Dear D. Parkes and co-authors,
Your manuscript received 3 detailed but critical reviews that both highlighted the importance of the general topic of the paper modelling global/regional and long-term (millennia scale) glacier-climate interaction and comparison with long-term length records. Further the topic of the manuscripts seems clearly within the scope of TC.
However, besides some minor editing corrections all 3 referees consistently thought that while the study is very interesting it does not analyse the results to the full potential and raised some further more major issues/weaknesses (of which some I already indicated in my access review) that I briefly list below:
1) It requires clarification of the description of the OGGM processes in particular the treatment of the forcings and calibration of the model,
2) It requires clarification focus and scope of study of the study (which the authors decided in the response (well justified) to put on the long-term glacier modelling rather than the temp. versus precip. forcing),
3) related to the above, clearer and more substantial conclusions are required that are supported by the results
4) Improve the analysis related interpretation (and thereby use the potential of the results better). This specifically should include: (i) a more quantitative analysis of the modelled length response against the observations, (ii) present, analyse and discuss the results beyond just a 'mean' regional response (which should also include the issue of varying response time), (iii) and to better discuss the representativness of the sampled glaciers and better consider the sampling issue (changing number of glaciers in observational length dataset through time and the related spikes).

Given these critical points and concerns above the manuscript requires major revisions. Based on your author response it seems that you seem to be able to address these issues in a revised version of the manuscript within the standard time given for revisions.
From the information I have from the response, I judge, that a revised manuscript (as indicated in the response) should be able to address the major points by the referees and have the potential to be accepted. However, as the asked revisions were rather major, the paper will likely be sent for review again.
Please address all the raised points by the referees point by point in the revised manuscript. Beyond this, I would like to specifically mention the point on just presenting 'mean' regional results. In the suggested revision 2 new figures are presented which address this issue at least partly (for trend etc), but I am not sure if this is enough given the comments by the referees. I understand stand that plotting the length response curves for all individual glaciers and regions for all models is not possible in one figure, but maybe an additional figure could be added in the supplements (similar to revised fig. 2) that shows the spread/variations (as a shaded band) for each model and region.

I would like to thank the authors for their careful and very thorough treatment and addressing of the referee comment in their response and look forward on the revised version of the manuscript.

Andreas Vieli, editor
19 March 2020

Dear D. Parkes and co-authors,
Your revised manuscript (in response to 3 detailed reviews) has been re-reviewed by 2 referees (one from the first round (Ref3), one new referee (Ref4)).
I agree with the view of the two referees that most substantial issues from the first set of reviews and some points I made as an editor have been well addressed which include:
" clarification focus and scope of study
" clearer and more substantial conclusions
" Improve the analysis related interpretation (more quantitative, additional figures beyond regional average, discussion of representativness).
The model and OGGM processes have also been somewhat improved compared to the first version but there still remain quite a few issues with regard to method and experimental setup which are in detail listed in Review4 and that need clarification. Essentially these include:
" clarification on the model type early on (flowline etc…)
" the treatment of precipitation, lapse rates, …
" clarification in use of CRU data
" structural aspects in model description and experimental setup
An additional point by referee 3 on the adjustment of the title should also be considered, but I think the authors should at the end be happy with their title.

Considering all the reviews and revisions, I consider this manuscript not far from being accepted for publication. However, some revisions, mostly with regard to model and experimental description as well as a few very minor editing issues/typos are still needed.
I therefore ask you to carefully consider to points made by the refrees 3 and 4 (see list below) and the very additional minor points made by myself as an editor when reading the revised manuscript. Please also provide a point by point response to the points with your revised manuscript.
When revised satisfactorily, this manuscript will provide a very valuable contribution to the general topic of global/regional/local glacier modelling and long-term glacier-climate interaction and I thank the authors for their good handling and cooperation in the reviewing process.
Andreas Vieli, editor
5 July 2020


Referee 3:
'Parkes and Goosse have substantially reworked their manuscript and have this so in a very elaborate in meticulous way. The issues that I raised during the first round of review, combined with suggestions from other reviewers (which also seem to have been addressed in great detail), have all been addressed. Through this, the quality of the manuscript has, in my opinion, greatly improved. The new figures (especially P3 and P4!), as well as the more elaborate explanation on the pre-frontal glacier dynamics, are a very welcome addition.

In summary, I am confident that this study will be of great interest to our scientific community. I have no remarks left on the text. The only issue that I still see relates to the authors referring to the study as 'regional' in the title of the publication. I agree with the authors that they are 'regional' as we can be', but I nevertheless find the title still misleading. As previously suggested, I would recommend omitting the 'regional' from the title, or replace 'regional' by something in the line of: 'in various regions around the Globe'.
I look forward to seeing this study in TC and congratulate the authors once again on their work.'


Referee 4:
I agree with the previous reviewers, that the questions that are being addressed in the
manuscript are very interesting and fit to the Cryosphere. Significant improvements have
been made based to the manuscript based on the previous reviews. Unfortunately the
improvements are in my opinion insufficient to recommend publication in its current state. I
like the revised introduction and most of it is easy to read, minor comments you will find
below. I have however trouble understanding the experimental set-up that was used. The
methods need to be clarified, as there are a couple of things that seem inconsistent and/or
confusing to me. This is being specified below, together with some other detailed comments
on the section. I will keep the remaining part of this review very brief, because based on the
method description, I am not certain about what the model set-up is and therefore on how to
value the results.
Without a version with marked changes of the manuscript it was very hard to track what
improvements have been maybe. However I found some comments that neither have been
addressed in the rebuttal letters nor implemented in the revised manuscript. I would
recommend the authors to go over the reviews again and check if they addressed all
comments. Though this maybe sounds all very negative, I think the changes that are required might be easy to implement and I see potential in the paper.

2. Data and model description
2.1 As you're analyzing changes in glacier length, I would recommend to spend a few words
on describing the dynamic part of the model in this section. (At it is current state the reader
will only find out it is a flowline model in the "Limitation and Extensions" section.)

Line 88: "OGGM is an open source model of glacier evolution, which couples a surface mass
balance (SMB) model based on solid precipitation and temperature with a model of glacier
dynamics."
This sentence makes me wonder, did you force your simulations with solid precipitation? If so, why, did you do so? (OGGM is calibrated with total precipitation and computes the solid
fraction based on that.) Please clarify the type of precipitation you used, either here or in line 111 for instance.

Line 94: " using CRU 3.21 temperature and precipitation data for the 20th century (Jones and Harris, 2013) " OGGM uses CRU TS v4.01 (Harris et al., 2014) anomalies with respect to the CRU CL (New et al., 2002) in its calibration procedure (see Maussion et al., 2019 Appendix A for the details).
In this context I am wondering, to which CRU data are you referring in the section Line
111-115. Is this the by OGGM preprocessed CRU climate data? If so this needs to be clearly
indicated. If not, why do you use a different climate than was used in the calibration
procedure, when applying the anomaly method to the GCM data?

Line 109: Again be more specific. What type of temperature (reference height temperature?) and precipitation (total/ solid) did you use?

Line 115-116: " The processing is done within OGGM, as its default setting. " and Line 119,
" using OGGM's default preprocessing at level 3" contradict what is being stated in line
108-110. " The GCMs provide gridded monthly records of temperature and precipitation, which OGGM uses to determine temperature and precipitation at each specific glacier location using locally-calculated lapse rates for each. "
OGGM uses by default a lapse rate of 6.5 C/km. This makes me wonder, did you use a
different lapse rate in your forward runs than was used in the calibration? Additionally I
wonder, where did you get the "locally calculated lapse rate" from? (To my knowledge this is not an implemented option in the OGGM function that process the GCM data.)

Line 118-123: It is odd to start the "Glacier observations" section with the model set-up. The
remaining part of this sections is in contrast to the first part of the method section written
less messy and more easy to follow.

Line 123: Maybe you could mention the inversion method somewhere.
Fig. 2: Consider adding the names of the regions in the legend.

Fig. 2: I think the word "number" is missing before "of glaciers".

3. Experimental description
Section 2 and 3: The order in which the methods are described, was inconvenient for me to
read, mostly because parts of the experimental set-up is described spread throughout the
"data and model description" section and the other parts are in the "experimental description".
I recommend to make section 2 strictly a description of the model and data, or merge the two sections. Either way I would just describe the model set-up that you used in one section, instead of spread over multiple sections with other descriptions in between.

Line 188: Do the spin-up runs with the different forcing results in very different glacier
lengths? Why did you decide to start from the different simulations from a different initial
state?

Line 190: " For all runs, a 300-year spinup using annual climate data selected randomly from a 51-year window of 875-925 CE from the same model is performed prior to the run, to allow the glacier to develop from the preprocessed glacier geometry (based on RGI data and
therefore representative of the year of the observation used) to a more realistic geometry for the climate near the start of the model run."
I find the statement quite strong. What is it based on?

4-7. Results - Conclusion
- How sensitive are the results to the year that was picked for normalization (line 148-149)?
- Line 215: Are there glaciers in the Leclercq dataset that are expected to have been
retreating throughout the last millennium?
- Fig 3&4. Ref table 1. It is not clear from the table which glaciers are included in the plot. Are the glaciers that have failed during one of the the simulations included in figure 3 and 4?
- I'm wondering, how do your results for Europe differ from those in Goose et al. (2018)?
- Line 347-365: Are you aware that there is a function in OGGM which can merge the glacier
flowlines?
( https://docs.oggm.org/en/latest/generated/oggm.workflow.merge_glacier_tasks.html ) Have you tested the claim that you make in the last part of this section?

- Line 376-378: Consider rewording.
- I noticed that very little references to the literature are being made in the Discussion and
Limitations and Extensions section.

Minor details:
- The comment on line 25 (previously line 21), by reviewer 3, hasn't been addressed nor
included: most relevant study from Oerlemans to support this statement is Oerlemans (2005)
- I got lost in the sentence that starts in line 24 and end in line 28. Where does ",however
(Zemp et al., 2015; Cogley, 2009)," refer too?
- Line 33: I think the reference to the website in redundant.
- Line 36: Update the reference to the CRU version that is being used by OGGM v1.1.
- Line 37: I think it would be nice to name the version of the RGI that you used some where.
(I know this can be looked up with the reference. However I think it is common practice to
name the version some where.)
- Line 46-47: I think the it would be appropriate to use a reference for the timing of the LIA and MWP.
- I would like to repeat the suggestion of reviewer 3, to try to avoid multiple repetitions of
brackets. (e.g. line 32 & 118). In addition please remover double brackets (e.g. Line 69).
- e.g. line 208: I would like to suggest to use the names of the regions and not their numbers




Further minor points by editor:

-line 26: the position of the citation seem odd in this sentence, should it not be: '…are relatively sparse (Zemp….), however,…'
-line 32: maybe delete one of the OGGMs in the brackets here, repetition.
-line 125/126: something wrong with the formatting of the citation, should be: '…compiled by Leclercq et al (2014).
-lines 137/138: I would first mention 'glacier thickness' here and cite Johannesson et al (1989), yes of course slope, etc… influence thickness but the simple theory by Jóhannesson
et al. (1989), and further developed by Harrison et al. (2001) really puts it down to thickness. I would refer to these original papers here. Thickness is to some degree of course linked to size (length) of a glacier, hence, the delay of large glacier.

Jóhannesson, T., Raymond, C., andWaddington, E.: Time-scale for adjustment of glaciers to changes in mass balance, Journal of Glaciology, 35, 355-369, https://doi.org/10.1017/S002214300000928X, 1989.

Harrison, W. D., Elsberg, D. H., Echelmeyer, K. A., and Krimmel, R. M.: On the characterization of glacier response by a single time-scale, Journal of Glaciology, 47, 659-664, https://doi.org/10.3189/172756501781831837, 2001.

-Fig 7: although the scale (limits) of the x and y axis are identical, a 1:1 line in the plot would help

Dear D. Parkes and co-authors,
Thank you for the detailed revisions of the last round of reviews which addressed all critical comments by the referees and myself well and in particular improved the model and method description satisfactorily. Thus, the manuscript can now be accepted for publication and will provide an interesting and valuable contribution to the general topic of global/regional/local glacier modelling and long-term glacier-climate interaction and I thank again the authors for their good handling and cooperation in the reviewing process.
There are only very few minor editing issues or typos that I spotted when rereading the manuscript and that should be corrected before publication.
should be corrected:
-line 431: there is a 'be' missing here, '…it will BE necessary…'
-caption Fig. 4, line 3: replace the 'full stop' with a 'comma': '…that year, resulting in some…'
-caption Fig 7, line 1: there are brackets missing around the year '2014' in the citation of Leclercq.
-please also acknowledge the refrees in your acknwoledgments

Andreas Vieli, editor, 1 Aug 2020