RATE: Respiratory Acoustics To estimate Energy in wild cetaceans

Project Information

RATE

Grant agreement ID: 706867

DOI

10.3030/706867

Project closed

EC signature date 7 March 2016

Start date 1 February 2017

End date 31 January 2019

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 200 194,80

EU contribution

€ 200 194,80

Coordinated by

AARHUS UNIVERSITET
Denmark

Periodic Reporting for period 1 - RATE (RATE: Respiratory Acoustics To estimate Energy in wild cetaceans)

Reporting period: 2017-02-01 to 2019-01-31

Summary of the context and overall objectives of the project

Respiratory rate is a vital sign often used to assess responses to natural and disturbed behavior in humans and in animals, but how much air do they exchange with every breath? Cetaceans (whales and dolphins) show considerable variation in tidal volume, the volume of air breathed in and out. This greatly affects estimates of metabolic rate from breathing frequency alone, especially in the short-term or during activity. Monitoring tidal volume in wild animals is challenging, especially in cetaceans. We sought to estimate free-ranging tidal volumes by applying techniques from human medicine to establish a relationship between recorded respiratory sounds and measured air-flow rates.
Through RATE, we have made the first near-continuous breath-by-breath estimates of tidal volume in wild cetaceans, with major implications on estimating short-term ventilation and gas exchange in natural and disturbed behaviors.

Objectives

I: Estimate respiratory flow rates from breath sounds recorded in bottlenose dolphins, calibrated to measurements.
II: Apply the method to monitor free-ranging tidal volumes of tagged wild dolphins.
III: Assess changes in respiratory variables in response to specific exposures or behaviors measured by tag inertial and acoustic sensors.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

We calibrated measurements of respiratory air-flow (with a pneumotach) to sound recorded on bio-logging tags behind the blowhole of bottlenose dolphins (Fig 1a). We developed the method with data from Dolphin Quest Oahu (in 2013 and 2017), where we used the pneumotach and tag on dolphins before, during, and after exercise. We focused on inhaled volume as cetaceans often begin to exhale before or as they surface.

Acoustic data from the tags were audited to detect breathing, and the quality of each marked breath reviewed to distinguish clear recordings with no interference with the breath sound. We time-aligned the air-flow (pneumotach) and acoustic (tag) records to match data for each breath. We removed the DC component, high-pass filtered, and removed transients in the acoustic record prior to taking the Hilbert envelope to obtain the processed sound record, which we then down-sampled to the same frequency as the pneumotach and cross-correlated to time-align (Fig 1C).

For each trial, we selected five random breaths as training data. We fit the model flow rate = a soundb between the time-matched measured flow rate (L/s) and sound envelope. We applied this fit to all breaths in the same tag deployment to estimate air-flow rates from the filtered sound envelope (testing data; Fig 1D). We assessed the method by comparing the measured and estimated volumes and completed various cross-validations.

To apply these methods to estimate inhaled tidal volumes of wild bottlenose dolphins tagged in Sarasota, FL in May 2014, we (1) estimated air-flow rates from sound, (2) estimated resulting tidal volume when the animals were held prior to release, and continued to estimate (3) inhale duration, (4) tidal volume, and (5) minute volume from acoustics after animals were released.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

We estimated tidal volumes of 768 inhaled breaths over 36 trials on 6 animals at Dolphin Quest (2013 and 2017). Estimated flow rates and volumes were within 10% of measured values. At rest, tidal volume was on average 8.0±1.5 L; tidal volumes increased to 13.4±1.0 L when swimming, and inhale duration decreased. After exercise, tidal volume decreased from 11.6 L to resting levels and inhale duration increased. This represents the first time that free-ranging tidal volumes of dolphins have been estimated and put in context with exercise physiology.

Over 22h of tag recordings of 7 wild dolphins, we detected 4699 breaths. 300 were before animals were released, of which 95 had the pneumotach over the blowhole. At rest, the measured tidal volume was 4.5±1.4 L across all individuals, or 19.1±8.2 ml/kg. Estimated volumes were no different from measured, and we found no difference in estimated VT between breaths when the pneumotach was on vs. off. Overall, estimated tidal volume varied from 3.6-5.6 L (IQR; 15-26 ml/kg) when dolphins were held in water.

Following release, we detected 4534 total breaths, 4474 (99%) of which we were able to estimate volume. Tidal volumes of free-swimming, wild bottlenose dolphins were 29±7 ml/kg, 7 ml/kg (33%) higher compared to when held. Tidal volumes varied up to 7.4 L between breaths. Capturing this variability in volume between breaths leads to a much-improved estimate compared to the state-of-the-art; the current standard of assuming a constant, fixed-value based on body size results in a 32% overestimate of ventilation (L/min).

With this method, we can speak to the natural variation in tidal volume through time, as well as specific changes in the context of sound exposure, quantifying biologically significant reactions. Through RATE, we have made the first near-continuous breath-by-breath estimates of tidal volume in wild cetaceans, with major implications on estimating short-term ventilation and gas exchange in natural and disturbed behaviors.

Figure 1. A: Pneumotach and tag on a bottlenose dolphin. B: Spectrogram of the breath sound. C: Time