Review Reports - Spatiotemporal Evolution and Drivers of Ecological Quality in the Tengger Desert (2001–2021)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Brief Summary

This study uses the Remote Sensing Ecological Index (RSEI) to thoroughly assess spatiotemporal variation in the ecological quality of the Tengger Desert from 2001 to 2021. The study makes an important contribution by conducting several analyses to assess changes in the ecological fragility of the Tengger Desert and to evaluate future outlooks. In particular, the authors conducted analyses to assess (a) temporal trends in annual mean RSEI; (b) changes in spatial distributions of RSEI; (c) the degree of clustering versus spatial randomness across time; (d) the persistence of temporal trends in RSEI; (e) future projections of trends in ecological quality; and (f) the key variables driving spatial variation in RSEI across time. These analyses are generally well-described in the Methods and Results sections.

General Concept Comments

The terms “ecological quality” and “environmental quality” are used prominently throughout the paper, but they are not specifically defined. What is meant by ecological and environmental quality, and what combination of conditions constitutes high versus low quality? For example, vegetation and drought indices are discussed in the second paragraph of the Introduction, but it’s not clear how these indices are used to define environmental or ecological quality. Moreover, the first sentence of the third paragraph reads, “The Remote Sensing Ecological Index (RSEI) integrates four key indicators: greenness, humidity, dryness, and heat.” It’s not specifically explained whether or how RSEI estimates environmental quality or ecological quality. This is briefly addressed in the Discussion (lines 406-407), but more details are needed in the Introduction and Methods sections.

Overall, the Methods section is clear, detailed, and well-written. The descriptions of the Theil-Sen slope estimation, Mann-Kendall trend analysis, spatial autocorrelation analysis, Hurst index, and Geodetector analysis are well-written, and the authors provide helpful information about interpreting the results from these analyses. However, Section 2.2.1 could be improved by adding further details. It’s not clear whether the RSEI dataset was produced for this study, or whether it was previously produced and published. It credits a dataset to Xu Dong et al., but there is no specific citation. Additionally, the acronyms LST and NDBSI are not defined. If the RSEI data was produced as part of this study, can the authors provide more details about the principal components analysis that was used to produce the RSEI data?

Regarding the assessment of driving mechanisms, it sounds as though the RSEI data was derived using a land use/land cover (LULC) input dataset. This study determined that LULC is a key factor influencing changes in RSEI. However, this result seems to be potentially due to LULC being an input to the RSEI dataset (note, the authors briefly addressed this in the Discussion in lines 419-420). Related to this point, lines 232-239 (Methods) describe how several factors were selected to explore the driving mechanisms of RSEI spatiotemporal variation. From this description, it’s not clear if these are the same datasets that were used to produce the RSEI index, or if these datasets are related to ecological quality but were not directly used to produced RSEI. Additionally, the use of “mechanisms” suggests a process-based analysis. The authors might consider using “variables” or “factors” rather than “mechanisms.”

Lastly, the authors briefly discussed the limitations of the 30 m spatial resolution in lines 450-452 (Discussion). The authors might consider also discussing whether changes in ecological quality are best detected at spatial resolutions of 30 m, or whether there might be ecological changes happening over smaller areas (i.e., sub-grid variation) that are not captured by analyses at 30 m resolution. Are there examples in the literature of changes in ecological quality over areas smaller than 30 m squared?

Specific Comments

Abstract: Overall, the abstract needs more detailed information. Lines 17-18 need further details about the Ecological and Environmental Quality Datasets as well as the anthropogenic indexes. Additionally, readers need more context to understand the locking engineering project described in line 21.

Line 69: The acronym RSEI can be used, rather than writing out the full words.

Line 72: It’s not clear how the sparse population and underdeveloped socioeconomic conditions contribute to the ecological challenges in the Tengger Desert.

Line 84: It’s not clear what is meant by “deeper, intrinsic drivers”. Please provide some examples.

Line 89: The RSEI acronym was already defined in line 54; however, in line 54 it was defined as a singular “Index”, rather than indices as written on line 89.

Line 94: It’s not clear how dynamic is being used in this sentence. Might it be clearer to simply say monitoring of regional ecological changes, rather than dynamic monitoring?

Line 95: It's not clear what is meant by “optimizing” here. Is a different word intended, perhaps "assessing"?

Line 100: Perhaps also state China, for readers unfamiliar with this geography.

Line 114: In Figure 1, in the map of China, there is an inset map of the southeastern part of China. It is not clear why the inset map is included.

Line 123: the RSEI acronym is defined again, although it has previously been defined.

Line 138: The CRU acronym is not defined.

Line 146: What LULC classes were included in this dataset?

Line 150: Which additional topographic metrics?

Line 219: Is “interpretation” intended here? Perhaps the authors intended to write “explanation” or another similar word.

Line 237: It’s not clear what is meant by the requirements of the study area.

Line 250: The last sentence of Section 3.1 describes that the results showing decreasing trends in RSEI underscore the need for more effective long-term measures. It’s not clear what the long-term measures would be. Additionally, the authors might consider discussing this point in the Discussion, rather than in the Results.

Line 264: In Figure 3, should there be a label on the Y-axis? Is this figure showing the percentage of pixels in each quality category?

Line 276-277: It’s not necessary to write it out this way. It can simply be stated, “for the years 2001, 2006, 2011, 2016, and 2021 (Figure 4). If panel F is the mean across all years, this should be stated.

Line 338: How were the Hurst Index and Sen Index combined? This should be described in the Methods section.

Line 341-343: Regarding the last sentence in this paragraph, the authors might consider discussing this in the Discussion, rather than in the Results.

Line 355: Section 3.4 is titled “Mechanisms driving RSEI”, but the word “mechanisms” suggests a process-based analysis. The authors might consider a title such as “Variables driving RSEI” or “Factors driving RSEI”. (Same point for Section 4.2).

Line 399: It’s not clear what is meant by “human-induced damage.” What kind of human activities and how did they affect ecological quality?

Line 414: The caption for Figure 10 describes linear trend information. The figure shows time series data but does not include information about linear trends.

Line 416: Does Figure 11 show annual mean NDVI for 2001 and 2021? If these datasets are from individual dates, do differences in seasonality contribute to differences in NDVI?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study shows an interesting procedure to evaluate environmental and human patterns on the performance of indices built with remote sensors

I think that the authors carried out a good analysis and good discussion; However, in my opinion the results section is difficult to understand. I suggest the following two points in order to improve this section:

The authors mention the origin of the meteorological, relief and land use data; However, they do not clearly explain which variables were evaluated on the RSEI.

I suggest changing the following text (lines 232-235) to the beginning of the methodology section.

Lines 232-235. “Considering the study area's conditions and data availability, key factors closely related to the RSEI and easily quantifiable—land use and land use cover (LULC), precipitation (PRE), temperature (TMP), wind speed (SWS), elevation (DEM), slope (SLOP), and relief (REL) were selected to explore the driving mechanisms of RSEI spatiotemporal evolution”

I also suggest to the authors to include a table with the mean, maximum, minimum, and standard deviation values for each of these variables for de study area (this for each year included in this study).

In order to be consistent with the methodology I suggest the following:

The authors must explain the results one by one, in the same order that the analysis procedures were declared in the methodology. According to the methodology, the order is as follows:

1. Mann-Kendall trend analysis

2. Spatial autocorrelation analysis

3. The Hurst index

4. Geodetector

In the results section, authors should first include the results of the Mann-Kendall trend analysis. Afterwards, I suggest to include all the information contained in sections 3.2.3, and also all the information in sections 3.3 and 3.4. I mean, to change the information included in the lines 311 to 385 from the end of the results section to the beginning of this section.

Once these changes are made, authors will be able to include the information in section 3.1...

Finally, it would be convenient to include a principal components analysis in order to make more evident the influence of the variables evaluated on the index. Although this analysis may be subject of other research related to the topic of this manuscript.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

1. Introduction

It is well developed and detailed but a comprehensive literature review has not been carried out, very centralised and other regions are not taken into account. In addition, only recent citations are used when the hierarchical analysis method o ecological footprint method, what are already long-established methodologies in ecology.

References 28 to 31 cited in the text are missing the year.

2. Materials and Methods

References 36, 37 and 38 do not seem relevant for the characterisation of the study area, you could select others that are.

Line 119, reference misquoted in the text

Lines 128 - 131 What criteria did you use to establish the RSEI class intervals? Do not just put the reference.

Line 146, reference cited in the text are missing the year.

I have left out citation errors, please review the entire text, Be careful because, a negative impression is created when references and citations in the text are not correct, as it is easy to implement it denotes a lack of thoroughness.

Line 146 How accurate is the Land Use and Land Cover Data (LULC) used?

Line 151 first full stop are incorrect

Line 162 The original Reference on the Theil-Sen Slope Estimation method is missing.

Line 183 the reference to Moran Index is missing

Line 199 the reference to origen of Moran Index is missing

3. Results

Line 295, 302, 305, 306 y 308 What statistical analysis as well as significance level has been applied to indicate that there are significant differences, significant decreases or no significant increases?

Line 329 which pixel size was used, 1 km for climate variables or the 30 m pixel size for uses?

Line 358 which test has been used? How and in which programme did you calculate the q value?

4. Discussion

You can't put results in the discussion, unless you do results and discussion together.In addition, you have more sections on results than on discussion, and they do not correspond to the objectives you set in the introduction.

You must rearrange and reorganise them.

5. Conclusions

You should write it in the form of a conclusion, not results.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors, the quality and intelligibility of the article has improved considerably. The work at this stage would be publishable in my opinion. I leave the decision to the editor.