Search for: All records

Abstract BackgroundDespite exhibiting one of the longest migrations in the world, half of the humpback whale migratory cycle has remained unexamined. Until now, no study has provided a continuous description of humpback whale migratory behavior from a feeding ground to a calving ground. We present new information on satellite-derived offshore migratory movements of 16 Breeding Stock G humpback whales from Antarctic feeding grounds to South American calving grounds. Satellite locations were used to demonstrate migratory corridors, while the impact of departure date on migration speed was assessed using a linear regression. A Bayesian hierarchical state–space animal movement model (HSSM) was utilized to investigate the presence of Area Restricted Search (ARS) en route. Results35,642 Argos locations from 16 tagged whales from 2012 to 2017 were collected. The 16 whales were tracked for a mean of 38.5 days of migration (range 10–151 days). The length of individually derived tracks ranged from 645 to 6381 km. Humpbacks were widely dispersed geographically during the initial and middle stages of their migration, but convened in two convergence regions near the southernmost point of Chile as well as Peru’s Illescas Peninsula. The state–space model showed almost no instances of ARS along the migratory route. The linear regression assessing whether departure date affected migration speed showed suggestive but inconclusive support for a positive trend between the two variables. Results suggestive of stratification by sex and reproductive status were found for departure date and route choice. ConclusionsThis multi-year study sets a baseline against which the effects of climate change on humpback whales can be studied across years and conditions and provides an excellent starting point for the investigation into humpback whale migration. 

Machine learning algorithms are often used to model and predict animal habitat selection—the relationships between animal occurrences and habitat characteristics. For broadly distributed species, habitat selection often varies among populations and regions; thus, it would seem preferable to fit region- or population-specific models of habitat selection for more accurate inference and prediction, rather than fitting large-scale models using pooled data. However, where the aim is to make range-wide predictions, including areas for which there are no existing data or models of habitat selection, how can regional models best be combined? We propose that ensemble approaches commonly used to combine different algorithms for a single region can be reframed, treating regional habitat selection models as the candidate models. By doing so, we can incorporate regional variation when fitting predictive models of animal habitat selection across large ranges. We test this approach using satellite telemetry data from 168 humpback whales across five geographic regions in the Southern Ocean. Using random forests, we fitted a large-scale model relating humpback whale locations, versus background locations, to 10 environmental covariates, and made a circumpolar prediction of humpback whale habitat selection. We also fitted five regional models, the predictions of which we used as input features for four ensemble approaches: an unweighted ensemble, an ensemble weighted by environmental similarity in each cell, stacked generalization, and a hybrid approach wherein the environmental covariates and regional predictions were used as input features in a new model. We tested the predictive performance of these approaches on an independent validation dataset of humpback whale sightings and whaling catches. These multiregional ensemble approaches resulted in models with higher predictive performance than the circumpolar naive model. These approaches can be used to incorporate regional variation in animal habitat selection when fitting range-wide predictive models using machine learning algorithms. This can yield more accurate predictions across regions or populations of animals that may show variation in habitat selection.