Suitable Habitat Prediction for African Wild Ass (Equus africanus) in the Danakil Desert of the Afar Region, Ethiopia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Please revise the manuscript according to the comments and suggestions. You can find the detaild of comments in the attached PDF file.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 1,

We (the authors) thank you for the useful suggestions and comments. Justifications have been made based on your comments. The changes made on the manuscript have been highlighted in yellow.

With best regards,

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The study holds substantial scientific value, addressing a highly relevant and pressing issue within contemporary conservation research: suitable habitat prediction for African wild ass (Equus africanus) in the Danakil Desert of Afar Region, Ethiopia.

The African wild ass (Equus africanus), is a critically endangered species native to the Danakil Desert in Ethiopia and Eritrea, which faces severe population decline in the last decades.

This study applied Maxent modeling to assess habitat suitability across dry and wet seasons using occurrence data and environmental variables. Key factors varied seasonally, with precipitation, temperature, and water access crucial during wet periods, while proximity to settlements and water shaped habitat suitability in dry conditions. The predictive models performed strongly, offering valuable insights into optimal conservation areas and guiding future ecological surveys in the Afar Region.

Several aspects of the manuscript warrant further clarification or supplementation:

1. In the Introduction, it would be valuable to include a brief overview of the population dynamics of Equus africanus over recent decades to provide readers with a clearer understanding of the extent of the population decline.
2. At line 76, we recommend the use of the term “locations” for improved precision and consistency.
3. In the Discussion section, a comparison with similar studies conducted on other herbivore species would be beneficial. Such an analysis could elucidate why vegetation cover (food availability), although a key limiting factor for most species, appears to play a relatively minor role in the tested habitat suitability model.

We also propose a stratified assessment of habitats based on favorability categories, with quantitative delineation of areas corresponding to optimal and moderately suitable habitats. Additionally, we recommend referencing these metrics to the total extent of the surveyed habitat patches to enable proportional analysis.

A comprehensive discussion on the evolution of the Maxent modeling approach is warranted, particularly in light of the methodological criteria outlined in the 2023 publication (line 588). This necessity arises from the fact that certain predictive variables have been substituted, while the relative weighting of others has undergone adjustment.

Author Response

Comments and suggestions:

This study applied Maxent modeling to assess habitat suitability across dry and wet seasons using occurrence data and environmental variables. Key factors varied seasonally, with precipitation, temperature, and water access crucial during wet periods, while proximity to settlements and water shaped habitat suitability in dry conditions. The predictive models performed strongly, offering valuable insights into optimal conservation areas and guiding future ecological surveys in the Afar Region.

Response to Reviewer 2,

Thank you very much for your thoughtful comments and positive feedback on our study. We are pleased that you found the seasonal application of the Maxent model and the identification of key environmental factors to be useful. Your recognition of the strong predictive performance and the conservation value of our findings is greatly appreciated. The changes made on the manuscript have been highlighted in yellow.

Comment 1: In the Introduction, it would be valuable to include a brief overview of the population dynamics of Equus africanus over recent decades to provide readers with a clearer understanding of the extent of the population decline.

Response 1: A sentence has been included about population dynamics of Equus africanus on line 43-48.

Comment 2: At line 76, we recommend the use of the term “locations” for improved precision and consistency.

Response 2: Justification has been made and the term “locations” is used on line 68 of this manuscript.

Comment 3: In the Discussion section, a comparison with similar studies conducted on other herbivore species would be beneficial. Such an analysis could elucidate why vegetation cover (food availability), although a key limiting factor for most species, appears to play a relatively minor role in the tested habitat suitability model.

Response 3: A sentence has been included about comparison with similar studies on other ungulate herbivores on line 324-327.

Comment 4: We also propose a stratified assessment of habitats based on favorability categories, with quantitative delineation of areas corresponding to optimal and moderately suitable habitats. Additionally, we recommend referencing these metrics to the total extent of the surveyed habitat patches to enable proportional analysis.

Response 4: The categories of delineation of areas corresponding to optimal and moderately suitable habitats were explained on line 227-231.

Comment 5: A comprehensive discussion on the evolution of the Maxent modeling approach is warranted, particularly in light of the methodological criteria outlined in the 2023 publication (line 588). This necessity arises from the fact that certain predictive variables have been substituted, while the relative weighting of others has undergone adjustment.

Response 5: Sentences have been included about Maxent modeling approach on line 204-210 and the predictor variable selection was explained on 2.3 predictor variable section and using the schematic illustration on fig.3.

We indeed appreciate your supportive comments and feedbacks.

 

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

Title: Suitable Habitat Prediction for African Wild Ass (Equus africanus) in the Danakil Desert of Afar Region, Ethiopia

 

General Overview

The manuscript presents an analysis of the best suitable areas for a critically endangered species of Equidae (African wild ass) that lives in Ethiopia. The species is down to approximately the last 600 individuals living in the wild, mostly in the Africa arid region of the Danakil desert that comprises Ethiopia, Eritrea and Djibouti. The authors applied a widely used method by modelling the species distribution through Maxent software for that purpose. They predict that the main factors driving their species´ choice of habitat is the distance to permanent water resources during the dry season and the amount of precipitation during the wet season. The English writing is good and clear, as is the overall manuscript structure. The authors also discuss the challenges this species may endure in the next years as it is struggling with habitat loss and competition with human livestock for water resources.

Detailed Review

Although Maxent is an admittedly well-suited tool for the purpose of identifying the factors that determine species distribution, it has some shortcomings that must be discussed. Maxent might produce a biased set of predictions if the factors are somewhat correlated (Lissovsky, A. A., & Dudov, S. V. (2021). Biology Bulletin Reviews, 11(3), 265-275.). The authors addressed this issue in the manuscript, albeit a matrix table expressing the collinearity between the predictor variables would be desirable. The manuscript would also benefit from a figure expressing the annual average month temperature and precipitation as both are considered important prediction factors by the authors.

Another important predictor variable that should be better presented is the dynamics of the human occupation in the area. The authors mention a reference with the very same purpose of this manuscript published 11 years ago, in the very same portion of the Danakil Desert (Kebede, F., Moehlman, P. D., Bekele, A., & Evangelista, P. H. (2014). Predicting seasonal habitat suitability for the critically endangered African wild ass in the Danakil, Ethiopia. African Journal of Ecology, 52(4), 533-542.). They emphasize in this manuscript that distance to human settlings is important to determine African wild ass choice of habitat as they tend to avoid these areas. They also discuss that the human occupation in the area has increased in the last 10 years, which reduced the habitat suitable habitats for the African wild ass. Unfortunately, no quantitative data was provided in this manuscript to support this statement. I recommend the authors to better describe the human occupation of the area in the last 10 years as it is a critical factor.

Regarding the African wild ass suitable habitats and the species distribution range, I recommend the authors to express in the manuscript how these animals move between Ethiopia and Eritrea as both countries share a border in the Danakil desert and African wild ass individuals are recorded in both countries. In fact, a very similar paper was published in 2023 discussing the best suitable areas for this species in Eritrea (Tesfai, R. T., Parrini, F., Moehlman, P. D., Young, N. E., & Evangelista, P. H. (2023). Predicting suitable habitat for the Critically Endangered African wild ass Equus africanus in the Danakil Desert of Eritrea. Oryx, 57(5), 592-599.). Unless there is some sort of fence in the border between these countries, these animals must transit unaware of any political border. Considering the critically endangered status of the species such movements should be addressed and be part of any conservationist effort.

Finally, it is important to consider that many animals distribution are determined not only by environment factors, but also by their biological interactions with other species. Such is that importance that many papers are now incorporating biological interactions in their prediction factors (1- Van der Putten, W. H., Macel, M., & Visser, M. E. (2010). Philosophical Transactions of the Royal Society B: Biological Sciences, 365(1549), 2025-2034;  2- Bateman, B. L., VanDerWal, J., Williams, S. E., & Johnson, C. N. (2012). Diversity and Distributions, 18(9), 861-872; 3- Lewin, A., Erinjery, J. J., de Waroux, Y. L. P., Tripler, E., & Iwamura, T. (2021). Journal of Arid Environments, 192, 104547 ; 4- Gherghel, I., Brischoux, F., & Papeş, M. (2018). Journal of Biogeography, 45(9), 2216-2225 ;   5- Cosentino, F., Seamark, E. C. J., Van Cakenberghe, V., & Maiorano, L. (2023). Ecology and Evolution, 13(3), e9855.)

Considering the human occupation of the area is increasing, the competition African wild ass is enduring with human livestock for water, one can predict that contact with diseases from domestic animals might be more frequent. Moreover, as long as the livestock increases in the area as do so the competition for the water resources, one might expect the increasing presence of natural predators in the proximity. Including these biological factors as predictor factors would enhance this manuscript greatly as the overall results presented so far do not differ much of the ones provided by a very similar research published with the same species in the same area in 2014.       

Author Response

General Overview

The manuscript presents an analysis of the best suitable areas for a critically endangered species of Equidae (African wild ass) that lives in Ethiopia. The species is down to approximately the last 600 individuals living in the wild, mostly in the Africa arid region of the Danakil desert that comprises Ethiopia, Eritrea and Djibouti. The authors applied a widely used method by modelling the species distribution through Maxent software for that purpose. They predict that the main factors driving their species´ choice of habitat is the distance to permanent water resources during the dry season and the amount of precipitation during the wet season. The English writing is good and clear, as is the overall manuscript structure. The authors also discuss the challenges this species may endure in the next years as it is struggling with habitat loss and competition with human livestock for water resources.

Response to Reviewer 3,

Thank you for your positive and encouraging comments on our manuscript. We are pleased that you found the English writing to be clear, the structure sound, and the overall topic and analysis to be of value. We have carefully considered your feedback and have incorporated all suggestions into the revised manuscript. The changes made on the manuscript have been highlighted in yellow. Thank you again for your time and constructive review.

Detailed Review

Comment 1: Although Maxent is an admittedly well-suited tool for the purpose of identifying the factors that determine species distribution, it has some shortcomings that must be discussed. Maxent might produce a biased set of predictions if the factors are somewhat correlated (Lissovsky, A. A., & Dudov, S. V. (2021). Biology Bulletin Reviews, 11(3), 265-275.). The authors addressed this issue in the manuscript, albeit a matrix table expressing the collinearity between the predictor variables would be desirable. The manuscript would also benefit from a figure expressing the annual average month temperature and precipitation as both are considered important prediction factors by the authors.

Response 1: A sentence has been added about caveats on Maxent modeling on line 207-210 and the table of correlation matrix has been included on the supplementary data for both seasons. The monthly temperature and precipitation values of the study areas were described on line 114-115 and 118-120.  

Comment 2: Another important predictor variable that should be better presented is the dynamics of the human occupation in the area. The authors mention a reference with the very same purpose of this manuscript published 11 years ago, in the very same portion of the Danakil Desert (Kebede, F., Moehlman, P. D., Bekele, A., & Evangelista, P. H. (2014). Predicting seasonal habitat suitability for the critically endangered African wild ass in the Danakil, Ethiopia. African Journal of Ecology, 52(4), 533-542.). They emphasize in this manuscript that distance to human settlings is important to determine African wild ass choice of habitat as they tend to avoid these areas. They also discuss that the human occupation in the area has increased in the last 10 years, which reduced the habitat suitable habitats for the African wild ass. Unfortunately, no quantitative data was provided in this manuscript to support this statement. I recommend the authors to better describe the human occupation of the area in the last 10 years as it is a critical factor.

Response 2: A justification has been taken about the human population size on line 74-81.

Comment 3: Regarding the African wild ass suitable habitats and the species distribution range, I recommend the authors to express in the manuscript how these animals move between Ethiopia and Eritrea as both countries share a border in the Danakil desert and African wild ass individuals are recorded in both countries. In fact, a very similar paper was published in 2023 discussing the best suitable areas for this species in Eritrea (Tesfai, R. T., Parrini, F., Moehlman, P. D., Young, N. E., & Evangelista, P. H. (2023). Predicting suitable habitat for the Critically Endangered African wild ass Equus africanus in the Danakil Desert of Eritrea. Oryx, 57(5), 592-599.). Unless there is some sort of fence in the border between these countries, these animals must transit unaware of any political border. Considering the critically endangered status of the species such movements should be addressed and be part of any conservationist effort.

Response 3: We definitely accept this comment. We have a future plan of doing the African wild ass cross boundary habitat distribution and to join hand to conserve the species where it thrive based on its suitable habitats.

Comment 4: Finally, it is important to consider that many animals distribution are determined not only by environment factors, but also by their biological interactions with other species. Such is that importance that many papers are now incorporating biological interactions in their prediction factors (1- Van der Putten, W. H., Macel, M., & Visser, M. E. (2010). Philosophical Transactions of the Royal Society B: Biological Sciences, 365(1549), 2025-2034;  2- Bateman, B. L., VanDerWal, J., Williams, S. E., & Johnson, C. N. (2012). Diversity and Distributions, 18(9), 861-872; 3- Lewin, A., Erinjery, J. J., de Waroux, Y. L. P., Tripler, E., & Iwamura, T. (2021). Journal of Arid Environments, 192, 104547 ; 4- Gherghel, I., Brischoux, F., & Papeş, M. (2018). Journal of Biogeography, 45(9), 2216-2225 ;   5- Cosentino, F., Seamark, E. C. J., Van Cakenberghe, V., & Maiorano, L. (2023). Ecology and Evolution, 13(3), e9855.)

Response 4: The variables that indicate biological interactions (peoples and livestock) were included in the temporary settlement variables as African wild asses avoid settlements with peoples and livestock. This is explained on line 193-196 and a statement has been included on line 87.

Comment 5: Considering the human occupation of the area is increasing, the competition African wild ass is enduring with human livestock for water, one can predict that contact with diseases from domestic animals might be more frequent. Moreover, as long as the livestock increases in the area as do so the competition for the water resources, one might expect the increasing presence of natural predators in the proximity. Including these biological factors as predictor factors would enhance this manuscript greatly as the overall results presented so far do not differ much of the ones provided by a very similar research published with the same species in the same area in 2014.      

Response 5: Predators and disease prevalence were not factors for the distribution of African wild ass so we found it irrelevant to use those variables on our modeling.

The current study has given more insight incorporating temporary water point data and biotic variables (peoples and livestock data) as these variables relate with mobile village huts of locals. This showed the extent of suitable habitat as our primary aim is to locate where to delineate a conservation area for African wild ass. In this regard, the current study showed new explicit findings than the previous one.

Author Response File: Author Response.docx