8月4日

查看全文
  • Jianan Zhao,
  • Yingru Pei,
  • Chonghao Liu

摘要

Structural fault frameworks impose major constraints on hydrothermal uranium metallogenic systems, calling for precise characterization of structural ore-controlling mechanisms to explore deep and concealed mineral resources. Conventional assessment methods are generally limited to qualitative or semi-quantitative analyses of surface traces and do not adequately quantify the deep-penetrating features of structural architectures. Furthermore, combining multi-source information often introduces biases due to subjective weighting methods. To surmount these limitations, a quantitative evaluation framework is proposed that integrates empirical geometric scaling with multivariate statistical analysis. The granite-hosted uranium field in the southern Zhuguang Mountain region of Northern Guangdong, South China, is utilized as a case study. The empirical displacement-length (D-L) scaling law is first applied to estimate fault vertical extensions, serving as an a priori architectural filter to verify the deep connectivity of surface fault traces. To address spatial multicollinearity, fault linear density, fault intersection kernel density, and Euclidean distances to faults and intersections are selected as key spatial parameters. Principal Component Analysis (PCA) elucidates the intrinsic correlations among these multidimensional variables and objectively determines factor weights, thereby facilitating the construction of a comprehensive Tectonic Control Index (TCI) model. The results show that favorable metallogenic zones, delineated using a statistical threshold based on the Central Limit Theorem, successfully capture the right-skewed spatial anomalies of known deposits. The principal innovation is the development of a quantitative methodological framework that combines empirical depth calibration with data-driven objective weighting, thereby enhancing predictive capacity to identify deep, concealed mineralization networks. These findings deliver a strong, data-driven spatial-statistical approach for the integrated exploration of concealed deposits in complex geological environments.