Diel CO2 cycles reduce severity of behavioural abnormalities in coral reef fish under ocean acidification

Elevated CO2 levels associated with ocean acidification (OA) have been shown to alter behavioural responses in coral reef fishes. However, all studies to date have used stable pCO2 treatments, not considering the substantial diel pCO2 variation that occurs in shallow reef habitats. Here, we reared juvenile damselfish, Acanthochromis polyacanthus, and clownfish, Amphiprion percula, at stable and diel cycling pCO2 treatments in two experiments. As expected, absolute lateralization of A. polyacanthus and response to predator cue of Am. percula were negatively affected in fish reared at stable, elevated pCO2 in both experiments. However, diel pCO2 fluctuations reduced the negative effects of OA on behaviour. Importantly, in experiment two, behavioural abnormalities that were present in fish reared at stable 750 µatm CO2 were largely absent in fish reared at 750 ± 300 µatm CO2. Overall, we show that diel pCO2 cycles can substantially reduce the severity of behavioural abnormalities caused by elevated CO2. Thus, past studies may have over-estimated the impacts of OA on the behavioural performance of coral reef fishes. Furthermore, our results suggest that diel pCO2 cycles will delay the onset of behavioural abnormalities in natural populations.


Experimental systems and CO 2 manipulation
Experiment one The experimental system used at JCU was an 11,000 L re-circulating system. Briefly, the system consisted of a large external 3,700 L sump tank connected to a bio-filter, protein skimmer, UV steriliser and a 1000 L algal bio-remediation tank. The external sump supplied water to four separate 1,600 L re-circulating systems (one per pCO2 treatment) made up from one 1000 L sump tank and fifteen 40 L holding tanks, contained within a temperature controlled room. Water was supplied at a rate of approximately 1,600 L per day allowing for a complete exchange with the external sump. Holding tanks were supplied with water at a rate of 1 L min -1 . Both the internal sumps and holding tanks were aerated with ambient air.
Elevated pCO2 treatments were achieved by dosing the 1000 L internal sumps with CO2. This was controlled by solenoid valves (M-Ventil Standard, Aqua Medic, Germany) connected to a pH control system (Aqua Medic AT Control System, Aqua Medic, Germany) with laboratory grade pH electrodes (Neptune Systems, USA). The Aqua Medic AT Control System has a curve function which allowed us to create fluctuating pCO2 profiles. pH profiles in the fluctuating pCO2 treatments were recorded every other day using a pH meter (InLab Expert Pro electrode and Seven2Go Pro meter, Mettler Toledo, Switzerland) set to take a reading every 15 min. For the stable pCO2 treatments pHNBS was measured twice daily using the same model of pH meter. Seawater pH on the total hydrogen ion concentration scale (total scale, pHT) was measured each week with a spectrophotometer following standard operating procedures 1 using the indicator dye meta/m-cresol purple (mCP) (m-cresol purple sodium salt 99%, non-purified, Acros Organic). Daily and fluctuating pHNBS measurements were converted to pHT based on the offset between weekly pHT and pHNBS measurements.
Temperature was recorded daily with a digital thermometer (Comark C26, Norfolk, UK).
Salinity readings were taken weekly using a conductivity sensor (HQ15d; Hach, USA). Total alkalinity was also measured weekly using Gran Titration (Metrohm 888 Titrando Titrator Metrohm AG, Switzerland) and certified reference material from Dr. A. G. Dickson (Scripps Institution of Oceanography). All seawater parameters were measured in randomly chosen holding tanks. pCO2 values were calculated as a function of pHT, temperature and salinity using CO2SYS 2 employing constants from 3 refit by 4 and the KHSO4 dissociation constant from 5 . Mean values for each of these seawater parameters are presented in Table 1.

Experiment two
The experimental system used at SeaSim was a flow-through system which comprised of multiple independent lines (duplicate independent lines per pCO2 treatment). The system used ultra-filtered seawater (0.04 µm), temperature controlled to 28.5°C. Each seawater line supplied three custom made 50 L tanks at the rate of 50 L h -1 . The experimental tanks were placed in individual temperature-controlled water baths to ensure temperature stability (± 0.1˚C). Treatments and tank replicates were randomly positioned in the experimental room. 3M, USA). Total alkalinity was also measured weekly as described above. Mean values for seawater parameters are presented in Table 2.
Throughout the experiment incoming coastal water had a pCO2 ranging between 500-550 µatm. Thus, to achieve a control pCO2 level closer to 460 µatm, membrane contactors (Membrana Liqui-Cel 4x28 Extra-Flow) were used to remove CO2, using CO2depleted air as sweep gas. This was only possible for the control treatments and consequently the lower pCO2 levels in the 750 ± 300 µatm treatment matched the pCO2 of the incoming seawater (500-550 µatm).   Profiles for the stable pCO2 treatments were based on measurements taken twice per day. In reality some minor daily variation would have likely occurred. Coloured sections are ± 1 SD.