Time-Varying Stock-Recruitment Model For Estimating Population Characteristics of Green Turtles

The stock-recruitment relationship (SR), customarily used in fisheries assessment, can be used to analyze demographic data of sea turtles to infer changes in hatchling production ( R ) as a function of nester abundance ( S ), recruitment rates and the influence of environmental conditions on these population features. The SR Cushing model (R=aS^b), where a and b are the model parameters) is well-suited for representing the dynamics of recovering populations, such as the green turtle ( Chelonia mydas ) in Campeche, Mexico. This study aimed to explore the SR Cushing model using a time series of the abundance of nesters and hatchlings (1984–2020). By applying local regressions (9-yr moving windows), we found that the time series of parameter b (the change in R as a function of S ) and the recruitment rate (hatchlings per nester) were inversely correlated with a 26-yr cycle of the Atlantic Multidecadal Oscillation –sea surface temperature (SST), over the Atlantic– (r^2=0.83) and (r^2=0.64), respectively, at a 3-yr lag). Model diagnostics using the time-dependent Cushing model substantiated that the log-normal distribution of hatchlings of C. mydas in Campeche depends on the abundance of nesting females and on a low frequency SST signal (r^2=0.98). The positive trend in nester numbers of green turtles in Campeche during the past 44 years may be the result of persistent conservation efforts, while the drastic and sporadic changes in the growth rate of annual arrivals and hatchling production are suggestive of population dynamics driven by low frequency, basin-wide environmental signals.