Population structure of humpback whales in the western and central South Pacific Ocean as determined by vocal exchange among populations

Abstract

2 Short title: Vocally derived population structure Word count: 5,851 (first word of Abstract through last word of Literature Cited) Abstract For cetaceans population structure is traditionally determined by molecular genetics or photographically identified individuals. Acoustic data, however, has provided information on movement and population structure with less effort and cost than traditional methods in an array of taxa. Male humpback whales (Megaptera novaeangliae) produce a continually evolving vocal sexual display, or song, that is similar among all males in a population. The rapid cultural transmission (the transfer of information/behaviour between conspecifics through social learning) of different versions of this display between distinct but interconnected populations in the western and central South Pacific region presents a unique way to investigate population structure based on the movement dynamics of a song (acoustic) display. Using 11 years of data, we investigated an acoustically based population structure for the region by comparing stereotyped song sequences among populations and years using the Levenshtein distance technique, to group previously defined populations into (vocally based) clusters using the overall similarity of their song display in space and time. We identified the following distinct vocal clusters: western cluster, 1 population off eastern Australia; central cluster, populations around New Caledonia, Tonga, and American Samoa; and eastern region, either a single cluster or two clusters, one around the Cook Islands and the other off French Polynesia. These results are consistent with the hypothesis that each breeding aggregation represents a distinct population (each occupied a single, terminal node) in a metapopulation, similar to the current understanding of population structure based on genetic and photo-identification studies. However, the central vocal cluster had higher levels of song-sharing among populations than the other clusters, indicating that levels of vocal connectivity varied within the region. Our results demonstrate the utility and value of using culturally transmitted vocal patterns as a way of defining connectivity to infer population structure. We suggest 4 vocal patterns be incorporated by the International Whaling Commission in conjunction with traditional methods in the assessment of structure. 5

Humpback whale, <i>Megaptera novaeangliae</i>, song during the breeding season in the Gulf of Tribugá, Colombian Pacific

“…In this report, the authors have put forth the first description of stock G's song in the Gulf of Tribugá during the breeding season of 2013. This paper illustrates that the song is highly organized (e.g., Winn 1978, Payne et al 1983) and shared by individuals of the specific population (e.g., Winn and Winn 1978, Winn et al 1981, Guinee et al 1983, Payne and Payne 1985, Garland et al 2015. It follows the same ordered and consistent theme repetition described in other populations (Payne and McVay 1971, Winn and Winn 1978, Payne et al 1983, Payne and Payne 1985, Jenkins et al 1995, Darling and Sousa-Lima 2005.…”

“…Many studies within the song literature compare the song of different populations (Cerchio et al 2001, Darling and Sousa-Lima 2005, Murray et al 2012, Garland et al 2011, 2015. Future studies should seek to expand the current report on Colombia's stock G song via collaborative global comparisons with whale stocks in other locations.…”

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

“…The degree of differentiation between BSE2 (New Caledonia) and BSE3 (Tonga; F ST = 0.009, Φ ST = 0.006, p ≤ 0.05) is lower than that between BSE1 (eastern Australia) and BSE2 (F ST = 0.023, Φ ST = 0.009, p ≤ 0.05) or BSE1 and BSE3 (F ST = 0.030, Φ ST = 0.004, p ≤ 0.05; Table 2). Genetic and acoustic breaks have previously been identified that separate BSE1 and BSE2 from the rest of Oceania (Anderson 2013, Garland et al 2015; however, photo-identification matches between BSE1 and BSE2 have been reported (Garrigue et al 2002), indicating some level of exchange, and significant longitudinal displacement of whales from American Samoa, eastern Australia, and Tonga has also been observed (Robbins et al 2011, Franklin et al 2014.…”

“…Some subpopulations may represent multiple demographic populations (e.g. BSB2, BSC2, and BSE2), in some cases, for example, as a mixed-stock migratory corridor, and there are continuing efforts to appropriately define the boundaries of these populations on both breeding and feeding areas (Rosenbaum et al 2009, Carvalho et al 2014, Garland et al 2015. Failure to incorporate underlying genetic structure into population status assessments can result in significant biases through over-or under-representation of discrete population units in management goals.…”

“…However, fully resolving the boundaries of these stocks and subpopulations will require the integration of acous tic, genetic, photo-identification, and satellite tele metry data (Garland et al 2015).…”