Estimates of field activity and metabolic rates of bonefish (Albula vulpes) in coastal marine habitats using acoustic tri-axial accelerometer transmitters and intermittent-flow respirometry

Abstract

We tested the utility of acoustic tri-axial acceleration transmitters in combination with ethogram and respirometry studies to quantify the activity patterns and field metabolic rates of free-swimming bonefish (Albula vulpes) in the coastal waters of Eleuthera, The Bahamas. Bonefish were found to exhibit relatively low activity levels in the field with no evidence of diel patterns or inter-sexual variation. Low activity levels reported by the accelerometers reflected low swimming speeds and intermittent swimming behaviours (i.e., swim-then-drift) that maximized energy efficiency. Such behaviours were also observed when conducting ethograms on bonefish held in a large field mesocosm. Laboratory derived standard metabolic rates and maximum metabolic rates during recovery from exercise were combined with acceleration transmitter calibrations to determine that individual bonefish typically operate at between 40 and 60% of their estimated metabolic scope in the field. However, occasionally acceleration values in the field were indicative of high intensity bursting activity that exhausted the majority of their estimated metabolic scope (0.7% of all field observations exceed 90% of estimated metabolic scope). Data gathered in this study provide a critical starting point for the development of a bioenergetics model for bonefish which will ultimately lend clues into how fish make a living in tropical tidal flats environments. Given that this paper is one of the first to use transmitters rather than archival loggers to collect data on tri-axial acceleration on fish, we also discuss the opportunities and limitations of using this new technology for marine fisheries research.

Introduction

Energy is an important commodity for all organisms, and is the currency most often employed in analyses of animal behaviour (Townsend and Calow, 1981). How animals partition energy into different life functions, and thus how they make a living, can be represented by balanced energy equations (Soofiani and Hawkins, 1985). Animals making energetic choices that increase survival will be favored, and as such, the balanced energy equation is strongly influenced by natural selection (Diana, 2004, Wilson et al., 2006). Estimating the complete energy budget of a free-living organism, however, has presented challenges to scientists, particularly when assessing the energetic costs of activity in fish (see Briggs and Post, 1997a). As the cost of activity may represent a large and variable component of the fish's energy budget (Boisclair and Sirois, 1993), the methods used to measure locomotion need to be effective in free-swimming fish in the wild (Briggs and Post, 1997b, Cooke et al., 2004).

Previous studies have explored the utility of a variety of biotelemetry sensors including heart rate (e.g., Lucas et al., 1991), tail-beat frequency (e.g., Ross et al., 1981), and axial muscle electromyograms (EMGs) (e.g., Briggs and Post, 1997a, Briggs and Post, 1997b, Cooke et al., 2004). However, both heart rate (electrocardiogram (ECG)) and EMG transmitters require precise surgical implantation of electrodes and significant handling of the animal (Whitney et al., 2007). More recently, the use of animal-borne acceleration data loggers for studying free-swimming fish and other animals is gaining popularity (Wilson et al., 2007, Shepard et al., 2008). Because locomotion occurs when animals expend energy to contract muscles which leads to body acceleration, the accurate measurement of acceleration should be a good proxy for energy expenditure during activity (Halsey et al., 2009). Indeed, acceleration data loggers have been successful in elucidating homing migrations and spawning behaviour in salmon (Tanaka et al., 2001, Tsuda et al., 2006; respectively), diel activity patterns in whitetip reef sharks (Triaenodon obesus) (Whitney et al., 2007), and general activity patterns in rainbow trout (Onchorhynchus mykiss) (Kawabe et al., 2003a) and Japanese flounder (Paralicthys olivaceus) (Kawabe et al., 2003b). Acceleration data loggers have their limits as well, requiring retrieval of the logger to access the data (Ropert-Coudert and Wilson, 2005). Only recently has the technology of onboard processing improved sufficiently to encode and transmit tri-axial accelerometer data efficiently. With acceleration transmitters the data are sent to acoustic hydrophone receivers, extending the use of these devices to species and/or environments where recaptures are difficult.

Knowledge of the activity patterns and energetic requirements of marine species is becoming increasingly important for modeling ecosystems and managing populations (Lowe, 2002, Fitzgibbon et al., 2007). This is particularly true for species occupying coastal habitats, since over half of the world's population lives in these areas (Barnabé and Barnabé-Quet, 2000). Habitat degradation is widespread where humans exploit resources such as mangrove forests (Alongi, 2002, Blaber, 2007). Studying the behaviour and activity patterns of a species that not only occupies coastal marine environments, but is also the object of an economically important recreational fishery may provide insight into individual and population level processes, which may ultimately influence the effectiveness of conservation and management strategies.

Bonefish (Albula spp.) are a group of benthivorous fish found in tropical tidal flats and tidal creeks (Colton and Alevizon, 1983a, Colton and Alevizon, 1983b, Humston et al., 2005). Throughout much of their circumtropical distribution bonefish also carry the distinction of being a popular sport fish and thus play an important role in many local economies (Pfeiler et al., 2000, Ault, 2008, Danylchuk et al., 2008). To date, there has been no known study which has examined bonefish activity patterns and behaviour beyond traditional positional biotelemetry studies (see Colton and Alevizon, 1983b, Humston et al., 2005, Friedlander et al., 2008, Larkin et al., 2008), all of which have had limited spatial (often on the order of 500 m accuracy of positioning) and temporal resolution (fish tracked at infrequent intervals and often for short duration), making it impossible to evaluate fine-scale activity patterns or estimate energy expenditure.

The objective of this study was to quantify the field activity and metabolic rates of bonefish (Albula vulpes) in tidal flats and tidal creek areas near Cape Eleuthera, Eleuthera, The Bahamas. Using acoustic tri-axial acceleration transmitters, we investigated the influence of sex and photoperiod on the activity patterns of wild bonefish and compared results to laboratory and field calibrations. Furthermore, we catalogued the discrete behaviours of bonefish held in a natural wetland mesocosm to produce an activity time budget. Static respirometry was used to determine standard metabolic rate and maximum metabolic rate during recovery after exercise. When combined with data from accelerometers, we estimated the field energetics of bonefish.