From fear to feast: Rattlesnakes navigate the landscape of fear to optimize foraging
Data files
Sep 08, 2025 version files 2.85 MB
-
Beh_Proba_all.csv
41.84 KB
-
DetectionY.csv
986 B
-
Game_Camera_DR_31AUG23.csv
1.15 MB
-
MergedModelPrey.csv
102.70 KB
-
OFT_Stats_Feb25.Rmd
39.47 KB
-
opreysummer.csv
49.08 KB
-
overlap.csv
306.78 KB
-
overlapFall.csv
150.89 KB
-
OverlapPrey.csv
135.68 KB
-
overlappreyfall.csv
63.06 KB
-
overlappreyspring.csv
23.63 KB
-
overlapSpring.csv
53.51 KB
-
overlapSummer.csv
102.48 KB
-
PredicSnakePred.csv
49.54 KB
-
PredicSnakePrey17JAN.csv
44.21 KB
-
PreyUsable.csv
475.42 KB
-
README.md
5.95 KB
-
RecordingDays_GC.csv
461 B
-
TrainTest_Pred_noNA2.csv
60.72 KB
Abstract
According to optimal foraging theory, mesopredators should forage in areas where their prey is abundant while avoiding high predation risk. Here, we investigate how environmental factors influence mesopredators’ abilities to minimize spatiotemporal overlap with predators while increasing spatiotemporal overlap with prey. We paired 30 western diamond-backed rattlesnake (Crotalus atrox) 3D-printed replicas with game cameras in West Texas for two years to quantify several spatiotemporal factors affecting prey availability and predation risk. Concurrently, 25 Crotalus atrox were radiotracked at the same site to gather activity and microhabitat selection data regarding free-ranging individuals. Random forest algorithms were trained using data obtained from the game camera and applied to predict the probability of predation and the probability of prey encounter for each radiotracking event. Time of day, month, vegetation structure, and concealment percentage all had a significant association with the probability of predation and the probability of prey encounter. Our results suggest that rattlesnakes choose to be active when and where the probability of prey encounter is significantly higher than the probability of predator encounter, thus following optimal foraging theory. Our results demonstrate that mesopredators increase the chances of prey capture while reducing predator detection in natural settings.
Dataset DOI: 10.5061/dryad.931zcrjz1
Description of the data and file structure
Da Cunha O., Dominguez R. P., Horne L. M., Mead J. J., Fournier C., Johnson J. D., and Seymoure B. M. From Fear to Feast: Rattlesnakes Navigate the Landscape of Fear to Optimize Foraging. In Press. Submitted to The American Naturalist on Nov. 15th, 2024.
Authors: Da Cunha O., Dominguez R. P., Horne L. M., Mead J. J., Fournier C., Johnson J. D., and Seymoure B. M. University of Texas at El Paso, TX, USA.
Corresponding authors: Oceane Da Cunha (odacunha2@utep.edu).
Statistical analyses conducted by Oceane Da Cunha.
Files and variables
Empty cells indicate missing or unavailable data.
File: RecordingDays_GC.csv
Description: File containing the number of recording days per game camera
Variables
- Camera: Camera ID
- Days: Number of recording days
- Location: Location of the camera
File: Game_Camera_DR_31AUG23.csv
Description: File containing all game camera observations and microhabitats associated with observations. NA stands for "Not applicable".
Variables
- Camera_Number: Camera ID
- Number: Camera number
- Date: Date of observation
- Time: Time of observation
- Time2: Adjusted/alternative time format
- Common_name: Common species name
- Scientific_name: Scientific species name
- Predation_Y_N: Yes or No
- Detection: Yes or No
- Notes: Additional notes
- Rock_per: Percentage of rock
- Rock_type: type of rock
- DNR: Distance to nearest rock (meters)
- Soil_per: Percentage of exposed substrate
- Soil_type: type of soil
- DNB: Distance to nearest burrow (meters)
- DNB_type: Type of burrow
- Vegetation_per: Percentage of vegetation
- Vegetation_type: vegetation species
- Vegetation_height: in meters
- DNV: Distance to nearest vegetation (meters)
- DNV_type: species of nearest vegetation
- DNV_height: height of nearest vegetation (meters)
- Concealment: Percentage of concealment
- Conc_source: Type of concealment
- Location: Location on the map
- Location2: Tank or desert
- Model_size: small or big
- Month: Month of observation
- TimeS: hours of observation
- Predator_Prey_Other: Classification of observed species
- Rodent: yes or no
- Prey: yes or no
File: DetectionY.csv
Description: Detection observations
Variables
- Scientific_name:
- n: number of observations
File: TrainTest_Pred_noNA2.csv
Description: File containing telemetry (test) and game camera observations (train) for predator detection to input in RF models. NA stands for "Not applicable".
Variables: see previous variables in Game_Camera_DR_31AUG23.csv, plus:
- ID: Snake identification or game camera number
- Set: training or test set for random forest
- LocationN: GPS coordinates
- LocationE: GPS coordinates
File: PredicSnakePred.csv
Description: File with predictions of detection by predators obtained from the RF model
Variables: See Game_Camera_DR_31AUG23.csv and TrainTest_Pred_noNA2.csv plus:
- LocationEN: additional GPS coordinate references
- N: Probabilities for no detection
- Y: Probabilities for detection
File: PreyUsable.csv
Description: File with prey observation to input in RF models. NA stands for "Not applicable".
Variables: see Game_Camera_DR_31AUG23.csv plus:
- Usable: Yes or No
File: MergedModelPrey.csv
Description: File containing telemetry (test) and game camera observations (train) for prey encounter to input in RF models
Variables: see previous variables in Game_Camera_DR_31AUG23.csv, TrainTest_Pred_noNA2.csv, and PreyUsable.csv
File: PredicSnakePrey17JAN.csv
Description: File with predictions of prey encounter obtained from RF models
Variables: see previous variables in see previous variables in Game_Camera_DR_31AUG23.csv, TrainTest_Pred_noNA2.csv, and PreyUsable.csv
Overlap Files (overlap.csv, opreysummer.csv, overlapFall.csv, OverlapPrey.csv, overlappreyfall.csv, overlappreyspring.csv, overlapSpring.csv, overlapSummer.csv)
Description: Observations and time stamps to calculate seasonal or prey-specific activity overlap.
Variables:
- Camera_Number : Camera ID
- Date: Date of observation
- Time2: Time of observation (adjusted format)
- Scientific_name: Scientific species name
- type: Type of animal (e.g., predator, prey, other)
File: Beh_Proba_all.csv
Description: File with probability of predator detection and prey encounter obtained from RF per rattlesnake behavior
Variables
- Individual_ID: snake identification
- Beh2: Type of behavior
- Type: predator or prey
- Proba: probability of detection or encounter
File: OFT_Stats_Feb25.Rmd
Description: R code for all analyses
Code/software
R version 4.2.2 (2022-10-31 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 26100)
attached base packages:
stats
graphics
grDevices
utils
datasets
methods
base loaded via a namespace (and not attached):
rstudioapi_0.14
knitr_1.42
magrittr_2.0.3
tidyselect_1.2.0
munsell_0.5.0
colorspace_2.1-0
R6_2.5.1
rlang_1.1.0
fastmap_1.1.0
fansi_1.0.4
dplyr_1.1.1
tools_4.2.2
ggdist_3.3.0
grid_4.2.2
gtable_0.3.3
xfun_0.36
utf8_1.2.3
cli_3.6.0
htmltools_0.5.4
yaml_2.3.7
digest_0.6.31
tibble_3.2.1
lifecycle_1.0.3
farver_2.1.1
ggplot2_3.5.1
gghalves_0.1.4
vctrs_0.6.1
evaluate_0.20
glue_1.6.2
rmarkdown_2.21
compiler_4.2.2
pillar_1.9.0
generics_0.1.3
scales_1.3.0
distributional_0.3.2
pkgconfig_2.0.3