Mapping the Thermal State of Permafrost in Northeast China Based on the Surface Frost Number Model
Round 1
Reviewer 1 Report
I think the authors need to present a detailed comprehensive model development in the methodology section. I recommend the manuscript for publication after the following major changes:
1. Why did you choose the surface frost number model in this study? What is the novelty of the proposed methodology over other established DL techniques(eg, MASK RCNN) which are also successfully used in different permafrost applications? Please introduce these studies and related others, and their potential impacts. The authors should explain this aspect in the introduction section. Otherwise, the readers cannot see the importance of your proposed methods over other techniques.
2. Can you explain ice degradation in warming permafrost? Is there any impact on hydrology over the study area? You can include the following studies:
https://www.nature.com/articles/ngeo2674
https://onlinelibrary.wiley.com/doi/abs/10.1002/hyp.8350
https://www.mdpi.com/2072-4292/13/4/558/htm
3. Can you provide high-resolution figures (4,9,10,13,14)
4. I didn't see any climatic information over the study area. You can discuss climatic differences over the study area. Can you add the Köppen-Geiger climatic zones(table/map) for the study area?
5. What is the fundamental theory about the frost number model? What software or programs are used to fit the model? You can provide a table with hyper parameters for the method used in this study.
6. How do the models perform spatially? Could you please provide a feature map for the frost number model which will help readers to understand how the model detects observation initially?
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In this manuscript, MODIS LST data were used to dynamically analyze the thermal state of frozen soil in northeastern China from 2003 to 2019, and to classify the frozen soil types in the study area based on a surface frost number model. The interesting point of this study is the improvement of MODIS LST using vegetation factors, but the current version of the experimental scheme is unconvincingly designed. In addition, there are numerous spelling errors in the article. Generally, I recommend the manuscript to be published after major revisions. Here are my comments:
Major comments:
1. Figure 1: According to the information of meteorological stations shown in the figure, the authors used only 11 meteorological stations located in the northern part of the study area, what is the reason? As far as I know the number of meteorological stations in northeastern China is large (Zhang et al., 2021).
2. Line 226-241 and Line 258-259: The improvement of MODIS LST according to the vegetation factor is the biggest innovation of this study, but it seems that the authors' design of this part of the experimental method is arbitrary at present. Please give a detailed reference for each step of this part of the operation, such as reference to other scholars' research results or statistical relationships based on the measured data?
3. Section 3.6: At present, other scholars have published studies related to permafrost development in northeastern China. It is suggested that the authors conduct an overall and local comparative analysis of the classification results of this paper with other scholars, to show the specific differences between the results of this study and other scholars, and it is difficult to judge the superiority of the final results from the accuracy index alone.
Minor comments:
Line 27: The relationship between precipitation and frozen soil thermal state does not seem to be discussed in the article, so it is difficult to support this conclusion.
Line 53: “thermal thawing lakes and ponds” should be revised to “thermokarst lakes and ponds”, and please check the rigor of the terminology throughout the article.
Line 114: “useing” should be revised to “using”, and please check the spelling errors in the whole article.
Line 174: “aie” should be revised to “air”. In addition, please explain the mean annual air temperature and mean annual precipitation in Northeast China.
Line 206: “mean annual air temperature(MAAT)”, the full spelling is only required when the term is first listed in the article (Line 174).
Line 207-209: Please give details of meteorological stations and frozen soil temperature monitoring sites and add them to a table, such as longitude, latitude, elevation, monitoring period, monitoring height (depth), etc.
Line 253: Please give the data source for Figure 2b and add a citation in the corresponding place in the article.
Figure 4 (a): The current presentation is confusing. Suppose the blue column represents the MAAT change rate for China (0.26°C/10a) and the purple column represents the MAAT change rate for each weather station. Why are most of the weather stations with MAAT change rates higher than China's MAAT change rates located in the southern part of northeastern China? This is not consistent with the description in the article.
[1] Zhang, Z. Q., Wu, Q. B., Hou, M. T., Tai, B. W., & An, Y. K. (2021). Permafrost change in Northeast China in the 1950s–2010s. Advances in Climate Change Research, 12(1), 18-28.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The submitted paper presents interesting and original results from modelling and field investigations concerning permafrost conditions affected by recent warming trends in northeast China.
After careful consideration, I recommend that the paper be published only after the authors address the issues listed below:
Lines 41-42: Permafrost is a thermal state of the ground and does not necessary contain ice (see Harris et al., 1988).
The text could be shortened in several places within the Introduction – for instance concerning the description of ``Surface temperature as a key factor for permafrost occurrence…you can delete phrase at line 62-63 since is a repetition of 49-50.
You are using the expression permafrost melting, but it might be wrong, since permafrost is thawing (e.g., line 94).
Citation 55 and 88 is the same!
Line 116 see if Nelson et al. is correct or should pe replaced with Nelson and Outcalt!
Line 378 replace Nelson with Nelson and Outcalt.
Phrase at lines 173-176 should be rewritten!
Study region:
Please describe a little bit more the climate of the region (or insert a MAAT/pp map).
It would also be appropriate to insert a permafrost distribution model of the region here. How much of the region is affected by permafrost? Maybe some general considerations on the permafrost characteristics would also be necessary.
It will also be interesting to mention the elevation of the 0ºC and of the lower boundary of permafrost.
Line 201-202: the DEM has a 500 m resolution? I doubt this is correct!
Fig 2b: there is no scalebar on the map. What is the difference between grassland and grass? Why you choose to create two separate classes? Replace shrubbery with shrubs!
Line 266: please replace melting with thawing
Lines 288-289. You calculated the change of MAAT in each year compared only to 1960? 1960 is the reference year and then all the others are compared with this particular one? If so, I see no relevance for this analysis and I suggest calculating moving average (9 or 11 years interval) or to compare at least decades or periods of 30yrs.
Line 304 replace climate warming with air temperature rising. Better avoid the popular but scientifically incorrect term “climate warming”.Climate is defined as an average of meteorological conditions and as such, cannot “warm”.
Do you think you can add on the figures 4,5,6,7 a polygon//line representing the extent of permafrost in the study region? You can use models of Gruber/Obu to extract this limit, if there are no regional models.
Lines 353-355 and 366-368: how can you now where is sporadic, discontinuous or continuous permafrost in the described regions?
Lines 383-387: is not clear if the classification is from the literature (in this case please describe it in the Methodology) or is your contribution (in this case you should explain why setting these thresholds).
Fig 8b,c – what is UF and UB? Please explicitly write in the figure caption.
Figure 9 caption: Are you sure that the entire region lies in permafrost conditions. If not you should refer only to regions with permafrost occurrence.
Fig 10: Please check if 10e is ok?
Fig. 10: In the caption please explain what does a,b,c…j means. Maybe you could highlight these boreholes on fig. 1.
Fig. 11 + interpretation is weird. Your model says that at some points permafrost is stable, but it seems that boreholes data reveal permafrost absence. I couldn`t understand very well the explanation, so maybe you could improve here.
Fig. 12. I don`t understand why there are two y axis named Area? Are you sure the results are ok? It seems that stable permafrost occur between 0 and 300m? Is this correct?
Lines 558-559:What about snow in this region? This factor is extremely important for atmosphere-ground thermal interactions.
It is not clear if NDVI trend is positive or negative and if this trend is in accordance with temp/pp evolution.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors significantly improved the quality of the paper by addressing most of the comments. This research work will be very effective in the remote sensing imagery community. I recommend the manuscript for publication. Congratulation to the authors!!
Reviewer 3 Report
The authors addressed all the comments and I consider that the paper can be published in the present form.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
