Can a present-day thermal niche be preserved in a warming climate by a shift in phenology? A case study with sea turtles

Abstract

How species respond to climate change may impact their extinction probability. Here we link climatology and ecology to tackle a globally important conservation question. For sea turtles, there are concerns that climate warming will cause both the feminization of populations as well as reduced hatchling survival. For 58 nesting sites across the world spanning all seven sea turtle species, we investigated whether warming might be avoided by shifts in nesting phenology to a cooler part of the year. We show that even with the most extreme phenological shift that has been reported to date-an 18-day advance in nesting per °C increase in sea surface temperature (SST)-temperatures will continue to increase at nesting sites with climate warming. We estimate that SST at nesting sites will rise by an average of 0.6°C (standard deviation = 0.9°C, n = 58) when we model a 1.5°C rise in SST combined with a best-case-scenario shift in nesting. Since sea turtles exhibit temperature-dependent sex determination, these temperature rises could lead to increasingly female-biased sex ratios as well as reduced hatchling production at sites across the world. These findings underscore concerns for the long-term survival of this iconic group.

Keywords: climate change adaptation; climatology; conservation; endangered species; marine turtles.

Figure 1.

Conceptual framework for how a rise in SST may be mitigated by a phenological shift of the nesting season. The green and brown lines represent the current and projected SST. The filled circles represent the peak of the sea turtle nesting season. In this case, a projected 1.5°C rise in SST translates to less than 1.5°C rise in SST during peak nesting season due to a phenological shift to earlier nesting (blue arrow). The turtle image was kindly provided by NOAA Fisheries (www.fisheries.noaa.gov).

Figure 2.

Phenological shifts of the nesting season have variable impacts at different sites. We modelled how a 1.5°C rise in SST combined with a 27-day advance of the nesting season would impact SST at 58 sea turtle nesting sites. Here we highlight three case examples: (a) On Saint Eustatius in the Caribbean, a shift of the nesting season does not mitigate any warming SST. (b) On Sal in the Northeast Atlantic, a shift of the nesting season mitigates approximately 60% of a 1.5°C rise in SST. (c) In Florida in the Northwest Atlantic (c), almost 100% of warming is mitigated by a best-case-scenario phenological shift. Open circles represent mean monthly SST and the black line represents the sine fit. The grey line represents projected conditions after a 1.5°C rise in SST. The filled circles represent a month during the peak of the nesting season in their respective scenarios. For easy comparison between subpanels, the vertical bars represent 1.5°C. The geographical location of these three study sites is highlighted in figure 3.

Figure 3.

Global patterns for the interaction of rising SST and shifting nesting phenology. We show how a 27-day shift of the nesting in response to a 1.5°C rise in SST would affect SST at all 58 sites used in our study. The filled slice of each pie chart represents the proportion of the 1.5°C rise in SST that occurred, such that a completely full pie indicates that no warming is mitigated. Colours represent the different turtle species. The sites highlighted in figure 2 are indicated here: (a) Saint Eustatius (the Netherlands) in the Caribbean, (b) Sal (Cape Verde) in the Northeast Atlantic and (c) Florida (United States of America) in the Northwest Atlantic.

Figure 4.

Phenological shifts in nesting at sites farthest from the equator have more impact on mitigating warming SST. The proportion of warming that was mitigated is plotted across sites. A 27-day shift in nesting in response to a 1.5°C rise in SST was most effective at maintaining current SST at sites greater than 30° latitude