Data from: Large carnivore feeding is resilient to human disturbance
Data files
Sep 10, 2025 version files 1.21 MB
-
Clus_Summary.csv
435.04 KB
-
Cluster_Investigations.csv
328.12 KB
-
Dryad_Disturbance_Code.R
14.20 KB
-
FeedRates.csv
106.02 KB
-
Forays_Undisturbed.csv
11.40 KB
-
Joined_All.csv
312.99 KB
-
README.md
7.19 KB
Abstract
Impacts of anthropogenic disturbance on wild populations have the potential to alter natural systems. Humans are a main source of mortality for terrestrial carnivores that far exceeds natural mortality, which may lead to commensurate antipredator responses of carnivores toward humans. These behavioural changes have been implicated in trophic cascades, resulting in ecological alterations due to changes in space use or altered kill rates by predators to offset resource losses. These implications highlight the importance of evaluating behavioural responses of carnivores to human disturbance. While studies have implemented both correlative analyses and experimental proxies for human disturbance to assess these questions, here we leveraged an approach using direct human encounters at feeding sites of mountain lions in central Wyoming, USA. Counter to previous studies, mountain lions responded to disturbances through extended time budgets dedicated to feeding sites, >80% return rates after disturbances, and ultimately a marked reduction in feeding rate estimates. Additionally, we found evidence of behavioural flexibility, as the probability of mountain lions returning to feeding sites positively correlated with the cumulative disturbances they incurred. We discuss factors likely to influence state-dependent antipredator responses to disturbance, potential discrepancies across studies, and question the ramifications of behavioural responses to carnivore fitness due to human disturbance via encounters in natural systems.
Dataset DOI: 10.5061/dryad.tb2rbp0c7
Description of the data and file structure
Files and variables
File: Clus_Summary.csv
Description: Complete summary output from GPSeqClus clustering algorithm
Variables
- AID: animal identification
- clus_ID3: cluster identification number
- clus_start: timestamp of cluster start
- clus_end: timestamp of cluster end
- clus_dur_hr: cluster duration in hours (clus_end – clus_start)
- n_clus_locs: number of locations in the cluster
- visits: number of times (bouts) the cluster was visited
- fix_succ_clus_dur: the fix success during the cluster duration (total / successful attempts)
- adj_clus_locs: locations adjusted for fix success (n_clus_locs / fix_succ_clus_dur)
- Fid: fidelity to the cluster (m) (n_clus_locs – locations away from cluster)
- max_foray: maximum foray from cluster centroid (m)
- clus_radius: maximum cluster radius (m)
- avg_clus_dist: average distance of cluster locations from the centroid (m)
- n_24_per: unique 24 hr periods with at least one cluster location
- bin_24hr: binary variable whether the cluster duration exceeded 24 hr (0 = less than or equal to 24 hr, 1 = greater than 24 hr)
- season: 0 = 15 Nov - 14 May , 1 = 15 May - 14 Nov
- season2: 0 = 1 Nov - 30 April, 1 = 1 May - 31 Oct
- night_pts: number of night locations
- night_prop: proportion of night locations (night_pts / n_clus_locs)
File: Joined_All.csv
Description: merged and aligned data from all model-predicted feeding events from cluster summary and other investigations. Represents unique investigations of each prey item (includes rows for multiple kills at a single site). Includes investigations where the cluster shifted from predicted kill to no kill with added data, but was already investigated, also includes 1 investigation where we captured the animal off a kill (reported as cluster number 0). The dataset includes variables from Clus_Summary.csv with additional data columns described below.
Variables
- Lion.Sex: mountain lion sex
- Ch_Date: investigation timestamp
- Ch_by: investigator
- clus_shift: timestamp of cluster shifts or multiple carcasses at a site
- Mod_pred: identifies if the cluster was ultimately predicted to be a feeding event
- DetectionTime: time difference from the first cluster location until model prediction of feeding event (hrs)
- Check_FirstLoc: time difference from the first cluster location until an investigation occurred (hrs)
- Check_Detected: time difference from model prediction until an investigation occurred (hrs)
- Check_LastLoc: time difference from the investigation until the last cluster location (hrs)
- Cluster.Description: what was found during cluster investigation (e.g., kill, scavenge, day bed, nothing)
- Carcass.Present: carcass present at cluster (Y/N)
- Species: prey species
- Sex: prey sex
- Final_Class: age classification of prey
- Observed_lion: was a mountain lion observed during investigation (Y/N)
- X..Consumed: estimate of how much of the carcass was consumed at the time of investigation (1-5)
- Cover.Amount: estimate of cover at the carcass (1-5)
- Cent_to_Carc: distance from the cluster centroid to the carcass (m)
- checked_during: investigation occurred within the handling duration (T/F)
- Lion.Age: estimated mountain lion age during the feeding event
- lion.class: estimated mountain lion age class during feeding event
- live_weight: estimated weight of prey (kg)
- Disturbance: estimated human disturbance during feeding event
- next_clus_dist: subsequent cluster location distance relative to investigation (m)
File: Cluster_Investigations.csv
Description: subset of Joined_All.csv restricted to only investigated sites. The dataset includes variables from Joined_All.csv with additional data columns described below.
Variables
- Disturbance: estimated human disturbance during the feeding event
- prev_loc_time: previous location time relative to investigation (hrs)
- prev_loc_dist: previous location distance relative to investigation (m)
- prev_clus_time: previous cluster location time relative to investigation (hrs)
- prev_clus_dist: previous cluster location distance relative to investigation (m)
- next_loc_time: subsequent location time relative to investigation (hrs)
- next_loc_dist: subsequent location distance relative to investigation (m)
- next_clus_time: subsequent cluster location time relative to investigation (hrs)
- next_clus_dist: subsequent cluster location distance relative to investigation (m)
- nearest_return_to_carcass: nearest distance returned to carcass after disturbance (m)
- Dist_Type: disturbance classification
- max_flight_dist: maximum distance from carcass before returning to the feeding site (m)
- n_disturb_feed: cumulative number of disturbances incurred by individuals during feeding events
- n_disturb_all: cumulative number of disturbances incurred by individuals
- live_weight: estimated weight of prey (kg)
- bio_rem: estimated biomass remaining at the time of investigation (kg)
File: FeedRates.csv
Description: The feeding interval data includes rows identifying sequential feeding events by mountain lions.
Variables
- AID: animal identification and cluster number of the first event
- Lion.Sex: sex of mountain lion
- Lion.Age: estimated age of mountain lion at time of feeding event
- first_kill: prey species of the first feeding event
- fk_sex: prey sex of first feeding event
- fk_class: prey age class of first feeding event
- fk_type: type of feeding identified for the first feeding event
- fk_cover: estimated cover of carcass at first feeding event
- fk_season: 0 = 15 Nov - 14 May , 1 = 15 May - 14 Nov of first feeding event
- fk_live_weight: live weight estimate of prey of the first feeding event
- fk_Disturb: human disturbance at first feeding event
- fk_time: first location time of first feeding event
- fk_end_time: last location time of first feeding event
- dur: handling duration of first feeding event
- second_kill: sequential feeding event prey species
- nk_time: first location time of sequential feeding event
- KI: feeding interval estimate (difference between first location times of sequential feeding events; hrs)
- KI_fix: fix the success rate during the feeding interval
- IKI: intra feeding interval (difference between the last feeding location and the sequential first feeding location; hrs)
- Satellite: did these feeding events overlap
File: Forays_Undisturbed.csv
Description: data includes rows identifying forays from feeding sites not disturbed by investigators.
Variables
- AID: animal identification
- clus_ID: cluster identification number
- time.diff.hrs: time spent away from the feeding site on the foray (hrs)
- foray_dist: maximum distance from the carcass during the foray (m)
Code and software
File: Dryad_Disturbance_Code.R
Description: Associated R script to procure summary metrics, model outputs, and figures. Analysis run in R v. 4.0.5.
