Spatial Analysis of the Suitability of Hass Avocado Cultivation in the Cauca Department, Colombia, Using Multi-Criteria Decision Analysis and Geographic Information Systems

Reviewer/

Editor

Reviewer's

Suggestions

Author's Corrections

Review Report 1 (Round 2)

Key results and conclusions are required.

Accepted: Thank you for your comments on the scientific paper. We accept the necessary correction and strive to include the key findings and conclusions in the abstract. We appreciate your time and effort to help us improve our work.

Rephrase the sentence for better understanding.

Accepted: The sentence was rewritten for better understanding.

This seems to be the part of methodology and aim of this study. This section demands the reflection of key conclusions made based on results ans discussion.

Accepted: While we recognise that there are various ways to write the conclusion section, we believe that the most accurate way is to first provide an executive summary of the main findings to put the reader in context. This is reported in several article writing guidelines. We have therefore rewritten this first paragraph so that it does not feel like a simple summary, but rather highlights our findings.

Review Report 2 (Round 2)

1. In phase 3 of figure 1, it shown that in weighting factor using FAHP-Fuzzy AHP? but in the description just using ordinary AHP, which is the correct one? Some other papers uses fuzzy AHP rather than just ordinary AHP.

Accepted: Thank you for your comments on our paper. We have made the change to indicate that the methodology used is AHP, not FAHP. While there is FAHP, we note that the changes in crop zoning are not significant. We appreciate your time and effort in helping us improve our work. If there are any further clarification you need, please let us know.

How many experts involved in weight evaluation by using AHP? How the experts evaluation were aggregated should be described in the paper

Accepted: A comprehensive description was made for the expert assessment with the factors.

It will be good if a sub-chapter, i.e 4.2. Managerial Implication --- be added in this paper, discussing the follow up actions for implementing the study results.

Not Accepted: From a rigorous and technical perspective, we appreciate the reviewer's suggestion to include a sub-chapter on management implications. However, its inclusion is not appropriate in the context of our study for several fundamental reasons.

First, adequately addressing management implications would require considering relevant factors such as existing regulations and public policies, which would significantly increase the complexity and scope of our study. Furthermore, the inclusion of a sub-chapter on management implications may lead to a misperception of the purpose and nature of our work, presenting it as a public policy report rather than a rigorous scientific study based on replicable results.

Therefore, to maintain the scientific and technical integrity of our study and ensure that the results obtained are consistent with our main objective, we consider excluding this sub-section appropriate. However, we appreciate the reviewer's suggestion and are open to considering other ways of including management implications in future work.

Review Report 3 (Round 2)

11-      The abstract must be more informative about the used method and data sets. Which method you use? Which data sets were used as input? In addition, ıt should contain the numerical results of accuracies or so on.

Accepted: The abstract was modified to better detail the materials and methods used in the research, taking into account that it is within the abstract.

22-      Introduction part should contain literature surveys (i.e. previous studies). Therefore, there is no need to separate “related works”. It must be a part of introduction.

Not Accepted: We appreciate the feedback provided by the reviewer. We understand that there are varying approaches to writing a scientific article, however, we believe that including a separate section for related works is crucial for effectively presenting our work and highlighting its distinctiveness. This approach also helps to maintain the clarity and scientific rigor of the article by avoiding an overly lengthy introduction section.

33-      In the study, different type of data sets were used such as vector, raster and attribute data sets. However, there are no any information about the scale or pixel size of the used data sets. This is very important for GIS studies. These factor affects the output scale. The name of data sets are not enough. Please explain the type of data sets with their scales in the text.

The input data used in the study consists of the following:

1.      Average annual temperature distribution obtained from Ideam (2014), with a multiannual average ranging from 1981-2010 and a scale of 1:100,000. The basic cartography was provided by IGAC (2012) with a raster information consisting of 8088 columns and 8736 rows, 1 band, a cell size of 30 (X) by 30 (Y), a pixel type of natural integer, and an 8-bit pixel depth.

2.      Average annual total rainfall, also obtained from Ideam (2014), with similar information to the temperature distribution.

3.      Average annual daily solar brightness obtained from Ideam (2014), with the same information as the temperature and rainfall distributions.

4.      Slope information was obtained from IGAC (2014) with a national soil correlation map with a scale of 1:100,000 and a digital terrain model with a spatial resolution of 90 m. The raster information consists of 7222 columns and 7181 rows, 1 band, a cell size of 31.2 (X) by 31.2 (Y), a pixel type of marked integer, and a 16-bit pixel depth.

5.      Road information was obtained from Mintransporte (2014) and IGAC (2012), with the sections of the national and departmental road network, with a scale of 1:100,000. The raster information consists of 1694 columns and 1903 rows, 1 band, a cell size of 137 (X) by 137 (Y), a pixel type of marked integer, and a 32-bit pixel depth.

44-      The authors used 5 factors to generate suitability map of avocado production. One of the factors, slope aspect, is also related to cultivation. Why this factor could not join in the analysis? Please consider this comment.

If it was included in the analysis, in fact it is one of the main factors used to determine the areas of suitability for cultivation.

55-      Classes of the input data sets are problematic.  For example, Average annual temperature are divided into 4 classes which are 15-18, 18-20, 13-15 and <13->20 in table 2. What is the last class? If the annual temperature is 18. So which classes is? Is it in high or Medium? Similarly, precipitation class has duplicated values. 2000mm/year is highly suitable or medium? Please arrange the classes and define clearly class values in the related classes. Please revised this table.

The temperature parameters were adjusted as follows:

Class 1: High ( ⩾ 15- < 18)

Class 2: Medium ( ⩾ 18- < 20)

Class 3: Low ( ⩾ 13- <15)

Class 4: Not suitable ( < 13- > 20)

- What is the last class? à It is the class "Not suitable".

- If the annual temperature is 18. Then which class is it? à If the temperature is 18, it belongs to class 2 ("Medium").

- The other ranges of the other factors were adjusted in the same way.

66-      Table 3 has same problem with adjacent classes. What is slope value 440? Is it percent? Is it degree? There is no any explanation. In addition, the table 3 is so strange table?? The average annual temperature is starting from 8 to 28, slope is starting from 0 to 440, and proximity also have minimum and maximum values. But precipitation seems as a loop. It is starting from a range between 1000 and 1500 and final value has a range between 1500 and 2000. It is very confusing. Please prepare more comprehensive table. And also arrange junction values (i.e. duplicated values).

·        What is the value of slope 440? Is it in percent? or degree? - A new column was created in Table 3 that displays the dimensional unit of each factor.

·        The average annual temperature ranges from 8 to 28, slope ranges from 0 to 440, and proximity also has minimum and maximum values. However, precipitation appears to be a loop. It starts with a range between 1000 and 1500 and the final value has a range between 1500 and 2000 - The precipitation factor parameters were ordered to facilitate the reader's understanding.

• And also arrange the bond values (i.e., duplicate values) - The table was reorganized considering the bond values of the ranges.

77-      Input data sets already classified into 4 classes which are high, moderate, low and not suitable. This means your inputs are categorical data sets. Therefore the quality of the output seems as categorical and non-realistic. Figure 3 is the resultant map of suitability for avocado cultivation. The boundaries of the suitability classes are drastically changed between extreme classes which are highly suitable and not suitable. It is not so logical to change of suitable and not suitable changes in a very close areas such as Totoro, North of Almaguer, south of Bolivar etc. The almost half of the weight of the input data sets is affected by temperature. Do you think the temperature is so changes in a short distance? Why do not use continuous scales for all inputs? In that case more realistic results can be obtained. In categorical input case you already judge the suitability of the data sets before starting to MCDA analysis. Please consider this crucial comment.

Not Accepted: We appreciate your comments and feedback on the article. We understand your concerns about the quality of the results obtained in the study. However, we would like to clarify a few points.

In the study, the input data were classified into four categories, which are high, moderate, low, and unfit. As a result, the quality of the output is presented as categorical. However, considering the rugged topography of the study region, it is straightforward to go from flat terrain to a peak, which directly affects the temperature. For example, in the case of Totoró, which is a cold area, this condition directly classifies it as unsuitable. On the other hand, in Almaguer, a warm zone, the temperature increases, so it is classified as an optimal zone. Indeed, in the study region, the temperature changes over short distances.

The reason why continuous variables were not used for all inputs is that the original data come from sources that have categorical variables. Attempting to convert them to continuous variables could affect the results.

We thank you again for your comment and are open to considering it in future studies.

88-      According to AHP matrix, Slope is more important than solar brightness. And why proximity of roads are more important factor than the slope? It is not logical. According to MCDA, slope should be one of the major environmental factors. Roads are manmade structures which is related about logistics. Therefore, please check the AHP matrix objectively.

Thank you for your feedback on the article. We would like to address your concerns regarding the AHP matrix.

As per the AHP matrix, slope and solar exposure have been determined to be of equal importance, as they both have a score of 1. However, the experts have noted that proximity to roads is five times more important than slope. The reason for this is that slope can be managed by implementing suitable facilities such as cables to harvest the fruit. These facilities have feasible costs. On the other hand, the cost of adapting type 1 and 2 roads to the crop is not within the control of the grower and depends on the actions of governmental bodies. Additionally, investment in road infrastructure is much more costly.

Temperature has been identified as the most critical environmental factor. If the area does not meet the appropriate parameters for Hass avocado cultivation, the tree may not survive, and if it does, its production will be nil.

We hope this information clarifies any confusion and provides a better understanding of the AHP matrix.

99-       I have not ever seen such a strange scale of the map which is “1:427.561,44”. How can I arrange this? Please revise the figure 3.

Accepted: The correct scale is 1:100.000

110-   The discussion part must contain advantageous and disadvantageous of your study. Please improve this part.

Accepted: Thank you for your comment. We have taken your suggestion on board and have improved the discussion section to include a detailed and accurate assessment of the advantages and disadvantages of our study. We hope that this improvement reflects our commitment to appropriate technical and scientific standards and ensures a rigorous and accurate assessment of our work. We appreciate the opportunity to improve and hope that we have met your expectations.

111-   In literature, there are such this type of suitability studies as you already cited some of them. So, what are the differences between the existing studies and your study? What is the novelty of your study? Please explain in the manuscript.

Accepted: We appreciate your comment and the opportunity to clarify the novelty and differences between our study and previous suitability studies. To our knowledge, no previous studies have applied engineering techniques to determine the ideal regions for Hass avocado cultivation in Colombia (taking into account that it is a country with little productive development in terms of avocado). Therefore, the main contribution of our research is to provide empirical evidence on the feasibility of different zones for Hass avocado cultivation in Cauca.

Farmers and policymakers can use our results to inform resource allocation and develop strategies to improve the sustainability of avocado production in the region. This information is valuable and unique, and therefore hope it will be of interest to the scientific community.

We add this information in the related work section to clarify the novelty and differences between our study and previous literature. Thank you again for your valuable comment and your contribution to our work.

Reviewer/

Editor

Reviewer's

Suggestions

Author's Corrections

Review Report 1 (Round 2)

Key results and conclusions are required.

Accepted: Thank you for your comments on the scientific paper. We accept the necessary correction and strive to include the key findings and conclusions in the abstract. We appreciate your time and effort to help us improve our work.

Rephrase the sentence for better understanding.

Accepted: The sentence was rewritten for better understanding.

This seems to be the part of methodology and aim of this study. This section demands the reflection of key conclusions made based on results ans discussion.

Accepted: While we recognise that there are various ways to write the conclusion section, we believe that the most accurate way is to first provide an executive summary of the main findings to put the reader in context. This is reported in several article writing guidelines. We have therefore rewritten this first paragraph so that it does not feel like a simple summary, but rather highlights our findings.

Review Report 2 (Round 2)

1. In phase 3 of figure 1, it shown that in weighting factor using FAHP-Fuzzy AHP? but in the description just using ordinary AHP, which is the correct one? Some other papers uses fuzzy AHP rather than just ordinary AHP.

Accepted: Thank you for your comments on our paper. We have made the change to indicate that the methodology used is AHP, not FAHP. While there is FAHP, we note that the changes in crop zoning are not significant. We appreciate your time and effort in helping us improve our work. If there are any further clarification you need, please let us know.

How many experts involved in weight evaluation by using AHP? How the experts evaluation were aggregated should be described in the paper

Accepted: A comprehensive description was made for the expert assessment with the factors.

It will be good if a sub-chapter, i.e 4.2. Managerial Implication --- be added in this paper, discussing the follow up actions for implementing the study results.

Not Accepted: From a rigorous and technical perspective, we appreciate the reviewer's suggestion to include a sub-chapter on management implications. However, its inclusion is not appropriate in the context of our study for several fundamental reasons.

First, adequately addressing management implications would require considering relevant factors such as existing regulations and public policies, which would significantly increase the complexity and scope of our study. Furthermore, the inclusion of a sub-chapter on management implications may lead to a misperception of the purpose and nature of our work, presenting it as a public policy report rather than a rigorous scientific study based on replicable results.

Therefore, to maintain the scientific and technical integrity of our study and ensure that the results obtained are consistent with our main objective, we consider excluding this sub-section appropriate. However, we appreciate the reviewer's suggestion and are open to considering other ways of including management implications in future work.

Review Report 3 (Round 2)

11-      The abstract must be more informative about the used method and data sets. Which method you use? Which data sets were used as input? In addition, ıt should contain the numerical results of accuracies or so on.

Accepted: The abstract was modified to better detail the materials and methods used in the research, taking into account that it is within the abstract.

22-      Introduction part should contain literature surveys (i.e. previous studies). Therefore, there is no need to separate “related works”. It must be a part of introduction.

Not Accepted: We appreciate the feedback provided by the reviewer. We understand that there are varying approaches to writing a scientific article, however, we believe that including a separate section for related works is crucial for effectively presenting our work and highlighting its distinctiveness. This approach also helps to maintain the clarity and scientific rigor of the article by avoiding an overly lengthy introduction section.

33-      In the study, different type of data sets were used such as vector, raster and attribute data sets. However, there are no any information about the scale or pixel size of the used data sets. This is very important for GIS studies. These factor affects the output scale. The name of data sets are not enough. Please explain the type of data sets with their scales in the text.

The input data used in the study consists of the following:

1.      Average annual temperature distribution obtained from Ideam (2014), with a multiannual average ranging from 1981-2010 and a scale of 1:100,000. The basic cartography was provided by IGAC (2012) with a raster information consisting of 8088 columns and 8736 rows, 1 band, a cell size of 30 (X) by 30 (Y), a pixel type of natural integer, and an 8-bit pixel depth.

2.      Average annual total rainfall, also obtained from Ideam (2014), with similar information to the temperature distribution.

3.      Average annual daily solar brightness obtained from Ideam (2014), with the same information as the temperature and rainfall distributions.

4.      Slope information was obtained from IGAC (2014) with a national soil correlation map with a scale of 1:100,000 and a digital terrain model with a spatial resolution of 90 m. The raster information consists of 7222 columns and 7181 rows, 1 band, a cell size of 31.2 (X) by 31.2 (Y), a pixel type of marked integer, and a 16-bit pixel depth.

5.      Road information was obtained from Mintransporte (2014) and IGAC (2012), with the sections of the national and departmental road network, with a scale of 1:100,000. The raster information consists of 1694 columns and 1903 rows, 1 band, a cell size of 137 (X) by 137 (Y), a pixel type of marked integer, and a 32-bit pixel depth.

44-      The authors used 5 factors to generate suitability map of avocado production. One of the factors, slope aspect, is also related to cultivation. Why this factor could not join in the analysis? Please consider this comment.

If it was included in the analysis, in fact it is one of the main factors used to determine the areas of suitability for cultivation.

55-      Classes of the input data sets are problematic.  For example, Average annual temperature are divided into 4 classes which are 15-18, 18-20, 13-15 and <13->20 in table 2. What is the last class? If the annual temperature is 18. So which classes is? Is it in high or Medium? Similarly, precipitation class has duplicated values. 2000mm/year is highly suitable or medium? Please arrange the classes and define clearly class values in the related classes. Please revised this table.

The temperature parameters were adjusted as follows:

Class 1: High ( ⩾ 15- < 18)

Class 2: Medium ( ⩾ 18- < 20)

Class 3: Low ( ⩾ 13- <15)

Class 4: Not suitable ( < 13- > 20)

- What is the last class? à It is the class "Not suitable".

- If the annual temperature is 18. Then which class is it? à If the temperature is 18, it belongs to class 2 ("Medium").

- The other ranges of the other factors were adjusted in the same way.

66-      Table 3 has same problem with adjacent classes. What is slope value 440? Is it percent? Is it degree? There is no any explanation. In addition, the table 3 is so strange table?? The average annual temperature is starting from 8 to 28, slope is starting from 0 to 440, and proximity also have minimum and maximum values. But precipitation seems as a loop. It is starting from a range between 1000 and 1500 and final value has a range between 1500 and 2000. It is very confusing. Please prepare more comprehensive table. And also arrange junction values (i.e. duplicated values).

·        What is the value of slope 440? Is it in percent? or degree? - A new column was created in Table 3 that displays the dimensional unit of each factor.

·        The average annual temperature ranges from 8 to 28, slope ranges from 0 to 440, and proximity also has minimum and maximum values. However, precipitation appears to be a loop. It starts with a range between 1000 and 1500 and the final value has a range between 1500 and 2000 - The precipitation factor parameters were ordered to facilitate the reader's understanding.

• And also arrange the bond values (i.e., duplicate values) - The table was reorganized considering the bond values of the ranges.

77-      Input data sets already classified into 4 classes which are high, moderate, low and not suitable. This means your inputs are categorical data sets. Therefore the quality of the output seems as categorical and non-realistic. Figure 3 is the resultant map of suitability for avocado cultivation. The boundaries of the suitability classes are drastically changed between extreme classes which are highly suitable and not suitable. It is not so logical to change of suitable and not suitable changes in a very close areas such as Totoro, North of Almaguer, south of Bolivar etc. The almost half of the weight of the input data sets is affected by temperature. Do you think the temperature is so changes in a short distance? Why do not use continuous scales for all inputs? In that case more realistic results can be obtained. In categorical input case you already judge the suitability of the data sets before starting to MCDA analysis. Please consider this crucial comment.

Not Accepted: We appreciate your comments and feedback on the article. We understand your concerns about the quality of the results obtained in the study. However, we would like to clarify a few points.

In the study, the input data were classified into four categories, which are high, moderate, low, and unfit. As a result, the quality of the output is presented as categorical. However, considering the rugged topography of the study region, it is straightforward to go from flat terrain to a peak, which directly affects the temperature. For example, in the case of Totoró, which is a cold area, this condition directly classifies it as unsuitable. On the other hand, in Almaguer, a warm zone, the temperature increases, so it is classified as an optimal zone. Indeed, in the study region, the temperature changes over short distances.

The reason why continuous variables were not used for all inputs is that the original data come from sources that have categorical variables. Attempting to convert them to continuous variables could affect the results.

We thank you again for your comment and are open to considering it in future studies.

88-      According to AHP matrix, Slope is more important than solar brightness. And why proximity of roads are more important factor than the slope? It is not logical. According to MCDA, slope should be one of the major environmental factors. Roads are manmade structures which is related about logistics. Therefore, please check the AHP matrix objectively.

Thank you for your feedback on the article. We would like to address your concerns regarding the AHP matrix.

As per the AHP matrix, slope and solar exposure have been determined to be of equal importance, as they both have a score of 1. However, the experts have noted that proximity to roads is five times more important than slope. The reason for this is that slope can be managed by implementing suitable facilities such as cables to harvest the fruit. These facilities have feasible costs. On the other hand, the cost of adapting type 1 and 2 roads to the crop is not within the control of the grower and depends on the actions of governmental bodies. Additionally, investment in road infrastructure is much more costly.

Temperature has been identified as the most critical environmental factor. If the area does not meet the appropriate parameters for Hass avocado cultivation, the tree may not survive, and if it does, its production will be nil.

We hope this information clarifies any confusion and provides a better understanding of the AHP matrix.

99-       I have not ever seen such a strange scale of the map which is “1:427.561,44”. How can I arrange this? Please revise the figure 3.

Accepted: The correct scale is 1:100.000

110-   The discussion part must contain advantageous and disadvantageous of your study. Please improve this part.

Accepted: Thank you for your comment. We have taken your suggestion on board and have improved the discussion section to include a detailed and accurate assessment of the advantages and disadvantages of our study. We hope that this improvement reflects our commitment to appropriate technical and scientific standards and ensures a rigorous and accurate assessment of our work. We appreciate the opportunity to improve and hope that we have met your expectations.

111-   In literature, there are such this type of suitability studies as you already cited some of them. So, what are the differences between the existing studies and your study? What is the novelty of your study? Please explain in the manuscript.

Accepted: We appreciate your comment and the opportunity to clarify the novelty and differences between our study and previous suitability studies. To our knowledge, no previous studies have applied engineering techniques to determine the ideal regions for Hass avocado cultivation in Colombia (taking into account that it is a country with little productive development in terms of avocado). Therefore, the main contribution of our research is to provide empirical evidence on the feasibility of different zones for Hass avocado cultivation in Cauca.

Farmers and policymakers can use our results to inform resource allocation and develop strategies to improve the sustainability of avocado production in the region. This information is valuable and unique, and therefore hope it will be of interest to the scientific community.

We add this information in the related work section to clarify the novelty and differences between our study and previous literature. Thank you again for your valuable comment and your contribution to our work.

Reviewer/

Editor

Reviewer's

Suggestions

Author's Corrections

Review Report 1 (Round 2)

Key results and conclusions are required.

Accepted: Thank you for your comments on the scientific paper. We accept the necessary correction and strive to include the key findings and conclusions in the abstract. We appreciate your time and effort to help us improve our work.

Rephrase the sentence for better understanding.

Accepted: The sentence was rewritten for better understanding.

This seems to be the part of methodology and aim of this study. This section demands the reflection of key conclusions made based on results ans discussion.

Accepted: While we recognise that there are various ways to write the conclusion section, we believe that the most accurate way is to first provide an executive summary of the main findings to put the reader in context. This is reported in several article writing guidelines. We have therefore rewritten this first paragraph so that it does not feel like a simple summary, but rather highlights our findings.

Review Report 2 (Round 2)

1. In phase 3 of figure 1, it shown that in weighting factor using FAHP-Fuzzy AHP? but in the description just using ordinary AHP, which is the correct one? Some other papers uses fuzzy AHP rather than just ordinary AHP.

Accepted: Thank you for your comments on our paper. We have made the change to indicate that the methodology used is AHP, not FAHP. While there is FAHP, we note that the changes in crop zoning are not significant. We appreciate your time and effort in helping us improve our work. If there are any further clarification you need, please let us know.

How many experts involved in weight evaluation by using AHP? How the experts evaluation were aggregated should be described in the paper

Accepted: A comprehensive description was made for the expert assessment with the factors.

It will be good if a sub-chapter, i.e 4.2. Managerial Implication --- be added in this paper, discussing the follow up actions for implementing the study results.

Not Accepted: From a rigorous and technical perspective, we appreciate the reviewer's suggestion to include a sub-chapter on management implications. However, its inclusion is not appropriate in the context of our study for several fundamental reasons.

First, adequately addressing management implications would require considering relevant factors such as existing regulations and public policies, which would significantly increase the complexity and scope of our study. Furthermore, the inclusion of a sub-chapter on management implications may lead to a misperception of the purpose and nature of our work, presenting it as a public policy report rather than a rigorous scientific study based on replicable results.

Therefore, to maintain the scientific and technical integrity of our study and ensure that the results obtained are consistent with our main objective, we consider excluding this sub-section appropriate. However, we appreciate the reviewer's suggestion and are open to considering other ways of including management implications in future work.

Review Report 3 (Round 2)

11-      The abstract must be more informative about the used method and data sets. Which method you use? Which data sets were used as input? In addition, ıt should contain the numerical results of accuracies or so on.

Accepted: The abstract was modified to better detail the materials and methods used in the research, taking into account that it is within the abstract.

22-      Introduction part should contain literature surveys (i.e. previous studies). Therefore, there is no need to separate “related works”. It must be a part of introduction.

Not Accepted: We appreciate the feedback provided by the reviewer. We understand that there are varying approaches to writing a scientific article, however, we believe that including a separate section for related works is crucial for effectively presenting our work and highlighting its distinctiveness. This approach also helps to maintain the clarity and scientific rigor of the article by avoiding an overly lengthy introduction section.

33-      In the study, different type of data sets were used such as vector, raster and attribute data sets. However, there are no any information about the scale or pixel size of the used data sets. This is very important for GIS studies. These factor affects the output scale. The name of data sets are not enough. Please explain the type of data sets with their scales in the text.

The input data used in the study consists of the following:

1.      Average annual temperature distribution obtained from Ideam (2014), with a multiannual average ranging from 1981-2010 and a scale of 1:100,000. The basic cartography was provided by IGAC (2012) with a raster information consisting of 8088 columns and 8736 rows, 1 band, a cell size of 30 (X) by 30 (Y), a pixel type of natural integer, and an 8-bit pixel depth.

2.      Average annual total rainfall, also obtained from Ideam (2014), with similar information to the temperature distribution.

3.      Average annual daily solar brightness obtained from Ideam (2014), with the same information as the temperature and rainfall distributions.

4.      Slope information was obtained from IGAC (2014) with a national soil correlation map with a scale of 1:100,000 and a digital terrain model with a spatial resolution of 90 m. The raster information consists of 7222 columns and 7181 rows, 1 band, a cell size of 31.2 (X) by 31.2 (Y), a pixel type of marked integer, and a 16-bit pixel depth.

5.      Road information was obtained from Mintransporte (2014) and IGAC (2012), with the sections of the national and departmental road network, with a scale of 1:100,000. The raster information consists of 1694 columns and 1903 rows, 1 band, a cell size of 137 (X) by 137 (Y), a pixel type of marked integer, and a 32-bit pixel depth.

44-      The authors used 5 factors to generate suitability map of avocado production. One of the factors, slope aspect, is also related to cultivation. Why this factor could not join in the analysis? Please consider this comment.

If it was included in the analysis, in fact it is one of the main factors used to determine the areas of suitability for cultivation.

55-      Classes of the input data sets are problematic.  For example, Average annual temperature are divided into 4 classes which are 15-18, 18-20, 13-15 and <13->20 in table 2. What is the last class? If the annual temperature is 18. So which classes is? Is it in high or Medium? Similarly, precipitation class has duplicated values. 2000mm/year is highly suitable or medium? Please arrange the classes and define clearly class values in the related classes. Please revised this table.

The temperature parameters were adjusted as follows:

Class 1: High ( ⩾ 15- < 18)

Class 2: Medium ( ⩾ 18- < 20)

Class 3: Low ( ⩾ 13- <15)

Class 4: Not suitable ( < 13- > 20)

- What is the last class? à It is the class "Not suitable".

- If the annual temperature is 18. Then which class is it? à If the temperature is 18, it belongs to class 2 ("Medium").

- The other ranges of the other factors were adjusted in the same way.

66-      Table 3 has same problem with adjacent classes. What is slope value 440? Is it percent? Is it degree? There is no any explanation. In addition, the table 3 is so strange table?? The average annual temperature is starting from 8 to 28, slope is starting from 0 to 440, and proximity also have minimum and maximum values. But precipitation seems as a loop. It is starting from a range between 1000 and 1500 and final value has a range between 1500 and 2000. It is very confusing. Please prepare more comprehensive table. And also arrange junction values (i.e. duplicated values).

·        What is the value of slope 440? Is it in percent? or degree? - A new column was created in Table 3 that displays the dimensional unit of each factor.

·        The average annual temperature ranges from 8 to 28, slope ranges from 0 to 440, and proximity also has minimum and maximum values. However, precipitation appears to be a loop. It starts with a range between 1000 and 1500 and the final value has a range between 1500 and 2000 - The precipitation factor parameters were ordered to facilitate the reader's understanding.

• And also arrange the bond values (i.e., duplicate values) - The table was reorganized considering the bond values of the ranges.

77-      Input data sets already classified into 4 classes which are high, moderate, low and not suitable. This means your inputs are categorical data sets. Therefore the quality of the output seems as categorical and non-realistic. Figure 3 is the resultant map of suitability for avocado cultivation. The boundaries of the suitability classes are drastically changed between extreme classes which are highly suitable and not suitable. It is not so logical to change of suitable and not suitable changes in a very close areas such as Totoro, North of Almaguer, south of Bolivar etc. The almost half of the weight of the input data sets is affected by temperature. Do you think the temperature is so changes in a short distance? Why do not use continuous scales for all inputs? In that case more realistic results can be obtained. In categorical input case you already judge the suitability of the data sets before starting to MCDA analysis. Please consider this crucial comment.

Not Accepted: We appreciate your comments and feedback on the article. We understand your concerns about the quality of the results obtained in the study. However, we would like to clarify a few points.

In the study, the input data were classified into four categories, which are high, moderate, low, and unfit. As a result, the quality of the output is presented as categorical. However, considering the rugged topography of the study region, it is straightforward to go from flat terrain to a peak, which directly affects the temperature. For example, in the case of Totoró, which is a cold area, this condition directly classifies it as unsuitable. On the other hand, in Almaguer, a warm zone, the temperature increases, so it is classified as an optimal zone. Indeed, in the study region, the temperature changes over short distances.

The reason why continuous variables were not used for all inputs is that the original data come from sources that have categorical variables. Attempting to convert them to continuous variables could affect the results.

We thank you again for your comment and are open to considering it in future studies.

88-      According to AHP matrix, Slope is more important than solar brightness. And why proximity of roads are more important factor than the slope? It is not logical. According to MCDA, slope should be one of the major environmental factors. Roads are manmade structures which is related about logistics. Therefore, please check the AHP matrix objectively.

Thank you for your feedback on the article. We would like to address your concerns regarding the AHP matrix.

As per the AHP matrix, slope and solar exposure have been determined to be of equal importance, as they both have a score of 1. However, the experts have noted that proximity to roads is five times more important than slope. The reason for this is that slope can be managed by implementing suitable facilities such as cables to harvest the fruit. These facilities have feasible costs. On the other hand, the cost of adapting type 1 and 2 roads to the crop is not within the control of the grower and depends on the actions of governmental bodies. Additionally, investment in road infrastructure is much more costly.

Temperature has been identified as the most critical environmental factor. If the area does not meet the appropriate parameters for Hass avocado cultivation, the tree may not survive, and if it does, its production will be nil.

We hope this information clarifies any confusion and provides a better understanding of the AHP matrix.

99-       I have not ever seen such a strange scale of the map which is “1:427.561,44”. How can I arrange this? Please revise the figure 3.

Accepted: The correct scale is 1:100.000

110-   The discussion part must contain advantageous and disadvantageous of your study. Please improve this part.

Accepted: Thank you for your comment. We have taken your suggestion on board and have improved the discussion section to include a detailed and accurate assessment of the advantages and disadvantages of our study. We hope that this improvement reflects our commitment to appropriate technical and scientific standards and ensures a rigorous and accurate assessment of our work. We appreciate the opportunity to improve and hope that we have met your expectations.

111-   In literature, there are such this type of suitability studies as you already cited some of them. So, what are the differences between the existing studies and your study? What is the novelty of your study? Please explain in the manuscript.

Accepted: We appreciate your comment and the opportunity to clarify the novelty and differences between our study and previous suitability studies. To our knowledge, no previous studies have applied engineering techniques to determine the ideal regions for Hass avocado cultivation in Colombia (taking into account that it is a country with little productive development in terms of avocado). Therefore, the main contribution of our research is to provide empirical evidence on the feasibility of different zones for Hass avocado cultivation in Cauca.

Farmers and policymakers can use our results to inform resource allocation and develop strategies to improve the sustainability of avocado production in the region. This information is valuable and unique, and therefore hope it will be of interest to the scientific community.

We add this information in the related work section to clarify the novelty and differences between our study and previous literature. Thank you again for your valuable comment and your contribution to our work.