More of the same: allopatric humpback whale populations share acoustic repertoire

Abstract

BackgroundHumpback whales (Megaptera novaeangliae) are a widespread, vocal baleen whale best known for producing song, a complex, repetitive, geographically distinct acoustic signal sung by males, predominantly in a breeding context. Humpback whales worldwide also produce non-song vocalizations (“calls”) throughout their migratory range, some of which are stable across generations.MethodsWe looked for evidence that temporally stable call types are shared by two allopatric humpback whale populations while on their northern hemisphere foraging grounds in order to test the hypothesis that some calls, in strong contrast to song, are innate within the humpback whale acoustic repertoire.ResultsDespite being geographically and genetically distinct populations, humpback whales in Southeast Alaska (North Pacific Ocean) share at least five call types with conspecifics in Massachusetts Bay (North Atlantic Ocean).DiscussionThis study is the first to identify call types shared by allopatric populations, and provides evidence that some call types may be innate.

Humpback whale call repertoire on a northeastern Newfoundland foraging ground

“…Although gradation in the humpback whale call repertoire likely plays a key role in classification success, it also seems that computer-generated or manual measurements of variables did not always capture variation apparent during AV classification (Chmelnitsky & Ferguson, 2012;Janik, 1999). The growls and whups present an example of this (Figures 2 and 5; Fournet et al, 2015;Fournet et al, 2018c), whereby the upsweep was likely not captured by manually extracted or computer-generated measurements. Not capturing the variation, compounded with small sample sizes (Fournet et al, 2015(Fournet et al, , 2018bIndeck et al, 2020), likely also contributed to the lower classification success of the rumbles and the growl-moans (Table 5).…”

“…Following AV classification, classification and regression tree (CART) and random forest (RF) analyses were run with the 16 extracted variables (Table 1) to assess classification to broad classes and to call types. CART and RF analyses have emerged as preferred methods for supervised classification in humpback whale repertoire studies (Fournet et al, 2018b(Fournet et al, , 2018cGarland et al, 2012;Rekdahl et al, 2013Rekdahl et al, , 2017. These methods are preferred because they are minimally affected by outliers, nonnormality, nonindependent data, correlated variables, and sample size (Armitage & Ober, 2010;Breiman, 2001;Breiman et al, 1984), all of which are common in humpback whale repertoire studies.…”

“…The CART and RF were run using the rpart and randomforest packages (R version 3.5.0; Liaw & Wiener, 2002;R Core Team, 2016). The CART and RF were initially run by broad class (LF, P, HF; hereafter "broad class") and then with all call types together, regardless of call class (hereafter "pooled"), and finally with the call types separated by broad class, where only call types within each class were included (hereafter "within class"; Fournet et al, 2018bFournet et al, , 2018cRekdahl et al, 2013Rekdahl et al, , 2017. The Gini index was used in the CART analysis to determine the "goodness of split" at each node (Breiman et al, 1984) and terminal nodes were set to a minimum size of 10 samples.…”

Exploring movement patterns and changing distributions of baleen whales in the western North Atlantic using a decade of passive acoustic data

“…Non‐song vocalizations, or social sounds, vary with some calls being similar to those found in song while others are completely different. These non‐song vocalizations are produced by both sexes and across ages (Dunlop, Cato, & Noad, 2008; Fournet, Jacobsen, Gabriele, Mellinger, & Klinck, 2018; Stimpert, 2010; Stimpert et al., 2011). Given that humpback whale song can be highly variable between years, the call library described in Baumgartner and Mussoline (2011) was expanded and improved for this analysis to include a wider variety of examples of humpback whale vocalizations, across all years, to increase detection probability (Table S1).…”

Detections of humpback whale (Megaptera novaeangliae) vocalizations on an acoustic sensor in the New York Bight

“…Calls are likely an attempt to communicate with nearby groups, while songs are a broadcast signal (Dunlop et al, 2008). A number of call types have been shown to be temporally stable and do not change year-to-year the way songs do, allowing for easier detection and analysis using automated detection-classification approaches (Fournet, Jacobsen, et al, 2018;Rekdahl et al, 2013Rekdahl et al, , 2016. Because calls have a lower source level and are produced at a lower rate than song, they have a smaller detection range and also have a lower overall detection probability.…”