Low levels of sibship encourage use of larvae in western Atlantic bluefin tuna abundance estimation by close-kin mark-recapture

Abstract

Globally, tunas are among the most valuable fish stocks, but are also inherently difficult to monitor and assess. Samples of larvae of Western Atlantic bluefin tuna Thunnus thynnus (Linnaeus, 1758) from standardized annual surveys in the northern Gulf of Mexico provide a potential source of “offspring” for close-kin mark-recapture (CKMR) estimates of abundance. However, the spatial patchiness and highly skewed numbers of larvae per tow suggest sampled larvae may come from a small number of parents, compromising the precision of CKMR. We used high throughput genomic profiling to study sibship within and among larval tows from the 2016 standardized Gulf-wide survey compared to targeted sampling carried out in 2017. Full- and half-siblings were found within both years, with 12% of 156 samples in 2016 and 56% of 317 samples in 2017 having at least one sibling. There were also two pairs of cross cohort half-siblings. Targeted sampling increased the number of larvae collected per sampling event but resulted in a higher proportion of siblings. The combined effective sample size across both years was about 75% of the nominal size, indicating that Gulf of Mexico larval collections could be a suitable source of juveniles for CKMR in Western Atlantic bluefin tuna.

Close-kin mark-recapture informs critically endangered terrestrial mammal status

“…The kinship analysis showed peaks in the PLOD kinship statistic that aligned well with theoretical expectations and variability. However, the PLOD distribution was not as well spread as for other species, e.g., Atlantic bluefin tuna 43 . For CKMR analyses of other species, studies have used a variant of DArTSeq namely DArTCap, which gives much higher sequencing read depth per locus than DArTSeq, and lower genotyping error rates, which can improve the spread of the PLOD distribution and in turn discrimination of kinship classes.…”

“…Kin clusters were more apparent for kin with different mt-DNA haplotypes, which implies that the pairwise independence assumption of the pseudo-likelihood is unreliable for male-derived parameters. This does not bias the point estimates but does affect the variance estimates of the male parameters 10 , 43 . Variances may be underestimated based on the Hessian and profile likelihood methods and should be interpreted with more caution.…”

“…Hessian-based standard errors are appropriate for CKMR provided approximate independence holds (see Sect. 4. of Bravington et al 10 and Supplement of McDowell et al 43 ). One heuristic for assessing approximate independence outlined in Bravington et al 10 inspects the ratio of kin-triads (within which the pairwise comparisons are clearly not independent) to kin-pairs.…”

“…However, pseudo-maximum-likelihood estimates are asymptotically unbiased, and variance estimates are accurate when sample sizes are small relative to the size of the population (see Sect. 4. of Bravington et al 10 for a discussion and Supplement of McDowell et al 43 ).…”

Sample contamination explains evidence of close kin proximity in yellowfin tuna (Thunnus albacares) in the Western and Central Pacific Ocean

“…That said, genetic assessments of larval or early juvenile stages may reveal more structure due to limited mobility and/or mixing opportunities during those life phases. Indeed, an excess of sibship was found among 5-11 day old larvae of the related Atlantic bluefin tuna samples by McDowell et al (2022). Further representative sampling of larvae and juveniles has long been identified as a research need to improve our understanding of tropical tuna population structure, and plans are afoot to obtain larval samples during upcoming collaborative research cruises.…”