Data from: Habitat selection and outdoor recreation explain human-carnivore conflict
Data files
Jul 16, 2025 version files 1.34 GB
-
1_morgan_et_al_issa_steps.csv
315.08 MB
-
2_morgan_et_al_issa_recreation.zip
1.01 GB
-
3_morgan_et_al_CA_conflict.csv
12.77 MB
-
README.md
7.92 KB
Abstract
Human-wildlife conflict occurs anywhere humans and wildlife overlap with significant socio-ecological consequences. As participation in outdoor recreation grows, so too does the potential for conflict between people and wildlife. This work investigates how outdoor recreation affects the movement behavior of a large carnivore and demonstrates the potential for human activity to lead to habituation in mountain lions (Puma concolor). Despite evidence of habituation, we found that human-tolerant individuals are not more likely to engage in conflict. Instead, mountain lion attacks on people, and other conflict events, are best explained by an overlap in high-quality mountain lion habitat and concentrated human presence. This work offers a mechanistic understanding of human-wildlife conflict and demonstrates behavioral pathways that contribute to human-wildlife coexistence.
Dataset DOI: 10.5061/dryad.q573n5tv4
Description of the data and file structure
The dataset provided can be used to reproduce results associated with the manuscript titled “Habitat selection and outdoor recreation explain human-carnivore conflict” by Morgan et al. 2025 (in review).
These data consist of .csv files originally analyzed using R (version 3.6.0). All GPS movement data were collected by the Santa Cruz Puma Project within California’s Santa Cruz Mountains, USA, between 2018 and 2023. GPS data have been stripped of location information due to the sensitive nature of mountain lion space use within the study area. Human outdoor recreation data were derived from Strava Metro data collected from 2018–2023. Recreation data represent on-trail recreation intensity at multiple unique spatial and temporal scales around each used and available GPS location. Further details on how recreation intensity was calculated can be found in the manuscript. Human-mountain lion conflict data are based on 23 confirmed mountain lion attacks on people occurring in California between 1986 and 2024. Additional types of human-mountain lion conflict, such as depredations, nuisance reports, potential conflict, and sightings, are included and were compiled by California Department of Fish and Wildlife from 2018–2024. Together these data can be used to model mountain lion behavioral response to outdoor recreation activity and to better understand the drivers of human-mountain lion conflict.
Files and variables
File: 1_morgan_et_al_issa_steps.csv
Description: This file contains all core data for fitting integrated step selection analyses (iSSA), including columns for pumaID, stepID, movement characteristics, and standardized habitat covariates. This file should be merged with the appropriate recreation file based on the recreation scale of interest. Files should be merged based on “unique_pt_id” column.
Variables
- unique_pt_id: unique identifier of each GPS relocation. Includes both used and available points.
- pumaID: individual mountain lion identifier
- sex: male/female
- case: used locations = 1, available locations = 0
- stepID: unique step identifier, each consisting of 1 used location and 20 available locations
- start_time_UTC: date and time at start of step, timezone = UTC
- end_time_UTC: date and time at end of step, timezone = UTC
- year: calendar year of movement step
- sl_km: step length in kilometers
- sl_log_km: natural log of step length in kilometers
- ta_rad: turning angle in radians; Note: this column contains ‘NA’ values (not applicable) for locations that do not have have a valid relocation immediately prior and therefore no turning angle could be calculated.
- hd_150_end_std: standardized housing density at end of step
- slope_end_std: standardized slope at end of step
- cover_end_std: standardized % cover at end of step
- dist_urban_end_std: standardized distance to nearest urban edge at end of step
- dist_trail_end_std: standardized distance to nearest trail at end of step
File: 2_morgan_et_al_issa_recreation.zip
Description: This folder contains outdoor recreation intensity data for all spatial and temporal scales tested. Each file is named according to the temporal scale at which recreation was calculated. Within each file there are columns named based on the spatial scale tested. For example, the file “rec_dat_01hr_AVG365da.csv” refers to the mean hourly recreation intensity based on the previous 365 days of trail use data. Within this file, the column “avg_rec_all_30_end” refers to the recreation intensity within 30 m of the end of each step. Merging this file with 1_morgan_et_al_issa_steps.csv will provide data necessary for fitting top performing iSSA model.
- rec_dat_12mo.csv - cumulative recreation intensity based on previous 12 months of trail use
- rec_dat_30da.csv - cumulative recreation intensity based on previous 30 days of trail use
- rec_dat_01da.csv - cumulative recreation intensity based on previous 1 day of trail use
- rec_dat_24hr.csv - cumulative recreation intensity based on previous 24 hours of trail use
- rec_dat_04hr.csv - cumulative recreation intensity based on previous 4 hours of trail use
- rec_dat_01hr.csv - cumulative recreation intensity based on previous 1 hour of trail use
- rec_dat_04hr_AVG07da.csv - average recreation intensity over previous 4 hours based on previous 7 days of trail use
- rec_dat_04hr_AVG30da.csv - average recreation intensity over previous 4 hours based on previous 30 days of trail use
- rec_dat_04hr_AVG90da.csv - average recreation intensity over previous 4 hours based on previous 90 days of trail use
- rec_dat_04hr_AVG180da.csv - average recreation intensity over previous 4 hours based on previous 180 days of trail use
- rec_dat_04hr_AVG365da.csv - average recreation intensity over previous 4 hours based on previous 365 days of trail use
- rec_dat_01hr_AVG07da.csv - average recreation intensity over previous 1 hour based on previous 7 days of trail use
- rec_dat_01hr_AVG30da.csv - average recreation intensity over previous 1 hour based on previous 30 days of trail use
- rec_dat_01hr_AVG90da.csv - average recreation intensity over previous 1 hour based on previous 90 days of trail use
- rec_dat_01hr_AVG180da.csv - average recreation intensity over previous 1 hour based on previous 180 days of trail use
- rec_dat_01hr_AVG365da.csv - average recreation intensity over previous 1 hour based on previous 365 days of trail use
File: 3_morgan_et_al_CA_conflict.csv
Description: This file contains human-mountain lion conflict data across the state of California and was used to fit generalized linear models predicting the drivers of conflict. Attack and non-attack locations are defined in the “case_attack” column, while wildlife incident report events are defined in the “case_wir” column. Recreation intensity for each cell is defined by “rec_std”, while “rsf_fix_std”, “rsf_min_std”, “rsf_max_std” refer to the predicted relative probability of mountain lion space use based on population-level fixed effects, as well random effect estimates representing shy and bold individuals in our top fitting iSSA model.
Variables
- cell_id: unique identifier for each used and available location across CA
- case_attack: attack locations = 1, available locations = 0
- case_wir: locations with wildlife incident report = 1, available locations = 1
- rec_std: standardized estimate of recreation intensity
- rsf_fix_std: predicted relative probability of mountain lion use based on fixed effects in top performing iSSA model
- rsf_min_std: predicted relative probability of mountain lion use based on lowest observed levels of human tolerance. Derived from individual-level random effects extracted from top performing iSSA model.
- rsf_max_std: predicted relative probability of mountain lion use based on highest observed levels of human tolerance. Derived from individual-level random effects extracted from top performing iSSA model.
Code/software
All models were run using R (version 3.6.0).
iSSA models were fit using the function ‘fitTMB’ in the glmmTMB R package.
Generalized linear models were fit using the function ‘glm’ in the stats R package.
Access information
Data related to mountain lion attacks on people were derived from California Department of Fish and Wildlife:
California statewide recreation intensity estimates were derived from Strava’s global heatmap: