Re-Review - Bevington et al. “Satellite telemetry of surface ablation to inform spatial melt modelling and event-scale monitoring, Place Glacier, Canada”, submitted to the Cryosphere.
The re-submitted manuscript presents a novel setup of automated ‘smart stakes’ that transmit temperature, humidity and ultrasonic surface height data via Iridium satellite telemetry from the Place Glacier in Canada. The manuscript presents results from a summer campaign during 2024 finding a range of melt factors for snow and ice, which is variable based upon the unique temperature records stemming from each of the four smart stakes. The study finds that such networks can be highly valuable to improve attribution of melt to specific warm events with a near-realtime transmission of data.
The authors now present a much improved modelling approach and consideration of lapse rate relevance and I appreciate their efforts in addressing several of the concerns I and other reviewers had. Nevertheless, I find that the manuscript still contains some unclear statements regarding ETI and lapse rates as well as the occurrence of inversions under heatwave events. The authors have produced some nice new figures of the data and modelling outputs, but the clarity of results and discussion sections still needs improvement to better fit the goal of the manuscript, in my opinion. There are several formatting errors that still need to be addressed and I feel that the authors did not directly answer one or two of the review questions from myself and others. I still consider minor corrections necessary before consideration for publication by the journal.
Specific comments
Original Comment - “The reported wake up times are naturally very short, due to battery life considerations, as is well reported by the authors. The equilibrium response times of the sensor according to the manufacturer are up to 30 seconds for the 63% range, but did the authors make any tests of these values compared to a continuously logging temperature sensor at the other weather stations, for example? Are the sensors also comparing well when placed together? Some short report of this would be beneficial.”
Author Response: "We apologize about this confusion. The battery consumption is a scenario with assumptions on timing. We reworded to add clarity. In reality, the startup time is about 10 seconds, and the Temp/RH measurement takes about 3 seconds, and the ultrasonic is sampled 10 times and the median value is reported. These points are now included in section 3.1 Sensors. "
This is fine, but does not really address my question, which perhaps was not clear. Does the wake up time have any notable impact on the air temperature accuracy when compared to continually logging instrumentation when tested before installation on the glacier? What about a ‘huddle test’ of all 4 smart stakes in the same space… are there large deviations in the recorded temperatures? Essentially, should the reader have faith in the accuracy of the derived temperatures and their differences across the glacier? The authors later remark in their responses (now L151-154) that comparison to other sensors show a correlation of 0.96, but refer to Fig. S2 which is not showing this. When measuring the same (or similar) space, the systematic bias is also important to know. Please revise and take care to check all figure references and formatting.
L318: Fig.S5 deals with the interpolated temperatures, not the interpolated DEMs.
L326-327: The authors should clarify that effects of the katabatic boundary layer are not explicitly resolved, but may be partially accounted for by using on-glacier stations for lapse rates (lapse rate options 2, 4, 5, 6).
Page 17-20 / L361-376: A formatting error has occurred whereby Figure 6 has been replicated 4 times over successive pages and section 5.2 has an absence of text that has also lost its correct formatting seemingly. Please check and revise. It is perhaps a compile error or artefact of file conversion.
L393-394: The authors need to define somewhere clearly what a negative lapse rate refers to (i.e. decrease of air temperature with elevation following the glaciological convention, but not the atmospheric one). The term for “our glacier stations.. “ is imprecise. State clearly if using combination 4 (e.g.) lapse rates experienced frequency temperature inversions (mean value + frequency) related to warm (non-glacier) temperatures (of => X°C). Can the authors show a scatter of lapse rates on glacier vs mean air temperatures off-glacier to make this point clearer?
L399: Dynamics of what?
Fig. 8: This provides a nice summary of how off-glacier data struggle to represent on-glacier temperatures under warm conditions (or “heat events”). I think it would be useful to give some numbers for the amount of melt estimated by the different methods for those conditions. E.g. how much melt would be over-estimate in a season/during heat events by using ECCC and a -6.5°C km1 lapse rate?
L427-429: This segment is very unclear and should be re-written to state clearly what was tested and how/why “We selected the ETI with the on–glacier linear lapse rate as a visual inspection of the ETI with the upper station polynomial lapse rate did not perform well outside of the air temperature observations in the higher elevations of the glacier. “
Figures 10/11: It is useful to see these evaluations, but they should be leveraged better to demonstrate the value of smart stakes. What happens if the authors use the better model (ETI) but using off-glacier AWS only? There has been an established need for representing on-glacier temperatures before (Greuell and Böhm, 1998; Shea and Moore, 2010; Ayala et al., 2015; Shaw et al., 2025), but the smart stakes provide a novel view over a time series to see under which days and which conditions does accounting for local temperatures over glaciers make the most difference, and crucially, by how much? I think exploring the residuals of model-observed mass loss (Fig 10) using a few key model-lapse rate combinations would provide much more wealth of information here.
Fig. 11 clarify that the model period is different in caption
Sect 5.6 can be more concise and clear with the comparison of models + forcing. What is the value of the manual stakes here? Once more, the authors present a clear and useful analysis of the sensor setup and its benefits and drawbacks, but the subsequent analyses like that presented in this section do not lend themselves to a clear result and statement that can be beneficial to the reader.
Fig. 13 - The authors should mask the off-glacier areas and nunataks.
L510: This is a very unclear statement that should be revised: “The importance of these heat events is facilitated by temperature inversions during the heat events (Figure 8), also reported by Ayala et al. (2015).” How have inversions facilitated the importance of heat events? If anything, such near-surface cooling complicates the relationship of glacier response to heat events.
Line 533-535: This sentence belongs at the end of the previous paragraph, as the ETI model described above does separate the sensible and radiative components.
L537: This may still result in a linear lapse rate… just one that is inverted. The authors should clarify that this is substantially different from a linear lapse rate applied from a single AWS off-glacier.
Supplementary
Fig. S1 and elsewhere: I think that some more context is required for the figure captions to meaningfully interpret what they are showing.
Fig S3, compile error for the citation in the figure caption. Is this lapse rate produced from all stations?
Cited works
Ayala, A., Pellicciotti, F., & Shea, J. (2015). Modeling 2m air temperatures over mountain glaciers: Exploring the influence of katabatic cooling and external warming. Journal of Geophysical Research: Atmospheres, 120, 1–19. https://doi.org/https://doi.org/10.1002/2015JD023137
Greuell, Wouter., & Böhm, Reinhard. (1998). 2 m temperatures along melting mid-latitude glaciers , and implications for the sensitivity of the mass balance to variations in temperature. Journal of Glaciology, 44(146), 9–20. https://doi.org/https://doi.org/10.3189/S0022143000002306
Shaw, T. E., Miles, E. S., McCarthy, M., Buri, P., Guyennon, N., Salerno, F., Carturan, L., Brock, B., & Pellicciotti, F. (2025). Mountain glaciers recouple to atmospheric warming over the twenty-first century. Nature Climate Change. https://doi.org/10.1038/s41558-025-02449-0
Shea, J. M., & Moore, R. D. (2010). Prediction of spatially distributed regional-scale fields of air temperature and vapor pressure over mountain glaciers. Journal of Geophysical Research, 115(D23), D23107. https://doi.org/10.1029/2010JD014351 |
Review in attached PDF