EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning‐Based Tracking of Subduction Zones in Mantle Convection Models

H. Choi, B. J. Foley

Journal of Geophysical Research: Solid Earth · 2026-02 · 131 (2)

DOI: 10.1029/2025jb032355 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Geophysical Research: Solid Earth, online erschienen am 24.02.2026. Verfasst von H. Choi und 1 weiteren Autor:innen. Die Arbeit nennt 78 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Directorate for Geosciences.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Subduction zones are central to Earth's tectonic evolution. Previously, identification and tracking of subduction zones in numerical mantle convection models have relied on methods such as detecting gradients in horizontal surface velocity. However, these approaches require the use of arbitrary thresholds and often lack sufficient spatial context, making robust and consistent detection challenging. This study adapts an established deep‐learning workflow for subduction zone (SZ) detection using Fully Convolutional Networks (FCNs), which perform semantic segmentation on RGB images constructed from temperature, vertical velocity, and fineness fields derived from 2D geodynamic simulations. Trained on a curated data set with high‐resolution ground truth masks, the FCN effectively identifies SZs with improved spatial coherence and robustness compared to traditional surface velocity convergence methods. Incorporating an attention‐based architecture and a custom loss function, the FCN captures complex SZ morphologies, reduces false detections, and enables consistent tracking of SZ evolution over time. Statistical analysis of tracked SZs reveals that those forming near continental margins are longer‐lived and less sensitive to mantle convective vigor, while oceanic SZs are shorter‐lived and strongly influenced by convective vigor. These findings support the hypothesis that continental lithosphere plays a stabilizing role in early Earth subduction dynamics. This FCN‐based method provides a threshold‐free, data‐driven alternative for SZ detection and offers a powerful tool for analyzing subduction processes in Earth and other planetary interiors.

Förderung

  • Directorate for Geosciences

Zitieren

APA

Choi, H., & Foley, B. J. (2026). Deep Learning‐Based Tracking of Subduction Zones in Mantle Convection Models. Journal of Geophysical Research: Solid Earth, 131(2). https://doi.org/10.1029/2025jb032355

Vancouver

Choi H, Foley BJ. Deep Learning‐Based Tracking of Subduction Zones in Mantle Convection Models. Journal of Geophysical Research: Solid Earth. 2026;131(2). doi:10.1029/2025jb032355

BibTeX
@article{Choi2026,
  author = {Choi, H. and Foley, B. J.},
  title = {Deep Learning‐Based Tracking of Subduction Zones in Mantle Convection Models},
  journal = {Journal of Geophysical Research: Solid Earth},
  year = {2026},
  volume = {131},
  number = {2},
  doi = {10.1029/2025jb032355}
}
Kostenlos starten →