Fig 1.
Experimental set-up and quantification of oyster response.
A, schematic view. Is, loudspeaker position; mpcg, multiplate current generator; o, oysters equipped with electrodes; tb, tennis ball; w, wooden board; sb, sandbox; tb, w and sb compose a vibration absorber. B1, commercial loudspeaker to produce tones at frequencies from 80–20000 Hz; B2, laboratory-made loudspeaker for frequencies from 10–80 Hz. D, typical valve closure response and measured values: ymax, daily maximum VOA (valve opening amplitude); ymin, daily minimum VOA; y1, VOA prior to valve closing; y2, minimal VOA during response; Δy, amplitude of the response expressed in %, see the Materials and methods section. C, set-up for shell acceleration measurement.
Fig 2.
Typical oyster responses to 3 min of pure tone (100 Hz).
A, relationship between shell acceleration and sound pressure level at various frequencies. B, from top to bottom, waveform and typical responses ranging from minimal to maximal responses as a function of time. Dashed lines, onset and offset of the stimulus; dotted line, end of the fade-in period; n = 4 individuals.
Fig 3.
Oysters responded to sound frequencies and exhibit two peaks of maximum sensitivity at 20 and 90–100 Hz.
A, a logistic regression described the relationship between the percentage of responding oysters in a group and sound frequency. For each frequency, the distribution is described by quartiles (bold line, median). B1, the measured sound pressure level, SPL, for the studied frequencies expressed in rms. B2, the measured shell accelerations at various frequencies expressed in rms. C, the relationship between the percentage of valve opening amplitude and sound frequency. At each frequency, the data distribution is described by quartiles. N = 16 oysters. D1 and D2, Principal Component Analysis describing the correlation between the percentage of responding oysters (% Rep, D1) and valve opening-amplitude decrease (VOA, D2) combined with frequency (Hz), shell acceleration (Acc) and sound pressure level (SPL).
Fig 4.
Response delay is function of sound frequency from 10–300 Hz.
A, The response delay was systematically shorter from 10–80 Hz and the variability smaller from 10–80 Hz, with an exception at 60 Hz, illustrating a particular sensitivity to the lowest frequencies, N = 16 oysters. B1 and B2 represent the measured sound pressure levels, SPL, and shell accelerations for frequencies from 10–300 Hz.
Fig 5.
Identifying thresholds at various sound pressure levels for various frequencies.
Four examples of logistic regression models describing the relationship between oyster group responses and sound pressure levels at 10, 40, 100 and 200 Hz. Sound pressure levels are expressed as dBrms.
Fig 6.
Behavioral thresholds (M. gigas) based on the percentage of responding oysters in a group.
At each sound frequency, the percentage of responding oysters increased with sound pressure level, SPL, and shell acceleration allowing a family of curves to be drawn. A, Relationship among SPL, frequency and percentage of responding oysters. a, b, c and d, are examples of in situ noise recordings (rms; a, background noise under laboratory conditions; b, background noise at Eyrac pier, Bay of Arcachon, France; c, background noise in Poole Harbour, UK [31]; d, noise level produced by a cargo boat at 10 m away [32]. B, Relationship among shell acceleration, frequency and percentage of responding oysters. a, laboratory background noise; e, water motion of breaking waves on a rocky shore [33]. The minimum acoustic energy required to evoke a response increased with frequency.