Surface melt and the importance of water flow –  an analysis based on high-resolution unmanned  aerial vehicle (UAV) data for an Arctic glacier

Bash, Eleanor A.; Moorman, Brian J.

doi:https://doi.org/10.5194/tc-14-549-2020

Articles | Volume 14, issue 2

https://doi.org/10.5194/tc-14-549-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/tc-14-549-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 14, issue 2

Research article

|

12 Feb 2020

Research article |

| 12 Feb 2020

Surface melt and the importance of water flow – an analysis based on high-resolution unmanned aerial vehicle (UAV) data for an Arctic glacier

Eleanor A. Bash and Brian J. Moorman

Download

Final revised paper (published on 12 Feb 2020)
Preprint (discussion started on 23 May 2019)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

SC1: 'Quick thoughts on the melt modelling', Shawn Marshall, 28 May 2019
- AC1: 'Comment Response', Eleanor Bash, 03 Sep 2019
RC1: 'Review of TC-2019-81', Anonymous Referee #1, 06 Jun 2019
- AC2: 'Response to Reviewer #1', Eleanor Bash, 03 Sep 2019
- AC4: 'Additional Response for Reviewer #1', Eleanor Bash, 05 Sep 2019
RC2: 'Review of TC-2019-81', Anonymous Referee #2, 07 Aug 2019
- AC3: 'Response to Reviewer #2', Eleanor Bash, 03 Sep 2019

Peer-review completion

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

ED: Reconsider after major revisions (further review by editor and referees) (09 Sep 2019) by Daniel Farinotti

AR by Eleanor Bash on behalf of the Authors (01 Nov 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (14 Nov 2019) by Daniel Farinotti

RR by Anonymous Referee #3 (29 Nov 2019)

Suggestions for revision or reasons for rejection

Thanks to the authors for the considerable extra work and analysis that went into these revisions. The manuscript is much improved, and I believe the energy-balance analysis, revised treatment of temperature in the ETI model, and revisited albedo model add a lot to the results and analysis. The conclusions - importance of stream channels to ablation and high local variations in ablation - seem robust and are well-supported by the modelling and statistical results. I have a few additional comments and requests for clarification. These are included in the annotated manuscript, where I have highlighted grammatical or minor points, but the most significant points are summarized here, for the authors' consideration.

1. I agree with the conclusion on large local variability in melt rates on glacier surfaces, and yet... when one spends time on a glacier in late summer, in both Arctic and temperate regions, the surface is not typically that variable. Stream channels certainly induce incision and additional melting, below the 'reference surface' of a glacier, but the reference surface is relatively even. Certainly not ~2:1 variations in cumulative summer melt. I wonder if the authors could think about and comment on this in the final conclusions, when it comes to heterogeneity of melt.

2. I like this result of the effects of meltwater features on additional ablation. e.g. p.1, l.14 and in the discussion, but I am wondering if it is possible to separate melting due to streamflow from that due to albedo effects. Water will absorb and also transmit more solar radiation to the ice (the channel bed), giving extra solar and possible thermal melting, if the water warms in the sun. This could contribute to the stream-associated ablation, so it is probably best to mention this as well as the kinetic/mechanical dissipation that is proposed here. Without specific measurements and modelling of these processes, I would recommend just to mention the different possible explanations and not pick a favorite.

3. p.12, This is a good analysis - it shows pretty convincingly that the roughness alone can't explain the discrepancy very well, although part of it.
I worry that much of the systematic underestimation of melt in the distributed models may come from the systematic underestimation of SWin. Is it possible to add a short sensitivity test of this? I would suggest using the observed (AWS) SWin, maybe just adjusting for slope and aspect effects across your study area, vs. the full modelled SW radiation which has significant biases. This could help to say how much of the AWS vs. distributed model results can be explained from this. As it sits now, it looks like the reference AWS site has ~15% more melt energy than the average grid cell in your study area, which is possible but could also just be a reflection of the SW radiation error.

Specific comments

p.1, l.10, it would be useful to also report the total melt over the study period, and/or to report the model confidence as a percentage of the total melt. i.e. the models agree within *% of the observations.

p.3, l.19, uncertainty reported for the melt rate - is this over 3 days? - suggest reporting a per-day rate, for better comparison with the numbers below (0.06 m/d)

p.3, l.21, there should be only 1 value for an average, Do you mean the range of daily melt rates? Or would it be better to report 0.042 +/- 0.012 m/d?

p.3, l.27 and p.4, l.2, suggest using "surface height" rather than "position" for the SR50 signal. Just as you are not looking at (x,y) here, which I think is implied by "position"

p.5, l.22, For clarity, do you mean MJ per 12 hour period, MJ per day, or MJ over your full period?

p.6, l2, suggest \sigma for Stefan-Boltzmann's constant, to be conformist.

p.10, l.20, Can you remind us here about the values of the observed melt at the AWS site, from a) the SR50 and b) the UAV results? This would help with the evaluation.

p.10, l.24, "all four models underestimate melt at the point" - but aren't they tuned at the AWS site, and with a mean 0 residual as noted above? Do you mean underestimate over the study area?

p.12, l.12, also there is a strong diurnal variation in the zenith angle - places with 24-hour light still have strong diurnal cycles, e.g. greater than 2:1 variations in incoming SW radiation at local noon vs. local midnight, depending on the latitude

p.12, l.13, "in. The strength of that variation is much greater in the EB models than the ETI models". This does not make sense to me - don't both models use incoming SW, using the same model for this? As this reads, the shortwave radiation varies more for the EB model than the ETI model. Do you mean the errors or residuals, rather than the SW radiation here?

p13, l.5, here, it is both EB and ETI models that come up short, non? Maybe refer to them as the "distributed melt models" ?

Figure 2 caption. For the SW, it would be helpful to note the duration of time used for the units here MJ/m2) as this is a cumulative value - are these hourly values? In which case it is cumulative or total hourly (vs. hourly averaged) SW. It is more like PDD with MJ.

Figure 4b - is there a reason to show just the second half of July here? it works fine, maybe just not clear why some things are all of July and some just the second half.

Referee Report: tc-2019-81-referee-report.pdf

Hide

ED: Publish subject to minor revisions (review by editor) (03 Dec 2019) by Daniel Farinotti

AR by Eleanor Bash on behalf of the Authors (16 Dec 2019) Author's response Manuscript

ED: Publish subject to technical corrections (12 Jan 2020) by Daniel Farinotti

Dear authors,

thank you for submitting your revised manuscript.
At this stage, I’m left with what TC calls “technical corrections”. They are outlined here below and should not require any major effort to address. The most important point is the wording of the new title.

New title: I’m happy to see the title changed but I’m not sure to understand what “gap” the new title is referring to. I particularly fear that the meaning will be hard to grasp for the readership. Is one of the following suggestions capturing your message? If not, can you please propose an alternative?
i) High-resolution UAV data from an Arctic glacier reveals the role of water flow in surface melt
ii) The role of melt induced by surface water flow: An assessment based on high-resolution UAV data from an Arctic glacier
iii) Surface melt and the importance of water flow – an analysis based on high-resolution UAV data for an Arctic glacier

Reply to reviewer’s comment 2: I understand the answer and agree with it. Yet, I was wondering whether adding “[…] including the potential role of the lower water albedo when compared to ice.” at the end of P15 L23 would allow you to accommodate the reviewer’s comment in a more “visible” manner?

Reply to reviewer’s comment 3: I’m ready to accept your answer and not to insist on the sensitivity analysis asked by the reviewer. It seems to make sense, however, to add your interpretation (something like your sentence “the discrepancy between measured and modelled melt is greater across the grid than at the AWS site, indicating the difference is not due to solar radiation alone” is all what is needed) somewhere in the manuscript. Can you do so?

P10 L16: “for the study period” (“the” is missing)

Table 2: Please use the table’s caption to define “Residual \sigma”. I assume it is the standard deviation of the residuals as defined at P9 L10?

I’m looking forward to see the above points clarified and the paper published.
Best wishes.
Daniel Farinotti

Hide

AR by Eleanor Bash on behalf of the Authors (14 Jan 2020) Author's response Manuscript

Short summary

High-resolution measurements from unmanned aerial vehicle (UAV) imagery allowed for examination of glacier melt model performance in detail at Fountain Glacier. This work capitalized on distributed measurements at 10 cm resolution to look at the spatial distribution of model errors in the ablation zone. Although the model agreed with measurements on average, strong correlation was found with surface water. The results highlight the contribution of surface water flow to melt at this location.