Review Reports

1. Summary

Thank you very much for taking the time to review this manuscript. We have studied the comments carefully and have made amendments to the paper that we hope will meet with your approval. Our point-by-point responses to each of the comments can be found in this document, and the revisions to the manuscript are highlighted.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: In the opinion of the reviewer, the classification of environmental quality should be clarified (Table 1). Why was a 5-level system chosen? Why is the scale uniform across RSEI values? Such a division is not valid for all indicators. For example, the division into ranges is uneven in Table 2.

Response 1: Thank you for pointing this out. Before conducting principal component analysis, we normalized the values of four indicators, mapping them to the range of [0,1], transforming them into dimensionless values. In order to make large values of PC1 represent good ecological conditions, further subtracted 1 from PCI to obtain the RSEI, which ranges between [0,1]. We added the following sentence as an explanation: The ecological quality is classified into five levels based on RSEI values with intervals of 0.2. The purpose is to simplify data interpretation and comparison, providing a more intuitive representation of differences in ecological conditions. (page 6, line 215-217) This uniform division is intended to assist researchers and decision-makers in better comprehending and communicating changes in ecological conditions while reducing complexity.

Comments 2: The selected boundaries in Table 2 also require explanation.

Response 2: We agree with this comment. This paper employs the natural breaks method to categorize the coefficient of variation (CV) into five levels. We added the following sentence as an explanation: The natural breaks method aims to categorize the study subject into groups with similar characteristics by identifying the natural breakpoints within the sequences, each of which holds statistical significance. (page 8, line 257-259) These natural breakpoints serve as effective boundaries for classification. It is worth noting, however, that the utilization of this method may result in uneven-scale divisions.

Comments 3: The RSEI values in Figure 3a show significant scatter and the superimposed trend line should have a low R² coefficient. Is this due to weather differences from year to year, the quality of the source data, or something else? Figure 3 itself relates to the results and discussion, so it makes more sense to move it and the associated text into the Results section.

Response 3:  Thank you for your question and suggestion. According to our investigation, the fitted trend line in Figure 3a does exhibit a relatively low R2 coefficient, but it remains within a reasonable range. This is considered to be attributed to factors such as annual precipitation and climate. Additionally, we have taken your suggestion into consideration and relocated Figure 3 along with the relevant text from the Methods section to Section 3.1 in the Results. We have also made some modifications to optimize the logical flow of the paper.

The annual average trend of RSEI from 1990 to 2021 is shown in Figure 3a. The mean of RSEI exhibits a trend of initially decreasing and then increasing, with 2008 representing the lowest point. The average RSEI value declined from 0.84 in 1990 to 0.59 in 2008 and increased to 0.80 in 2021. The t-value of the time series exceeds the significant interval in 2004 and 2013, indicating these two years as breakpoints in the time series (Figure 3b). Consequently, we selected specific years as temporal nodes for segmenting the long time series data in the study area: 1990 (start of the study period), 2004 (breakpoint), 2008 (minimum value), 2013 (breakpoint), and 2021 (end of the study period). (page 8-9, line 273-280)

Comments 4: The chart in Figure 5f shows that there are virtually no areas with poor RSEI in 2021. However, in Figure 5e we see a large red spot in the southeastern part, which is several times larger in area than the previously observed problem area. Please explain.

Response 4: Thank you for pointing out the error in our figure. In 2021, areas with a 'poor' classification do indeed exist in the southeastern part of the study area, precisely where the reservoir project is located. The issue with the figure lies in the chronological order of the years on the Y-axis in Figure 5f. We have verified and rectified this problem, and the corrected version is now available in Section 3.2.1 on page 10 of the manuscript.

Comments 5: Authors should include higher resolution maps in Figures or attach them as supplementary materials.

Response 5: Thank you for your suggestion. We have made corrections to some of the figures in the paper, displaying them at a higher resolution for a smoother reading experience for the readers.

Comments 6: "i" should be a subscript for "Ei" (line 202)

Response 6: Thank you for your correction. We have made amendments to rectify this error.

3. Response to Comments on the Quality of English Language

Point 1: Minor editing of English language required.

Response 1: Thanks so much for your useful comments. The manuscript has been thoroughly revised and edited by a native speaker.

1. Summary

Thank you very much for taking the time to review this manuscript. We have studied the comments carefully and have made amendments to the paper that we hope will meet with your approval. Our point-by-point responses to each of the comments can be found in this document, and the revisions to the manuscript are highlighted.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: There are no complaints about the article. But the list of references consists of 90% exclusively of Chinese works? Is it really true that no similar research is being carried out in the whole world except China? Authors are encouraged to add similar foreign works.

Response 1: Thank you for your valuable suggestion. As our study area is a specific reservoir region in China, we have primarily referenced literature from Chinese scholars. We are willing to accept your suggestion and have added some relevant international literature to the references, thereby balancing the sources in the references. The additional references are as follows:

1.       Fischer, J.; Gardner, T.A.; Bennett, E.M.; Balvanera, P.; Biggs, R.; Carpenter, S.; Daw, T.; Folke, C.; Hill, R.; Hughes, T.P.; et al. Advancing sustainability through mainstreaming a social-ecological systems perspective. Curr. Opin. Environ. Sustain. 2015, 14, 144–149.

2.       Nathaniel, S.P.; Adeleye, N. Environmental preservation amidst carbon emissions, energy consumption, and urbanization in selected African countries: Implication for sustainability. J. Clean Prod. 2021, 285, 125409.

3.       Khan, I.; Hou, F.; Le, H.P. The impact of natural resources, energy consumption, and pollution growth on environmental quality: Fresh evidence from the United States of America. Sci. Total Environ. 2021, 754, 142222.

4.       Speidel, J.J.; Weiss, D.C.; Ethelston, S.A.; Gilbert, S.M. Population Policies, Programmes and the Environment. Philos. Trans. R. Soc. B Biol. Sci. 2009, 364, 3049–3065.

5.       Borrelli, P.; Robinson, D.A.; Fleischer, L.R.; Lugato, E.; Ballabio, C.; Alewell, C.; Meusburger, K.; Modugno, S.; Schütt, B.; Ferro, V.; et al. An assessment of the global impact of 21st century land use change on soil erosion. Nat. Commun. 2017, 8, 1–13.

6.       Sterk, G.; Stoorvogel, J. Desertification–scientific versus political realities. Land. 2020, 9, 156.

3. Response to Comments on the Quality of English Language

Point 1: English language fine. No issues detected.

Response 1: Thank you for your recognition.

1. Summary

Thank you very much for taking the time to review this manuscript. We have studied the comments carefully and have made amendments to the paper that we hope will meet with your approval. Our point-by-point responses to each of the comments can be found in this document, and the revisions to the manuscript are highlighted.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: In Section 3.1.1, "Spatial Distribution of Ecological Environment," you provided illustrative explanations for the spatial distribution of four remote sensing indicators and RSEI in the study area across different time periods. However, I find several points of confusion in Fig5 (P10):

(1) Could you please provide an explanation for the two distinctly red (indicating poor) areas in the figure, namely the eastern and southeastern regions?

Response (1): Thank you for pointing this out. We added the following sentence as an explanation: It is worth noting that the regions classified as "poor" in the eastern and southeastern parts correspond to the locations of Laoyahe Reservoir and Chushandian Reservoir, respectively. The construction of these reservoirs has caused significant disturbance to the ecological environment, resulting in the conversion of forest and cropland into build-up and water. Consequently, this has had detrimental effects on the ecological quality of the reservoir area. (page 10, line 311-316)

The construction of Chushandian Reservoir in the southeast began in 2015, which explains why there were no red areas in the southeastern region of the map before 2013.

(2) Why did the RESI levels in the eastern "poor" region consistently improve from 1990 to 2008 but experienced a deterioration again in 2013?

Response (2): Thank you for pointing this out. The "poor" region in the east refers to the Laoyahe Reservoir area. Through investigation, we found that the annual precipitation in 2013 was relatively low, below the multi-year average. In contrast, other years, such as 2004 and 2008, had higher annual precipitation compared to 2013. Precipitation likely influences the variation in RSEI levels in the reservoir area. Additionally, the ecological quality of the reservoir area is also affected by factors such as reservoir water levels, capacity, and temperature. In subsequent research, we will consider additional driving factors to conduct a more detailed analysis and interpretation of environmental changes within the study area.

(3) We can clearly observe that the red "poor" areas in Figure (e) occupy a larger area compared to Figures (a), (b), (c), and (d). However, why is there no representation of the proportional area of "poor" in Figure (f) for the year 2021?

Response (3): Thank you for pointing out the error in our figure. In 2021, areas with a 'poor' classification do indeed exist in the southeastern part of the study area, precisely where the reservoir project is located. The issue with the figure lies in the chronological order of the years on the Y-axis in Figure 5f. We have verified and rectified this problem, and the corrected version is now available in Section 3.2.1 on page 10 of the manuscript.

Comments 2: Regarding the autocorrelation analysis issue:

(1) On line 238 of page 7, it is suggested to provide explicit explanations for the clustering results: high-high, high-low, low-high, low-low, or non-significant, elucidating their respective implications.

Response (1):  Thank you for your suggestion. We added the following explanation to the manuscript: High-High and Low-Low denote agglomeration in areas with relatively high (low) RSEI values, while High-Low and Low-High signify the coexistence of areas with high and low RSEI values, highlighting a contrasting phenomenon. “Not significant” indicates that the spatial correlation between RSEI values lacks significance in geographical space. (page 7, line 239-243)

(2) Furthermore, in Section 3.1.2 on autocorrelation analysis on page 10, rather than a description of Fig6, I am interested in an explanation of the underlying factors contributing to the observed phenomenon.

Response (2): Thank you for your suggestion. We have made modifications to Section 3.2.2 on page 10, replacing descriptions of the phenomena with an analysis of underlying factors: The High-High clustering areas are predominantly situated in the mountainous and hilly terrain surrounding the study region. These areas exhibit minimal distribution fluctuations, characterized by high ecological quality and relatively minor disturbances from human activities. In contrast, the Low-Low clustering areas are mainly found in urban built-up areas, cropland, and reservoir regions. These areas face substantial disruptions from human activities, especially notable seasonal changes affecting farmland, contributing to the unstable distribution of the Low-Low clustering regions. (page 10-11, line 327-334)

Comments 3: In Figure 9 on page 13, why is the southeastern reservoir area predicted to undergo persistent degradation in the future?

Response 3: Thank you for your question. The predictive principle of this study involves overlaying multi-year RSEI distribution maps, constructing time series on a per-pixel basis, and calculating the Hurst index for each sequence. This process allows us to forecast future trends. The Chushandian Reservoir in the southeastern part of the study area began construction in 2015, leading to the extensive conversion of forest and farmland into construction land, disrupting the ecological environment of the reservoir area. Therefore, after 2015, the ecological environment in the reservoir area exhibited a declining trend. Predicting the future trend based on data from the past few years, considering the aforementioned trend, inevitably forecasts a continued degradation of the ecological quality in the reservoir area.

Comments 4: There is a question that requires contemplation: in several figures within the manuscript (P10 Fig5, P12 Fig8, P13 Fig9), the reservoir area on the southeastern side of the study area consistently demonstrates lower ecological quality levels. However, the construction of reservoirs constitutes a human intervention aimed at optimizing the ecological environment. The consideration of only four remote sensing data, namely NDVI, WET, NDBSI, and LSI, as indicators for ecological assessment, resulting in significant outcomes in the reservoir area, raises the question of whether there is an issue of insufficient consideration of indicator factors.

Response 4:  Thank you for your valuable feedback. The research methodology of this paper is rooted in the Remote Sensing Ecological Index (RSEI), which was first proposed by Chinese scholar Hanqiu Xu in 2013. This index couples four indicators—NDVI, WET, NDBSI, and LST—enabling accurate monitoring and assessment of regional ecological environments, and it has gained widespread recognition among scholars. For the lower ecological quality level in the reservoir area, the reason considered is the ecological disturbance caused by the reservoir construction project starting in 2015, which has been confirmed in the temporal stability analysis in Section 3.3.1.

The construction of RSEI is highly flexible, allowing for improvements based on the specific conditions of the study area. We elaborated on this aspect in Section 4.3 (Limitations and Future Work). Therefore, in future research, we will adhere to your suggestion and no longer be confined to the four indicators of NDVI, WET, NDBSI, and LST. Instead, we will consider introducing new indicators to construct an RSEI more tailored to the study area, thereby obtaining more precise results.

Comments 5: The entire manuscript dedicates significant attention to exploring the spatiotemporal heterogeneity of the ecological environment, with only a limited portion devoted to the section on land use change response. It is recommended to augment the analytical content in this regard.

Response 5: Thank you for your suggestion. In Section 3.4 on pages 14 and 15 of the manuscript, we have enriched the analysis of the impact of land use changes on the ecological environment. Additionally, we have relocated some relevant content to Section 4.2 in the Discussion, focusing on the impact of human activities on the ecological environment.

Comments 6: Please increase the font size of the text inside Fig3 on Page 7 and Fig7 on Page 12; it appears too small.

Response 6: Thank you for pointing this out. We have made modifications to Figure 3 and Figure 7, and the revised versions have been included on pages 9 and 12 in the manuscript.

3. Response to Comments on the Quality of English Language

Point 1: English language fine. No issues detected.

Response 1: Thank you for your recognition.