The impacts of climate change in Antarctica and the Southern Ocean are not uniform and ice‐obligate species with dissimilar life‐history characteristics will likely respond differently to their changing ecosystems. We use a unique data set of Weddell
Satellites Over Seals (SOS), a project initiated in late 2016, is a crowdsourced method to determine factors behind the presence/absence patterns and to ultimately determine the global population of the Weddell seal (
- NSF-PAR ID:
- 10457570
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- Remote Sensing in Ecology and Conservation
- Volume:
- 6
- Issue:
- 1
- ISSN:
- 2056-3485
- Page Range / eLocation ID:
- p. 70-78
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract Leptonychotes weddellii and crabeater seals' (CESs)Lobodon carcinophaga breeding season distribution in the Weddell Sea, determined from satellite imagery. We contrast the theoretical climate impacts on both ice‐obligate predators who differ in life‐history characteristics: CESs are highly specialized Antarctic krillEuphausia superba predators and breed in the seasonal pack ice; Weddell seals (WESs) are generalist predators and breed on comparatively stable fast ice. We used presence–absence data and a suite of remotely sensed environmental variables to build habitat models. Each of the environmental predictors is multiplied by a ‘climate change score’ based on known responses to climate change to create a ‘change importance product’. Results show CESs are more sensitive to climate change than WESs. Crabeater seals prefer to breed close to krill, and the compounding effects of changing sea ice concentrations and sea surface temperatures, the proximity to krill and abundance of stable breeding ice, can influence their post‐breeding foraging success and ultimately their future breeding success. But in contrast to the Ross Sea, here WESs prefer to breed closer to larger colonies of emperor penguins (Aptenodytes forsteri ). This suggests that the Weddell Sea may currently be prey‐abundant, allowing the only two air‐breathing Antarctic silverfish predators (Pleuragramma antarctica ) (WESs and emperor penguins) to breed closer to each other. This is the first basin‐scale, region‐specific comparison of breeding season habitat in these two key Antarctic predators based on real‐world data to compare climate change responses. This work shows that broad‐brush, basin‐scale approaches to understanding species‐specific responses to climate change are not always appropriate, and regional models are needed—especially when designing marine protected areas. -
more » « less
Abstract We introduce a semiautomated machine learning method that employs high‐resolution imagery for the species‐level classification of Antarctic pack‐ice seals. By incorporating the spatial distribution of hauled‐out seals on ice into our analytical framework, we significantly enhance the accuracy of species identification. Employing a Random Forest model, we achieved 97.4% accuracy for crabeater seals and 98.0% for Weddell seals. To further refine our classification, we included three linearity measures: mean distance to a group's regression line, straightness index, and sinuosity index. Additional variables, such as the number of neighboring seals within a 250 m radius and distance of individual seals to the sea ice edge, also contributed to improved accuracy. Our study marks a significant advancement in the development of a cost‐effective, unified Antarctic seal monitoring system, enhancing our understanding of seal spatial behavior and enabling more effective population tracking amid environmental changes.
-
Pack-ice seals are key indicator species in the Southern Ocean. Their large size (2–4 m) and continent-wide distribution make them ideal candidates for monitoring programs via very-high-resolution satellite imagery. The sheer volume of imagery required, however, hampers our ability to rely on manual annotation alone. Here, we present SealNet 2.0, a fully automated approach to seal detection that couples a sea ice segmentation model to find potential seal habitats with an ensemble of semantic segmentation convolutional neural network models for seal detection. Our best ensemble attains 0.806 precision and 0.640 recall on an out-of-sample test dataset, surpassing two trained human observers. Built upon the original SealNet, it outperforms its predecessor by using annotation datasets focused on sea ice only, a comprehensive hyperparameter study leveraging substantial high-performance computing resources, and post-processing through regression head outputs and segmentation head logits at predicted seal locations. Even with a simplified version of our ensemble model, using AI predictions as a guide dramatically boosted the precision and recall of two human experts, showing potential as a training device for novice seal annotators. Like human observers, the performance of our automated approach deteriorates with terrain ruggedness, highlighting the need for statistical treatment to draw global population estimates from AI output.more » « less
-
more » « less
Abstract The Weddell seal (
Leptonychotes weddellii ) thrives in its extreme Antarctic environment. We generated the Weddell seal genome assembly and a high-quality annotation to investigate genome-wide evolutionary pressures that underlie its phenotype and to study genes implicated in hypoxia tolerance and a lipid-based metabolism. Genome-wide analyses included gene family expansion/contraction, positive selection, and diverged sequence (acceleration) compared to other placental mammals, identifying selection in coding and non-coding sequence in five pathways that may shape cardiovascular phenotype. Lipid metabolism as well as hypoxia genes contained more accelerated regions in the Weddell seal compared to genomic background. Top-significant genes wereSUMO2 andEP300 ; both regulate hypoxia inducible factor signaling. Liver expression of four genes with the strongest acceleration signals differ between Weddell seals and a terrestrial mammal, sheep. We also report a high-density lipoprotein-like particle in Weddell seal serum not present in other mammals, including the shallow-diving harbor seal. -
State-of-the-art deep learning technology has been successfully applied to relatively small selected areas of very high spatial resolution (0.15 and 0.25 m) optical aerial imagery acquired by a fixed-wing aircraft to automatically characterize ice-wedge polygons (IWPs) in the Arctic tundra. However, any mapping of IWPs at regional to continental scales requires images acquired on different sensor platforms (particularly satellite) and a refined understanding of the performance stability of the method across sensor platforms through reliable evaluation assessments. In this study, we examined the transferability of a deep learning Mask Region-Based Convolutional Neural Network (R-CNN) model for mapping IWPs in satellite remote sensing imagery (~0.5 m) covering 272 km2 and unmanned aerial vehicle (UAV) (0.02 m) imagery covering 0.32 km2. Multi-spectral images were obtained from the WorldView-2 satellite sensor and pan-sharpened to ~0.5 m, and a 20 mp CMOS sensor camera onboard a UAV, respectively. The training dataset included 25,489 and 6022 manually delineated IWPs from satellite and fixed-wing aircraft aerial imagery near the Arctic Coastal Plain, northern Alaska. Quantitative assessments showed that individual IWPs were correctly detected at up to 72% and 70%, and delineated at up to 73% and 68% F1 score accuracy levels for satellite and UAV images, respectively. Expert-based qualitative assessments showed that IWPs were correctly detected at good (40–60%) and excellent (80–100%) accuracy levels for satellite and UAV images, respectively, and delineated at excellent (80–100%) level for both images. We found that (1) regardless of spatial resolution and spectral bands, the deep learning Mask R-CNN model effectively mapped IWPs in both remote sensing satellite and UAV images; (2) the model achieved a better accuracy in detection with finer image resolution, such as UAV imagery, yet a better accuracy in delineation with coarser image resolution, such as satellite imagery; (3) increasing the number of training data with different resolutions between the training and actual application imagery does not necessarily result in better performance of the Mask R-CNN in IWPs mapping; (4) and overall, the model underestimates the total number of IWPs particularly in terms of disjoint/incomplete IWPs.more » « less
