Academia.eduAcademia.edu

Figure from:Effect of elevated carbon dioxide on shoal familiarity and metabolism in a coral reef fish

See full PDFdownloadDownload figure
Figure 2: Effect of CO, on shoal preference and activity. (a) Proportion of time spent with each shoal (familiar and unfamiliar). (b) Initial shoal choice after removal of the barrier (in number of trials). (¢) Mean activity per trial (number of shoal visits). Error bars are SEM, and n = 18 for all treatments. Asterisks indicate statistical significance (*P < 0.05).

Figure 2 Effect of CO, on shoal preference and activity. (a) Proportion of time spent with each shoal (familiar and unfamiliar). (b) Initial shoal choice after removal of the barrier (in number of trials). (¢) Mean activity per trial (number of shoal visits). Error bars are SEM, and n = 18 for all treatments. Asterisks indicate statistical significance (*P < 0.05).

Related Figures (5)

The estimated partial pressure of CO. (Estimated pCO.) was calculated in the program CO2SYS using the other measured parameters. /n situ pCO2 was measured using a portable CO, equilibrator with non-dispersive infrared (NDIR) sensor. Seawater pH was measured on the NBS (National Bureau of Standards) scale (pHNBS). Error is SEM. Table 1: Summary of seawater chemistry parameters in control and elevated-carbon dioxide treatments for experiments 1 and 2
The estimated partial pressure of CO. (Estimated pCO.) was calculated in the program CO2SYS using the other measured parameters. /n situ pCO2 was measured using a portable CO, equilibrator with non-dispersive infrared (NDIR) sensor. Seawater pH was measured on the NBS (National Bureau of Standards) scale (pHNBS). Error is SEM. Table 1: Summary of seawater chemistry parameters in control and elevated-carbon dioxide treatments for experiments 1 and 2
Figure 1: Schematic diagrams of the two experimental set-ups. (a) Choice test tank used in experiment 1 (90 cm length x 30cm width x 30cm height). The dark ovals on either end of the tank represent the shoal holding containers, and the dark oval in the centre of the tank illustrates the container used for the focal fish during the pre-trial acclimation period. White dots represent the porosity of the containers (each container contained 50 5 mm holes). (b) Side view of the respirometry chamber. The experimental set-up was composed of an inner respirometry chamber (length, 13.5 cm; inner diameter, 3.24 cm; volume of chamber and associated gas-impermeable tubing, 100 ml) and an outer shoal-mate holding chamber (length, 12.0 cm; inner diameter, 11.4. cm; volume of chamber, 1.10 litres). Arrows indicate the direction of water flow through tubing. Each X indicates a water pump used for mixing the inner chamber and flushing both chambers. The outer shoal- mate holding chamber was flushed with its own pump. The outflow port for this outer chamber was connected to the flush pump for the inner respirometry chamber, to provide olfactory cues of shoal-mates to the focal individual. In order to ensure proper mixing in the inner respirometry chamber, a pump ran continuously in a closed loop. Deoxygenated water in the inner chamber was discarded during on phases of the flush pump. All focal individuals were tested in both an alone-testing treatment and a shoal-testing treatment (with six shoal-mates). The VIE tags were administered 24-48 h before placement in the CO, treatment. Shoaling preference was established using a choice test, using methodology adapted from Griffiths and Magurran (1997). An elongate testing tank (Fig. 1a) was filled to a depth of 20cm with seawater at the same CO, level as the relevant treatment. Two 1 litre plastic containers (height, 24 cm x diameter, 10cm) were placed at each end of the tank, 6 cm from the side-wall. The plastic containers were transparent and made porous to olfactory cues by holes drilled around the cir- cumference (50 5 mm holes per container). Shoals composed of 7 fish of either the familiar or an unfamiliar group were placed in these bottles. The location of the familiar shoal (right or left bottle) was randomized. The shoal used as unfamiliar was also randomized, to ensure that each shoal within a treatment was used as the unfamiliar shoal a uniform number of times and that a different unfamiliar shoal was used when testing each of the two focal fish from a shoal. The focal fish was placed in a clear, porous container in the centre of the tank. This container Nine experimental C. viridis shoals, each composed of eight fish, were acclimated to each CO, treatment for a period of 4-7 days before experimentation. This time period is sufficient for elevated CO, to induce behavioural changes in reef fishes, and previous studies indicate that longer acclimation periods do not change results (Munday et al., 2013a, 2014a; Welch et al., 2014). Two individuals per group were chosen ran- domly for testing for shoal association preferences (7 = 18 indi- viduals per treatment). These individuals were distinguished from each other and their shoal-mates using unique visible implant elastomer (VIE) tags (Hoey and McCormick, 2006).
Figure 1: Schematic diagrams of the two experimental set-ups. (a) Choice test tank used in experiment 1 (90 cm length x 30cm width x 30cm height). The dark ovals on either end of the tank represent the shoal holding containers, and the dark oval in the centre of the tank illustrates the container used for the focal fish during the pre-trial acclimation period. White dots represent the porosity of the containers (each container contained 50 5 mm holes). (b) Side view of the respirometry chamber. The experimental set-up was composed of an inner respirometry chamber (length, 13.5 cm; inner diameter, 3.24 cm; volume of chamber and associated gas-impermeable tubing, 100 ml) and an outer shoal-mate holding chamber (length, 12.0 cm; inner diameter, 11.4. cm; volume of chamber, 1.10 litres). Arrows indicate the direction of water flow through tubing. Each X indicates a water pump used for mixing the inner chamber and flushing both chambers. The outer shoal- mate holding chamber was flushed with its own pump. The outflow port for this outer chamber was connected to the flush pump for the inner respirometry chamber, to provide olfactory cues of shoal-mates to the focal individual. In order to ensure proper mixing in the inner respirometry chamber, a pump ran continuously in a closed loop. Deoxygenated water in the inner chamber was discarded during on phases of the flush pump. All focal individuals were tested in both an alone-testing treatment and a shoal-testing treatment (with six shoal-mates). The VIE tags were administered 24-48 h before placement in the CO, treatment. Shoaling preference was established using a choice test, using methodology adapted from Griffiths and Magurran (1997). An elongate testing tank (Fig. 1a) was filled to a depth of 20cm with seawater at the same CO, level as the relevant treatment. Two 1 litre plastic containers (height, 24 cm x diameter, 10cm) were placed at each end of the tank, 6 cm from the side-wall. The plastic containers were transparent and made porous to olfactory cues by holes drilled around the cir- cumference (50 5 mm holes per container). Shoals composed of 7 fish of either the familiar or an unfamiliar group were placed in these bottles. The location of the familiar shoal (right or left bottle) was randomized. The shoal used as unfamiliar was also randomized, to ensure that each shoal within a treatment was used as the unfamiliar shoal a uniform number of times and that a different unfamiliar shoal was used when testing each of the two focal fish from a shoal. The focal fish was placed in a clear, porous container in the centre of the tank. This container Nine experimental C. viridis shoals, each composed of eight fish, were acclimated to each CO, treatment for a period of 4-7 days before experimentation. This time period is sufficient for elevated CO, to induce behavioural changes in reef fishes, and previous studies indicate that longer acclimation periods do not change results (Munday et al., 2013a, 2014a; Welch et al., 2014). Two individuals per group were chosen ran- domly for testing for shoal association preferences (7 = 18 indi- viduals per treatment). These individuals were distinguished from each other and their shoal-mates using unique visible implant elastomer (VIE) tags (Hoey and McCormick, 2006).
Figure 3: Effect of CO. and testing treatment on the minimal metabolic rate (MR»pini in milligrams of Oz per hour; a) and initial stress response (ISR; in milligrams of O2 per hour; b). In these panels, metabolic rate measures were mass corrected by using residuals of the relationship between logarithm of the body mass and logarithm of whole-animal metabolic rate added to the fitted value for mass = 1.29 g, the mean mass of all fish used in the study. Error bars are SEM, and n = 10 for all treatments. Asterisks indicate statistical significance (*P < 0.05). Statistical analysis was conducted on whole- animal metabolic rates, with body mass as a covariate.
Figure 3: Effect of CO. and testing treatment on the minimal metabolic rate (MR»pini in milligrams of Oz per hour; a) and initial stress response (ISR; in milligrams of O2 per hour; b). In these panels, metabolic rate measures were mass corrected by using residuals of the relationship between logarithm of the body mass and logarithm of whole-animal metabolic rate added to the fitted value for mass = 1.29 g, the mean mass of all fish used in the study. Error bars are SEM, and n = 10 for all treatments. Asterisks indicate statistical significance (*P < 0.05). Statistical analysis was conducted on whole- animal metabolic rates, with body mass as a covariate.
Figure 4: Effect of CO, and testing treatment on the routine metabolic rate (RMR; in milligrams of O2 per hour; a) and activity (number of 180° turns/min; b). In these panels, metabolic rate measures were mass corrected by using residuals of the relationship between the logarithm of the body mass and logarithm of whole- animal metabolic rate added to the fitted value for mass = 1.29 g, the mean mass of all fish used in the study. Error bars are SEM, and n= 10 for all treatments. Asterisks indicate statistical significance (*P < 0.05). Statistical analysis was conducted on whole-animal metabolic rates, with body mass as a covariate. The RMRgroup was significantly lower than RMRalone (Fig. 4a; F126 = 33.84, P < 0.001). Respirometry treatment had a comparable effect on RMR in individuals from all CO; treat- ments (F)27 = 0.73, P = 0.490), and there was no significant interaction between testing and CO, treatment (Fr2¢5 = 0.43, P = 0.714). Neither testing (F; 26 = 0.31, P = 0.583) nor CO, treatment (F)57 = 0.32, P = 0.732) exerted a significant effect
Figure 4: Effect of CO, and testing treatment on the routine metabolic rate (RMR; in milligrams of O2 per hour; a) and activity (number of 180° turns/min; b). In these panels, metabolic rate measures were mass corrected by using residuals of the relationship between the logarithm of the body mass and logarithm of whole- animal metabolic rate added to the fitted value for mass = 1.29 g, the mean mass of all fish used in the study. Error bars are SEM, and n= 10 for all treatments. Asterisks indicate statistical significance (*P < 0.05). Statistical analysis was conducted on whole-animal metabolic rates, with body mass as a covariate. The RMRgroup was significantly lower than RMRalone (Fig. 4a; F126 = 33.84, P < 0.001). Respirometry treatment had a comparable effect on RMR in individuals from all CO; treat- ments (F)27 = 0.73, P = 0.490), and there was no significant interaction between testing and CO, treatment (Fr2¢5 = 0.43, P = 0.714). Neither testing (F; 26 = 0.31, P = 0.583) nor CO, treatment (F)57 = 0.32, P = 0.732) exerted a significant effect
Figure 5: Effect of CO. on maximal metabolic rate (MMR; in milligrams of O2 per hour; a) and aerobic scope (AS; in milligrams of O2 per hour; b). In these panels, metabolic rate measures were mass corrected by using residuals of the relationship between the logarithm of body mass and logarithm of whole-animal metabolic rate added to the fitted value for mass = 1.29 g, the mean mass of all fish used in the study. Error bars are SEM, and n = 10 for all treatments. Statistical analysis was conducted on whole-animal metabolic rates, with body mass as a covariate.
Figure 5: Effect of CO. on maximal metabolic rate (MMR; in milligrams of O2 per hour; a) and aerobic scope (AS; in milligrams of O2 per hour; b). In these panels, metabolic rate measures were mass corrected by using residuals of the relationship between the logarithm of body mass and logarithm of whole-animal metabolic rate added to the fitted value for mass = 1.29 g, the mean mass of all fish used in the study. Error bars are SEM, and n = 10 for all treatments. Statistical analysis was conducted on whole-animal metabolic rates, with body mass as a covariate.