SOM-加权FCM不确定性聚类的奇村地热水化学分类与成因识别

    Geothermal hydrohemistry classification and genesis identification of Qicun using SOM-Weighted FCM uncertainty clustering

    • 摘要: 为探明奇村地热区地热流体化学特征、控制因素以及成因机制,选取区内2024年5~6月采集的24份地热水样品,通过Piper三线图、Gibbs图和离子端元图分析地热水化学特征,采用熵权水质指数法对水质进行评价。同时,引入自组织映射网络(SOM)与加权FCM模糊聚类相结合的方法识别水化学类型,通过Bootstrap重采样对聚类稳定性进行检验,揭示地热水主要物质来源、水质影响因素与地热流体成因机制。结果表明:奇村地区地热水呈弱碱性,水化学类型以SO4·Cl-Na型、HCO3-Ca型和HCO3-Ca·Mg型为主,地下水水质整体较好,Ⅰ类、Ⅱ类水样共占比91.67%,局部地区存在NO3⁻超标。SOM-加权FCM将水样划分为3类:C1受浅层补给影响,C2为端元混合与中等水–岩作用影响的过渡型,C3代表深循环高演化端元。Bootstrap结果显示聚类整体稳定(均值0.86),识别核心井17个,过渡井1个。研究区地热水水化学特征受深部地热水与浅层地下冷水混合作用影响,水–岩作用是深部地热水的主要影响因子,硅酸盐类矿物溶解是其主要类型。本研究为奇村地区水热型地热资源的可持续开发利用与水质保护提供依据。

       

      Abstract: To investigate the chemical characteristics, control factors, and genesis mechanisms of geothermal fluids in the Qicun geothermal area, 24 geothermal water samples collected from May to June 2024 were analyzed. Piper ternary diagrams, Gibbs diagrams, and ion endpoint diagrams were employed to characterize the geothermal hydrochemistry, while the entropy-weighted water quality index method was used to evaluate water quality. Simultaneously, a combined approach of Self-Organizing Map (SOM) and weighted Fuzzy Compound Membership Function (FCM) clustering was employed to identify hydrochemical types. Cluster stability was verified through Bootstrap resampling, revealing the primary sources of geothermal water constituents, influencing factors on water quality, and the genesis mechanisms of geothermal fluids. The results indicate that the geothermal water in Qicun exhibits weak alkalinity, with the predominant hydrochemical types being SO4·Cl-Na, HCO3-Ca and HCO3-Ca·Mg. Overall groundwater quality is favorable, with Class I and Class II water samples accounting for 91.67% of the total. However, localized areas show nitrate (NO3⁻) concentrations exceeding regulatory standards. SOM-weighted FCM classified water samples into three categories: C1 influenced by shallow recharge, C2 representing a transitional type affected by endpoint mixing and moderate water–rock interaction, and C3 representing deep-circulating, highly evolved endpoints. Bootstrap results indicated overall cluster stability (mean 0.86), identifying 17 core wells and 1 transitional well. The hydrochemical characteristics of geothermal water in the study area are influenced by the mixing of deep geothermal water with shallow groundwater. Water-rock interaction is the primary influencing factor for deep geothermal water, with silicate mineral dissolution being the predominant type. This study provides a basis for the sustainable development and utilization of hydrothermal geothermal resources in the Qicun area, as well as for water quality protection.

       

    /

    返回文章
    返回