A Bayesian assessment of the conservation status of humpback whales (Megaptera novaeangliae) in the western South Atlantic Ocean

Résumé

The population of humpback whales (Megaptera novaeangliae) wintering off the eastern coast of South America is referred to by the International Whaling Commission as ‘Breeding Stock A’ (BSA). This population was heavily exploited in 20th century modern commercial whaling operations. After more than 30 years of protection, its present status remains unknown. A deterministic sex and age-aggregated population dynamics model was used to estimate the pre-exploitation population size (K), the maximum net recruitment rate (rmax), the maximum depletion level (Nmin/K), and other quantities of interest of BSA. Input data included modern whaling catch series, absolute estimates of abundance, observed growth rates and indices of relative abundance. A Bayesian statistical method was used to calculate probability distributions for the model parameters. Prior distributions were set on rmax – an uninformative (Uniform [0, 0.106]) and an informative (Normal [0.067, 0.042]) – and on the population size in 2005 – N2005 (Uniform [500, 22,000]). A total of 10,000 samples were used to compute the joint posterior distribution of the model parameters using the Sampling-Importance-Resampling algorithm. Sensitivity of model outputs to the priors on rmax, a genetic constraint, data inclusion and catch allocation scenarios was investigated. Medians of the posterior probability distributions of quantities of interest for the base case scenario were: rmax = 0.069 (95% probability intervals [PI] = 0.013–0.104), K = 24,558 (95% PI = 22,791–31,118), Nmin/K = 2% (PI = 0.31%–12.5%), N2006/ K = 27.4% (PI = 18.3%–39.5%), N2020/K = 61.8% (PI = 23.8%–88.6%), and N2040/K = 97.3% (PI = 31.6%–99.9%). Despite apparent recovery in the past three decades, the western South Atlantic humpback whale population is still low relative to its pre-exploitation size and requires continued conservation efforts.