Calving Front Machine (CALFIN): glacial termini dataset and automated deep learning extraction method for Greenland, 1972–2019

Cheng, Daniel; Hayes, Wayne; Larour, Eric; Mohajerani, Yara; Wood, Michael; Velicogna, Isabella; Rignot, Eric

doi:https://doi.org/10.5194/tc-15-1663-2021

Articles | Volume 15, issue 3

https://doi.org/10.5194/tc-15-1663-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/tc-15-1663-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 15, issue 3

Research article

|

01 Apr 2021

Research article |

| 01 Apr 2021

Calving Front Machine (CALFIN): glacial termini dataset and automated deep learning extraction method for Greenland, 1972–2019

Daniel Cheng, Wayne Hayes, Eric Larour, Yara Mohajerani, Michael Wood, Isabella Velicogna, and Eric Rignot

Download

Final revised paper (published on 01 Apr 2021)
Supplement to the final revised paper
Preprint (discussion started on 14 Oct 2020)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of „Calving Front Machine (CALFIN): Glacial Termini Dataset and Automated Deep Learning Extraction Method for Greenland, 1972-2019”', Anonymous Referee #1, 09 Nov 2020
- AC1: 'Reply on Reviewer 1', Daniel Cheng, 18 Jan 2021
RC2: 'Interactive comment on “Calving Front Machine (CALFIN): Glacial Termini Dataset and Automated Deep Learning Extraction Method for Greenland, 1972–2019” by Daniel Cheng et al.', Anonymous Referee #2, 11 Nov 2020
- AC2: 'Reply on Reviewer 2', Daniel Cheng, 18 Jan 2021
AC3: 'Revised manuscript with changes highlighted', Daniel Cheng, 18 Jan 2021

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (further review by editor and referees) (18 Jan 2021) by Stef Lhermitte

AR by Daniel Cheng on behalf of the Authors (18 Jan 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (18 Jan 2021) by Stef Lhermitte

RR by Anonymous Referee #2 (20 Jan 2021)

RR by Anonymous Referee #1 (25 Jan 2021)

Suggestions for revision or reasons for rejection

The revised version of the manuscript from Cheng et al. improved significantly. The authors addressed all concerns appropriately and were able to increase the quality of the paper in terms of visualization and written explanation. The revised abstract addresses a broader audience and the restructuring of the chapters allows for a better understanding of the work performed. Especially, the added Flowchart (Figure 3) as well as Figures 13 & 14 highlight the developed methodology and amount of processed data. Finally, the revised manuscript is in line with the great methodological approach of this study.
Only some minor remarks remain which are indicated below:
P1L6: change “average 86.76 meters ± 1.43 m” to “average 86.76± 1.43 meters”
P2L31: Also mention TerraSAR-X validation data used for the comparison with the study from Zhang et al.
P6L5: Why were rotations up to 12° chosen?
P8L9: can be used
Figure 13: This Figure represents a great improvement and shows the amount of processed data. Unfortunately, it is difficult to read the legends. Maybe you could put a larger version of this figure in the supplements. For example, split up the figure and present each glacier individually.
Figure 14: This is a great overview of glacier changes for entire Greenland. In my opinion, the overall differences between the basins count more than the seasonal variability in this Figure. Therefore, I would recommend to use the same vertical axis scaling. This would highlight the differences in retreat for each basin at first glance (e.g. 6 km in SE versus 1.6 km in SW).
Conclusion: Most readers tend to read the Abstract and Conclusion only. You could consider including the main findings of your study in the conclusion (e.g. accuracy 86.76± 1.43, numberof studied glaciers, good performance compared to validation data, absolute retreat differences in all Greenlandic basins etc.).

Hide

ED: Publish subject to minor revisions (review by editor) (28 Jan 2021) by Stef Lhermitte

AR by Daniel Cheng on behalf of the Authors (10 Feb 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (10 Feb 2021) by Stef Lhermitte

AR by Daniel Cheng on behalf of the Authors (18 Feb 2021) Manuscript

Short summary

Tracking changes in Greenland's glaciers is important for understanding Earth's climate, but it is time consuming to do so by hand. We train a program, called CALFIN, to automatically track these changes with human levels of accuracy. CALFIN is a special type of program called a neural network. This method can be applied to other glaciers and eventually other tracking tasks. This will enhance our understanding of the Greenland Ice Sheet and permit better models of Earth's climate.