Description: Passive acoustic data provide a prime source of information on marine mammal distribution and behaviour. Particularly in the Southern Ocean, where ship-based data collection can be severely hampered by weather and ice conditions, passive acoustic monitoring (PAM) of marine mammals forms an important source of year-round information on acoustic presence. Array data can be used to obtain directional information on the species present in the recordings to derive movement patterns. Acoustic arrays furthermore allow spatial comparisons of marine mammal distribution patterns and habitat affinities when the acoustic presence information is linked to local environmental parameters. Here we present two passive acoustic monitoring arrays that have been implemented by the Alfred Wegener Institute’s Ocean Acoustic Lab and serve the investigation of marine mammals on different spatial scales. During the austral summer season 2012/2013 a local scale array of sea ice-based time-synchronized passive acoustic recorders was deployed in Atka Bay, Antarctica. The PASATA (PASsive Acoustic Tracking of Antarctic marine mammals) project investigates coastal local habitat usage and communication ranges of marine mammals by integrating positional information from triangulation of calling animals and information from environmental parameters. For studies on marine mammals over larger spatial scales, 23 passive acoustic recorders were deployed in oceanographic moorings in the Southern Ocean, reaching from the Greenwich meridian throughout the Weddell Sea to the Western Antarctic Peninsula. The inter-disciplinary nature of this mooring array allows combining in-situ oceanographic measurements with passive acoustic data on marine mammal occurrence. It furthermore forms the first basin-wide, long term array, at least in the Southern Ocean. Here, we describe both arrays, the recorder types used, and technical and logistic requirements for PAM in a polar environment.

Description: Passive acoustic data provide a prime source of information on marine mammal distribution and behaviour. Particularly in the Southern Ocean, where ship-based data collection can be severely hampered by weather and ice conditions, passive acoustic monitoring (PAM) of marine mammals forms an important source of year-round information on acoustic presence. Array data can be used to obtain directional information on the species present in the recordings to derive movement patterns. Acoustic arrays furthermore allow spatial comparisons of marine mammal distribution patterns and habitat affinities when the acoustic presence information is linked to local environmental parameters. Here we present two passive acoustic monitoring arrays that have been implemented by the Alfred Wegener Institute’s Ocean Acoustic Lab and serve the investigation of marine mammals on different spatial scales. During the austral summer season 2012/2013 a local scale array of sea ice-based time-synchronized passive acoustic recorders was deployed in Atka Bay, Antarctica. The PASATA (PASsive Acoustic Tracking of Antarctic marine mammals) project investigates coastal local habitat usage and communication ranges of marine mammals by integrating positional information from triangulation of calling animals and information from environmental parameters. For studies on marine mammals over larger spatial scales, 23 passive acoustic recorders were deployed in oceanographic moorings in the Southern Ocean, reaching from the Greenwich meridian throughout the Weddell Sea to the Western Antarctic Peninsula. The inter-disciplinary nature of this mooring array allows combining in-situ oceanographic measurements with passive acoustic data on marine mammal occurrence. It furthermore forms the first basin-wide, long term array, at least in the Southern Ocean. Here, we describe both arrays, the recorder types used, and technical and logistic requirements for PAM in a polar environment.

Description: Water is an excellent medium for sound transmission and many aquatic animals actively produce sound for communication, orientation or foraging purposes. The underwater environment is filled with abiotic and biotic sounds of variable intensity over a wide range of frequencies and showing high levels of variation in space and time. Evidence is increasing that the passive use of such environmental sounds also provides an important information source for aquatic organisms to orient themselves and obtain information about the environment. All fishes can detect and process acoustic particle motion, including species that do not produce sound, indicating that sound is likely to be of critical importance in the lives of many fish species. Over the last century, human activities in and near the water have increasingly added artificial sounds to the underwater environment. To date, information on the behavioral and ecological implications of increasing underwater sound levels for fish is generally lacking. However, comparative evidence from other vertebrate species suggests that impeding the ability of fish to hear biologically relevant sounds can interfere with critical functions such as acoustic communication, predator avoidance and use of information from underwater soundscapes. Sounds produced by offshore windpower plants comprise loud impulse sounds (construction) and more moderate underwater noises of longer duration (operation and maintenance), which differ substantially in the type of impact and the spatio-temporal scale over which effects may occur. Short impact sounds, such as from pile driving can have dramatic effects on nearby fish, resulting in physical damage and death. However, long-term moderate anthropogenic noise exposure may have a greater impact on fish behaviour and ecology, potentially affecting whole ecosystems. Before noise-exposure criteria and mitigation measures can be established for fish, experimental studies in the field and in the laboratory are needed to investigate if and how fish are impacted behaviorally and physiologically by chronic and impact noise. Comparative evidence from terrestrial studies may provide guidance in experimental design and in taking a more holistic approach to develop a better understanding of the ecological impact of anthropogenic noise.

Description: An understanding of marine mammal distribution patterns forms the basis of the design and implementation of effective management measures. Habitat modeling offers a valuable approach to combine information on species presence (or absence) with local environmental parameters to explore species-specific habitat affinities. Most habitat modeling approaches require marine mammal presence-absence data which can only be obtained during dedicated visual surveys. However, in the Southern Ocean, the collection of visual data is complicated by the region’s remoteness, limited seasonal accessibility and the dependency on favorable light and weather conditions to conduct visual observations. Passive acoustic monitoring, by contrast, is highly suitable for long-term monitoring of marine mammals as they use sound in many behavioural contexts and species can be readily identified by their acoustic signatures. Passive acoustic data provide accurate information on temporal patterns in acoustic presence and time spent in the vicinity of the recorders. Furthermore, knowledge on the behavioral context in which specific sound types are produced can be used to derive information on habitat usage. Here we describe an approach for combining multi-year, year-round marine mammal presence data from passive acoustic recorders with a selected set of relevant environmental parameters to develop species-specific habitat models. Our project comprises multi-year passive acoustic data collected in Antarctic coastal as well as offshore areas throughout the Weddell Sea. Some of the species recorded are sighted only rarely during visual surveys, but are acoustically abundant in our recordings, such as the Antarctic blue whale (Balaenoptera musculus intermedia), humpback whale (Megaptera novaeangliae), fin whale (B. physalus), leopard seal (Hydrurga leptonyx), crabeater seal (Lobodon carcinophaga) and Ross seal (Ommatophoca rossii). The model will incorporate both static environmental variables, such as depth or slope, and dynamic variables, such as sea surface temperature, sea surface height, sea ice concentration and their derivatives. The project aims at furthering our current understanding of marine mammal habitat affinities in the Southern Ocean by constructing species-specific habitat models at yet unprecedented spatial and temporal time scales.

Description: The Southern Ocean provides an important habitat for marine mammals, both residential and migratory, yet long term studies of their habitat usage are hampered by the region’s seasonal inaccessibility. To overcome this problem, two autonomous underwater passive acoustic recorders were deployed in the Weddell Sea in 2008 to collect multiyear passive acoustic data. The recorders were retrieved in 2010 and the acoustic recordings were analyzed in terms of broad- and narrow-band noise. Noise in this context is defined as the acoustic energy not assignable to a specific singular source. It comprises both biotic as well as abiotic components. Noise levels were determined by selecting the quietest 10 s of each 5 min recording to exclude energetic contributions from nearby singular acoustic sources. The respective sound pressure levels (SPL) and spectra were correlated with time series of environmental covariates. The ambient noise levels of both recorders were found to be highly variable in time, ranging from 102 to 115 dB re 1 μPa (broadband SPL 5th and 95th percentile), and were correlated with the sea ice cover and wind speed. The annual variation of the ice cover caused a bimodal distribution of broadband SPL. In winter the SPL mode was 106 dB re 1 μPa. By contrast, storms over the open ocean in summer resulted in an SPL mode of 111 dB dB re 1 μPa. Variation in the ambient noise spectra could be correlated to wind speed and ice coverage. The acoustic presence of several mysticete (Antarctic blue whale, Balaenoptera musculus intermedia, fin whale, Balaenoptera physalus) and pinniped (leopard seal, Hydrurga leptonyx, crabeater seal, Lobodon carcinophaga) species created distinct bands in the spectra that contributed considerably to ambient noise levels. Comparison of the timing of these noise bands between the two acoustic data sets revealed offsets in the occurrence of acoustic activity between both recorders, suggestive of marine mammal latitudinal migration. At 66°S (the northern recorder position) fin whales were acoustically present earlier and longer in summer than at 69°S. Similarly, the blue whale chorus was more intense at 66°S than at 69°S. This might be related to the response of these species to the seasonal variation in the extension and density of sea ice. Seasonal cycles were also detected in the noise band attributed to crabeater seal vocalisations. They were annually present in September and November, followed by the leopard seals noise band, which is discernible between December and January. Results from this latitudinal recorder pair give a first impression on possible marine mammal migration patterns as well as the spatial and temporal distribution of marine mammal acoustic presence in the Southern Ocean. Additional recorders deployed in the basin wide HAFOS array will expand the spatial and temporal resolution of the acoustic dataset and allow conducting detailed multiyear studies of marine mammal acoustic presence and behavior throughout the Weddell Sea.

Description: Killer whales (Orcinus orca) are highly social top predators distributed throughout the worldʼs oceans. They are divided into different ecotypes according to foraging specializations, phenotype, and social organization. For Northern Hemisphere killer whale ecotypes, acoustic behaviour has been shown to relate to foraging strategies and social organization. In contrast to the intensively studied Northern Hemisphere ecotypes, distribution patterns, social structures, and acoustic behaviour of the Southern Hemisphere killer whale ecotypes are poorly known. One of the Southern Hemisphere ecotypes, the Antarctic Ecotype C killer whale, is known to occur in regions with dense pack ice. The limited accessibility of these areas make passive acoustic monitoring (PAM) methods a very effective investigation tool to derive information on ecotype-specific abundance and distribution. During 2 d in February 2013, it was possible to collect concurrent visual and acoustic information of Ecotype C killer whales off the Antarctic continent. From these events, a call type catalogue was compiled. The 2,238 examined calls were subjectively classified into 26 discrete call types. Ten percent of the examined calls were re-classified by two additional independent observers to examine robustness of the classification. Mean classification accordance among observers was 68%. Most call types were composed of more than one call part. Sixty-five percent of all call types were monophonic, and 35% were biphonic. Almost two-third of all call types started with a short, broadband pulse. The variability within call types was relatively high. The Ecotype C vocal repertoire contained typical acoustic features such as biphonation, high call complexity, and generally high variability in frequency modulation. For future studies, the distinct characteristics of some of the call types described herein could potentially serve as acoustic markers for PAM-based differentiation of killer whale ecotypes in the Southern Ocean.

Description: Evaluation of the performance of computer-based algorithms to automatically detect mammalian vocalizations often relies on comparisons between detector outputs and a reference data set, generally obtained by manual annotation of acoustic recordings. To explore the reproducibility of these annotations, inter- and intra-analyst variability in manually annotated Antarctic blue whale (ABW) Z-calls are investigated by two analysts in acoustic data from two ocean basins representing different scenarios in terms of call abundance and background noise. Manual annotations exhibit strong inter- and intra-analyst variability, with less than 50% agreement between analysts. This variability is mainly caused by the difficulty of reliably and reproducibly distinguishing single calls in an ABW chorus made of overlaying distant calls. Furthermore, the performance of two automated detectors, based on spectrogram correlation or subspace-detection strategy, is evaluated by comparing detector output to a “conservative” manually annotated reference data set, which comprises only analysts' matching events. This study highlights the need for a standardized approach for human annotations and automatic detections, including a quantitative description of their performance, to improve the comparability of acoustic data, which is particularly relevant in the context of collaborative approaches in collecting and analyzing large passive acoustic data sets.

Description: The eastern Atlantic Ocean is considered to provide important breeding and wintering habitats for several migratory cetacean species. The spatio-temporal distributions and migratory behaviors of cetaceans off southern Africa are nevertheless still poorly understood. This study investigated the temporal patterns of acoustic occurrence of baleen whales in a presumed baleen whale breeding area off Namibia using passive acoustic recordings collected between November 2011 and May 2013. Our results show seasonal acoustic presence of humpback whales Megaptera novaeangliae, fin whales Balaenoptera physalus and Antarctic minke whales B. bonaerensis from November to January and from June to August. Their acoustic absence from February to May possibly indicates that most animals migrated to other areas (presumably in higher latitudes) in austral summer to feed. By contrast, Antarctic blue whales B. musculus intermedia were acoustically present throughout the recording period, indicating that part of the population remains at lower latitudes year-round. Our findings support the presumed ecological importance of the oceanic area off Namibia, providing (part of) a suitable cetacean wintering and, possibly, breeding range or migratory corridor. Furthermore, the occurrence of Antarctic blue and minke whales off Namibia, concurrent with their reported acoustic presence in high-latitude feeding areas, adds to growing evidence that baleen whale migration is not obligate but much more dynamic than has long been assumed.

Description: Since 2001, hundreds of thousands of hours of underwater acoustic recordings have been made throughout the Southern Ocean south of 60° S. Detailed analysis of the occurrence of marine mammal sounds in these circumpolar recordings could provide novel insights into their ecology, but manual inspection of the entirety of all recordings would be prohibitively time consuming and expensive. Automated signal processing methods have now developed to the point that they can be applied to these data in a cost-effective manner. However training and evaluating the efficacy of these automated signal processing methods still requires a representative annotated library of sounds to identify the true presence and absence of different sound types. This work presents such a library of annotated recordings for the purpose of training and evaluating automated detectors of Antarctic blue and fin whale calls. Creation of the library has focused on the annotation of a representative sample of recordings to ensure that automated algorithms can be developed and tested across a broad range of instruments, locations, environmental conditions, and years. To demonstrate the utility of the library, we characterise the performance of two automated detection algorithms that have been commonly used to detect stereotyped calls of blue and fin whales. The availability of this library will facilitate development of improved detectors for the acoustic presence of Southern Ocean blue and fin whales. It can also be expanded upon to facilitate standardization of subsequent analysis of spatiotemporal trends in call-density of these circumpolar species.