SealNet: A fully-automated pack-ice seal detection pipeline for sub-meter satellite imagery

Gonçalves, Bento Collares; Spitzbart, B.; Lynch, Heather J.

doi:10.1016/j.rse.2019.111617

Cited by 32 publications

(39 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The roughness of the terrain in the Antarctic pack ice and the available quality of satellite images still makes it difficult to discern between two seals lying next to each other and to distinguish pups from adults. Although making use of citizen scientist votes and comparing it to an expert voter is biased, automated image recognition only detects 30% of seals, even during summer when the ice floes are smaller and flatter, and seals haul out in high numbers close to each other (Gonçalves, Spitzbart, & Lynch, 2020).…”

Section: Caveats Of the Datamentioning

confidence: 99%

Citizen science and habitat modelling facilitates conservation planning for crabeater seals in the Weddell Sea

Wege

Salas

LaRue

2020

Diversity and Distributions

View full text Add to dashboard Cite

Aim Creating a network of marine protected areas in the Southern Ocean requires extensive knowledge on species’ abundances, distributions and population trends especially in the Weddell Sea where year‐round pack ice makes most of the Weddell Sea inaccessible. We combine satellite images and citizen science to model habitat suitability for crabeater seals (Lobodon carcinophaga) throughout the Weddell Sea. Location Weddell Sea, Antarctica. Methods High‐resolution satellite images covering 18,219 km2 of the Weddell Sea during crabeater seal breeding season (October—November) were hosted on the crowd‐sourcing platform Tomnod (DigitalGlobe). Citizen scientists marked “maps” where seals were present/absent and these votes were compared with the votes of an experienced observer. Correction factors were used to correct votes to either a continuous probability of seal presence, or a binary seal presence/absence value. We modelled probability of seal presence using ensemble models of Random Forests (RF), Boosted Regression Trees (BRT) and Support Vector Machines (SVM), and used fitted Maxent models to model seal presence/absence data. Results Model predictive power was low (RF: R2 = 0.076 ± 0.002: BRT: R2 = 0.086 ± 0.0008; SVM: R2 = 0.082 ± 0.003) to average (Maxent: AUC = 0.71 ± 0.004). Distance to the ice edge and bathymetry were the most important variables that influenced crabeater seal distribution. Main conclusions Crabeater seals were more likely to be present over abyssal water, which coincides with typical adult Antarctic krill habitat — crabeater seal preferred prey. Where ice concentrations were more variable, that is more accessible, crabeater seals were also more likely to occur. Results agreed with the known ecology of crabeaters seals and the abundance, distribution and ecology of Antarctic krill. We were able to survey the largest area ever surveyed in the Weddell Sea and provide a model to assist furthering policy around the proposed protected area.

show abstract

Section: Caveats Of the Datamentioning

confidence: 99%

Citizen science and habitat modelling facilitates conservation planning for crabeater seals in the Weddell Sea

Wege

Salas

LaRue

2020

Diversity and Distributions

View full text Add to dashboard Cite

show abstract

“…While remote sensing is a promising tool to better monitor these predators, it does not currently yield as detailed information as traditional visual surveys made from ship or manned aircraft as it is still difficult to reliably detect or distinguish species from remote sensing platforms. Nonetheless, further advances in remote sensing platforms and automated image processing (Gonçalves et al 2020) will make regional and Southern Ocean-wide spatiotemporal monitoring of pack-ice seal population trends possible.…”

Section: Discussionmentioning

confidence: 99%

Habitat-based density models of pack-ice seal distribution in the southern Weddell Sea, Antarctica

Oosthuizen

Reisinger

Bester

et al. 2021

Mar. Ecol. Prog. Ser.

View full text Add to dashboard Cite

Climate variability and changes in sea ice dynamics have caused several ice-obligate or krill-dependent populations of marine predators to decline, eliciting concern about their demographic persistence and the indirect ecological consequences that predator depletions may have on marine ecosystems. Pack-ice seals are dominant ice-obligate predators in the Antarctic marine ecosystem, but there is considerable uncertainty about their abundance and population trends. We modelled the density and distribution of pack-ice seals as a function of environmental covariates in the southern Weddell Sea, Antarctica. Our density surface modelling approach used data from aerial surveys of pack-ice seals collected in the 2013/14 austral summer. Crabeater seals Lobodon carcinophaga, the most numerous pack-ice seal we observed, occurred at the highest densities in areas with extensive sea ice near the continental shelf break, but were almost absent in areas of similar sea ice concentration in the southern extent of the Weddell Sea. The highest densities of Weddell seals Leptonychotes weddelli, which were less abundant than crabeater seals within the pack-ice habitat, were predicted to occur over the continental shelf, near the shelf break. The distribution of both seal species broadly corresponded with the distribution and relative abundance of their main prey (Antarctic krill Euphausia superba and Antarctic silverfish Pleuragramma antarctica) obtained from concurrent ecosystem surveys. Ross seals Ommatophoca rossii and leopard seals Hydrurga leptonyx were not detected at all and are apparently rare within the southern Weddell Sea. These results can contribute to biodiversity assessments in the context of marine protected area planning in this region of the Southern Ocean.

show abstract

“…Remote observation might improve this (e.g., Labrousse et al, 2019) and satellite surveillance can enable remote area assessment, at large scale and much reduced cost (LaRue and Knight, 2014). Counting animals from satellite images manually can be time-consuming and researchers are harnessing Deep Learning (e.g., Gonçalves et al, 2020) and citizen science through crowdsourcing, where online volunteers identify animals in the images (LaRue et al, 2019).…”

Section: Population Monitoringmentioning

confidence: 99%

Marine Ecosystem Assessment for the Southern Ocean: Birds and Marine Mammals in a Changing Climate

et al. 2020

View full text Add to dashboard Cite

The massive number of seabirds (penguins and procellariiformes) and marine mammals (cetaceans and pinnipeds) -referred to here as top predators -is one of the most iconic components of the Antarctic and Southern Ocean. They play an important role as highly mobile consumers, structuring and connecting pelagic marine food webs and are widely studied relative to other taxa. Many birds and mammals establish dense breeding colonies or use haul-out sites, making them relatively easy to study. Cetaceans, however, spend their lives at sea and thus aspects of their life cycle are more complicated to monitor and study. Nevertheless, they all feed at sea and their reproductive success depends on the food availability in the marine environment, hence they are considered useful indicators of the state of the marine resources. In general, top predators have large body sizes that allow for instrumentation with miniature datarecording or transmitting devices to monitor their activities at sea. Development of scientific techniques to study reproduction and foraging of top predators has led to substantial scientific literature on their population trends, key biological parameters, migratory patterns, foraging and feeding ecology, and linkages with atmospheric or oceanographic dynamics, for a number of species and regions. We briefly summarize

show abstract

SealNet: A fully-automated pack-ice seal detection pipeline for sub-meter satellite imagery

Cited by 32 publications

References 43 publications

Citizen science and habitat modelling facilitates conservation planning for crabeater seals in the Weddell Sea

Citizen science and habitat modelling facilitates conservation planning for crabeater seals in the Weddell Sea

Habitat-based density models of pack-ice seal distribution in the southern Weddell Sea, Antarctica

Marine Ecosystem Assessment for the Southern Ocean: Birds and Marine Mammals in a Changing Climate

Contact Info

Product

Resources

About