Daily GPS movement and activity of free-roaming male cats before and after gonadectomy — a longitudinal within-individual study (Barlovento, La Palma)
Description
This dataset contains daily GPS movement and accelerometer-based activity data for twelve free-roaming cats from a managed (trap–neuter–return) community-cat colony in Barlovento, northern La Palma, Canary Islands, Spain (UNESCO Biosphere Reserve). It supports a longitudinal, within-individual study of how gonadectomy affects ranging in male cats: three intact males (two adults and one juvenile) were tracked before and after neutering, while nine already-neutered cats served as a contemporaneous control band. Cats were fitted with Tractive GPS-and-accelerometer collars (median fix interval ≈ 5 min). Monitoring was intermittent, comprising a 15-day pre-surgical baseline (12–26 July 2025) and three post-surgical windows (6 Aug–8 Sep 2025; 20 Dec 2025–9 Jan 2026; 25 Mar–3 Apr 2026), giving 80 monitored days per cat (960 daily records). The package includes: the daily data in tidy long format (CSV) and documented Excel; per-animal metadata; the raw GPS tracks (KML/GPX) used for home-range estimation, renamed to anonymised codes; derived outputs (linear mixed-model results, home-range areas); and the analysis scripts (Python and R) that reproduce the published results. Individuals are identified by anonymised codes; all values are real, device-recorded measurements. Ethical approval: OEBA_ULPGC_35/2023 (University of Las Palmas de Gran Canaria).
Files
Steps to reproduce
1. Site and animals. A managed (trap-neuter-return) free-roaming community-cat colony with individually identifiable cats and a fixed feeding station as spatial reference (Barlovento, La Palma, Canary Islands, Spain; UNESCO Biosphere Reserve). Record sex, baseline neuter status and age. 2. Groups. Three intact males to be neutered during the study (two adults, one juvenile) and nine cats already neutered before the study, which control for season and weather. 3. Devices. GPS-and-accelerometer collars (Tractive GPS CAT Mini; IP68; 28 g), standard tracking with a fix about every 5 min. 4. Baseline (PRE). Monitor all cats ~2 weeks before surgery (12-26 July 2025; 15 days). 5. Intervention. Gonadectomise the selected intact males (~29 July 2025); leave controls unchanged. 6. Post (POST). Resume after recovery. As battery life is short in these active cats, monitor in discrete windows and rotate two collars per animal where needed. Three windows used (6 Aug-8 Sep 2025; 20 Dec 2025-9 Jan 2026; 25 Mar-3 Apr 2026), giving 65 POST and 80 total monitored days per cat. 7. Variables. Per cat and day, distance (km) and active time (min), each split into day (07:00-21:00 h) and night, plus daily temperature and rainfall. GPS tracks were exported for home-range estimation, but raw coordinates are not released (privacy/colony safety); home ranges are provided as areas only. 8. Ethics. Approval OEBA_ULPGC_35/2023 (University of Las Palmas de Gran Canaria). REPRODUCING THE ANALYSIS FROM THE DEPOSITED FILES (Python 3 with numpy and pandas; run inside scripts/) a. Mixed model per response variable: python run_lmm.py km_total km_night km_day min_total min_night min_day Uses data_long.csv; writes lmm_results.json (period x intervention estimate, 95% CI, p, R-squared), reproducing derived_outputs/Table2_LMM.csv. b. Age at neutering vs home range: python corr_age_homerange.py (uses homerange_all12.csv, areas in hectares). c. Optional, R (lme4/lmerTest/MuMIn) for Satterthwaite p-values: LMM_reproducible.R on data_long.csv. Model: response ~ period * intervention + temperature + rainfall + (1 | animal); the period x intervention interaction estimates the neutering effect. Home-range areas were computed with hr.py from the per-fix tracks; recomputation needs those coordinates, withheld for privacy and available de-identified from the corresponding author on request. All values are real device measurements; none imputed or simulated.
Institutions
- Universidad de Las Palmas de Gran CanariaCanary Islands, Las Palmas de Gran Canaria