Passive acoustic monitoring of the decline of Mexico's critically endangered vaquita

An assessment of potential interventions to reduce the totoaba illegal trade market

The illegal trade in totoaba (Totoaba macdonaldi) is causing adverse social, ecological, and economic impacts. This illegal activity is accelerating the overexploitation of totoaba and pushing the critically endangered vaquita (Phocoena sinus) closer to extinction. Despite extensive efforts to recover vaquita populations, scant attention has been given to the totoaba trade as an independent issue. As a result, data on the totoaba trade are limited, which hampers robust analyses and development of effective interventions to reduce illegal harvesting. We used a previously developed framework specifically designed to examine dynamics of illegal markets and guide measures to mitigate illegal use of totoaba. This framework separates markets into 3 analytical levels: characterization of participating actors (e.g., fishers, intermediaries); examination of how actors interact within the market (e.g., organization of supply chains); and assessment of the overall market dynamics that result from these interactions (e.g., factors determining price and quantity). We reviewed existing literature (108 initial articles) and interviewed key market actors, academics, and nongovernmental organization experts (14) to obtain data for this framework. Our findings offer an overview of the totoaba illegal market operation, highlighting intervention points (e.g., customs agents) and areas where additional information is required to decrease information gaps (e.g., US local market). We describe the structure and complexity of this market, emphasizing the influential role of organized crime in shaping its dynamics (e.g., controlling prices paid to fishers and stockpiling). By providing a systematic and in-depth understanding of the market operation, we aimed to establish a benchmark for effective interventions and future research aimed at reducing uncertainties. Our results provide a crucial step toward addressing this critical issue and can help facilitate development of effective strategies to combat the illegal totoaba trade and promote biodiversity conservation more broadly.

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.

Ecological inferences about marine mammals from passive acoustic data

Monitoring on the basis of sound recordings, or passive acoustic monitoring, can complement or serve as an alternative to real-time visual or aural monitoring of marine mammals and other animals by human observers. Passive acoustic data can support the estimation of common, individual-level ecological metrics, such as presence, detection-weighted occupancy, abundance and density, population viability and structure, and behaviour. Passive acoustic data also can support estimation of some community-level metrics, such as species richness and composition. The feasibility of estimation and certainty of estimates is highly context dependent, and understanding the factors that affect the reliability of measurements is useful for those considering whether to use passive acoustic data. Here, we review basic concepts and methods of passive acoustic sampling in marine systems that often are applicable to marine mammal research and conservation. Our ultimate aim is to facilitate collaboration among ecologists, bioacousticians, and data analysts. Ecological applications of passive acoustics require one to make decisions about sampling design, which in turn requires consideration of sound propagation, sampling of signals, and data storage. One also must make decisions about signal detection and classification and evaluation of the performance of algorithms for these tasks. Investment in the research and development of systems that automate detection and classification, including machine learning, are increasing. Passive acoustic monitoring is more reliable for detection of species presence than for estimation of other species-level metrics. Use of passive acoustic monitoring to distinguish among individual animals remains difficult. However, information about detection probability, vocalisation or cue rate, and relations between vocalisations and the number and behaviour of animals increases the feasibility of estimating abundance or density. Most sensor deployments are fixed in space or are sporadic, making temporal turnover in species composition more tractable to estimate than spatial turnover. Collaborations between acousticians and ecologists are most likely to be successful and rewarding when all partners critically examine and share a fundamental understanding of the target variables, sampling process, and analytical methods.

What the F-POD? Comparing the F-POD and C-POD for monitoring of harbor porpoise (Phocoena phocoena)

Passive acoustic monitoring (PAM) is a cost-effective method for monitoring cetacean populations compared with techniques such as aerial and ship-based surveys. The Cetacean POrpoise Detector (C-POD) has become an integral tool in monitoring programs globally for over a decade, providing standardized metrics of occurrence that can be compared across time and space. However, the phasing out of C-PODs following the development of the new Full waveform capture POD (F-POD) with increased sensitivity, improved train detection, and reduced false-positive rates represents an important methodological change in data collection, particularly when being introduced into existing monitoring programs. Here, we compare the performance of the C-POD with that of its successor, the F-POD, co-deployed in a field setting for 15 months, to monitor harbor porpoise (Phocoena phocoena). While similar temporal trends in detections were found for both devices, the C-POD detected only 58% of the detection-positive minutes (DPM), recorded by the F-POD. Differences in detection rates were not consistent through time making it difficult to apply a correction factor or directly compare results obtained from the two PODs. Generalized additive models (GAMs) were used to test whether these differences in detection rates would have an effect on analyses of temporal patterns and environmental drivers of occurrence. No differences were found in seasonal patterns or the environmental correlates of porpoise occurrence (month, diel period, temperature, environmental noise, and tide). However, the C-POD failed to detect sufficient foraging rates to identify temporal patterns in foraging behavior, which were shown by the F-POD. Our results suggest that the switch to F-PODs will have little effect on determining broad-scale seasonal patterns of occurrence but may improve our understanding of fine-scale behaviors such as foraging. We highlight how care must be taken interpreting F-POD results as indicative of increased occurrence when used in time-series analysis.

More vaquita porpoises survive than expected

In 2018, it was estimated that fewer than 20 of Mexico’s endemic vaquita porpoise Phocoena sinus remained, and the species was declining by 47% yr-1. Entanglement in gillnets is the sole threat to the species, and since the last population size estimate, gillnetting has increased in the small area where most vaquitas remain—a 12 × 24 km area in the Gulf of California near San Felipe, Mexico. We conducted research efforts in 2019 and 2021 in that area to estimate the minimum numbers of adults and calves and look for any signs that vaquitas are unhealthy. Through expert elicitation, we estimated between 7 and 15 unique individuals were seen in 2019 and 5-13 were seen in 2021. Calves were seen in both years, and all vaquitas appeared healthy. Population projections from the last full survey indicated that more vaquitas have survived than expected. We suggest that these surviving adult vaquitas may have learned to avoid entanglement in gillnets. These vaquitas and their calves provide hope that the species can survive. However, given the high levels of illegal gillnetting and the theft of equipment which hindered our monitoring efforts, and with only around 10 individuals remaining, survival can only be assured if vaquita habitat is made gillnet-free.

Estimating cetacean population trends from static acoustic monitoring data using Paired Year Ratio Assessment (PYRA)

The cetacean conservationist is often faced with evaluating population trends from abundance data that are either sparse or recorded at different times in different years. The presence of diel or seasonal patterns in the data together with unplanned gaps is often problematic. Such data are typical of those obtained from static acoustic monitoring. We present a simple and transparent non-parametric trend evaluation method, ‘Paired Year Ratio Assessment (PYRA)’ that uses only whole days of data wherever they are present in each of successive pairs of periods of 365 days. We provide a quantitative comparison of the performance of PYRA with traditional generalised additive models (GAMS) and nonparametric randomisation tests that require a greater level of skill and experience for both application and interpretation. We conclude that PYRA is a powerful tool, particularly in the context of identifying population trends which is often the main aim of conservation-targeted acoustic monitoring.

Estimating the abundance of the critically endangered Baltic Proper harbour porpoise (Phocoena phocoena) population using passive acoustic monitoring

Knowing the abundance of a population is a crucial component to assess its conservation status and develop effective conservation plans. For most cetaceans, abundance estimation is difficult given their cryptic and mobile nature, especially when the population is small and has a transnational distribution. In the Baltic Sea, the number of harbour porpoises (Phocoena phocoena) has collapsed since the mid-20th century and the Baltic Proper harbour porpoise is listed as Critically Endangered by the IUCN and HELCOM; however, its abundance remains unknown. Here, one of the largest ever passive acoustic monitoring studies was carried out by eight Baltic Sea nations to estimate the abundance of the Baltic Proper harbour porpoise for the first time. By logging porpoise echolocation signals at 298 stations during May 2011-April 2013, calibrating the loggers' spatial detection performance at sea, and measuring the click rate of tagged individuals, we estimated an abundance of 71-1105 individuals (95% CI, point estimate 491) during May-October within the population's proposed management border. The small abundance estimate strongly supports that the Baltic Proper harbour porpoise is facing an extremely high risk of extinction, and highlights the need for immediate and efficient conservation actions through international cooperation. It also provides a starting point in monitoring the trend of the population abundance to evaluate the effectiveness of management measures and determine its interactions with the larger neighboring Belt Sea population. Further, we offer evidence that design-based passive acoustic monitoring can generate reliable estimates of the abundance of rare and cryptic animal populations across large spatial scales.

Accounting for the Lombard effect in estimating the probability of detection in passive acoustic surveys: Applications for single sensor mitigation and monitoring

The detection range of calling animals is commonly described by the passive sonar equations. However, the sonar equations do not account for interactions between source and ambient sound level, i.e., the Lombard effect. This behavior has the potential to introduce non-linearities into the sonar equations and result in incorrectly predicted detection ranges. Here, we investigate the relationship between ambient sound and effective detection ranges for North Atlantic right whales (Eubalaena glacialis) in Cape Cod Bay, MA, USA using a sparse array of acoustic recorders. Generalized estimating equations were used to model the probability that a call was detected as a function of distance between the calling animal and the sensor and the ambient sound level. The model suggests a non-linear relationship between ambient sound levels and the probability of detecting a call. Comparing the non-linear model to the linearized version of the same model resulted in 12 to 25% increases in the effective detection range. We also found evidence of the Lombard effect suggesting that it is the most plausible cause for the non-linearity in the relationship. Finally, we suggest a simple modification to the sonar equation for estimating detection probability for single sensor monitoring applications.

Low-frequency sampling rates are effective to record bottlenose dolphins

Acoustic monitoring in cetacean studies is an effective but expensive approach. This is partly because of the high sampling rate required by acoustic devices when recording high-frequency echolocation clicks. However, the proportion of echolocation clicks recorded at different frequencies is unknown for many species, including bottlenose dolphins. Here, we investigated the echolocation clicks of two subspecies of bottlenose dolphins in the western South Atlantic Ocean. The possibility of recording echolocation clicks at 24 and 48 kHz was assessed by two approaches. First, we considered the clicks in the frequency range up to 96 kHz. We found a loss of 0.95-13.90% of echolocation clicks in the frequency range below 24 kHz, and 0.01-0.42% below 48 kHz, to each subspecies. Then, we evaluated these recordings downsampled at 48 and 96 kHz and confirmed that echolocation clicks are recorded at these lower frequencies, with some loss. Therefore, despite reaching high frequencies, the clicks can also be recorded at lower frequencies because echolocation clicks from bottlenose dolphins are broadband. We concluded that ecological studies based on the presence-absence data are still effective for bottlenose dolphins when acoustic devices with a limited sampling rate are used.

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.

Deep neural networks for automated detection of marine mammal species

Deep neural networks have advanced the field of detection and classification and allowed for effective identification of signals in challenging data sets. Numerous time-critical conservation needs may benefit from these methods. We developed and empirically studied a variety of deep neural networks to detect the vocalizations of endangered North Atlantic right whales (Eubalaena glacialis). We compared the performance of these deep architectures to that of traditional detection algorithms for the primary vocalization produced by this species, the upcall. We show that deep-learning architectures are capable of producing false-positive rates that are orders of magnitude lower than alternative algorithms while substantially increasing the ability to detect calls. We demonstrate that a deep neural network trained with recordings from a single geographic region recorded over a span of days is capable of generalizing well to data from multiple years and across the species’ range, and that the low false positives make the output of the algorithm amenable to quality control for verification. The deep neural networks we developed are relatively easy to implement with existing software, and may provide new insights applicable to the conservation of endangered species.

Let's Agree to Disagree: Comparing Auto-Acoustic Identification Programs for Northeastern Bats

With the declines in abundance and changing distribution of white-nose syndrome–affected bat species, increased reliance on acoustic monitoring is now the new “normal.” As such, the ability to accurately identify individual bat species with acoustic identification programs has become increasingly important. We assessed rates of disagreement between the three U.S. Fish and Wildlife Service–approved acoustic identification software programs (Kaleidoscope Pro 4.2.0, Echoclass 3.1, and Bat Call Identification 2.7d) and manual visual identification using acoustic data collected during summers from 2003 to 2017 at Fort Drum, New York. We assessed the percentage of agreement between programs through pairwise comparisons on a total nightly count level, individual file level (e.g., individual echolocation pass call file), and grouped maximum likelihood estimate level (e.g., probability values that a species is misclassified as present when in fact it is absent) using preplanned contrasts, Akaike Information Criterion, and annual confusion matrices. Interprogram agreement on an individual file level was low, as measured by Cohen's Kappa (0.2–0.6). However, site-night level pairwise comparative analysis indicated that program agreement was higher (40–90%) using single season occupancy metrics. In comparing analytical outcomes of our different datasets (i.e., how comparable programs and visual identification are regarding the relationship between environmental conditions and bat activity), we determined high levels of congruency in the relative rankings of the model as well as the relative level of support for each individual model. This indicated that among individual software packages, when analyzing bat calls, there was consistent ecological inference beyond the file-by-file level at the scales used by managers. Depending on objectives, we believe our results can help users choose automated software and maximum likelihood estimate thresholds more appropriate for their needs and allow for better cross-comparison of studies using different automated acoustic software.

Decline towards extinction of Mexico's vaquita porpoise (Phocoena sinus)

The vaquita (Phocoena sinus) is a small porpoise endemic to Mexico. It is listed by IUCN as Critically Endangered because of unsustainable levels of bycatch in gillnets. The population has been monitored with passive acoustic detectors every summer from 2011 to 2018; here we report results for 2017 and 2018. We combine the acoustic trends with an independent estimate of population size from 2015, and visual observations of at least seven animals in 2017 and six in 2018. Despite adoption of an emergency gillnet ban in May 2015, the estimated rate of decline remains extremely high: 48% decline in 2017 (95% Bayesian credible interval (CRI) 78% decline to 9% increase) and 47% in 2018 (95% CRI 80% decline to 13% increase). Estimated total population decline since 2011 is 98.6%, with greater than 99% probability the decline is greater than 33% yr^-1. We estimate fewer than 19 vaquitas remained as of summer 2018 (posterior mean 9, median 8, 95% CRI 6-19). From March 2016 to March 2019, 10 dead vaquitas killed in gillnets were found. The ongoing presence of illegal gillnets despite the emergency ban continues to drive the vaquita towards extinction. Immediate management action is required if the species is to be saved.

Last call: Passive acoustic monitoring shows continued rapid decline of critically endangered vaquita

The vaquita is a critically endangered species of porpoise. It produces echolocation clicks, making it a good candidate for passive acoustic monitoring. A systematic grid of sensors has been deployed for 3 months annually since 2011; results from 2016 are reported here. Statistical models (to compensate for non-uniform data loss) show an overall decline in the acoustic detection rate between 2015 and 2016 of 49% (95% credible interval 82% decline to 8% increase), and total decline between 2011 and 2016 of over 90%. Assuming the acoustic detection rate is proportional to population size, approximately 30 vaquita (95% credible interval 8-96) remained in November 2016.

Categorizing click trains to increase taxonomic precision in echolocation click loggers

Passive acoustic monitoring is an efficient way to study acoustically active animals but species identification remains a major challenge. C-PODs are popular logging devices that automatically detect odontocete echolocation clicks. However, the accompanying analysis software does not distinguish between delphinid species. Click train features logged by C-PODs were compared to frequency spectra from adjacently deployed continuous recorders. A generalized additive model was then used to categorize C-POD click trains into three groups: broadband click trains, produced by bottlenose dolphin (Tursiops truncatus) or common dolphin (Delphinus delphis), frequency-banded click trains, produced by Risso's (Grampus griseus) or white beaked dolphins (Lagenorhynchus albirostris), and unknown click trains. Incorrect categorization rates for broadband and frequency banded clicks were 0.02 (SD 0.01), but only 30% of the click trains met the categorization threshold. To increase the proportion of categorized click trains, model predictions were pooled within acoustic encounters and a likelihood ratio threshold was used to categorize encounters. This increased the proportion of the click trains meeting either the broadband or frequency banded categorization threshold to 98%. Predicted species distribution at the 30 study sites matched well to visual sighting records from the region.

Preliminary estimates of the abundance and fidelity of dolphins associating with a demersal trawl fishery

The incidental capture of wildlife in fishing gear presents a global conservation challenge. As a baseline to inform assessments of the impact of bycatch on bottlenose dolphins (Tursiops truncatus) interacting with an Australian trawl fishery, we conducted an aerial survey to estimate dolphin abundance across the fishery. Concurrently, we carried out boat-based dolphin photo-identification to assess short-term fidelity to foraging around trawlers, and used photographic and genetic data to infer longer-term fidelity to the fishery. We estimated abundance at ≈ 2,300 dolphins (95% CI = 1,247–4,214) over the ≈ 25,880-km² fishery. Mark-recapture estimates yielded 226 (SE = 38.5) dolphins associating with one trawler and some individuals photographed up to seven times over 12 capture periods. Moreover, photographic and genetic re-sampling over three years confirmed that some individuals show long-term fidelity to trawler-associated foraging. Our study presents the first abundance estimate for any Australian pelagic dolphin community and documents individuals associating with trawlers over days, months and years. Without trend data or correction factors for dolphin availability, the impact of bycatch on this dolphin population’s conservation status remains unknown. These results should be taken into account by management agencies assessing the impact of fisheries-related mortality on this protected species.

Using paired visual and passive acoustic surveys to estimate passive acoustic detection parameters for harbor porpoise abundance estimates

Passive acoustic monitoring is a promising approach for monitoring long-term trends in harbor porpoise (Phocoena phocoena) abundance. Before passive acoustic monitoring can be implemented to estimate harbor porpoise abundance, information about the detectability of harbor porpoise is needed to convert recorded numbers of echolocation clicks to harbor porpoise densities. In the present study, paired data from a grid of nine passive acoustic click detectors (C-PODs, Chelonia Ltd., United Kingdom) and three days of simultaneous aerial line-transect visual surveys were collected over a 370 km² study area. The focus of the study was estimating the effective detection area of the passive acoustic sensors, which was defined as the product of the sound production rate of individual animals and the area within which those sounds are detected by the passive acoustic sensors. Visually estimated porpoise densities were used as informative priors in a Bayesian model to solve for the effective detection area for individual harbor porpoises. This model-based approach resulted in a posterior distribution of the effective detection area of individual harbor porpoises consistent with previously published values. This technique is a viable alternative for estimating the effective detection area of passive acoustic sensors when other experimental approaches are not feasible.