Preprints
https://doi.org/10.5194/gc-2021-22
https://doi.org/10.5194/gc-2021-22

  29 Jul 2021

29 Jul 2021

Review status: this preprint is currently under review for the journal GC.

GC Insights: Identifying conditions that sculpted bedforms – Human insights to build an effective AI

John K. Hillier1, Chris Unsworth2, Luke De Clerk3, and Sergey Savel'ev3 John K. Hillier et al.
  • 1Geography and Environment, Loughborough University, Loughborough, LE1 3TU, UK
  • 2School of Ocean Sciences, Bangor University, Bangor, LL59 5AB, UK
  • 3Dept. Physics, Loughborough University, Loughborough, LE1 3TU, UK

Abstract. 42 survey participants demonstrate that it is visually possible to recognise the type of flow that created bedforms (e.g. sand dunes, riverbed ripples) from short distance-depth profiles, but this is much harder for individual forms. An interpreter's geoscience expertise does not help, indicating a machine learning or 'AI' algorithm might be trained well from the data alone, especially if multiple bedforms are used.

John K. Hillier et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gc-2021-22', Anonymous Referee #1, 09 Aug 2021
    • AC1: 'Reply on RC1', John K. Hillier, 28 Sep 2021
  • RC2: 'Comment on gc-2021-22', Dylan Ward, 01 Sep 2021
    • AC2: 'Reply on RC2', John K. Hillier, 28 Sep 2021

John K. Hillier et al.

John K. Hillier et al.

Viewed

Total article views: 514 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
389 109 16 514 31 5 4
  • HTML: 389
  • PDF: 109
  • XML: 16
  • Total: 514
  • Supplement: 31
  • BibTeX: 5
  • EndNote: 4
Views and downloads (calculated since 29 Jul 2021)
Cumulative views and downloads (calculated since 29 Jul 2021)

Viewed (geographical distribution)

Total article views: 451 (including HTML, PDF, and XML) Thereof 451 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 26 Nov 2021
Special issue
Download
Short summary
It is an aspiration to infer flow conditions from bedform morphology (e.g. riverbed ripples) where sedimentary structures preserve the geological past, or in inaccessible environments (e.g. Mars). This study was motivated by the idea of better designing an 'AI' algorithm by using lessons from non-AI (i.e. human) abilities. Survey participants demonstrate that the task is visually possible, but geoscience expertise does not help, so an algorithm might be trained well from the data alone.
Altmetrics