An official website of the United States government
There is limited information about the biology and seasonal distribution of bearded seals (Erignathus barbatus) in Greenland. The species is highly ice-associated and depends on sea ice for hauling out and giving birth, making it vulnerable to climate change. We investigated the seasonality and distribution of bearded seal vocalizations at seven different locations across southern Baffin Bay and Davis Strait, West Greenland. Aural M2 and HARUphone recorders were deployed on the sea bottom during 2006–2007 and 2011–2013. Recordings were analyzed for presence/absence of bearded seal calls relative to location (including distance to shore and depth), mean sea ice concentration and diel patterns. Calling occurred between November and late June with most intense calling during the mating season at all sites. There was a clear effect of depth and distance to shore on the number of detections, and the Greenland shelf (< 300 m) appeared to be the preferred habitat for bearded seals during the mating season. These results suggest that bearded seals may retreat with the receding sea ice to Canada during summer or possibly spend the summer along the West Greenland coast. It is also possible that, due to seasonal changes in bearded seal vocal behavior, animals may have been present in our study area in summer, but silent. The number of detections was affected by the timing of sea ice formation but not sea ice concentration. Diel patterns were consistent with patterns found in other parts of the Arctic, with a peak during early morning (0400 local) and a minimum during late afternoon (1600 local). While vocalization studies have been conducted on bearded seals in Norwegian, Canadian, northwest Greenland, and Alaskan territories, this study fills the gap between these areas.
more »
« less
Computer Vision for Bioacoustics: Detection of Bearded Seal Vocalizations in the Chukchi Shelf Using YOLOV5
Year-round recordings of bearded seal calls were collected in the northeastern edge of the Chukchi Continental Slope (Alaska, within the Arctic Circle) in 2016–2017, 2018–2019, and 2019–2020. While the underwater vocalizations of bearded seals are often analyzed manually or using automatic detections manually validated, in this article, a detection and classification system (DCS) based on the You Only Look Once Version 5 (YOLOV5) algorithm is proposed. With YOLOV5, the network learns how to detect and classify these marine mammals’ calls using the principle of computer vision for object detection in images where bounding boxes enclose the objects of interest. During training, validation, and testing, YOLOV5 achieved an accuracy of 96.54%, 93.36%, and 93.87%, respectively. The DCS was applied to the three-yearlong dataset, and an analysis of the vocal behavior of the bearded seals showed that there exists a geographical dependence where this species prefers shallower water depths in the Chukchi Continental Slope. Another advantage of using YOLOV5 over other typical DCS is that the predicted bounding boxes have embedded statistical information about the vocalization, such as the duration, bandwidth, and center frequency of the signals. This additional information equips biologists with statistical data that facilitate the analysis of animal vocal behavior.
more »
« less
- Award ID(s):
- 2322365
- PAR ID:
- 10535944
- Publisher / Repository:
- IEEE Journal of Oceanic Engineering
- Date Published:
- Journal Name:
- IEEE Journal of Oceanic Engineering
- Volume:
- 49
- Issue:
- 1
- ISSN:
- 0364-9059
- Page Range / eLocation ID:
- 133 to 144
- Subject(s) / Keyword(s):
- Seals Spectrogram Object detection Computer vision Arctic Oceans Prediction algorithms Arctic bearded seals computer vision deep learning marine mammals You Only Look Once Version 5 (YOLOV5)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Domoic acid (DA) and saxitoxin (STX)‐producing algae are present in Alaskan seas, presenting exposure risks to marine mammals that may be increasing due to climate change. To investigate potential increases in exposure risks to four pagophilic ice seal species (Erignathus barbatus, bearded seals;Pusa hispida, ringed seals;Phoca largha, spotted seals; andHistriophoca fasciata, ribbon seals), this study analyzed samples from 998 seals harvested for subsistence purposes in western and northern Alaska during 2005–2019 for DA and STX. Both toxins were detected in bearded, ringed, and spotted seals, though no clinical signs of acute neurotoxicity were reported in harvested seals. Bearded seals had the highest prevalence of each toxin, followed by ringed seals. Bearded seal stomach content samples from the Bering Sea showed a significant increase in DA prevalence with time (logistic regression,p = .004). These findings are consistent with predicted northward expansion of DA‐producing algae. A comparison of paired samples taken from the stomachs and colons of 15 seals found that colon content consistently had higher concentrations of both toxins. Collectively, these results suggest that ice seals, particularly bearded seals (benthic foraging specialists), are suitable sentinels for monitoring HAB prevalence in the Pacific Arctic and subarctic.more » « less
-
Using unmanned aerial vehicles (UAVs) to track multiple individuals simultaneously in their natural environment is a powerful approach for better understanding the collective behavior of primates. Previous studies have demonstrated the feasibility of automating primate behavior classification from video data, but these studies have been carried out in captivity or from ground-based cameras. However, to understand group behavior and the self-organization of a collective, the whole troop needs to be seen at a scale where behavior can be seen in relation to the natural environment in which ecological decisions are made. To tackle this challenge, this study presents a novel dataset for baboon detection, tracking, and behavior recognition from drone videos where troops are observed on-the-move in their natural environment as they move to and from their sleeping sites. Videos were captured from drones at Mpala Research Centre, a research station located in Laikipia County, in central Kenya. The baboon detection dataset was created by manually annotating all baboons in drone videos with bounding boxes. A tiling method was subsequently applied to create a pyramid of images at various scales from the original 5.3K resolution images, resulting in approximately 30K images used for baboon detection. The baboon tracking dataset is derived from the baboon detection dataset, where bounding boxes are consistently assigned the same ID throughout the video. This process resulted in half an hour of dense tracking data. The baboon behavior recognition dataset was generated by converting tracks into mini-scenes, a video subregion centered on each animal. These mini-scenes were annotated with 12 distinct behavior types and one additional category for occlusion, resulting in over 20 hours of data. Benchmark results show mean average precision (mAP) of 92.62% for the YOLOv8-X detection model, multiple object tracking precision (MOTP) of 87.22% for the DeepSORT tracking algorithm, and micro top-1 accuracy of 64.89% for the X3D behavior recognition model. Using deep learning to rapidly and accurately classify wildlife behavior from drone footage facilitates non-invasive data collection on behavior enabling the behavior of a whole group to be systematically and accurately recorded. The dataset can be accessed at https://baboonland.xyz.more » « less
-
Pavement surface distresses are analyzed by transportation agencies to determine section performance across their pavement networks. To efficiently collect and evaluate thousands of lane-miles, automated processes utilizing image-capturing techniques and detection algorithms are applied to perform these tasks. However, the precision of this novel technology often leads to inaccuracies that must be verified by pavement engineers. Developments in artificial intelligence and machine learning (AI/ML) can aid in the progress of more robust and precise detection algorithms. Deep learning models are efficient for visual distress identification of pavement. With the use of 2D/3D pavement images, surface distress analysis can help train models to efficiently detect and classify surface distresses that may be caused by traffic loading, weather, aging, and other environmental factors. The formation of these distresses is developing at a higher rate in coastal regions, where extreme weather phenomena are more frequent and intensive. This study aims to develop a YOLOv5 model with 2D/3D images collected in the states of Louisiana, Mississippi, and Texas in the U.S. to establish a library of data on pavement sections near the Gulf of Mexico. Images with a resolution of 4096 × 2048 are annotated by utilizing bounding boxes based on a class list of nine distress and non-distress objects. Along with emphasis on efforts to detect cracks in the presence of background noise on asphalt pavements, six scenarios for augmentation were made to evaluate the model’s performance based on flip probability in the horizontal and vertical directions. The YOLOv5 models are able to detect defined distresses consistently, with the highest mAP50 scores ranging from 0.437 to 0.462 throughout the training scenarios.more » « less
-
Abstract A regional coupled sea ice‐ocean model and mooring/shipboard measurements are used to investigate the origins, seasonality, and downstream fate of the Chukchi Slope Current (CSC). Three years (2013–2015) of model integration indicates that, in the mean, the model slope current transports ∼0.45 Sv of Pacific water northwestward along the Chukchi continental slope. Only 62% of this water emanates from Barrow Canyon, while the rest (38%) is fed by a westward jet extending from the southern Beaufort Sea. The jet merges with the outflow from the canyon, forming the CSC. Due to these two distinct origins, the slope current in the model has a double velocity core at times. This is consistent with the double‐core structure of the slope current seen in ship‐based observations. Seasonal changes in the volume, heat, and freshwater transports by the slope current appear to be related to the changes in the upstream flows. A tracer diagnostic in the model suggests that the part of the slope current over the upper continental slope continues westward toward the East Siberian Sea, while the portion of the current overlying deeper isobaths flows northward into the Chukchi Borderland, where it ultimately gets entrained into the Beaufort Gyre. Our study provides a detailed and complete picture of the slope current.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1109/JOE.2023.3307175
