Mapping Geochemical Anomalies Through Integrating Random Forest and Metric Learning Methods

Extracting geochemical anomalies from geochemical exploration data is one of the most important activities in mineral exploration. Geochemical anomaly detection can be regarded as a binary classification problem. The similarity between geochemical samples can be measured by their distance. The key issue of this classification is to find the intrinsic relationship and distance between geochemical samples to separate geochemical anomalies from background. In this paper, a hybrid method that integrates random forest and metric learning (RFML) is used to identify geochemical anomalies related to Fe-polymetallic mineralization in Southwest Fujian Province of China. RFML does not require any specific statistical assumption on geochemical data, nor does it depend on sufficient known mineral occurrences as the prior knowledge. The geochemical anomaly map obtained by the RFML method showed that the known Fe deposits and the generated geochemical anomaly area have strong spatial association. Meanwhile, the receiver operating characteristic curves for the results of RFML and another method, namely maximum margin metric learning, indicated that the RFML method exhibited better performance, suggesting that RFML can be effectively applied to recognize geochemical anomalies.

     

Enhancement and Mapping of Weak Multivariate Stream Sediment Geochemical Anomalies in Ahar Area,NW Iran

In this study, stream sediment geochemical data have been subjected to robust principal components analysis (RPCA) and singularity mapping (SM) to enhance and map significant multivariate geochemical anomalies (i.e., mineralization-related) in Ahar area, NW Iran. The RPCA was applied to (a) account for the compositional nature of stream sediment geochemical data using suitable log-ratio transformation, (b) modulate the effect of outliers in component estimation and (c) derive a multivariate geochemical footprint of mineralization. The SM was applied to extract anomalous patterns of the multivariate geochemical footprint of mineralization. The exploration targets were then delineated using Student’s t-statistics analysis. The correlations of mapped exploration targets with the known mineral occurrences and mineralization-related patterns were further evaluated using normalized density index and overall accuracy analyses.     

Assessing Uncertainty Associated with the Delineation of Geochemical Anomalies

A pedogeochemical exploratory survey of gold deposits was carried out in the region of São Sepé (southernmost Brazil). The region comprises a predominantly metamorphosed belt of volcanoclastics, sediments, serpentinites, basalts, gabbros, chert, tuffs, and banded iron formation of the Proterozoic age. The anomalies were identified first by stream sediment heavy mineral survey at the regional scale of exploration. Once spatial continuity was modeled, ordinary block kriging was performed to generate geochemical maps. Indicator block kriging also was used as an alternative in analyzing and interpreting geochemical data. A novel approach is proposed, which combines both ordinary and indicator kriging for delineating geochemical anomalies. Probability maps proved to be appropriate for selecting new sites for further exploration. Gold anomalies in soils trending NE were well defined by geostatistical analysis and subsequently confirmed by drilling.     

Multi-parameter Analysis of Local Singularity Mapping and Its Application to Identify Geochemical Anomalies in the Xishan Gold Deposit,North China

Natural Resources Research - Identification of geochemical anomalies from geological background is of great significance in the exploration of complex mineralization systems. For a 2D problem, the...     

Assessment of Various Fuzzy c-Mean Clustering Validation Indices for Mapping Mineral Prospectivity: Combination of Multifractal Geochemical Model and Mineralization Processes

This paper describes the application of an unsupervised clustering method, fuzzy c-means (FCM), to generate mineral prospectivity models for Cu?±?Au?±?Fe mineralization in the Feizabad District of NE Iran. Various evidence layers relevant to indicators or potential controls on mineralization, including geochemical data, geological–structural maps and remote sensing data, were used. The FCM clustering approach was employed to reduce the dimensions of nine key attribute vectors derived from different exploration criteria. Multifractal inverse distance weighting interpolation coupled with factor analysis was used to generate enhanced multi-element geochemical signatures of areas with Cu?±?Au?±?Fe mineralization. The GIS-based fuzzy membership function MSLarge was used to transform values of the different evidence layers, including geological–structural controls as well as alteration, into a [0–1] range. Four FCM-based validation indices, including Bezdek’s partition coefficient (V_Pc) and partition entropy (V_Pe) indices, the Fukuyama and Sugeno (V_FS) index and the Xie and Beni (V_XB) index, were employed to derive the optimum number of clusters and subsequently generate prospectivity maps. Normalized density indices were applied for quantitative evaluation of the classes of the FCM prospectivity maps. The quantitative evaluation of the results demonstrates that the higher favorability classes derived from V_FS and V_XB (N_d?=?9.19) appear more reliable than those derived from V_Pc and V_Pe (N_d?=?6.12) in detecting existing mineral deposits and defining new zones of potential Cu?±?Au?±?Fe mineralization in the study area.

     

Machine Learning of Mineralization-Related Geochemical Anomalies: A Review of Potential Methods

Research on processing geochemical data and identifying geochemical anomalies has made important progress in recent decades. Fractal/multi-fractal models, compositional data analysis, and machine learning (ML) are three widely used techniques in the field of geochemical data processing. In recent years, ML has been applied to model the complex and unknown multivariate geochemical distribution and extract meaningful elemental associations related to mineralization or environmental pollution. It is expected that ML will have a more significant role in geochemical mapping with the development of big data science and artificial intelligence in the near future. In this study, state-of-the-art applications of ML in identifying geochemical anomalies were reviewed, and the advantages and disadvantages of ML for geochemical prospecting were investigated. More applications are needed to demonstrate the advantage of ML in solving complex problems in the geosciences.     

Unsupervised Machine Learning for Lithological Mapping Using Geochemical Data in Covered Areas of Jining,China

Natural Resources Research - Application of (supervised and unsupervised) machine learning algorithms to big geoscience data can facilitate intelligent lithological mapping and interpretation in a...     

Co-simulated Size Number: An Elegant Novel Algorithm for Identification of Multivariate Geochemical Anomalies

Natural Resources Research - Identification of geochemical anomalies is of particular importance for tracing the footprints of anomalies. This can be implemented by advanced techniques of...     

城郊聚落景观的集聚特征分析方法选择研究

高度集聚是城郊聚落景观最明显的空间格局特征,是郊区城市化的直观反映。针对目前景观集聚程度研究空间计算方法缺乏筛选、不具可比性等问题,通过4种集聚空间算法在不同角度分析典型区域的城郊聚落景观集聚特征,并在西安市长安区作以简单应用。结果表明:1景观聚集度适于区分出同类连续大斑块和不同类破碎小斑块,核密度适于宏观上集聚组团的识别,空间关联算法适于空间定位具体要素点的集聚特征,Ripley’s L函数适于识别空间距离以确定搜索半径;2根据核密度计算结果,从研究区聚落景观中提取出3个大型组团,分别命名为"政府商业中心聚落组团""沣渭新区聚落组团"和"旅游度假区聚落组团",其划分模式符合研究区各经济板块的未来发展方向。     

面向地学应用的不确定数据聚类算法比较研究

不确定数据聚类分析已成为空间数据挖掘领域的一个研究热点。近年来,在传统划分与基于密度的聚类算法基础上,一系列不确定数据聚类算法相继被提出。虽然这些算法在地学领域的应用已经得到了广泛关注,然而其实际应用的有效性尚缺乏客观的评价。为此,选取当前具有代表性的6种算法进行实验对比分析。首先,设计40组包含预设模式的模拟数据进行测试。进而,采用亚洲气候数据集对6种方法识别气候区的能力进行比较分析,以Kppen-Geiger气候分类结果为基准对各种方法的实际应用效果进行评价。借助准确率和召回率对各种方法的聚类质量定量度量后,发现:1)对于同类型聚类算法,采用相对熵距离的算法聚类质量总体优于采用期望距离和模糊距离函数的算法;2)采用相对熵距离的划分算法聚类质量优于基于密度的算法,其中采用相对熵距离的KMedoids-KL算法的聚类质量最好。     

Water Resources Management Through Flood Spreading Project Suitability Mapping Using Frequency Ratio,k-nearest Neighbours,and Random Forest Algorithms

Natural Resources Research - Lack of water resources is a common issue in many countries, especially in the Middle East. Flood spreading project (FSP) is an artificial recharge technique, which is...     

Regionalized Classification of Geochemical Data with Filtering of Measurement Noises for Predictive Lithological Mapping

Natural Resources Research - A method for predictive lithological mapping is proposed, which combines geostatistical simulation of geochemical concentrations with coregionalization analysis and...     

A Novel Hybrid Technique of Integrating Gradient-Boosted Machine and Clustering Algorithms for Lithology Classification

Natural Resources Research - The significant body of research on lithology identification in recent years has laid emphasis on the improvement of classification performance using hybrid machine...     

多尺度空间分层聚类算法在土地利用与土地覆被研究中的应用

利用遥感数据,综合最大似然法监督分类、多尺度空间分层聚类的部分监督分类方法和主成分方法,分析黄河上游龙羊峡水库库区1987～1999年间土地利用土地覆盖变化.提取专题信息,不同要素采用不同方法;具体分类中,土地利用类型的一级类型耕地、水体及未利用土地类型采用主成分分析和最大似然法监督分类方法;对一级类型草地采用多尺度分层聚类算法的部分监督分类方法.结果表明,草地信息利用SSHC方法提取结果较好,与Bayes分类方法相比,精度提高4.2%,SSHC所获结果数据Kappa系数为0.84,Bayes所获结果数据Kap-pa系数为0.78.对某专题要素分类,此方法结果较优.     

Generating Surface Models of Population Using Dasymetric Mapping*

Aggregated demographic datasets are associated with analytical and cartographic problems due to the arbitrary nature of areal unit partitioning. This article describes a methodology for generating a surface‐based representation of population that mitigates these problems. This methodology uses dasymetric mapping and incorporates areal weighting and empirical sampling techniques to assess the relationship between categorical ancillary data and population distribution. As a demonstration, a 100‐meter‐resolution population surface is generated from U.S. Census block group data for the southeast Pennsylvania region. Remote‐sensing‐derived urban land‐cover data serve as ancillary data in the dasymetric mapping.     

Analysis of Zoning Pattern of Geochemical Indicators for Targeting of Porphyry-Cu Mineralization: A Pixel-Based Mapping Approach

In this paper, a pixel-based mapping of geochemical anomalies is proposed to avoid estimation errors resulting from using interpolation methods in the modeling of anomalies. The pixel-based method is a discrete field modeling of geochemical landscapes for mapping lithogeochemical anomalies. In this method, the influence area of each composite rock sample is the whole area covered by a pixel where the materials of the sample were taken from. In addition to the pixel-based method, because delineation of mineral exploration target areas using geochemical data is a challenging task, the application of metal zoning concept is demonstrated for vectoring into porphyry mineralization systems. In this regard, different geochemical signatures of the deposit-type sought were mapped in a model. Application of the proposed pixel-based method and the metal zoning concept is a powerful tool for targeting areas with potential for porphyry copper deposits.     

Compositional Balance Analysis: An Elegant Method of Geochemical Pattern Recognition and Anomaly Mapping for Mineral Exploration

Natural Resources Research - Geochemical pattern recognition and anomaly mapping are always involved in the fields of environmental and exploration geochemistry. Principal component analysis (PCA)...     

Future Scenarios of Environmental Vulnerability Mapping Using Grey Analytic Hierarchy Process

Any sustainable resource utilization plan requires evaluation of the present and future environmental impact. The present research focuses on future scenario generation of environmental vulnerability zones based on grey analytic hierarchy process (grey-AHP). Grey-AHP combines the advantages of grey clustering method and the classical analytic hierarchy process (AHP). Environmental vulnerability index (EVI) considers twenty-five natural, environmental and anthropogenic parameters, e.g. soil, geology, aspect, elevation, slope, rainfall, maximum and minimum temperature, normalized difference vegetation index, drainage density, groundwater recharge, groundwater level, groundwater potential, water yield, evapotranspiration, land use/land cover, soil moisture, sediment yield, water stress, water quality, storage capacity, land suitability, population density, road density and normalized difference built-up index. Nine futuristic parameters were used for EVI calculation from the Dynamic Conversion of Land-Use and its Effects, Model for Interdisciplinary Research on Climate 5 and Soil and Water Assessment Tool. The resulting maps were classified into three classes: “high”, “moderate” and “low”. The result shows that the upstream portion of the river basin comes under the high vulnerability zone for the years 2010 and 2030, 2050. The effectiveness of zonation approach was between “better” and “common” classes. Sensitivity analysis was performed for EVI. Field-based soil moisture point data were utilized for validation purpose. The resulting maps provide a guideline for planning of detailed hydrogeological studies.

     

Modeling the Influence of Groundwater Exploitation on Land Subsidence Susceptibility Using Machine Learning Algorithms

Natural Resources Research - Groundwater over-exploitation in arid and semiarid environments has led to many land subsidence cases. Immense economic losses incurred from land subsidence occurrences...