This special issue on marine mammals marks the 20th anniversary of the European Cetacean Society (ECS). The Society's 2006 conference was held in Gdynia, Poland in April, with the theme of ‘Marine mammals and man in coastal ecosystems: can they co-exist?’ Several of the papers contained in this volume describe work that was presented at this conference, and most relate to research carried out by ECS members.

This paper presents the new concept of passive acoustic tomography which allows ocean data collection with a passive acoustic remote sensing process. The originality lies in using acoustic sources of opportunity such as surface noise, radiated ship noise and marine mammal calls. Such use of passive tomography is a promising way to reduce acoustic emissions in oceans. A review is first presented, including the description of new concepts of covert active, assisted passive and autonomous tomography, followed by applications on real world data. Under the assumptions of multipath propagation and measurements performed by a sparse network of hydrophones, a time–frequency processor is proposed to simultaneously estimate the source location and the impulse response of the propagation channel for marine mammal calls used as opportunistic sources (multipath structure, time delay and attenuation are estimated). Promising results are obtained on real data coming from the Laurentian channel where wideband beluga calls (1 to 3 kHz) are measured by a sparse network of 6 bottom hydrophones.

We investigated the feasibility of using a T-POD, a passive acoustic dolphin detector system, to monitor bottlenose dolphins in the Shannon Estuary, Ireland, from 27 June to 18 August 2005. A v.3 T-POD, logging alternate minutes, was moored in view of an observation site. Land-based theodolite tracking was used to record the position of the closest animal, school size and activity of the closest dolphin school to the T-POD. All cetacean detections on the T-POD synchronous with shore watches in sea states ≤2 were analysed. A total of 94 schools were observed at distances up to 8000 m from the observer. Acoustic detections corresponded well with visual detections, with 82% of the dolphin schools observed within 500 m of the T-POD detected acoustically. The furthest distance dolphins were observed from the T-POD during periods of acoustic detection was 1246 m. Twelve acoustic encounters were logged without corresponding visual detections, four of which may represent false positives. School size did not affect the acoustic detectability of the dolphins, as there was no difference in the size of schools observed with or without corresponding acoustic detections (Kruskal–Wallis, P=0.64). Similarly no relationship was found between acoustic detections and school sizes at different distances to the T-POD (linear regression P=0.5, r2=0.01). Acoustic encounters did not vary in relation to diel patterns (Mann–Whitney, P=0.13) but were related to tidal state (χ2=40.2, P=0.00, df=11) with more encounters logged in the 4 h after high water, probably reflecting prey-related changes in habitat use. T-POD detections correlated well with visual observations and although detection is likely to vary according to T-POD specification, sensitivity and the conditions at the deployment site, the T-POD is a useful tool that has been shown to offer the possibility of continuous monitoring, something that is difficult to achieve with visual methods alone.

The vocal repertoire of many delphinid odontocetes includes narrowband tonal whistles used mainly for communication. The aim of this study was to describe the whistle repertoire of short-beaked common dolphins, Delphinus delphis, recorded in the Celtic Sea between May and August 2005. The 1835 whistles recorded were classified into six broad categories and 30 sub-types, of which simple upsweeps and downsweeps were the most common. Furthermore, the parameters duration, inflections, steps and various frequency variables were measured. The whistles covered a frequency span from 3.56 kHz to 23.51 kHz and had durations between 0.05 and 2.02 seconds. Whistle parameters varied with behavioural context, group size and between encounters. The whistle repertoire of Celtic Sea common dolphins was compared to that of D. delphis from the Western Approaches of the English Channel, recorded during a survey between January and March 2004. The relative abundances of the broad whistle types did not differ between the two locations, but most whistle parameters were significantly different: almost all frequency variables measured were significantly higher in English Channel whistles. This may indicate some degree of population structuring of short-beaked common dolphins around Britain. Alternatively, the common dolphins in the English Channel may have shifted the frequencies of their vocalizations up to avoid masking by low-frequency ambient noise produced by high levels of vessel traffic in this area.

We tried to find discriminating features for sperm whale clicks in order to distinguish between clicks from different whales, or to enable unique identification. We examined two different methods to obtain suitable characteristics. First, a model based on the Gabor function was used to describe the dominant frequencies in a click, and then the model parameters were used as classification features. The Gabor function model was selected because it has been used to model dolphin sonar pulses with great precision. Additionally, it has the interesting property that it has an optimal time–frequency resolution. As such, it can indicate optimal usage of the sonar by sperm whales. Second, the clicks were expressed in a wavelet packet table, from which subsequently a local discriminant basis was created. A wavelet packet basis has the advantage that it offers a highly redundant number of coefficients, which allow signals to be represented in many different ways. From the redundant signal description a representation can be selected that emphasizes the differences between classes. This local discriminant basis is more flexible than the Gabor function, which can make it more suitable for classification, but it is also more complex. Class vectors were created with both models and classification was based on the distance of a click to these vectors. We show that the Gabor function could not model the sperm whale clicks very well, due to the variability of the changing click characteristics. Best performance was reached when three subsequent clicks were averaged to smoothen the variability. Around 70% of the clicks classified correctly in both the training and validation sets. The wavelet packet table adapted better to the changing characteristics, and gave better classification. Here, also using a 3-click moving average, around 95% of the training sets classified correctly and 78% of the validation sets. These numbers lowered by only a few per cent when single clicks, instead of a moving average, were classified. This indicates that, while the features may show too much variability to enable unique identification of individual whales on a click by click basis, the wavelet approach may be capable of distinguishing between a small group of whales.

Recordings of a group of foraging sperm whales usually result in a mixture of clicks from different animals. To analyse the click sequences of individual whales these clicks need to be separated, and for this an automatic classifier would be preferred. Here we study the use of a radial basis function network to perform the separation. The neural network's ability to discriminate between different whales was tested with six data sets of individually diving males. The data consisted of five shorter click trains and one complete dive which was especially important to evaluate the capacity of the network to generalize. The network was trained with characteristics extracted from the six click series with the help of a wavelet packet-based local discriminant basis. The selected features were separated in a training set containing 50 clicks of each data set and a validation set with the remaining clicks. After the network was trained it could correctly classify around 90% of the short click series, while for the entire dive this percentage was around 78%.

Sperm whales (Physeter macrocephalus) typically produce sharp onset, broadband pulse sounds at varying repetition rates. Acoustic recordings of different social units of sperm whales in the Mediterranean Sea included apparent non-click sounds of tonal quality, termed ‘squeals’. Quantitative analysis of the spectral signal and waveform indicate that although squeals are perceived as tonal and appear spectrally as narrowband frequency-modulated structures with harmonics, they actually consist of pulses at high repetition rates exceeding 1600 clicks/s. Squeals contained energy at between 400 Hz and 22 kHz, with mean peak energy at the relatively low frequency of 700 Hz. Five spectral forms of squeal could be recognized, with the dominant form (45%) of squeals showing a decrease in frequency along the squeal contour. Mean click repetition rate ranged between 713 and 1385 clicks/s for individual squeals, and also varied within squeals at rates of between 64 and 444 clicks/s. Variation in click repetition rate was reflected in the frequency spacing of the spectral sidebands, in a statistically significant inverse relationship. Squeals were recorded only during bouts of sperm whale social behaviour, consistent with their having a communicative social function. Sperm whale squeals are structurally and audibly similar to the burst-pulse sounds produced by many smaller odontocete species, and might fall on the continuum between distinct click trains and pure-tone whistles.

Cetaceans are prone to collisions with fast vessels, and in areas of high cetacean and vessel density such as in the Canary Islands, the sperm whale (Physeter macrocephalus) is of great concern. Sperm whales are highly vocal and can be localized with passive sonar, but, when at or near the surface, they tend to stop vocalizing, i.e. when they are most at risk. Regrettably, ship-borne active solutions have proven inefficient due to the short detection range and the ships' high-speeds. Our objective in this paper is to evaluate the efficiency of an original passive sonar solution that would use vocalizing whale clicks at depth as acoustic sources to detect silent whales. This solution could be a non-invasive complementary component of a more complex passive localization whale anti-collision system. To meet this aim, a simulation tool for 3D acoustic propagation was designed in which a wideband Nx2D ray solution of the wave equation simulates a passive solution consisting of an arbitrary number of active acoustic sources, an illuminated object, and a receiver, all positioned in a three-dimensional space with arbitrary bathymetry. Both curved and straight ray solutions were implemented, the latter providing greater computational speeds at the expense of temporal and angular precision. The software recreates the resulting sound mixture of direct, reverberated and target back-scattered signals arriving at the array sensors for any array configuration, any number of sources and one target. Simulations show the application of the concept for the Canary Islands, with a detection range upper bound of the order of one kilometre.

The sonar capabilities of the sperm whale, Physeter macrocephalus, have been the subject of speculation for a long time. While the usual clicks of this species are considered to support mid-range echolocation, no physical characteristics of the signal have clearly confirmed this assumption nor have they explained how sperm whales forage on squid. The recent data on sperm whale on-axis recordings have allowed us to simulate the propagation of a 15 kHz pulse as well as its received echoes from different targets, taking into account the reflections from the bottom and the sea surface. The analysis was performed in a controlled environment where the oceanographic parameters and the acoustic background could be modified. We also conducted experimental measurements of cephalopod target strength (TS) (Loligo vulgaris and Sepia officinalis) to further investigate and confirm the TS predictions from the geometric scattering equations. Based on the results of the computer simulations and the TS experimental measurements (TS squid=−36.3±2.5 dB), we were able to determine the minimum requirements for sperm whale sonar, i.e. range and directional hearing, to locate a single 24.5 cm long squid, considered to be (from stomach contents) the major size component of the sperm whale diet. Here, we present the development of the analysis which confirms that sperm whale usual clicks are appropriate to serve a mid-range sonar function, allowing this species to forage on individual organisms with low sound-reflectivity at ranges of several hundreds of metres.

Describing the auditory periphery of odontocetes is a key conservation issue to further assess the effect of acoustic pollution. Because all odontocetes produce species-specific frequency ranges, differences in echolocation signals should reflect anatomical differences in the auditory pathways. Here, we studied the ears of 15 odontocete species through 3D reconstructions from computerized tomography scans to extract standard measurements (bullae lengths/volumes and cochlear volume) and investigate the discriminatory weight of each of these variables as well as their relation to the species' hearing specificity. Any of the measurements appeared to be a good indicator of the species and could therefore be used to classify them. All the ear lengths and most volumes were strongly linearly correlated (r >0.9) in all species and the proportion between the tympanic and periotic bones appeared to remain constant. This constant ratio could be an indication of a functional relationship between both structures, and might suggest an active role of the odontocete middle ear during target acoustic detection, providing new information on the odontocete sound reception mechanism. Our results are generally consistent with previous studies, although here the coefficients of correlation between animal lengths and the total volume and lengths of the bullae were lower (0.77< r <0.86), indicating that the length of the animals may not be a primary parameter to take into account when defining ear measurements. These results suggest that the measurements described characterize standard ears which could be used as a morphological basis for further species-specific acoustic comparison.

Deuterium oxide was used to measure isotope dilution space 11 times in seven free living, adult, male Atlantic walruses (Odobenus rosmarus rosmarus) (930–1597 kg) in north-east Greenland in the month of August 2000 and August 2001. Telemetry instruments, which were attached to the animals as part of a long term study, provided data on activity levels. Equilibration of the intravenously injected isotope was complete after 2–3 h. The application of a general correction factor developed for all pinniped species allowed the estimation of total body water content as 56.8% (range: 36.6–73.3%) of total body mass (TBM). Water turnover averaged 44.5 g kg-1 d-1. For one animal it was possible to estimate water influx to 4.8 g kg-1 d-1 and water efflux to 21.8 g kg-1 d-1. The estimates of body fat and body protein averaged respectively 24.7% and 18.4% of TBM. Similarly, for an ‘average’ walrus with a TBM of 1300 kg body ash was calculated to average 26 kg and body gross energy to 18,300 MJ. The assessment of body fat by isotope dilution did not differ significantly from the estimates of blubber content obtained from the dissection of walruses in the wild.

The minke whale (Balaenoptera acutorostrata) is a widespread, opportunistic species showing spatial and seasonal variations in diet according to local availability of prey. Although previous research has been conducted on the foraging strategies of this small rorqual whale, in terms of prey aggregation and assimilation, relatively little has been published on the foraging association of this species with coastal seabirds. Over the past five years, minke whales occurring along the outer coastline of the southern Moray Firth in north-east Scotland during the summer and autumn months have been recorded foraging in the presence of seabirds, such as kittiwakes (Rissa tridactyla), herring gulls (Larus argentatus), guillemots (Uria aalge) and shearwaters, which form dense feeding rafts at the water's surface. The formation of bird rafts notably occurs independently of the presence of B. acutorostrata, believed instead to be the successive result of prey concentrated at the surface by predatory schooling fish from below rather than by activities of the whales themselves. In this area of the North Sea, schooling mackerel (Scomber scombrus) constitute the most significant component of the summer fish biomass and are believed to perform the role of compacting targeted sandeel (Ammodytes spp.) prey into concentrated bait balls almost exclusively. The resulting ball of prey is consequently available to the foraging whale, which can be seen opportunistically utilizing successive bird rafts rather than expending unnecessary energy corralling the Ammodytes prey by traditional, active entrapment methods. The role of the mackerel in increasing both the rate and density of sandeel ball formation (as indicated by the presence and activity of associated bird rafts) is therefore thought to be very significant in this inshore Scottish location. Changes in oceanographic variables such as water temperature have been directly correlated with the migration of S. scombrus. The observed inter-annual variability in B. acutorostrata distribution in the outer Moray Firth may subsequently be related to the respective distribution and abundance of these migratory, pelagic fish species.

To characterize temporal patterns of humpback group structure around Abrolhos Archipelago, Brazil, seven years (1998–2004) of data obtained from July through November were analysed. During one-hour scans, observers determined group composition within 9.3 km around a land-based station. A total of 930 scans, comprising 4288 groups were analysed. Seven group categories were identified and their frequencies were 14.6% of 1AD (lone whale), 25.2% of DYAD (two adult whales), 7.3% of TRIO (three adult whales), 5.2 of TRIO+ (more than three adults), 24.9% of MOC (mother and calf), 19.6% of MOCE (mother, calf and one escort), and 3.1% of MOCE+ (mother, calf and more than one escort). Proportions of whale-group categories did not change between morning and afternoon, nor among years. However, as the season progressed, groups with calves increased while groups without calves decreased. This progression may be explained by the segregated migration patterns of humpback whales and also by changes in the social status of individuals. The Abrolhos Reef provides protection from the prevailing winds, which may explain the higher proportion of groups with calves in the area.

This paper presents baseline information on maternal behaviour and lactation in the Mediterranean monk seal, with particular focus on the age at which pups are weaned. The study was conducted in the western Saharan population, the only surviving colony of the species. The first moult finished at a mean pup age of 72.3±17 d (N=17) and, in contrast to other taxonomically-related phocids, this process was not associated with weaning. Lactation lasted a mean of 119.4 d (N=9; range: 103–149 d) in the pups that could be monitored until full weaning had taken place. This period almost doubles the maximum lactation length reported in other phocid species. During the first week after birth the mother–pup bond was well developed and mothers always remained with their pups. The time invested in nursing (17%, SD: ±36) and in mother–pup interactions (14%, SD: ±32) was higher during this period than afterwards (8%, SD: ±23 and 4%, SD: ±19, respectively). After the first week, nursing continued but mothers started to leave their pups in order to feed at sea. Weaning occurred gradually. Already since birth, pups were active and mobile, and swam frequently before moulting or weaning occurred. Fostering and milk stealing were common patterns of behaviour for both lactating females and pups. In 26.6% of the suckling episodes observed in mother–pup pairs of known identity, pups suckled from females other than their mothers. Some females nursed more than one pup, at least occasionally, and in some cases a pup was fostered long-term by an alien female. The Mediterranean monk seal exhibits maternal-care characteristics that are more like otarids than phocids. This observation contradicts previous proposals that a short lactation period is a phylogenetic characteristic of phocids. Several of the unusual maternal traits observed may be favoured by year-round access to abundant food supply, availability of breeding sites, and mild climatic conditions. This information should be taken into account when designing conservation strategies for the species and, very particularly, in the implementation of pup rehabilitation programmes.

Several types of relationships link organisms to each other, including competition, predation and various types of associations. This paper presents the first case of association/interaction between dugongs, Dugong dugon, the only strictly marine herbivorous mammal, and three species of tropical dolphins in and around the lagoon of Mayotte (45°10'E 12°50'S), in the western tropical Indian ocean. Data were collected opportunistically from 1999 to 2005 in the surrounding waters of Mayotte from boat (N=2 observations) and ULM (ultra-light motorized vessel, N=2). The dolphin species involved in associations with dugongs were Indo-Pacific bottlenose dolphin, Tursips aduncus, Indo-Pacific humpback dolphin, Sousa chinensis, and spinner dolphin, Stenella longirostris. In one case, dugongs were associated with both bottlenose and humpback dolphins. Associations were observed in the protected waters of the lagoon and outside, along the external slope of the barrier reef. Group size, activity and group structure of each species were recorded. Behavioural observations suggest that dugongs and dolphins were engaged in similar activities, such as travelling, on several occasions and were clearly associated when group formation was tight. If dolphins and dugongs may not associate for feeding purposes, then these interactions may occur: (1) for predation avoidance toward sharks; or (2) without any ecological reasons due to dolphin and dugong habitat overlap around the island.

Between 1993 and 2005, 68 sightings of southern right whales were recorded along 400 km of coastline between Paraty Bay (23°13'S 44°42'W) and Macaé (22°22'S 41°47'W), Rio de Janeiro State, south-eastern Brazilian coast. Mother/calf pairs represented 66.1% of sightings. Only solitary individuals showed a distinct pattern of occurrence. Females with calf did not show a distinct pattern of occurrence suggesting their widespread distribution in the area. Analysis of the records and detailed accounts of the sightings of southern right whales reveal that there exist conflicts between the animals and human activities such as harassment and collision with boats during attempted rescues, accidental entanglement in fishing nets and intentional mortality, which may have a bearing in the conservation of the species in Brazilian waters.

Hawaiian spinner dolphins (Stenella longirostris) feed offshore at night on mesopelagic prey and move to protected shallow waters in early morning to rest. We hypothesized that this behaviour would make them particularly sensitive to anthropogenic factors in these rest areas and would affect their overall behaviour and their population frequency. To test our hypothesis we collected data in a known rest area along the north-west coast of the island of Oahu during August of 2001, 2002 and 2003. Using land- and water-based surveys, we evaluated dolphin group size, counted boats, kayaks and swimmers in the vicinity of the dolphins (<40 m) and reported any changes in dolphin behaviour. Our results demonstrated a stable dolphin habitat frequency over the three summers, which emphasizes the critical value of this rest area for this spinner dolphin population. The results also showed an increase in dolphin-oriented activities, but we could not draw any definitive conclusion on their real impacts on spinner dolphin ecology and ethology.

On the north-eastern coast of Sardinia, from November 2004 to January 2006, the first attempt in the Mediterranean basin to obtain information on encounter rate, group size and incidental capture of bottlenose dolphins in a marine fish farm was assessed, combining direct observations from fish farm boats with photo-identification studies. During 15 months of research, 79 d (65.3% of the total monitored days) were spent in direct observation of 146 groups of bottlenose dolphins around the fish farm cages. There was a peak in bottlenose dolphin presence during winter. Photographs were taken in the fish farm area during 79 encounters on 61 different days (totalling 34 marked individuals). The regular occurrence of some dolphins suggests individual preferences for the fish farm area. The incidental bottlenose dolphin capture observed in large, loose predator nets (1 dolphin per month) is cause for concern, as it is questionable whether or not the bottlenose dolphins in the area can sustain incidental capture of this magnitude. The information gained from this study shows the necessity for further regulations to be established, both in the use of predator nets and management of marine fish farms.

Fine-scale habitat selection of Chilean dolphins was studied between January and April 2002 through shore-based theodolite tracking in order to investigate the environmental and behavioural determinants of habitat use, and to evaluate the interactions between this species and aquaculture activities in Yaldad Bay, southern Chile. During 293.5 h of effort, movement and habitat selection patterns of dolphins exhibited a significantly concentrated use of only 21% of the entire study area. Correspondence analysis showed that shallow waters (5–10 m), proximity to coast and rivers were the most significant environmental parameters determining fine-scale dolphin distribution patterns, with foraging the most frequently observed activity. Aquaculture activities in the area were observed to affect dolphin habitat use patterns by restricting space available for biologically important dolphin behaviours.

Trials were carried out in the Shannon estuary, Ireland, to test the effects of continuous (CPs) and responsive pingers (RPs) on bottlenose dolphin behaviour. In controlled trials, active and control pingers were deployed on fixed moorings, with T-PODs—acoustic monitoring devices to detect cetacean activity. In a separate trial, pingers were deployed from a moving boat which actively located dolphin groups in the estuary, and dolphin behaviour was recorded. In the static trials, overall detection rates of dolphin vocalizations on the T-POD were significantly lower in the presence of active CPs, but this was not the case for RPs. Mean inter-click interval values were longer for click trains produced in the presence of inactive RPs than for active RPs, active or inactive CPs. In boat-based trials, both active CPs and RPs appeared to affect bottlenose dolphin behaviour, whereby dolphins immediately left the area at speed and in a highly directional manner, involving frequent leaps.