Effects of Land Use/Cover on Regional Habitat Quality under Different Geomorphic Types Based on InVEST Model

Round 1

Reviewer 1 Report

Comments to the Authors:

The authors applied InVEST model to estimate habitat quality in Altay Region of China and analyzed the effect of land use/cover to habitat quality and combined with a view of geomorphic types. A combination of habitat quality and geomorphic type was a  good perspective to observe the characteristics of habitat quality in vertical distribution, but there still remained some inadequacies. And then, I shall give the suggestions and questions were as following.

 

Questions:

1. habitat quality is an indicator for mirroring biodiversity, how to directly determines regional biodiversity? the authors should establish a correct relationship of habitat quality and biodiversity and revise the main text in total manuscript.
2. in line 272, the authors had stacked a series of steps to extract surface relief. Was the first step calculate the altitude difference and just need elevation value? How and What are the use of other topographic factors in table 3?
3. how to select the threat factors in this study? The more details should supply.
4. why took the habitat suitability scores as the weight of contribution index? Will it confuse the ratio of land use between different geomorphic types?
5. the connection of hot-spot analysis between other analyses in this manuscript showed a comparatively low. What’s the purpose of hot-spot analysis?
6. the parameter of habitat quality based on InVEST model was land use/type data, and the spatial distribution of land use/cover and the spatial distribution of habitat quality were high correlation. In line 539-549 and line 768-811, the author had analyzed the changes and distribution mechanisms of habitat were interpreted to the spatial distribution and transition of land use/cover. Did the authors consider the influence of this high correlation in parameter inputting and result outputting?

 

Suggestions:

- line 72-81

There were used too many conj., may the authors rephrase these sentences.

- line 105-110

Habitat assessment method should add the corresponding literatures, and, the second method which the authors referred that was using model, but the models could assess habitat quality were not only based on land use type, such as species distribution model. the authors should mention it.

- line 120

How to “maintain” biodiversity?

- line 149

Altay region was not a study object, it should be referred as a study area.

- the section of introduction could restructure and make the connection of several paragraphs show more logical and tight, such as the paragraph of surface morphology could shift with the paragraph of Altay region.

- line 190

“one belt one road” was an incorrect word, please revise it and check total manuscript.

- Section 2.2

“Gaofao-1” need to check spelling in line 208. The authors should supply a concise description of the accuracy of the LULC results based on GF-1.

- line 289, The equation (2) should check.

- line 295, the text ‘Table 5’ need check.

- table 4 should added units.

 

- Section 2.3.2

1. The manuscript mentioned the threat factor was referred to the relevant literature, but more details also should be introduced instead of a series of “cite”.
2. the principle and equation of Habitat Quality could revise to a concise introduction, because the InVEST user’s guide had explained very well which the reader could easily find it. But the parameter setting need a detailed explanation.

 

- figure 9, the legend of Low-High and Not Significant were same. The authors should change the color.

 

- Section 3.4.2, the correlation analysis is not the traditional correlation analysis which were tested the linear correlationship of one variation between another one variation. May the subtitle should be revised.

 

- the discussion section in the manuscript was stacked a series of the analysis of the results. It is suggested that the authors could move the content of the analysis to the results. The discussion should show your findings, the significance of findings, how the findings can be applied, and whether other authors have found the findings that similar with yours or different with yours.

Author Response

Response to Reviewer 1 Comments

Dear Prof.

First of all, thank you for your careful reading and valuable comments, which have a great effect on the improvement of my article, thank you!

Questions:

1. habitat quality is an indicator for mirroring biodiversity, how to directly determines regional biodiversity? the authors should establish a correct relationship of habitat quality and biodiversity and revise the main text in total manuscript.

Thank you for your advice.

This paper focuses on the impact of different land use types on habitat quality, and divides different geomorphic types to analyze habitat quality. Quoting "habitat quality has a great impact on biodiversity" is only used to lead to the research on habitat quality in this paper. The focus is not to summarize the relationship between habitat quality and biodiversity, which will be the work to be carried out in the future. In addition, in order to explain that habitat quality has a certain impact on biodiversity, this paper also adds relevant literature in line 77-82 and line 922-923.

1. in line 272, the authors had stacked a series of steps to extract surface relief. Was the first step calculate the altitude difference and just need elevation value? How and What are the use of other topographic factors in table 3?

Thank you for your advice.

First of all, I want to explain the division of geomorphic types in this paper. Referring to the previously published classification scheme of geomorphic types at all levels and scales in China and the latest 1:1 million digital geomorphic classification method, the basic geomorphic form types are mainly divided by three basic indicators: surface slope, surface undulation and altitude.

Then, combined with the actual situation of the study area, firstly, according to the surface slope combination, the study area is divided into two major geomorphic units: great plain area and mountainous area. Then according to the change of surface relief, combined with the elevation variation coefficient reflecting the elevation change and the surface roughness reflecting the surface relief change, the great plain area and mountainous area are subdivided into six basic geomorphic types: plain, platform, hill, small relief mountain, medium relief mountain and large relief mountain, that is, the medium-scale geomorphic type in this paper(Figure 4a).

And then, considering the ladder characteristics of the terrain in the study area, the study area is divided into three geomorphic surface altitude levels: low altitude (< 1000m), medium altitude (1000m-2400m) and high altitude (> 2400m). so that, the above 6 basic geomorphic types and 3 geomorphic surface altitude levels are combined into 17 basic geomorphic morphological types (Table 2). Therefore, line 272(line 303 in the revised draft) is mainly used to extract the surface relief, which is used in the best statistical unit.

Thanks to the expert's reminder, line272 is consistent with the surface relief in Table 3. There is no need to introduce it separately again, so the steps of line 272 (line 298-307 in the revised draft) have been deleted.

In addition, according to the opinions of experts, the terrain factors not used in Table 3, aspect, Gradient rate have been deleted.

1. how to select the threat factors in this study? The more details should supply.

Thank you for your advice.

More detailed description of how to select threat factors in this study has been added in line 393-435 of the revised manuscript.

 

1. why took the habitat suitability scores as the weight of contribution index? Will it confuse the ratio of land use between different geomorphic types?

Thank you for your advice.

It will not confuse the ratio of land use between different geomorphic type.

Habitat suitability scores were determined based on literature data and experts’ knowledge which had certain subjectivity and objectivity, so that it can not only take into account the subjective preference of decision makers, but also reduce the subjective randomness of empowerment. In addition, different geomorphic types included different land use types, the proportion of each land use type was different, and the habitat suitability of different land use types was also very different. The InVEST model believes that the more natural habitat types are, the more the sensitivity to threat factors, that is, the greater the impact of threat factors on habitat degradation, while human management factors can enhance the environment's ability to restore and reduce sensitivity. Therefore, when calculating the index contribution of different land use types to habitat quality within different geomorphic types, we need to take into account the proportion of land types and habitat suitability.

It is also supplemented in the revised manuscript in line 507-518.

1. the connection of hot-spot analysis between other analyses in this manuscript showed a comparatively low. What’s the purpose of hot-spot analysis?

Thank you for your advice. In this paper, hot-spot was used to describe the cold/hot spots of habitat quality in the study area to explore the local spatial agglomeration of habitat quality under the control of different geomorphic types. So as to analyze the impact of geomorphic types on habitat quality. This is also an innovation of this paper. Geomorphology is the most basic natural background. Exploring the impact of different land use types under geomorphic control on regional habitat quality can help people make rational use of land resources according to local conditions.

In order to increase the connection between hot spot analysis and other analysis, this paper adds a section (4.2 Effects of geomorphic types on habitat quality and suggestions for ecological management) in the discussion part.

1. the parameter of habitat quality based on InVEST model was land use/type data, and the spatial distribution of land use/cover and the spatial distribution of habitat quality were high correlation. In line 539-549 and line 768-811, the author had analyzed the changes and distribution mechanisms of habitat were interpreted to the spatial distribution and transition of land use/cover. Did the authors consider the influence of this high correlation in parameter inputting and result outputting?

Thank you for your advice. Yes, I have considered the influence of this high correlation in parameter inputting and result outputting. That's why we invited local experts to participate in the survey. The purpose of this paper is to use the invest model to evaluate the habitat quality of different land use types. In order to determine which land use types seriously threaten the quality of habitat, we listened to the opinions of local experts, asked experts to score (give weight to) relevant indicators, and then took the land types that may pose a threat to habitat as input data. The ultimate goal is to determine the land use types that seriously threaten the habitat quality, so as to change the future land resource use. More details has been added in line 354-360 and line 393-435 of the revised manuscript.

Suggestions:

- line 72-81

There were used too many conj., may the authors rephrase these sentences.

Thank you for your advice. I have adjusted and revised the content of the article.

- line 105-110

Habitat assessment method should add the corresponding literatures, and, the second method which the authors referred that was using model, but the models could assess habitat quality were not only based on land use type, such as species distribution model. the authors should mention it.

According to the experts’ suggestions, the corresponding literatures on habitat assessment method were added in the revised manuscript, and the second method of using the model was also supplemented. For more details, please refer to line 77-98 and line 106-122 in the revised manuscript.

- line 120

How to “maintain” biodiversity?

Thank you for your advice. Relevant descriptions have been added in the revised manuscript in line107-122.

- line 149

Altay region was not a study object, it should be referred as a study area.

Thank you for your advice. It has been changed into a study area in the revised manuscript.

- the section of introduction could restructure and make the connection of several paragraphs show more logical and tight, such as the paragraph of surface morphology could shift with the paragraph of Altay region.

Thank you for your advice. The paragraph has been adjusted, and the paragraph of surface morphology has been adjusted with the paragraph of Altay region. For more details, please refer to line 133-180 in the revised manuscript.

- line 190

“one belt one road” was an incorrect word, please revise it and check total manuscript.

The "one belt, one road" has been deleted once and the full text has been checked.

- Section 2.2

“Gaofao-1” need to check spelling in line 208. The authors should supply a concise description of the accuracy of the LULC results based on GF-1. “

Thank you for your advice. The misspelled " GAOFAO-1" has been changed to " Gaofen-1" (line 220), and the accuracy of LULC results are described in lines 239-258 of the revised manuscript.

- line 289, The equation (2) should check.

Thank you for your advice. The formula has been modified in the revised manuscript in line 317. Change - to+.

- line 295, the text ‘Table 5’ need check.

Thank you for your careful viewing. Inserted in the wrong place, table5 has been removed.

- table 4 should added units.

Thank you for your careful viewing. Due to the addition of a table in the previous modification, Table4 is changed to table5, and the unit (km) has been increased.

 - Section 2.3.2

1. The manuscript mentioned the threat factor was referred to the relevant literature, but more details also should be introduced instead of a series of “cite”.
2. the principle and equation of Habitat Quality could revise to a concise introduction, because the InVEST user’s guide had explained very well which the reader could easily find it. But the parameter setting need a detailed explanation.

 Section 2.3.2 has been revised following expert’s opinion. More details has been introduced and the parameter settings are also supplemented with more details. For more details, please refer to Section 2.3.2 in the revised manuscript.

- figure 9, the legend of Low-High and Not Significant were same. The authors should change the color.

Thank you for your careful viewing. The colors in figure 9 have been change.

- Section 3.4.2, the correlation analysis is not the traditional correlation analysis which were tested the linear correlationship of one variation between another one variation. May the subtitle should be revised.

 Thank you for your advice. The subtitle has been revised to：Effects of land use change on habitat quality.

- the discussion section in the manuscript was stacked a series of the analysis of the results. It is suggested that the authors could move the content of the analysis to the results. The discussion should show your findings, the significance of findings, how the findings can be applied, and whether other authors have found the findings that similar with yours or different with yours.

Thank you for your advice. According to expert’s opinion, the discussion section and results sections have been adjusted, and moved analysis content into results. In addition, the discussion section is divided into three parts to discuss. For more details, please refer to Section 4 Discussion.

Thank you again for reviewing my article and giving your valuable comments. This revision has enabled me to learn a lot of things needing attention in paper writing, which is of great help to my future paper writing. Thank you again.

Best regards

Author Response File: Author Response.docx

Reviewer 2 Report

Dear authors

Thank you for giving me the opportunity to read your work. The manuscript “Effects of land use/cover on regional habitat quality under different geomorphic types based on InVEST model” is very interesting and denotes a lot of work. The study of land use land cover, especially for studying habitat quality is a relevant topic in research. I really like the theme. Congrats on that. References are generally adequate and up-to-date. However, I have to say that in the end of the introduction is not clear what kind of novelty your work brings.

As main opinions I have to say that, despite this is an article about remote sensing and submitted to a high impact factor remote sensing journal, there is almost no information about the image classification procedure. One does not know what was the algorithm used for classification, how many and the extension of the training areas and worst we have not so ever data about the classification accuracy and the number of points used for validation. This becomes even more dangerous as the study performs LULC change analysis. The error in changes analysis is cumulative; therefore, if we have two images with 86% accuracy the change analysis between them will have a 73.96% accuracy (86*86/100). So, 26% of change can be due to classification errors.

The problems that come from combining data with 2m, 30m resolution and 1:1000000 scale must be discussed.

Line 367 – “A large number of simulation calculations show that the global Moran's I index is more accurate in identifying the center of the aggregation area, while the Getis-Ord G* coefficient is more accurate in detecting the aggregation area[62-64].” This is common knowledge and the characteristics if the two indicators. If your research question is about measuring the similarity of nearby features, you should use Moran's I. The measure only indicates that similar values occur together. It does not indicate whether any cluster is composed of high or low values. General G statistic can be used to indicate whether high or low values are concentrated over the study area. Hence, if you wish to find out whether your data is clustered in general (auto correlated) use Moran's I. If you want to know more specifically whether or not there are clusters of high/low values use G stat. Tt might be a good idea to use both indicators, global and local at the same time, easy to implement in e.g., GeoDa or following the methodology of: Abdulhafedh, A. (2017) A Novel Hybrid Method for Measuring the Spatial Autocorrelation of Vehicular Crashes: Combining Moran’s Index and Getis-Ord Gi* Statistic. Open Journal of Civil Engineering, 7, 208-221. doi: 10.4236/ojce.2017.72013. However, authors conducted an analysis of both measures but not a combined analysis of the two. Also, there is a lack of information about what distance measure was used, which can greatly affect the results.

On line 416, table 6, one should not perform spatial statistics analysis with a moving window with a even number of lines and columns. This means that cell being processed uses data that is not collected at the same distance in all directions. This is why moving windows are always odd values, 3x3, 5x5, 7x7 …..

In my modest opinion the discussion section is just more about results. It just a description of the obtained results. Discussion must compare the results with other works (preferably already mentioned in the texts), analysand the differences and similitudes, and advancing possible reasons to justify it.

As more smaller comments I would use preferably equations or expressions instead of formulas (usually reserved for chemistry).

Line 175 the area is 1.18×105 km, km² is the result

Line 263, table 3, slope is × 100 and the result is in percentage, if it is × 100% is x100/100 and continues equal

Line 436-437 – “6 medium-scale geomorphic types (Figure 4a) and 14 small-scale geomorphic types (Figure 4b). More classes, more detail, bigger scale, not the opposite. Perhaps authors meant detail or resolution.

And that’s all , sorry, for being so picky and continue the good work

Best regards

Author Response

Response to Reviewer 2 Comments

Dear Prof.

First of all, thank you for your careful reading and valuable comments, which have a great effect on the improvement of my article, thank you!

Q1: Thank you for giving me the opportunity to read your work. The manuscript “Effects of land use/cover on regional habitat quality under different geomorphic types based on InVEST model” is very interesting and denotes a lot of work. The study of land use land cover, especially for studying habitat quality is a relevant topic in research. I really like the theme. Congrats on that. References are generally adequate and up-to-date. However, I have to say that in the end of the introduction is not clear what kind of novelty your work brings.

R1: After being revised as suggested by experts, the innovations of this paper have been highlighted at the back of the introduction in line 143-163.

This paper aims to explore the impact of land use types on habitat quality under the control of different geomorphic types, so as to analyze the impact of physiognomy on habitat quality. This is also an innovation of this paper, because the related work on habitat quality research on different geomorphic types has not been carried out. Geomorphology is the most basic natural background, and exploring the influence of different land use types under the control of geomorphology on habitat quality of regional can help people to rationally utilize land resources according to local conditions.

Q2: As main opinions I have to say that, despite this is an article about remote sensing and submitted to a high impact factor remote sensing journal, there is almost no information about the image classification procedure. One does not know what was the algorithm used for classification, how many and the extension of the training areas and worst we have not so ever data about the classification accuracy and the number of points used for validation. This becomes even more dangerous as the study performs LULC change analysis. The error in changes analysis is cumulative; therefore, if we have two images with 86% accuracy the change analysis between them will have a 73.96% accuracy (86*86/100). So, 26% of change can be due to classification errors.

The problems that come from combining data with 2m, 30m resolution and 1:1000000 scale must be discussed.

R2: Thank you for your advice. In fact, we spent a month on-site investigation of 175 sample points in the study area to determine the accuracy of classification results. The accuracy of LULC results are described in lines 239-258 of the revised manuscript.

The 2 m resolution remote sensing image data is used to classify land use/cover, the 30m SRTM_ DEM data is used to classify geomorphic types, and the Chinese 1:1000000 digital geomorphic classification data is only for reference. Their purposes are different. Based on the existing classification system and classification results of 1: 1000000 digital geomorphic classification data in China, this paper sets the geomorphic classification system in combination with the actual situation of the study area, ,and uses the existing 1:1 million National geomorphic classification results to verify whether the geomorphic classification of the study area is accurate.

Q3: Line 367 – “A large number of simulation calculations show that the global Moran's I index is more accurate in identifying the center of the aggregation area, while the Getis-Ord G* coefficient is more accurate in detecting the aggregation area[62-64].” This is common knowledge and the characteristics if the two indicators. If your research question is about measuring the similarity of nearby features, you should use Moran's I. The measure only indicates that similar values occur together. It does not indicate whether any cluster is composed of high or low values. General G statistic can be used to indicate whether high or low values are concentrated over the study area. Hence, if you wish to find out whether your data is clustered in general (auto correlated) use Moran's I. If you want to know more specifically whether or not there are clusters of high/low values use G stat. Tt might be a good idea to use both indicators, global and local at the same time, easy to implement in e.g., GeoDa or following the methodology of: Abdulhafedh, A. (2017) A Novel Hybrid Method for Measuring the Spatial Autocorrelation of Vehicular Crashes: Combining Moran’s Index and Getis-Ord Gi* Statistic. Open Journal of Civil Engineering, 7, 208-221. doi: 10.4236/ojce.2017.72013. However, authors conducted an analysis of both measures but not a combined analysis of the two. Also, there is a lack of information about what distance measure was used, which can greatly affect the results.

R3: Thank you for your careful reading and comments. The meaning of this article is consistent with the experts' description of Moran's I and General G statistic. It may be due to language reasons, resulting in ambiguity. Thanks for the reference articles recommended by the experts (Abdulhafedh, A. ,2017). The contents of this paper have been modified according to the opinions of the experts. For more details, please refer to section 2.3.3 in the revised manuscript.

Q4: On line 416, table 6, one should not perform spatial statistics analysis with a moving window with a even number of lines and columns. This means that cell being processed uses data that is not collected at the same distance in all directions. This is why moving windows are always odd values, 3x3, 5x5, 7x7 …..

R4: Thank you for your careful reading and comments. In order to save space, I only show a part of moving windows, which may lead to misunderstanding that only moving windows with even rows and columns are used to perform spatial statistical analysis. Table 6 (Table 7 in the revised manuscript) has been supplemented completely. For more details, please refer to line 525 in the revised manuscript.

Q5: In my modest opinion the discussion section is just more about results. It just a description of the obtained results. Discussion must compare the results with other works (preferably already mentioned in the texts), analysand the differences and similitudes, and advancing possible reasons to justify it.

R5: Thank you for your advice. According to expert’s opinion, the discussion section and results sections have been adjusted, and moved analysis content into results. In addition, the discussion section is divided into three parts to discuss. For more details, please refer to Section 4 Discussion in the revised manuscript.

Q6: As more smaller comments I would use preferably equations or expressions instead of formulas (usually reserved for chemistry).

R6: Thank you for your advice. The smaller comments have been modified.

Q7: Line 175 the area is 1.18×105 km, km² is the result

R7: Thank you for your careful reading. I have changed the area to 1.18×10⁵ km² in line 188 in the revised manuscript.

Q8: Line 263, table 3, slope is × 100 and the result is in percentage, if it is × 100% is x100/100 and continues equal

R8: Thank you for your careful reading. The slope in table3 has been modified to × 100.

Q9: Line 436-437 – “6 medium-scale geomorphic types (Figure 4a) and 14 small-scale geomorphic types (Figure 4b). More classes, more detail, bigger scale, not the opposite. Perhaps authors meant detail or resolution.

R9: Yes, I mean more classes, more details, so that habitat quality changes at different geomorphic types scales can be analyzed.

Thank you again for reviewing my article and giving your valuable comments. This revision has enabled me to learn a lot of things needing attention in paper writing, which is of great help to my future paper writing. Thank you again.

 

Best regards

Author Response File: Author Response.docx