Special Issue: Data-Driven Discovery in Geosciences: Opportunities and Challenges

Abstract

With the rapid expansion in big data and artificial intelligence (AI), Earth sciences are undergoing unprecedented advances in data processing and interpretation techniques, as well as in facilitating data-driven discoveries of complex Earth systems. This special collection explores scientific research related to data-driven discoveries in geosciences and provides a timely presentation of progress in developments and/or applications of AI and big data approaches to multiple aspects of geosciences. These include geohazards monitoring, mineral resource exploration, and environmental assessments. We hope this collection will inspire researchers and will transform the work undertaken in the field of data-driven Earth science. While many challenges remain, including the formidable tasks of transforming the deluge of geoscience data into useable information and furthering knowledge via cutting-edge AI techniques, we envision that data-driven discovery will revolutionize conventional methods of observation, analysis, modeling, and prediction in geosciences, and will further advance scientific understanding of our complex Earth system. KeywordsData-driven discovery • Big data • Artificial intelligence • Geosciences B Guoxiong Chen

(D)rifting in the 21st century: Key processes, natural hazards and geo-resources

“…Such satellite imagery (e.g., topography data) for instance allows for automatic interpretation using state-of-the-art algorithms and artificial intelligence (AI) to generate fault maps (Gayrin et al, 2023). Machine learning and AI in general will expand our research options in many different ways (Chen et al, 2023). Even so, such advanced methods will rely on detailed and correct data from the real world, which remains a challenge.…”

A New Anisotropic Singularity Algorithm to Characterize Geo-Chemical Anomalies in the Duolong Mineral District, Tibet, China

“…In this context, the field of fractal/multifractal models has made significant advancements in recent years, due to continuous efforts dedicated to improving these algorithms. By upgrading models and fusing algorithms, there has been progressive improvement in understanding the complexity and precision of mineralization preserved in these mathematical structures [41][42][43][44][45]. Using anisotropy as an example, it represents variations of properties measured along different directions.…”