Evaluation of Three Air Temperature Reanalysis Datasets in the Alpine Region of the Qinghai–Tibet Plateau
Round 1
Reviewer 1 Report
Review
The paper intends to evaluate three air temperature reanalysis datasets, ERA5L, GLDAS, and CLDAS, in the Tibetan Plateau using 17 in-situ observations over the period 2017-2018. The evaluation involves the spatial distribution and temporal changes at daily, monthly, and seasonal scales. The paper is well written overall, however, several issues and concerns need to be addressed to improve the paper quality prior to the recommendation of paper acceptance. Major revision is needed.
Major
1. Abstract needs to be concise, too many details and general expressions. For example, there are four sentences to elaborate the study significance, and after three exhibited results, major conclusions are shown, thus authors did not well summarize the major highlights.
2. Conclusion: it is too lengthy in the present form. Please summarizing the main findings and understandings.
Minor
1. Figure 13a can be combined with Figure 1
2. L37, “Third Pole”, “Asian Water Tower”
3. L52-55, the reference (Wang, X., Pang, G., & Yang, M. (2018). International Journal of Climatology, 38(3), 1116-1131.) could be more appropriate for describing the distribution of meteorological stations in the TP.
4. L58-60, two papers including the use of reanalysis dataset to reveal surface temperature warming over the TP (Gao, K., Duan, A., Chen, D., & Wu, G. (2019). Science Bulletin, 64(16), 1140-1143.) and regionalizing seasonal precipitation using ERA5 over the TP (Lai, H. W., Chen, H. W., Kukulies, J., Ou, T., & Chen, D. (2021). Journal of Climate, 34(7), 2635-2651.) should be covered.
5. L83-88, It is good to fully follow the already done research about the evaluation of reanalysis datasets, however, the previous study of Huang et al. has verified the three reanalyses, the authors obtained conclusion generally resemble those of Huang et al., so what is the innovation of your study?
6. L158-159, grammar issue in the sentence, please correct it.
7. L210, why |CC| <0.4 is regarded as low linear correlation, and higher than 0,7 means high linear correlation? The CC is obviously related to the sample number.
8. L225, the subtitle is not appropriate, it that data characteristic described in the section?
9. Captions of Figures 2-6, there should be annual mean, winter, spring, summer, autumn, respectively. Please modify them.
10. L288, should be datasets.
11. Table 3, which time scale? Please clarify. From the results, it seems to be the monthly or daily, but here you are doing the analysis for annual mean.
12. Figures 8-9b, how are the daily/monthly CCs generated? Are they performed based on the 17 sites between reanalyses and observations? In that sense, it could be spatial correlation, since temporal correlation could be not like this, it should be one value based on figures 8-9a, this also applies to Figures 8-9 (c-e) and Figure 10. Please clarify the figure captions.
13. L487-489, this should be in introduction.
14. L512-515, the statement is not all correct. Since the lack of observations in the TP, how CLDAS can integrate the number of stations instead of the other analyses. ERA5 indeed assimilated several station data in the TP.
15. L533-534, this study could test the errors caused by the different interpolation methods you are discussing.
16. L602, you are evaluating the reanalyses in the alpine area, but why permafrost area appears so suddenly?
17. Reference format is not consistent, please check.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Please find the review in the attached file.
Comments for author File: Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
The study with a title of “Evaluation of Three Air Temperature Reanalysis Datasets in the Alpine Region of the Qinghai-Tibet Plateau” has been seriously reviewed. Overall, this manuscript is well organized and written in English. However, I still find that there exist some major issues in this manuscript, and I would like to recommend to be conducted a major revision. The detailed comments can be found below.
Comments:
1. In this study, the authors would like to use the independent air temperature datasets to evaluate the performance of three reanalysis product. However, I can only get less information about the in situ observations. Therefore, the authors have to show more detailed information about these observations, especially for the height of the measurements. If these observations is not installed on 2m height, I think that the evaluation may be not reliable. As I know, these in situ observations may be from the different flux net systems. So, there may be many gaps, how did you fill these gaps? Please clarify.
2. Please the authors carefully and seriously check the used validation metrics. Why you select the NSE metric? As I know, this metric is mostly used in the hydrology researches. Moreover, this equation of NSE is wrongly expressed here, and the NSE is larger 1 (NSE should be between -∞ and 1). Please corrected this fault, and the related descriptions and figures. This fault represents that the authors have not rigorous attitude. Maybe, the combined validation metric of KGE is better here.
3. How did get the grid values when you compare their values against the observations? Reanalysis product have different spatial resolution. If you did not resampled them into the same spatial resolution, I think that it is unfair to compare their performance. So, I would like to suggest the authors to resample all the products into the same spatial resolution, i.e., 0.25o x 0.25o.
4. It is confusing that different CC values existed between Table 3 and figure 7. Why? Please explain, and revise the related expressions for reducing the confusing.
5. What is BIAS in some figures? You did not define the metric of BIAS in the manuscript. The equation of ME is wrong, please correct it and be careful.
6. It is so subjective to select several transects for evaluating temperature. I suggest the authors to do the comparisons following the procedures below: you can classified all the sites into different types according to altitude (land cover), and then calculate the validation metrics and conduct comparisons.
7. Line 494, please add the reference number for Ding et al.
8. Please carefully check the list of literatures. For example, there is no the journal of name in the 38th literature.
9. You should acknowledge the contribution of the in situ data collectors.
10. In Line 433, NSEE should be NSE. There are some similar errors in this manuscript, and please the authors seriously check.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Although the authors have well addressed the last concerns of review, minor revision is needed to modify the following issues.
Abstract: “the influence of interpolation and spatial resolution have little influence on the evaluation results.” The expression has a little bit issue in soundness. The readers maybe misunderstand that spatial resolution does not have influence on model evaluation if they do not go through the whole paper. Please clarify it.
Sections 3.4 & 4.1 subtitle should be consistent with others, such as Evaluation at Individual Sites.
L681-682, several grammar issues in the sentence.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors have revised the MS in accordance with my comments and am fine with their response and revised MS.
Author Response
Thank you very much for you appreciation.
