Spatial and Temporal Variations in the Occurrence and Foraging Activity of Coastal Dolphins in Menai Bay, Zanzibar, Tanzania

Validation of the F-POD-A fully automated cetacean monitoring system

The F-POD, an echolocation-click logging device, is commonly used for passive acoustic monitoring of cetaceans. This paper presents the first assessment of the error-rate of fully automated analysis by this system, a description of the F-POD hardware, and a description of the KERNO-F v1.0 classifier which identifies click trains. Since 2020, twenty F-POD loggers have been used in the BlackCeTrends project by research teams from Bulgaria, Georgia, Romania, Türkiye, and Ukraine with the aim of investigating trends of relative abundance in populations of cetaceans of the Black Sea. Acoustic data from this project analysed here comprises 9 billion raw data clicks in total, of which 297 million were classified by KERNO-F as Narrow Band High Frequency (NBHF) clicks (harbour porpoise clicks) and 91 million as dolphin clicks. Such data volumes require a reliable automated system of analysis, which we describe. A total of 16,805 Detection Positive Minutes (DPM) were individually inspected and assessed by a visual check of click train characteristics in each DPM. To assess the overall error rate in each species group we investigated 2,000 DPM classified as having NBHF clicks and 2,000 DPM classified as having dolphin clicks. The fraction of NBHF DPM containing misclassified NBHF trains was less than 0.1% and for dolphins the corresponding error-rate was 0.97%. For both species groups (harbour porpoises and dolphins), these error-rates are acceptable for further study of cetaceans in the Black Sea using the automated classification without further editing of the data. The main sources of errors were 0.17% of boat sonar DPMs misclassified as harbour porpoises, and 0.14% of harbour porpoise DPMs misclassified as dolphins. The potential to estimate the rate at which these sources generate errors makes possible a new predictive approach to overall error estimation.

Multi-year passive acoustic monitoring of coastal dolphins along the Israeli Mediterranean shallow shelf reveals the impact of marine fish farms and trawling patterns on their habitat utilization

Dolphin preference and usage of various habitats along the Israeli shallow coastal shelf were investigated between 2019 and 2021 with passive acoustic monitoring devices. A hurdle model was used to examine the dolphins' visiting probability (chance of detection) and visit duration (length of stay once detected) across habitats, with diel cycle and season as explanatory variables. The influence of spatiotemporal prohibitions placed on trawler activity was also examined. It was found that dolphins exhibited higher presence in the vicinity of fish farms, up to three orders of magnitude, and even more so during periods when trawler activity was halted. The study also found a higher presence during the winter season and nighttime. Modeling did not find significant differences in the visiting probability or the visit duration between any non-farm-associated sites, including areas where trawling is prohibited. Further restrictions on the fishing industry may induce recovery of the benthic ecosystem and lower competition for resources, thus promoting higher dolphin presence in natural habitats along the shelf.

Seasonal and diel influences on bottlenose dolphin acoustic detection determined by whistles in a coastal lagoon in the southwestern Gulf of California

Marine mammals in subtropical coastal habitats are sentinels of the health of the ecosystem and offer important ecosystem services. They rely on prey that pursues feeding opportunities, while both avoid unfavorable conditions. In many cases, these predator-prey dynamics fluctuate seasonally and are regulated by lunar, tidal, and/or diel cycles (hour). However, these rhythmical patterns may vary under different seasonal conditions. Bottlenose dolphins (Tursiops truncatus) in the Ensenada de La Paz in Baja California Sur, Mexico, were detected acoustically over the course of an annual cycle on 21 separate occasions, covering 640 h from June 2017 to May 2019. The presence of bottlenose dolphins was examined using Generalized Additive Models (GAM) including variables that are related directly to their habitat (direct variables: hour, distance, depth) and to their prey (indirect variables: SST, moon phase and tides). Seasonal differences in the presence of bottlenose dolphins were influenced more by indirect variables (explained deviance: 34.8% vs. 37.7%). Hourly acoustic detections occurred less frequently when SST exceeded 27.4 °C (Aug-End of Nov.) and more frequently at moderate temperatures (22.7 °C to 26.3 °C) in May through July. Moreover, bottlenose dolphins were detected more frequently during waning and new moon phases, at the onset of flood and ebb tides, and during day (04:00 to 20:00). The seasonal differences in acoustic detections rates were highlighted by the global GAM and hierarchical clustering. The strong seasonal pattern indicated possible interactions with rhythmic pattern of bottlenose dolphins. Four candidate variables (SST, moon, tide, and hour) were tested for plausible interaction terms additional to their individual consideration, out of which only hour changed significantly between seasons. The patterns of presence likely increase feeding opportunities or may favor other behaviors such as socializing, resting, or nursing. These might prove responsible for the distinct occurrence and hourly patterns of bottlenose dolphins.

Characterization and comparison of echolocation clicks of white-beaked dolphins (Lagenorhynchus albirostris) off the Northumberland coast, UK

Odontocetes produce ultrasonic clicks for navigation and foraging. These are commonly categorized as regular or buzz clicks based on the inter-click interval. Buzz clicks are linked to foraging behaviors and may be subdivided into slow buzz clicks for prey chase, and regular buzz clicks for prey capture. This study recorded these three click types produced by white-beaked dolphins (Lagenorhynchus albirostris) off the Northumberland coast, UK. Acoustic parameters (including duration, centroid frequency, and root-mean-squared bandwidth) were calculated and compared across the three click types. The results showed that the regular clicks had shorter durations and higher frequencies than both the buzz click types. The regular buzz clicks had longer durations, lower frequencies, and narrower bandwidths than the slow buzz clicks. Additionally, regardless of click type, about 30% of the clicks had high-frequency (200-250 kHz) secondary peaks and >90% of the clicks displayed spectral peak and notch patterns between 20 and 80 kHz. These findings are useful for future quantitative assessment of the echolocation performance of white-beaked dolphins in the wild. The patterns of spectral peaks and notches identified may facilitate for acoustic identification of this species.

Distribution of the acoustic occurrence of dolphins during the summers 2011 to 2015 in the Upper Gulf of California, Mexico

Baseline knowledge of spatial and temporal distribution patterns is essential for cetacean management and conservation. Such knowledge is particularly important in areas where gillnet fishing occurs, as the Upper Gulf of California, which increases the probability of bycatch of cetaceans. In this area, the vaquita porpoise (Phocoena sinus) has been widely studied, but the knowledge of other cetaceans is scarce and based on traditional visual survey methods. We used data collected by an array of acoustic click detectors (C-PODs) during the summers 2011 to 2015 to analyze the distribution of dolphins in the Vaquita Refuge in the Upper Gulf of California. We recorded 120,038 echolocation click trains of dolphins during 12,371 days of recording effort at 46 sampling sites. Based on simultaneous visual and acoustic data, we estimated a false positive acoustic detection rate of 19.4%. Dolphin acoustic activity varied among sites, with higher activity in the east of the Vaquita Refuge. Acoustic activity was higher at night than during the day. We used negative binomial generalized linear models to study the count of clicks of dolphins in relation to spatial, temporal, physical, biological and anthropogenic explanatory variables. The best model selected for the response variable included sampling site, day-night condition, and vertical component of tide speed. Patterns in the spatial distribution of predicted acoustic activity of dolphins were similar to the acoustic activity observed per sampling season. Higher acoustic activity was predicted at night, but the tide speed variable was not relevant under this condition. Acoustic activity patterns could be related to the availability of prey resources since echolocation click trains are associated with foraging activities of dolphins. This is the first study of the distribution of dolphins in Mexico using medium-term systematic passive acoustic monitoring, and the results can contribute to better management to the natural protected area located in the Upper Gulf of California.

Description and classification of echolocation clicks of Indian Ocean humpback (Sousa plumbea) and Indo-Pacific bottlenose (Tursiops aduncus) dolphins from Menai Bay, Zanzibar, East Africa

Passive acoustic monitoring (PAM) is a powerful method to study the occurrence, movement and behavior of echolocating odontocetes (toothed whales) in the wild. However, in areas occupied by more than one species, echolocation clicks need to be classified into species. The present study investigated whether the echolocation clicks produced by small, at-risk, resident sympatric populations of Indian Ocean humpback dolphin (Sousa plumbea) and Indo-Pacific bottlenose dolphin (Tursiops aduncus) in Menai Bay, Zanzibar, East Africa, could be classified to allow species specific monitoring. Underwater sounds of S. plumbea and T. aduncus groups were recorded using a SoundTrap 202HF in January and June-August 2015. Eight acoustic parameters, i.e. -10 dB duration, peak, centroid, lower -3 and lower -10 dB frequencies, and -3 dB, -10 dB and root-mean-squared bandwidth, were used to describe and compare the two species’ echolocation clicks. Statistical analyses showed that S. plumbea clicks had significantly higher peak, centroid, lower -3 and lower -10 dB frequencies compared to T. aduncus, whereas duration and bandwidth parameters were similar for the two species. Random Forest (RF) classifiers were applied to determine parameters that could be used to classify the two species from echolocation clicks and achieved 28.6% and 90.2% correct species classification rates for S. plumbea and T. aduncus, respectively. Both species were classified at a higher rate than expected at random, however the identified classifiers would only be useful for T. aduncus monitoring. The frequency and bandwidth parameters provided most power for species classification. Further study is necessary to identify useful classifiers for S. plumbea. This study represents a first step in acoustic description and classification of S. plumbea and T. aduncus in the western Indian Ocean region, with potential application for future acoustic monitoring of species-specific temporal and spatial occurrence in these sympatric species.

Eavesdropping on dolphins: Investigating the habits of bottlenose dolphins (Tursiops truncatus) through fixed acoustic stations

This study investigates the bottlenose dolphin (Tursiops truncatus, Montagu 1821) habitat use in the Portofino marine protected area (NW Italy) and adjacent waters, a core area for the dolphins and a highly touristic area in the Mediterranean Sea. A permanent automated real-time passive acoustic monitoring system, able to detect and track dolphins continuously, was tested in the area within the activities of the Life+ Nature project ARION. The habits of bottlenose dolphins was investigated considering the resident rate inside the area, which quantifies the amount of time dolphins spent in these waters, by means of random forest regression. The dependency of dolphin resident rate was analyzed in relation to four explanatory variables: sea surface temperature, season, time of day, and proximity to the coast. Dolphins spent more time in the area during spring and when sea surface temperature ranged between 15–16°C. Summer resulted the season with lower dolphin residency with significant difference between working day and weekend, in the last the lowest residency was recorded. Main findings provide important information to properly manage the area in order to protect bottlenose dolphins.

Fine scale spatial variability in the influence of environmental cycles on the occurrence of dolphins at coastal sites

Environmental cycles often influence the presence of animals, creating patterns at different temporal scales, which may mean that their effects overlap and/or interact. Interactions between diel and seasonal cycles have been reported to influence fish behaviour but little is known about such interactions in marine top predators. Here, we studied the combined effect of seasonal, tidal and diel cycles on the occurrence of bottlenose dolphins (Tursiops truncatus) within a Marine Protected Area in Scotland. Our analyses were based on echolocation detections from passive acoustic devices (CPODs) deployed at three coastal sites between 2010 and 2016. We described patterns of dolphins' occurrence using circular statistics and then used generalised additive mixed models to explore the relative importance of each cycle and any interactions between them. We found site-specific cyclical patterns of presence that remained constant across years. There was a highly significant interaction between seasonal and diel cycles at two sites around deep channels, where occurrence was diurnal in summer but became nocturnal in autumn. The study demonstrates the highly plastic behaviour of bottlenose dolphins and shows a previously unreported behaviour that has management implications for this and other marine protected areas.

Validating automated click detector dolphin detection rates and investigating factors affecting performance

Passive acoustic monitoring (PAM) is a widely used technique for studying the distribution and habitat use of cetaceans. The C-POD, an acoustic sensor with an onboard automated click detector, has been deployed in diverse acoustic environments, but studies verifying its offshore detection rates and factors affecting detection probability are scarce. To empirically evaluate the performance of C-PODs in detecting bottlenose dolphins (Tursiops truncatus), C-PODs were deployed alongside archival acoustic recorders 12-30 km offshore in the Northwest Atlantic Ocean. The C-POD and acoustic recordings, post-processed using PAMGUARD software, were compared for a period of 6852 h. C-POD false positive rates were very low (mean 0.003%), and positive hourly detection accuracy was very high (mean 99.6%). Analysis of the acoustic environment and dolphin click characteristics revealed that true positive detections by C-PODs were significantly more likely to occur when PAMGUARD detected more clicks and there was increased high frequency noise (>20 kHz), likely from distant or unclassified clicks. C-PODs were found to be reliable indicators of dolphin presence at hourly or greater time scales. These results support the application of C-PODs in PAM studies that aim to investigate patterns of dolphin occurrence, such as those related to offshore windfarms.

Long-Term Monitoring of Dolphin Biosonar Activity in Deep Pelagic Waters of the Mediterranean Sea

Dolphins emit short ultrasonic pulses (clicks) to acquire information about the surrounding environment, prey and habitat features. We investigated Delphinidae activity over multiple temporal scales through the detection of their echolocation clicks, using long-term Passive Acoustic Monitoring (PAM). The Istituto Nazionale di Fisica Nucleare operates multidisciplinary seafloor observatories in a deep area of the Central Mediterranean Sea. The Ocean noise Detection Experiment collected data offshore the Gulf of Catania from January 2005 to November 2006, allowing the study of temporal patterns of dolphin activity in this deep pelagic zone for the first time. Nearly 5,500 five-minute recordings acquired over two years were examined using spectrogram analysis and through development and testing of an automatic detection algorithm. Echolocation activity of dolphins was mostly confined to nighttime and crepuscular hours, in contrast with communicative signals (whistles). Seasonal variation, with a peak number of clicks in August, was also evident, but no effect of lunar cycle was observed. Temporal trends in echolocation corresponded to environmental and trophic variability known in the deep pelagic waters of the Ionian Sea. Long-term PAM and the continued development of automatic analysis techniques are essential to advancing the study of pelagic marine mammal distribution and behaviour patterns.