底青云,赵亮,郑忆康,耿智,喻志超,单小彩,李超,徐志尧,吕鹏飞.人工智能赋能深地科学的关键挑战、重大场景与发展路径[J].中国科学院院刊,2026,41(6):1193-1205.

人工智能赋能深地科学的关键挑战、重大场景与发展路径

Artificial intelligence for deep Earth science: Key challenges, major application scenarios and development pathways

底青云^1,2*
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
DI Qingyun^1,2*
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
赵亮^1,2
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
ZHAO Liang^1,2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
郑忆康^1,2
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
ZHENG Yikang^1,2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
耿智^1,2
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
GENG Zhi^1,2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
喻志超^1,2
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
YU Zhichao^1,2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
单小彩^1,2
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
SHAN Xiaocai^1,2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
李超^1,2
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
LI Chao^1,2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
徐志尧^1,2
中国科学院地质与地球物理研究所 北京 100029;中国科学院大学 地球与行星科学学院 北京 100049
XU Zhiyao^1,2
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China;College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
吕鹏飞¹
中国科学院地质与地球物理研究所 北京 100029
LV Pengfei¹
Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China

         深地科学;多圈层耦合;人工智能;研究范式;资源探测;深地工程;地月系统

        deep Earth science;multi-sphere interactions;artificial intelligence;research paradigm;resource exploration;deep underground engineering;Earth-Moon system

        深地科学是揭示地球内部结构与多圈层耦合演化规律、保障国家能源与关键矿产资源安全，以及提升重大灾害防控能力的战略前沿领域。然而，受限于极端环境下原位观测能力不足、多源异构数据难以高效融合、复杂多物理场耦合过程建模能力薄弱等瓶颈，传统研究范式已难以支撑深地科学向更深层次发展。人工智能技术的快速发展为突破上述瓶颈提供了新的路径：以数据驱动与机理认知融合为核心的新型研究范式，正推动深地科学从“经验解释”向“智能认知”转变，从“单点分析”向“系统推演”演进，并加速构建“感知—建模—推演—决策—控制”的全链条体系。在此背景下，文章系统梳理人工智能赋能深地科学的研究范式变革与关键技术体系，分析国内外发展态势，重点探讨其在深地资源探测与开发、深地工程建造与运维、深地灾害感知与预警，以及地月空间探测与认知等典型场景中的应用潜力。面向未来发展需求，从深地探测数据支撑设施、学科基础理论、关键技术与装备、应用场景平台和组织实施机制等方面提出发展布局建议，旨在为相关领域的科技决策和战略部署提供参考。

        Deep Earth science is central to understanding Earth’s internal architecture and the coupled evolution of its major spheres, while also underpinning energy security, the supply of critical mineral resources, and resilience to major geohazards. Nevertheless, the advancement of deep Earth science is currently hindered by insufficient in situ observations under extreme conditions, the difficulty of integrating multi-source heterogeneous data, and the limited capability to model complex multiphysics coupling processes. Recent advances in artificial intelligence offer a potential route beyond these limitations. By integrating data-driven learning with physical and geological understanding, AI is reshaping deep Earth science from empirical interpretation to intelligent discovery, and from local analysis to system-level inference. This emerging paradigm is accelerating the development of a comprehensive full-chain system spanning sensing, modelling, prediction, decision-making, and control. This study systematically examines the paradigm shift and technological foundations of AI-enabled deep Earth science, assesses global research trends, and highlights emerging opportunities in deep resource exploration and development, deep underground engineering, deep geohazard monitoring and early warning, and Earth-Moon exploration. We further outline strategic priorities for future development across deep Earth exploration data infrastructure, foundational theory, enabling technologies and instrumentation, application platforms, and implementation mechanisms. Together, these perspectives provide a framework for advancing deep Earth science in the era of AI.