Electronic Supplementary Material
Ocean acidification erodes crucial auditory behaviour in a marine fish
Stephen D. Simpson, Philip L. Munday, Matthew L. Wittenrich,
Rachel Manassa, Danielle L. Dixson, Monica Gagliano, Hong Y. Yan
1. Water chemistry in housing and test environments
Ambient or CO2-enriched air was bubbled into each tank at 1.5 l m-1 through a fine-pore airstone. The pHNBS of treatment seawater was measured daily with a portable meter (HQ11D, Hach Co., Loveland, CO) calibrated with fresh buffers (Merck KGaA, Darmstadt, Germany). Total alkalinity was determined weekly by titration. pCO2 concentration in seawater was estimated using the program CO2SYS ([1]; Supplementary Table 1) from measured values of pH and total alkalinity, and cross-validated by in situ sampling using a submerged CO2-permeable membrane connected to an aspirated CO2 probe (GM222, Vaisala Oyj, Helsinki, Finland) in a closed loop (N = 4-6 30 minute observations per treatment) [2]. The CO2 concentration in the test arena was matched to the rearing environments for each treatment group. Water temperature in the housing and test environments was maintained at 30°C ± 0.5 and oxygen levels were maintained at >90% saturation.
Supplementary Table 1: Summary of mean seawater parameters in control and acidification treatments.
pHNBSmean ± SD / Total alkalinity
(µmol kg-1) / pCO2 (seawater) (µatm) / IPCC Emissions Scenario
8.15 ± 0.05 / 1984 / 391 / Present day
7.99 ± 0.07 / 1992 / 613 / low range – B2
7.93 ± 0.09 / 1986 / 718 / intermediate – A1B
7.86 ± 0.05 / 2015 / 876 / high range – A1FI
2. Details of daytime reef recording used
The daytime recording of a high quality reef in a marine protected area included both snapping shrimp and multiple sources of fish calls (0840 hrs 14/06/2007, Balicasag Black Forest, Bohol, Philippines, N 9°31.006’ E 123°41.337’, details of equipment and calibrations in [3]). Although taken outside the natural species range of A. percula, this recording was selected as it had elicited a clear negative response by reef fish larvae in a previous experiment [4]; a result we hypothesised was due to it being a daytime, predator-rich recording. Since the larvae used in this experiment were reared from eggs in captivity, they had no prior experience of local reef noise, so we did not expect the origin of the recording to compromise the likelihood of observing a response to the sound by larvae. Higher frequency sounds above the hearing range of clownfish were filtered out in Avisoft SASLab Pro (Avisoft Bioacoustics, Berlin, Germany) using a 2 kHz low pass filter (IIR Time Domain, Tschebyscheff 8th order). A single sound file was used as this study focused on the difference in response of fish from different CO2 groups to a sound, rather than exploring the general response of fish to multiple recordings [5].
3. Electrophysiological assessment of the hearing abilities of juvenile Amphiprion ephippium
Clownfish larvae (Amphiprion ephippium), received as eggs from a commercial hatchery (Vince Rado at Oceans, Reefs and AquariumsTM), were reared and tested in controlled temperature and light conditions, with approval by the University of Kentucky Institutional Animal Care and Use Committee (00217L2001). Standard Auditory Brainstem Response (ABR) procedures were followed [6,7]: fish were restrained under mesh in a harness, the head and electrodes were submerged below the surface, the signal was measured using a calibrated hydrophone, all handling was by pipette and water pressure to minimise mechanical stress to the fish. Using six randomised-order test frequencies (100, 300, 400, 700, 1000, 1500 Hz), electrophysiological recordings were made of synchronous neural activity in the eighth cranial nerve and brainstem auditory nuclei in response to an acoustic stimulus (20 ms tone bursts, 2000 sweeps per test) presented at different sound pressure levels to determine thresholds of hearing. Fish were tested at 17, 20, 24 and 36 days post-hatching (Supplementary Fig. 1).
Supplementary Figure 1: Audiogram of juvenile clownfish (A. ephippium) hearing. “d PH” indicates days post-hatching, and mean total length is in parentheses. Three subjects were tested at 17 and 20 d PH, and two at 24 and 36 d PH. Maximum SE between subjects was 4.83 dB (mean <2 dB), so for clarity only mean threshold values are shown.
4. Analysis of otolith morphology
Methods: Left and right sagittal otoliths were removed from every fish used in the behavioural trials and photographed to produce calibrated grey-scale images from which the following dimensions and shape descriptors were obtained: maximum otolith length (mm), perimeter (mm), area (mm2), rectangularity, and circularity. The shape of the otolith was also characterised using Fast Fourier analysis of 128 equidistant points around the outline [8]. The resulting five dimensions and first 20 standardised shape Fast Fourier descriptors were analysed, as well as the signed asymmetry value (R–L) of the otolith pair. To test for asymmetry, the distribution of these values in each otolith pair character was compared to a distribution with a mean equal to 0 and normal variation, and signs of different forms of asymmetry, such as directional asymmetry (i.e. skewed distributions with kurtosis values smaller than 0) and antisymmetry (i.e. bimodality) were tested [9]. Otolith morphology was compared between groups using a factorial multivariate analysis of variance (MANOVA). Stepwise discriminant function analysis was used to determine which otolith shape descriptors contributed the most to separation between groups and statistical significance of those otolith descriptors was identified using univariate analysis of variance (ANOVA with Bonferroni HSD) corrected for multiple comparisons (a=0.05/k, where k is the number of selected otolith descriptors).
Results: Analysis of otolith size, shape and symmetry did not identify any clear patterns that related to rearing in the elevated CO2 treatments. The right (but not left) otoliths from juvenile fish in the 600ppm group were significantly larger than those from the 700ppm group (length: F3,49 = 5.319, P = 0.003; perimeter: F3,49 = 5.075, P = 0.004), with greater rectangularity (right only: F3,49 = 6.41, P = 0.001) and less circularity (right: F3,49 = 7.028, P<0.001; left: F3,49 = 5.643, P = 0.002), but neither group differed with the 390 or 900ppm group. This is likely to be a random effect and does not associate with the behavioural result of fish in 390ppm behaving differently to those from all of the elevated-CO2 conditions. Considering that 25 potentially variable metrics were compared between groups, and these were the only differences detected, it appears that CO2 did not significantly modify the growth, size or shape of the otoliths in any consistent way, and there was no relationship between the few identified changes and the elicited behaviour in the choice chambers.
5. References
1. Lewis, E. L. & Wallace, D. W. R. 1995 Basic programs for the CO2 system in seawater. Brookhaven National Laboratory Informal Report, BNL# 61827.
2. Hari, P., Pumpanen, J., Huotari, J., Kolari, P., Grace, J., Vesala, T. & Ojala, A. 2008 High-frequency measurements of productivity of planktonic algae using rugged nondispersive infrared carbon dioxide probes. Limnol. Oceanogr.: Methods 6, 347-354.
3. Vermeij, M. J. A., Marhaver, K. L., Huijbers, C. M., Nagelkerken, I. & Simpson, S. D. 2010 Coral larvae move toward reef sounds. PLoS ONE 5, e10660.
4. Heenan, A., Simpson, S. D. & Braithwaite, V. A. 2009 Testing the generality of acoustic cue use at settlement in larval coral reef fish. Proceedings of the 11th International Coral Reef Symposium, 554-558.
5. Slabbekoorn, H. & Bouton, N. 2008 Soundscape orientation: a new field in need of sound investigation. Anim. Behav. 76, E5-E8.
6. Kenyon, T. N., Ladich, F. & Yan, H. Y. 1998 A comparative study of hearing ability in fishes: the auditory brainstem response approach. J. Comp. Physiol. A 182, 307-318.
7. Parmentier, E., Colleye, O. & Mann, D. 2009 Hearing ability in three clownfish species. J. Exp. Biol. 212, 2022-2025.
8. Gagliano, M., Depczynski, M., Simpson, S. D. & Moore, J. A. Y. 2008 Dispersal without errors: symmetrical ears tune into the right frequency for survival. Proc. R. Soc. London Ser. B. 275, 527-534.
9. Palmer, A. R. & Strobeck, C. 1986 Fluctuating asymmetry: measurement, analysis, patterns. Annu. Rev. Ecol. Syst. 17, 391-421.