基于大数据的成因矿物学研究思考

在国际地科联启动深时数字地球大科学计划背景下,开展成因矿物学大数据平台建设和数据深度挖掘和研究,具有十分重要的意义。建议优先考虑矿物系统发生史、矿物标型和矿物成因分类3个方面的数据平台建设,开展3个方面的大数据模型、大数据处理的方法研究和大数据结果的信息提取。部署开展基于大数据的若干国家战略性关键金属元素矿物(如锂矿物、铀矿物、镓矿物、铈矿物、铂矿物等)的系统发生史和矿物类的系统发生史研究,分析其在不同地质历史时期和地球不同构造单元中的聚散规律,为战略性关键金属找矿预测提供思路。根据矿物大数据库的深度挖掘和矿物系统发生史研究所提供的有关矿物演化的规律,揭示地球历史中重要地球物理、地球化学和地球生物事件的性质、分布、规模及辐射效应。借助矿物成因分类的大数据平台,完善“显性”成因矿物族分类和区划图编制,在此基础上,开展基于大数据的“隐性”成因矿物族和找矿矿物族的分类和区划图研制。要重视矿物学和地质大数据研究的复合型人才培养,在开设地质大数据本科专业基础上,在研究生层次开设“成因矿物学大数据”研究方向。

Considerations on big data-based genetic mineralogical research

With the launching of the Deep-time Digital Earth program (DDE) by the International Union of Geosciences, it is of great significance to carry out the construction of big data platforms for genetic mineralogy and in-depth data mining research. We suggest that priority should be given to constructing big data platform for mineral phylogenetic history, mineral typomorphism and mineral genetic classification researches, which involves building big data models, developing big data processing methods, and extracting information from the big data processing results. Big data-based phylogenetic history (or mineral evolution) research should be conducted on several strategic key metals (e.g, lithium, gallium, uranium, cerium and platinum, etc.) and mineral classes to analyze their accumulation and dispersion patterns in different tectonic units and geological eras as a basis for key metal ore prediction. It is possible to reveal the nature, distribution, scale and radiation effect of important geochemical, geophysical and biological events during the Earth’s geologic history by studying mineral evolution through in-depth mining of mineralogical big data and researching mineral phylogenetic history. Big data platform for genetic mineralogical research should be employed in finalizing the classification of “explicit” genetic mineral groups and related map compilation and, later on, in carrying out the classification of “implicit” genetic mineral groups and related map compilation. Attention should be paid to cultivating talents for big data-based interdisciplinary mineralogical research, while big data-based genetic mineralogical research should be pursued at the graduate level.