GENERAL INFORMATION Title of dataset: Cow space use, activity, and THI Authors involved with sample collection, processing and/or analysis: Kareemah Chopra, Chenyu Zhang, Chunde Liu, Zhiwen Luo, Christopher K. Reynolds, Jonathan R. Amory, Zoe E. Barker, Andy Thurman, Edward A. Codling Contact Author: Kareemah Chopra ORCID ID: 0009-0003-1427-3384 Institution: University of Essex Address: Wivenhoe Park, Colchester, Essex CO4 3SQ, UK Email: km19088@essex.ac.uk Date of data collection: 2024-06-01 to 2024-12-01. Geographic location of data collection: The Centre for Dairy Research, Reading, England, UK. Funding: This research was funded by BBSRC (grant numbers: BB/X00824X/1 to JRA, BB/X008517/1 to EAC and BB/X009564/1 to CKR). SHARING/ACCESS INFORMATION Title of publication that uses the data: Associations between space-use behaviour and temperature-humidity index in barn-housed dairy cows METHODOLOGICAL INFORMATION Positioning data: A real-time local positioning system was used to track the spatial position and activity of a commercial dairy herd (c100 cows) in a barn continuously at high temporal resolution (0.125 Hz/ every 8 seconds). The time cows spent near key resource areas was calculated: fans, water troughs, cubicles, and in the feeding zone (FZ). Bunching was measured using four different spatial measures determined for each of four time periods (as below): full and core range size, inter-cow distance (ICD), and nearest neighbour distance (NND). Activity was also calculated using accelerometry data (recorded every 10 minutes). TIME PERIODS (TP) Four-hour blocks were used to summarise space-use and activity metrics. Milking periods were excluded (04:00–05:59 h and 14:00–15:59 h). | TP | Time Range (h) | |----|----------------| | 1 | 00:00–03:59 | | 2 | 06:00–09:59 | | 3 | 10:00–13:59 | | 4 | 16:00–19:59 | | 5 | 20:00–23:59 | Temperature-humidity index: Wall-mounted temperature and humidity sensors automatically recorded the ambient barn temperature and relative humidity continuously throughout the study period. A mean measurement was calculated across all sensors (n = 16 temperature-only sensors and n = 20 humidity sensors, recording every 10 minutes). Data selection: Extended interruption occurred on 23 of the study days (05/06/2024, 09/07/2024, 19/07/2024, 22/07/2024 to 26/07/2024, 12/08/2024 to 24/08/2024 and 27/10/2024) due to the system malfunctioning and resetting part-way through the day; these days were excluded from the subsequent analysis. The minimum and maximum numbers of unique cows with sensors present per day during the study period were 85 and 107, respectively. A total of 125,567,618 location data points were collected from these cows. Subsequent data pre-processing consisted of: 1) Removing location data outside of the cubicle yard due to minor positional inaccuracies (12,756,417 data points removed), 2) Removing nonsensical positional data (e.g., sensors were stuck in exactly the same or a similar, point location for multiple consecutive time points; 22 data points removed), 3) Plots of trajectories were visualised manually and any further nonsensical data was removed (344,011 data points removed points removed). A total of 112,467,168 data points remained for the analysis. The time spent near key resources (3m radius from the centre of any water trough, 3m radius from the centre of any fan, within the feeding zone and within 1m of the cubicles was calculated. The time spent within each area was summed across each time period, and a mean was calculated across cows. The time spent near the fans was also calculated using a secondary method accounting for the direction of airflow. Activity data determine overall activity by computing the mean of a filtered VeDBA (Vectorial Dynamic Body Acceleration). The mean was calculated using a sliding window of 32 data points for each sensor. The mean and variance in activity were calculated across cows for each time period. The distance travelled by each cow during high activity periods (activity > 2.15) was also determined for each time period, then a mean was calculated for each time period. Bunching metrics were derived from the raw positioning data. Range size was calculated by overlaying a virtual grid (2m x 2m = 4m2 cells) over the barn map. At each timestep, location data were used to assign each individual cow within the herd to a given cell. The number of individual data points assigned to each cell for each individual and time period were counted giving a final total for each virtual cell within the barn. The highest density cells cumulatively adding to 50% or 95% were used to create an hourly utility distribution corresponding to the core range (CR) and full range (FR), respectively. The number of unique cells included in each core and full range distribution were then defined as the range size. The mean and variance were calculated across individuals. Core and full range sizes were also calculated across the herd for each time period. Inter-cow distance (ICD) was calculated by finding the mean inter-cow distance at each time step for each of the possible dyad pairs across the herd. An hourly value was calculated as the arithmetic mean of all (360) values recorded over a given time period. The variance across all ICD values was also calculated for each time period. Nearest neighbour distance (NND) was calculated as the distance between a given individual and its closest neighbour in the herd. This corresponded to the smallest inter-cow distance when considering all dyad pairs involving that individual. A value was calculated across the full herd for each time period. Variance in ICD was also calculated, by calculating a mean distance for each timestep across cows, then the variance of these values was calculated for each time period. Similarly, variance in NND was calculated. DATA-SPECIFIC INFORMATION Number of variables: 22. Number of rows: 800 (including column names). Each row represents a date x TP observation. Variable List (note that all values are averages per day and time period): Column 1. date (DD/MM/YYYY). Column 2. time_period. Four-hour block time period (see table above) (integer (1–5)). Column 3. mean_THI. Mean temperature-humidity index (unitless) Column 4. mean_time_near fans. Mean time spent within 3m proximity of any fan across cows (minutes). Column 5. mean_time_near_fan_shapes. Mean time spent in areas around fans across cows (minutes). Column 6. mean_time_near_water_troughs. Mean time spent within 3m proximity of any water trough across cows (minutes). Column 7. mean_time_near_cubicles. Mean time spent within 1m of cubicles across cows (minutes). Column 8. mean_time_in_FZ. Mean time spent within the feeding zone across cows (minutes). Column 9. mean_activity. Mean activity index across cows (unitless). Column 10. variance_activity. Variance in activity index across cows (unitless). Column 11. mean_distance_high_activity. Mean distance travelled during high activity (metres). Column 12. mean_z. Mean z value across cows (metres). Column 13. mean_individual_CR. Mean individual core range (50%) (4m2 virtual cells) Column 14. variance_individual_CR. Variance in individual core range (50%) (4m2 virtual cells) Column 15. mean_individual_FR. Mean individual full range (95%) (4m2 virtual cells) Column 16. variance_individual_FR. Variance in mean individual full range (95%) (4m2 virtual cells) Column 17. herd_CR. Herd core range (50%) (4m2 virtual cells) Column 18. herd_FR. Herd full range (95%) (4m2 virtual cells) Column 19. mean_ICD. Mean inter-cow distance (ICD) across cows (metres). Column 20. variance_ICD. Variance in inter-cow distance (ICD) across cows (m2) Column 21. mean_NND. Mean nearest-neighbour distance (NND) across cows (metres). Column 22. variance_NND. Variance in nearest-neighbour distance (NND) across cows (m2)