CHELSA-TraCE21k – high-resolution (1&thinsp;km) downscaled transient temperature and precipitation data since the Last Glacial Maximum

Karger, Dirk Nikolaus; Nobis, Michael P.; Normand, Signe; Graham, Catherine H.; Zimmermann, Niklaus E.

doi:https://doi.org/10.5194/cp-19-439-2023

Articles | Volume 19, issue 2

https://doi.org/10.5194/cp-19-439-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/cp-19-439-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 19, issue 2

Research article

|

20 Feb 2023

Research article |

| 20 Feb 2023

CHELSA-TraCE21k – high-resolution (1 km) downscaled transient temperature and precipitation data since the Last Glacial Maximum

Dirk Nikolaus Karger, Michael P. Nobis, Signe Normand, Catherine H. Graham, and Niklaus E. Zimmermann

Download

Final revised paper (published on 20 Feb 2023)
Supplement to the final revised paper
Preprint (discussion started on 03 May 2021)

Interactive discussion

Status: closed

CC1:
'Comment on cp-2021-30: Lorenz et al & adding a caveat emptor for 1km downscaling', John Williams, 11 May 2021

A minor suggestion: adding a citation for Lorenz et al. 2016, which was (I think) the first published downscaling of SynTrace data, though this was for North America only, not global: Lorenz, D.J., Nieto-Lugilde, D., Blois, J.L., Fitzpatrick, M.C., Williams, J.W., 2016. Downscaled and debiased climate simulations for North America from 21,000 years ago to 2100AD. Scientific Data 3, 160048.
A more general comment/concern is that downscaling to 1km at the LGM (and, really, any pre-instrumental time period) has always seemed to me to be an instance of false precision. Given a) the good but imperfect data-model correlations shown for the historical data, b) the known uncertainties and biases in paleoclimatic simulations, and c) the intent of using these downscaled climates for paleobiogeographic simulations, there is a real risk that these downscaled data products will be used uncritically by biogeographic modelers. The 1km downscaling is already done, of course, so if the ms. is published here I strongly suggest adding a 'caveat emptor' paragraph or two in the discussion (with corresponding sentences in Abstract and Conclusions).
That said, this is a technically impressive work and is a useful addition to the suite of downscaled SynTrace simulations now available.

Citation: https://doi.org/10.5194/cp-2021-30-CC1
- AC1: 'Reply on CC1', Dirk N. Karger, 14 May 2021
  
  Thanks for the comment. Indeed Lorenz et al is a good example we have not mentioned in the manuscript yet, and will certainly do so during the revision.
  With respect to the false precision: It is indeed a problem that paleo climatic simulations are uncertain to a large degree. We do not know for example if at a specific location, the annual mean temperature was really -20°C or -24°C. What we can however be rather certain about is that temperatures at higher elevation are cooler than those of a lower elevation (although lapse rates can also be flexible in this respect). We also know that certain orographic effects for precipitation are similiar (rain shadows of moutains for example). These assumptions are based on simple physical relationships.
  The CHELSA_TraCE21k model is really rather "general" in this case, then "precise". If we would have a 100% precision during the historical period, we most likely overfitted the model and it would loose generality over the long term and wont be able to produce reasonable results anymore. Thats why we are less concerned with the not perfect fit during the historical period, as this is really something rather wanted for such a model. Precision, realism, and generality are really a tradeoff in this case.
  We argue that its not really a false precision, as we only estimate simple relationships here between e.g. elevation and temperature. Nevertheless, we fully agree that these points might be not neccessarily present in everybodies mind when appling the data to impact studies, and we therefore agree it will be very useful to add a paragraph concerning the caveats of the data. We will certainly add this paragraph. Thanks for brining it up.
  
  Citation: https://doi.org/10.5194/cp-2021-30-AC1
RC1:
'Comment on cp-2021-30', Anonymous Referee #1, 04 Jul 2021

Review of “CHELSA-TraCE21k v1.0. Downscaled transient temperature and precipitation data since the last glacial maximum” by Karger and others

Summary

-------

The manuscript describes an approach to produce high-resolution climate data by downscaling the output of a long simulation of a general circulation model using additional data sources. Most of the text is rather technical, a description of the downscaling process and validation of the resulting temperature and precipitation data. The paper ends with a potential use case of the produced dataset, the application to a problem in paleo-biology.

General comments

----------------

Overall, I appreciate the manuscript as an interesting contribution to facilitate paleo modelling work, which relies on high resolution past climate data. In my view, however, there is a number of severe shortcomings in the paper that need to be addressed before it can be published.

My first question was if the manuscript is well placed in the context of CP. The manuscript largely reads like a model description paper and may be better placed in a journal specialised for such content (e.g. GMD). The comments I will raise further on will not depend on this decision. But I will suggest revisions that bring out the modelling aspect even more, asking for further details that are currently lacking.

The manuscript is giving a good overview of 'what' is done, but has severe shortcomings in explaining 'why' and 'how'. I believe there is need to improve on describing the motivation for most of the decisions and clarifying the details of the processes (see specific comments below). The use of symbols is confusing and inconsistent and should be improved. The aim should be to put interested readers in the position to understand and reproduce the work that has been done. Additional figures/illustrations may help to achieve that.

There may be a conceptual problem with the reconstruction of past surface elevations for glaciated regions. It is not clear to me why past sea surface elevation (i.e. global sea-level) is needed to correct the elevation (l109). The surface elevation of a glaciated region is the result of changing ice thickness and changing bedrock elevation. Neither of these changes is related (linearly) to sea-level changes. A better explanation is needed to justify the presented approach.

The same applies to the coupling with temperature, which appears to modify the elevation estimate. Again, this is not well motivated and described.

Specific comments

-----------------

Abstract

l12 Suggest to start a new sentence after (ICE6G) and lay out in simple terms what the temperature coupling entails. This has not become clear to me throughout the manuscript. The amount of ice at a certain place is not determined solely by local temperature, if that is what is happening here.

l13 At this stage the reader will not know what CCSM3-TrCE21k and CHELSA stand for. This requires a bit more explanation already in the abstract.

l16 Here the species distribution is described as a validation of the dataset, while later it is an application of the forcing data. Which one is it?

l31 It is not clear to me how satellite data can be used to 'bride the gap between the coarse GCM output and the high resolution needed'. The two sources are distinct and have their own biases. Can you explain?

l43 'ice shields' --> 'ice sheets'. Correct this also in the rest of the manuscript.

l43 'along the poles' --> 'in polar regions'

l58 How does a simulation that starts at 21k-BP with 100 yr time steps come out at 1990?

l59 Isn't paleo-orography an *input* to the downscaling procedure? Here it looks like an output.

l62 It would be useful to distinguish between the model (CCSM3), the specific simulation (TraCE-21k) and the output of that model for a specific simulation (CCSM3 TraCE-21k).

l69 Should add here that the model is run with a fixed topography (which one? PD, LGM) and fixed land-sea mask (if that is so). Is the fixed land-sea mask not a problem for the downscaling? How does your process deal with regions that change from land to ocean with deglacial sea-level rise?

l73 What does the acronym CHELSA stand for?

l76 Explain what GPCC stands for.

l78 ICE-7G appears to be available since 2018. Can you explain why you are using ICE6G? What is the difference between ICE6G and ICE6G_C and why did you chose the 'C' variant?

l78 Same point as above, is ICE6G_C the model the simulation or the data?

l80 To my knowledge ICE6G is not a dynamic ice sheet model and does not explicitly model changes in ice thickness.

l82 Topography update is every 500 years, but according to l57 climate updates every 100 years. How do you deal with this issue?

l85 It could be useful to explain here or elsewhere that the given extent mask is not necessarily in agreement with ICE6G at LGM. When is LGM defined in Ehlers et al., (2011). Is there a possible temporal mismatch with ICE6G?

l88 What year do you assign to this dataset, what does 'current' mean specifically?

l94 What does 'derived' mean. What is different and what is the same compared to the original CHELSA V1.2. Again, is CHELSA V1.2 the algorithm (as stated here) or the dataset originating from it?

l96 Why is GMTED2010 not described as input data in section 2? It should.

l98 Why is the Miller data not described as an input dataset in section 2? It should.

l100 Could state here that the details 1-4 are described in the following sub-sections. It would be useful to describe the overall process in a flow diagram or other schematic to make it easier to understand the different steps. Add motivation at every step why things are done the way they are and how in detail.

l102-104 It seems that this part still belongs to the general intro section 3 assuming that 3.1 is only about orography.

l105 Explain what the purpose of combing Ehlers and ICE6G is. Motivate this by laying out your assumptions (do you trust Ehlers more than ICE6G in terms of accuracy?). What does Ehlers give you that ICE6G doesn't and vice versa? It may be useful to illustrate the whole process with a figure for one or several example location. Maybe a cross section through the margin of an ice sheet?

l106 Explain the choice and significance of taking 100 samples? How many samples are left (on average, max) after removing the outliers. Is ICE6G distributed at 1 degree resolution? If so mention it in section 2.2. If not, why work at that resolution?

l106 Not clear what "extracted the height of the glacier plus the surface elevation" means. Maybe 'height of the glacier' is ice thickness? Or is it height as in surface heigh? Is e_{t}^{ice} the surface elevation or the surface elevation + glacier height?

l107 What does the subscript 't' stand for?

l107 Explain why these points are omitted. What is the reason for further extracting the point locations on the boundaries?

l108 What does 'DEM' stand for and what is the data source for it?

l109 Is 'past sea surface elevation' from Miller? Explain why this correction to sea-level is needed. As mentioned in the general comments, I don't understand why this is done and suspect a conceptional problem. Please explain this.

l116 What does subscript 'c' stand for.

I have tried to give exhaustive comments on page 4 to show the level of detail that is in my mind required to make this a useful description. Similar comments could be made in the sections that follow.

l120 Explain up front why the B-spline interpolation is needed and what the main ideas of the iterative approach are.

l131 What is this 'change factor'? Explain what it serves for in the approach.

l131 What time period is tas_{cur}^{mod} averaged over? Why do you resample to 0.5 degree resolution?

l138 What is the significance of levels 26 and 20. What pressure/altitude do they represent?

l146 The main ideas of that temperature coupling process have to be explained. What assumptions go into that approach? How is temperature assumed to modify orography?

l155 Where does 'glacial melt' come from in this approach?

l166 What process is assumed to modify orography?

l195 in l184 the grid is described to have 4km resolution. Why the change to 3km?

l225 text here appears to be repeated in l227.

Figure 1

Suggest to (additionally) show anomalies relative to the present day. For now, it is difficult to make out clear differences in these plots.

Suggest to show the present day reference temperature field for comparison.

The '-' in 22k-BP should re removed. It reads like a minus sign.

Is this a perceptually uniform colour map? If not, consider using one (e.g. https://www.nature.com/articles/s41467-020-19160-7)

Figure 2

Results should be shown relative to a long-term average rather than one year (1990). If this is the case, what period is the data averaged over?

l288 Motivate why it is needed to project the data to another map? Details like projection parameters can be presented elsewhere (table, appendix).

l290-294 What is the underlying physical assumption for the 1/0 assignment? Clarify.

l314 Is the strong correlation maybe related to the fact that the data was bias corrected to a similar product? How do you explain such impressive match?

l334 'idiosyncratic' Strange choice of word. Reformulate?

Figure 5

The strong mismatch at LGM could suggest that the lapse rate correction plays out in an unexpected way. It should be checked if that mismatch arises from climate model bias, lapse rate corrections or the bias corrections that are applied in the process.

l344 It is not clear to me how glacier extent is meant to validate the downscaling process. It may serve to validate ICE6G and may reveal a mismatch between ICE6G and Dyke, but that is not really at stake here. Could you explain how that comparison can constrain your approach? How does unmodified ICE6G compare to Dyke. Is that improved with your modifications?

l360 This section comes with unexpected new concepts and models (GLM, KISSMig) that were not introduced before. After going through the technicalities of the sections before, this is a steep change of register. In the abstract this part is introduced as another aspect of the model validation, while here it is written as a use case for the produced dataset. In either case, I suggest this part has to be better linked with the rest of the paper or, maybe better, extended and conceived as a separate paper.

l381 Why is yet another projection needed in this case? Motivate.

l418 What trends are to be preserved? Clarify. Reference to (Hempel et al., 2013) is probably better placed in the description in section 3.2.

Citation: https://doi.org/10.5194/cp-2021-30-RC1
- AC2: 'Reply on RC1', Dirk N. Karger, 17 Nov 2021
  
  REVIEWER 1:
  Review of “CHELSA-TraCE21k v1.0. Downscaled transient temperature and precipitation data since the last glacial maximum” by Karger and others
  Summary
  -------
  The manuscript describes an approach to produce high-resolution climate data by downscaling the output of a long simulation of a general circulation model using additional data sources. Most of the text is rather technical, a description of the downscaling process and validation of the resulting temperature and precipitation data. The paper ends with a potential use case of the produced dataset, the application to a problem in paleo-biology.
  General comments
  ----------------
  Overall, I appreciate the manuscript as an interesting contribution to facilitate paleo modelling work, which relies on high resolution past climate data. In my view, however, there is a number of severe shortcomings in the paper that need to be addressed before it can be published.
  My first question was if the manuscript is well placed in the context of CP. The manuscript largely reads like a model description paper and may be better placed in a journal specialised for such content (e.g. GMD). The comments I will raise further on will not depend on this decision. But I will suggest revisions that bring out the modelling aspect even more, asking for further details that are currently lacking.
  Response:     The manuscript has actually been transferred from GMD from the editors as the fit was considered higher with CP.
  The manuscript is giving a good overview of 'what' is done, but has severe shortcomings in explaining 'why' and 'how'. I believe there is need to improve on describing the motivation for most of the decisions and clarifying the details of the processes (see specific comments below). The use of symbols is confusing and inconsistent and should be improved. The aim should be to put interested readers in the position to understand and reproduce the work that has been done. Additional figures/illustrations may help to achieve that.
  Response: We will include a figure highlighting the downscaling procedure in a revised manuscript.
  There may be a conceptual problem with the reconstruction of past surface elevations for glaciated regions. It is not clear to me why past sea surface elevation (i.e. global sea-level) is needed to correct the elevation (l109). The surface elevation of a glaciated region is the result of changing ice thickness and changing bedrock elevation. Neither of these changes is related (linearly) to sea-level changes. A better explanation is needed to justify the presented approach.
  Response:    The model we present includes the ice thickness from ICE6G as indicated in line 108. The sea level is included as the high resolution DEM is based on a current bathymetric DEM. Not including the sea level changes would result in an orography that still has today's sea level, which is incorrect for past time steps. The model does not include a changing bedrock on the high resolution (1km) elevation however. We will add a comment on this in the revised manuscript.
  The same applies to the coupling with temperature, which appears to modify the elevation estimate. Again, this is not well motivated and described.
  Specific comments
  -----------------
  Abstract
  l12 Suggest to start a new sentence after (ICE6G) and lay out in simple terms what the temperature coupling entails. This has not become clear to me throughout the manuscript. The amount of ice at a certain place is not determined solely by local temperature, if that is what is happening here.
  Response:    The amount of ice is of course a balance of precipitation fluxes and temperature changes, which can be physically modelled with a numerical model. This is however not feasible at 1km resolution due to computational limitations and not due to the approach presented here. We use mean annual temperature as a proxy for an interpolation approach. The entire approach is still an interpolation and needs a local correction which we describe in lines 149ff. We see from the reviewer comments however that there have been several misunderstandings on the interpolation procedure and our focus on a mathematical description may not have been easy to understand. We therefore will also include a figure highlighting the interpolation algorithm better.
  l13 At this stage the reader will not know what CCSM3-TrCE21k and CHELSA stand for. This requires a bit more explanation already in the abstract.
  Response:    Agreed. We will include a short description.
  l16 Here the species distribution is described as a validation of the dataset, while later it is an application of the forcing data. Which one is it?
  Response:    Both. Paleoclimate models often get validated by specific proxy variables, as we are lacking direct measurements. Appling the dataset to model the distribution of species is the ‘application’. Checking if the results make sense is not a ‘validation’ as such, but a ‘plausibility test`.
  l31 It is not clear to me how satellite data can be used to 'bride the gap between the coarse GCM output and the high resolution needed'. The two sources are distinct and have their own biases. Can you explain?
  Response:    Maybe this is misleading at this point. Can be deleted.
  l43 'ice shields' --> 'ice sheets'. Correct this also in the rest of the manuscript.
  Response:    Will be changed
  l43 'along the poles' --> 'in polar regions'
  Response:    Will be changed
  l58 How does a simulation that starts at 21k-BP with 100 yr time steps come out at 1990?
  Response:    This is based on the timesteps of CCSM3 TraCE-21k. The last timestep for the 20th century goes until 1990. We will add the information in a revised manuscript.
  l59 Isn't paleo-orography an *input* to the downscaling procedure? Here it looks like an output.
  Response:    The output of one timestep (t) is the input for the next time step (t+1).
  l62 It would be useful to distinguish between the model (CCSM3), the specific simulation (TraCE-21k) and the output of that model for a specific simulation (CCSM3 TraCE-21k).
  Response:    Will be changed
  l69 Should add here that the model is run with a fixed topography (which one? PD, LGM) and fixed land-sea mask (if that is so). Is the fixed land-sea mask not a problem for the downscaling? How does your process deal with regions that change from land to ocean with deglacial sea-level rise?
  Response:    No, there is no fixed topography. The modelling of the paleo-ography is explained in paragraph 3.1 and 3.2. This is also why we include the sea level in the estimation of paleo-orography.
  l73 What does the acronym CHELSA stand for?
  Response:    Climatologies at High Resolution for the Earth’s Land Surface Areas. Will be included.
  l76 Explain what GPCC stands for.
  Response:     Will be included.
  l78 ICE-7G appears to be available since 2018. Can you explain why you are using ICE6G? What is the difference between ICE6G and ICE6G_C and why did you chose the 'C' variant?
  Response:     ICE6G_C was the latest version when we made the calculations. ICE-7G was not available at that point. An update is not feasible at this point due to the high computational demand.
  l78 Same point as above, is ICE6G_C the model the simulation or the data?
  Response:     The output of the model (the data).
  l80 To my knowledge ICE6G is not a dynamic ice sheet model and does not explicitly model changes in ice thickness.
  Response:     That is not what is stated on their webpage: “Each file contains information for points on a 1x1 degree global grid. The model's ice thickness field is given by the variable stgit.” https://www.atmosp.physics.utoronto.ca/~peltier/data.php
  l82 Topography update is every 500 years, but according to l57 climate updates every 100 years. How do you deal with this issue?
  Response:    This is indeed not described. The nearest time step has been used. We will add this at the respective point.
  l85 It could be useful to explain here or elsewhere that the given extent mask is not necessarily in agreement with ICE6G at LGM. When is LGM defined in Ehlers et al., (2011). Is there a possible temporal mismatch with ICE6G?
  Response:    This is correct. We will mention this.
  l88 What year do you assign to this dataset, what does 'current' mean specifically?
  Response:    This cannot be clearly stated as GLIMS does not give this information.
  l94 What does 'derived' mean. What is different and what is the same compared to the original CHELSA V1.2. Again, is CHELSA V1.2 the algorithm (as stated here) or the dataset originating from it?
  Response:    As stated.
  l96 Why is GMTED2010 not described as input data in section 2? It should.
  Response:    Will be included.
  l98 Why is the Miller data not described as an input dataset in section 2? It should.
  Response:    Will be included.
  l100 Could state here that the details 1-4 are described in the following sub-sections. It would be useful to describe the overall process in a flow diagram or other schematic to make it easier to understand the different steps. Add motivation at every step why things are done the way they are and how in detail.
  Response:    We will include such a diagram in the revision and make sure to better describe each step.
  l102-104 It seems that this part still belongs to the general intro section 3 assuming that 3.1 is only about orography.
  Response:    Orography is both bedrock elevation + glacier thickness.
  l105 Explain what the purpose of combing Ehlers and ICE6G is. Motivate this by laying out your assumptions (do you trust Ehlers more than ICE6G in terms of accuracy?). What does Ehlers give you that ICE6G doesn't and vice versa? It may be useful to illustrate the whole process with a figure for one or several example location. Maybe a cross section through the margin of an ice sheet?
  Response:    We will include such a figure. Ehlers gives a high resolution that can be used to delineate glacial boundaries at 1km. ICE6G gives a temporal signal of the elevation only of the major ice sheets.
  l106 Explain the choice and significance of taking 100 samples? How many samples are left (on average, max) after removing the outliers. Is ICE6G distributed at 1 degree resolution? If so mention it in section 2.2. If not, why work at that resolution?
  Response:    It’s the native resolution. 100 samples is arbitrary in this case. It seemed a good compromise between size of the resulting sample and accuracy.
  l106 Not clear what "extracted the height of the glacier plus the surface elevation" means. Maybe 'height of the glacier' is ice thickness? Or is it height as in surface heigh? Is e_{t}^{ice} the surface elevation or the surface elevation + glacier height?
  Response:    Will be clarified.
  l107 What does the subscript 't' stand for?
  Response:    Timestep. Will be clarified.
  l107 Explain why these points are omitted. What is the reason for further extracting the point locations on the boundaries?
  Response:    Otherwise there would be a surface elevation of a glacier where there is no glacier.
  l108 What does 'DEM' stand for and what is the data source for it?
  Response:    Digital Elevation Model. Will be described. Data source is GMTED2010.
  l109 Is 'past sea surface elevation' from Miller? Explain why this correction to sea-level is needed. As mentioned in the general comments, I don't understand why this is done and suspect a conceptional problem. Please explain this.
  Response:    Please see our response above. We don’t understand what the conceptual problem could be, please elaborate what you expect. Our intent was to a) provide evidence for land where there was land in the past, and b) to use accurate and realistic elevation data for modelling climate.
  l116 What does subscript 'c' stand for.
  Response:    Will be explained.
  I have tried to give exhaustive comments on page 4 to show the level of detail that is in my mind required to make this a useful description. Similar comments could be made in the sections that follow.
  l120 Explain up front why the B-spline interpolation is needed and what the main ideas of the iterative approach are.
  Response: A B-Spline interpolates data to higher resolutions, the iterative process (hence: Multilevel B-Spline) makes sure that the error of the interpolation decreases.
  l131 What is this 'change factor'? Explain what it serves for in the approach.
  Response:    Will be explained.
  l131 What time period is tas_{cur}^{mod} averaged over? Why do you resample to 0.5 degree resolution?
  Response:    Will be explained.
  l138 What is the significance of levels 26 and 20. What pressure/altitude do they represent?
  Response:    Will be explained.
  l146 The main ideas of that temperature coupling process have to be explained. What assumptions go into that approach? How is temperature assumed to modify orography?
  Response:    The assumption is that increasing temperatures are related to melting of glaciers (and vice-versa).
  l155 Where does 'glacial melt' come from in this approach?
  Response:    Probably better explained as reduction in glacier thickness.
  l166 What process is assumed to modify orography?
  Response:    In this case only the changes in size/thickness of the glaciers.
  l195 in l184 the grid is described to have 4km resolution. Why the change to 3km?
  Response:    Simply a typo, will be corrected.
  l225 text here appears to be repeated in l227.
  Response:    Will be corrected.
  Figure 1
  Suggest to (additionally) show anomalies relative to the present day. For now, it is difficult to make out clear differences in these plots.
  Response:    This will be included in a revision
  Suggest to show the present day reference temperature field for comparison.
  Response:    If we show anomalies, that’s not needed anymore
  The '-' in 22k-BP should re removed. It reads like a minus sign.
  Response:     Will be removed
  Is this a perceptually uniform colour map? If not, consider using one (e.g. https://www.nature.com/articles/s41467-020-19160-7)
  Figure 2
  
  Results should be shown relative to a long-term average rather than one year (1990). If this is the case, what period is the data averaged over?
  Response:    1950-1990
  l288 Motivate why it is needed to project the data to another map? Details like projection parameters can be presented elsewhere (table, appendix).
  Reponse:    Will be excluded
  l290-294 What is the underlying physical assumption for the 1/0 assignment? Clarify.
  Response:    Will be clarified. 0 means there is no glacier, 1 means there is glacier. This binary information is then tested. Why does this need a physical assumption? Please elaborate.
  l314 Is the strong correlation maybe related to the fact that the data was bias corrected to a similar product? How do you explain such impressive match?
  Response:    The bias correction certainly has an influence as well, but so does the downscaling. The CHELSA model decreases the bias and increases the correlation usually (for more details see. e.g. Karger et al. 2017, 2020, 2021 Scientific Data);
  l334 'idiosyncratic' Strange choice of word. Reformulate?
  Response:    Will be changed
  Figure 5
  
  The strong mismatch at LGM could suggest that the lapse rate correction plays out in an unexpected way. It should be checked if that mismatch arises from climate model bias, lapse rate corrections or the bias corrections that are applied in the process.
  Response:    It could be all of the above plus inaccuracies in the proxies. It cannot be disentangled at this point, as we have no data to check the degree of bias in LGM climate simulations (the coarse resolution model output, which is input to the downscaling algorithm) or the proxy data.
  l344 It is not clear to me how glacier extent is meant to validate the downscaling process. It may serve to validate ICE6G and may reveal a mismatch between ICE6G and Dyke, but that is not really at stake here. Could you explain how that comparison can constrain your approach? How does unmodified ICE6G compare to Dyke. Is that improved with your modifications?
  Response:    We compare the 1km predictions estimated using our algorithm, not ICE6G. We can include ICE6 G as well, but only for limited time steps.
  l360 This section comes with unexpected new concepts and models (GLM, KISSMig) that were not introduced before. After going through the technicalities of the sections before, this is a steep change of register. In the abstract this part is introduced as another aspect of the model validation, while here it is written as a use case for the produced dataset. In either case, I suggest this part has to be better linked with the rest of the paper or, maybe better, extended and conceived as a separate paper.
  Response:    Its part of the validation with the idea that it simultaneously presents a case study as a plausibility test. We will add more basic information about the reasons for applying this case study and about the underlying models used such as KISSMig.
  l381 Why is yet another projection needed in this case? Motivate.
  Response:    Since KISSMig is a grid based dynamic species migration model, it uses an equal area projection to avoid spatial bias in the simulations..
  l418 What trends are to be preserved? Clarify. Reference to (Hempel et al., 2013) is probably better placed in the description in section 3.2.
  Response:    Will be changed.
  
  Citation: https://doi.org/10.5194/cp-2021-30-AC2
RC2:
'Comment on cp-2021-30', Anonymous Referee #2, 20 Oct 2021

Review

“CHELSA-TraCE21k v1.0. Downscaled transient temperature and precipitation data since the last glacial maximum”, Karger et al.

General comments:

This manuscript presents a method of downscaling climate model data, using an algorithm, with objective to create high-resolution global monthly climatology for precipitation and temperature for the period of last 21000 years. The manuscript consists of description of the method, evaluation of obtained results and an example of potential use of created high-resolution climate data set in paleo-ecology to detect refugia of plant species at the end.

The final product of this research, in format of high-resolution climate data set, presents an important result, that should hopefully find its application in several scientific fields. I highly appreciate efforts to get the final data set, which, I believe, was a difficult, time-consuming and very technical task. That technical part is well presented and documented within the manuscript. However, I find that the rest of the manuscript has serious shortcomings, especially the evaluation of the obtained data set, what is expected to be the central part of this research. There are also serious issues in manuscript structure. Some figures are presented without any deeper analysis, while on the other hand there are chapters that describe validation of the obtained data set, but without figures, that actually follow in subsequent chapters, which affects significantly the readability of the manuscript. At the and, there is a well described chapter about potential application of the high-resolution data set, however, without clear connection with the rest of the manuscript. It could represent a highlight of this research and be a crucial proof for efficient application of the data set, but it is not even mentioned in the main objective (although it exists in abstract and is mentioned in introduction). There are also inconsistencies in use of terms and symbols throughout the text that, for example, lead to confusion in understanding of some parts of the manuscript, of some figures and even in understanding of correct name of the algorithm in the abstract.

This manuscript has some good material and important results, but it requires significant improvements and better structure in order to be considered and at the end accepted for publication in this journal. Therefore, I would suggest a major revision, to give the authors a chance to improve it, but with caution to stay within the scope of this journal. My further specific comments are listed as follows:

Specific comments:

Line 9, Line 14: What is the name of the algorithm? Is it "CHELSA-TraCE21k downscaling algorithm" or "CHELSA V1.2 algorithm"? Please, be consistent in using specific terms throughout the text. In Line 17 it says "CHELSA TraCE21k output" (without hyphen), which leads to confusion since the very beginning. In addition, in the title of the manuscript it says: “CHELSA-TraCE21k v1.0", and that part "v1.0" does not appear at all in any part of the manuscript.

Lines 26-29: There are several applications mentioned, where temporal and spatial variability of temperature and precipitation matter. I would like to see at least one more of these applications described in detail, where your high-resolution data set can be used. I believe it could demonstrate the added value of created high-resolution data set. However, that would probably lead to writing of a completely new manuscript, possibly out of the scope of this journal.

Line 58: How do you end up with the year 1990, when you start from 21K BP and use 100-years time steps? And, please, do not use hyphen in "21K BP", it is not correct, unless it is an adjective.

Lines 60-72: What is the main reason to use exactly this model? Please, justify.

Lines 67-72: If you say in Lines 66-67 that CCSM3 is global climate with coupled ocean, atmosphere, sea-ice and land surface components, then try to maintain the same order of the Earth system components when you describe characteristics of each one, in order to maintain consistency.

Line 74: What is CHELSA? Is it a data set or an algorithm? It is very confusing. What does this acronym stand for?

Lines 245-248: Very confusing, at the and, I don't understand what is presented in Figure 1. Especially due to use of hyphen in the figures, that gives impression it is a "minus" (22k-BP). Please, avoid that in all other figures, too. Also, there is no any discussion about that figure, only the statement in the legend that it shows "exceptional climate dynamics". Please, avoid use of such strong words, especially if they are not supported by any explanation.

Lines 258-264: It is not very clear what is shown in this figure. Did you calculate difference of all mentioned 100-year BP periods from 1990 year only? Or from some annual mean of 1960-1990 period, or 1900-1990? Also, there is no any discussion about this figure and the same comments stand as for the previous one. In addition, I see some strange separation in anomaly sign in southern hemisphere, approximately around 10 S and 40 S. Is there maybe some problem with the downscaling algorithm for that region? Or there is some physical explanation for this pattern?

Lines 265-359: I would suggest to reorder and rewrite chapters 4.1, 4.2, 4.3, 5.1, 5.2 and 5.3. In a current form, it is difficult to follow. It would look much better and improve readability if you could merge 4.1 with 5.1, 4.2 with 5.2 and 4.3 with 5.3.

Line 266: What is the resolution of GHCN? Is it comparable with your data set?

Lines 360-409: This whole chapter does not seem to have a good connection with the rest of the manuscript, although it gives an important application of the obtained high-resolution data set. A suggestion could be to remove it from this manuscript and to try to improve the rest with more profound and more comprehensive evaluation of the data set. Current chapter could be used with several other examples of potential applications of the high-resolution data set with objective to create another manuscript.

Technical corrections:

Line 25: Spatial resolution should not be expressed in square kilometers. I would rather say "at spatial resolutions lower than 1 km", for example

Lines 73 and 78: Repeated chapter number

Line 76: Acronyms ERA and GPCC are mentioned for the first time in manuscript, therefore, it is expected to write their meaning.

Line 177: It seems there is an extra space between the words "level resulting"

Lines 184, 186, 195, 196: Please, use the hyphen when you have number, followed by unit when it is an adjective (4-km grid resolution, 3-km grid cell) and spacing when you have number and unit when it is not an adjective (1 km, not 1km). Try to maintain consistency throughout manuscript in all other similar cases.

Line 210: It is not understandable, is it a continuation of the sentence, that should be separated by comma and followed by small letter or something else?

Line 215: One "being" extra, please, remove it.

Line 216 and 240: 30-arc sec. resolution/grid

Line 229: windward-leeward equations

Line 234: Are these 2 dots instead of comma? Please, correct it.

Line 245: 1-km paleoclimatic dataset

Lines 255 and 263: Why is it written "8.2 kiloyear", when in all other cases you use only "k"? Please, maintain consistency in use of symbols throughout the manuscript.

Lines 282, 344, 373, 380, 395, etc.: 18k PB and 1k PB. Please, maintain consistency throughout the manuscript by correcting other similar cases

Lines 302-303: Another different way of writing 18k BP and 1k BP; 1-km resolution

Line 305, 308 and 316: It is RMSE, not RSME.

Line 320: Taylor diagrams; typing error with an extra "f"

Lines 328-343: "CHELSA_TraCE21k model", "CHELSA_TraCE", "TraCE21k", "TraCE", "CHELSA-TraCE21k time series data", "CHELSA V2.1" - so many similar and confusing names in this short paragraph, that it is impossible to follow. Please, rewrite it and try to be consistent in using specific terms.

Lines 354-359: ice sheet, not ice shield

Line 390: Typing error "the"

Line 420: "comparably well when compared". Please, try to find better words

Citation: https://doi.org/10.5194/cp-2021-30-RC2
- AC3: 'Reply on RC2', Dirk N. Karger, 17 Nov 2021
  
  CHELSA-TraCE21k v1.0. Downscaled transient temperature and precipitation data since the last glacial maximum”, Karger et al.
  
  General comments:
  
  This manuscript presents a method of downscaling climate model data, using an algorithm, with objective to create high-resolution global monthly climatology for precipitation and temperature for the period of last 21000 years. The manuscript consists of description of the method, evaluation of obtained results and an example of potential use of created high-resolution climate data set in paleo-ecology to detect refugia of plant species at the end.
  The final product of this research, in format of high-resolution climate data set, presents an important result, that should hopefully find its application in several scientific fields. I highly appreciate efforts to get the final data set, which, I believe, was a difficult, time-consuming and very technical task. That technical part is well presented and documented within the manuscript. However, I find that the rest of the manuscript has serious shortcomings, especially the evaluation of the obtained data set, what is expected to be the central part of this research. There are also serious issues in manuscript structure. Some figures are presented without any deeper analysis, while on the other hand there are chapters that describe validation of the obtained data set, but without figures, that actually follow in subsequent chapters, which affects significantly the readability of the manuscript. At the and, there is a well described chapter about potential application of the high-resolution data set, however, without clear connection with the rest of the manuscript. It could represent a highlight of this research and be a crucial proof for efficient application of the data set, but it is not even mentioned in the main objective (although it exists in abstract and is mentioned in introduction). There are also inconsistencies in use of terms and symbols throughout the text that, for example, lead to confusion in understanding of some parts of the manuscript, of some figures and even in understanding of correct name of the algorithm in the abstract.
  This manuscript has some good material and important results, but it requires significant improvements and better structure in order to be considered and at the end accepted for publication in this journal. Therefore, I would suggest a major revision, to give the authors a chance to improve it, but with caution to stay within the scope of this journal. My further specific comments are listed as follows:
  Response:     Thank you for your judgment on the value of the generated climate datasets. The generation of this dataset was our main aim, and the evaluation of its usefulness will be improved.
  
  Specific comments:
  
  Line 9, Line 14: What is the name of the algorithm? Is it "CHELSA-TraCE21k downscaling algorithm" or "CHELSA V1.2 algorithm"? Please, be consistent in using specific terms throughout the text. In Line 17 it says "CHELSA TraCE21k output" (without hyphen), which leads to confusion since the very beginning. In addition, in the title of the manuscript it says: “CHELSA-TraCE21k v1.0", and that part "v1.0" does not appear at all in any part of the manuscript.
  Response:     Will be made more consistent by better explaining the terms and by cleaning the text accordingly.
  Lines 26-29: There are several applications mentioned, where temporal and spatial variability of temperature and precipitation matter. I would like to see at least one more of these applications described in detail, where your high-resolution data set can be used. I believe it could demonstrate the added value of created high-resolution data set. However, that would probably lead to writing of a completely new manuscript, possibly out of the scope of this journal.
  Response:    It seems odd that this is mentioned here, as we gave one application an entire chapter (Chapter 6). We are not sure how we can add more here. What we can add is a short paragraph on possible other applications. Actually performing such applications would render the manuscript too heavy and hard to read. It would deviate readers from the main aim of the manuscript.
  Line 58: How do you end up with the year 1990, when you start from 21K BP and use 100-years time steps? And, please, do not use hyphen in "21K BP", it is not correct, unless it is an adjective.
  Response:    This is based on the timesteps of CCSM3 TraCE-21k. The last timestep for the 20th century goes until 1990. We will clarify this in a revised manuscript.
  Lines 60-72: What is the main reason to use exactly this model? Please, justify.
  Response:    The model is transient and can therefore also be used for dynamic modelling, as for example highlighted in chapter 6.
  Lines 67-72: If you say in Lines 66-67 that CCSM3 is global climate with coupled ocean, atmosphere, sea-ice and land surface components, then try to maintain the same order of the Earth system components when you describe characteristics of each one, in order to maintain consistency.
  Response:    Will be changed.
  Line 74: What is CHELSA? Is it a data set or an algorithm? It is very confusing. What does this acronym stand for?
  Response:    Climatologies at high resolution for the earths land surface areas. It’s both the name of the output dataset and the algorithm. Will be explained in a revised version.
  Lines 245-248: Very confusing, at the and, I don't understand what is presented in Figure 1. Especially due to use of hyphen in the figures, that gives impression it is a "minus" (22k-BP). Please, avoid that in all other figures, too. Also, there is no any discussion about that figure, only the statement in the legend that it shows "exceptional climate dynamics". Please, avoid use of such strong words, especially if they are not supported by any explanation.
  Response:    Will be changed.
  Lines 258-264: It is not very clear what is shown in this figure. Did you calculate difference of all mentioned 100-year BP periods from 1990 year only? Or from some annual mean of 1960-1990 period, or 1900-1990? Also, there is no any discussion about this figure and the same comments stand as for the previous one. In addition, I see some strange separation in anomaly sign in southern hemisphere, approximately around 10 S and 40 S. Is there maybe some problem with the downscaling algorithm for that region? Or there is some physical explanation for this pattern?
  Response:    The anomalies are actually taken directly from the CCSM3-TraCE21k dataset and therefore cannot be caused by the downscaling approach.
  Lines 265-359: I would suggest to reorder and rewrite chapters 4.1, 4.2, 4.3, 5.1, 5.2 and 5.3. In a current form, it is difficult to follow. It would look much better and improve readability if you could merge 4.1 with 5.1, 4.2 with 5.2 and 4.3 with 5.3.
  Response:    Thank you for the suggestion. We will reorder the manuscript accordingly by merging the respective chapters.
  Line 266: What is the resolution of GHCN? Is it comparable with your data set?
  Response:    GHCN consists of point measurements.
  
  Lines 360-409: This whole chapter does not seem to have a good connection with the rest of the manuscript, although it gives an important application of the obtained high-resolution data set. A suggestion could be to remove it from this manuscript and to try to improve the rest with more profound and more comprehensive evaluation of the data set. Current chapter could be used with several other examples of potential applications of the high-resolution data set with objective to create another manuscript.
  Response:    In the beginning you stated that a potential highlight for an application would be needed. The chapter is included to show exactly such a use case. The application also serves as a plausibility check of the downscaled data. If the glacial refugia cannot be modelled correctly, the downscaled data might be wrong, which is not the case here.
  
  Technical corrections:
  
  Line 25: Spatial resolution should not be expressed in square kilometers. I would rather say "at spatial resolutions lower than 1 km", for example
  Response:     Will be changed.
  Lines 73 and 78: Repeated chapter number
  Response:     Will be changed.
  Line 76: Acronyms ERA and GPCC are mentioned for the first time in manuscript, therefore, it is expected to write their meaning.
  Response:     Will be changed.
  Line 177: It seems there is an extra space between the words "level resulting"
  Response:     Will be changed.
  Lines 184, 186, 195, 196: Please, use the hyphen when you have number, followed by unit when it is an adjective (4-km grid resolution, 3-km grid cell) and spacing when you have number and unit when it is not an adjective (1 km, not 1km). Try to maintain consistency throughout manuscript in all other similar cases.
  Response:     Will be changed.
  Line 210: It is not understandable, is it a continuation of the sentence, that should be separated by comma and followed by small letter or something else?
  Response:     Will be changed.
  Line 215: One "being" extra, please, remove it.
  Response:     Will be changed.
  Line 216 and 240: 30-arc sec. resolution/grid
  Response:     Will be changed.
  Line 229: windward-leeward equations
  Response:     Will be changed.
  Line 234: Are these 2 dots instead of comma? Please, correct it.
  Response:     Will be changed.
  Line 245: 1-km paleoclimatic dataset
  Response:     Will be changed.
  Lines 255 and 263: Why is it written "8.2 kiloyear", when in all other cases you use only "k"? Please, maintain consistency in use of symbols throughout the manuscript.
  Response:     Will be changed.
  Lines 282, 344, 373, 380, 395, etc.: 18k PB and 1k PB. Please, maintain consistency throughout the manuscript by correcting other similar cases
  Response:     Will be changed.
  Lines 302-303: Another different way of writing 18k BP and 1k BP; 1-km resolution
  Response:     Will be changed.
  Line 305, 308 and 316: It is RMSE, not RSME.
  Response:     Will be changed.
  Line 320: Taylor diagrams; typing error with an extra "f"
  Response:     Will be changed.
  Lines 328-343: "CHELSA_TraCE21k model", "CHELSA_TraCE", "TraCE21k", "TraCE", "CHELSA-TraCE21k time series data", "CHELSA V2.1" - so many similar and confusing names in this short paragraph, that it is impossible to follow. Please, rewrite it and try to be consistent in using specific terms.
    Response:     Will be changed.
  Lines 354-359: ice sheet, not ice shield
  Response:     Will be changed.
  Line 390: Typing error "the"
  Response:     Will be changed.
  Line 420: "comparably well when compared". Please, try to find better words
  Response:     Will be changed.
  
  Citation: https://doi.org/10.5194/cp-2021-30-AC3

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Reconsider after major revisions (29 Nov 2021) by Irina Rogozhina

AR by Dirk N. Karger on behalf of the Authors (18 Feb 2022) Author's response Author's tracked changes Manuscript

ED: Reconsider after major revisions (09 Apr 2022) by Irina Rogozhina

ED: Referee Nomination & Report Request started (13 Apr 2022) by Irina Rogozhina

RR by Anonymous Referee #1 (20 May 2022)

Suggestions for revision or reasons for rejection

Re-review of “CHELSA-TraCE21k v1.0. Downscaled transient temperature and precipitation data since the last glacial maximum” by Karger and others

General comments
----------------
I have reviewed an earlier version of this manuscript as REVIEWER 1. While I acknowledge important improvements in the current version of the manuscript, it requires major revisions to make it suitable for publishing. A few of my earlier comments have not been addressed, so I am reiterating them once more.

With the given documentation, it is still not possible to judge if the applied methodology is flawed. This concerns 1) the way only sea level (and not bedrock changes) are used to calculate elevation changes and 2) the interpolation procedure using temperature as a proxy for ice sheet retreat. I have suggested possibilities to validate the methods with available reconstructions, which I consider fundamental to support the choices presented in this paper.

Specific comments
-----------------

l12.
Repeating my earlier comment: ICE6G is not a dynamic ice sheet model. I have looked up the description for you: "the ICE-6G_C (VM5a) model that is under discussion in this paper is based upon the ‘GIA only’ methodology. In this methodology the ice thickness history as a function of position is simply adjusted iteratively in order to satisfy all of the available constraints". Please reformulate.

l26.
Here you state "climatic conditions at spatial resolutions < 1 km", but later (e.g. 48) it is "~1km". Should be made consistent.

l27.
"run at much coarser grains" --> "run at much coarser resolution"

l34.
"has be bridged" --> "has been bridged"

l44.
"on earth" --> "on Earth"

l48.
It is somewhat trivial to state that 0.043 SYPD is 25-fold lower than 1 SYPD. What would be an example of a typical 1 SYPD simulation?

l60.
Repeating comments for "time steps of 100 years from 21k-BP to 1990"
Here and elsewhere, this should be written as "21 kyr BP"
BP typically refers to years before 1950. If you go in steps of 100 from 21 kyr BP forward, you never end up at 1990. Either you used a different time step than 100 years, or a different starting point. That is why you have to modify this statement.

l60.
(TraCE-21k) should be defined the first time you use it (outside of the abstract).

l64.
It would be good to add some information about what topography and ice sheet boundary condition (topography, mask, albedo) was used to produce this simulation. Is the land-sea mask constant? Is the topography changing? If yes, is the topography change consistent with the ICE6G_C reconstruction?

l69.
Is "DGVM" the name of the specific model or the shorthand for all dynamic global vegetation models (as line 71-72 may suggest)? Clarify!

l70.
Is the "land [...] model" the same as the vegetation model? If not, introduce the land model here.

l76.
What is "mean monthly daily 2m mean"? Maybe "monthly mean, minimum and maximum temperature and precipitation fields"?

l78.
You call it "mechanistic climate downscaling". I am familiar with terms statistical and dynamical downscaling. Which category does yours belong to?

l84.
Remove "and dynamics". ICE6G is not a dynamic ice sheet model.

l85.
"from the Last Glacial Maximum"
What is the first available year in the time series?

l87.
LGM is already defined. Remove "(LGM)"

l88.
LGM is already defined. Modify description.

l89.
How "up-to-date" is this dataset from more than 10 years ago?
Suggest to reformulate.

l95.
What time does "the 'current' extent of the glaciers" refer to? I understand that this may not be clearly defined by the dataset providers (late 90s, early 2000 maybe), but you are assigning it eventually to a certain time in your modelling (maybe 1990?). This should be mentioned here or elsewhere as in "We are assigning the dataset to the year xxx in our modelling". The same applies to all the other data sources (GMTED, GEBCO), which may be best done by a summary statement in the end.

L105.
"we keep as land altimetric data that of the CHELSA V1.2 procedure"
I didn't find a description what topography is used in CHELSA. Should be added in 2.2.

Section 2 in general
Since this has become a long list of rather short subsections, it could be an idea to display the information about the different datasets in a table (name, description, time coverage, reference, ...) complemented with a summary paragraph.

l117.
"As the orography at different time steps between 21k BP and current times is not available"
I think that ICE6G_C would give you that information. Maybe add "... at the high resolution required for our downscaling method" to make sense of this statement.

Figure 1, l124.
Terminology. I am not an expert on this, but it seems that the distinction made here between topography (as relative to present day sea-level) and orography (relative to current global sea-level) is not supported by common definitions of the two terms. Also, I could not find a description of bathymetry in the ocean and surface elevation over land in one generalised term. I would suggest to using your own symbols and describing what they mean, rather than using established terms that mean something else.

I doubt that the combination of GMTED2010 and GEBCO really gives you bedrock topography (upper right in figure 1). I think GMTED2010 provides surface elevation over glaciated areas (Greenland/Antarctica), which is the upper ice surface. The bedrock topography is a few thousand meters below that. The same applies to 'bedrock' in the downstream box 'bedrock orography' to the left.

The idea to correct present-day topography/bathymetry with global sea-level does not make sense to me. Surface elevation over this time scale does not only change due to changing sea-level, also due to isostatic changes of the bedrock. In the periphery of the ice sheets where it may be most important for your biological application, the bedrock change may well be the dominant signal. I see in l169 that you acknowledge that bedrock changes are not taken into account. But why not?. ICE6G_C will give you a consistent set of data for sea level and bedrock elevation, in addition to ice thickness. In fact, the bedrock change is likely the most reliable output of ICE6G_C, because ice thickness is prescribed to get the right loading history. Why are you not using it? If you think your sea-level correction method gives a better representation of surface elevation, you should at least compare your results (with appropriate figures) against ICE6G_C. This will show if your method is an appropriate approximation for the full solution. My intuition is that surface elevation will be off by a few hundred meters in proximity of an ice sheet. If I am overlooking something obvious here, please explain better in the manuscript why you chose to not use the full information provided by ICE6G_C.

l128.
The following description still misses clear motivations for why things are done the way they are done. You should start by making clear what information you have (e.g. present day seabed and land topography), what you are trying derive, and how you are making approximations/interpolations to get there.

l130.
I think from "We first combined ..." you are no longer at the LGM. That would be good to make clear. E.g. with "that provides the surface elevation e at the present day"

l136.
Back to LGM? Confusing!
I think from l136-l147 (maybe even l149) you are describing how to produce a smooth ice surface elevation for the LGM on a high-resolution grid. Could be good to say that upfront.
The then following transition into a time-dependent estimate is confusing to me, because we don't know at this point how the interpolation between LGM and present will work. Maybe it would be better to leave the time dependence out of this for now?

l170.
Without any validation/comparison against reconstructed glacial extents from the literature, it is not possible to judge whether this interpolation approach is an innovative method or a bad idea. See also comment l220.

l171.
"As high-resolution estimates of glacial surface elevation are not available for timesteps t other than the LGM"
For the LGM you had Ehlers2011 to delineate glacier extents, but the resolution of the surface elevation data is the same for all time steps in ICE6G_C, isn't it?.

l173.
"at each time step 𝑡 ≠ 0"
But also not at present day, right?

l181.
"resampled to a 0.5° grid resolution"
Can you motivate the need for this intermediate grid? Why not interpolate one of the products to the grid of the other?

l220.
There are large data collections available to constrain the LGM-present glacial extent based on geomorphological constraints (e.g. https://doi.org/10.1016/j.quascirev.2015.09.016; https://doi.org/10.1016/S1571-0866(04)80209-4). How does your approximation compare to those reconstructions? Please include a validation for your method.

l240.
'idiosyncratic'
I have commented this before and supposedly you had changed it. Here it is again.

l299.
Could you motivate the parameter choices for c and h? Have you tried other values? Are they taken from past experience with the model or from published values?

l304.
Remind us what reference period the bias correction is calculated over and that you are using an intermediate grid.

l331.
"tas being interchangeable for tasmax and tasmin in Eq. 22 and Eq. 23"
This doesn't work, because you have already defined tas = (tasmax+tasmin)/2 in line 323. Need to find another symbol.

l335.
Figure2. It seems strange to switch region in the middle of the description from e) to f). It would be useful to continue with the first region to the end and add another figure for the second region if needed. For the validation of interpolated ice extent, it would also be good practice to show your reconstructed ice mask for a number of time slices through the deglaciation of the NH ice sheets.

l335.
Figure2. Fine to have a figure for the LGM, but it would be useful to also have a figure some time during the deglaciation. That is when interpolation to a different topography happens and eventual problems with the method could be inspected. I would suggest to document three time slices with figures similar to Fig2 for the LGM, halfway into the deglaciation and present day. Depending on how different they are, one or two could be pushed to a supplement.

Hide

RR by Anonymous Referee #2 (26 May 2022)

Suggestions for revision or reasons for rejection

The latest version of the manuscript leaves much better impression when compared with the previous one. I appreciate authors’ efforts to improve the manuscript based on both reviewers’ suggestions. It obviously required a lot of work, but the manuscript now has significantly improved structure and consequently, notably improved readability. Also, a newly added validation chapter crucially contributes to the quality of the manuscript. However, I still find there is a space and need for further improvements, so my observations and suggestions in this regard are as follows:
1) The title does not correspond adequately to what is presented in the manuscript. It seems to me that the data set is developed with intention to be used in paleo-ecology (at least the introduction indicates so), therefore, there is no reason not to put that into the title. Also, I would remove “V1.0” from the title, I find it useless. I would rather put “high-resolution data set” or even more specifically: “1-km data set” if that is something what will differentiate it from other potential similar datasets. A suggested title could be: “CHELSA-TraCE21k high-resolution (or “1-km”) data set - downscaled transient temperature and precipitation data since the last glacial maximum – development, validation and application in paleo-ecology”, or something similar, maybe shorter. If you want to maintain the current title, then I am afraid you would have to change the bigger part of the introduction.
2) It was really an unfortunate decision to name equally both the data set and the algorithm/model - CHELSA V1.2. Regrettably, I don’t see significant progress in clarification in that context throughout the manuscript. Only between the lines 60 and 121, a reader can find the next phrases: “CHELSA V1.2 algorithm”, “CHELSA V1.2 climate data set”, “CHELSA V1.2 mechanistic downscaling model”, “CHELSA V1.2 procedure”, “CHELSA downscaling model”, “CHELSA V1.2 model”. It is just unacceptable and extremely confusing. I find it necessary to add a paragraph, for example, somewhere at the beginning of chapter 2, clarifying that and informing the reader there are the dataset and the algorithm/model with the same name. It has to be very clear. And try to use only 2 words in its description throughout the text, for example “data set” and “model/algorithm”. In addition, once again, please, maintain consistency throughout the text and decide whether you want to use “TraCE21k” or “TraCE-21k”.
3) In abstract, you say:”High resolution, downscaled climate model data are used in a wide variety of applications across environmental sciences”. Then in chapter 6, you start with:”Transient long-term climatic data have a wide range of possible applications”. Please, add 2-3 examples where exactly, for example, just continuing the sentence by:”..., such as A, B and C, for example”.
4) Are there other comparable downscaled data sets available on the market? If yes, then, what is the advantage of CHELSA-Trace21k, why is it different/better then the other ones, why it should deserve attention, why is it unique? Please, clarify in a sentence/paragraph, in conclusion, for example.
5) Before submitting a final version of the manuscript, I suggest a thorough inspection regarding the consistency of the use of terms throughout the manuscript, as well as typing and other errors

Specific comments:

Line 8:
High-resolution
Lines 9-10:
Here we introduce a new, high-resolution dataset, named CHELSA-TraCE21k. It is obtained by downscaling TraCE-21k data, using CHELSA V1.2 algorithm with objective to create global monthly climatologies for temperature and precipitation at 30-arc sec spatial resolution in 100-year time steps for the last 21,000 years.
Line 18:
Validations show that CHELSA-TraCE21k V1.0 dataset reasonably represents the distribution...
Line 27:
GCM states for general circulation model, not global
coarser grain=>coarser resolutions
Line 34:
has been bridged, or had to be bridged
Line 59:
Here we present paleo-climatic data, downscaled from the CCSM3_TraCE21k model output (or TraCE21k dataset) to a 30-arc sec. resolution using the CHELSA V1.2 algorithm
Line 67:
in various parts of the world from
Line 68:
The TraCE-21k simulation has T31_gx3v5 resolution (…..)
Line 70-71:
Which resolution??
Please, specify the resolution per model component, for example CAM has resolution 3,75°x3,75°, which is important information for this manuscript, because you are downscaling atmospheric variables
Line 76:
It includes mean monthly daily 2m mean, minimum, and maximum temperature ?!
Line 78:
ERA stands for 'ECMWF Re-Analysis’
Lines 332-335:
Figure 2, axis labels not very clear, should be improved
Line 361, 364, 373:
RMSE, not RSME
Lines 444-450:
Figure 6, a) and b) missing on the figure; a) is also missing in the legend
Line 455:
...if the transient…
Line 498:
Separate Figure 7 legend from the text below

Referee Report: PDF

Hide

ED: Reconsider after major revisions (27 Jul 2022) by Irina Rogozhina

AR by Dirk N. Karger on behalf of the Authors (02 Aug 2022) Author's response Author's tracked changes Manuscript

ED: Reconsider after major revisions (15 Sep 2022) by Irina Rogozhina

AR by Dirk N. Karger on behalf of the Authors (13 Oct 2022) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (13 Oct 2022) by Irina Rogozhina

RR by Anonymous Referee #2 (09 Nov 2022)

ED: Publish subject to minor revisions (review by editor) (28 Nov 2022) by Irina Rogozhina

Dear authors,
With this letter, I would like to inform you that I have made a decision to publish your manuscript after minor revisions. The review process related to this manuscript has been extremely rocky and lengthy due to differences in opinions among reviewers, authors, and the editor (I). Even though it is tempting to opt for a quick completion of such a long process, we (reviewer 2 and I) see the need for some final adjustments in both the manuscript and communications with reviewer 1. Our recommendations are outlined below.
In his final report, reviewer 2 admits that the earlier points raised by his review have been mostly addressed, and he only has minor suggestions for the remaining changes in the manuscript included in their recommendations visible to the authors. Following these suggestions, please, do a thorough proof-read of the manuscript in its entirety, including the text, all figure captions, and legends. Please, ensure that the suggested changes are included in the manuscript’s title.
Furthermore, I have received a more detailed report from reviewer 2 discussing to which extent the criticism in review 1 has been addressed and whether the lack of responses to some major points is justified. I am combining this assessment by reviewer 2 with my own assessment in this final decision.
First, we do not feel that the authors have mastered the art of diplomacy required by the official peer-review process. Some of the responses to review 1 are at least indelicate if not arrogant or even borderline rude. I would like to eliminate such communication style from the official review process, even if you disagree with the reviewer’s line of argument. Not only does it obstruct the timely review process, since everyone gets irritated and confrontational, but it also undermines the fact that reviewers invest significant time into reading manuscripts and writing their reports. While doing so, they act as allies of the journal (if the review process is fair), with this suggesting that rudeness towards reviewers is also rudeness towards the editor whose choice of reviewers is far from random (you can be absolutely sure in this case). Taken together, these actions can be treated as disrespectful towards the journal as a whole. Although my final decision is generally in favor of the publication, the above issues will have to be corrected as part of the required minor revisions.
We have concluded that the authors have put significant additional work into improvements of their manuscript, following many suggestions of both reviewers. However, we would also like to stress that only due to reviewers’ time investment and ideas, this study is now close to publication. At the initial stage, it rather resembled work-in-progress and not a very thorough one, albeit with some potential that justified its initial publication in discussions. I believe that this energy-saving effort and the lack of diplomacy were the main reasons for tensions, which could be avoided altogether if the manuscript were stronger from the start.
Regarding some of the criticism from review 1 that has not been fully addressed, I encourage the authors to look at the first round of reviews to find the comment 2) mentioned by reviewer 1 in the second round (i.e., the interpolation procedure using temperature as a proxy for ice sheet retreat). It is not enough to simply state that you could not find it and thus could not explain why it was not fully addressed. Regarding your response to reviewer 1’s comment related to l12, as explained by the reviewer, ICE6G is not derived from an ice sheet model, and you have corrected this misunderstanding but only after the reviewer 1 has mentioned it twice (in the two rounds of reviews!). However, it is still too vague to write “ice sheet data” – please, look it up in the original article for a more precise definition.
Although I agree with reviewer 1 that GIA-related changes in the topographic forcing, in addition to changes due to sea level fluctuations, would improve the quality of your final dataset (~few hundreds of meters), I understand that this might not be possible at this stage. Even though it should not be that difficult based on my experience, I am ready to pass on this issue. Finally, your statement that reviewer 1 has ignored section 4.3 when talking about the validation against geological evidence is only partly valid. Why on Earth did you pick the ancient dataset of Dyke et al. (2003) to validate your model and why did you do it over North America only? What a strange choice considering the focus of your other analyses on Europe. There are extremely well resolved geochronological datasets for the British Isles, Scandinavia and even the European Alps. I understand the comments of the authors regarding their Section 4.3 being ignored, but I also emphasize that the way it is handled now is suboptimal from the point of view of the fields of geochronology and geomorphology. You need to at least discuss your choice of a region (and dataset) for validation. Otherwise, it leaves a heavy feeling.
Given the above, I would like the authors to walk through the remaining issues mentioned about and find elegant ways to address them at a limited labor cost. Furthermore, I am asking the authors to go back to their responses to review 1 and remove all the irritation-filled comments and overall apply good practices of diplomacy throughout their updated responses to reviewer 1. This effort will pay off handsomely with a manuscript publication.
Kind regards,
Irina

Hide

AR by Dirk N. Karger on behalf of the Authors (02 Dec 2022) Author's response Author's tracked changes Manuscript

ED: Publish as is (29 Jan 2023) by Irina Rogozhina

AR by Dirk N. Karger on behalf of the Authors (30 Jan 2023)

Post-review adjustments

AA: Author's adjustment | EA: Editor approval

AA by Dirk N. Karger on behalf of the Authors (14 Feb 2023) Author's adjustment Manuscript

EA: Adjustments approved (14 Feb 2023) by Irina Rogozhina

Short summary

Here we present global monthly climate time series for air temperature and precipitation at 1 km resolution for the last 21 000 years. The topography at all time steps is created by combining high-resolution information on glacial cover from current and Last Glacial Maximum glacier databases with the interpolation of an ice sheet model and a coupling to mean annual temperatures from a global circulation model.