MEPS 389:257-269 (2009)  -  DOI: https://doi.org/10.3354/meps08206

ABSTRACT: A generalized additive mixed modeling approach was used to assess somatic growth for juvenile green turtles Chelonia mydas at 4 sites in 3 ecologically distinct foraging habitats along the east central coast of Florida, USA. The 3 habitats were a man-made nuclear submarine turning basin (Trident Submarine Basin), an estuary (Indian River Lagoon), and oceanic sabellariid worm rock reefs (Sebastian Inlet and St. Lucie Power Plant). Turtles from the Indian River Lagoon site grew significantly faster than turtles from the Trident Submarine Basin and sabellariid worm rock reef sites. There were no significant differences in growth rates between the sabellariid worm rock reef and Trident Submarine Basin sites. Non-monotonic or dome-shaped growth rate functions reflecting an immature peak in growth rates were observed for all 3 habitats. Growth rates peaked in 1998 for turtles in the Trident Submarine Basin and sabellariid worm rock reef habitats; since then growth rates have declined. This temporal decline in growth rates may reflect density-dependent effects on growth as more juveniles recruit to Florida foraging grounds, a direct result of increases in nest production at the primary rookeries (Costa Rica, Florida and Mexico). Developmental habitats are important for the survival of juvenile marine turtles. This study illustrates the degree to which juvenile growth rates vary among developmental habitats, which ultimately can affect the rate of growth and recovery potential of nesting stocks.

KEY WORDS: Chelonia mydas · Density-dependent effects · Developmental habitat · Florida · Foraging grounds · Generalized Additive Mixed Model · Green turtles · Growth rate