Local differences in robustness to ocean acidification

Abstract

Ocean acidification (OA) caused by increased atmospheric carbon dioxide is affecting marine systems globally and is more extreme in coastal waters. A wealth of research to determine how species will be affected by OA, now and in the future, is emerging. Most studies are discrete and generally do not include the full life cycle of animals. Studies that include the potential for adaptation responses of animals from areas with different environmental conditions and the most vulnerable life stages are needed. Therefore, we conducted experiments with the widely distributed blue mussel, Mytilus edulis, from populations regularly exposed to different OA conditions. Mussels experienced experimental conditions prior to spawning, through embryonic and larval development, both highly vulnerable stages. Survivorship to metamorphosis of larvae from all populations was negatively affected by extreme OA conditions (pH 7.3, Ωar, 0.39, pCO2 2479.74), but, surprisingly, responses to mid OA (pH 7.6, Ωar 0.77, pCO21167.13) and low OA (pH 7.9, Ωar 1.53, pCO2 514.50) varied among populations. Two populations were robust and showed no effect of OA on survivorship in this range. One population displayed the expected negative effect on survivorship with increased OA. Unexpectedly, survivorship in the fourth population was highest under mid OA conditions. There were also significant differences in development time among populations that were unaffected by OA. These results suggest that adaptation to OA may already be present in some populations and emphasizes the importance of testing animals from different populations to see the potential for adaptation to OA.

Keywords: Mytilus edulis; Blue mussel; Larval survivorship; Ocean acidification.

Fig. 1.

Map of Long Island Sound and sites where blue mussels were collected (Google Earth image). Milford CT, 41°12′35″N 73°03′03″W; Groton CT, 41°19′04″N 72°03′51″W, Orient NY 41°09′34″N 72°14′05″W; Stony Brook NY, 40°55′16″N 73°09′00″W.

Fig. 2.

Survivorship of larvae from each of the four test populations reared under each of the three OA conditions. Shown are means of five replicates (error bars not included for clarity). Low OA (pH 7.9, Ω_ar 1.53, pCO₂ 514.50), Mid OA (pH 7.57, Ω_ar 0.77, pCO₂ 1167.13) High OA (pH 7.26, Ω_ar 0.39, pCO₂ 2479.74).

Fig. 3.

Survivorship of larvae among populations and OA treatments. (A) The probability of survivorship for larvae from each of the four test populations under each of the three test conditions. Test populations were Groton (G), Milford (M), Orient (O) and Stony Brook (S). The three OA treatments were: low OA (pH 7.9, Ω_ar 1.53, pCO₂ 514.50), mid-OA (pH 7.57, Ω_ar 0.77, pCO₂ 1167.13) high-OA (pH 7.26, Ω_ar 0.39, pCO₂ 2479.74) (Table 3). No larvae survived to metamorphosis in the high-OA treatment, but there was a difference among populations in length of larval survival. Larvae survived to metamorphosis in the mid-OA treatment for all populations except Stony Brook. (B) The hazard ratio compares the rate of an event happening between two or more treatments. In this case, overall, the rate of death for the mid-OA treatment was 0.28 times less than that of the high-OA treatment (and was significant). The rate of death for larvae from Stony Brook (S) was 1.87 times greater than for larve from the Groton (reference) population (significant). Error bars are hidden by markers.

Fig. 4.

Cumulative numbers of larvae that metamorphosed for the four test populations reared under each of the three OA conditions. Shown are means of five replicates for each population. Low-OA (pH 7.9, Ω_ar 1.53, pCO₂ 514.50), mid-OA (pH 7.57, Ω_ar 0.77, pCO₂ 1167.13) high OA (pH 7.26, Ω_ar 0.39, pCO₂ 2479.74).

Fig. 5.

The hazard ratio compares the rate of an event happening between two or more treatments. The rate of metamorphosis under low-OA conditions was 1.05 times faster than that of mid-OA, but not significantly different. The rate of metamorphosis for larvae from Orient (O) was 0.78 times slower than that of larvae from Groton (G), and this difference was significant.