Whales
in Space:
Surround Audio
Recordings of the Vocalizations of Humpback Whales in the Inside Passage of
Southeast Alaska
James P. Crutchfield,
David Dunn, and Alex Jurgens
This is a binaural recording. Please use headphones for full
spatial immersion.
Whales in Space is a documentary recording of the
first successful surround recordings made of the vocalizations of Humpback
Whales in the Inside Passage of southeast Alaska. During a two week, 300-mile
voyage from Juneau to Ketchikan, Alaska (2023), over 50 hours of recordings
were made using the hydroambiphone, a new type
of underwater hydrophone array [1]. This voyage was the first field test of
this novel audio technology based upon ambisonics
theory as designed
and fabricated by physicist James P. Crutchfield.
Introduced only in the 1970s, ambisonics is a mathematical theory of 3D spatial
acoustic fields [2] that has antecedents in the theory of molecular electron
orbitals developed in the 1920s as quantum “wave” mechanics was being invented.
To date, ambisonics has been largely
developed for representing sound in air and has become a predominant standard
for capturing and reproducing spatial sound beyond conventional monophonic or
stereo playback. Its successful application to recording underwater sound,
however, is relatively novel. Since the theory affords a precise way to capture
a 3D sound field, a primary benefit is to allow for a listener-centric
representation of the sound field experienced by underwater animals that move
and vocalize in three dimensions.
The hydroambiphone
consists of four hydrophone transducers mounted in a tetrahedral
configuration on the surface of a 12-inch diameter hollow sphere. To properly
“shadow” each hydrophone from sounds arriving from opposite directions (and so
improve directionality) the sphere was 2mm thick stainless steel, which has a
high specific acoustic impedance. Audio signals from the array were connected
to a multi-channel digital recorder through conventional audio cables and a
steel cable support. The raw high-resolution four-channel field recordings were
stored as conventional digital audio files and subsequently subjected to
appropriate signal processing and editing software to achieve the final
surround playback files. These required hundreds of hours of post-production
that included noise reduction, editing and sequencing, adjusting for the speed
of sound in water (approximately five times faster than in air) and correcting
for the specific transducer positions on the 12-inch diameter spherical housing
of the array. These adjustments were essential to account for the propagation
of sound in the ocean environment and the relative arrival times of sound waves
to the individual hydrophone positions as necessary to maintain the precise
phase accuracy of the ambisonics encoding. This was achieved by correcting the
value of constants within the wave equation: ⃗u = c2∇2⃗u,
where c is the speed of sound and ⃗u(x,y,z) is local state of the medium.
Over the 300-mile voyage, a wide
variety of whale sounds (social calls, bubble-net feeding, breaching,
tail-lobbing, and fin-slapping) were recorded that are very different from the
more familiar so-called “songs” made by Humpback males during their mating
season. We also recorded the constant anthropogenic noise made by cruise ships,
fishing boats, shipping barges, and whale watching tours. Humpback social calls
have previously been recorded and studied but not using audio technology
appropriate for surround playback and research. Our agenda was to capture and
represent the perceptual perspective of these marine mammals in their actual
aquatic reality. While these recordings certainly inform questions about the
nature and purpose of whale communication, they are also intended to
communicate both the extraordinary complexity (and beauty) of that
communication as well as the acoustic challenges that these creatures must
endure in their current fight to survive against human sonic encroachment and
climate change. The best way to capture that reality is through the inclusion
of spatiotemporal information¾ information
missing from both traditional monophonic hydrophone recordings and conventional
media presentations where the use of surround audio is merely a contrived
enhancement to the audience experience. Instead, we hope to capture and convey
the real acoustic world of the whale, what in biology has been called its Umwelt.
It is important to point out just how
different undersea audio recording is from that in air. Since the medium of
water is denser than that of air, the speed of sound in water is five times
faster. This can lead to echoes as sounds bounce off the seabed. The effect is
conspicuous in these recordings made while traversing the Alaskan Inside
Passage where surface conditions and fierce underwater currents can change
suddenly. The undersea terrain is complex in its differing morphology and
depth. Seawater density also increases with depth and can contribute to sounds
from distant sources being detected more easily than in air and over much
greater distances. Surface wave motion can also sonically propagate to
surprising depths. The velocity of sound increases with water pressure (and so
depth) and decreases with water temperature and salinity. Unlike sound in air,
underwater sound at different frequencies can also propagate at different
speeds. One result is a channeling of acoustic waves over dozens over even
hundreds of kilometers.
These recordings were all made from the same
vessel, the M/Y Blue Pearl (Vancouver, British Columbia), owned and piloted by
Don and Denise Bermant. The hydroambiphone
was generally deployed at a depth of between 30 to 40 feet. For public
presentation, cross-fading between selective event segments was employed to
represent as much sonic and behavioral diversity as possible within a
reasonable listening time frame. These events are therefore presented in a
truncated sequence representative of the 300-mile voyage from start to finish.
No other equalization, filtering, signal processing (other than what has
already been described), or juxtaposed mixing was used.
Bibliography:
[1] F. Zotter and M. Frank, Ambisonics: A Practical 3D Audio
Theory for Recording, Studio Production, Sound Reinforcement, and Virtual
Reality, Springer, Springer Topics in Signal Processing, volume 19 (Cham,
Switzerland), 2019.
[2] Whales in Space
project webpage.
Credits:
James Crutchfield designed
and built the HAP underwater hydrophone recording array. J.C. and David Dunn
conducted the underwater field recording and its associated calibration and processing.
Alex Jurgens was responsible for field photography and observations of
correlated surface behavior of the whales to monitor their presence.
Photography by James Crutchfield.
Audio editing and mastering
by David Dunn.
Design by Martin Back.
For a more in-depth discussion of this
project, go to: [2312.16662] Whales in Space:
Experiencing Aquatic Animals in Their Natural Place with the Hydroambiphone.
Or listen to the Podcast.
Compact disc of hydroambiphone
recordings: Whales in Space