Development of a novel physiology tag to measure oxygen consumption in free-ranging seabirds: research

Introduction

Ambitious renewable energy targets have been developed by many countries to mitigate potential impacts of climate change (Jay 2010; Toke 2011). This has led to the proposed installation of several thousand wind turbines throughout coastal areas of Europe. In Scotland, offshore renewables have the potential to make a significant contribution to the Scottish Government's target of being Net Zero Carbon by 2045.

Proposed wind farms are often located on offshore sandbanks, which are also important habitats for many seabird species (Wanless, Harris & Greenstreet 1998). This has led to concerns about potential impacts on seabirds, including injury or mortality as a result of collisions with turbine blades (for review see: Drewitt & Langston 2006), displacement to less favourable habitats and/or increased flight costs due to a change in trajectories to avoid the wind farm (Searle et al. 2014; Searle et al. 2018). For example, Krijgsveld et al. (2011) quantified barrier effects of the 'Noordzeewind' wind farm near Egmond aan Zee for northern gannets and derived a macroavoidance rate of 0.64. Similarly, Welcker and Nehls (2016) found significant displacement of 5 species of seabird with 75-92% lower abundance inside the German offshore wind farm 'alpha ventus' compared to outside the wind farm.

Displacement and/or changes in flight trajectories to avoid offshore wind farms may have energetic consequences for individuals. However, at present there is a paucity of empirical data linking energetic changes associated with displacement and barrier effects and the survival and reproduction of seabirds (Searle et al. 2018). This has led to attempts to model the potential energetic consequences of avoiding offshore wind farms by a range of seabird species. For example, Masden et al. (2010) developed an energetic model of flight based on species-specific morphometrics, and typical foraging distances and trips per day, and investigated the effects of increased travel distances as a result of avoiding a simulated wind farm. Their predictions suggested that for European shags (Phalacrocorax aristotelis), great cormorants (Phalacrocorax carbo), common guillemots (Uria aalge), and Atlantic puffins (Fratercula arctica), daily energetic expenditure could increase by up to 170% (Masden et al. 2010).

However, uncertainties about the true energetic consequences of displacement remain due to, in part, a lack of suitable technology for measuring energy expenditure at an appropriately high resolution in free-ranging seabirds. Doubly-labelled water techniques have been used previously to measure the field metabolic rates of seabirds (e.g. Gabrielsen, Mehlum & Nagy 1987; Weimerskirch et al. 2003) but this only provides single measurements over days or weeks. Alternatively, heart rate measurements, based on the physiological relationship between heart rate (fH) and oxygen consumption rate (V̇O₂), can provide high resolution estimates of energy expenditure in free-living animals; however, this generally requires species-specific calibration in controlled laboratory conditions (Butler et al. 2004).

Overall dynamic body acceleration (ODBA) is used increasingly as a calibrated proxy for energy expenditure in free-ranging animals (e.g. Hicks et al. 2017). Energetic costs of movement constitute the majority of energy expended by individual animals (Karasov 2015); in theory therefore, body acceleration, which can be measured using animal-borne accelerometer tags (Johnson & Tyack 2003), should correlate with energy expenditure and provide an index of V̇O₂ (Wilson et al. 2006; Gleiss, Wilson & Shepard 2011; Elliott et al. 2013). However, the efficacy of ODBA as a measure of energy expenditure in diving species is currently open to debate given the equivocal results from marine mammal and seabird studies in captivity (Fahlman et al. 2008; Hindle et al. 2010; Halsey et al. 2011). For example, Fahlman et al. (2008) reported a significant relationship between V̇O₂ and ODBA for diving Steller sea lions (Eumetopias jubatus). However, re-analyses of these data by Halsey et al. (2011) suggests that the relationship between ODBA and V̇O₂ is relatively poor for Steller sea lions, particularly when compared to results from terrestrial species. Halsey et al. (2011) present further data showing that ODBA is also a poor predictor of V̇O₂ in double-crested cormorants (Phalacrocorax auratus); they conclude that, while ODBA shows promise for estimating energy expenditure in many terrestrial and fully aquatic species, it may well be that other methods of measuring field metabolic rate will prove more suitable for at least some diving species (Halsey et al. 2011).

An alternative method that is used widely in human research to provide real-time measurements of blood oxygenation is near-infrared spectroscopy (NIRS) (e.g. Mancini et al. 1994). NIRS uses infrared light from LEDs shone onto the skin, which passes through the underlying tissue and is scattered back to a receiver located next to the LEDs. As oxygenated [HbO₂] and deoxygenated haemoglobin [HHb] have different absorption spectra, spectral analysis of the refracted light determines the relative concentrations of these chromophores. Real time oxygen consumption can therefore be measured by comparing rate of change in [HbO₂] and [HHb]; this information can be used to effectively calculate oxygen consumption within the underlying tissue.

NIRS has been used successfully to measure haemoglobin (Hb) O₂ saturation levels in the locomotory muscle of emperor penguins (Aptenodytes forsteri) (Williams, Meir & Ponganis 2011). This previous study developed a NIRS instrument which was attached to feathers on the back of the birds, with the sensor head implanted into the locomotor muscle to make measurements of Hb saturation (Williams, Meir & Ponganis 2011). This provided confirmation that NIRS can be used to accurately measure Hbsaturation levels in muscles of birds as well as providing important insights into the dive response in penguins; however, the approach was invasive, requiring full anaesthesia to implant sensors under the skin with sutures to muscles (Williams, Meir & Ponganis 2011). More recently, research has shown that an animal-borne NIRS sensor can be used to measure [HbO₂] and [HHb] in freely diving, captive harbour seals (Phoca vitulina) without the need for sensor implantation (McKnight et al. 2019). With appropriate development, this has the potential to provide a non-invasive means of measuring muscle V̇O₂ in free-ranging birds.

Here, we describe the initial development of an animal-borne NIRS sensor and recording logger to measure muscle O₂ saturation (SmO₂) in free-ranging birds. We then report on the results of deployments of the logger on free-ranging European shags to determine the efficiency of the loggers, and to estimate activity-specific (e.g. diving, flying, resting) energy expenditure in shags. We then discuss the implications of the results of the development, deployment and future perspectives for the technology.