A Process-Based Model for Arctic Coastal Erosion Driven by Thermodenudation and Thermoabrasion Combined and including Nearshore Morphodynamics
Round 1
Reviewer 1 Report
Major comments:
The manuscript presents a model of coastal erosion based on several modules: thermal abrasion, thermal denudation, wave action and sediment transport. Authors then parameterize the model using observations along two transects and then validate the model at other transects. The authors are going into details to discuss each module, but it is unlikely that the presented model can reproduced at the given or other sites.
The lack of reproducibility is likely a main shortcoming of the presented manuscript. For example, the module responsible for slumping, discussed in Appendix C2, make a bullet point list of steps, but does not specify how quickly the slumping occurs. No mentioning of sediment creep dynamics. A similar argument is applicable to Appendix C3 – slope failure conditions. The governing equations for modeling the storm surge are not closed, i.e., the value of V is not clear how to compute or parameterize. It would be optimal to discuss the physics, rheology (plastic vs elastic vs viscos-plastic), hydrodynamical assumptions, if possible, to present governing equations and then present step-based algorithms as in the Appendix.
Authors are encouraged to go over each presented equation, e.g. (1) and (3), and conduct a thorough analysis of units. The units of the left- and right-hand sides do not match in many equations. There are typos in the equations and units presented in Table 5. Please double checked the formulae and Table 5, if necessary to add more parameters and values to the latter. Authors list units for some variables and parameters, while omitting units for the other variables. It is encouraged to list units for all variables as soon as they appear in the manuscript.
After presenting the model, authors calibrate the model to match to observations, however the uncertainty in values of the recovered parameters is not discussed. Which parameters can be recovered with a small and large uncertainty? A method used to estimate the values and proceeding was not discussed in the details, or it was obscured.
Although a sensitivity analysis has been conducted for inclusion/omission of main modules, the sensitivity analysis of modeling results with respect to recovered values is not conducted. Please consider adding a sensitivity analysis with respect to parameter values.
Minor comments:
Page 4; Lines 105-106: “The boundary BC2 is defined where the mean sea level (hm) and wavelength (λ) have a ratio of less than 0.5.” Why this ratio?
Page 4: Line 113: “permafrost melting”, permafrost is thawing.
Page 6: “The thawing depth (xt) is defined as the depth of permafrost melting or freezing face“, Probably you want to say the seasonal thaw depth. Please use an official permafrost glossary about different terms and definitions
Figure 3: “Permafrost line”, Permafrost table!
Equation 3: There should be no “dt” on the left-hand side. This formula assumes that all incoming heat flux is used to thaw ice or freeze water and that the ground temperature always stays at 0C. Which is not entirely correct. It is suggested to consider proper ground temperature modeling, where the heat flux on the surface is given according to (2). Some comparisons on this simplified methodology for cold and warm climates is deemed necessary.
Equation 4: Coefficients are rather arbitrary. Why a 0.05m threshold? Probably depends on the soil material, i.e. clay vs silt vs sand vs consolidated
Page 15: Lines 363-365: “The remaining parameters of the model are calibrated by iteration to find the combinations that yield the closest estimate of the erosion volume. For the iteration, the initial values and upper/lower limits for the different parameters are assessed based on theoretical estimates and site observations” What exactly method was used to estimate the parameters? Are they unique? What is the uncertainty?
Page 21: 467-468: “Applying the environmental forcing iterations are made with the calibrated parameters to match the outcome of the simulation to the three targets: net erosion volume, crest retreat and bluff slope as closely as possible.” More details…. Non-uniqueness Uncertainty in the values
Author Response
We, the authors, like to express our sincere appreciation to the reviewer for the critical review of the work and for pointing out the shortcomings of the manuscript. We addressed the issues raised by the reviewer to our best understanding.
Attached are our responses to the comments from the reviewer.
regards
Authors
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Reviewed manuscript dedicated to the very crucial problem of simulation of coastal erosion, which is one of the most critical processes in the Arctic. The model developed by the authors takes into consideration both thermal abrasion and thermal denudation processes. It is essential to notice, that the latter is still an understudied process. I recommend accept the paper for publication with minor edits.
First of all, I recommend to the authors provide comprehensive justification for such model parameters as timestep (3 hours as it is noticed at the page 9), boundary values of ice concentration 20% and water level a the bluff 10 cm (page 10) as well as soil properties accepted as input parameters for the model. The authors validated the model at one observation site in Western Siberia but did not reference any data about permafrost characteristics other than temperature and thaw depth, while physical properties (especially ice content) are extremely crucial for understanding the modeled processes.
As for the terminology, the authors use the words "melting permafrost", which is incorrect. permafrost does not melt, it thaws, but ground ice melts. Please, use the words "thawing permafrost" or "melting ground ice".
Here are a few more small suggestions:
ln. 42 Replace "non-erodable" to "slow-erodable". There are no absolutely non-erodable shores.
Table 1 - wrong units for the latent heat of permafrost.
ln. 272 I would specify that the model had been validated only at one! research site. So far this sentence sounds like the authors verified the model for the entire Arctic coast.
Table 2. Mixed up columns of Erosion and Accretion for case#2.
Author Response
We thank the reviewer for the insightful comments and suggestions to improve the journal paper. We revised the manuscript based on the comments received from the reviewer. Our responses are attached as a pdf.
regards
Authors
Author Response File: Author Response.pdf
Reviewer 3 Report
The research is nice. The writing is standardized. It is worthy of acceptance for publication.
Of course, there are many shortcomings in the manuscript.
But the paper must be properly revised.
Comments for author File: Comments.pdf
Author Response
We, the authors, thank the reviewer for agreeing to assess the quality of the paper. He/she provided valuable comments, drew attention to some shortcomings, and guided us to improve the quality of the scientific work. We modified the journal to reflect the comments and resubmit it for his/her perusal.
Attached are our responses to the comments from the reviewer.
regards
Authors
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The manuscript is improved, but there is still some work to do.
1) I am not sure if the current manuscript can lead to the reproducible results. The authors are willing to share the code, but the manuscript itself should be clearly written. The code will be long gone, but the paper will be survive. Besides, the code is not being reviewed.
The section on the storm surge modeling now has a clear link, where the model is described. However, the sub-module 'slumping' needs more attention. It is not clear how the addition "The models by Simpson et al. [47], Sun and Huang [48] or similar can be used to determine the mcr." can be used to find the value of m_cr. The provided references do not directly deal with finding m_cr.
2) On the topic of model calibration and recovery parameters, the authors provided the reply "We do not have enough measurements to estimate the uncertainties in the model input parameters, e.g. to fit a probability density function for each parameter. Hence, we used instead an upper and lower limit values for each parameters. These limits were deducted based on a theoretical evaluation or from field observations. Once the limits where identified, a trial-and-error procedure were used to calibrate some of the parameter" But it is not clear what method/procedure was employed to estimate the parameters? What exactly stands under a trial-and-error procedure? What metric was employed to judge if a fit is satisfactory? It is know that there is no unique set of recovered parameters and there are multiple solutions. Why the authors selected this solution and not the other?
3) Sensitivity study with respect to main parameters. Thank you for adding the plots to Appendix I. Unfortunately, I do not understand the results related to the Air Temperature. It looks like in both cases -20% and +20% increase of the air temperature, the erosion increases, with respect to the base case. But, for the colder climate the erosion should be slower.
Also, if the water temperature is increasing the coast should be eroding quickly, but the quickest erosion rate is for -10% case (colder water)
Something strange with the model and it needs to be explained.
Other comments and suggestions seem to be addressed.
Author Response
We thank the reviewer for carefully examining the manuscript and for his/her comments. We prepared answers to the comments and attached them as pdf.
best regards
Authors
Author Response File: Author Response.pdf
Reviewer 3 Report
The author carefully revised all the questions.
The author responded to all comments.
The quality of the manuscript has been greatly improved.
It is recommended to accept and publish.
Author Response
We thank the reviewer for his/her positive assessment and for approving the manuscript.
best regards
Authors.
