Derivation of Vegetation Optical Depth and Water Content in the Source Region of the Yellow River using the FY-3B Microwave Data

Round 1

Reviewer 1 Report

Overall: this is an attempt to develop a more accurate method to retrieve VOD for areas with large water bodies. Although the method develop can provide a good addition to this subject, the results are poorly discussed and the paper is not contextualised with the literature, and so it does not provide enough information on its main contributions to the subject.
Images must be reworked in order to be published.
Sections are somewhat disorganised: results section also provides discussion, and the discussion section is a summary section.
The results also need to be better presented to show the improvements of this method over the current method applied.
English and some style errors also need to be corrected.

Comments:

last paragraph of introduction: the text here is the delineation of the methodology, and should be in that section. Here you should summarise the problem you are addressing, the objetive of your manuscript, and the way it is organised.

line 133: u not mu

figure 1: the images are too small considering the details presented, and should be reworked. Also add a box containing the zoomed area of the map containing the monitoring network

Line 266: not sure what you meant to say here? did you mean that there were 5 sample collections over the period from June to October?

Line 264-266: this information is better explained in the next paragraph, so consider removing it to avoid repetition

Figure 3: this graph needs to be better presented for publication (labels, design,...)

Section 4: i like the way this section is organised, however its name does not reflect its content. I recommend to name this "Results and discussion", and section 5 "Summary and conclusions".

Section 4.1: here you should present your results, and discuss them, similar to section 4.2. The images are not enough. For example, the images look pretty similar, what are the main differences? How are these results important for the next sections?

Figures 5 and 6: a figure with the difference between VOD and VODe would improve visualisation of results.

Sections 4.2.3: correlation values are a bit difficult to visualise in Figure 8, so perhaps adding a few metrics of these coefficients will improve understanding (e.g. mean values for each soil cover).
Here, I presume you are using VOD considering water bodies... how do the "default" VOD performs in these correlations (NDVI, MVI and VWC)?

Figure 9: is there a spatial/pixel pattern in this correlation?

Line 491: why was it that VOD considering the water bodies was superior?

Section 5: there is very little discussion and contextualisation of your results with the literature. How do they compare? What is the main achievements of your paper? This is poorly stated here.

Figure 11: same as figure 3

Author Response

Response to Reviewer 1 Comments

 

Overall: this is an attempt to develop a more accurate method to retrieve VOD for areas with large water bodies. Although the method develop can provide a good addition to this subject, the results are poorly discussed and the paper is not contextualised with the literature, and so it does not provide enough information on its main contributions to the subject.

Images must be reworked in order to be published.

Sections are somewhat disorganised: results section also provides discussion, and the discussion section is a summary section.

The results also need to be better presented to show the improvements of this method over the current method applied.

English and some style errors also need to be corrected.

 

Response: We thank the reviewer for the constructive comments and suggestions as well as the careful reading. We reorganised the structure and the result section has been rewritten in the revised version.  As a result, we believe that the quality of the revised version has been improved significantly. In the paragraphs that follow, we include point-to-point responses.

 

Comments:

 

last paragraph of introduction: the text here is the delineation of the methodology, and should be in that section. Here you should summarise the problem you are addressing, the objetive of your manuscript, and the way it is organised.

Response: Thank the reviewer for questioning this issue. This paragraph has been redesigned in the revised version.

 

 

line 133: u not mu

Response: We look back the manuscript and do not find ‘mu’ in line 133.

 

 

figure 1: the images are too small considering the details presented, and should be reworked. Also add a box containing the zoomed area of the map containing the monitoring network

Response: Thanks for the reviewer to point this. We re-designed the Figure 1 and added the location of the catchment in the revised manuscript to clearly explain the catchment and the soil moisture monitoring network located.

 

 

Line 266: not sure what you meant to say here? did you mean that there were 5 sample collections over the period from June to October?

Response: Sorry for this confusion. There are 20 monitoring sites chosen for vegetation samples according to the SMST monitoring network, and each site sampled 5 times. 100 samples were obtained initially. Unfortunately, 5 samples were scorched by misoperation. So, a total of 95 samples are obtained during the entire experiment period, including a total of 87 samples from the monitoring network, and 8 samples from the source region required for validation. This sentence has been added into the revised manuscript.

 

 

Line 264-266: this information is better explained in the next paragraph, so consider removing it to avoid repetition

Response: We fully agree with the reviewer’s opinion here. This sentence has been deleted in the next paragraph.

 

 

Figure 3: this graph needs to be better presented for publication (labels, design,...)

Response: Thanks for the reviewer to point this. We reworked this figure and the date form on x axis has been changed in the revised version.

 

 

 

Section 4: i like the way this section is organised, however its name does not reflect its content. I recommend to name this "Results and discussion", and section 5 "Summary and conclusions".

Response: We fully agree with the reviewer’s opinion here. Section headline has been rewritten in the revised manuscript.

 

 

Section 4.1: here you should present your results, and discuss them, similar to section 4.2. The images are not enough. For example, the images look pretty similar, what are the main differences? How are these results important for the next sections?

Response: Sorry for this typo. The missing paragraph has been added into the revised manuscript.

 

 

 

Figures 5 and 6: a figure with the difference between VOD and VODe would improve visualisation of results.

Response: We thank the reviewer for the constructive comments and suggestions. Figure 5 and Figure 6 present the spatial distribution of VOD. Compare with these two figures, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 6(b), (c) and (d). This demonstrates that during the July to September, the VOD from previous method was invalid in some area at least in our study.  Moreover, the temporal variation of retrieved VOD in figure 7 (b) also show failed from previous method. The VODr (gray trend line) represents the variation trends of the algorithm from previous method and this gray trend line missed most of the time in figure 7 (b). In this case, VOD retrieved from previous method was failed in our study area. Thus, difference maps can’t be presented in manuscript due to the previous method is invalid in some part of our study area.

 

 

Sections 4.2.3: correlation values are a bit difficult to visualise in Figure 8, so perhaps adding a few metrics of these coefficients will improve understanding (e.g. mean values for each soil cover).

Here, I presume you are using VOD considering water bodies... how do the "default" VOD performs in these correlations (NDVI, MVI and VWC)?

Response: As last explanation, difference maps can’t be presented in manuscript due to the previous method is invalid in some part of our study area.

 

 

 

Figure 9: is there a spatial/pixel pattern in this correlation?

Response: Maqu regional scale soil moisture and soil temperature monitoring network is set up in the water source region of the Yellow River to validate the accuracy of the satellite-derived soil moisture products for the Tibetan Plateau. This network has the capability to monitor the spatial and temporal soil moisture variability of the area with a high degree of accuracy (Dente et al., 2012; Zheng et al., 2017). It consists of 20 stations distributed. The VWC sampling is collected according to the 20 site.  Dente (Dente et al., 2012) proposed the weighted spatial average of measured soil moisture was successfully used as ground reference for the validation of the AMSR-E soil moisture products and ASCAT soil wetness index products. The results of Dente’s study are very encouraging for our applications. So, the match problem can be ignore in our opinion. These information has been added into the Validation section.

 

Dente, L.; Vekerdy, Z.; Wen, J.; Su, Z. Maqu network for validation of satellite-derived soil moisture products. Int. J. Appl. Earth Obs. 2012. 17, 55-65. DOI: org/10.1016/j.jag.2011.11.004

Zheng, D.H.; Wang, X.; van der Velde, R.; Zeng, Y.; Wen, J.; Wang, Z.; Schwank, M.; Ferrazzoli, P.; Su, Z. L-Band Microwave Emission of Soil Freeze-Thaw Process in the Third Pole Environment. IEEE Trans. Geosci. Remote Sens.2017, 99, 1-15. DOI:10.1109/TGRS.2017.2705248

 

 

Line 491: why was it that VOD considering the water bodies was superior?

Response: More detailed discussion was added the revised version. Unlike VWC or soil moisture, VOD can’t be observed or verify directly by field work. Figure 5 and Figure 6 present the spatial distribution of VOD. Compare with these two figures, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 6(b), (c) and (d). This demonstrates that during the period from July to September, the results without including the influence of water bodies (namely, the wetland coverage) are poor in some part of the region, while the algorithm with considering water bodies gives better results. This also means the VOD method without considering the influence of water bodies was invalid in some area at least in our study.  Moreover, the temporal variation of retrieved VOD in figure 7 (b) also shows that the method without considering the influence of water bodies was failed. The VODr (gray trend line) represents the variation trends of the algorithm without the influence of water bodies and this gray trend line missed most of the time in figure 7 (b). This also demonstrates VOD method considering the water bodies is superior.

 

Section 5: there is very little discussion and contextualisation of your results with the literature. How do they compare? What is the main achievements of your paper? This is poorly stated here.

Response: We fully agree with the reviewer’s opinion here. The result section has been rewritten in the revised manuscript.

 

 

Figure 11: same as figure 3

Response: Thanks for the reviewer to point this. We reworked this figure.

 

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

This paper proposes deriving vegetation optical depth and vegetation water content from source region of the Yellow River using Fengyun-3B Microwave Radiation Image (FY-3B MWRI), passive microwave satellite data set. I did not review the part related to the theory since my background in mathematics and passive microwave theory are rather weak. Besides that, I enjoyed reading the text and I did not find any major flaw in the manuscript. The minor comments are:

1. Title: I suggest the authors complement the title of the paper with “Using Fengyun-3B Microwave Radiation Image (FY-3B MWRI) data set”.

2. Figure 4 was not cited neither discussed in the text. A paragraph or even a short sentence should be provided about this fractional coverage of open water bodies.

3. Gyaring Lake and Ngoring Lake appear several times in the text. The location of these lakes in Figure 4 or Figure 5 is important for those don´t know the study area.

4. L441: Authors said that VOD is usually assumed to be linearly proportional to VWC, however, a logarithmic relation is presented right after this sentence. I suggest chance the paragraph to: “Although VOD is usually assumed to be linearly proportional to VWC (citations), a relatively good logarithmic correspondence between these two parameters were obtained in this study”.

Author Response

Response to Reviewer 2 Comments

 

This paper proposes deriving vegetation optical depth and vegetation water content from source region of the Yellow River using Fengyun-3B Microwave Radiation Image (FY-3B MWRI), passive microwave satellite data set. I did not review the part related to the theory since my background in mathematics and passive microwave theory are rather weak. Besides that, I enjoyed reading the text and I did not find any major flaw in the manuscript. The minor comments are:

 

Response: We thank the reviewer for the constructive comments and suggestions as well as the careful reading. As a result, we believe that the quality of the revised version has been improved significantly. In the paragraphs that follow, we include point-to-point responses.

 

1. Title: I suggest the authors complement the title of the paper with “Using Fengyun-3B Microwave Radiation Image (FY-3B MWRI) data set”.

Response: We are very grateful for your comments on our manuscript. We revised the title as ’Derivation of Vegetation Optical Depth and Water Content in the Source Region of the Yellow River using the FY-3B microwave data’.

 

 

2. Figure 4 was not cited neither discussed in the text. A paragraph or even a short sentence should be provided about this fractional coverage of open water bodies.

Response: Sorry for this typo. The missing paragraph has been added into the revised manuscript.

 

 

3. Gyaring Lake and Ngoring Lake appear several times in the text. The location of these lakes in Figure 4 or Figure 5 is important for those don´t know the study area.

Response: Sorry for this confusion. The first time Gyaring and Ngoring Lake appeared was in Figure 1(Line 211) with ‘The two blue regions in the upper left corner of Figure 1 denote Gyaring Lake and Ngoring Lake, respectively’. The geographic coordinates of the two lakes were added into the the revised manuscript.

 

4. L441: Authors said that VOD is usually assumed to be linearly proportional to VWC, however, a logarithmic relation is presented right after this sentence. I suggest chance the paragraph to: “Although VOD is usually assumed to be linearly proportional to VWC (citations), a relatively good logarithmic correspondence between these two parameters were obtained in this study”.

Response: Thank the reviewer for raising this issue. We fully agree with the reviewer’s comments and this paragraph was rewritten.

 

Author Response File: Author Response.docx

Reviewer 3 Report

Review of “Derivation of Vegetation Optical Depth and Water Content in the Source Region of the Yellow River” for possible publication in the journal Remote Sensing.

 

In this study, the authors used the FY-3B satellite’s MWRI sensor to improve the tau-omega model with regard to the radiative contribution from water bodies to the pixels over the wetland. I enjoyed this paper. In addition, I found some issues that should be explained or improved before publication. The followings are my major comments:

1) Figures must be appropriately edited. The current figures are very low quality and some are unreadable, 2) Explain which correlation coefficient equation is used. If the authors used the Pearson-correlation coefficient, the entire manuscript and results should be re-written because the NDVI, MVW, and VOD values do not share the same variance and data ranges; so the ranked Pearson correlation coefficient would be more appropriate, 3) because the authors used field observation-based VWC, a possible  mismatch in spatial resolution between the field data and satellite-based data should be investigated. What are the possible errors caused by spatial mismatch? 4) Without considering the fractional seasonal wetlands in the source regions, what are the biases and RMSE? How much improvement could be made by considering the fractional seasonal wetlands? 5) Since VOD and VWC do not use the same units of measurement, comparing them directly does not make sense. As an example, the authors have written, “The results indicate that the VOD is in the range of 0.20 to 1.20 and the VWC is in the range of 0.20kg/m2 to 1.40kg/m2 in the entire source region of the Yellow River in 2012” – this does not have any meaning. It would be better to provide the anomaly of these values. 6) The authors should include previous research to support their findings. For example, why are the RMSE values of VWC 0.12 km/m2 considered to be successful retrievals of VWC? Which previous research are they using for comparison? 7) The method section should be rewritten. As currently written, the methods proposed here are impossible for other researchers to recreate. And 8) The entire abstract should be rewritten. The current abstract omits too much information. For instance, what is the major finding of the current research? What is the goal of the current research?

 

 

Specific comments:

L26, 30:  m2 -> m²

What is the exact meaning of “source region of” ? Why not just say “the Yellow River?”

The method sections are unnecessarily long. Much of this information has already been published in previous research. I recommend that the authors reduce the methodology section and use proper citations instead.

L115: references should be added

L122: Г^P

L124: Please provide detailed information regarding how to obtain the e^p_s and Fresnel coefficient.

L127: Why was the vegetation single scattering albedo fixed to 0.05?

L133: How did you obtain the value of optical thickness of vegetation?

L139: Shielded?

L155-159: Please be more specific since this is the main assumption and the novel portion of this study.

L172: What does “typical” indicate here?

How was equation 5 obtained?

L181-198: References are needed.

L194-196: Specific information is required. The method shown in this section should be replicable by other researchers. How are the emissivity, albedo, and rs,h(v) obtained? 

L205: km²

L288-293: km² (please check the manuscripts for any other instances of this error)

L310: References are needed.

L316: Why was the algorithm valid for non-precipitating conditions?

Move Figures 3 and 9 to the supplementary information

Please describe the status of RFI in this study area

L344-349: This is hard to follow or to find the specific reason regarding the results. What is the specific reason for having no VOD in JAS months?

L355: Relative to what?

L357: Why are only these regions sensitive to T and SM?

L364-370: It seems that not all the discussion is supported by the results. The authors should be more specific or else conduct further research to support all the assertions made here.

4.2.2: Why were these two sites selected?

As with Figure 10, I recommend that the authors make difference maps such as VWC from the previous method – VWC from the current method.

Throughout the entire manuscript, the authors have misused the term “relative”. Please delete incorrect uses of “relative”. 

Author Response

Response to Reviewer 3 Comments

In this study, the authors used the FY-3B satellite’s MWRI sensor to improve the tau-omega model with regard to the radiative contribution from water bodies to the pixels over the wetland. I enjoyed this paper. In addition, I found some issues that should be explained or improved before publication. The followings are my major comments:

 

1) Figures must be appropriately edited. The current figures are very low quality and some are unreadable, 2) Explain which correlation coefficient equation is used. If the authors used the Pearson-correlation coefficient, the entire manuscript and results should be re-written because the NDVI, MVW, and VOD values do not share the same variance and data ranges; so the ranked Pearson correlation coefficient would be more appropriate, 3) because the authors used field observation-based VWC, a possible  mismatch in spatial resolution between the field data and satellite-based data should be investigated. What are the possible errors caused by spatial mismatch? 4) Without considering the fractional seasonal wetlands in the source regions, what are the biases and RMSE? How much improvement could be made by considering the fractional seasonal wetlands? 5) Since VOD and VWC do not use the same units of measurement, comparing them directly does not make sense. As an example, the authors have written, “The results indicate that the VOD is in the range of 0.20 to 1.20 and the VWC is in the range of 0.20kg/m2 to 1.40kg/m2 in the entire source region of the Yellow River in 2012” – this does not have any meaning. It would be better to provide the anomaly of these values. 6) The authors should include previous research to support their findings. For example, why are the RMSE values of VWC 0.12 km/m2 considered to be successful retrievals of VWC? Which previous research are they using for comparison? 7) The method section should be rewritten. As currently written, the methods proposed here are impossible for other researchers to recreate. And 8) The entire abstract should be rewritten. The current abstract omits too much information. For instance, what is the major finding of the current research? What is the goal of the current research?

 

 

Response: We thank the reviewer for the constructive comments and suggestions as well as the careful reading. As a result, we believe that the quality of the revised version has been improved significantly. In the paragraphs that follow, we include point-to-point responses.

(1)We re-design the Figures to be better present in the revised manuscript.

(2)The ranked Spearman correlation coefficient is used in our study and some information has been added in the revised manuscript.

(3)Maqu regional scale soil moisture and soil temperature monitoring network is set up in the water source region of the Yellow River to validate the accuracy of the satellite-derived soil moisture products for the Tibetan Plateau. This network has the capability to monitor the spatial and temporal soil moisture variability of the area with a high degree of accuracy (Dente et al., 2012; Zheng et al., 2017). It consists of 20 stations distributed. The VWC sampling is collected according to the 20 site.  Dente (Dente et al., 2012) proposed the weighted spatial average of measured soil moisture was successfully used as ground reference for the validation of the AMSR-E soil moisture products and ASCAT soil wetness index products. The results of Dente’s study are very encouraging for our applications. So, the match problem can be ignore in our opinion. These information has been added into the Validation section.

(4) Unlike VWC or soil moisture, VOD can’t observe or verify directly by field work. Compare with the figure 5 and figure 6, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 6(b), (c) and (d). This demonstrates that the results from previous method are poor in some part of the region. Moreover, the temporal variation of retrieved VOD in figure 7 (b) also show the method without considering the influence of water bodies was failed. The VODr (gray trend line) represents the variation trends of the algorithm without the influence of water bodies and this gray trend line missed most of the time in figure 7 (b). Due to the previous method was failed and invalid, it's unnecessary to present the biases and RMSE.

(5) The anomaly of the VOD or VWC values are important for indicate the abnormal variability. There are only 5 months VOD and VWC data, and these can’t capture the anomaly information. Further research would be conduct to provide the long-time series variability.

(6) Following your suggestion, the comparison between other similar regions has been added into the section 4

(7) The method section has been modified in the revised manuscript.

(8) The abstract has been rewritten.

 

Dente, L.; Vekerdy, Z.; Wen, J.; Su, Z. Maqu network for validation of satellite-derived soil moisture products. Int. J. Appl. Earth Obs. 2012. 17, 55-65. DOI: org/10.1016/j.jag.2011.11.004

Zheng, D.H.; Wang, X.; van der Velde, R.; Zeng, Y.; Wen, J.; Wang, Z.; Schwank, M.; Ferrazzoli, P.; Su, Z. L-Band Microwave Emission of Soil Freeze-Thaw Process in the Third Pole Environment. IEEE Trans. Geosci. Remote Sens.2017, 99, 1-15. DOI:10.1109/TGRS.2017.2705248

 

 

Specific comments:

 

L26, 30:  m2 -> m2

Response: Sorry for this typo. The superscript is used in the revised version.

 

What is the exact meaning of “source region of” ? Why not just say “the Yellow River?”

Response: The source region of the Yellow River is located in the north-eastern part of the Tibetan Plateau of 95°30'-103°30'E and 31°30'-36°30'N, while the Yellow River located in  90°-120°E and 30°-40°N approximately. Figure 1 shows the location of the source region of the Yellow River catchment, red line in upper right corner in figure 1 also shows the Yellow River located in China. Sorry for this confusion. Figure 1 has been redesigned into the revised manuscript.

                                             

Figure 1. Location of the source region of the Yellow River catchment

 

The method sections are unnecessarily long. Much of this information has already been published in previous research. I recommend that the authors reduce the methodology section and use proper citations instead.

Response: This section has been edited and condensed for clarity. Some unnecessary process about the method has been deleted.

 

 

L115: references should be added

Response: The reference has been added into the revised manuscript.

 

 

 

L122: ГP

Response: Sorry for this typo. The missing p has been superscripted and added in the revised version.

 

L124: Please provide detailed information regarding how to obtain the ep_s and Fresnel coefficient.

Response: Sorry for this confusion. We look back the manuscript and rewrite this sentence.

 

 

L127: Why was the vegetation single scattering albedo fixed to 0.05?

Response: Here, 0.05 is cited by Jones’(2011) work and we find that el,h and el,v typically converge to 0.95 for high biomass vegetation cover, which is also consistent with previous studies (English, 2008; Peterson et al., 2000); therefore the vegetation single scattering albedo is set to 0.05. Yang et al (2007) also analysed single scattering albedo, size factor X and complex refractive index m based on the Mie theory, and find that the single scattering albedo appears gradually large while X ascends, but finally equals to 0.5 approximately. So, 0.05 is proper in our study in our opinion.

 

English, S. J. The importance of accurate skin temperature in assimilating radiances from satellite sounding instruments. IEEE Transactions on Geoscience and Remote Sensing. 2008, 46(2), 403−408.

Peterson, T. C., Bassist, A. N., Williams, C. N., Grody, N. C. A blended satellite-in situ near-global surface temperature dataset. Bulletin of the American Meteorological Society. 2000, 81(9), 2157−2164.

Jones, M.O.; Jones, L.A.; Kimball, J.S.; Mcdonald, K.C. Satellite passive microwave remote sensing for monitoring global land surface phenology. Remote Sens. Environ. 2011, 115, 1102-1114. https://doi.org/10.1016/j.rse.2010.12.015

Yang C.P., Wu J., Wan M. Calculation of single scattering albedo in the radiative transfer equation. Laser Journal, 28(4), 45-, 2007.

 

L133: How did you obtain the value of optical thickness of vegetation?

Response: L135: ‘Equations (1) and (2) are usually called the τ-ω model. Given that the other parameters in equations (1) and (2) are known, one can derive VOD’. In previous method, VOD can be obtained directly by equations (1) and (2), meanwhile the fraction of open water is considered in the τ-ω model (equations 7) in our method.

 

 

L139: Shielded?

Response: This sentence has been deleted according to one reviewer’s opinion.

 

 

L155-159: Please be more specific since this is the main assumption and the novel portion of this study.

Response: Thanks for the reviewer to point this. This paragraph has been rewritten according to one reviewer’s opinion.

 

 

L172: What does “typical” indicate here?

Response: We use this value according to the previous studies for high biomass vegetation cover. This word has been deleted in the revised version.

 

 

How was equation 5 obtained?

Response: Eq. (5) is simplified and linearized from its usual quadratic form by ignoring atmospheric emissions reflected by the surface (Jones et al., 2010). This reference has been deleted in the revised version

 

Jones, L.A.; Ferguson, C.R.; Kimball, J.S.; Zhang, K.; Chan, S.T.K.; McDonald, K.C.; Njoku, E.G.; Wood, E.F. Satellite microwave remote sensing of daily land surface air temperature minima and maxima from AMSR-E. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2010, 3, 111–123. 10.1109/JSTARS.2010.2041530

 

 

L181-198: References are needed.

Response: We fully agree with the reviewer’s opinion here. The references have been added into the revised manuscript.

 

 

 

L194-196: Specific information is required. The method shown in this section should be replicable by other researchers. How are the emissivity, albedo, and rs, h(v) obtained?

Response: We agree with the reviewer’s suggestions and some specific information has been added. We find that el,h and el,v typically converge to 0.95 for high biomass vegetation cover, which is also consistent with previous studies (English, 2008; Peterson et al., 2000). Thus, the vegetation single scattering albedo (ω) is fix to 0.05. The Fresnel reflectivities of the soil assumed constant 0.05 in some research, but in order to improve the accuracy of VOD, the microwave emissivity of surface soil is obtained by the measurements from the Maqu regional scale soil moisture and soil temperature monitoring network by using the Fresnel reflection equation ( ).

English, S. J. The importance of accurate skin temperature in assimilating radiances from satellite sounding instruments. IEEE Transactions on Geoscience and Remote Sensing. 2008, 46(2), 403−408.

Peterson, T. C., Bassist, A. N., Williams, C. N., Grody, N. C. A blended satellite-in situ near-global surface temperature dataset. Bulletin of the American Meteorological Society. 2000, 81(9), 2157−2164.

 

 

L205: km2

Response: Sorry for this typo. The superscript is used in the revised version.

 

 

 

L288-293: km2 (please check the manuscripts for any other instances of this error)

Response: Sorry for this typo. The superscript is used in the revised version.

 

 

 

L310: References are needed.

Response: We fully agree with the reviewer’s opinion here. The references have been added into the revised manuscript.

 

 

 

L316: Why was the algorithm valid for non-precipitating conditions?

Response: This sentence is improper here, we try to express that the observations are processed under non-precipitating conditions when the satellite passed over the sites. This has been deleted in the revised version.

 

 

Move Figures 3 and 9 to the supplementary information

Response: The observed VWC during the experimental period (Figure 3) is our novel step different from previous method. In the previous researches, the parameter b () for different underlying vegetation surfaces is determined as constant and the dependence of the vegetation parameter b on various physical and sensor variables was not determined exactly. In our study, parameter b is obtained by the measured VWC from our field work and VOD from Jone’s method. So the Figure 3 is the key step to obtain the parameter b. Also, in Figure 9 (black line represents logarithm trendline), there is a good corresponding relationship between VWC and VOD in the source region during the summer season. According to the Figure 9, the parameter b is determined by the logarithm trendline . With this relationship the spatial VWC is obtained during the entire growth period (Figure 10). So figure 3 and 9 are important step in the manuscript and improper move to other sections.

 

 

Please describe the status of RFI in this study area

Response: The status of RFI in this study area has been added into the revised manuscript. VOD retrieved based on τ-ω model is widely used for interpreting L-band measurements, and VOD retrievals at lower (e.g. 6.9 and 10.7 GHz) microwave frequencies increase potential VOD sensitivity at higher biomass levels. however, AMSR-E Tb retrievals at these frequencies are reported to have significant RFI contamination and are more sensitive to surface soil moisture at lower VOD levels (Njoku et al., 2005), while higher frequencies are increasingly sensitive to atmosphere effects. Water vapor and cloud liquid water in various atmospheric layers also absorb and emit scattered radiation for higher frequencies ＞18.7 GHz. So, the 18.7 GHz frequency that we used in our study has relatively low RFI and reduced atmosphere and soil moisture sensitivity relative to other channels (Njoku and Li, 1999).

 

Njoku, E. G., Li, L. Retrieval of land surface parameters using passive microwave measurements at 6–18 GHz. IEEE Transactions on Geoscience and Remote Sensing. 1999. 37(1), 79.93.

Njoku, E.G., Ashcroft, P., Chan, T.K., Li, L. Global survey and statistics of radiofrequency interference in AMSR-E land observations. IEEE Transactions on Geoscience and Remote Sensing. 2005. 43, 938.947.

 

 

L344-349: This is hard to follow or to find the specific reason regarding the results. What is the specific reason for having no VOD in JAS months?

Response: More detailed discussion was added the revised version. Unlike VWC or soil moisture, VOD can’t be observed or verified directly by field work. Figure 5 and Figure 6 present the spatial distribution of VOD. Compare with these two figures, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 6(b), (c) and (d). This demonstrates that during the period from July to September, the results from previous method are poor in some part of the region, while the algorithm with considering water bodies gives better results. This also means the VOD method without considering the influence of water bodies was invalid and failed in some area at least in our study.  Moreover, the temporal variation of retrieved VOD in figure 7 (b) also shows the method without considering the influence of water bodies was failed. The VODr (gray trend line) represents the variation trends of the algorithm without the influence of water bodies and this gray trend line missed most of the time in figure 7 (b). This also demonstrates VOD method considering the water bodies was superior.

 

 

L355: Relative to what?

Response: Sorry for the incorrect word here, we want to express that VOD variation was obvious. This sentence has been rewritten.

 

L357: Why are only these regions sensitive to T and SM?

Response: Sorry for this confusion. According to the DEM in figure 1, the east of the source region where Maqu located has the lowest elevation in the study area, with higher temperature and moisture. The VOD in this part has the highest values, and this indicates that the vegetation in this area was sensitive to surface temperature and soil moisture.

 

 

L364-370: It seems that not all the discussion is supported by the results. The authors should be more specific or else conduct further research to support all the assertions made here.

Response: We fully agree with the reviewer’s opinion here. The words we used in old version are improper. This section has been rewritten.

 

4.2.2: Why were these two sites selected?

Response: In order to improve measurement accuracy, convenience of transportation to the sites must be considered to ensure that the collected vegetation can be immediately sent back to the base station to be weighed and dried to avoid any loss of VWC. Both the sites have our laboratory for measuring fresh weight. The fresh vegetation bag labeled with the vegetation type, sampling time must be immediately returned to the laboratory. Maqu area representing the southern part and the other is the Maduo area representing the northwest part. These have been added into the manuscript.

 

As with Figure 10, I recommend that the authors make difference maps such as VWC from the previous method – VWC from the current method.

Response: VWC is obtained by equation (3)   . Figure 5 and Figure 6 present the spatial distribution of VOD(). Compare with these two figures, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 6(b), (c) and (d). This demonstrates that during the July to September, the VOD from previous method was invalid in some area at least in our study. Moreover, the temporal variation of retrieved VOD in figure 7 (b) also show failed from previous method. The VODr (gray trend line) represents the variation trends of the algorithm from previous method and this gray trend line missed most of the time in figure 7 (b). In this case, VOD from previous method was failed in our study area, so VWC can’t obtain from previous VOD method. Thus, difference maps can’t present in manuscript due to the old method is invalid in some part of our study area.

 

 

Throughout the entire manuscript, the authors have misused the term “relative”. Please delete incorrect uses of “relative”.

Response: We fully agree with the reviewer’s opinion here. These sentences have been rewritten in the revised manuscript.

Author Response File: Author Response.docx

Reviewer 4 Report

Manuscript: ‘Derivation of Vegetation Optical Depth and Water Content in the Source Region of the Yellow River’

In this work have been investigated the brightness temperature at the given microwave frequency (18.7 GHz) from the Microwave Radiation Imager (MWRI) on board satellite Fengyun-3B (FY-3B) to improve the τ-ω model by considering the radiative contribution from water body in the pixels over the wetland of the Yellow River source region, China. Unlike other similar works: i) a dual polarization slope parameter was defined to express the surface emissivity in the τ-ω model as the sum of soil emissivity and water body emissivity; ii) with use of the field observation data of the vegetation water content (VWC) in the source region of the Yellow River during the summer of 2012, a regression relationship between vegetation optical depth (VOD) and VWC was established and then the vegetation parameter b is estimated (based on Jackson and O'Neill, 1990); and iii) the relationship derived from (2) was employed to derive the spatial VWC during the entire vegetation growing period. The study addresses to a quite interesting topic, in term of operational applications focused on retrieve the VWC in the source region of the Yellow River; therefore I think that the results of this study could contribute to obtain more accurate information on seasonal vegetation growth in the source region of the Yellow River. Furthermore, this is a relevant topic lies within the scope of the MDPI remote sensing journal. The article is well organized and neatly written with the appropriate scientific content. I did not detect plagiarism. I have listed my comments that need to be considered by authors for the next round.

********************************

Title: it fits perfectly the paper content. 

Abstract: it is quite adjusted to the paper content.

Line 26: it is suggested to use superscript for the VWC; e.g., 0.20kg/m² to 1.40kg/m²

Line 30: it is suggested to use superscript for the root-mean-square error; e.g., 0.12kg/m²

Introduction: this section provides sufficient background and includes relevant references about the main features of flash droughts and some of the approaches that have been used. Objectives and the novelty of the study are also clearly stated.

Liner 89: delete comma ‘Consequently,, the results showed…’

Study area and data source: the description of the study area and datasets are clearly stated. In my opinion, the research shows a design appropriated and its methods have been adequately described, but for the sake of clarity, I think that this section could be significantly improved if the authors add a flow chart with the different methods described in text, highlighting inputs, applied analysis/procedure, and outputs so that readers could understand this section easier.  

Line 205: it is suggested to use superscript for the area, e.g., 122,000 km².

Line 220: add ‘from’. ‘The FY-3B MWRI is derived from MVI by…’

Lines 220-232: this part should be moved at ‘Results’.

Figure 1: add a scale bar.

Figure 2a: change “Grass” and “Shrub” class colors. It is difficult to distinguish between “Grass” and “Shrub”

Figure 3: It is not clear to me if the x axis shows Month-Day. Please verify (e.g., Date [Month-Day]).

Line 309-310: ‘Few studies have been conducted about the vegetation detection by using the FY-3B microwave data sets.’ Here, add at least a study as a reference.

Results: the results have been presented clearly, but I have some specific comments:

Line 325: figures should be placed in the main text near to the first time they are cited. Where was cited Figure 4? 

Figure 8: it is not clear in the figure.  It is suggested to write: ‘Correlations between retrieved VOD against (a) NDVI or (b) MVI during the experimental period’

Figure 9: what does the dash line mean? What does the black line mean? Please, clarify in the caption.

Line 456 and 462: it is suggested to use superscript for the VWC; e.g., 0.20-1.20kg/m²

Line 474 and 475: it is suggested to use superscript for the root-mean-square error; e.g., 0.12kg/m²

Figure 11: what does the red line mean? Please, clarify in the caption.

Discussion and Conclusions

Line 498 and 500: it is suggested to use superscript for the VWC; e.g., 0.20-1.40kg/m²

It is strongly advised to compare the results quantitatively with results from works over similar regions.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 4 Comments

In this work have been investigated the brightness temperature at the given microwave frequency (18.7 GHz) from the Microwave Radiation Imager (MWRI) on board satellite Fengyun-3B (FY-3B) to improve the τ-ω model by considering the radiative contribution from water body in the pixels over the wetland of the Yellow River source region, China. Unlike other similar works: i) a dual polarization slope parameter was defined to express the surface emissivity in the τ-ω model as the sum of soil emissivity and water body emissivity; ii) with use of the field observation data of the vegetation water content (VWC) in the source region of the Yellow River during the summer of 2012, a regression relationship between vegetation optical depth (VOD) and VWC was established and then the vegetation parameter b is estimated (based on Jackson and O'Neill, 1990); and iii) the relationship derived from (2) was employed to derive the spatial VWC during the entire vegetation growing period. The study addresses to a quite interesting topic, in term of operational applications focused on retrieve the VWC in the source region of the Yellow River; therefore I think that the results of this study could contribute to obtain more accurate information on seasonal vegetation growth in the source region of the Yellow River. Furthermore, this is a relevant topic lies within the scope of the MDPI remote sensing journal. The article is well organized and neatly written with the appropriate scientific content. I did not detect plagiarism. I have listed my comments that need to be considered by authors for the next round.

 

Response: We really appreciate the positive comments for our work and have considered your suggestions into our revised manuscript.

********************************

 

Title: it fits perfectly the paper content.

Response: We really appreciate the positive comments for our manuscript.

 

Abstract: it is quite adjusted to the paper content.

Response: We really appreciate the positive comments for our manuscript.

 

Line 26: it is suggested to use superscript for the VWC; e.g., 0.20kg/m2 to 1.40kg/m2

Response: We agree with the reviewer’s suggestions and these units have been rewritten.

 

Line 30: it is suggested to use superscript for the root-mean-square error; e.g., 0.12kg/m2

Response: We agree with the reviewer’s suggestions and these units have been rewritten.

 

Introduction: this section provides sufficient background and includes relevant references about the main features of flash droughts and some of the approaches that have been used. Objectives and the novelty of the study are also clearly stated.

Response: We really appreciate the positive comments for our manuscript.

 

Liner 89: delete comma ‘Consequently,, the results showed…’

Response: Sorry for this typo. This sentence has been rewritten.

 

Study area and data source: the description of the study area and datasets are clearly stated. In my opinion, the research shows a design appropriated and its methods have been adequately described, but for the sake of clarity, I think that this section could be significantly improved if the authors add a flow chart with the different methods described in text, highlighting inputs, applied analysis/procedure, and outputs so that readers could understand this section easier. 

Response: Thank you for pointing out. We have redesigned the ‘Study area and data source’ section including the figures, and content. Most VOD retrieval is conducted simultaneously with soil moisture retrieval. Dual-polarization, multi-frequency data from passive-microwave radiometers provide an estimate of VWC during retrievals of soil moisture content. The retrieval algorithm based on the τ-ω model uses the horizontal and vertical polarizations for channels at 10, 18.7 and 37 GHz to solve simultaneously for soil moisture content, VWC, surface temperature, and surface roughness (Li et al., 2010). Under conditions of seasonal wetland coverage in our study area, the brightness temperatures from the microwave radiometer measurements contain the signal of water bodies, except for those of exposed soil and vegetation. So, we presumed that mixed pixels are composed of water bodies and vegetation, with the vegetation component containing both vegetation canopy and soil covered by vegetation, and a dual polarization slope parameter is defined. By combining the τ-ω model and the dual-polarized slope parameter, the effective optical depth of land fraction is determined by considering the influence of water bodies. In brief, the dual polarization slope parameter is the most different between our methods and others. Sorry for this confusion.

 

Li, L., Gaiser, P. W., Gao, B. -C., Bevilacqua, R. M., Jackson, T. J., Njoku, E. G., et al. (2010). WindSat global soil moisture retrieval and validation. IEEE Transactions on Geoscience and Remote Sensing, 48, 2224–2241.

 

 

 

Line 205: it is suggested to use superscript for the area, e.g., 122,000 km2.

Response: We agree with the reviewer’s suggestions and these units have been rewritten.

 

 

 

Line 220: add ‘from’. ‘The FY-3B MWRI is derived from MVI by…’

Response: Thanks for the reviewer to point this. The correction has been made in the revised version.

 

Lines 220-232: this part should be moved at ‘Results’.

Response: We agree with the reviewer’s suggestions and this section has been re-designed.

 

 

Figure 1: add a scale bar.

Response: Thanks for pointing out. Figure 1(right) is cited from Zheng et al’s(2017) literature and focus on the SMST(soil moisture and soil temperature) network samples  distributed  over  an  area  of  40  km×80  km. We re-design the Figure 1 and added the location of the catchment in the revised manuscript to clearly explain the catchment and the soil moisture monitoring network located.

Zheng, D.H.; Wang, X.; van der Velde, R.; Zeng, Y.; Wen, J.; Wang, Z.; Schwank, M.; Ferrazzoli, P.; Su, Z. L-Band Microwave Emission of Soil Freeze-Thaw Process in the Third Pole Environment. IEEE Trans. Geosci. Remote Sens.2017, 99, 1-15. DOI:10.1109/TGRS.2017.2705248.

 

Figure 2a: change “Grass” and “Shrub” class colors. It is difficult to distinguish between “Grass” and “Shrub”

Response: We reworked the land cover map in the revised manuscript.

 

 

Figure 3: It is not clear to me if the x axis shows Month-Day. Please verify (e.g., Date [Month-Day]).

Response: Thanks for the reviewer to point this. The date form on x axis has been changed in the revised version.

 

Line 309-310: ‘Few studies have been conducted about the vegetation detection by using the FY-3B microwave data sets.’ Here, add at least a study as a reference.

Response: We fully agree with the reviewer’s opinion here. The references have been added into the revised manuscript.

 

 

Results: the results have been presented clearly, but I have some specific comments:

Line 325: figures should be placed in the main text near to the first time they are cited. Where was cited Figure 4?

Response: Sorry for this mistake. The missing paragraph has been added into the revised manuscript.

 

 

 

Figure 8: it is not clear in the figure.  It is suggested to write: ‘Correlations between retrieved VOD against (a) NDVI or (b) MVI during the experimental period’

Response: Thank you for point out. The title has been edited according to the reviewer’s opinion.

 

Figure 9: what does the dash line mean? What does the black line mean? Please, clarify in the caption.

Response: The graph has been reworked and the information has been added in the revised version.

 

 

Line 456 and 462: it is suggested to use superscript for the VWC; e.g., 0.20-1.20kg/m2

Response: Sorry for this typo. The superscript is used in the revised version.

 

 

Line 474 and 475: it is suggested to use superscript for the root-mean-square error; e.g., 0.12kg/m2

Response: Sorry for this typo. The superscript is used in the revised version.

 

 

Figure 11: what does the red line mean? Please, clarify in the caption.

Response: The red line represents reference line of zero bias. The explanation has been added into the revised version.

 

 

Discussion and Conclusions

Line 498 and 500: it is suggested to use superscript for the VWC; e.g., 0.20-1.40kg/m2

Response: Sorry for this typo. The superscript is used in the revised version.

 

 

It is strongly advised to compare the results quantitatively with results from works over similar regions.

Response: We fully agree with the reviewer’s opinion here. Following your suggestion, the comparison between other similar regions has been added into the section 4. In the revised version, "Results and discussion" and "Summary and conclusions" are the headline of section 4 and 5, respectively.

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Most of my comments have been answered by the authors, however, there were two points which were not (feels like the responses were meant to answer other questions, as they do not address what I asked):

Sections 4.2.3: correlation values are a bit difficult to visualise in Figure 8, so perhaps
adding a few metrics of these coefficients will improve understanding (e.g. mean values for
each soil cover). Here, I presume you are using VOD considering water bodies... how do the
"default" VOD performs in these correlations (NDVI, MVI and VWC)?

Response: As last explanation, difference maps can’t be presented in manuscript due to the
previous method is invalid in some part of our study area.

Obs: This answer refers to Figure 8, in the revised manuscript, however my question was meant for Figure 8 of the previous version (Figure 9 in the revised version)

Figure 9: is there a spatial/pixel pattern in this correlation?

Response: Maqu regional scale soil moisture and soil temperature monitoring network is set up
 in the water source region of the Yellow River to validate the accuracy of the satellite-
 derived soil moisture products for the Tibetan Plateau. This network has the capability to
 monitor the spatial and temporal soil moisture variability of the area with a high degree
 of accuracy (Dente et al., 2012; Zheng et al., 2017). It consists of 20 stations distributed.
The VWC sampling is collected according to the 20 site. Dente (Dente et al., 2012) proposed
the weighted spatial average of measured soil moisture was successfully used as ground
reference for the validation of the AMSR-E soil moisture products and ASCAT soil wetness
index products. The results of Dente’s study are very encouraging for our applications. So,
the match problem can be ignore in our opinion. These information has been added into the
Validation section.

Obs: question not answered.

The are a few English errors that need to be corrected.

Lastly, most figures from Figure 3 onwards need an aesthetical improvement in order to be fit for publishing. For example, in Figure 2 legend is not symmetrical, and in Figure 3 the axis labels are poor: x: Date, y: (kg/m²).

Author Response

Most of my comments have been answered by the authors, however, there were two points which were not (feels like the responses were meant to answer other questions, as they do not address what I asked):

 

Sections 4.2.3: correlation values are a bit difficult to visualise in Figure 8, so perhaps adding a few metrics of these coefficients will improve understanding (e.g. mean values for each soil cover). Here, I presume you are using VOD considering water bodies... how do the "default" VOD performs in these correlations (NDVI, MVI and VWC)?

 

Response: As last explanation, difference maps can’t be presented in manuscript due to the previous method is invalid in some part of our study area.

 

Obs: This answer refers to Figure 8, in the revised manuscript, however my question was meant for Figure 8 of the previous version (Figure 9 in the revised version)

 

Response: We apologize for the confusion. Some information about the figure 9(new version) has been added into the manuscript.

 

 

 

Figure 9: is there a spatial/pixel pattern in this correlation?

 

Response: Maqu regional scale soil moisture and soil temperature monitoring network is set up  in the water source region of the Yellow River to validate the accuracy of the satellite- derived soil moisture products for the Tibetan Plateau. This network has the capability to  monitor the spatial and temporal soil moisture variability of the area with a high degree  of accuracy (Dente et al., 2012; Zheng et al., 2017). It consists of 20 stations distributed. The VWC sampling is collected according to the 20 site. Dente (Dente et al., 2012) proposed the weighted spatial average of measured soil moisture was successfully used as ground reference for the validation of the AMSR-E soil moisture products and ASCAT soil wetness index products. The results of Dente’s study are very encouraging for our applications. So, the match problem can be ignore in our opinion. These information has been added into the Validation section.

 

 

 

Obs: question not answered.

 

Response: We apologize for the confusion. There is a spatial/pixel pattern in this correlation. According to the Figure 10, the parameter b is determined by the logarithm trend line. With this relationship the spatial VWC is obtained during the entire growth period (Figure 11). This has been added into the manuscript.

 

 

 

There are a few English errors that need to be corrected.

Response: The previous manuscript has been carefully revised to minimize typographical, grammatical, and bibliographical errors.

 

Lastly, most figures from Figure 3 onwards need an aesthetical improvement in order to be fit for publishing. For example, in Figure 2 legend is not symmetrical, and in Figure 3 the axis labels are poor: x: Date, y: (kg/m²).

Response: The figures 2 and 3 have been reworked in the new version.

Author Response File: Author Response.docx

Reviewer 3 Report

It is very difficult to review the revised version of the manuscript. The authors have written a lot of notes in response to my concerns; however, it is very hard to find out how they solved my concerns within the actual revised manuscript. The authors should find ways to demonstrate their work more efficiently. For example, specific lines in the updated version of the manuscript should be included in the Response to Reviewer file. Author responses such as “Response: Thanks for the reviewer to point this. This paragraph has been rewritten according to one reviewer’s opinion.

” are certainly not a good way to demonstrate their changes; rather, they put much more responsibility on the reviewer.

 

For this reason and more, it is very hard to review the revised version of the manuscript properly. I strongly recommend updating and resubmitting the Response to Reviewer file again.

Author Response

Response to Reviewer 3 Comments

 

It is very difficult to review the revised version of the manuscript. The authors have written a lot of notes in response to my concerns; however, it is very hard to find out how they solved my concerns within the actual revised manuscript. The authors should find ways to demonstrate their work more efficiently. For example, specific lines in the updated version of the manuscript should be included in the Response to Reviewer file. Author responses such as “Response: Thanks for the reviewer to point this. This paragraph has been rewritten according to one reviewer’s opinion.” are certainly not a good way to demonstrate their changes; rather, they put much more responsibility on the reviewer.

 

 

 

For this reason and more, it is very hard to review the revised version of the manuscript properly. I strongly recommend updating and resubmitting the Response to Reviewer file again.

 

 

Response: We apologize for the confusion. Here we rewrite the responses and highlight the alteration.

 

 

In this study, the authors used the FY-3B satellite’s MWRI sensor to improve the tau-omega model with regard to the radiative contribution from water bodies to the pixels over the wetland. I enjoyed this paper. In addition, I found some issues that should be explained or improved before publication. The followings are my major comments:

 

1) Figures must be appropriately edited. The current figures are very low quality and some are unreadable, 2) Explain which correlation coefficient equation is used. If the authors used the Pearson-correlation coefficient, the entire manuscript and results should be re-written because the NDVI, MVW, and VOD values do not share the same variance and data ranges; so the ranked Pearson correlation coefficient would be more appropriate, 3) because the authors used field observation-based VWC, a possible  mismatch in spatial resolution between the field data and satellite-based data should be investigated. What are the possible errors caused by spatial mismatch? 4) Without considering the fractional seasonal wetlands in the source regions, what are the biases and RMSE? How much improvement could be made by considering the fractional seasonal wetlands? 5) Since VOD and VWC do not use the same units of measurement, comparing them directly does not make sense. As an example, the authors have written, “The results indicate that the VOD is in the range of 0.20 to 1.20 and the VWC is in the range of 0.20kg/m2 to 1.40kg/m2 in the entire source region of the Yellow River in 2012” – this does not have any meaning. It would be better to provide the anomaly of these values. 6) The authors should include previous research to support their findings. For example, why are the RMSE values of VWC 0.12 km/m2 considered to be successful retrievals of VWC? Which previous research are they using for comparison? 7) The method section should be rewritten. As currently written, the methods proposed here are impossible for other researchers to recreate. And 8) The entire abstract should be rewritten. The current abstract omits too much information. For instance, what is the major finding of the current research? What is the goal of the current research?

 

 

Response: We thank the reviewer for the constructive comments and suggestions as well as the careful reading. As a result, we believe that the quality of the revised version has been improved significantly. In the paragraphs that follow, we include point-to-point responses.

(1)We re-design the Figures to be better present in the revised manuscript.

(2)The ranked Spearman correlation coefficient is used in our study and some information has been added in the revised manuscript(L461).

(3)Maqu regional scale soil moisture and soil temperature monitoring network is set up in the water source region of the Yellow River to validate the accuracy of the satellite-derived soil moisture products for the Tibetan Plateau. This network has the capability to monitor the spatial and temporal soil moisture variability of the area with a high degree of accuracy (Dente et al., 2012; Zheng et al., 2017). It consists of 20 stations distributed. The VWC sampling is collected according to the 20 site.  Dente (Dente et al., 2012) proposed the weighted spatial average of measured soil moisture was successfully used as ground reference for the validation of the AMSR-E soil moisture products and ASCAT soil wetness index products. The results of Dente’s study are very encouraging for our applications. So, the match problem can be ignore in our opinion. These information has been added into the Validation section(L529-537).

(4) Unlike VWC or soil moisture, VOD can’t observe or verify directly by field work. Compare with the figure 5 and figure 6, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 6(b), (c) and (d). This demonstrates that the results from previous method are poor in some part of the region. Moreover, the temporal variation of retrieved VOD in figure 7 (b) also show the method without considering the influence of water bodies was failed. The VODr (gray trend line) represents the variation trends of the algorithm without the influence of water bodies and this gray trend line missed most of the time in figure 7 (b). Due to the previous method was failed and invalid, it's unnecessary to present the biases and RMSE.

(5) The anomaly of the VOD or VWC values are important for indicate the abnormal variability. There are only 5 months VOD and VWC data, and these can’t capture the anomaly information. Further research would be conduct to provide the long-time series variability.

(6) Following your suggestion, the comparison between other similar regions has been added into the section 4.3 (L523).

(7) The method section has been modified in the revised manuscript.

(8) The abstract has been rewritten.

 

Dente, L.; Vekerdy, Z.; Wen, J.; Su, Z. Maqu network for validation of satellite-derived soil moisture products. Int. J. Appl. Earth Obs. 2012. 17, 55-65. DOI: org/10.1016/j.jag.2011.11.004

Zheng, D.H.; Wang, X.; van der Velde, R.; Zeng, Y.; Wen, J.; Wang, Z.; Schwank, M.; Ferrazzoli, P.; Su, Z. L-Band Microwave Emission of Soil Freeze-Thaw Process in the Third Pole Environment. IEEE Trans. Geosci. Remote Sens.2017, 99, 1-15. DOI:10.1109/TGRS.2017.2705248

 

 

Specific comments:

 

L26, 30:  m2 -> m2

Response: Sorry for this typo. The superscript is used in the revised version(L 29, 33).

 

What is the exact meaning of “source region of” ? Why not just say “the Yellow River?”

Response: The source region of the Yellow River is located in the north-eastern part of the Tibetan Plateau of 95°30'-103°30'E and 31°30'-36°30'N, while the Yellow River located in  90°-120°E and 30°-40°N approximately. Figure 1 shows the location of the source region of the Yellow River catchment, red line in upper right corner in figure 1 also shows the Yellow River located in China. Sorry for this confusion. Figure 1 has been redesigned into the revised manuscript(L 225).

                                             

Figure 1. Location of the source region of the Yellow River catchment

 

The method sections are unnecessarily long. Much of this information has already been published in previous research. I recommend that the authors reduce the methodology section and use proper citations instead.

Response: This section has been edited and condensed for clarity. Some unnecessary process (Line 135-146) about the method has been deleted.

 

 

L115: references should be added

Response: The reference has been added into the revised manuscript (L113).

 

 

 

L122: ГP

Response: Sorry for this typo. The missing p has been superscripted and added in the revised version (L119).

 

L124: Please provide detailed information regarding how to obtain the ep_s and Fresnel coefficient.

Response: Sorry for this confusion. We look back the manuscript and rewrite this sentence (L121-125).

 

 

L127: Why was the vegetation single scattering albedo fixed to 0.05?

Response: Here, 0.05 is cited by Jones’(2011) work and we find that el,h and el,v typically converge to 0.95 for high biomass vegetation cover, which is also consistent with previous studies (English, 2008; Peterson et al., 2000); therefore the vegetation single scattering albedo is set to 0.05. Yang et al (2007) also analysed single scattering albedo, size factor X and complex refractive index m based on the Mie theory, and find that the single scattering albedo appears gradually large while X ascends, but finally equals to 0.5 approximately. So, 0.05 is proper in our study in our opinion.

 

English, S. J. The importance of accurate skin temperature in assimilating radiances from satellite sounding instruments. IEEE Transactions on Geoscience and Remote Sensing. 2008, 46(2), 403−408.

Peterson, T. C., Bassist, A. N., Williams, C. N., Grody, N. C. A blended satellite-in situ near-global surface temperature dataset. Bulletin of the American Meteorological Society. 2000, 81(9), 2157−2164.

Jones, M.O.; Jones, L.A.; Kimball, J.S.; Mcdonald, K.C. Satellite passive microwave remote sensing for monitoring global land surface phenology. Remote Sens. Environ. 2011, 115, 1102-1114. https://doi.org/10.1016/j.rse.2010.12.015

Yang C.P., Wu J., Wan M. Calculation of single scattering albedo in the radiative transfer equation. Laser Journal, 28(4), 45-, 2007.

 

L133: How did you obtain the value of optical thickness of vegetation?

Response: L134-135: ‘Equations (1) and (2) are usually called the τ-ω model. Given that the other parameters in equations (1) and (2) are known, one can derive VOD’. In previous method, VOD can be obtained directly by equations (1) and (2), meanwhile the fraction of open water is considered in the τ-ω model in our method.

 

 

L139: Shielded?

Response: This sentence has been deleted according to one reviewer’s opinion.

 

 

L155-159: Please be more specific since this is the main assumption and the novel portion of this study.

Response: Thanks for the reviewer to point this. This paragraph has been rewritten according to reviewer’s opinion (L 152-159).

 

 

L172: What does “typical” indicate here?

Response: We use this value according to the previous studies for high biomass vegetation cover. This word has been deleted in the revised version.

 

 

How was equation 5 obtained?

Response: Eq. (5) is simplified and linearized from its usual quadratic form by ignoring atmospheric emissions reflected by the surface (Jones et al., 2010). This reference has been deleted in the revised version (L179).

 

Jones, L.A.; Ferguson, C.R.; Kimball, J.S.; Zhang, K.; Chan, S.T.K.; McDonald, K.C.; Njoku, E.G.; Wood, E.F. Satellite microwave remote sensing of daily land surface air temperature minima and maxima from AMSR-E. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2010, 3, 111–123. 10.1109/JSTARS.2010.2041530

 

 

L181-198: References are needed.

Response: We fully agree with the reviewer’s opinion here. The references have been added into the revised manuscript (L185-202).

 

L194-196: Specific information is required. The method shown in this section should be replicable by other researchers. How are the emissivity, albedo, and rs, h(v) obtained?

Response: We agree with the reviewer’s suggestions and some specific information has been added (L198-202). We find that el,h and el,v typically converge to 0.95 for high biomass vegetation cover, which is also consistent with previous studies (English, 2008; Peterson et al., 2000). Thus, the vegetation single scattering albedo (ω) is fix to 0.05. The Fresnel reflectivities of the soil assumed constant 0.05 in some research, but in order to improve the accuracy of VOD, the microwave emissivity of surface soil is obtained by the measurements from the Maqu regional scale soil moisture and soil temperature monitoring network by using the Fresnel reflection equation ( ).

English, S. J. The importance of accurate skin temperature in assimilating radiances from satellite sounding instruments. IEEE Transactions on Geoscience and Remote Sensing. 2008, 46(2), 403−408.

Peterson, T. C., Bassist, A. N., Williams, C. N., Grody, N. C. A blended satellite-in situ near-global surface temperature dataset. Bulletin of the American Meteorological Society. 2000, 81(9), 2157−2164.

 

 

L205: km2

Response: Sorry for this typo. The superscript is used in the revised version(L206).

 

 

 

L288-293: km2 (please check the manuscripts for any other instances of this error)

Response: Sorry for this typo. The superscript is used in the revised version(L278-280).

 

 

 

L310: References are needed.

Response: We fully agree with the reviewer’s opinion here. The references have been added into the revised manuscript(L298).

 

 

 

L316: Why was the algorithm valid for non-precipitating conditions?

Response: This sentence is improper here, we try to express that the observations are processed under non-precipitating conditions when the satellite passed over the sites. This has been deleted in the revised version.

 

 

Move Figures 3 and 9 to the supplementary information

Response: The observed VWC during the experimental period (Figure 3) is our novel step different from previous method. In the previous researches, the parameter b () for different underlying vegetation surfaces is determined as constant and the dependence of the vegetation parameter b on various physical and sensor variables was not determined exactly. In our study, parameter b is obtained by the measured VWC from our field work and VOD from Jone’s method. So the Figure 3 is the key step to obtain the parameter b. Also, in Figure 9 (black line represents logarithm trendline), there is a good corresponding relationship between VWC and VOD in the source region during the summer season. According to the Figure 9, the parameter b is determined by the logarithm trendline . With this relationship the spatial VWC is obtained during the entire growth period (Figure 10). So figure 3 and 9 are important step in the manuscript and improper move to other sections.

 

 

Please describe the status of RFI in this study area

Response: L 309-311:’The channel at 18.7 GHz frequency used in this study has low RFI and reduced atmosphere and soil moisture sensitivity relative to other channels [6].’ VOD retrieved based on τ-ω model is widely used for interpreting L-band measurements, and VOD retrievals at lower (e.g. 6.9 and 10.7 GHz) microwave frequencies increase potential VOD sensitivity at higher biomass levels. however, AMSR-E Tb retrievals at these frequencies are reported to have significant RFI contamination and are more sensitive to surface soil moisture at lower VOD levels (Njoku et al., 2005), while higher frequencies are increasingly sensitive to atmosphere effects. Water vapor and cloud liquid water in various atmospheric layers also absorb and emit scattered radiation for higher frequencies ＞18.7 GHz. So, the 18.7 GHz frequency that we used in our study has relatively low RFI and reduced atmosphere and soil moisture sensitivity relative to other channels (Njoku and Li, 1999).

 

Njoku, E. G., Li, L. Retrieval of land surface parameters using passive microwave measurements at 6–18 GHz. IEEE Transactions on Geoscience and Remote Sensing. 1999. 37(1), 79.93.

Njoku, E.G., Ashcroft, P., Chan, T.K., Li, L. Global survey and statistics of radiofrequency interference in AMSR-E land observations. IEEE Transactions on Geoscience and Remote Sensing. 2005. 43, 938.947.

 

 

L344-349: This is hard to follow or to find the specific reason regarding the results. What is the specific reason for having no VOD in JAS months?

Response: More detailed discussion was added the revised version(L372-377). Unlike VWC or soil moisture, VOD can’t be observed or verified directly by field work. Figure 6 and Figure 7 present the spatial distribution of VOD. Compare with these two figures, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 7(b), (c) and (d). This demonstrates that during the period from July to September, the results from previous method are poor in some part of the region, while the algorithm with considering water bodies gives better results. This also means the VOD method without considering the influence of water bodies was invalid and failed in some area at least in our study.  Moreover, the temporal variation of retrieved VOD in figure 8 (b) also shows the method without considering the influence of water bodies was failed. The VODr (gray trend line) represents the variation trends of the algorithm without the influence of water bodies and this gray trend line missed most of the time in figure 8 (b). This also demonstrates VOD method considering the water bodies was superior.

 

 

L355: Relative to what?

Response: Sorry for the incorrect word here, we want to express that VOD variation was obvious. This sentence has been rewritten (L389: In particular, in the region from the west of Maqing to the east of Ngoring Lake, VOD variation was obvious, with strong monthly variation.)

 

L357: Why are only these regions sensitive to T and SM?

Response: Sorry for this confusion. According to the DEM in figure 1, the east of the source region where Maqu located has the lowest elevation in the study area, with higher temperature and moisture. The VOD in this part has the highest values, and this indicates that the vegetation in this area was sensitive to surface temperature and soil moisture.

 

 

L364-370: It seems that not all the discussion is supported by the results. The authors should be more specific or else conduct further research to support all the assertions made here.

Response: We fully agree with the reviewer’s opinion here. The words we used in old version are improper. This section has been rewritten(L399-406: Because the surface cover in the source region of the Yellow River is dominated by alpine meadow and the coverage type is simple, the spatiotemporal distribution type of the VOD that includes the influence of water bodies may be characterizes the seasonal timing of vegetation growing seasons, canopy growth and senescence. The timing, rate and duration of these events are vital on vegetation photosynthesis, carbon sequestration and land–atmosphere water and energy exchange. Therefore, in addition to acting as a criterion for evaluating the optical depth in the derivation of soil moisture, VOD may be used as a characteristic variable of the vegetation to characterize comprehensive features of vegetation canopy.)

 

4.2.2: Why were these two sites selected?

Response: In order to improve measurement accuracy, convenience of transportation to the sites must be considered to ensure that the collected vegetation can be immediately sent back to the base station to be weighed and dried to avoid any loss of VWC. Both the sites have our laboratory for measuring fresh weight. Maqu area representing the southern part and the other is the Maduo area representing the northwest part. These have been added into the manuscript (L432-434: The fresh vegetation bag labeled with the vegetation type, sampling time must be immediately returned to the laboratory and both the sites have the laboratories for measuring and drying fresh vegetation).

 

As with Figure 10, I recommend that the authors make difference maps such as VWC from the previous method – VWC from the current method.

Response: VWC is obtained by equation (3)   . Figure 6 and Figure 7 present the spatial distribution of VOD(). Compare with these two figures, for some part of the area, VOD without considering the contribution of water bodies cannot obtain any results in figure 7(b), (c) and (d). This demonstrates that during the July to September, the VOD from previous method was invalid in some area at least in our study. Moreover, the temporal variation of retrieved VOD in figure 8 (b) also show failed from previous method. The VODr (gray trend line) represents the variation trends of the algorithm from previous method and this gray trend line missed most of the time in figure 8 (b). In this case, VOD from previous method was failed in our study area, so VWC can’t obtain from previous VOD method. Thus, difference maps can’t present in manuscript due to the old method is invalid in some part of our study area.

 

 

Throughout the entire manuscript, the authors have misused the term “relative”. Please delete incorrect uses of “relative”.

Response: We fully agree with the reviewer’s opinion here. These sentences have been rewritten in the revised manuscript.

 

 

Author Response File: Author Response.docx

Reviewer 4 Report

The authors have revised the manuscript based on most of my comments. Therefore, I support this revised version for publication in Remote Sensing.

Author Response

The authors have revised the manuscript based on most of my comments. Therefore, I support this revised version for publication in Remote Sensing.

 

 

Response: We really appreciate the reviewer for the constructive comments and the positive comments for our work.

Author Response File: Author Response.docx

Round 3

Reviewer 3 Report

I am satisfied with the reply of the authors and their proposed improvements. The manuscript reads much better now. 

One suggestion:

The introduction should be started with a vegetation-related topic not soil moisture since the main topic of the current manuscript "vegetation". I recommend the authors to move the first paragraph and make proper changes in the introduction section.

Cheers,

Author Response

Response: We really appreciate the reviewer for the constructive comments and the positive comments for our work. According to reviewer’s opinion this paragraph has been rewritten as below,

 

        Vegetation water content, vegetation canopy biomass and soil moisture strongly affect fluxes of the surface water vapor and energy, control the proportion of surface sensible heat and latent heat allocated to the usable surface energy as well as the proportion of rainfall among the surface runoff, infiltration, and evapotranspiration, and thus can have significant impacts on numerical weather, climate and hydrologic predictions [1,2]. Optical-infrared satellite remote sensing, such as the Normalized Difference Vegetation Index (NDVI) and Leaf-Area Index (LAI), is commonly used to determine vegetation water content. However, clouds, aerosols, and solar illumination effects degrade the ability to monitor vegetation information. Satellite microwave observations are insensitive to clouds at lower frequencies and independent of solar illumination [3]. Currently, soil moisture retrievals using microwave radiometers have a high accuracy over bare soil or sparse vegetation coverage area, whereas ongoing challenges remain in high vegetated regions, due to lack of knowledge about the properties of vegetation structure, vegetation optical depth (VOD), vegetation water content (VWC) and biomass of canopy [3-5]. To accurately retrieve soil moisture, passive soil moisture retrievals must properly account for the effect of VOD, so developing a feasible VOD algorithm is necessary for soil moisture retrievals.

 

 

 

 

 

Author Response File: Author Response.docx