Quantifying Southern Pacific Rattlesnake (Crotalus oreganus helleri) Hunting Behavior through Community Science

Round 1

Reviewer 1 Report

The authors present their evaluation of data on predation behavior of a rattlesnake species they extracted from two community based platforms. The mansucript is well written and well structured. It presents the results in a plain and straight forward manner. I am mostly working on taxonomic and phylogenetic issues and maybe I am the wrong person to judge the scientific merit of this manuscript but while reading I had difficulties to understand the significance of their findings for conservation or even as an important contribution to the knowledge of the natural history of the studied species. Anyways, I have made a few comments and corrections in the electronic file for the authors to consider.

Comments for author File: Comments.pdf

Author Response

We thank the reviewers for their comments and suggestions, and overall positive reviews. We have addressed each one in bold text below. Upon revision, we have fixed information in the introduction regarding snake species and added more discussion on biases within community science data. We have also made the suggested English and grammar changes. We thank you again for your time and helpful suggestions.

 

Reviewer 1

The authors present their evaluation of data on predation behavior of a rattlesnake species they extracted from two community based platforms. The manuscript is well written and well structured. It presents the results in a plain and straight forward manner. I am mostly working on taxonomic and phylogenetic issues and maybe I am the wrong person to judge the scientific merit of this manuscript but while reading I had difficulties to understand the significance of their findings for conservation or even as an important contribution to the knowledge of the natural history of the studied species. Anyways, I have made a few comments and corrections in the electronic file for the authors to consider.

Response: We thank the reviewer for their positive comments on our writing. We rewrote areas of the manuscript to reduce focus on conservation efforts, acknowledging that our study does not offer direct conservation management recommendations (see changes to second paragraph of introduction). We also made an effort to highlight the novelty of the research – which really is collecting a large dataset on juvenile observations, which are notoriously difficult to study using traditional field methods. We accepted all suggested wording corrections on the pdf file provided by the reviewer, and we respond to the few comments made on the document below.

 

Line59-60 C. cerastes climbing up into trees? really?

Response: Our apologies, we were not clear that this line was referring to the Santa Catalina Rattlesnake (Crotalus catalinensis). Thank you for commenting on our misinformation, we have corrected our writing. Please see changes in lines 60-61:

“Juvenile Santa Catalina Rattlesnakes (C. catalinensis) can often be found in arroyos where lizard activity is high, while adult rattlesnakes are more often found in areas with high mammal activity [26].”

 

Line 145-146 who confirmed the respective IDs?

Response: iNaturalist is a web-based community science data center. Users upload their photos to the website, and other users can confirm or change taxonomic IDs as needed. An observation needs a 2/3rds agreement of community science members in order to confirm a taxonomic ID. HerpMapper uses a different method. Users can choose the taxonomic ID, and other users can request a revision if it is incorrect. If one is deemed incorrect, a panel of taxonomic reviews can confirm or deny the ID. See changes on lines147-151:

“Observations on iNaturalist become ResearchGrade, when they have media (photo or sound), a location, a date, and a community consensus on a precise taxonomic identification. More than two-thirds of the community identifiers must agree on the taxonomic ID. HerpMapper uses a similar method, as users can choose the taxonomic ID, but it can be corrected or changed by a panel of taxonomic reviewers who are experienced herpetologists.”

 

Figure 1. This diagram is superfluous! This easy to understand procedure is well explained in the text.

Response: We respectfully disagree with this comment, and believe it is an informative diagram because it demonstrates the filtering process in a step-by-step matter and emphasizes the amount of observations that were irrelevant to our study. Furthermore, this detailed information (i.e. the specific numbers) isn’t listed in our text. It illustrates the filtering process of ~4,000 observations, and includes the number of photos that were removed for different reasons. We think the diagram may emphasize the need for better observations on community science platforms.

Reviewer 2 Report

Overview:  

This is an interesting and well-written paper on the behavioral ecology of Pacific Rattlesnakes. Such data are important for understanding snake behavior and are needed badly in the literature. In the study the authors used “crowd-sourced” data from online social media sites to collect their data, a new and innovative approach.

Overall, I have a positive evaluation of the paper. However, their are two critical issues that detract from this evaluation:

1. The paper is justified as being important to snake conservation and management efforts. I see little basis for that link and do not see why the authors want to go there. Data on snake behavior are important in their own right and don’t require a functional significance in terms of management. See “Detailed Comments” below.
2. I am very uneasy about the biases built into these types of data sources. See “Detailed Comments” below for more on this. 

Detailed Comments (linked to line #)

Lines 38-45: At least some of the information in this paragraph is either incorrect or outdated. For example, the information on Massasaugas is flat-out wrong. Not only is the citation used 70 years old, but Massasaugas do not exist in Mississippi. 

Lines 48-49: While fully supportive of the need for more and better data on rattlesnake behavior, I fail to see the link between ontogenetic changes in foraging behavior and “effective conservation strategies”. Please provide examples from the literature to support this contention.

Lines 79-108: The authors present a well-written overview of the advantages of crowd-sourced data and I agree with much of what they say. However, to be objective, the authors must also review the pitfalls of such data. What these observations represent are classic “accessibility sampling”, where only behaviors that are shown by individuals accessible to untrained observers are recorded. They are NOT a random or representative sample and should not be treated as such. We learned this lesson in snake biology with body temperature data, where temperature readings on active individuals were used to represent temperature profiles of snakes in nature. That, we learned, was a very incomplete picture of what was going on, once we compared that to data from telemetry. 

Lines 200-229: I was glad to see that the authors did address some of the bias issues here and agree with what they did in terms of reducing lack of independence. However, that does not address the comments above about accessibility sampling. 

Lines 316-317: This is a good example of where these kinds of data should be regarded with much skepticism when using accessibility sampling as was done here. 

Lines 336-340: Again, glad to see the authors recognize the apparent bias in their data set. However, I see no basis for the statement “these biases should be standardized across all age classes”. Indeed, I could argue that the opposite would be true. If, for example, small snakes are more active during the day, then they will be over-represented in the data set. 

In general, first, there are some errors in the Introduction (on Massasaugas, of all things), but that can be rectified easily. Second, the link between these data and management are weak at best and I think are misplaced.

Most importantly, however, is whether these "crowd-sourced" data are so inherently biased as to preclude their analysis.

 

 

Author Response

We thank the reviewers for their comments and suggestions, and overall positive reviews. We have addressed each one in bold text below. Upon revision, we have fixed information in the introduction regarding snake species and added more discussion on biases within community science data. We have also made the suggested English and grammar changes. We thank you again for your time and helpful suggestions.

Reviewer 2

Overview:  

This is an interesting and well-written paper on the behavioral ecology of Pacific Rattlesnakes. Such data are important for understanding snake behavior and are needed badly in the literature. In the study the authors used “crowd-sourced” data from online social media sites to collect their data, a new and innovative approach.

Response: We thank the reviewer for their overall positive reception of the manuscript.

 

Overall, I have a positive evaluation of the paper. However, there are two critical issues that detract from this evaluation:

1. The paper is justified as being important to snake conservation and management efforts. I see little basis for that link and do not see why the authors want to go there. Data on snake behavior are important in their own right and don’t require a functional significance in terms of management. See “Detailed Comments” below.

Response: Thank you for this comment. We agree that understanding behavior is important on its own and have therefore reduced wording related to direct conservation actions. We also highlight the novelty of the research which is using community science to collect a large number of juvenile observations.

 

1. I am very uneasy about the biases built into these types of data sources. See “Detailed Comments” below for more on this. 

Response: We have addressed this concern below.

 

Detailed Comments (linked to line #)

Lines 38-45: At least some of the information in this paragraph is either incorrect or outdated. For example, the information on Massasaugas is flat-out wrong. Not only is the citation used 70 years old, but Massasaugas do not exist in Mississippi. 

Response: We appreciate the reviewer bringing the incorrect information to our attention. We have updated our references and information in our introduction section. Please see changes in lines 38-42:

“For example, Red Diamond Rattlesnakes (Crotalus ruber), Timber Rattlesnakes (C. horridus), Twin-spotted Rattlesnakes (C. pricei), Banded Rock Rattlesnakes (C. lepidus. klauberi), and Ridgenosed Rattlesnakes (C. willardi) have lost much of their suitable habitats due to human activity and land development [8–12]. Eastern Massasaugas (Sistrurus catenatus) are considered endangered in all of their range [13], and Eastern Diamondbacks (C. adamanteus) are on the verge of becoming listed as a threatened species [14,15].”

Lines 48-49: While fully supportive of the need for more and better data on rattlesnake behavior, I fail to see the link between ontogenetic changes in foraging behavior and “effective conservation strategies”. Please provide examples from the literature to support this contention.

Response: We have updated our information and references to further illustrate how ontogeny can provide better conservation strategies. Please see corrections in lines 43-49:

“Understanding rattlesnake behaviors could be key for better understanding the ecologies of these species. Yet, very few studies focus on wild juvenile rattlesnake behavior, so conservation efforts may gear towards adult tendencies, which could be ineffective. Understanding ontogeny of behavior is often important. For example, Hellbenders (Cryptobranchus alleganiensis) show ontogeny in diet, suggesting the importance of aquatic insect microhabitats to protect Hellbenders at their larval stage [16]. American Alligators (Alligator mississippiensis) have been shown to utilize different microhabitats throughout their lifetime, therefore, all these microhabitats should be recognized in conservation plans [17]. Studies on juvenile individuals could reveal different needs of rattlesnakes throughout their life stages.”

 

 

Lines 79-108: The authors present a well-written overview of the advantages of crowd-sourced data and I agree with much of what they say. However, to be objective, the authors must also review the pitfalls of such data. What these observations represent are classic “accessibility sampling”, where only behaviors that are shown by individuals accessible to untrained observers are recorded. They are NOT a random or representative sample and should not be treated as such. We learned this lesson in snake biology with body temperature data, where temperature readings on active individuals were used to represent temperature profiles of snakes in nature. That, we learned, was a very incomplete picture of what was going on, once we compared that to data from telemetry. 

Response: We agree that snakes or behaviors that are most easily observed will likely make up the majority of the dataset and have reworded parts of the MS to acknowledge this. We disagree, however, with the notion that community science generated data is more biased in these regards to studies using traditional field methods (as indicated by the reviewer) and explained in more detail below. We have expanded upon the bias issues in a new paragraph in the Discussion on lines 342-352.

 

Lines 200-229: I was glad to see that the authors did address some of the bias issues here and agree with what they did in terms of reducing lack of independence. However, that does not address the comments above about accessibility sampling. 

Response: See above for our comments to this.

 

Lines 316-317: This is a good example of where these kinds of data should be regarded with much skepticism when using accessibility sampling as was done here. 

Response: See responses above and below.

 

Lines 336-340: Again, glad to see the authors recognize the apparent bias in their data set. However, I see no basis for the statement “these biases should be standardized across all age classes”. Indeed, I could argue that the opposite would be true. If, for example, small snakes are more active during the day, then they will be over-represented in the data set. 

Response: This sentence has been reworded following revisions. We agree that human activity biases could change how often juveniles vs. adults are seen based on differences in activity times between the snake age classes, but this should not greatly affect age differences in coil direction, microhabitat use, or perch use. We show that observations of adults and juveniles by human observers are relatively the same throughout the day suggesting one age class is not more likely to be observed at certain times of day than the other. It is possible that one age class is more active during the day vs. night, but again, we do not see how this would greatly affect coil directions of the snakes, microhabitat use and/or perch use during the day (we only looked at daytime observations). We have added some wording to clarify that our results are only of daytime observations of snake behaviors, and these could potentially change during the night.

 

As a note, we also want to mention that traditional field studies often use similar techniques as those conducted by community scientists such as surveying an area and noting behaviors of all observed individuals (e.g. on perch vs on ground), or only capturing individuals that are easy to find and capture (i.e. the bold individuals). Even in rattlesnake telemetry studies, results could be biased toward individuals that have body sizes large enough for surgical implantation of transmitters or towards individuals that were just easy to find. Every study will have some sampling bias based on the species’ biology and limitations of the field team. We have at least two paragraphs in the Discussion acknowledging these biases and have done our best to account for spatial biases and biases associated with observers. We argue that a community science dataset, that often has hundreds (sometimes thousands) of observations, will be more robust to noise within the data because of the inherently large sample size. Studies using traditional field methods also have sampling biases but low sample sizes.

 

In general, first, there are some errors in the Introduction (on Massasaugas, of all things), but that can be rectified easily. Second, the link between these data and management are weak at best and I think are misplaced.

Most importantly, however, is whether these "crowd-sourced" data are so inherently biased as to preclude their analysis

Response: We have addressed these comments above.

Round 2

Reviewer 2 Report

a very brief review of the author's responses made it appear that they made the appropriate changes.