Bruno Leroy

Age, growth and preliminary estimates of maturity of bigeye tuna, Thunnus obesus, in the Australian region

Biological parameters such as age, growth and age (or size) at maturity are vital for accurate stock assessments and management plans to ensure that fisheries develop sustainably. Despite this, very few validated age studies have been conducted for large tropical pelagic species within the Australian region. Age and growth parameters were estimated for bigeye tuna, Thunnus obesus (Lowe, 1839), sampled from longline fisheries in the Australian region using validated techniques based on counts of annual increments. Poor increment clarity reduced the number of otoliths included in the final analysis to only 50% of the 3200 selected for reading (39–178-cm fork length). Microincrement analysis confirmed the position of the first two annual increments in these otoliths.A maximum age of 16 years was obtained, but over 80% of fish in the Australian catch were <5 years old. Growth is most rapid in the first few years of life and asymptotic length is reached at about age 9 to 10 years. The von Bertalanffy growth parameters were estimated at L∞ =169.09, k=0.238, and to=−1.706 for the south-west Pacific Ocean and L∞ =178.41, k=0.176, and to=−2.500 for the eastern Indian Ocean. These parameters were significantly different, suggesting that there is little mixing between populations in the Pacific and Indian Oceans. Length at 50% maturity for females sampled in northern Queensland was estimated to be 102.4-cm fork length.

Methodological assessment of 2b-RAD genotyping technique for population structure inferences in yellowfin tuna (Thunnus albacares)

Global population genetic structure of yellowfin tuna (Thunnus albacares) is still poorly understood despite its relevance for the tuna fishery industry. Low levels of genetic differentiation among oceans speak in favour of the existence of a single panmictic population worldwide of this highly migratory fish. However, recent studies indicated genetic structuring at a much smaller geographic scales than previously considered, pointing out that YFT population genetic structure has not been properly assessed so far. In this study, we demonstrated for the first time, the utility of 2b-RAD genotyping technique for investigating population genetic diversity and differentiation in high gene-flow species. Running de novo pipeline in Stacks, a total of 6772 high-quality genome-wide SNPs were identified across Atlantic, Indian and Pacific population samples representing all major distribution areas. Preliminary analyses showed shallow but significant population structure among oceans (FST=0.0273; P-value<0.01). Discriminant Analysis of Principal Components endorsed the presence of genetically discrete yellowfin tuna populations among three oceanic pools. Although such evidence needs to be corroborated by increasing sample size, these results showed the efficiency of this genotyping technique in assessing genetic divergence in a marine fish with high dispersal potential.

Vertical Behavior and the Observation of FAD Effects on Tropical Tuna in the Warm-Pool of the Western Pacific Ocean

Archival and acoustic tagging were carried out in the exclusive economic zone (EEZ) of Papua New Guinea (PNG) in the western Pacific Ocean during 2006–07 to investigate the vertical behavior of tropical tuna found in association with large arrays of anchored FADs. Industrialized purse-seine fishing on anchored FADs has existed in the PNG EEZ for more than a decade. Archival tags were implanted in bigeye (n = 40; length 40–90 cm FL) and yellowfin (214; length 42–126 cm FL) tuna in the Bismarck and Solomon Seas. Acoustic tags were released in the same areas in 195 tuna (10 bigeye, 116 yellowfin, 69 skipjack). In addition, 27 tuna (eight bigeye, 19 yellowfin) received both an archival and an acoustic tag. Archival tag data from 32 recaptures were categorized into the three distinct vertical behavior modes for bigeye, and the three distinct modes for yellowfin that have been described in the published literature. The depth distribution for each of the categories was then calculated to examine potential vulnerability to industrial purse-seine capture in this region. A region-specific analysis was considered important as oceanographic conditions in this region are distinctly different to the conditions reported in the published literature from other locations. Analysis of acoustic data reveals short residence times at FADs and strong school cohesion. Vertical behavior of skipjack, yellowfin and bigeye tuna that were simultaneously present at the same FAD, as determined by depth transmitting acoustic tags, suggested some vertical separation of these species. However, there was a high degree of depth overlap, particularly during early morning hours when purse seining on floating objects normally occurs in this region. This overlap limits the potential for targeting particular species or size classes of tuna through fishing depth selection. This observation was confirmed from the archival tag depth records for yellowfin and bigeye during the same time period regardless of behavior type exhibited. The recapture of bigeye and yellowfin tuna implanted with both acoustic and archival tags allowed the observation of the natural vertical behavior of these fish when they were close to anchored FADs equipped with acoustic receivers. Occupation of shallow (<100 m) depths over a 24 h period was identified as the dominant behavior exhibited on FADs.

Removing Bias in Latitude Estimated from Solar Irradiance Time Series

Latitude estimates from light recorded by electronic data storage tags typically contain large errors, particularly at times near the equinoxes. We employ error propagation analysis to the fundamental equation relating latitude to solar elevation and time of day. Large latitude errors are caused by mathematical “amplification” of small errors in relating solar irradiance to solar elevation. Furthermore, the sign of these errors is such that the estimated latitude is badly biased. This analysis leads directly to a method of removing systematic error (bias) from latitude estimates in state-space track reconstruction models. Preliminary implementation of this method effectively removes all bias from latitude estimated from solar irradiance recorded by archival tags deployed both on moorings and on freely swimming tuna.

Objective classification of latent behavioral states in bio‐logging data using multivariate‐normal hidden Markov models

Analysis of complex time-series data from ecological system study requires quantitative tools for objective description and classification. These tools must take into account largely ignored problems of bias in manual classification, autocorrelation, and noise. Here we describe a method using existing estimation techniques for multivariate-normal hidden Markov models (HMMs) to develop such a classification. We use high-resolution behavioral data from bio-loggers attached to free-roaming pelagic tuna as an example. Observed patterns are assumed to be generated by an unseen Markov process that switches between several multivariate-normal distributions. Our approach is assessed in two parts. The first uses simulation experiments, from which the ability of the HMM to estimate known parameter values is examined using artificial time series of data consistent with hypotheses about pelagic predator foraging ecology. The second is the application to time series of continuous vertical movement data from yellowfin and bigeye tuna taken from tuna tagging experiments. These data were compressed into summary metrics capturing the variation of patterns in diving behavior and formed into a multivariate time series used to estimate a HMM. Each observation was associated with covariate information incorporating the effect of day and night on behavioral switching. Known parameter values were well recovered by the HMMs in our simulation experiments, resulting in mean correct classification rates of 90-97%, although some variance-covariance parameters were estimated less accurately. HMMs with two distinct behavioral states were selected for every time series of real tuna data, predicting a shallow warm state, which was similar across all individuals, and a deep colder state, which was more variable. Marked diurnal behavioral switching was predicted, consistent with many previous empirical studies on tuna. HMMs provide easily interpretable models for the objective classification of many different types of noisy autocorrelated data, as typically found across a range of ecological systems. Summarizing time-series data into a multivariate assemblage of dimensions relevant to the desired classification provides a means to examine these data in an appropriate behavioral space. We discuss how outputs of these models can be applied to bio-logging and other imperfect behavioral data, providing easily interpretable models for hypothesis testing.

Revisiting the vulnerability of juvenile bigeye (Thunnus obesus) and yellowfin (T. albacares) tuna caught by purse-seine fisheries while associating with surface waters and floating objects

Tuna fisheries catch over three million tonnes of skipjack tuna (Katsuwonus pelamis) each year, the majority of which come from purse-seine vessels targeting fish associated with man-made fish aggregating devices (FADs). A significant challenge for fisheries management is to maximize the efficiency of skipjack tuna catches whilst minimizing the bycatch of small and immature bigeye (Thunnus obesus) and yellowfin (T. albacares) tuna, for which long-term sustainability is uncertain in 75% of the world’s stocks. To better manage the issues common with this fishing method, an improved understanding of tuna behaviour around FADs is necessary. We probabilistically classified the vertical behavioural patterns of 50 bigeye and 35 yellowfin tuna (mean fork length 72cm and 70cm, respectively) electronically tagged throughout the western and central Pacific Ocean into shallow and deep states, using a state-space modelling approach. The occurrence of surface-association behaviours, defined as an individual remaining in a shallow state for 24-hours, was examined in relation to known capture events and FAD density. In general, surface-association events for both species were short and lasted on average less than three days, although events as long as 28 days were observed, and were more common in yellowfin when in archipelagic waters. Events were longest immediately following tagging in 62% and 17% of bigeye and yellowfin, respectively. Surface-association behaviour was not generally estimated just prior to recapture, being either non-existent or shorter than two days for 85% of bigeye and 74% of yellowfin. Current management measures in purse-seine tuna fisheries involve periodic or spatial closures for FAD use. If the chief benefit to purse-seine fishers of surface-association around floating objects is in locating schools in horizontal space at short-term time-scales, rather than holding fish near the surface for extended periods, controlling the number of sets made on FADs should be explored further as an additional management tool.