Influence of Moon Luminosity, Seasonality, Sex and Weather Conditions on the Activity Levels of the Nocturnal Javan Slow Loris
Round 1
Reviewer 1 Report
The authors work with an impressive data set, working with very update to methods to achieve their objectives. They suggest a very interesting approach to make up the low visibility we often encounter in our field when collecting data on animals, especially in the wild. Although direct behavioural observations might provide more detail on their activity budget, low visibility potentially does obscure important aspects of the animal’s activity. By using accelerometers, they were able to collect a massive amount of information regarding the animal’s movements over a prolonged time frame. In fact, such an amount would nearly be impossible to collect through traditional observations. This generally seems a very interesting approach, but especially in nocturnal species which are all the more difficult to observe seems very effective. I want to congratulate the authors for successfully demonstrating that efficiency and usefulness of this method.
Throughout the first time a read through the manuscript I wrote down several doubts and questions, which pleasantly I found to be answered in following sections. The authors use a very comfortable to follow and engaging writing style which both animates and satisfies the reader curiosity.
Although this study only focuses on 12 individuals, this is very typical in studies regarding primates. Furthermore, it has been compensated by providing long term data and thus should not be a reason against publishing this article.
Other than congratulating the authors I have very little to add, as in my opinion this draft could be published right away.
Some minor suggestions:
Line 85-86: The meaning is clear but I would rephrase the sentence. "recording in automatic" does not sound correct. Suggestion "...with the capacity to automatically record animal activities"
Line 233-240: You mention here the possibility of lactation, gestation and infant raising to be a possible explanation of the reduced activity of females compared to males. I definingly agree and I think it is important to mention, but it would be helpful if you could state if and how often the females had offspring during the observation period. Like this, the reader gains a better understanding how this might have affected your data.
Materials and Methods: I assume that all 12 individuals were adults already at the start of the data collection phase. I would suggest to mention this in 2.1 Study Site and Subjects.
Result section: I believe it would be a great addition if you could provide general information about the activity levels (descriptive simple data). Apart from the very interesting results you present, I often find that these days we forget to also publish the very basic insights of our data. In order for other researches working with other populations in the wild or in captivity to have a baseline for “normal” activity levels it would be very interesting to have your data as a reference point. A refer to a simple table listing the average activity levels of Javan slow lorises. This might be as simple as averaging your whole data set or could be mor detailed (separate avengers for seasonality, sex, etc.). If you feel that this would interrupt the reading flow as you might not want to discuss this afterwards in the discussion section you could simply add this to the supplementary material.
Author Response
The authors work with an impressive data set, working with very update to methods to achieve their objectives. They suggest a very interesting approach to make up the low visibility we often encounter in our field when collecting data on animals, especially in the wild. Although direct behavioural observations might provide more detail on their activity budget, low visibility potentially does obscure important aspects of the animal’s activity. By using accelerometers, they were able to collect a massive amount of information regarding the animal’s movements over a prolonged time frame. In fact, such an amount would nearly be impossible to collect through traditional observations. This generally seems a very interesting approach, but especially in nocturnal species which are all the more difficult to observe seems very effective. I want to congratulate the authors for successfully demonstrating that efficiency and usefulness of this method.
Throughout the first time a read through the manuscript I wrote down several doubts and questions, which pleasantly I found to be answered in following sections. The authors use a very comfortable to follow and engaging writing style which both animates and satisfies the reader curiosity.
Although this study only focuses on 12 individuals, this is very typical in studies regarding primates. Furthermore, it has been compensated by providing long term data and thus should not be a reason against publishing this article.
Other than congratulating the authors I have very little to add, as in my opinion this draft could be published right away.
We thank the reviewer for the very kind words, and we appreciated the suggestions provided to ameliorate this piece of work. We have considered all of them.
Some minor suggestions:
Line 85-86: The meaning is clear but I would rephrase the sentence. "recording in automatic" does not sound correct. Suggestion "...with the capacity to automatically record animal activities"
Changed
Line 233-240: You mention here the possibility of lactation, gestation and infant raising to be a possible explanation of the reduced activity of females compared to males. I definingly agree and I think it is important to mention, but it would be helpful if you could state if and how often the females had offspring during the observation period. Like this, the reader gains a better understanding how this might have affected your data.
We have now added this information, plus added additional considerations based on the comments from other reviewers as well.
Materials and Methods: I assume that all 12 individuals were adults already at the start of the data collection phase. I would suggest to mention this in 2.1 Study Site and Subjects.
We have now added this important information
Result section: I believe it would be a great addition if you could provide general information about the activity levels (descriptive simple data). Apart from the very interesting results you present, I often find that these days we forget to also publish the very basic insights of our data. In order for other researches working with other populations in the wild or in captivity to have a baseline for “normal” activity levels it would be very interesting to have your data as a reference point. A refer to a simple table listing the average activity levels of Javan slow lorises. This might be as simple as averaging your whole data set or could be mor detailed (separate avengers for seasonality, sex, etc.). If you feel that this would interrupt the reading flow as you might not want to discuss this afterwards in the discussion section you could simply add this to the supplementary material.
We agree with the reviewer that the reference activity scores should be presented. We have now added the mean score in the results and the means and 95% CI for the relevant categories as Table A1
Reviewer 2 Report
The Authors provide here an accurate analysis of the factors influencing activity patterns of a wild population of Javan slow loris. The Manuscript is overall well structured and detailed, the introduction section is balanced and the conclusions are supporte by the results.
I only have some minor concerns to address to the Authors, as follows:
LL.72: please remove the comma after the word "small"
LL.135: The Authors state that they extracted raw data on the activity counts from the software. All the analyses reported in the work refer anyway to the activity scores. Is there a difference between the two? As I can see in Figure 2, the activity score could have different values. If yes, how where the scores calculated?
LL.164: Did the Authors test for model concurvity? Please provide details about it.
LL. 184: Have you tested for interactions between the predictors, e.g. temperature*humidity?
Fig.1: I suggest the authors to change the colors of the heatmap. The figure is useful but the color variation is difficult to be interpreted, especially figure b.
Table 1: It is not clear from the table if the authors also tested for individual identity as a predictor. In LL. 153 the Authors state that individuals were modeled as random factors, but they inserted the parameter as predictor in table 1. I also suggest to change the predictors name in the Table, using only the name reported in the brackets. I know this is the model output, but for the reader it will be easier.
Figure 2: What does the number 9.93 refers to in the Gaussian quartiles box?
LL.238: Do the sexes show dimorphism in feeding rates and ecology? Could differences in activity scores be related to different energy intake by the sexes? Could differences be discussed at the light of differences in spatial ecology?
As a general comment, do the Authors considered reduced visual acuity as a possible explanation for a drop in activity scores during more luminous nights?
Author Response
The Authors provide here an accurate analysis of the factors influencing activity patterns of a wild population of Javan slow loris. The Manuscript is overall well structured and detailed, the introduction section is balanced and the conclusions are supporte by the results.
We thank the reviewer for the very kind words, and we appreciated the suggestions provided to ameliorate this piece of work. We have considered all of them.
I only have some minor concerns to address to the Authors, as follows:
LL.72: please remove the comma after the word "small"
Removed
LL.135: The Authors state that they extracted raw data on the activity counts from the software. All the analyses reported in the work refer anyway to the activity scores. Is there a difference between the two? As I can see in Figure 2, the activity score could have different values. If yes, how where the scores calculated?
We have now clarified that activity scores are the same as activity counts. In Figure 2 there is a representation of the GAM output, meaning those are variations over the reference value (median predicted response value) but the raw data are activity scores given by the accelerometer.
LL.164: Did the Authors test for model concurvity? Please provide details about it.
We have now added this information in the Data Analysis section
- 184: Have you tested for interactions between the predictors, e.g. temperature*humidity?
We have now added this information in the Data Analysis section
Fig.1: I suggest the authors to change the colors of the heatmap. The figure is useful but the color variation is difficult to be interpreted, especially figure b.
We have now changed the colours we hope the figure is easier to be interpreted
Table 1: It is not clear from the table if the authors also tested for individual identity as a predictor. In LL. 153 the Authors state that individuals were modeled as random factors, but they inserted the parameter as predictor in table 1. I also suggest to change the predictors name in the Table, using only the name reported in the brackets. I know this is the model output, but for the reader it will be easier.
The random effects are included as an output of GAMs as they are modelled with the smooth term bs=”re”, which does not mean that they are fixed factors. All the ones with s are smooth terms, that is why we would like to keep them like that to differentiate from the fixed factor Sex.
Figure 2: What does the number 9.93 refers to in the Gaussian quartiles box?
That number represents the edf (also reported in Table 1). That is what is usually included in the output of a GAM for random effects
LL.238: Do the sexes show dimorphism in feeding rates and ecology? Could differences in activity scores be related to different energy intake by the sexes? Could differences be discussed at the light of differences in spatial ecology?
We have included a section in the discussion
As a general comment, do the Authors considered reduced visual acuity as a possible explanation for a drop in activity scores during more luminous nights?
We have now added two lines in the discussion
Reviewer 3 Report
This study analyzed activity of Javan slow loris and found that several variables influence activity. The manuscript is clear, well written and the topic is relevant.
I have only few comments:
Page 3, line 118: What is the activity level? refers to periods of time when the animal is moving? what type of movements? I think it is not clear in the manuscript how activity is measured.
Page 3, lines 124-125. It says ‘the placement of collars around the neck is efficient for recording activity of slow lorises’. Then why the activity was not analyzed with the data from the VHF collars? I think it is necessary to explain in the manuscript the advantage of accelerometers.
Author Response
This study analyzed activity of Javan slow loris and found that several variables influence activity. The manuscript is clear, well written and the topic is relevant.
We thank the reviewer for the very kind words, and we appreciated the suggestions provided to ameliorate this piece of work. We have considered all of them.
I have only few comments:
Page 3, line 118: What is the activity level? refers to periods of time when the animal is moving? what type of movements? I think it is not clear in the manuscript how activity is measured.
We have now clarified that activity levels are based on activity scores from accelerometers.
Page 3, lines 124-125. It says ‘the placement of collars around the neck is efficient for recording activity of slow lorises’. Then why the activity was not analyzed with the data from the VHF collars? I think it is necessary to explain in the manuscript the advantage of accelerometers.
We refer the reviewer to the introduction (lines 82-88) where we clarified why it is important to consider the accelerometer data for nocturnal animals, plus the last paragraph of the conclusion where we explain the importance of bio-logging devices such as accelerometers to collect long term data, especially on cryptic animals.
