Multiple processes drive genetic structure of humpback whale (Megaptera novaeangliae) populations across spatial scales

Abstract

Elucidating patterns of population structure for species with complex life histories, and disentangling the processes driving such patterns, remains a significant analytical challenge. Humpback whale (Megaptera novaeangliae) populations display complex genetic structures that have not been fully resolved at all spatial scales. We generated a data set of nuclear markers for 3575 samples spanning the seven breeding stocks and substocks found in the South Atlantic and western and northern Indian Oceans. For the total sample, and males and females separately, we assessed genetic diversity, tested for genetic differentiation between putative populations and isolation by distance, estimated the number of genetic clusters without a priori population information and estimated rates of gene flow using maximum-likelihood and Bayesian approaches. At the ocean basin scale, structure is governed by geographical distance (IBD P < 0.05) and female fidelity to breeding areas, in line with current understanding of the drivers of broadscale population structure. Consistent with previous studies, the Arabian Sea breeding stock was highly genetically differentiated (F 0.034-0.161; P < 0.01 for all comparisons). However, the breeding stock boundary between west South Africa and east Africa was more porous than expected based on genetic differentiation, cluster and geneflow analyses. Instances of male fidelity to breeding areas and relatively high rates of dispersal for females were also observed between the three substocks in the western Indian Ocean. The relationships between demographic units and current management boundaries may have ramifications for assessments of the status and continued protections of populations still in recovery from commercial whaling.

First circumglobal assessment of Southern Hemisphere humpback whale mitochondrial genetic variation and implications for management

“…Population and subpopulation comparisons confirm previous observations of complex patterns of genetic differentiation and gene flow, supporting assertions regarding the hierarchical nature of humpback whale population genetic structure (Rosenbaum et al 2009, Baker et al 2013, Kershaw et al 2017). In addition, the global context of the study provided an opportunity to consider potential drivers of circumpolar genetic structure, thus providing the large- Table 1 scale geographic context required for the interpretation of existing regional and local analyses.…”

“…8cs4f), indicating that population structure at this scale is driven by maternally directed site fidelity to breeding areas (Mackintosh 1942, Baker & Medrano-Gonzalez 2002, Baker et al 2013. The dominant influence of maternally directed site fidelity at the regional scale has also been confirmed by comparative assessments of bi parentally inherited nuclear markers (Ruegg et al 2013, Schmitt et al 2014, Kershaw et al 2017. Although significant genetic differentiation was observed at the regional scale, patterns of estimated gene flow between breeding stocks were found to be relatively high and asymmetric in all cases (Fig.…”

“…BSB1) substructure. Two substocks, BSB1, which breeds off Gabon, and BSB2, which seasonally feeds off and migrates past western South Africa (and which shows lower than expected direct connectivity with, and some genetic differentiation from, BSB1 based on the mtDNA control region and 10 microsatellites), are recognized (Rosenbaum et al 2009, Carvalho et al 2014, Kershaw et al 2017. Fine-scale temporal genetic substructure has been observed within BSB, consistent with the temporal segregation of migrating whales on the basis of age, sex, and reproductive status (Carvalho et al 2014).…”

“…2). At local spatial scales within breeding stocks, more complex patterns of population structure have been observed using nuclear markers (Kershaw et al 2017). A future comparative study employing microsatellite markers would therefore be useful to further elucidate circumpolar-scale population structure.…”

“…Such variation may influence the interpretation of the number of demographically distinct populations due to temporal sampling effects (Carvalho et al 2014). BSC is currently divided by the IWC into 4 substocks (BSC1−BSC4), although evidence of considerable migrant exchange between 3 of the substocks has emerged based on mtDNA control region and microsatellite data as well as satellite telemetry (Rosenbaum et al 2009, Ersts et al 2011, Fossette et al 2014, Kershaw et al 2017, suggesting relatively fluid exchange between the Mascarene region (BSC4), the Madagascar ridge (BSC3), and the Mozambique channel (BSC2). Gene flow between BSB and BSC is also apparent (Best et al 1998, Pomilla & Rosenbaum 2005, Rosenbaum et al 2009, Kershaw et al 2017, and further sampling is needed to explore the possibility of a westward directional bias (i.e.…”