A Socio-Ecological Approach to Understanding How Land Use Challenges Human-Elephant Coexistence in Northern Tanzania

Round 1

Reviewer 1 Report

Dear authors, 

Thank you for the opportunity to review this interesting and very useful piece of work. I thought the work was generally clear and well written, however there is some clarity needed in relation to the surveys. 

My major concern/confusion is surrounding the survey data: it is unclear what the purpose of the surveys was, what data was captured from the questionnaires, nor how this contributed to the results. A copy of the questions used for these interviews is needed, and it would be beneficial to briefly describe the main topics of conversation in the methods. An overview of these findings is then needed in the results. It would also be interesting to report how the reports from local people match up with the official incident logs from the same time periods. Some consideration is then needed (maybe in a study limitations section) as to how reliable this data is, and why you think it is more reliable than other surveys (L76/77). You also need to include ethical approval for these surveys - as you are using human data ethical approval is required. 

Discussion would benefit from a brief section on study limitations (as mentioned above) and then also a section on implications of the research/a brief summary of recommendations for mitigating the issues of HEC

Line by line comments:

L21: seems to be a word missing between 'analyzed' and 'dataset'

L22: the years

L28: sentence beginning 'elephant speed' doesn't read quite right, think it needs slight rephrasing

L32: unclear what you mean by 'ensure human livelihoods'?

L77: excaggerated by survey respondents

L91: you only looked at male elephants?

L97: obj iv seems a repeat of i, suggest dropping speed from objective i

Fig 1. Move to after it is first detailed in the text

Section 2.4 - is this all of the HEC data or is this likely to be an underrepresentation of actual encounters? Would be beneficial to briefly comment on how representative it is

Section 2.5 - were the elephants collared for this study? This section reads like they were but the acknowledgements read like you were given the data. This needs to be clarified. If collared for this study then ethical approval is also needed for this. If just using data that has been provided then this needs to be clearer. 

L206: unclear how field questionnaires were used here

L222: which groups? which parameters?

L312: last sentence in figure legend not needed

L316: remove 'by'

L389: male/bull elephants

L396: not clear how your data shows this?

Author Response

Dear Reviewer,

Thank you very much for reviewing our Manuscript and constructive comments

we have revised comments line by line and uploaded the revised comments, manuscript and the copy of questionnaires we used as recommended.

best regards

John Sanare

corresponding author.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript "Understanding how land use challenges human-elephant coexistence in Northern Tanzania" is suitable to publish in Diversity.

In Table 3 the percentage should be the exact number when the areas were not 0.

Author Response

Dear Reviewer,

Thank you very much for reviewing our manuscript and constructive comments

we have revised the comments line by line and uploaded the revised manuscript

With kind regards,

John Sanare

corresponding author.

Point 1: In Table 3 the percentage should the exact number, when the areas were not 0.

Response 1: We thank the reviewer for this comment, we have revised table 3, find attached document

Author Response File: Author Response.docx

Reviewer 3 Report

The authors studied how land use change affected human-elephant conflicts (HEC) in Enduimet Wildlife Management Area (EWMA). The manuscript is interesting, tackling an important issue. However, some improvement is needed before it can be accepted for publication.

The authors must better associate their methods, results and discussion with their study area (EWMA). Also, they must firstly and explicitly emphasize and discuss their findings about the effects of land use land change (LULC) on HEC in the study area. Right now, this has not happened, making the manuscript weak, hard to convey any significant outcome and is not in line with the stated aims.

The English language needs some improvement.

Comments

---Line 21 – We analyzed the dataset…

---Line 22 – from 2016 t0 2020…

---Lines 23-30 – LULC changes were studied, however the authors did not refer to how these changes affected HEC, the main objective of the study, as reflected in the title. Please add and emphasize this information.

---Lines 27-28 – Authors state here that elephants prefer other the study area. The reader could infer that HEC might not be intense because of this. However, later on, the authors propose measures for mitigating the intense conflict. Please revise and clearly convey what is really important.

---Lines 28-30 – Other factors were also found important, besides speed. Please add them in the Abstract.

---Line 29 - farmland, likely reflecting the availability of foraging opportunities, …

---Lines 30-33 – Do the authors mean that settlements need to be relocated? Or the boundaries of protected areas should be revised? Please be clear about suggestions.

---Line 41 – Humans…

---Line 55 – Aryal et al. [10] found…

---Line 63 – Delete ”A2abd ver 3.1”. Also, “humans have become…”.

---Lines 90-91 – But the authors did not estimate the home range of female elephants. Please revise aims.

---Line 110 – How many elephants used the study area during the wet season? Give this information.

---Lines 127-128 – But, according to Figure D1, June-November is dry season. Please check and revise.

---Line 136, Figure 1 – According to Figure 1, most farmland was found outside the EWMA, the stated study area. However, the authors talk, throughout the manuscript, about HEC in EWMA. This is very confusing and must be revised. Also, the authors must be consistent throughout.

---Lines 287-291 – We cannot talk about “highest number” when there are statistically significant differences…

---Lines 303-304 – Again here, we cannot talk about “highly variable” when there are statistically significant differences…

---Lines 304-305 – Table A2 and Figs A1-A3 do not exist…

---Lines 306-310 – According to Figures 1 and 4, most HECs occurred both within village boundaries and in or close to farmland. Also, the observation from these maps that the two HEC hotspots of Tingatinga and Ngereyani fall within village boundaries where farmland is scarce, while HECs are scarce outside village boundaries where farmand is more emphasize the importance of settlements in HEC. This is a major observation that needs to be explained in the discussion. Also, most farmland fells outside the boundaries of EWMA which was supposed to be the study area. This has to be clarified or revised.

---Lines 318-321 – This is discussion.

---Lines 330-333 – Do not repeat results in the Discussion.

---Lines 340-342 – But, according to maps, most farmland was outside the EWMA. Please revise and explain.

---Lines 346-347 – Change from 2% to 0% is not severe. Although forest no longer exists. Revise appropriately.

---Line 375 – But May is wet and June dry according to Figure D1. Please revise.

---Lines 381-383 – So the reason of conflict was not the increase in farmland in these to hotspots (?). Plantations (what kind of?), bushland, tree and grassland systems where? Be more specific and relate it with LULC, the main purpose of the manuscript.

---Lines 383-386 – This is ambiguous and confusing. How these habitats and their arrangement have affected elephant movements? Please elaborate.

---Lines 388-389 – Since movements did not differ between the season no need to discuss.

--Lines 398-399 – That elephants spend most of their time away from farms reads the reader to infer that HEC would not be important. Please further explain this point and emphasize the significance of HECs for the study area.

---Lines 406-407 – Why were HECs lower in lower human population density areas? Any explanation?

Author Response

Dear Reviewer,

Thank you very much for reviewing our manuscript and constructive comments

we have revised the comments line by line and uploaded a revised manuscript

Best, Regards

John Sanare

Author Response File: Author Response.docx

Reviewer 4 Report

Human-wildlife conflict is a very important issue in changing world, especially that concerns endangered species. The manuscript entitled: ‘Understanding how land use challenges human-elephant coexistence in Northern Tanzania’ consider such issue. The Authors analyzed crop foraging events done by elephants, combined with data on changes in land use/land cover, environmental variables, and data collected from collared elephants, showing interesting results. Therefore I recommend MS to be published, after some changes, listed below:

Abstract

Please add the aim of the research to the Abstract. So far, I don’t understand why the Authors used data received from GPS collars. Do you mean positions? (l. 24)

l. 23-24 Is it LULC changes from 1989-2019 or 1989, 1999, 2009, and 2019?

l. 27 Based on Abstract, I don’t know what Amboseli is, and how it’s different from Enduimet. I suggest rewriting that sentence, for example like this: Elephants spent most of their time in human-modified habitat (Amboseli) compared to natural area of Enduimet, increasing the risk of human-elephant conflicts.

l. 27 I don’t understand what ‘area under land use’ means here, all areas more or less are under land use. Is it modified by humans? Settlements? Farmlands? Please specify.

l. 28 Please write clearly in the Abstract how and what for did the Authors describe elephants' speed.

l. 30-31 We conclude that communities in Enduimet urgently need support to increase the effective distance between settlements and protected areas – I don’t understand this conclusion. How the Authors want to increase that distance? By moving settlements or PAs?

l.31 Based on the Abstract, I don’t understand why Authors pointed out the community in Enduimet, while They showed earlier, that higher risk of human-elephant conflicts occurred in Amboseli (l. 27).

Keywords

l. 35 - as I understood, the analysis concerned only one specie, it would be a good idea to put the (Latin) name of the species in keywords; delete ‘Kernel estimation density’; please delete ‘remote sensing’? Maybe add ‘human-wildlife conflicts’; I suggest changing ‘Elephant collars’ into ‘GPS collars’

Introduction

l. 68 I cannot see the link between previous sentences in the paragraph and description of the method and usage of elephants' speed

l. 91 As I understood from other parts of MS, it should be only ‘male’ here.

Materials and Methods

l. 168 Pleas add the questionnaire (as Appendix).

l.181 Could you explain more what OIKOS does? Moreover, I cannot understand how the result from the questionnaire was compared/linked to data obtained from OIKOS.

l. 219 Kikoti et al. [24]

l. 241 I don’t understand the first result of the research. Although it is listed in the aims of the MS, I rather see it as Supplementary material, or it should be combined with HEC. But data on historical HEC should be needed here.

l. 258-259 What do you mean by ‘near’, ‘close’? 1-10 meters? 50 meters? 500 meters?

l. 261 What do you mean by ‘higher average’? 0.55 or 0.7?

l. 258-264 All values given here were taken from Table 4, so there is no need to show them twice. Table 4 is enough. The Authors could only list variables that significantly influenced HEC occurrence. I would rather like to see how all important variables influenced HEC occurrence. Moreover, most of the values were shown as continuous, so I expect to see them as continuous on graphs. Please change Figure 2.

l. 269 Could you explain how to define interaction between river and road and from farmland and PAs, shown in Table 4 ?

l. 285 Although crop losses caused by elephants occurred throughout

l. 286-287 most incidents (55%) were recorded during the dry season, from June to November, when wild forage resources for elephants decrease in quality
The second part of the sentence should be written in Discussion chapter.

l. 289 I couldn’t find Figure A2

l. 303-304 There is lack of Table A2 and Fig A1-A3 in the MS.

I don’t know if farmlands are equaled agriculture areas. In different part of manuscript one of those names was given. Please stick to one word in the MS.

Figure 4. What is the difference between colours of home range 31-60 and 61-90. I cannot see different home ranges for three different individuals.

Based on results, I disagree with Authors conclusion showed in Abstract. Even the speed of elephants was lower in farmlands/agricultural areas, their speed in settlements was similar compared to other areas.

l. 412-413 ‘which has increased the competition between elephants and humans’ it is not a conclusion based on MS

l. 413-414 As the majority of farms in EWMA were located close to the protected areas – it was not shown in the MS

Author Response

Dear Reviewer,

Thank you very much for reviewing our manuscript and constructive comments

we have revised comments line by line and uploaded the revised manuscript

Best Regards.

John

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear John & colleagues,

Thank you for taking the time to respond to my comments and to make edits to the manuscript. I have made some further comments on your revision which I think are still needed to improve the clarity of the work.

L77 – 79: this is a repeat of the sentence below

L130: citation needs formatting

L146: citation needs formatting

L150: m – not masl

L192/193: I am still unclear on the surveys – I don’t think they contribute to an understanding of land use cover (given that wasn’t covered in the survey). What was included in the ‘participatory group discussions’? why was this done in addition to the conversations with local people? And what was asked of the EWMA staff? Clarify please over what time period you were asking people to remember the incidents. The supplementary material also needs to be referred to.

L224: this is too precise to be ‘about’

L236: how did you integrate this data and why did you integrate? Would this not lead to a double representation of some incidents if village people also reported HECs through official channels? I think it would be better to treat these as separate sets of data and then compare the outputs with the official data you captured.

L290: not sure what ‘users accuracy and producers accuracy’ means?

L297: this no longer makes sense as a sentence

Table 4: split into two tables – with model outputs as a separate table. Presume % in first part is % of raids? This needs clarifying

Figure 2: graphs appear to be repeated

L320: statistics missing – again state whether the information provided in this area is reported incidents or from interviews – I think it would be clearer to separate these

L343: either bull elephants or male elephants – one or the other, not elephants/bull

L348: not sure what the codes mean – is this supplementary material? If so then clarify

L360: needs stats output

L373: sentence doesn’t make sense

L432: sentence does not make sense

L469: change has to ‘may hav’

L479: the same speeds

L480: bull elephants

L484: areas other than farms

L486: where did your results show this?

L514: this doesn’t make sense as it is currently written. Limitations also don’t usually get added to the conclusion. They would be included as part of the discussion. At the moment they feel like an ‘add on’ to the end of the manuscript.

Author Response

Dear Reviewer,

Thank you very much for reviewing our Manuscript.

we have revised comments line by line and uploaded the revised comments , together with revised manuscript version, moreover  a questionnaire forms as recommended

Best wishes.

John Sanare

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors failed while addressing suggestions:

a) to connect conflicts with the study area.

b) to relate conflicts with land use (their major stated aim)

c) to clarify methodology and make clear and meaningful conclusions.

Author Response

Dear Reviewer,

Thank you very much for reviewing and provides the good comments in our manuscript.

we have revised all points line by line and uploaded the adressed comments , together with the revised manuscript

Author Response File: Author Response.pdf

Reviewer 4 Report

I thank the Authors for all their responses and clarifications. However, I still have some comments to the MS (point # corresponds to the #  given by Authors).

Point 1:  After reading the whole manuscript I understand all goals of the study. Nevertheless, abstract should be a separate paragraph, giving overview of study. In presented version of Abstract, I can’t see the aim of the study. Lines 21-23 presented rather materials and methods. I also suggest presenting methods and results in the same order.

Point 15: Thank you for your explanations, but I still can’t understand it from the MS. For example, the distance from settlements is given as continuous variable (up to 60 km), and the result from logistic regression only showed that the smaller the distance, the higher the HEC occurrence. You did not define “close” or “near” in MS.

Point 25: Based on Figure 5 you can tell, that speed of elephants was lower in farmland, but higher in settlements and other areas. What does ‘other areas’ mean?

Point 26: In results, you showed that “(…)forests and grasslands had been converted into agricultural land and settlements over the last three decades (…)” but based on that you cannot tell, that “which has increased the competition between elephants and humans”.

Author Response

Please find attached

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Dear authors,

Thank you for taking the time to respond to and action as appropriate the comments I gave you on the manuscript.

I am still confused on the data collected during the OIKOS surveys and your surveys – the OIKOS surveys you said in response #7 were from 2016 – 2019. But in the revision at L187 – 198 you say that you were asking villagers to remember incidents from the last 5 years. So there will be a duplication in that data surely? I am also still unclear how you combined the data, you said you have changed that in the text to be clearer but this needs to show what data you extracted and how it was combined/what did you do with that data.

I cannot see the survey in the supplementary material?

Minor comments:

L71 – the IUCN red list

L77 – occurrences

L102 – I now understand (after your latest response) that the current interviews basically provide the same information as the data from OIKOS but I don’t think the sentence at line 102 reflects this. The sentence at L102 I think could do with revising to match your new methods information. I think it is the reference to grey literature which is confusing.

L112/113 – suggest saying ‘would increase’ as these are hypotheses/predictions

L117 – of not on

L188 – delete ‘was used’

L198 – only after verbal consent to participate was given

L454 – not sure what this means ‘But not significant, so show other studies where this was significant and explain why you think we did not see these patterns.’

Author Response

Dear Reviewer,

Thank you very much for reviewing our manuscript.

we have revised comments line by line and uploaded together with revised Manuscript as well as questionnaire used as supplementry (S1) material

Author Response File: Author Response.docx

Reviewer 3 Report

The authors state in the Abstract that “41% of the area had been converted into farmlands and settlements within the last three decades but human-elephant conflict did not significantly increase.” (lines 29-31) and “human-induced LULC changes and the encroachment into elephant habitats have resulted in spatially and temporally predictable increases in HEC in EWMA” (lines 35-36). These statements are contradicting. Also, they are not supported by results.

Again, aims do not correspond to hypotheses regarding land use land cover (LULC) (lines 108-116).

The above inconsistencies emanated from the inability to compare long-term LULC change with human-elephant conflicts (HEC) due to a lack of data, as the authors now recognize in lines 498-505.

However, besides these inconsistencies in the dataset the authors insist to make “comparisons”, which lead to further inconsistencies, confusion and logical leaps.

Having said the above, the manuscript could be of great value if the authors present only the data than can be compared. I suggest to remove long-term LULC change comparisons from the manuscript since this is the greatest source of confusion. Keep only land use types for 2019 that can be compared with 2016-2020 HEC data.

More particularly:

Keep Figures 1-5. Clarify what you mean “Distance from protected areas” in Figure 2. Is it Enduimet Wildlife Management Area (EWMA) or all nearby protected areas? Be careful and consistent throughout the manuscript.

Move Table 1 to supplement.

Delete Table 3 and all relevant methods, results, discussion.

Keep Tables 2 and 4.

Delete Table S1. Give information about 2019 in the text.

Keep Table S2.

Keep Figures S1-S3.

Rename Figure D1 to Figure S4 for consistency. Add months December and January even if no HEC were detected. This will better show the year-round picture.

S1, S2, etc. in Tables and Figures refer to online supplementary materials. If the authors choose to keep them in the manuscript, they should be treated as Appendices. Please advise journal style for Appendices.

Following the above suggestions will give the manuscript direction, allowing the authors for making meaningful conclusions and proposing management actions. In this view, a title like “Identifying important human-elephant conflict hotspots in Northern Tanzania” would be most relevant and informative.

In the discussion, the authors should explain all their significant findings. Historical data from trusted sources could aid in this direction here. This would allow for making informed and meaningful management recommendations. Also, be specific where it is possible. For example, when proposing changes in crop type state what crops exactly. Have these crops used for mitigating HEC elsewhere? And with what success?

Author Response

Dear Reviewer,

Thank you for reviewing and provides the good comments in our Manuscript. we have revised all points line by line and uploaded the adressed comments and a revised manuscript

Author Response File: Author Response.docx