Seasonal energetic challenges may constrain an animal’s ability to respond to changing individual and environmental conditions. Here we investigated variation in heart rate, a well-established proxy for metabolic rate, in Svalbard reindeer, a species with strong seasonal changes in foraging and metabolic activity. In 19 adult females we recorded heart rate, subcutaneous temperature and activity using biologgers. Mean heart rate more than doubled from winter to summer. Typical drivers of energy expenditure, such as reproduction and activity, explained a relatively limited amount of variation (2–6% in winter and 16–24% in summer), compared to seasonality which explained 75% of annual variation in heart rate. The relationship between heart rate and subcutaneous temperature depended on individual state via body mass, age and reproductive status, and the results suggested that peripheral heterothermy is an important pathway of energy management in both winter and summer. While the seasonal plasticity in energetics make Svalbard reindeer well-adapted to their highly seasonal environment, intraseasonal constraints on modulation of their heart rate may limit their ability to respond to severe environmental change. This study emphasizes the importance of encompassing individual state and seasonal context when studying energetics in free-living animals.

Adult females (ages 5–8 years, marked as calves) were captured in March–April 2018 for biologger deployment and in April 2019 for biologger retrieval. On both occasions, animals were caught by net using snowmobiles [47]: we recorded their body mass (± 0.5 kg) and checked for pregnancy using an ultrasound scanner (Kaixin Electronic Instrument Co., Xuzhou, China). In August 2018, surveys were conducted on foot to relocate marked animals and assess calf status.

We fitted each animal with a combined heart rate and temperature logger (DST centi-HRT, Star-Oddi, Gardabaer, Iceland; ~19 g), which was implanted subcutaneously on the left side of the sternum or behind the left axilla, while animals were under anaesthesia. Surgical procedures are described in ESM section 1.2. Heart rate was automatically calculated from a 4 s electrocardiogram (ECG) at 150 Hz measurement frequency and stored alongside a quality index of signal clarity. We programmed the loggers to record heart rate and subcutaneous body temperature (T_sc) every 15 min, and to store a raw electrocardiogram signal every 6 hrs for manual validation.

The animals were also fitted with a collar (Vertex Plus, Vectronic Aerospace GmbH, Berlin, Germany, ~750 g) containing an activity sensor. The activity sensor measured acceleration along two orthogonal axes representing back-forward and right-left movements at 4 Hz intervals. An internal algorithm calculates activity as the difference in acceleration between two consecutive measurements and is given within a relative range between 0 and 255, providing a mean value of acceleration in each axis every 5 min.

Further details about data processing are provided in the article and its electronic supplementary material.

The data sets attached contain the final data which has been used for analyses in Trondrud et al. 2021 and has been through validation, filtering and processing. The steps taken are provided in the methods. The data therefore contains missing values where poor-quality readings of heart rate have been removed as described in the electronic supplementary material of Trondrud et al. 2021.

Further information about the data is provided in the readme-file accompanying the data sets.