Identifying seasonal distribution patterns of fin whales across the Scotia Sea and the Antarctic Peninsula region using a novel approach combining habitat suitability models and ensemble learning methods

Abstract

Following their near extirpation by industrial whaling of the 20th century, the population status of Southern Hemisphere fin whales (SHFW) remains unknown. Systematic surveys estimating fin whale abundance in the Southern Ocean are not yet available. Records of fin whale sightings have been collected by a variety of organisations over the past few decades, incorporating both opportunistic data and dedicated survey data. Together, these isolated data sets represent a potentially valuable source of information on the seasonality, distribution and abundance of SHFW. We compiled records across 40 years from the Antarctic Peninsula and Scotia Sea from multiple sources and used a novel approach combining ensemble learning and a maximum entropy model to estimate abundance and distribution of SHFW in this region. Our results show a seasonal distribution pattern with pronounced centres of distribution from January-March along the West Antarctic Peninsula. Our new approach allowed us to estimate abundance of SHFW for discrete areas from a mixed data set of mainly opportunistic presence only data.

Historical Mitogenomic Diversity and Population Structuring of Southern Hemisphere Fin Whales

“…Fin whale recovery since the whaling is now evident with observations of fin whales increasing in the high latitudes of the western South Atlantic and eastern South Pacific in summer. High densities have been reported particularly in the southwest of the Scotia Arc, around the South Orkney Islands, and the Elephant Island near the Antarctic Peninsula, a location where they were also historically abundant based on catch records [7,12,110,116]. Fin whales, among other baleen whales, play a key role in nutrient cycling, through defecation, that can subsequently result in positive feedback loops and increased primary productivity [1].…”

Short Communication: Lateralised lungefeeding behaviour in Southern Hemisphere fin whales at Antarctic feeding grounds

“…We did not sample for prey, but echosounder records, as well as visual observations, indicated that Euphausia superba (Antarctic krill) was present in the study area (Meyer & Wessels, 2018). The presence of other krill predators, such as Arctocephalus gazella (Antarctic fur seals) and multiple species of birds, further indicated the availability of prey in the upper water column (Herr et al, 2022).…”

“…In the Northern Hemisphere, fin whales have been shown to display a strong right-biased lateralisation of ~90% during surface-feeding lunges (Tershy & Wiley, 1992), while information on foraging lateralisation has yet to be documented for Southern Hemisphere fin whales (SHFW). SHFWs are known to congregate at Antarctic feeding grounds during austral summers to forage on Euphausiids (krill) (Santora et al, 2014;Herr et al, 2022a;Viquerat et al, 2022), often forming large feeding aggregations involving up to 300 individuals (Herr et al, 2022a;2022b). B. physalus possess an asymmetrically pigmented cephalic region, the left side being entirely dark slate and the right side being white on the ventral side and light grey in the dorsal cephalic area (Wursig et al, 2017).…”

Evidence from interpretable machine learning to inform spatial management of Palau's tuna fisheries

“…We then used ML‐based modeling approaches (Kuhn & Johnson, 2013), coupled with recent developments in explainable artificial intelligence (AI) tools (Lundberg et al, 2020; Qiu et al, 2022; Scholbeck et al, 2020), to derive interpretable species‐specific and spatially resolved catch predictions for pelagic longline fishing fleets that operate in the Palau EEZ. ML approaches are increasingly used in a wide range of knowledge domains including medicine, finance, geoscience, ecology, paleobiology, climatology, fisheries, marine spatial planning, and economics to derive informed predictions from data that could include spatial–temporal structures, nonlinear predictor functional form, and complex predictor interactions (Bergen et al, 2023; Dedman et al, 2017; Effrosynidis et al, 2020; Foster et al, 2022; Gerassis et al, 2021; Sokhansanj & Rosen, 2022; Viquerat et al, 2022; Yang et al, 2022). ML‐based approaches are powerful tools for applied predictive modeling and make few assumptions about data structures (Kuhn & Johnson, 2013).…”