Processes influencing heat transfer in the near-surface ice of Greenland's ablation zone

Hills, Benjamin H.; Harper, Joel T.; Meierbachtol, Toby W.; Johnson, Jesse V.; Humphrey, Neil F.; Wright, Patrick J.

doi:https://doi.org/10.5194/tc-12-3215-2018

Articles | Volume 12, issue 10

https://doi.org/10.5194/tc-12-3215-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/tc-12-3215-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 12, issue 10

Research article

|

08 Oct 2018

Research article |

| 08 Oct 2018

Processes influencing heat transfer in the near-surface ice of Greenland's ablation zone

Benjamin H. Hills, Joel T. Harper, Toby W. Meierbachtol, Jesse V. Johnson, Neil F. Humphrey, and Patrick J. Wright

Download

Final revised paper (published on 08 Oct 2018)
Supplement to the final revised paper
Preprint (discussion started on 02 May 2018)
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 Hills et al. 2018: Processes influencing near-surface heat transfer in Greenland's ablation zone, MS No.: tc-2018-51', Anonymous Referee #1, 24 Jun 2018
- AC1: 'Response to Anonymous Referee #1', Benjamin Hills, 30 Jul 2018
RC2: 'review Hills et al', Anonymous Referee #2, 02 Jul 2018
- AC2: 'Response to Anonymous Referee #2', Benjamin Hills, 30 Jul 2018

Peer-review completion

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

AR by Benjamin Hills on behalf of the Authors (30 Jul 2018) Author's response

ED: Referee Nomination & Report Request started (14 Aug 2018) by Tobias Sauter

RR by Anonymous Referee #1 (02 Sep 2018)

Suggestions for revision or reasons for rejection

Comment to revised manuscript tc-2018-51-AC1-supplement
Hills et al. 2018: Processes influencing heat transfer in the near-surface ice of Greenland’s ablation zone
The main concerns of both reviewers addressed the definition of the metric T0, consistent depth referencing of observed and calculated vertical profiles, aspects of forcing data and lower boundary condition and enhanced discussion of subsurface refreezing processes. Overall, the the manuscript was substantially improved with respect to these issues, thanks for that effort. The following comments address some remaining issues which I suggest to reconsider in more detail.
1) Line above P2L15: I am still not confident regarding “T0”, which is the essential metric used in this work. Here the authors state “Seasonal air temperature oscillations are diminished with depth into the ice, until they are negligible (i.e. ~1%) at a 'depth of zero annual amplitude'. On the other hand it is stated that “The mean value from the lowermost sensor (analogous to T0 ) is -3.2 at 27-km, … (P4L19, which refers to actual evaluation practice) “. Methodically seen this leaves kind of gap i.e., to show that at this lowermost sensor postion the 1% criterium is acutally matched (on average at least). Additionally this issue also questions in what extent the pragmatic choice of the lowermost sensor position is justified in view of the fact that right this level is intrinsically influenced by the lower boundary condition (where no temperature variability can occur by definition). Hence this question indirectly leads back to the one whether the simulations should not build on a deeper domain. I still recommend performing a sensitivity study addressing these issues and related uncertainties.

2) I recognize the added details concerning the meteorological data. Concerning the ice temperature measurements close to the surface some critical remarks may be given, though. If thermistors were housed in a “black casing” (as stated now, P3L30), than not only sensors lying at the surface may have been affected by solar induced heating. Dark cables/casing can experience significantly enhanced temperature and effect sensors (also by conduction within cables due to temperature gradients). Glendinning and Morris (1999) demonstrated for snow that corresponding effects can be of order >2°C @70cm depth. The indicated "discarding" procedure helps identifying problems close to 0°C, but does not identify/correct effects on sensors having negative temperatures. Overall one might expect that thermistors in the upper ca. 50cm below the surface are prone to a warm bias during summer (which can not be excluded by observations that at a certain point of time cables were found frozen into the ice.

3) The authors multiply mention that the paper is not focussing on meteorological aspects, which is fine. However, this not acceptable regarding obseved air temperature which constitutes essential model input and is used in context of interpretation of the results. I reckognize more detailed information on used instruments and their setup. But the potential influence of a likely inefficient radiation shield on the temperature measurements is still understated. According to the now given Fig. S2, a rather ineffective shield was used and significant radiation errors may be expected (also because being mounted close i.e., ca. 0.5m to the strongly reflecting surface and the low incidence angle of solar radiaton during transitional seasons). There are several studies incl. manufacturer statements, that this kind of screen can induce significant errors in temperature measurements (several °C depending on wind speed, too). Unfortunately, these effects are hard to quantify or to correct. At least, however, one expects some more critical comments that such uncertainties are inherent in the data and were not corrected. The currently used air temperature data are likely to be too high, which shall be discussed in the interpretation of the results, too.
Comparison to PROMICE data is ok (Fig. S4) is not really valuable in this context (due large distance between sites and respective need to correct for elevation differences). Regarding calculation of surface temperatures, the used emissivity shall be specified.

4) Revision of Fig. 2 is acknowledged, however, I still can not fully agree to the argument that different i.e. inconsistent record lengths do not affect calculation of T0 (“… should be comparable because it is below any seasonal variation ..”)
Further the caption mentions " For field sites at which the air temperature was measured for at least a full year, a dashed line shows the mean air temperature". Why then still showing dashed lines for b, de and f, which do not cover a full year?
Overall, I do not fully support diverse argumetns why uncertainties in air temperature measurements and its use as model input is not an issue in context of this investigation. But I also see the weak point that the vailable data hardly allow a better approach. In this perspective, corresponding sensitivity studies could have been valuable. This is menat in sense of disturbing input (i.e., air temperature) and investigate the impact on simulatin results (T0 basically).

5) P5L16: …”Net radiation is less than zero in the winter (net outgoing because of thermal emission in the infrared wavelengths)”, may be reformulated to account for the fact that not only emission counts, but that this component emission prevails over atmospheric input)

6) P6L8: “…. model uses measured meteorological variables as the surface boundary condition and simulates ice temperature to 21 m, a depth chosen for consistency with measured data. The ice temperature at the depth of zero annual amplitude, T0 , is output from the bottom of the domain for each model experiment and used as a metric….” Admittently, I am still not convinced about several aspects in this context as long as respective uncertainties are not addressed quantitatively. In particular this concerns use of air temperature as forcing at the upper boundary (ignoring measurements uncertainties and stratification effects) and the implementation of the lower boundary condition at a depth close to the depth of average T0. Both is still rather superficial treated.

P9L4: I need some help how of Phi(rad) in given dimensions is compatible with equ.1

P10L15: “The limiting cases show that this bottom boundary condition strongly controls the near-surface temperature, with a range in the resulting T0 values from -17.0°C to -2.0°C. In summary, measured ice temperatures are consistently warmer than both the measured air temperature and simulated ice temperature …” This is an expected and most important result, which has to be re-emphasized in the discussions, too. And again, it would be most interesting to know in what extent this issue depends on alternative depth of the model domain and corresponding specification of the lower boundary condition.

Hide

ED: Publish subject to minor revisions (review by editor) (06 Sep 2018) by Tobias Sauter

AR by Benjamin Hills on behalf of the Authors (16 Sep 2018) Author's response

ED: Publish as is (21 Sep 2018) by Tobias Sauter

AR by Benjamin Hills on behalf of the Authors (23 Sep 2018)

Short summary

At its surface, an ice sheet is closely connected to the climate. Assessing heat transfer between near-surface ice and the overlying atmosphere is important for understanding how the ice sheet is melting at the surface. We measured ice temperature within 20 m of the surface of the Greenland Ice Sheet. Resulting ice temperatures are warmer than the air, a peculiar result which implies the role of some nonconductive heat transfer processes such as latent heating by refreezing meltwater.