【作者】 黄静宁；

【导师】 赵鹏大；

【作者基本信息】 中国地质大学（北京） ， 矿产普查与勘探， 2010， 博士

【摘要】 矿床作为一定异常地质过程的产物,成矿的原因是“地质异常事件的耦合”,研究地质异常的目的是揭示“矿致因素”和“找矿标志”,“地质异常找矿方法”的关键技术是研究“致矿地质异常”,提高对深层次致矿异常信息的识别成为研究地质异常的重点和难点所在。同时,基于地质异常的成矿预测是一种有序性很强,突出了致矿异常信息定量提取的成矿预测方法,在成矿预测的过程中,查明不同尺度、不同类型的致矿地质异常与不同级别找矿地段之间的关系是基于地质异常的成矿预测方法的基础,也是找矿的前提和选靶的依据。本次研究是在地质异常致矿理论的指导下,通过不同的信息提取方法,获取不同层次的致矿异常信息,重点探索和研究了多元统计分析方法、地质统计学方法、多重分形滤波和二维经验模分解(BEMD)等非线性方法在深层次致矿异常信息提取中的应用,并在此基础上开展了基于地质异常的多尺度聚焦找矿方法的实际探索应用,研究结果表明:1.以我国Cu-Au-PGE重要的成矿远景区滇东地区为研究区,采用因子分析、地质统计学和多重分形S-A滤波等方法对滇东地区复杂地质背景下的不同层次的Pt-Cu-Au矿化异常进行定量提取与评价。研究结果表明:(1)因子分析结果能够揭示成矿背景和致矿异常元素组合及其空间分布特征;(2)变差函数分析能有效揭示成矿元素在区域上的空间变化特征,泛克立格法能有效揭示区域异常;(3)多重分形滤波法能有效提取深层次局部矿化异常信息。2.系统论述了二维EMD方法原理,编制了MATLAB环境下的二维EMD处理模块,并以鲁西铜石金矿田为研究区,成功的将二维EMD方法应用于铜石金矿田重力数据分解,获取了比傅氏变换处理结果更精细、更客观反映铜石金矿田深部地质结构、与金矿化有关的致矿异常信息。然后结合其地质特征,构建了铜石金矿田地质地球物理成矿模型,该成矿模型显示具有环形重力异常特征的接触交代矿化带是寻找金矿的远景地段,而在铜石杂岩体北东侧,早白垩世火山沉积岩之下,应是寻找隐伏金矿床的潜在地段。3.不同的数学方法可用以解决不同的地质成矿问题,因子分析和地质统计学等线性方法在研究区域成矿地质背景、揭示成矿元素及其元素组合在区域上的空间变化性特征上具有显著优势;多重分形滤波方法对提取局部矿化异常信息是有效的;二维EMD方法可多层次分解复杂的叠加场信号从而获取深部地质构造与矿化有关的致矿异常信息,以上方法可在深层次致矿异常信息提取中得到推广应用。4.根据基于地质异常的“5P”找矿地段逐步逼近法的思路,将大尺度下圈定的滇东Pt、Pd地球化学省定义为寻找Pt矿的找矿可行地段,对该地段实施多尺度聚焦找矿方法的实际应用,应用变差函数分析不同尺度下Pt含量的空间变化特征及控制因素,依据不同尺度的勘查数据,定量圈定了寻找Pt矿的找矿有利地段和资源潜在地段,在实施聚焦找矿过程中,多尺度的勘查数据揭示了多层次的控矿因素组合,获取致矿信息量逐渐增多,找矿靶区逐渐缩小而资源量显著浓集,靶区级别逐渐增高。因此,多尺度聚焦找矿战略具有从全球到成矿省尺度的高性能选靶能力,该方法具有快速易实施的特点,适合在勘查程度较低地区推广应用。更多还原

【Abstract】 As the result of certain geological anomalous process, deposit is located at the area where the spatio-temporal coincidence of a series of geological anomalies. The aim of geological anomaly research is to reveal the mineral deposit controlling factors and indictor, and the key of quantitative mineral resources assessment based on geological anomaly is researching the geoanomaly associated with mineralization. Therefor, how to recognize the conceal geoanomaly information associated with mineralization effectively is becoming the difficulty and the key point of geological anomaly research. Meanwhile, quantitative mineral resources assessment based on geological anomaly is the theory in sequence and emphasized on quantitative extraction of geoanomaly information associated with mineralization. In mineral resources prediction process, how to identify the relationship between multi-scale and multi-type geoanomaly associated with mineralization and different stage of ore-finding process is the basic of quantitative mineral resources assessment based on geological anomaly, the precondition of ore-finding and the evidence of target selection.This study is based on the geological anomaly ore-forming theory, using different information processing techniques to obtain geoanomaly information associated with mineralization in different level, specially focus on researching and applying multi-statistical analysis, geostatistics, multi spectrum and area filtering (S-A) technique, and Bidimensional empirical mode decomposition (BEMD) method on extracting the conceal geoanomaly information associated with mineralization, then, explore application of the telescoping ore targets at multiple scales base on geological anomaly. The studied results show as follows.1. Eastern Yunnan area-the important Cu-Au-PGE ore-forming perspective area in China is selected as the study area. The factor analysis, geostatistical and multifractal S-A filtering method are applied to quantitatively extract and estimate the Pt-Cu-Au mineralization anomalies from the complicated geological background. The research shows that: (1) The results of factor analysis illustrate that there are three element associations existing in Eastern Yunnan area; (2) The semivariation analysis illustrates that the concentrations of Pt, Cu and Au are continuous in NE trend (9°~18.6°) and the maximal ranges are about 95km. The geochemical maps obtained by the Universal Kringing imply the concentration distributions of Pt, Cu and Au are mainly controlled by the faults and magmatism; (3) The multifractal S-A filtering method characterizes the detailed information of local mineralization, extract the conceal mineralization information.2. A bidimensional empirical mode decomposition (BEMD) program on a MATLAB platform was effectively used to handle gravity signals for the Tongshi gold field. This yielded a three-dimensional intrinsic mode function image that meticulously depicts the spatial distribution relationship between various gold deposits and the different geological units of the gold field rather than Fourier transform. By combining the IMF image with geological features yields a geological–geophysical pattern for the Tongshi gold field showing the formation and distribution of gold deposits, which shows that concealed gold deposits might be discovered on the northeastern side of the Tongshi intrusive complex and the contact metasomatic zone covered by early Cretaceous volcanic sedimentary rocks.3. Different geological and ore-forming problems can be solved by different geomathematic methods. The linear geomathematic methods such as the factor analysis and geostatistics can be used to explore the regional ore-forming background and the regional spatial variability of the ore-forming elements and the element associations, while the nonlinear geomathematic methods such as S-A method are efficient in extracting the local ore-forming information and the BEMD method are efficient in decomposing the complex superimposed field singles such as gravity data into several components to obtain the geoanomaly information associated with mineralization under cover. These methods are of benefit to the concealed ore-forming anomaly information extraction.4. Following the method of quantitative delineation of“5P”ore finding area, the Pt-Pd geochemical province in Eastern Yunnan was defined as permissive ore-finding area for Pt and the telescoping ore targets at multiple scales was applied in this area. The semi-variograms were used to quantitatively describe the variability of Pt anomalies and further analyze the factors controlling the variability, preferable ore-finding area and potential mineral resource area were quantitatively delineated based on multi-scale exploration data. During the process of telescoping ore targets, multi level of ore-control factors were revealed while the amount of geoanomaly information gradually increased, the target areas gradually decreased while the Pt resource amount concentrated singularly. So, the telescoping ore targets at multiple scales has a good exploration function that efficiently focuses on ore targets and easily and quick to apply, especially in the green field exploration.更多还原