Evaluation of CAMEL over the Taklimakan Desert Using Field Observations
, Wei Han 5,6,*, Zhenglong Li 7, E. Eva Borbas 7, Ali Mamtimin 1,2,3,4 and Yongqiang Liu 8Round 1
Reviewer 1 Report
The manuscript ‚Evaluation of CAMEL over the Taklimakan Desert using field observations’ submitted by Ma et al. utilizes PCA-filtered in-situ emissivity spectra, which were collected at 10 sites by Liu et al. during a field campaign in 2013, to evaluate the corresponding monthly land surface emissivities (LSE) spectra obtained from the CAMEL data set. The manuscript is generally well written and structured and the data and methods are presented in sufficient detail. However, the text contains too many errors and omissions and some of the conclusions drawn by the authors need to be backed up by additional research or references to already published research. Therefore, it is recommended to perform a major revision and re-review the manuscript.
Specific points:
- Throughout the paper the ‘mu’ sign in micrometer is missing
- A considerable number of references is missing in the list of references, e.g. in the introduction (Li et al., 2012), in section 2.1 (Liu et al., 2014), (Kora et al., 1996, 1999) – probably should be ‘Korb’, (Hook et al., 1996), and all references in section ‘4. Discussion’.
à I recommend to use automatic referencing, e.g. Zotero (https://www.zotero.org/).
- Line 97: please provide some examples / references for ‘… many other applications related to the surface.’
- Line 100: do you mean ‘0.01’, i.e. 1%? (I could not find the reference)
- Line 142: … and the hot blackbody temperature 10°C higher …
- Lines 142-144: this sentence is confusing. Please reformulate. In my opinion, section 2.1 needs to clearer (and shorter) and some of the sentences would benefit from a revision by an English native speaker.
- In the text you use ‘land-surface radiation temperature’, ‘LST’, ‘Ts’ (in eq. 1) and ‘GST’ (fig. 12). In my opinion, these are all the same, namely ‘LST’. Please justify the use of four different variables or use ‘LST’ everywhere.
- Figure 2 needs to be larger.
- Figure 5: in my opinion, fig 5 shows the residuals (original – reconstructed spectra). However, in the caption it says ‘bias’ and on the y-axis ‘Filtered EOBS’. Please explain or correct.
- Line 267: ‘The CAMEL imagery for Oct 2013 …’
- Figure 7b: in my opinion it should say ‘Hetian River’ and not ‘the Tarim River’ … Some of the other labels in the figure are also inconsistent with the text.
- Figure 8: I thought CAMEL are monthly data – how did you extract the LSE for 16 to 18 Oct 2013? Please explain.
- Figure 9: since the in in-situ LSE spectra are between 8 and 14 micron, I recommend to limit the CAMEL LSE to the same spectral range and adjust the y-axis so that the readers can see the spikes and other features in the spectra.
- Figure 10: labels of the sub-plots a, b, c are missing. In my opinion, the bars show differences (and not a bias).
- Lines 346 – 350: in my opinion it would be better to choose a representative CAMEL pixel rather than taking the pixel collocated with the in-situ measurement. This way you would avoid the influence of the – clearly unrepresentative – oasis and its vegetation.
- Lines 353-355: in my opinion, the effect of adsorption on LSE is relatively small. Please see Hulley et al. (2010; doi: 10.1016/j.rse.2010.02.002): ‘Laboratory measurements showed that the LSE at 8.6 μm of two different sand sources increased by 0.17 (17%) for Coral Pink Sands and 0.05 (5%) for Great Sands after wetting and returned to within 1% of the dry equilibrium state within one hour of drying.‘
and Masiello et al. (2015; doi:10.5194/amt-8-2981-2015), particularly figure 13, where LSE in SEVIRI channel 8.7 has a daily amplitude of about 0.005, which also may be an artifact from the diurnal cycle of LST.
à Please refer to / discuss these with other previous results.
à You may also want to reconsider your conclusions (lines 484-486).
- Line 361: do you mean ‘broader’ (instead of ‘longer’)?
- Figure 11: despite the PCA filtering, to me the ‘spikes’ look like an influence of the atmosphere / meteorological conditions on the EOBS, i.e. an artifact of the measurements. If the features are real, could you suggest a surface material that would correspond to them?
- Lines 392-395: see my comments above on adsorption.
- Figure 12: which sensor was used to measure GST (= LST)? I could not find this information in the manuscript.
- Lines 416-418: Usually their lack of temporal LSE variation makes deserts ideal LSE validation sites: why is that a problem here? I do not understand why CAMEL should have more difficulties over a quasi-static deserts than elsewhere. Please explain.
Lines 431-433 and 435-441: Ermida et al. (2018; doi: 10.3390/rs10071114) developed and implemented a model to produce LST with angular LST & LSE corrections. The correction has been implemented into LSA SAF’s operational chain for SEVIRI LST: https://landsaf.ipma.pt/en/data/products/land-surface-temperature-and-emissivity/.
Perez-Planells et al. (2022; doi: 10.1109/TGRS.2022.3224639) investigated several IR LSE angular variation models for vegetation canopies (i.e., such models already exist).
The manuscript is generally well written; however, section 2.1 would benefit from proofreading by an English native speaker.
Author Response
Response to Reviewer 1 Comments
The manuscript ‚Evaluation of CAMEL over the Taklimakan Desert using field observations’ submitted by Ma et al. utilizes PCA-filtered in-situ emissivity spectra, which were collected at 10 sites by Liu et al. during a field campaign in 2013, to evaluate the corresponding monthly land surface emissivities (LSE) spectra obtained from the CAMEL data set. The manuscript is generally well written and structured and the data and methods are presented in sufficient detail. However, the text contains too many errors and omissions and some of the conclusions drawn by the authors need to be backed up by additional research or references to already published research. Therefore, it is recommended to perform a major revision and re-review the manuscript.
Specific points:
Point 1: Throughout the paper the ‘mu’ sign in micrometer is missing.
Response 1: All missed ‘mu’ sign in micrometer has been added. Thanks for your kind hint.
Point 2: Line 142: … and the hot blackbody temperature 10°C higher …
Response 2: The unit °C in line 142 has been replaced with K.
Point 3: A considerable number of references is missing in the list of references, e.g. in the introduction (Li et al., 2012), in section 2.1 (Liu et al., 2014), (Kora et al., 1996, 1999) – probably should be ‘Korb’, (Hook et al., 1996), and all references in section ‘4. Discussion’.
à I recommend to use automatic referencing, e.g. Zotero (https://www.zotero.org/).
Response 3: Thanks. All the missed references has been added.
Point 4: Line 97: please provide some examples / references for ‘… many other applications related to the surface.’
Response 4: Two references have been added as is suggested.
Point 5: Line 100: do you mean ‘0.01’, i.e. 1%? (I could not find the reference)
Response 5: Yes, we do mean 0.01 in Line 100, and the reference is in Chinese and not a SCI paper.
Point 6: Line 142: … and the hot blackbody temperature 10°C higher …
Response 6: The unit °C in line 142 has been replaced with K.
Point 7: Lines 142-144: this sentence is confusing. Please reformulate. In my opinion, section 2.1 needs to clearer (and shorter) and some of the sentences would benefit from a revision by an English native speaker.
Response 7: Thanks. These couple of sentences has been rerote as is suggested.
Point 8: In the text you use ‘land-surface radiation temperature’, ‘LST’, ‘Ts’ (in eq. 1) and ‘GST’ (fig. 12). In my opinion, these are all the same, namely ‘LST’. Please justify the use of four different variables or use ‘LST’ everywhere.
Response 8: Thanks for your careful review and kind hints. All the phrases ‘land-surface radiation temperature’ and ‘GST’ have been replaced with LST.
Point 9: Figure 2 needs to be larger.
Response 9: Thanks. Figure 2 has been magnified by 1.5 times as is suggested.
Point 10: Figure 5: in my opinion, fig 5 shows the residuals (original – reconstructed spectra). However, in the caption it says ‘bias’ and on the y-axis ‘Filtered EOBS’. Please explain or correct.
Response 10: Thanks for your kind hint. Y-axis labie of Figure 5 has been corrected as residuals, and the exact meaning of residual has also been expleained in thie title of Figure 5.
Point 11: Line 267: ‘The CAMEL imagery for Oct 2013 …’
Response 11: The referred sentence has been revised as ‘Out of the 13 hinge points of the CAMEL for Oct 2013, only seven overlap with the spectral coverage of EOBS.’, with the suggested phrases added.
Point 12: Figure 7b: in my opinion it should say ‘Hetian River’ and not ‘the Tarim River’ … Some of the other labels in the figure are also inconsistent with the text.
Response 12: Thanks. Figure 7b has been updated with all referred errors corrected.
Point 13: Figure 8: I thought CAMEL are monthly data – how did you extract the LSE for 16 to 18 Oct 2013? Please explain.
Response 13: Yes, CAMEL are monthly averaged data. We extract the LSE to the observation 10 sites used in this paper, and the spatially interpolated CAMEL are simply used as the target LSE for evaluation using the experimentally observed LSED as the truth.
Point 14: Figure 9: since the in in-situ LSE spectra are between 8 and 14 micron, I recommend to limit the CAMEL LSE to the same spectral range and adjust the y-axis so that the readers can see the spikes and other features in the spectra.
Response 14: Out of the 13 hinge points of the CAMEL for Oct 2013, only seven overlap with the spectral coverage of EOBS. However, in the process of expanding CAMEL 13 hinge points to high spectral resolution (HSR) spectra, all the 13 hinges points of CAMEL are used, that’s why we remained all the wavelengths of HSR CAMEL in figure 9. The CAMEL LSE are limited to the same spectral range with in-situ LSE spectra are between 8 and 14 micron in Figure 11 when comparing with in-situ LSE as is suggested.
Point 15: Figure 10: labels of the sub-plots a, b, c are missing. In my opinion, the bars show differences (and not a bias).
Response 15: Thanks. Figure 10 has been revised as is suggested.
Point 16: Lines 346 – 350: in my opinion it would be better to choose a representative CAMEL pixel rather than taking the pixel collocated with the in-situ measurement. This way you would avoid the influence of the – clearly unrepresentative – oasis and its vegetation.
Response 16: Thanks for your valuable suggestion. In this study, we’ve aimed to evaluate the credibility of CAMEL using the in-situ measurement as the truth, and thus selected the spatial collocated pixel of CAMEL for its objective accessment. We’ll consider it in the following field experiments as well the further study concerning land surface emissivity.
Point 17: Lines 353-355: in my opinion, the effect of adsorption on LSE is relatively small. Please see Hulley et al. (2010; doi: 10.1016/j.rse.2010.02.002): ‘Laboratory measurements showed that the LSE at 8.6 μm of two different sand sources increased by 0.17 (17%) for Coral Pink Sands and 0.05 (5%) for Great Sands after wetting and returned to within 1% of the dry equilibrium state within one hour of drying.‘
and Masiello et al. (2015; doi:10.5194/amt-8-2981-2015), particularly figure 13, where LSE in SEVIRI channel 8.7 has a daily amplitude of about 0.005, which also may be an artifact from the diurnal cycle of LST.
à Please refer to / discuss these with other previous results.
à You may also want to reconsider your conclusions (lines 484-486).
Response 17: Thanks a lot for your scientific points. As is suggested, some discussion of the refered previous results has been added around Lines 353-355, and some of the conclusions in lines 484-486 have also been reconsidered. By the way, in order to validate this opinion, we are now designing another set of field experiments, which will be conducted around Jun 10th, 2023. Corresponding results will be presented after that.
Point 18: Line 361: do you mean ‘broader’ (instead of ‘longer’)?
Response 18: Thanks. The words ‘longer’ in line 361 has been replaeced with ‘broader’.
Point 19: Figure 11: despite the PCA filtering, to me the ‘spikes’ look like an influence of the atmosphere / meteorological conditions on the EOBS, i.e. an artifact of the measurements. If the features are real, could you suggest a surface material that would correspond to them?
Response 19: The EOBS has been obtained in two days when weahter conditions there is sunny and less cloudy. Such spikes has also been detected by ELAB as is showed by the two red lines for two typical sandy samples in Figure 3.
Point 20: Lines 392-395: see my comments above on adsorption.
Response 20: Thanks for your kind hints, and discussion concerning impact of absoption upon LSE has been added in proper position as referred in Response 17. By the way, the phyrases ‘surface skin temperature’ have also been replaced with LST.
Point 21: Figure 12: which sensor was used to measure GST (= LST)? I could not find this information in the manuscript.
Response 21: The GST in Figure 12 has been replaced with LST. The sensor used to measure LST is the 109 Temperature Probe, which uses a thermistor to measure temperature in air, soil, and water.
Point 22: Lines 416-418: Usually their lack of temporal LSE variation makes deserts ideal LSE validation sites: why is that a problem here? I do not understand why CAMEL should have more difficulties over a quasi-static deserts than elsewhere. Please explain.
Response 22: In our opinion, the temporal LSE variation might be larger over a quasi-static deserts than elsewhere, since the temporal varation of LST, soil moisture as well as evaporation are larger in desert region than elsewhere as well, and all these three factors are correlated with LSE.
Point 23: Lines 431-433 and 435-441: Ermida et al. (2018; doi: 10.3390/rs10071114) developed and implemented a model to produce LST with angular LST & LSE corrections. The correction has been implemented into LSA SAF’s operational chain for SEVIRI LST: https://landsaf.ipma.pt/en/data/products/land-surface-temperature-and-emissivity/.
Perez-Planells et al. (2022; doi: 10.1109/TGRS.2022.3224639) investigated several IR LSE angular variation models for vegetation canopies (i.e., such models already exist)
Response 23: Thanks a lot for provding detailed consturction of IR LSE models. Acoording to this comments, the referred content has been added to the second paragraph of the discussion section.
Thanks so much again for your detailed comments, which do help us in improving this paper. In the future, we will further conduct relevant study fllowing your precious suggestions and scitific opinions, and further design relevant field experiments to investigate these viewpoints in depth.
Reviewer 2 Report
The manuscript intends to evaluate CAMEL data over TD through comparison them with ground observation data. Although the idea of the manuscript looks interesting, there are still several issues that need to be addressed before publication. Therefore, I would suggest major revision. The detailed comments are as follow:
1: In the whole manuscript when you talk about wavelength should be µm, please revise all.
2: L 21: use full expression for LSE.
3: In the abstract please in brief mention the usefulness of your study.
4: L52: you can use the following reference as it discusses the source of uncertainties in NWP:
https://doi.org/10.5194/isprs-annals-IV-4-W2-175-2017
10.1155/2020/9730129
5: L92: You can use the following reference for soil moisture:
10.3390/app7060566
6: Please clearly mention the innovation of your study in the last paragraph of the introduction.
L124: Please revise this sentence.
7: Explain about the site selection stage.
8: L 207: Please mention the source of noise in the observation data.
9: Please explain about your study period, why only two days in October? Do you think this short period can give us good overview about CAMEL feasibility?
10: As it was mention CAMEL is monthly averaged data however you compared these data with only averaged two days’ data in October. How we can verify your results and your final conclusion?
11. Why CAMEL yield a U-shaped graph? Please explain it more considering the spatial features of each station.
Comments for author File: 
Comments.pdf
Please put effort in English proofing.
Author Response
Response to Reviewer 2 Comments
The manuscript intends to evaluate CAMEL data over TD through comparison them with ground observation data. Although the idea of the manuscript looks interesting, there are still several issues that need to be addressed before publication. Therefore, I would suggest major revision. The detailed comments are as follow:
Point 1: In the whole manuscript when you talk about wavelength should be µm, please revise all..
Response 1: All missed ‘mu’ sign in micrometer has been added. Thanks for your kind hint.
Point 2: L 21: use full expression for LSE
Response 2: Thanks. The full espression of LSE has been provided in the beginning part of the abstract.
Point 3: In the abstract please in brief mention the usefulness of your study.
Response 3: The usefulness of our study has been briefed in the abstract as follows:
Infrared (IR) land surface emissivity (LSE) plays an important role in numerical weather prediction (NWP) models through the satellite radiances assimilation. However, due to the large uncertainties in LSE over the desert, many land-surface sensitive channels of satellite IR sensors are not assimilated. This calls for further assessments of the satellite retrieved LSE quality in these desert regions
Point 4: L52: you can use the following reference as it discusses the source of uncertainties in NWP:
https://doi.org/10.5194/isprs-annals-IV-4-W2-175-2017, 10.1155/2020/9730129.
Response 4: Thanks. We’ve added the suggested reference in the context.
Point 5: L92: You can use the following reference for soil moisture: 10.3390/app7060566
Response 5: The suggested reference has been added as follows:
Cai, J., Zhang, Y., Li, Y., Liang, X., & Jiang, T. (2017). Analyzing the Characteristics of Soil Moisture Using GLDAS Data: A Case Study in Eastern China. Applied Sciences, 7(6), 566. https://doi.org/10.3390/app7060566
Point 6: Please clearly mention the innovation of your study in the last paragraph of the introduction.
L124: Please revise this sentence.
Response 6: Thaks for precious advises. The innovation of our study has been mentioned in the last paragraph of the introduction, with the setence in Line 124 revised.
Point 7: Explain about the site selection stage.
Response 7: The site selection stage has been further explained in the third and forth line of section 2.1.
Point 8: L 207: Please mention the source of noise in the observation data.
Response 8: The source of noise in the observation data has been mentioned in Line 213-215.
Point 9: Please explain about your study period, why only two days in October? Do you think this short period can give us good overview about CAMEL feasibility?.
Response 9: You are right. The currently available in-situ LSE measurement are refined to these two days, that’s why we could only use them as the reference to evaluate CAMEL.
Point 10: As it was mention CAMEL is monthly averaged data however you compared these data with only averaged two days’ data in October. How we can verify your results and your final conclusion?
Response 10:
Point 11: Why CAMEL yield a U-shaped graph? Please explain it more considering the spatial features of each station
Response 11: Thanks. The spatial features of each station has been introduced in Table 1. And the cauces of U-shape distribution feature of both CAMEL and EOBS has been further explained in section 3.1 as is suggested.
Author Response File: 
Author Response.docx
Round 2
Reviewer 1 Report
Your response 2 missed my point: in lines 149 -150 please also write 'hot blackbody' (which is 10 K hotter than the environment).
Lines 139-140: the spectrometer is sensitive to TIR radiance, not just to that from a blackbody - please correct.
Lines 151 - 152: the end of this sentence does not look right (erase the '... , the.').
Line 152: the precision of the measurement is +-0.002. What you provide is mean LSE (0.998) and precision (stdev). Please correct.
General comment: the meaning of uncertainty, bias and precision is specified in GUM (https://www.bipm.org/en/committees/jc/jcgm/publications) and - with special focus on LST - in Guillevic et al. (2018), https://doi.org/10.5067/doc/ceoswgcv/lpv/lst.001
Line 448: it should say Perez-Planells (not 'Lluis')
Point 3 has not been addressed: '
A considerable number of references is missing in the list of references, e.g. in the introduction (Li et al., 2012), in section 2.1 (Liu et al., 2014), (Kora et al., 1996, 1999) – probably should be ‘Korb’, (Hook et al., 1996), and all references in section ‘4. Discussion’. I recommend to use automatic referencing, e.g. Zotero (https://www.zotero.org/).'
---> the above literature still needs to be included in the 'References'
Author Response
Response to Reviewer 1 Comments
Point 1: Your response 2 missed my point: in lines 149 -150 please also write 'hot blackbody' (which is 10 K hotter than the environment).
Response 1: Thanks. It has been corrected.
Point 2: Lines 139-140: the spectrometer is sensitive to TIR radiance, not just to that from a blackbody - please correct.
Response 2: Thanks. It has been corrected.
Point 3: Lines 151 - 152: the end of this sentence does not look right (erase the '... , the.').
Response 3: Thanks for your hint. The refered ’, the’ has been deleted.
Point 4: Line 152: the precision of the measurement is +-0.002. What you provide is mean LSE (0.998) and precision (stdev). Please correct.
Response 4: Thanks. It has been corrected.
Point 5: General comment: the meaning of uncertainty, bias and precision is specified in GUM (https://www.bipm.org/en/committees/jc/jcgm/publications) and - with special focus on LST - in Guillevic et al. (2018), https://doi.org/10.5067/doc/ceoswgcv/lpv/lst.001
Response 5: Thanks. I’ve visited https://www.bipm.org/en/committees/jc/jcgm/publications and downloaded the reffered reference concerning LST as follows:
Guillevic, P., Göttsche, F., Nickeson, J., Hulley, G., Ghent, D., Yu, Y., Trigo, I., Hook, S., Sobrino, J.A., Remedios, J., Román, M. & Camacho, F. (2018). Land Surface Temperature Product Validation Best Practice Protocol. Version 1.1.0 In P. Guillevic, F. Göttsche, J. Nickeson & M. Román (Eds.), Good Practices for Satellite-Derived Land Product Validation (p. 58): Land Product Validation Subgroup (WGCV/CEOS), doi:10.5067/doc/ceoswgcv/lpv/lst.001
According to this reference, all the words ‘Bias’ in the paper has been revise as difference or residual. The difference in study mean CAMEL minus EOBS, while the residule means the incrment of EOBS before and after filtering (Figure 5).
Point 6: Line 448: it should say Perez-Planells (not 'Lluis')
Response 6: Thanks. It has been corrected.
Point 7: Point 3 has not been addressed: '
A considerable number of references is missing in the list of references, e.g. in the introduction (Li et al., 2012), in section 2.1 (Liu et al., 2014), (Kora et al., 1996, 1999) – probably should be ‘Korb’, (Hook et al., 1996), and all references in section ‘4. Discussion’. I recommend to use automatic referencing, e.g. Zotero (https://www.zotero.org/).'
Response 7: Thanks. The reference (Korb et al., 1999) has been removed, and all other referred refereces have been added as follows:
Li, Z., Li, J., Li, Y., Zhang, Y., Schmit, T.J., Zhou, L., Goldberg, M.D., & Menzel, W.P., (2012). Determining diurnal varia-tions of land surface emissivity from geostationary satellites. Journal of Geophysical Research: Atmospheres, 117(D23), n/a-n/a. https://doi.org/10.1029/2012JD018279
Liu Y.,Ali M.,Huo W., Yang X.,Liu X., Meng X.,He Q. Estimation of the land surface emissivity in the hinterland of Taklimakan Desert. J. Mt. Sci. 2014, 11, 1143-1151.
Korb, A.R., Dybwad, P., Wadsworth, W., & Salisbury, J.W., (1996). Portable Fourier transform infrared spectroradiome-ter for field measurements of radiance and emissivity. Appl. Optics(35), 1679-1692. https://doi.org/10.1364/AO.35.001679
Francois, C., Ottle´, C., & Pre´vot, L. (1997). Analytical parametrisation of canopy emissivity and directional radiance in the thermal infrared: Application on the retrieval of soil and foliage temperatures using two directional measure-ments. International Journal of Remote Sensing, 12, 2587 – 2621.
McAtee, B. K., A. J. Prata, and M. J. Lynch, (2003). The angular behavior of emitted thermal infrared radiation (8–12 μm) at a semiarid site. J. Appl. Meteor. Climatol, 42, 1060–1071, https://doi.org/10.1175/1520-0450(2003)042<1060:TABOET>2.0.CO;2.
Ermida, S., Trigo, I., DaCamara, C., & Pires, A. (2018). A methodology to simulate lst directional effects based on parametric models and landscape properties. Remote Sensing, 10(7), 1114. https://doi.org/10.3390/rs10071114
Lluís Pérez-Planells, Raquel Niclòs, Enric Valor, Frank-Michael Göttsche. (2022). Retrieval of Land Surface Emissivi-ties Over Partially Vegetated Surfaces From Satellite Data Using Radiative Transfer Models. IEEE Trans. Geosci. Remote. Sens. 60, 1-21. https://doi.org/10.1109/TGRS.2022.3224639
Banghua Yan, J. Le Marshall. JCSDA Community Radiative Transfer Model (CRTM) [J]. American Geophysical Union, 2005, 122.
Torresani, M.; Masiello, G.; Vendrame, N.; Gerosa, G.; Falocchi, M.; Tomelleri, E.; Serio, C.; Rocchini, D.; Zardi, D. (2022). Correlation analysis of evapotranspiration, emissivity contrast, and water deficit indices: a case study in four eddy covariance sites in italy with different environmental habitats. Land, 11, 1903. https://doi.org/ 10.3390/land11111903
Masiello, G., Serio, C., Venafra, S., Liuzzi, G., Poutier, L., & Göttsche, F.-M. (2018). Physical retrieval of land surface emissivity spectra from hyper-spectral infrared observations and validation with in situ measurements. Remote Sensing, 10(6), 976. https://doi.org/10.3390/rs10060976
Point 8: ---> the above literature still needs to be included in the 'References'
Response 8: Thanks. All the above literature have been added into the reference list.
Reviewer 2 Report
Thank you for your effort in first revision. However, as you ignored to address some of my questions and comments, I would suggest major revision. All comments must be addressed carefully. Moreover, some of your explanations are not satisfying. So please provide more comprehensive response to all comments.
1: In the whole manuscript when you talk about wavelength should be µm, please revise all.
2: L 21: use full expression for LSE.
3: In the abstract please in brief mention the usefulness of your study.
4: L52: you can use the following reference as it discusses the source of uncertainties in NWP:
https://doi.org/10.5194/isprs-annals-IV-4-W2-175-2017
10.1155/2020/9730129
5: L92: You can use the following reference for soil moisture:
10.3390/app7060566
6: Please clearly mention the innovation of your study in the last paragraph of the introduction.
L124: Please revise this sentence.
7: Explain about the site selection stage.
8: L 207: Please mention the source of noise in the observation data.
9: Please explain about your study period, why only two days in October? Do you think this short period can give us good overview about CAMEL feasibility?
10: As it was mention CAMEL is monthly averaged data however you compared these data with only averaged two days’ data in October. How we can verify your results and your final conclusion?
11: Why CAMEL yield a U-shaped graph? Please explain it more considering the spatial features of each station.
Please edit the English of the manuscript as well.
Author Response
Response to Reviewer 2 Comments
Thank you for your effort in first revision. However, as you ignored to address some of my questions and comments, I would suggest major revision. All comments must be addressed carefully. Moreover, some of your explanations are not satisfying. So please provide more comprehensive response to all comments.
Point 1: In the whole manuscript when you talk about wavelength should be µm, please revise all.
Response 1: Thanks. All unit of wavelength in this manuscript has been revised to be µm.
Point 2: L21: use full expression for LSE.
Response 2: Thanks. The full espression of LSE has been provided in L21 as follows:
Infrared (IR) land surface emissivity (LSE) plays an important role in numerical weather prediction (NWP) models through the satellite radiances assimilation.
Point 3: In the abstract please in brief mention the usefulness of your study.
Response 3: The usefulness of our study has been briefed in the abstract as follows:
Infrared (IR) land surface emissivity (LSE) plays an important role in numerical weather prediction (NWP) models through the satellite radiances assimilation. However, due to the large uncertainties in LSE over the desert, many land-surface sensitive channels of satellite IR sensors are not assimilated. This calls for further assessments of the satellite retrieved LSE quality in these desert regions.
Point 4: L52: you can use the following reference as it discusses the source of uncertainties in NWP:
https://doi.org/10.5194/isprs-annals-IV-4-W2-175-2017
10.1155/2020/9730129
Response 4: Thanks. The suggested references have been referred while discussing the souce of uncertainties in NWP around L52, and they have also been added in tha reference list.
H.Karimian, Q.Li, C.C.Li, J.fan, C.Gong, L,Jin, Y.Mo. Daily Estimation of fine particulate matter mass concentration through satellite based aerosol optical depth. ISSC, 2017.
Liu, Y., Wang, R., Gao, J., & Zhu, P. (2020). The Impact of Different Mapping Function Models and Meteorological Parameter Calculation Methods on the Calculation Results of Single-Frequency Precise Point Positioning with Increased Tropospheric Gradient. Mathematical Problems in Engineering, 2020, 1–12. https://doi.org/10.1155/2020/9730129.
Point 5: L92: You can use the following reference for soil moisture:10.3390/app7060566
Response 5: Thanks. The suggested reference has been referred for soil moisture in L92, its has also been added in tha reference list.
Cai, J., Zhang, Y., Li, Y., Liang, X., & Jiang, T. (2017). Analyzing the Characteristics of Soil Moisture Using GLDAS Data: A Case Study in Eastern China. Applied Sciences, 7(6), 566. https://doi.org/10.3390/app7060566
Point 6: Please clearly mention the innovation of your study in the last paragraph of the introduction.
L124: Please revise this sentence.
Response 6: Thanks. The innovation of our study has been mentioned in the last paragraph of the introduction, with the sentence in Line 124 revised as follows:
It is emphasized that the previous validation of CAMEL are conducted using satellite products as the reference (Li et al. 2012), this study continues the efforts to compare and validate CAMEL using the highly accurate in situ emissivity measurements in desert regions as the reference instead.
Point 7: Explain about the site selection stage.
Response 7: The site selection stage has been further explained in the third and forth line of section 2.1 with the following sentence added:
The sites were selected along a south/north desert road in TD every 50 km (Liu et al. 2014) with their exact locations and land-use category shown in Table 1.
Point 8: L207: Please mention the source of noise in the observation data.
Response 8: The source of noise in the observation data has been further explained in the manuscript as follows:
The maximum value of EOBS is even larger than 1.0 for some sites. Such a phenomenon is likely caused by excessive noise in the observing process, which is common in LSE measurements (Liu et al., 2014). The sample temperature was measured using thermo-couples. The method of using thermocouples to measure sample temperature is not suitable for samples in natural environments with high roughness, poor thermal con-ductivity, and small thermal inertia, such as soil. The inaccuracy of temperature measurement will directly lead to a large error in the final calculated specific Radiance.
Point 9: Please explain about your study period, why only two days in October? Do you think this short period can give us good overview about CAMEL feasibility?
Response 9: You are right. The currently available in-situ LSE measurement are refined to these two days, that’s why we could only use them as the reference to evaluate CAMEL. we tried to filter out the possible obervation errors of in-situ measurements by using the first six principle components analized from CAMEL in this study, so that some climatic averaged information of LSE has also been added to the in-situ measurement, with majority features of the in-situ LSE has been remained.
Point 10: As it was mention CAMEL is monthly averaged data however you compared these data with only averaged two days’ data in October. How we can verify your results and your final conclusion?
Response 10: The in-situ LSE measurements have been obtained along a desert road in the Taklimankan Desert in China in October, 2014. The natural conditions there are extremely harsh, that’s why the time of the in-situ LSE are limited to two days. In addition, possibly due to the harah natural condition, one of our portable Fourier transform infrared spectrometer (FTIR) malfunctioned, and the exact causees of the malfunction are unkonwn till now. We’ve ever contacted with the factory that produced the FTIR, they were also unable to cope with it. Several years later, when we needed to conduct the LSE field obervation experiment again, we had to rental another FTIR from a university nearby. Considering all this objective factors above, we tried to filter out the possible obervation errors of in-situ measurements by using the first six principle components analized from CAMEL in this study, so that some climatic averaged information of LSE has also been added to the in-situ measurement, with majority features of the in-situ LSE has been remained.
Point 11: Why CAMEL yield a U-shaped graph? Please explain it more considering the spatial features of each station.
Response 11: Thanks. The spatial features of each station has been introduced in Table 1. And the cauces of U-shape distribution feature of both CAMEL and EOBS has also been further explained in section 3.1 as follows:
Both sites have LSE substantially larger than inner desert sites, due to the fact that, these bands, located in the fundamental vibrational stretching modes in the IR range, often have high reflectance (Hapke, 1993) and low emissivity according to Kirchhoff’s law of thermal radiation; they are extremely sensitive to the surface soil quartz contents. The surface soil at site 5 is quicksand, with fine sand accounts for over 98% of the total weight, silt accounts for 1.5%, clay accounts for 0.5%, and the average particle size of fine sand is 136.0 μm. There are two types of surface soil at site 9, namely quicksand and ancient river channel silt. The average particle size of quicksand varies greatly, ranging from 15.6 to 250.0 μm, which is a mixture of extremely fine sand, fine sand, medium silt, and coarse silt. Fine sand accounts for over 93% of the total weight, silt accounts for 4%, and clay accounts for 0.5%, which is close to the soil composition at observation site 5. The surface silt of ancient river channels is different from quicksand, with fine sand accounting for 77% of the total weight, silt accounting for 10%, and clay accounting for 13%. In the clay surface of observation point 10, fine sand accounts for 73% of the total weight, silt accounts for 9%, and clay accounts for 18%. The determi-nation of organic matter content in soil at a depth of 0-10 cm shows that quicksand contains 4.3 g/kg of organic matter, and ancient river silt contains 10.5 g/kg of organic matter (Korb et al., 1999; Hook et al., 1996). Unfortunately, the organic matter content of clay in Populus euphratica forest was not determined, and its organic matter content should be higher. The soil properties of other observation points on the underlying surface of quicksand are basically the same as those of observation site 5.
Author Response File: Author Response.docx
