Representation at the 22nd Biennial Conference on the Biology of Marine Mammals in Halifax, Nova Scotia

Home / Latest News / Representation at the 22nd Biennial Conference on the Biology of Marine Mammals in Halifax, Nova Scotia

MUCRU members are gearing up to attending the 22nd Biennial Conference on the Biology of Marine Mammals in Halifax, Nova Scotia (22-27 October 2017). During the conference, we will be presenting three oral talks, four speed talks and three posters (see below for details). In addition, some MUCRU members will be hosting a workshop focusing on Bayesian capture-recapture (see below for details).

We look forward to seeing many of our colleagues, friends and collaborators and hearing about your exciting work!!

Oral presentations

1. Title: Investigating the feasibility and potential applications of non-invasive laser photogrammetry in obtaining morphometric data on free-ranging coastal dolphins

Authors: Martin van Aswegen, John Symons, Fredrik Christiansen, Janet Mann, Krista Nicholson, Kate Sprogis and Lars Bejder

Abstract: Morphometric data plays a pivotal role in understanding key life history traits to elucidate biological, ecological and evolutionary processes. Obtaining morphometric data from free-ranging cetaceans is problematic, as traditional methods rely on either post-mortem or highly-invasive techniques. The present study evaluated the feasibility of remote laser photogrammetry as a non-invasive technique to obtain morphometric data on free-ranging coastal dolphins. First, we used simulation models and post-mortem specimens to investigate potential sources of measurement error and quantified their influence on the accuracy and precision of the morphometric data. These sources include horizontal angle, distance, poor calibration, lens distortion and body flexion. Second, to demonstrate the potential applications of this technique, laser photogrammetry measurements were obtained during boat-based photo-identification surveys on Indo-Pacific bottlenose dolphins (Tursiops aduncus) from Western Australia (Bunbury, Shark Bay and Mandurah). Laser-derived, blowhole-to-dorsal fin (BH-DF) measurements were obtained from individuals of known ages in Bunbury (n=103) and Shark Bay (n=76), in addition to individuals of unknown ages in Mandurah (n=124). Our laser-derived measurement data developed population growth curves in conjunction with longitudinal demographic data from Bunbury (~10 years) and Shark Bay (~33 years). These growth curves characterise not only the relationship between age and length, but also the significant morphological differences between these geographically-isolated populations. The Gompertz growth function provided the best fit for Bunbury and Shark Bay’s known-age populations and was consequently used to predict the age of dolphins in Mandurah. This study demonstrates the value of remote laser photogrammetry as an effective tool to investigate individual and population-based growth and life-history parameters. This non-invasive technique will provide unique opportunities to better understand the ecological, demographic and life-history characteristics of a population and so better inform conservation management strategies for free-ranging cetacean populations.

2. Title: Maternal size and body condition determine calf growth rates in Southern right whales: Repeated individual sampling using unmanned aerial vehicles

Authors: Fredrik Christiansen, Fabien Vivier, Claire Charlton, Rhianne Ward, Stephen Burnell, Alicia Amerson and Lars Bejder

Abstract: Baleen whales exhibit the fastest offspring growth rates of any mammal, suggesting considerable energetic costs for the mothers, particularly during the lactation phase. With lactating females fasting during the first three-four months after parturition, the absolute amount of energy stored in their body reserves should dictate how much energy a female can invest in her offspring which, in turn, should determine the growth rates of their calves. We tested these hypotheses on Southern right whales (Eubalaena australis) on a breeding ground in South Australia, by measuring the growth rates of individual calves in relation to the rate of loss in the body condition of their mothers, using unmanned aerial vehicles and photogrammetry. We obtained repeated measurements of the body size (body volume calculated from length and width measurements) of 39 mother and calf pairs over periods ranging from 40 to 89 days. Calves grew in size (volume) at a rate of 0.32m3 day-1(SE=0.041), while lactating females decreased in size at a rate of 0.51m3 day-1 (SE=0.146). The growth rate of the calves were positively related to the rate of loss in maternal body volume (R2=0.22), suggesting that maternal volume loss is proportional to the energy investment into her calf. Calf investment (rate of loss in maternal volume) was, in turn, determined by the mothers absolute size (length) and relative body condition (R2=0.47), with both longer and wider females investing more energy into their calves. Over a three-month breeding season, lactating females lost on average 25% of their initial body volume, while calves increased by 56.4% in length and as much as 600% in volume. This study demonstrates the considerable energetic costs that Southern right whale females face during the lactation period, and highlights the importance of good maternal condition for the growth and survival of their dependent calves.

3. Title: False positives in capture-recapture: why a philosophy of model selection matter

Author: Robert Rankin

Abstract: A false positive is the declaration that an effect is “significant” when, in truth, it has no effect; e.g. falsely concluding that “covariate X has a significant effect on survival”. This study looked at the rates of false positives and false negatives in a model capture-recapture system under temporary emigration (PCRD), using three different approaches to model confirmation and/or hypothesis testing: i) model confirmation by the AICc, ii) Bayesian model confirmation, and iii) the Neyman-Pearson hypothesis testing paradigm. In the PCRD system, survival and temporary emigration are so perniciously confounded, that there is a very strong tendency to overfit models and declare false positives. This is most apparent when using the AICc, contrary to most capture-recapture practitioners’ preference. The Bayesian confirmation paradigm does better, but requires a lot of data to guarantee consistent selection. Also discussed are the philosophies underpinning the Akaikean, Neyman-Pearson and Bayesian confirmation paradigms, and practical guidance for when practitioners should use them.

Speed talk presentations

1. Title: The first broad scale distribution models for coastal dolphins in northern Western Australia

Authors: Daniella Hanf, Amanda Hodgson, Lars Bejder, Halina Kobryn and Josh Smith.

Abstract: We present results from the first broad scale distribution models for coastal dolphins – Australian humpback (Sousa sahulensis), Australian snubfin (Orcaella heinsohnii) and Indo-Pacific bottlenose (Tursiops aduncus) – in northern Western Australia (WA). The 10,600 km2 western Pilbara study area is flat and shallow over a wide area and has few rivers entering it, a stark contrast to the east coast of Australia where most research has been undertaken for humpback and snubfin dolphins. Dolphin sightings were collected during 10 dugong aerial surveys that were conducted in May, July, October and December 2012, and May and October in 2013, 2014 and 2015. Snubfin dolphin sightings were too few to model but their occurrence in Exmouth Gulf was further south than previously recorded, suggesting that their known range be extended. Models to investigate inter- and intra- seasonal patterns in humpback and bottlenose dolphin distribution were built using Maximum Entropy (MaxEnt) software. Bottlenose and humpback dolphins were sympatric, with overlap in occurrence across the area. Humpback dolphins were associated with intertidal areas near the mainland and also islands, including 50 km offshore.  Both species had a seasonal shift in distribution. In May they were nearer to the 20 m bathymetric contour, likely in response to the warm Leeuwin Current, WA’s prevailing oceanographic feature. Occurrence was greater in Exmouth Gulf during October, a period of higher productivity that also attracts other marine megafauna. These new findings are important for the species’ protection as they are faced with an additive threat from vessel traffic, exploration activity, port development and fishing. They also highlight the need to start investigating the potential effect of a climate-driven regime shift in this subtropical environment that is prone to cyclones, dependant on the Leeuwin Current and has already suffered an extreme marine heating event in recent years.

 2. Title: Piecing together the foraging ecology of bottlenose dolphins (Tursiops aduncus) in southwestern Australia

Authors: Shannon McCluskey, Kate Sprogis, Lars Bejder and Neil Loneragan

Abstract: To investigate the foraging ecology of a population of bottlenose dolphins (Turciops aduncus), complimentary approaches were applied to provide information for the conservation of dolphins and their food resources. Stomach contents of stranded dolphins (n=14) were identified and compared with relative prey availability, measured through prey sampling, as well as quality of prey (measured as KJ/g; inferred from bomb-calorific measurements). Stable isotope ratios (dC13 and dN15) of live dolphins (n=14) obtained through biopsy samples, fish (n=26 species), and invertebrates (including crabs and zooplankton) were also compared to stomach contents and relative prey composition from sampling efforts. Boat-based, photo-identification surveys were conducted year-round between 2007-2013, to determine the proportion of time biopsied dolphin spent in inshore (bay and estuary) and outer (ocean) waters. Dolphins selectively foraged on prey of medium caloric value (4-6 KJ/g) and did not consume prey in proportion to their availability in the environment (McCluskey et al. 2016). Overall, smaller fish species (e.g., members of the Gobiidae) than expected were found in dolphin stomachs, suggesting that the catchability of prey is a better indicator of prey value than size. Stomach contents of individuals varied widely, with numbers of unique prey groups ranging from 1 to 24, as did the dN15 signatures, which ranged between 9.9-13.7 ‰. Dolphins that remained in the inshore waters (bay and estuary) year-round had higher dN15 signatures (mean=13.7 ‰) than those in offshore waters (mean=9.9 ‰), possibly due to higher levels of source nitrogen. As expected, dolphins occupying outer waters generally had lower dC13 signatures (mean=17.5 ‰) than those from the inner waters (mean=13.0‰), indicating differences between pelagic and benthic sources of carbon. Using complimentary methods to investigate foraging ecology yields more information than any single method, and is therefore a valuable approach to answering questions related to resource management.

3. Title: Implications of survey effort on estimating reproductive output, home range size and abundance of a long-lived marine top predator

Authors: John Symons, Kate Sprogis and Lars Bejder

Abstract: Potential long-term cumulative impacts of non-lethal interactions between cetaceans and human activities are poorly understood and require quantification of demographic parameters. The bottlenose dolphin (Tursiops aduncus) population off Bunbury, Western Australia, is forecast to decline by 6% annually (Manlik et al. 2016; Ecology and Evolution). Knowledge on the influence of survey effort on the ability to estimate demographic parameters is needed. To address this, we utilised a three-year capture-recapture dataset. Following Pollock’s Robust Design, seasons (summer, autumn, winter, spring) were used to define ‘primary periods’. Each primary period, consisted of six ‘secondary periods’ during which photo-identification transects were completed. Capture-recapture models were used to assess the effect of survey effort on the ability to detect trends in abundance. Dolphin sighting frequencies were used to assess limitations for home-range size modelling. We examined female calving histories to assess survey effort on quantifying reproductive output (calving events). A 50% reduction in effort was simulated by removing secondary sampling periods from each primary period, and demographic parameter estimates were compared between the two scenarios. A 50% reduction in survey effort: a) increased the time required to detect a 6% annual change in abundance from 2.25 to 4.75 years (at 80% power); b) limited our ability to analyse home range sizes, by increasing the time for individuals to be sighted >30 occasions (an often used cut-off metric for home-range analyses) from 7.74 to 15.46 years, which also reduced the number of individuals with adequate sighting frequencies from 13 to 0; and c) decreased our ability to detect calving events, resulting in 26.8%, 13.4% and 25.8% of calving events each year going undocumented in the year the calf was born. These results highlight the importance of survey effort on the ability to assess demographic parameters with clear implications for population viability analyses and population forecasting.

4. Title: Long-term effects of food provisioning on the reproductive success of free-ranging bottlenose dolphin population

Authors: Valeria Senigaglia, John Symons, Kate Sprogis, Fredrik Christiansen and Lars Bejder

Abstract: Human-wildlife interactions are frequently mediated by food-provisioning which facilitates predictable spatial and temporal up-close encounters. However, food provisioning may alter the natural behaviour of an animal, encouraging potentially adverse behavior, (e.g. begging for food handouts) and thus may affect the population’s survival and reproductive success. In Bunbury, Western Australia, a state-licensed provisioning program offers fish handouts to a limited number of free-ranging, bottlenose dolphins (Tursiops aduncus). The resident dolphin population is small (ca 120 individuals) and subjected to multiple anthropogenic stressors. Sensitivity analyses shows that reproduction greatly affect the viability of this population which is forecasted to decline at current level of reproductive output (Manlik et al. 2016, Ecology and Evolution). Thus, it is necessary to understand potential negative effects of food provisioning on reproductive success to ensure the long-term survival of the population. Dedicated individual focal follows, indicated a high incidence of dolphin-human interactions, including begging behaviours and illegal food provisioning. These activities were recorded during 19 of 28 surveys (68%), which is a higher percentage than observed in Savanna, Georgia, and Sarasota, Florida (64% and 22%, respectively) in the USA, and Cockburn Sound, Western Australia (13%). Using GLM, we investigated the effects of food provisioning on female dolphin reproductive output (i.e. total number of calves born per female) and success (i.e. total number of successfully weaned calves). Based on the long-term data available (> 10 years), results from the best-fitted models showed that the reproductive output was positively correlated with food provisioning, compared to non-provisioned females. In contrast, after accounting for the total number of calves per female, results showed that the number of successfully weaned calves (i.e. reproductive success) was negatively influenced by provisioning. Our findings highlight that wildlife provisioning in Bunbury lead to a decrease in calves’ survival, which can have significant effects on population dynamics.

Poster presentations

1. Title: Genetic structure of socially and spatially discrete subpopulations of dolphins (Tursiops aduncus) in metropolitan waters of Perth, Western Australia

Authors: Delphine Chabanne, Celine Frere, Simon Allen, Lars Bejder and William Sherwin

Abstract: Indo-Pacific bottlenose dolphins (Tursiops aduncus) in Perth metropolitan waters, Western Australia, are spatially and socially segregated into four discrete subpopulations. Contemporary and historical genetic differences among these socio-geographic subpopulations were assessed using ten microsatellite loci and mitochondrial DNA control region sequence data. Pairwise estimates of genetic differentiation (FST) based on microsatellite alleles revealed some significant differences between the subpopulations, supported by little or no contemporary gene flow. Differences were, however, too weak and recent to assign individuals to their original subpopulations using the Bayesian clustering implemented in STRUCTURE. Historically, the socio-geographic subpopulations originated from two ancestral lineages, although no apparent geographic clustering was detected. The Tajima’s D analysis suggested the occurrence of a historical bottleneck event or selection in at least one socio-geographic subpopulation associated with the semi-enclosed embayment of Cockburn Sound, although none of the other bottleneck tests agreed. Likewise, high genetic diversity was maintained by moderate asymmetric gene flow (m > 0.10) and could have erased any bottleneck or selection signal. Within this source-sink dynamic, it is important to conserve the source subpopulation on which the others depend. Also warranting particular conservation and management attention is an estuarine subpopulation, which acts as a sink and is vulnerable to extinction because of potential inbreeding, its small population size and impacts from multiple anthropogenic threats.

2. Title: Characteristics of a resident Indo-Pacific bottlenose dolphin (Tursiops aduncus) population in a microtidal temperate estuary

Authors: Krista Nicholson, Neil Loneragan and Lars Bejder

Abstract: Along the south-west coastline of Western Australia, dolphin populations occupy estuarine and riverine systems. We conducted an assessment of the Indo-Pacific bottlenose dolphin (Tursiops aduncus) population in the Peel-Harvey estuary, a shallow (< 2 meters) system with two openings to the Indian Ocean. In 2016, boat-based photo-identification surveys were conducted in winter, spring summer and autumn seasons along 19 transects designed to estimate population abundance and demographic parameters using Pollock’s Closed Robust Design; transects covered extensive areas in both the Peel-Harvey (c. 130 km2) and along the adjacent coast (c. 440km2). Information on dolphin strandings was also collated and assessed. We encountered 490 groups including 288 in the Peel-Harvey, resulting in the identification of c. 400 individuals. We identified a year-round resident population of c. 90 individuals occupying the Peel-Harvey (1.5 dolphins/km2) and recorded seasonal visitors to the system. Of the 81 known individuals alive in December 2016, 37 were considered adults, 26 juveniles/subadults and 18 dependent calves of which nine were born in 2016. The resident population uses the entire extent of the estuary, including the river systems. Some residents were also observed along the coast at the mouths of the estuary. Since 1987 at least 35 live stranding events have been recorded in the Peel-Harvey, involving c. 60 dolphins. At least 13 of the current resident dolphins have stranded since 1990, some on multiple occasions. All live strandings have been in locations regularly occupied by the resident dolphins with individuals being in good health at the time of stranding. One hypothesis to such dolphin stranding relates to foraging strategies during extreme tides. Because of the stranding frequency and the potential health implications, appropriate stranding response measures should be a management priority for this population.

3. Title: Will the spinner dolphins of Hawaii Island ever be able to rest peacefully?

Authors: Julian Tyne, Heather Heenehan, David Johnston and Lars Bejder

Abstract: The potential consequences of frequent human interactions with Hawaiian spinner dolphins have been of concern to managers and policymakers in the National Oceanic and Atmospheric Administration (NOAA), for more than a decade. In 2005, NOAA published an advanced notice of proposed rulemaking aimed at reducing impacts of human activity on spinner dolphins. In 2010, we commenced research aimed to inform management on the demographics, resting habitats, and exposure to human activities of the Hawaii Island spinner dolphin stock, the most genetically isolated stock in the Hawaiian archipelago. Between 2014-2016, we produced eight peer-reviewed manuscripts  documenting: i) spinner dolphin abundance is lower than all previous estimates, indicating a possible long term decline; ii) the dolphins are less likely to rest outside of their preferred resting bays, indicating their importance to resting spinner dolphins; iii) the dolphins exhibit a constrained behavioural schedule, which may make them less resilient to disturbance and iv) are exposed to human activities (<100m) for 82.7% of day time hours within resting bays. NOAA recently (2016) published their proposed rule, a 50-yard no approach distance rule within 2 nm of the coastline to mitigate effects of human disturbance. While other management options were noted, including NOAA’s previously preferred option of time area-closures, the distance rule was ultimately their preferred management option. Based on our peer reviewed research, we recommend compulsory time-area closures in the preferred spinner resting habitats during important resting periods combined with a 50-yard no approach distance rule out to 2 nm.   Our recommendations are consistent with the recent proposal by the IUCN Marine Mammal Protected Areas Task Force to designate spinner dolphin resting bays as Important Marine Mammal Areas (IMMA). This presentation will discuss not only the scientific basis for our recommendations, the need for unambiguous enforceable policy, and the socio-political reasoning for alternative options.

Workshops

Title: Bayesian and Hierarchical Bayesian models for Capture-Recapture: Introduction to theory and practical learning in R & BUGS.

Hosted by Robert Rankin and Krista Nicholson

The workshop will include an introduction to Bayesian theory and beginner/intermediate tutorials on R and JAGS. JAGS is the popular and flexible Bayesian scripting language that can code CR models and much more! The workshop is suitable for students, researchers, professors, veteran program MARK users, and anyone who is already familiar with CR and would like to learn how to implement Bayesian CR models. Intermediate familiarity with R and CR is expected.

Outline of the workshop: i) Bayesian philosophy; ii) introduction to the BUGS/JAGS language; iii-v) practical tutorials to implement common CR models for cetaceans (POPAN, PCRD, individual-covariate models); vi) Bayesian model-selection; vii) Hierarchical Bayesian models; viii) open-session to be determined by pre-workshop participant outreach (spatial capture-recapture, cloud-computing and high-performance computing, multi-event).

Please visit HERE for further detail on the workshop.

Lars Bejder PhD
Lars Bejder PhD
Professor Lars Bejder PhD is the Research Leader of the Murdoch University Cetacean Research Unit.
Recommended Posts
Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.