Sofie Van Parijs | Northeast Fisheries Science Center, NOAA (original) (raw)

Papers by Sofie Van Parijs

Research paper thumbnail of AM 41.1 Gulf of Mexico

Research paper thumbnail of Effects of ice cover on the behavioural patterns of aquatic-mating male bearded seals

Animal Behaviour, 2004

We used vocalizations to examine the behaviour of male bearded seals, Erignathus barbatus, in rel... more We used vocalizations to examine the behaviour of male bearded seals, Erignathus barbatus, in relation to ice cover over 2 consecutive years. Site fidelity varied between males. Territorial males returned to the same display locations in successive years, whereas roaming males returned to the study area but not to specific display locations. Males did not vocalize under the land-fast-ice edge and were therefore limited by access to areas of drift ice and open water for the purpose of displaying. Between-year fluctuations in ice cover restricted the number of displaying males during the early part of the mating season in April. Decreased ice cover during May resulted in more males displaying, but the number of males declined in June irrespective of the continued decline in ice cover. Roaming males were not heard when the ice cover was more than 60%, whereas territorial males were present during all ice conditions. Yearly fluctuations in ice cover may alter the long-term mating success of individuals and predictably available display areas may be preferred to areas that are more likely to be enclosed by fast-ice in some years.

Research paper thumbnail of Individuality in harp seal, Phoca groenlandica, pup vocalizations

Research paper thumbnail of Phase 4: DCL System Using Deep Learning Approaches for Land-Based or Ship-Based Real-Time Recognition and Localization of Marine Mammals - Distributed Processing and Big Data Applications

While the animal bioacoustics community at large is collecting huge amounts of acoustic data at a... more While the animal bioacoustics community at large is collecting huge amounts of acoustic data at an unprecedented pace, processing these data is problematic. Currently in bioacoustics, there is no effective way to achieve high performance computing using commericial off the shelf (COTS) or government off the shelf (GOTS) tools. Although several advances have been made in the open source and commercial software community, these offerings either support specific applications that do not integrate well with data formats in bioacoustics or they are too general. Furthermore, complex algorithms that use deep learning strategies require special considerations, such as very large libraiers of exemplars (whale sounds) readily available for algorithm training and testing. Detection-classification for passive acoustics is a data-mining strategy and our goals are aligned with best practices that appeal to the general data mining and machine learning communities where the problem of processing la...

Research paper thumbnail of R script for Zero Inflated Model and Likelihood Ratio

Research paper thumbnail of Summary of the detector performance data

Research paper thumbnail of Summary of the analysis of the southern right whale vocalisation data

Research paper thumbnail of See Profile

Vocalizations and movements suggest alternative mating tactics in male bearded seals

Research paper thumbnail of National Passive Acoustics Workshop Developing a Strategic Program Plan for NOAA’s Passive Acoustics

National Marine Fisheries Service William T. Hogarth, Assistant Administrator for FisheriesThis d... more National Marine Fisheries Service William T. Hogarth, Assistant Administrator for FisheriesThis document will initially be issued only online. It will be available at:

Research paper thumbnail of National Oceanic and Atmospheric Adminis… 113 PUBLICATIONS 1,773 CITATIONS SEE PROFILE

Male reproductive behaviour in the aquatic mating bearded seal (Erignathus barbatus).

Research paper thumbnail of Ecology of Beaked Whales and Sperm Whales in the Western North Atlantic Ocean: Insights from Passive Acoustic Monitoring

Ecology of Beaked Whales and Sperm Whales in the Western North Atlantic Ocean: Insights from Pass... more Ecology of Beaked Whales and Sperm Whales in the Western North Atlantic Ocean: Insights from Passive Acoustic Monitoring

Research paper thumbnail of Appendices 1-3 from The importance of migratory connectivity for global ocean policy

MiCO aims to coalesce existing knowledge on connectivity and migratory corridors of marine migrat... more MiCO aims to coalesce existing knowledge on connectivity and migratory corridors of marine migratory species that utilize areas beyond national jurisdiction. Numerous definitions of "migration" and "migratory species" exist, confusing efforts to develop a specific list of species to include within MiCO. Lascelles et al. (2014) assessed the current status and management needs of over 800 migratory marine species across the same four taxonomic groups as MiCO. They define migratory species as animals that move between "at least two jurisdictions during the course of their annual cycles," with specific species selected by expert groups. The list of species to be evaluated by MiCO was assembled from several sources: including Lascelles et al. (2014) (n = 829), the CMS Migratory Shark Species (n = 94), fish species managed by Regional Fisheries Management Organization (RFMO; n = 40), seabirds of the Migratory Bird Treaty Act (n = 171), and BirdLife International (n = 280). The initial 892 species to be evaluated by the MiCO system includes 440 fish, 346 seabird, 99 marine mammal, and 7 sea turtle species. Family Scientific Name Common Name CMS Annex (as of Aug 7 2018) IUCN Red List Status

Research paper thumbnail of Supplementary material from "The importance of migratory connectivity for global ocean policy

The distributions of migratory species in the ocean span local, national and international jurisd... more The distributions of migratory species in the ocean span local, national and international jurisdictions. Across these ecologically interconnected regions, migratory marine species interact with anthropogenic stressors throughout their lives. Migratory connectivity, the geographical linking of individuals and populations throughout their migratory cycles, influences how spatial and temporal dynamics of stressors affect migratory animals and scale up to influence population abundance, distribution and species persistence. Population declines of many migratory marine species have led to calls for connectivity knowledge, especially insights from animal tracking studies, to be more systematically and synthetically incorporated into decision-making. Inclusion of migratory connectivity in the design of conservation and management measures is critical to ensure they are appropriate for the level of risk associated with various degrees of connectivity. Three mechanisms exist to incorporate migratory connectivity into international marine policy which guides conservation implementation: site-selection criteria, network design criteria and policy recommendations. Here, we review the concept of migratory connectivity and its use in international policy, and describe the Migratory Connectivity in the Ocean system, a migratory connectivity evidence-base for the ocean. We propose that without such collaboration focused on migratory connectivity, efforts to effectively conserve these critical species across jurisdictions will have limited effect.

Research paper thumbnail of Studying marine mammal social systems

Pinnipeds and cetaceans often aggregate, sometimes in their thousands, and on occasion in excepti... more Pinnipeds and cetaceans often aggregate, sometimes in their thousands, and on occasion in exceptional densities. Many of these gatherings are not obviously the result of attraction to environmental features. Thus, these are actively maintained groups, not passive aggregations. Groups imply social structure, and social structure can affect ecology, genetics, population biology, and thus issues of conservation and management (Wilson 1975; Sutherland 1998). Hence, both for interest in the social lives of marine mammals, as well as a general understanding of the biology of the animals, their places in ecosystems, and the effects of human activities on them, we need to study social structure. Additionally, because the habitat of marine mammals is so different from that of terrestrial species, an understanding of the social structure of marine mammals provides an important comparative perspective on the forces of mammalian social evolution (Connor et al. 1998). Those who study the social ...

Research paper thumbnail of Phase 3: DCL System Using Deep Learning Approaches for Land-based or Ship-based Real-Time Recognition and Localization of Marine Mammals - Bioacoustic Applicaitons

Goals of this research phase is to investigate advanced detection and classification pardims usef... more Goals of this research phase is to investigate advanced detection and classification pardims useful for data-mining passive large passive acoustic archives. Technical objectives are to develop and refine a High Performance Computing, Acoustic Data Accelerator (HPC-ADA) along with MATLAB based software based on time series acoustic signal Detection cLassification using Machine learning Algorithms, called DeLMA. Data scientists and biologists integrate to use the HPC-ADA and DeLMA technologies to explore data using newly developed techniques aimed at inspection of data extracted at large spatial and temporal scales.

Research paper thumbnail of Fourth International Conference on the Effects of Noise on Aquatic Life

Hawaiian spinner dolphins display predictable daily behavior, using shallow bays to rest during t... more Hawaiian spinner dolphins display predictable daily behavior, using shallow bays to rest during the daytime. The frequency and intensity of the interactions between humans and dolphins in these bays has prompted concern for the animals. All previous research on the dolphins’ response to human activity has been conducted visually. In contrast, here we evaluate whether dolphins acoustically respond to human activity. Previous research found that the dolphins were silent during rest and that acoustic activity matched their general behavioral state. Therefore, we expected to see higher dolphin whistle activity before and after peak resting time and low activity to indicate rest. The daily pattern of dolphin whistle activity in two of the four bays matched what we expected. However, in two bays we did not find a drop in dolphin whistle activity. With regards to the acoustic response of the animals to human activity, we found relationships between the number of vessels and dolphin acousti...

Research paper thumbnail of Phase 1: DCL System Research Using Advanced Approaches for Land-based or Ship-based Real-Time Recognition and Localization of Marine Mammals - HPC System Implementation

We aim to investigate advancing the state of the art of detection, classification and localizatio... more We aim to investigate advancing the state of the art of detection, classification and localization (DCL) in the field of bioacoustics. The two primary goals are to develop transferable technologies for detection and classification in: (1) the area of advanced algorithms, such as deep learning and other methods; and (2) advanced systems, capable of real-time and archival and processing. This project will focus on long-term, continuous datasets to provide automatic recognition, minimizing human time to annotate the signals. Effort will begin by focusing on several years of multi-channel acoustic data collected in the Stellwagen Bank National Marine Sanctuary (SBNMS) between 2006 and 2010. Our efforts will incorporate existing technologies in the bioacoustics signal processing community, advanced high performance computing (HPC) systems, and new approaches aimed at automatically detecting-classifying and measuring features for species-specific marine mammal sounds within passive acoust...

Research paper thumbnail of Phase 2: DCL System Using Deep Learning Approaches for Land-based or Ship-based Real-Time Recognition and Localization of Marine Mammals - Machine Learning Detection Algorithms

ArXiv, 2016

Overarching goals for this work aim to advance the state of the art for detection, classification... more Overarching goals for this work aim to advance the state of the art for detection, classification and localization (DCL) in the field of bioacoustics. This goal is primarily achieved by building a generic framework for detection-classification (DC) using a fast, efficient and scalable architecture, demonstrating the capabilities of this system using on a variety of low-frequency mid-frequency cetacean sounds. Two primary goals are to develop transferable technologies for detection and classification in, one: the area of advanced algorithms, such as deep learning and other methods; and two: advanced systems, capable of real-time and archival processing. For each key area, we will focus on producing publications from this work and providing tools and software to the community where/when possible. Currently massive amounts of acoustic data are being collected by various institutions, corporations and national defense agencies. The long-term goal is to provide technical capability to an...

Research paper thumbnail of NOAA and BOEM Minimum Recommendations for Use of Passive Acoustic Listening Systems in Offshore Wind Energy Development Monitoring and Mitigation Programs

Frontiers in Marine Science, 2021

Offshore wind energy development is rapidly ramping up in United States (U.S.) waters in order to... more Offshore wind energy development is rapidly ramping up in United States (U.S.) waters in order to meet renewable energy goals. With a diverse suite of endangered large whale species and a multitude of other protected marine species frequenting these same waters, understanding the potential consequences of construction and operation activities is essential to advancing responsible offshore wind development. Passive acoustic monitoring (PAM) represents a newer technology that has become one of several methods of choice for monitoring trends in the presence of species, the soundscape, mitigating risk, and evaluating potential behavioral and distributional changes resulting from offshore wind activities. Federal and State regulators, the offshore wind industry, and environmental advocates require detailed information on PAM capabilities and techniques needed to promote efficient, consistent, and meaningful data collection efforts on local and regional scales. PAM during offshore wind co...

Research paper thumbnail of Evaluating the Efficacy of Acoustic Metrics for Understanding Baleen Whale Presence in the Western North Atlantic Ocean

Frontiers in Marine Science, 2021

Soundscape analyses provide an integrative approach to studying the presence and complexity of so... more Soundscape analyses provide an integrative approach to studying the presence and complexity of sounds within long-term acoustic data sets. Acoustic metrics (AMs) have been used extensively to describe terrestrial habitats but have had mixed success in the marine environment. Novel approaches are needed to be able to deal with the added noise and complexity of these underwater systems. Here we further develop a promising approach that applies AM with supervised machine learning to understanding the presence and species richness (SR) of baleen whales at two sites, on the shelf and the slope edge, in the western North Atlantic Ocean. SR at both sites was low with only rare instances of more than two species (out of six species acoustically detected at the shelf and five at the slope) vocally detected at any given time. Random forest classification models were trained on 1-min clips across both data sets. Model outputs had high accuracy (>0.85) for detecting all species’ absence in b...

Research paper thumbnail of AM 41.1 Gulf of Mexico

Research paper thumbnail of Effects of ice cover on the behavioural patterns of aquatic-mating male bearded seals

Animal Behaviour, 2004

We used vocalizations to examine the behaviour of male bearded seals, Erignathus barbatus, in rel... more We used vocalizations to examine the behaviour of male bearded seals, Erignathus barbatus, in relation to ice cover over 2 consecutive years. Site fidelity varied between males. Territorial males returned to the same display locations in successive years, whereas roaming males returned to the study area but not to specific display locations. Males did not vocalize under the land-fast-ice edge and were therefore limited by access to areas of drift ice and open water for the purpose of displaying. Between-year fluctuations in ice cover restricted the number of displaying males during the early part of the mating season in April. Decreased ice cover during May resulted in more males displaying, but the number of males declined in June irrespective of the continued decline in ice cover. Roaming males were not heard when the ice cover was more than 60%, whereas territorial males were present during all ice conditions. Yearly fluctuations in ice cover may alter the long-term mating success of individuals and predictably available display areas may be preferred to areas that are more likely to be enclosed by fast-ice in some years.

Research paper thumbnail of Individuality in harp seal, Phoca groenlandica, pup vocalizations

Research paper thumbnail of Phase 4: DCL System Using Deep Learning Approaches for Land-Based or Ship-Based Real-Time Recognition and Localization of Marine Mammals - Distributed Processing and Big Data Applications

While the animal bioacoustics community at large is collecting huge amounts of acoustic data at a... more While the animal bioacoustics community at large is collecting huge amounts of acoustic data at an unprecedented pace, processing these data is problematic. Currently in bioacoustics, there is no effective way to achieve high performance computing using commericial off the shelf (COTS) or government off the shelf (GOTS) tools. Although several advances have been made in the open source and commercial software community, these offerings either support specific applications that do not integrate well with data formats in bioacoustics or they are too general. Furthermore, complex algorithms that use deep learning strategies require special considerations, such as very large libraiers of exemplars (whale sounds) readily available for algorithm training and testing. Detection-classification for passive acoustics is a data-mining strategy and our goals are aligned with best practices that appeal to the general data mining and machine learning communities where the problem of processing la...

Research paper thumbnail of R script for Zero Inflated Model and Likelihood Ratio

Research paper thumbnail of Summary of the detector performance data

Research paper thumbnail of Summary of the analysis of the southern right whale vocalisation data

Research paper thumbnail of See Profile

Vocalizations and movements suggest alternative mating tactics in male bearded seals

Research paper thumbnail of National Passive Acoustics Workshop Developing a Strategic Program Plan for NOAA’s Passive Acoustics

National Marine Fisheries Service William T. Hogarth, Assistant Administrator for FisheriesThis d... more National Marine Fisheries Service William T. Hogarth, Assistant Administrator for FisheriesThis document will initially be issued only online. It will be available at:

Research paper thumbnail of National Oceanic and Atmospheric Adminis… 113 PUBLICATIONS 1,773 CITATIONS SEE PROFILE

Male reproductive behaviour in the aquatic mating bearded seal (Erignathus barbatus).

Research paper thumbnail of Ecology of Beaked Whales and Sperm Whales in the Western North Atlantic Ocean: Insights from Passive Acoustic Monitoring

Ecology of Beaked Whales and Sperm Whales in the Western North Atlantic Ocean: Insights from Pass... more Ecology of Beaked Whales and Sperm Whales in the Western North Atlantic Ocean: Insights from Passive Acoustic Monitoring

Research paper thumbnail of Appendices 1-3 from The importance of migratory connectivity for global ocean policy

MiCO aims to coalesce existing knowledge on connectivity and migratory corridors of marine migrat... more MiCO aims to coalesce existing knowledge on connectivity and migratory corridors of marine migratory species that utilize areas beyond national jurisdiction. Numerous definitions of "migration" and "migratory species" exist, confusing efforts to develop a specific list of species to include within MiCO. Lascelles et al. (2014) assessed the current status and management needs of over 800 migratory marine species across the same four taxonomic groups as MiCO. They define migratory species as animals that move between "at least two jurisdictions during the course of their annual cycles," with specific species selected by expert groups. The list of species to be evaluated by MiCO was assembled from several sources: including Lascelles et al. (2014) (n = 829), the CMS Migratory Shark Species (n = 94), fish species managed by Regional Fisheries Management Organization (RFMO; n = 40), seabirds of the Migratory Bird Treaty Act (n = 171), and BirdLife International (n = 280). The initial 892 species to be evaluated by the MiCO system includes 440 fish, 346 seabird, 99 marine mammal, and 7 sea turtle species. Family Scientific Name Common Name CMS Annex (as of Aug 7 2018) IUCN Red List Status

Research paper thumbnail of Supplementary material from "The importance of migratory connectivity for global ocean policy

The distributions of migratory species in the ocean span local, national and international jurisd... more The distributions of migratory species in the ocean span local, national and international jurisdictions. Across these ecologically interconnected regions, migratory marine species interact with anthropogenic stressors throughout their lives. Migratory connectivity, the geographical linking of individuals and populations throughout their migratory cycles, influences how spatial and temporal dynamics of stressors affect migratory animals and scale up to influence population abundance, distribution and species persistence. Population declines of many migratory marine species have led to calls for connectivity knowledge, especially insights from animal tracking studies, to be more systematically and synthetically incorporated into decision-making. Inclusion of migratory connectivity in the design of conservation and management measures is critical to ensure they are appropriate for the level of risk associated with various degrees of connectivity. Three mechanisms exist to incorporate migratory connectivity into international marine policy which guides conservation implementation: site-selection criteria, network design criteria and policy recommendations. Here, we review the concept of migratory connectivity and its use in international policy, and describe the Migratory Connectivity in the Ocean system, a migratory connectivity evidence-base for the ocean. We propose that without such collaboration focused on migratory connectivity, efforts to effectively conserve these critical species across jurisdictions will have limited effect.

Research paper thumbnail of Studying marine mammal social systems

Pinnipeds and cetaceans often aggregate, sometimes in their thousands, and on occasion in excepti... more Pinnipeds and cetaceans often aggregate, sometimes in their thousands, and on occasion in exceptional densities. Many of these gatherings are not obviously the result of attraction to environmental features. Thus, these are actively maintained groups, not passive aggregations. Groups imply social structure, and social structure can affect ecology, genetics, population biology, and thus issues of conservation and management (Wilson 1975; Sutherland 1998). Hence, both for interest in the social lives of marine mammals, as well as a general understanding of the biology of the animals, their places in ecosystems, and the effects of human activities on them, we need to study social structure. Additionally, because the habitat of marine mammals is so different from that of terrestrial species, an understanding of the social structure of marine mammals provides an important comparative perspective on the forces of mammalian social evolution (Connor et al. 1998). Those who study the social ...

Research paper thumbnail of Phase 3: DCL System Using Deep Learning Approaches for Land-based or Ship-based Real-Time Recognition and Localization of Marine Mammals - Bioacoustic Applicaitons

Goals of this research phase is to investigate advanced detection and classification pardims usef... more Goals of this research phase is to investigate advanced detection and classification pardims useful for data-mining passive large passive acoustic archives. Technical objectives are to develop and refine a High Performance Computing, Acoustic Data Accelerator (HPC-ADA) along with MATLAB based software based on time series acoustic signal Detection cLassification using Machine learning Algorithms, called DeLMA. Data scientists and biologists integrate to use the HPC-ADA and DeLMA technologies to explore data using newly developed techniques aimed at inspection of data extracted at large spatial and temporal scales.

Research paper thumbnail of Fourth International Conference on the Effects of Noise on Aquatic Life

Hawaiian spinner dolphins display predictable daily behavior, using shallow bays to rest during t... more Hawaiian spinner dolphins display predictable daily behavior, using shallow bays to rest during the daytime. The frequency and intensity of the interactions between humans and dolphins in these bays has prompted concern for the animals. All previous research on the dolphins’ response to human activity has been conducted visually. In contrast, here we evaluate whether dolphins acoustically respond to human activity. Previous research found that the dolphins were silent during rest and that acoustic activity matched their general behavioral state. Therefore, we expected to see higher dolphin whistle activity before and after peak resting time and low activity to indicate rest. The daily pattern of dolphin whistle activity in two of the four bays matched what we expected. However, in two bays we did not find a drop in dolphin whistle activity. With regards to the acoustic response of the animals to human activity, we found relationships between the number of vessels and dolphin acousti...

Research paper thumbnail of Phase 1: DCL System Research Using Advanced Approaches for Land-based or Ship-based Real-Time Recognition and Localization of Marine Mammals - HPC System Implementation

We aim to investigate advancing the state of the art of detection, classification and localizatio... more We aim to investigate advancing the state of the art of detection, classification and localization (DCL) in the field of bioacoustics. The two primary goals are to develop transferable technologies for detection and classification in: (1) the area of advanced algorithms, such as deep learning and other methods; and (2) advanced systems, capable of real-time and archival and processing. This project will focus on long-term, continuous datasets to provide automatic recognition, minimizing human time to annotate the signals. Effort will begin by focusing on several years of multi-channel acoustic data collected in the Stellwagen Bank National Marine Sanctuary (SBNMS) between 2006 and 2010. Our efforts will incorporate existing technologies in the bioacoustics signal processing community, advanced high performance computing (HPC) systems, and new approaches aimed at automatically detecting-classifying and measuring features for species-specific marine mammal sounds within passive acoust...

Research paper thumbnail of Phase 2: DCL System Using Deep Learning Approaches for Land-based or Ship-based Real-Time Recognition and Localization of Marine Mammals - Machine Learning Detection Algorithms

ArXiv, 2016

Overarching goals for this work aim to advance the state of the art for detection, classification... more Overarching goals for this work aim to advance the state of the art for detection, classification and localization (DCL) in the field of bioacoustics. This goal is primarily achieved by building a generic framework for detection-classification (DC) using a fast, efficient and scalable architecture, demonstrating the capabilities of this system using on a variety of low-frequency mid-frequency cetacean sounds. Two primary goals are to develop transferable technologies for detection and classification in, one: the area of advanced algorithms, such as deep learning and other methods; and two: advanced systems, capable of real-time and archival processing. For each key area, we will focus on producing publications from this work and providing tools and software to the community where/when possible. Currently massive amounts of acoustic data are being collected by various institutions, corporations and national defense agencies. The long-term goal is to provide technical capability to an...

Research paper thumbnail of NOAA and BOEM Minimum Recommendations for Use of Passive Acoustic Listening Systems in Offshore Wind Energy Development Monitoring and Mitigation Programs

Frontiers in Marine Science, 2021

Offshore wind energy development is rapidly ramping up in United States (U.S.) waters in order to... more Offshore wind energy development is rapidly ramping up in United States (U.S.) waters in order to meet renewable energy goals. With a diverse suite of endangered large whale species and a multitude of other protected marine species frequenting these same waters, understanding the potential consequences of construction and operation activities is essential to advancing responsible offshore wind development. Passive acoustic monitoring (PAM) represents a newer technology that has become one of several methods of choice for monitoring trends in the presence of species, the soundscape, mitigating risk, and evaluating potential behavioral and distributional changes resulting from offshore wind activities. Federal and State regulators, the offshore wind industry, and environmental advocates require detailed information on PAM capabilities and techniques needed to promote efficient, consistent, and meaningful data collection efforts on local and regional scales. PAM during offshore wind co...

Research paper thumbnail of Evaluating the Efficacy of Acoustic Metrics for Understanding Baleen Whale Presence in the Western North Atlantic Ocean

Frontiers in Marine Science, 2021

Soundscape analyses provide an integrative approach to studying the presence and complexity of so... more Soundscape analyses provide an integrative approach to studying the presence and complexity of sounds within long-term acoustic data sets. Acoustic metrics (AMs) have been used extensively to describe terrestrial habitats but have had mixed success in the marine environment. Novel approaches are needed to be able to deal with the added noise and complexity of these underwater systems. Here we further develop a promising approach that applies AM with supervised machine learning to understanding the presence and species richness (SR) of baleen whales at two sites, on the shelf and the slope edge, in the western North Atlantic Ocean. SR at both sites was low with only rare instances of more than two species (out of six species acoustically detected at the shelf and five at the slope) vocally detected at any given time. Random forest classification models were trained on 1-min clips across both data sets. Model outputs had high accuracy (>0.85) for detecting all species’ absence in b...