Artificial light at night during early development directly affects embryonic but not larval quality in a wild coral reef fish

Abstract

Artificial light at night (ALAN) is an emergent yet already global form of sensory pollution. However, its effects on marine environments remain poorly understood compared to those on terrestrial ecosystems. Low-latitude ecosystems such as shallow coral reefs might be at greater risk as they experience little change in annual day length and reef organisms rely on moonlight illumination as a zeitgeber for critical biological processes. Moreover, many coral reef fish are demersal spawners, making them vulnerable to the effects of ALAN from early life. We performed a field experiment to determine whether artificial light affects the quality of fish embryos and newly hatched larvae by exposing wild nests of the orange-fin anemonefish (Amphiprion chrysopterus) to white light emitting diode (LED) light (22 ± 2.0 lx; 4000 K) throughout the 6-day embryonic development period. We also explored whether light pollution indirectly influences offspring traits by measuring parental care investment. Exposure to ALAN altered embryo quality, leading to a reduction in egg volume (2.40%) and yolk reserves (6.11%) alongside an increase in heart rate (7.42%) a few hours before hatching. These changes reflect higher metabolic demands of embryos developing under light-polluted conditions. As parental care investment was unaffected by light pollution, our results suggest that these effects are more likely the consequence of a direct effect of ALAN on embryogenesis. In contrast, there was no influence of artificial light on the larval morphology or swimming performance, suggesting that the direct effects of ALAN on fish embryos do not cascade onto the larval stage immediately after hatching. These results may suggest that embryos compensated for ALAN exposure to maintain their early post-hatching larval performance. Further studies are needed to investigate whether light pollution exposure during embryonic development has delayed effects on larval performance during the dispersal phase or on larval survival.

Keywords: ALAN; Pomacentridae; cascading effects; coral reef fish; early-life traits; ecological light pollution; offspring quality; parental care.

Figure 1

(a) Location of the anemonefish breeding pairs used for the study on the north coast of Mo’orea, French Polynesia. (b) Photograph of a custom-built underwater light system at night 1–1.5 m from an anemonefish nest (Frederic Zuberer). (c) Photograph of wild anemonefish embryos a few hours prior to hatching (Frederic Zuberer).

Figure 2

Embryonic quality traits measured prior to hatching before (first spawning) and after (second spawning) exposure of embryos to control (Ctrl) or light pollution (Alan) treatments. (a) Egg volume, (b) yolk sac area, (c) embryonic heart rate. Large dots represent means and smaller dots represent individual measures. Significance codes for Before–After contrasts within treatments and their differences among treatments are ^*** for P-value <0.001, ^** for P-value <0.01, ^* for P-value <0.05 and NS for P-value >0.05.

Figure 3

Larval quality traits measured at 0 days post-hatch before (first spawning) and after (second spawning) exposure of embryos to control (Ctrl) or light pollution (Alan) treatments. (a) Overall larval morphology expressed by the first principal component, (b) maximum swimming speed. Large dots represent means and smaller dots represent individual measures. Significance codes for Before–After contrasts within treatments and their differences among treatments are ^*** for P-value <0.001, ^** for P-value <0.01, ^* for P-value <0.05 and NS for P-value >0.05.

Figure 4

Mean proportion of time spent providing parental care for breeding pairs before (first spawning) and after (second spawning) exposure to control (Ctrl) or light pollution (Alan) treatments. Vertical error bars represent standard error of the mean, dots represent individual measures. Significance code for Before–After contrasts within treatments and their differences among treatments is NS for P-value >0.05.