Skip to main content
Dryad

Interacting effects of surface water and temperature on wild and domestic large herbivore aggregations and contact rates

Cite this dataset

Titcomb, Georgia et al. (2024). Interacting effects of surface water and temperature on wild and domestic large herbivore aggregations and contact rates [Dataset]. Dryad. https://doi.org/10.5061/dryad.02v6wwqc5

Abstract

Earth’s climate is rapidly changing, bringing forth questions of how domestic and wild animals will alter their behavior in response to increasing temperatures and dryland expansion. Dwindling water availability will likely impact animal behavior and water foraging, potentially increasing animal aggregations and interspecific contacts. These interspecific contacts are especially important for competition, predation, and disease transmission among wildlife and domestic animals.

In this study, we analyzed interspecific wildlife and cattle contacts using two years of camera trap data at an experimental water manipulation site at a conservancy in central Kenya.

We found that on average, the hourly probability of any interspecific contact was approximately 3.4 times higher at water sources versus drained water sources, and 18 times higher than surrounding matrix areas, and that this relationship was amplified by dry and hot conditions.

Species-specific analyses revealed variation in the magnitude of responses across wildlife and domestic cattle, although all animals had approximately 2-3 times higher interspecific contact probability with other species at water in hot conditions versus other conditions. Notably, we observed the largest behavioral changes for relatively water-independent species, such as giraffe, which had 3.6 times higher interspecific contact probability at water sources in hot versus other conditions.

Synthesis and applications: These findings show how elevated temperatures that will become increasingly common with future climate changes can increase interspecific contacts around critical water resources. In mixed wildlife-livestock systems, maintaining wildlife-only water sources may be a practical management tool to mitigate human-wildlife conflict and disease transmission at this interface, especially during dry and hot conditions.

README: Interacting effects of surface water and temperature on wild and domestic large herbivore aggregations and contact rates

https://doi.org/10.5061/dryad.02v6wwqc5

This dataset includes data derived from camera traps set at Ol Pejeta Conservancy in Laikipia County, Kenya. The contents of the package are as follows:

  1. avg_hrly_temp: a dataframe containing mean hourly temperatures across sites.
  2. contact_analysis_2024_functions:* *R functions to fit species-specific models (called by contact_analysis_2024_publication)
  3. contact_analysis_2024_publication: Code used to produce the results presented in the main manuscript
  4. contact_by_species_by_hour: a dataframe containing average hourly activity across sites at Ol Pejeta Conservancy for all periods that cameras were running in a given location. These data are used to fit species-specific models.
  5. contact_data: a dataframe containing animal activity data for all camera triggers that were produced by an animal.
  6. opc_monthly_rainfall: a dataframe containing monthly rainfall data at Ol Pejeta Conservancy
  7. trap_nights: a dataframe containing information on the number of trap nights for each camera trapping deployment.

Description of the data and file structure

The dataframes are called by the main R script contact_analysis_2024_publication. This script will also call additional functions available as source code. Running the R script will produce all plots and tables reported in the publication. Further details on dataframes are provided below:

  • avg_hrly_temp
    • Location: One of five locations at Ol Pejeta Conservancy where the study took place.
    • Treatment: One of three experimental water treatments {"Matrix" = a random location 1km from water; "Drained" = an experimentally-drained water pan; "Filled" = a filled water pan}
    • status: One of three experimental water statuses {"Pre" = the period prior to draining water from the "Drained" pan; "During" = the period when the "Drained" pan was drained; "Post" = the period after water was reinstated in the "Drained" pan}
    • Day: The calendar day of the month {1:31}
    • Month: The calendar month {1:12}
    • Year: The year {2016:2018}
    • Hour: The hour of the day {0:23}
    • tempC_cam1: The mean temperature recorded across camera traps for a given hour. NA values indicate that no temperature was available. Units are degrees Celsius.
    • tempC_cam_imputed: The imputed temperature for all hours over the study period. Units are degrees Celsius.
    • bias_corrected: The bias-corrected temperature for all hours over the study period. Units are degrees Celsius.
  • contact_by_species_by_hour
    • Location, Treatment, Year, Month, Day, Hour, status: descriptions as above
    • trigger: the unique trigger ID for each trigger event. Note that trigger ID is NA when there were hours with no trigger.
    • BABOON:ZEBRAPLAINS: A binary indicator {0 or 1} of animal presence for each hour/location/treatment combination.
    • n_sp: The number of unique species that were present for each hour/location/treatment combination.
    • inter_contact: A binary indicator {0 or 1} of interspecific contact for each hour/location/treatment combination.
    • bias_corrected: the bias-corrected temperature in degrees Celsius.
    • rainfall_mm: the monthly rainfall (mm)
    • rainfall_mm100: the monthly rainfall divided by 100 for interpretability in model-fitting. (mm/100)
    • hot_day: a binary indicator {0 or 1} of whether or not the day was in the top 25% of warmest days in the dataset.
  • contact_data
    • subject_set_id: The identifier for the group of photographs uploaded and analyzed on Zooniverse
    • Deployment_Name: The descriptive identifier of photographs taken without interruption at a given site.
    • Location: One of five locations at Ol Pejeta Conservancy where the study took place.
    • Treatment: One of three experimental water treatments {"CONT", equivalent to "Matrix", a random location 1km from water; "WPM", equivalent to "Drained", an experimentally-drained water pan; "WPC", equivalent to "Filled", a filled water pan}
    • status: One of three experimental water statuses {"Pre" = the period prior to draining water from the "Drained" pan; "During" = the period when the "Drained" pan was drained; "Post" = the period after water was reinstated in the "Drained" pan}
    • Trap_Nights: The number of trap nights for the given deployment
    • Start_Date: The start date of the deployment
    • End_Date: The end date of the deployment
    • Year, Month, Day, Hour, trigger: As described above
    • triggerID: The trigger ID combined with its corresponding deployment
    • WARTHOG:ORYX: A binary indicator {0 or 1} of animal presence for each trigger set (triggerID).
    • n_p: The number of species present in a trigger set (triggerID).
    • anycontact: A binary indicator {0 or 1} of interspecific contact for each trigger set (triggerID).
  • opc_monthly_rainfall:
    • Year, Month: As described above
    • rainfall_mm: The monthly rainfall (mm) collected at Ol Pejeta Conservancy.
  • trap_nights:
    • Location: One of five locations at Ol Pejeta Conservancy where the study took place.
    • Treatment: One of three experimental water treatments {"CONT", equivalent to "Matrix", a random location 1km from water; "WPM", equivalent to "Drained", an experimentally-drained water pan; "WPC", equivalent to "Filled", a filled water pan}
    • status: One of three experimental water statuses {"Pre" = the period prior to draining water from the "Drained" pan; "During" = the period when the "Drained" pan was drained; "Post" = the period after water was reinstated in the "Drained" pan}
    • n: The number of trap nights.

Code/Software

There are two R scripts provided to run all the analyses in the manuscript. contact_analysis_2024_publication calls contact_analysis_2024_functions to fit various models. Scripts should be compatible with R version 4.2.1 and later. The required packages and versions are:

  • ggpubr 0.4.0
  • tidyverse 2.0.0
  • emmeans 1.8.2
  • glmmTMB 1.1.4
  • lubridate 1.9.2
  • DHARMa 0.4.5
  • igraph 1.3.4
  • ggthemes 4.2.4

Funding

National Science Foundation, Award: 1556786, DEB

National Geographic Society, Award: EC-33R-18