Data from: Effects of summer weather and heatwaves on wild boar activity
Data files
Jun 13, 2025 version files 1.49 MB
-
GAM1_data_daily.csv
1.35 MB
-
GAM2_data_seasonal.csv
112.92 KB
-
GLM_data_heatwaves.csv
1.88 KB
-
README.md
3.53 KB
-
Rscript.R
23.55 KB
-
temp_precip_history.csv
523 B
Abstract
Climate change threatens wildlife species, negatively affecting their fitness through environmental change, i.e., through increased severity of droughts and summer heatwaves. Wild boar, a species with limited physiological thermoregulation abilities, is potentially vulnerable to high temperatures during summer. Yet, little is known about the behavioural reactions of this species to heat stress. A detailed understanding of wild boar behavioural adaptations to their environment might help understand their future population growth and change in the geographical range. We used multisensory collars on 24 individual wild boar in the Czech Republic, calculating the dynamic body acceleration as a proxy for energy expenditure to detect activity changes in response to high temperatures on two temporal scales (daily and seasonal) and heatwaves. Our results revealed that overall, under higher temperatures, wild boar reduce their activity, unless it rained. Heatwave intensity did not affect wild boar activity. We suggest that wild boars adapt their activity to weather conditions and presume the importance of sufficient precipitation for thermoregulation in this species. Additionally, this research shows the potential of remote-sensing technologies to monitor wildlife behaviour, particularly in challenging observational scenarios, offering valuable insights into the behavioural responses of wildlife in the face of a changing climate.
Authors
Justine Güldenpfennig, Niccolò Fattorini, Miloš Ježek, Kevin Morelle, Tomasz Podgórski
Description of the data and file structure
This README describes the columns of the CSV files used to generate statistics and figures in the above-mentioned publication.
“GAM1_data_daily.csv” collects the data used in the first model to predict the effects of daily temperature and time on the hourly vectorial sum of dynamic body acceleration (VeDBA).
“GAM2_data_seasonal.csv” collects the data used in the second model to predict the effects of seasonal temperature and time on the daily vectorial sum of dynamic body acceleration (VeDBA).
“GLM_data_heatwaves.csv” collects the data used in the third Model used to examine the effect of heatwave intensity on the vectorial sum of dynamic body acceleration (VeDBA).
“Rscript” analyzes how animal activity (measured as VeDBA) is influenced by temperature, time of day, seasonality, and precipitation using generalized additive models (GAMs) and GLMs. It includes model fitting, predictions, visualizations, and post-hoc analyses to assess behavioral responses to weather and heatwaves.
“temp_precip_history.csv” collects the data used to visualise historical data about temperature and precipitation in the Czech Republic.
Cells with “NA” represent not unavailable data.
The columns are as follows:
“GAM1_data_daily.csv”
animalID - Unique identifier for each tracked animal
datetime - Date and time the VeDBA is summarised for
VeDBA_sum - The hourly vectorial sum of dynamic body acceleration in g
temp - The temperature measurement in °C
precip - The measurement of precipitation volume in mm
daynight - Categorises if the hour falls within daytime (“day”; sunrise to sunset) or nighttime (“night”; sunset to sunrise)
year - The year of the date
“GAM2_data_seasonal.csv”
animalID - Unique identifier for each tracked animal
date - The date the VeDBA is summarised for
VeDBA_mean_daily - The daily average of the hourly vectorial sum of dynamic body acceleration in g
Tmax - The daily maximum temperature in °C
Tmean - The daily average temperature in °C
precip_total - The daily total precipitation volume in mm
Julian - Day of the Year
heatwave - Binary category for within heatwave (1) or outside of heatwave (0)
seq_id - Continuous identifier for each heatwave
seq_nr - Continuous number of days within each heatwave (per seq_id)
seq_length - Maximum number of days within each heatwave (per seq_id)
SL-sum - daily distance traveled calculated using the sum of step length (distance between consecutive GPS locations)
NSD_max - daily maximum net squared displacement calculated using GPS locations
speed_mean - daily average speed calculated using GPS locations
“GLM_data_heatwaves.csv”
animalID - Unique identifier for each tracked animal
seq_id - Continuous identifier for each heatwave
VeDBA_mean_mean - The mean daily average of the hourly vectorial sum of dynamic body acceleration within each heatwave in g
Tmax_mean - The average daily maximum temperature within each heatwave in °C
precip_total - The total precipitation volume within each heatwave in mm
seq_length - Maximum number of days within each heatwave (per seq_id)
year - The year each heatwave was registered
“temp_precip_history.csv”
month - The month of the year
year - The year of measurement
Tmean - The average temperature in °C
precip - The total precipitation in mm
Code/Software
R environment version 4.2.2
We collected tri-axial acceleration data from multisensory collars mounted on free-ranging wild boar. The tags (Wildbytes Technologies Ltd.) recorded data in 10 Herz frequency. The raw data was then processed using the supporting DDMT software (Wildbytes Technologies Ltd.) and exported as 30-minute sums of the vectorial sum of the dynamic body acceleration (VeDBA). All further processing was performed within the R environment (R version 4.2.2). Weather data was downloaded from the Visual Crossing Weather Query Builder. We first build a data set containing the hourly sums of the VeDBA, the hourly temperature, and the hourly sum of precipitation volume. In the second data set, we generated the average daily VeDBA, adding the daily maximum and mean temperature, the sum of precipitation, and if the date fell within or outside of heatwaves (days over 24 degrees Celsius). The third data set contains the average daily mean of VeDBA only within heatwaves, as well as the length of the heatwave, the maximum temperature within each heatwave, and the total precipitation.
Ultimately, we used Generalised Additive Mixed Models (GAMMs) and Generalised Linear Mixed Models (GLMMs) to statistically test the effects of weather on the VeDBA in two temporal scales and the intensity of heatwaves.