Quantitative Analysis of Factors Influencing Spatial Distribution of Soil Erosion Based on Geo-Detector Model under Diverse Geomorphological Types
Round 1
Reviewer 1 Report
I would suggest the authors consider minor edits to the manuscript in order to enhance the clarity of the presentation of their work.
************************************
Notes for the authors of the manuscript n. land-1235885
In the Abstract, consider improving manuscript’s readability by minor edits like the ones suggested in the following:
[line 11] most severe erosion in the world
[line 14] independently or in pairs by interacting? {it would be a more precise statement in statistical terms, also since interaction could happen at higher degrees too}
[line 21] effect degree size
[line 22] The interactive effect degree size of interaction between land-
[lines 28-29] different accuracy caused sparkled the problem issue of “data coupling problem”
In the Introduction, consider improving manuscript’s readability by minor edits like the ones suggested in the following:
[lines 35-37] system, and as a medium … basic materials, such as foods, organic matters, etc [1,2].
[line 48] soil erosion [16, 17]. Some authors [18] concluded
[line 56] independently or in pairs by interacting. (3) How to identify the risk areas at higher risk of soil erosion?
[line 59] including correlation and regression analysis
[lines 62-70] soil erosion. By estimating a multiple regression to investigate the effect of modified rainfall erosivity, land use and soil erodibility on soil erosion, the authors of [15] concluded that rainfall erosivity and soil erodibility were the major factors. As alternative to classical regression, network and redundancy analysis has also been applied [28] as a practical method to quantify the interactions co-dependencies between soil erosion and influential factors as well as the interactions among driving factors. Nonetheless these conventional statistical methods are based on a set of statistics and can only be used to quantify the relationship between soil erosion and its driving factors, but often fail to demonstrate the spatial distribution characteristics of these those influencing factors [15,29]. And although Besides, spatial analysis methods have offer the advantages to identify zones with high or low erosive intensity, they cannot directly quantify the degree effect of factors driving soil erosion.
[lines 71-72] The Geo-detector model is a quite new statistical approach [31,32]. It is based primarily on the clear assessment of spatial heterogeneity of in geographical phenomena.
[lines 78-81] of two interacting independent explanatory variables on a specific dependent target variable. Besides Specifically, it can also detect high-risk areas of soil erosion in the study area by superimposing the maximum soil erosion modulus of each influencing factor [33,34]. In theory, the model can answer all the questions discussed above.
[line 89-90] and the Geo-detector model will be was applied to determine the dominant factor(s) and their interactions driving soil erosion, and to identify the high
[line 99] The region belongs is subject to an arid to semi-arid climateic conditions
[line 107] A reference study [37] divided
[line 118] In the first row of \Table S1 I would suggest to make explicit that the 2nd, 3rd and 5th columns report average values
[line 169] I would suggest using a typeface for the symbol f consistent with the one in eq. (10)
[line 170-] in eq. (10) I would recommend using log to denote the base 10 logarithm function
[line 177] the P values of different land-use types were are shown in the following tables
[line 186] the influence of two interacting independent explanatory variables on a specific dependent target variable.
[line 187] for the sake of clarity I would suggest sticking with the English names of outcomes of the Geo(g)-detector tool in ref. [A2]: The model outcome includes factor detection, interactive factor detection of interactions, risk detection, and ecological detection [31, 32]
[line 192-] please, review the notation in eq. (11): the summation has to be taken from h=1 to L
[lines 194-5] h is the index to denote each of the strata of related to the dependent variable Y or as well as the independent variables X; Nh and N are the units of layer stratum h and whole study areas respectively and similarly in the following for the variances
[line 200] Interactive factor Detection of interactions
[line 201] enhances or weakens their respective explanatory power
[line 208] The input variables to the Geo-detector model are required to be categorical data stratified; thus, continuous variables needed to be discretized
[line 226-7] The proportion of study area classified by erosion intensity level as “Slight” increased by 16.09%, whereas the proportion of related to other erosion intensity levels
[line 232-3] the severity degree of soil erosion in Yan’an City has greatly improved appreciably reduced.
[line 237] areas of “Slight” erosion
[line 237] areas of “Extreme” erosion
[line 237] areas of “Intense, Very Intense, and Extreme” erosion
[line 253] 3.2.1. Effect degree Analysis on of single simple effect by influencingtial factors of soil erosion.
[line 279] 3.2.2. Effect degree Analysis on of single interaction effect of factors of soil erosion.
[line 280] Interactive detection Results of detection of interactions show
[line 305-6] Based on the principle of risk detection, high-risk stratification of influencing factors by high-risk of soil erosion (soil erosion modulus ≥ 15000 t/km2·a) of soil erosion can be determined
[line 331] 3.3. Temporal analysis on dominant factors of soil erosion.
I would recommend to explain if the temporal analysis is based on only two dates of observation (2010 and 2017) and explain how the correlation between precipitation and soil erosion mentioned in the closing sentence of the section has been evaluated (in my understanding, correlation in spatial terms has already been considered in the previous section 3.2.1, and in temporal ones would be meaningless when just two points in time are available).
[line 359] has been greatly improved reduced
[line 367-8] It was Intervention to alleviate soil erosion have focused on vegetation and slope gradients that have been taken seriously in the study area.
[line 376] was compared with previously published results studies. Authors of a similar study [49] estimated the average erosion modulus
[line 377] Recently, another study [50] suggested that
[line 381-2] Therefore, the results of the RUSLE model in the study area have certain are attached to good reliability.
[line 400] And Moreover, precipitation data with the macroscopic spatial distribution and less microscopic differences brought by fewer sampling points, and as well as the interpolation techniques adopted also weaken the explanatory power of this factor on soil erosion.
[line 406-10] please, clarify the difference observed in the results
[line 410-12] please, specify which statistics supply the information reported and the types the two figure are related
[line 439] pflat platforms
[line 441] singler factors
Suggested additional references:
[A1] Liao, Y., Zhang, Y., He, L., Wang, J., Liu, X., Zhang, N., & Xu, B. (2016). Temporal and Spatial Analysis of Neural Tube Defects and Detection of Geographical Factors in Shanxi Province, China. PLoS One, 11(4), e0150332. doi: 10.1371/journal.pone.0150332
[A2] Wang, J.-F., & Hu, Y. (2012). Environmental health risk detection with GeogDetector. Environ. Model. Software, 33, 114–115. doi: 10.1016/j.envsoft.2012.01.015
Author Response
Thank you very much for your comments.
A detailed response to the reviewer 1
Thank you very much for the reviewers’ evaluation and comments on our paper. Your comments were highly insightful and enabled us to greatly improve the quality of our manuscript. We have substantially revised our manuscript after reading the comments provided by the reviewer 1. The explanation of what we have changed in response to the reviewers’ concerns is given point by point in the following pages.
Response to Reviewer 1 Comments
Point 1: [line 11] most severe erosion in the world
Response 1: Thank you for your kind suggestion. We have revised the sentence according to your and another expert’s kind suggestions in this revised manuscript, which can be found in lines 11-12.
Point 2: [line 14] independently or in pairs by interacting? {it would be a more precise statement in statistical terms, also since interaction could happen at higher degrees too}
Response 2: Thank you for your kind suggestion. We have replaced "in pairs" with "by interacting" in this revised manuscript, which can be found in line 14.
Point 3:[line 21] effect degree size
Response 3: Thank you for your kind suggestion. We have replaced "effect degree" with "effect size" in this revised manuscript, which can be found in line 21.
Point 4: [line 22] The interactive effect degree size of interaction between land-
Response 4: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 22-23.
Point 5: [lines 28-29] different accuracy caused sparkled the problem issue of “data coupling problem”
Response 5: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 28-29.
Point 6: [lines 36-38] system, and as a medium … basic materials, such as foods, organic matters, etc [1,2].
Response 6: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 36-38.
Point 7: [line 49] soil erosion [16, 17]. Some authors [18] concluded
Response 7: Thank you for your kind suggestion. We have revised this type of error according to your kind suggestion in this revised manuscript, which can be found in line 49.
Point 8:[lines 58-59] independently or in pairs by interacting. (3) How to identify the risk areas at higher risk of soil erosion?
Response 8: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 58-59.
Point 9: [line 61] including correlation and regression analysis
Response 9: Thank you for your kind suggestion. We have replaced "correlation regression analysis" with "correlation and regression analysis" in this revised manuscript, which can be found in line 61.
Point 10: soil erosion. By estimating a multiple regression to investigate the effect of modified rainfall erosivity, land use and soil erodibility on soil erosion, the authors of [15] concluded that rainfall erosivity and soil erodibility were the major factors. As alternative to classical regression, network and redundancy analysis has also been applied [28] as a practical method to quantify the interactions co-dependencies between soil erosion and influential factors as well as the interactions among driving factors. Nonetheless these conventional statistical methods are based on a set of statistics and can only be used to quantify the relationship between soil erosion and its driving factors, but often fail to demonstrate the spatial distribution characteristics of these those influencing factors [15,29]. And although Besides, spatial analysis methods have offer the advantages to identify zones with high or low erosive intensity, they cannot directly quantify the degree effect of factors driving soil erosion.
Response 10: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 64-72.
Point 1: [lines 76-77] The Geo-detector model is a quite new statistical approach [31,32]. It is based primarily on the clear assessment of spatial heterogeneity of in geographical phenomena.
Response 1: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 76-77.
Point 11: [lines 82-86] of two interacting independent explanatory variables on a specific dependent target variable. Besides Specifically, it can also detect high-risk areas of soil erosion in the study area by superimposing the maximum soil erosion modulus of each influencing factor [33,34]. In theory, the model can answer all the questions discussed above.
Response 11:Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 82-86.
Point 12: [lines 94-96] and the Geo-detector model will be was applied to determine the dominant factor(s) and their interactions driving soil erosion, and to identify the high
Response 12: Thank you for your kind suggestion. We have replaced simple future tense with simple past tense according to your kind suggestion in this revised manuscript, which can be found in lines 94-96.
Point 13: [line 106] The region belongs is subject to an arid to semi-arid climateic conditions
Response 13: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in line 106.
Point 14: [line 115] A reference study [37] divided
Response 14: Thank you for your kind suggestion. We have revised this type of error according to your kind suggestion in this revised manuscript, which can be found in lines 115-117.
Point 15: [line 127] In the first row of \Table S1 I would suggest to make explicit that the 2nd, 3rd and 5th columns report average values
Response 15: Thank you for your kind suggestion. We have added the word “average” in the front of the 2nd, 3rd and 4th columns in Table S1 according to your kind suggestion in this revised manuscript, which can be found in line 127.
Point 16: [lines 177-] I would suggest using a typeface for the symbol f consistent with the one in eq. (10)
in eq. (10) I would recommend using log to denote the base 10 logarithm function
Response 16: Thank you for your kind suggestion. We have revised the eq. (10) and eq. (11) according to your kind suggestion in this revised manuscript, which can be found in lines 177-.
Point 17: [line 184] the P values of different land-use types were are shown in the following tables
Response 17: Thank you for your kind suggestion. We have replaced “simple past tense” with “simple present tense” according to your kind suggestion in this revised manuscript, which can be found in line 184.
Point 18: [line 193] the influence of two interacting independent explanatory variables on a specific dependent target variable.
Response 18: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 193-194.
Point 19: [lines 194-195] for the sake of clarity I would suggest sticking with the English names of outcomes of the Geo(g)-detector tool in ref. [A2]: The model outcome includes factor detection, interactive factor detection of interactions, risk detection, and ecological detection [31, 32]
Response 19: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 194-195.
Point 20: [lines 199-] please, review the notation in eq. (12): the summation has to be taken from h=1 to L
Response 20: Thank you for your kind suggestion. We have revised this eq. (12) according to your kind suggestion in this revised manuscript, which can be found in lines 199-.
Point 21: [lines 201-203] h is the index to denote each of the strata of related to the dependent variable Y or as well as the independent variables X; Nh and N are the units of layer stratum h and whole study areas respectively and similarly in the following for the variance
Response 21: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 201-203.
Point 22: [line 207] Interactive factor Detection of interactions
Response 22: Thank you for your kind suggestion. We have replaced "Interactive factor detection" with "Detection of Interactions" in this revised manuscript, which can be found in line 207.
Point 23: [line 208] enhances or weakens their respective explanatory power
Response 23: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in line 208.
Point 24: [lines 215-216] The input variables to the Geo-detector model are required to be categorical data stratified; thus, continuous variables needed to be discretized
Response 24: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 215-216.
Point 25: [lines 233-235] The proportion of study area classified by erosion intensity level as “Slight” increased by 16.09%, whereas the proportion of related to other erosion intensity levels
Response 25: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 233-235.
Point 26: [lines 239-240] the severity degree of soil erosion in Yan’an City has greatly improved appreciably reduced.
Response 26: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 239-240.
Point 27: [line 244] areas of “Slight” erosion
[line 248] areas of “Extreme” erosion
[line 252] areas of “Intense, Very Intense, and Extreme” erosion
Response 27: Thank you for your kind suggestion. We have revised these wrong expressions according to your kind suggestion in this revised manuscript, which can be found in lines 244, 248 and 252.
Point 28: [line 261] 3.2.1. Effect degree Analysis on of single simple effect by influencingtial factors of soil erosion.
Response 28: Thank you for your kind suggestion. We have revised the title in Section 3.2.1 according to your kind suggestion in this revised manuscript, which can be found in line 261.
Point 29: [line 287] 3.2.2. Effect degree Analysis on of single interaction effect of factors of soil erosion.
Response 29: Thank you for your kind suggestion. We have revised the title in Section 3.2.2 according to your kind suggestion in this revised manuscript, which can be found in line 287.
Point 30: [line 288] Interactive detection Results of detection of interactions show
Response 30: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in line 288.
Point 31: [lines 314-315] Based on the principle of risk detection, high-risk stratification of influencing factors by high-risk of soil erosion (soil erosion modulus ≥ 15000 t/km2·a) of soil erosion can be determined
Response 31: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 314-315.
Point 32: [line 340] 3.3. Temporal analysis on dominant factors of soil erosion.
I would recommend to explain if the temporal analysis is based on only two dates of observation (2010 and 2017) and explain how the correlation between precipitation and soil erosion mentioned in the closing sentence of the section has been evaluated (in my understanding, correlation in spatial terms has already been considered in the previous section 3.2.1, and in temporal ones would be meaningless when just two points in time are available).
Response 32: Thank you for your kind suggestion. The reasons for using only two dates of observation (2010 and 2017) are as follows: land-use type and vegetation coverage were the key factors affecting soil erosion in this paper. From 2010 to 2017, the area of forest land, cultivated land and unused land continued to increase, while the area of grassland continued to decrease, and the vegetation coverage showed corresponding changes; Wolman.MG (1967) has revealed the relationship between the process of urban construction and soil erosion-the intensity of soil erosion often increased significantly during the onset of urbanization and then decreased thereafter. Between 2000 and 2017, the urban population of Yan'an City has more than tripled, and the urbanization level has jumped from 20.84% in 2000 to 60.79% in 2017; The study area was in the mid-term development stage of urbanization. It can be seen that changes in land-use types, vegetation coverage and population density were continuous.
Therefore, choosing only two dates of observation (2010 and 2017) to study the correlation between influential factors and soil erosion has a certain degree of reliability.
The question about the correlation between precipitation and soil erosion was carefully considered. The following tables (Table5-6) can only illustrate the changes of the explanatory power of influential factors on soil erosion, but not enough to demonstrate the correlation between them. Thus, we corrected the incorrect expression of the relationship between precipitation and soil erosion (lines 346-8). It should be noted that precipitation has annual variability. Here we only illustrate the changes of the explanatory power of precipitation on soil erosion at two dates in 2010 and 2017.
Point 33: [line 368] has been greatly improved reduced
Response 33: Thank you for your kind suggestion. We have replaced "improved" with "reduced" in this revised manuscript, which can be found in line 368.
Point 34: [lines 377-378] It was Intervention to alleviate soil erosion have focused on vegetation and slope gradients that have been taken seriously in the study area.
Response 34: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 377-378.
Point 35: [lines 386-388] was compared with previously published results studies. Authors of a similar study [49] estimated the average erosion modulus
Response 35: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 386-388.
Point 36: [lines 388-390] Recently, another study [50] suggested that
Response 36: Thank you for your kind suggestion. We have revised this type of error according to your kind suggestion in this revised manuscript, which can be found in line 388-390.
Point 37:[ lines 392-393] Therefore, the results of the RUSLE model in the study area have certain are attached to good reliability.
Response 37: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 392-393.
Point 38: [lines 411-414] And Moreover, precipitation data with the macroscopic spatial distribution and less microscopic differences brought by fewer sampling points, and as well as the interpolation techniques adopted also weaken the explanatory power of this factor on soil erosion.
Response 38: Thank you for your kind suggestion. We have revised the sentences according to your kind suggestion in this revised manuscript, which can be found in lines 411-414.
Point 39: [lines 419-24] please, clarify the difference observed in the results.
Response 39: Thank you for your kind suggestion. We corrected the contents in this revised manuscript to improve the clarity and readability of the manuscript, which can be found in lines 420-424 .
Point 40: [lines 424-426] please, specify which statistics supply the information reported and the types the two figure are related.
Response 40: Thank you for your kind suggestion. The types of the two figures are urban land and cultivated land in mid-elevation plains. This manuscript has added Table S5 (titled "Average erosion modulus of different land-use types in mid-elevation plains (t/km2·a).") to increase the clarity and readability of the manuscript.
Point 41: [line 454] pflat platforms
Response 41: Thank you for your kind suggestion. We have revised the wrong word according to your kind suggestion in this revised manuscript, which can be found in line 454.
Point 42: [line 456] singler factors
Response 42: Thank you for your kind suggestion. We have revised the wrong word according to your kind suggestion in this revised manuscript, which can be found in line 456.
We acknowledge your comments and suggestions very much, which are valuable in improving the quality of our manuscript.
Reviewer 2 Report
Dear authors,
your paper is focused on a very interesting and debated issue, i.e., soil erosion.
Starting from the existing model, equation and tool, the paper searches for implementing some novelties (e.g., Geo-detector Model). The paper offers an appropriate overview of the critical conditions of the investigated areas while highlighting a considerable improvement from 2010 to 2017.
I suggest to review accurately the text to remove lexical inaccuracies and some minor methodological gaps. In the attached pdf I have indicated most of the detected formal and methodological gaps.
kind regards
Comments for author File: 
Comments.pdf
Author Response
Thank you very much for your comments.
A detailed response to the reviewer 2
Thank you very much for the reviewers’ evaluation and comments on our paper. Your comments were highly insightful and enabled us to greatly improve the quality of our manuscript. We have substantially revised our manuscript after reading the comments provided by the reviewer 2. The explanation of what we have changed in response to the reviewers’ concerns is given point by point in the following pages.
Response to Reviewer 2 Comments
Point 1: [line 11] please remove these words that sound confusing
Response 1: Thank you for your kind suggestion. We have remove these words according to your kind suggestion in this revised manuscript, which can be found in line 11.
Point 2: [line 29] please remove the word problem which is redundant
Response2: Thank you for your kind suggestion. We have remove the word according to your kind suggestion in this revised manuscript, which can be found in line 29.
Point 3: [line 31]for a more ecologically sound development
Response 3: Thank you for your kind suggestion. We have replaced "for the more ecologically sound development" with "for a more ecologically sound development" in this revised manuscript, which can be found in line 31.
Point 4: [line 41]healthy
Response 4: Thank you for your kind suggestion. We have replaced "healthier" with "healthy" in this revised manuscript, which can be found in line 41.
Point 5: [line 48]the verb "lead" takes also this clause. Please remove them
Response 5: Thank you for your kind suggestion. We have remove the word according to your kind suggestion in this revised manuscript, which can be found in line 48.
Point 6: [lines 52, 66, 69, 116, 388 and 390]At the beginning of a sentence the citation shuold be placed in a different way. Please, read the guideline
Response 6: Thank you for your kind suggestion. We have revised this type of error according to your kind suggestion in this revised manuscript, which can be found in lines 52, 66, 69, 116, 388 and 390.
Point 7: [lines 50-51]from 98.0% to 99.9%
Response 7: Thank you for your kind suggestion. We have added the word “from” according to your kind suggestion in this revised manuscript, which can be found in lines 50-51.
Point 8: [lines 51-52]can greatly intensify soil loss from 2.1 to 15.6 times.
Response 8: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 51-52.
Point 9: [lines 55-56]To make the sense of sentence more clear, I suggest To effectively reduce soil erosion, the following key and difficult points need to be answered: "
Response 9: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 55-56.
Point 10: [lines 94-98] I am perplexed about the use of the future in these sentences
Response 10: Thank you for your kind suggestion. We have replaced simple future tense with simple past tense according to your kind suggestion in this revised manuscript, which can be found in lines 94-98.
Point 11: [lines 102-103]are located at relatively high elevations
Response 11: Thank you for your kind suggestion. We have revised the sentence according to your kind suggestion in this revised manuscript, which can be found in lines 102-103.
Point 12: [lines 104-106]please verify the adjective for loss, i.e., the clastic and predominantly silt-sized sediment that is formed by the accumulation of wind-blown dust
Response 12: Thank you for your kind suggestion. We have added the characterization of loessal soil according to your kind suggestion in this revised manuscript, which can be found in lines 104-106.
Point 13: [line 122]please, define them
Response 13: Thank you for your kind suggestion. We have added the description of terrain niche index to improving understanding according to your kind suggestion in this revised manuscript, which can be found in line 122.
Point 14: [line 129]In this list of used information, the word "data" is cited at the beginning: there is no need to add them for all the enumerated elements
Response 14: Thank you for your kind suggestion. We have deleted the sentence “The data used in this analysis include daily precipitation data, selected soil parameter data, Digital Elevation Model (DEM) data, Normalized Difference Vegetation Index (NDVI) data, the land-use and coverage data, population density data, geomorphic distribution data, and vector boundary data.” according to your kind suggestion in this revised manuscript, which can be found in line 129.
Point 15: [line 156]Please, provide the meaning of this acronym.
Response 15: Thank you for your kind suggestion. We have added the description of EI30 to improving understanding according to your kind suggestion in this revised manuscript, which can be found in line 156.
Point 16: [line 235]Minor
Response 16: Thank you for your kind suggestion. We have revised the wrong word according to your kind suggestion in this revised manuscript, which can be found in line 235.
Point 17: [lines 238]result
Response 17: Thank you for your kind suggestion. We have replaced “characteristic” with “result” according to your kind suggestion in this revised manuscript, which can be found in line 238.
Point 18: [lines 238, 244]Slight
Response 18: Thank you for your kind suggestion. We have revised the wrong word according to your kind suggestion in this revised manuscript, which can be found in lines 238 and 244.
Point 19: [line 385]carry out
Response 19: Thank you for your kind suggestion. We have replaced “detect” with “carry out” according to your kind suggestion in this revised manuscript, which can be found in line 385.
Point 20: [line 423]full stop
Response 20: Thank you for your kind suggestion. We have replaced “;” with “.” according to your kind suggestion in this revised manuscript, which can be found in line 423.
Point 21: [line 425]lower
Response 21: Thank you for your kind suggestion. We have replaced “slighter” with “lower” according to your kind suggestion in this revised manuscript, which can be found in line 425.
Point 22: [line 429]in soil erosion studies
Response 22: Thank you for your kind suggestion. We have added the word “studies” according to your kind suggestion in this revised manuscript, which can be found in line 429.
Point 23: [line 437]please, remove "as follows"
Response 23: Thank you for your kind suggestion. We have removed the words “as follows” according to your kind suggestion in this revised manuscript, which can be found in line 437.
We acknowledge your comments and suggestions very much, which are valuable in improving the quality of our manuscript.
