Comparative Evaluation of Microwave L-Band VOD and Optical NDVI for Agriculture Drought Detection over Central Europe

Round 1

Reviewer 1 Report

The current approach aims to investigate  the combined or separate contribution of two vegetation indices, (MODIS NDVI and SMOS VOD) to detection of  agricultural droughts through  the statistically comparing the average values of their anomalies during dry and wet periods derived through employing a meteorological drought index (SPI-12).

Regarding the introduction section, the authors are suggested to reconstruct the introduction section in terms of presentation and quality information. One of the main roles of the introduction section is to address and stress the problem as stated through previous scientific researches on the field. It will also of high importance, to provide an enriched introduction section with more recent and relevant studies that have been previously conducted on the field, rather than referring generally to the high importance of plant disease detection Moreover, an introduction regarding the utility and background supported by the relative literature is absent. A short description of the presented scientific approach is not provided.

In the discussion section, the results are briefly provided. There is short to no further explanation concerning how the employed indices, MODIS normalized difference vegetation index (NDVI) and SMOS vegetation optical depth (VOD) contribute either separately or combined to the detection of droughts. The two employed indices behavior are poorly compared and contrasted. More specifically, there is only reference to the results of the current study. There is also no further discussion on the contribution of the presented approach and its general potential. The authors are suggested to indicate how the presented study differentiate from the previously published studies and compare the results of the current study with previous relevant approaches to highlight their contribution to the agricultural droughts detection.

The authors are suggested to cover the above-mentioned issues improving and enlightening important aspects of the work in order to provide a more detailed and complete presentation of their work that also meets the quality standards of the journal.

There are some recommendations that authors are suggested to take into consideration given as follows:

• Line 35 “interactions with different industries” To which industries are you referring to? Please reform the sentence or elaborate more.
• Line 53 “unhealthy or water stressed vegetation conditions.” Perhaps you mean either diseased or infected. There in no condition called unhealthy in the terminology of crop monitoring and protection.
• Lines 69-70 Reference missing.
• Lines 76-82 To which particular studies are you referring to. Please elaborate more.
• Line 93-95 The sentence is suggested to be removed since it states only a personal opinion and not stated by the relevant signature.
• Figures 2 and 3 Figures of higher resolution is needed. In its current condition is too difficult to be read.
• Lines 327-328 Please explain the reason of this observation.
• Line 334 This conclusion is suggested to be accompanied by a previously contacted research. Alternatively, you could possibly contrast the outcomes of your study to those outcomes of previous studies in the field.
• Lines 375-377 How could be possibly this avoided, for example in some other attempt in the near future?
• Line 404-408 Please elaborate more.

Author Response

You may please find our response letter in the attached pdf file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors, thanks for your paper. I would like ask you to discussed more detail influence of spatial resolution of SMOS data. Resolution is mentioned only in end of the paper in one sentence, but it seems, that this could have real influence on the results. What is influence of mixels (mixed pixels with different vegetation). In end you mentioned Sentinel 1 and 2, but these are working on completely different spatial resolution. How this information can be combined. What is in indexes influence of different vegetation. Please try to add more information in this direction.

Author Response

You may please find our response letter in the attached pdf file.

Author Response File: Author Response.pdf

Reviewer 3 Report

This study aimed at the drought detection by comparing satellite-based VOD and NDVI. The manuscript is generally fairly well organized, and the methods are decently conceptualized, with the moderate scientific contribution.

My largest issue with the current version of the manuscript is the lack of information regarding the applied data sources and software, which is crucial for the repeatability of these methods. A minor English language check is recommended. There are also some specific corrections that should be performed before this manuscript is ready for publication:

Figures 1-4: This figures should be in higher DPI, as some segments are not clearly visible.

2.1. Study area: I suggest inserting a table containing used data sources, with the description of auxiliary data and their measurement units. There are some specifications regarding that (lines 223-225) but are misplaced in the text.

 2.2.1. Section: The equations in this subsection require their respective references, most notably this refers to the NDVI. While it is a common vegetation index, it should still contain an adequate reference.

Lines 424-426: This sentence should be rephrased as you only tested one "actual" vegetation index.

Author Response

You may please find our response letter in the attached pdf file.

Author Response File: Author Response.pdf

Reviewer 4 Report

Ch.1. Introduction: -At the begin is very general, and the same phrases can be readable in lots of publication.

-Are citations suitable in every case? Instead of generally SPI citations there are some self-citations.

-The paragraph (from the line 76) needs some citations. You write here about reviews, but what reviews?

-line 86 and 101: write with comma;  24,442

Ch.2. Materials and methods: -Fig.1.: wrong in printed form, please use much better resolution!

-Ch.2.2.1.: Why did you use exactly the SPI 12 version?

Ch.3.: -Such a spatial comparison it is interresting the difference that is appeared between the area bordered by NDVI and VOD. eg in France, at Iberian penninsula, in Gross-Brittanien or in Turkey. These geographical descriptions are missing.

-You wrtite about the same time lag, but large areas show difference, like in Iberian penninsula

-Fig.2. Use much better resolution. Legend shows large differences (from 1 to 6 months), but such a color scale does not show properly it.

-Why do not make spatial comparison based on Fig.2.? Where are same VOD and NDVI values and where are not?

-From the line 259 the editing is wrong. Too much space at left part.

-Fig.3.: -please rewrite panel B to NDVI

-Fig.3. shows the pixels analyzed as drought during the study period (by VOD, and NDVI). That means there are some geographical area, where VOD shows more drought (France, GBr) or NDVi shows more drought (Turkey)

-Panel C is interresting, because there are large red area (eg. western Europe) and large blue area and just in some cases can be seen the area covered with some pixels.

-Fig.4.: -do not use double title at panel A-E. We see on the Y axis the subject of analysis (Temp, Precip., etc.)

-lines 309 and 313: (between 0.25 amd 0.75 quantiles). One time is enough.

-Where are the values mentioned  in the lines 330, 331, 334 on the Fig.4.

-line 346. You mentioned 'eg. CfB' here, but only this column can be mentioned here. There is not anything else.

line 353. you mention bare land and sparse vegetation, but NDVI is also egal with VOD at forests and cropland.

References: -you use 12 self citations of 56 references. It seems too much, but it is true some articles has been written by lots of authors.  

Author Response

You may please find our response letter in the attached pdf file.

Author Response File: Author Response.pdf

Reviewer 5 Report

The authors address in this manuscript the question of whether the vegetation indices NDVI (normalized difference vegetation index) and VOD (vegetation optiocal depth) perform efficiently during drought events. For this purpose, they observe 24442 pixels (0.25 degree spatial resolution) over Central Europe and compare a wet and a drought period. They assess drought as SPI (standardized precipitation index). The analysis takes in account climate and land conditions for each pixel. This approach seems correct to me.

The authors found that NDVI is more effective in capturing agricultural droughts over regions with lower monthly average NDVI values and VOD can significantly contribute in drought monitoring over dense vegetation covers.

The introduction gives a good overview of the topic of the mansucript and is well written. It's length is appropriate.

The methodology is well written and explained.

- Eq. 3. Could they authods write the exact wavelengths used for NDVI?
See: Remote Sens. 2020, 12, 2104; doi:10.3390/rs12132104

The results section reads well. The figures are appropriate, and the discussion is supported by the data.

Author Response

You may please find our response letter in the attached pdf file.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Dear authors, thanks for clarification. In any case I would like ask you to extend conclusion and provide some suggestions how different land use covers and climate classes combine  with a specific VI

Author Response

We sincerely thank the reviewer for considering our article for publication. Following the reviewer’s minor comment, now we have improved our conclusion by mentioning the interactions of relative performance of NDVI and VOD with variations over land cover and climate classifications, as following (lines 460 to 462):

“While over pixels located in tree covered areas (e.g., forests) VOD is more sensitive to meteorological droughts, over areas with sparse vegetation or bare land covers NDVI can capture agricultural droughts more significantly.”

Reviewer 4 Report

Thanks for the answers and all the modifications in the whole text that I asked before. I see you wouldn't do more spatial analysis that I wanted to mention at my first opinions, but I understand, that the article is long enough in this form as well.

Author Response

We sincerely thank the reviewer for considering our article for publication. We also thank reviewer for raising this comment. However, as reviewer also confirmed, currently the manuscript is long and we afraid adding further analysis and text would hinder the readability of the manuscript.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.