地学前缘 ›› 2025, Vol. 32 ›› Issue (6): 396-410.DOI: 10.13745/j.esf.sf.2025.9.68-en

• 非主题来稿选登 • 上一篇    下一篇

Earth Science in the Era of Foundation Models: How AlphaEarth is Reshaping Quantitative Geoscience

CHENG Qiuming1,2(), YANG Yilin3, ZHOU Yuanzhi1,2, ZHANG Yuanzhi1,4   

  1. 1.State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences (Beijing), Beijing 100083, China
    2.Frontiers Science Center for Deep-time Digital Earth, China University of Geosciences (Beijing), Beijing 100083, China
    3.School of Earth Sciences and Engineering, SunYat-sen University, Zhuhai 519080, China
    4.School of Marine Sciences, Nanjing University of Information Science & Technology, Nanjing 210044, China
  • 收稿日期:2025-09-18 接受日期:2025-09-21 出版日期:2025-11-25 发布日期:2025-11-12

Earth Science in the Era of Foundation Models: How AlphaEarth is Reshaping Quantitative Geoscience

CHENG Qiuming1,2(), YANG Yilin3, ZHOU Yuanzhi1,2, ZHANG Yuanzhi1,4   

  1. 1. State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences (Beijing), Beijing 100083, China
    2. Frontiers Science Center for Deep-time Digital Earth, China University of Geosciences (Beijing), Beijing 100083, China
    3. School of Earth Sciences and Engineering, SunYat-sen University, Zhuhai 519080, China
    4. School of Marine Sciences, Nanjing University of Information Science & Technology, Nanjing 210044, China
  • Received:2025-09-18 Accepted:2025-09-21 Online:2025-11-25 Published:2025-11-12
  • About author:CHENG Qiuming (1960—), male, professor, Ph.D. supervisor, and academician of the Chinese Academy of Sciences, specializes in research on mathematical geosciences. E-mail: qiuming.cheng@iugs.org
  • Supported by:
    National Natural Science Foundation of China Key Project(42050103);Higher Education Disciplinary Innovation Program(B25052);Guangdong Pearl River Talent Program Innovative and Entrepreneurial Team Project(2021ZT09H399);Ministry of Education’s Frontiers Science Center for Deep-Time Digital Earth (DDE)(2652023001);Geological Survey Project of China Geological Survey(DD20240206201)

RichHTML

1

PDF (PC)

10

摘要：

Since the beginning of the 21st century, advances in big data and artificial intelligence have driven a paradigm shift in the geosciences, moving the field from qualitative descriptions toward quantitative analysis, from observing phenomena to uncovering underlying mechanisms, from regional-scale investigations to global perspectives, and from experience-based inference toward data- and model-enabled intelligent prediction.

AlphaEarth Foundations (AEF) is a next-generation geospatial intelligence platform that addresses these changes by introducing a unified 64-dimensional shared embedding space, enabling—for the first time—standardized representation and seamless integration of 12 distinct types of Earth observation data, including optical, radar, and lidar. This framework significantly improves data assimilation efficiency and resolves the persistent problem of “data silos” in geoscience research. AEF is helping redefine research methodologies and fostering breakthroughs, particularly in quantitative Earth system science. This paper systematically examines how AEF’s innovative architecture—featuring multi-source data fusion, high-dimensional feature representation learning, and a scalable computational framework—facilitates intelligent, precise, and real-time data-driven geoscientific research. Using case studies from resource and environmental applications, we demonstrate AEF’s broad potential and identify emerging innovation needs. Our findings show that AEF not only enhances the efficiency of solving traditional geoscientific problems but also stimulates novel research directions and methodological approaches.

关键词: large-scale models, artificial intelligence, mineral prospectivity mapping, AlphaEarth, knowledge graphs, deep and covered mineral exploration

Abstract:

Since the beginning of the 21st century, advances in big data and artificial intelligence have driven a paradigm shift in the geosciences, moving the field from qualitative descriptions toward quantitative analysis, from observing phenomena to uncovering underlying mechanisms, from regional-scale investigations to global perspectives, and from experience-based inference toward data- and model-enabled intelligent prediction.

AlphaEarth Foundations (AEF) is a next-generation geospatial intelligence platform that addresses these changes by introducing a unified 64-dimensional shared embedding space, enabling—for the first time—standardized representation and seamless integration of 12 distinct types of Earth observation data, including optical, radar, and lidar. This framework significantly improves data assimilation efficiency and resolves the persistent problem of “data silos” in geoscience research. AEF is helping redefine research methodologies and fostering breakthroughs, particularly in quantitative Earth system science. This paper systematically examines how AEF’s innovative architecture—featuring multi-source data fusion, high-dimensional feature representation learning, and a scalable computational framework—facilitates intelligent, precise, and real-time data-driven geoscientific research. Using case studies from resource and environmental applications, we demonstrate AEF’s broad potential and identify emerging innovation needs. Our findings show that AEF not only enhances the efficiency of solving traditional geoscientific problems but also stimulates novel research directions and methodological approaches.

Key words: large-scale models, artificial intelligence, mineral prospectivity mapping, AlphaEarth, knowledge graphs, deep and covered mineral exploration

中图分类号: