| Citation: |
Per Alström, Pamela C. Rasmussen, Canwei Xia, Magnus Gelang, Yang Liu, Guoling Chen, Min Zhao, Yan Hao, Chao Zhao, Jian Zhao, Chengte Yao, James A. Eaton, Robert Hutchinson, Fumin Lei, Urban Olsson. 2018: Taxonomy of the White-browed Shortwing (Brachypteryx montana) complex on mainland Asia and Taiwan: an integrative approach supports recognition of three instead of one species. Avian Research, 9(1): 34. DOI: 10.1186/s40657-018-0125-6
|
The songs of oscine passerines (hereafter ‘songbirds’) are complex and among the most notable communication signals in the animal kingdom, owing to the great diversity of song systems in the over 4500 extant species (Catchpole and Slater, 2003; Williams, 2004). The organization of song systems are highly variable among species. For instance, Song Sparrows (Melospiza melodia) and Northern Mockingbirds (Mimus polyglottos) have repertoires composed of multiple songs (Wildenthal, 1965; Searcy et al., 1985), while other species, like Black-capped Chickadees (Poecile atricapillus) and Chipping Sparrows (Spizella passerina) typically sing only one song (Smith, 1991; Liu and Kroodsma, 2006). In songbirds, young birds (tutees) learn by copying the song of a conspecific adult (tutor) (Catchpole and Slater, 2003; Kroodsma, 2004), although the precise learning window and tutor varies from species to species (Beecher and Brenowitz, 2005; Wheelwright et al., 2008; Thomas et al., 2021). An intriguing consequence of learning from tutors is that imprecise copying or improvisation by the tutees can result in multiple song ‘types’ developing within a population over time (Slater and Ince, 1979; Nelson et al., 2004; Naguib and Riebel, 2014), and in rare cases where a tutee learns from multiple tutors, ‘hybrid’ songs may also arise (Heinemann, 1981; Bell et al., 2003). Intraspecific variation in song types is common among species with repertoires but is also observed in single song species. For instance, Chipping Sparrows and Kentucky Warblers (Geothlypis formosa) both sing a single song, but populations may contain multiple song types (Tsipoura and Morton, 1988; Liu and Kroodsma, 2006).
We often see song-type clustering in local populations (i.e., song neighborhoods), where individuals are more likely to share songs with direct neighbors than with distant birds within their population (e.g., Hensel et al., 2022). Song sharing is hypothesized to have several social and fitness advantages (McGregor, 1991; Beecher and Brenowitz, 2005); for example, in male Song Sparrows, the probability of survival and maintaining territory tenure throughout the breeding and non-breeding season is greater when the fraction of song sharing amongst neighboring males in the population is high (Wilson et al., 2000). In some cases, however, song sharing occurs no more often than expected by chance between neighbors and is largely attributed to song learning mechanisms and/or breeding or dispersing phenology (Slater, 1989), as is the case of Kentucky Warblers where song neighborhoods do not exist (Tsipoura and Morton, 1988). Furthermore, in some cases, patterns of song clustering within a population may be more ambiguous or unstable. In Great Tits (Parus major), for instance, some song types are shared between neighbors while others are not (McGregor and Krebs, 1982), while in Chipping Sparrows, the degree of song sharing varies between years (Liu and Kroodsma, 2006).
Macro-geographic variation in song types is also sometimes observed. Many species have song dialects that are characterized by sharp geographical boundaries that are thought to stem from a combination of individuality and conformity towards local song variants during learning (Lemon, 1975). Perhaps the most notable example is that of White-crowned Sparrows (Zonotrichia leucophrys), a species that sings a single song that varies acoustically across multiple populations with sharp boundaries between one song type and another (Marler and Tamura, 1962). In some cases, patterns of macro-geographic variation are not always distinct. In MacGillivray's Warblers (Geothlypis tolmiei), for instance, geographic patterns of song clustering are more gradual (i.e., clinal) in distribution (Pitocchelli et al., 2018). Furthermore, in other species, there are no clear boundaries between songs. Connecticut Warblers (Oporornis agilis), for instance, sing a single song, but multiple song types are distributed throughout their broader range without any clear clustering or boundaries (Hannah et al., 2020). The multi-part songs of some birds show different geographic patterns for different parts of the song such that one part may show macro-dialects and vary between populations while the other may form mirco-dialects within populations (e.g., Savannah Sparrows (Passerculus sandwichensis); Williams et al., 2019; Hensel et al., 2022).
Clearly there is great variation in the song system of songbirds, and among wood warblers (Family Parulidae) in particular, in which most species fit into two groups. The first group of warblers sing two distinct categories of songs, one that is generally used in mate attraction and the other used primarily in countersinging bouts (Spector, 1992). The second group of warblers sing a single primary song that serves both mate attraction and territory defense functions (Spector, 1992). In addition, many (but not all) birds in this second group sing an “extended” (or flight) song that includes notes of the primary song (Spector, 1992). Ovenbirds (Seiurus aurocapilla) belong to the second group and are a common ground-nesting wood warbler that are widely distributed across the woodlands of North America during the breeding season (Porneluzi et al., 2020). Ovenbirds have a single primary song, heard by humans as “teacher-teacher-teacher”, that is sung frequently throughout the day and a flight song, which is sung infrequently between dusk and dawn (Hann, 1937; Lein, 1981; Thompson et al., 2020). Ovenbirds are highly territorial, and they respond aggressively towards unfamiliar males’ songs (Weeden and Falls, 1959), suggesting that neighbor-stranger discrimination plays a critical role in the territory maintenance of this species.
The early work by Lein (1981) showed that each male had a song composed of a single repeated phrase and that multiple phrase types were found in local populations. Additionally, similar phrases were recorded in both New Hampshire and Massachusetts, USA. However, the extent of variation in song has not been described, i.e., how many songs Ovenbirds (as a species) sing and whether or not there are song neighborhoods within local populations and/or macro-dialects within their breeding range. Given that bird songs are used for both mate attraction and territory defense (Searcy and Andersson, 1986) and are therefore important signals for survival and reproduction (e.g., Lambrechts and Dhondt, 1986), understanding Ovenbird song structure, song type distribution, and song use within the context of neighbor interactions, will be informative for understanding the communication system of this species. In addition, given the diversity of warbler song systems, adding to our understanding of geographic structuring of warbler song will be informative for better understanding the factors associated with dialects in this large and diverse group of songbirds. With recent improvements in autonomous recording units (Shonfield and Bayne, 2017), and the advent of media sharing platforms such as eBird (Sullivan et al., 2014; ebird.org), it is now possible to investigate these questions at a comprehensive geographic scale.
In this study, we characterized Ovenbird song types in a dense northern Ontario population. Using a simulation analysis, we then asked whether neighboring Ovenbirds were more likely to have the same song type or whether song types were distributed randomly. We predicted that Ovenbirds do not have song neighborhoods as previous work hints at the fact that only some Ovenbirds share songs with their neighbors (Lein, 1981). Next, we characterized song types of Ovenbirds across Canada and the United States to see if the same song types found in our study population are found elsewhere in their breeding range (or if there are different types not found in our study site), and whether Ovenbirds exhibit macro (or clinal) dialects anywhere across their breeding range. Based on previous work that suggested Ovenbirds may not have dialects at least in the eastern portion of their range (Lein, 1981), we predicted that Ovenbirds would not exhibit macro-dialects anywhere across their breeding range or that dialects in Ovenbirds would be present across several hundred kilometers with song types clustered in eastern or western portions of the range.
We made focal recordings of Ovenbirds between May and July 2021 in the Hiawatha Highlands Conservation Area, which is located north of Sault Ste Marie, Ontario's urban core (46°35′17″ N, 84°16′52″ W; Fig. 1). We recorded singing males (n = 158) using a Marantz PMD661 solid-state recorder and a Sennheiser MKH-70 microphone at 44.1 kHz sampling rate and 16-bit depth in wav format. When recording individuals, we took GPS waypoints using a Garmin GPS Map 64s. We made notes regarding the direction and distance of any countersinging birds and then moved systematically between birds attempting to record all individuals and their neighbors. We made repeat recordings of most individuals to confirm territory locations (n = 636 recordings of 158 individuals). As part of another study, we banded 14 individuals with a unique combination three colour bands and a numbered USFW aluminum band. These birds were all re-recorded singing the same song type. The remaining 144 individuals were not banded and individuals singing the same song type at a previous recording location were assumed to be the same individual.
Spectrogram images of each Ovenbird's primary song were generated using Syrinx PC (John Burt, Seattle, WA). One example of the clearest (i.e., sharpest, little background noise) song from each bird was analyzed by both authors. Prior to analysis, each song was assigned a random number to eliminate the possibility of observer bias. Ovenbird song structure is fairly simple and variable enough to discern different types visually using spectrogram images (Lein, 1981; Ehnes and Foote, 2015) and experienced ornithologists can even discern the different types aurally (Weeden and Falls, 1959; Lein, 1981; JRF pers. obs.). We visually inspected the recordings separately and organized each song's phrase into distinct ‘types’. Song types were defined based on having repeated phrases composed of the same number of elements or notes, in the same order, and with the same general shape (as in Tsipoura and Morton, 1988; Janes and Ryker, 2016; Pitocchelli et al., 2018; Hannah et al., 2020). We identified hybrid songs (n = 2 males in local site), which we defined as songs that include 2 distinct phrases, where each phrase is repeated within the song and the singer switches among phrases one or several times within a song (Appendix Fig. S1). Both observers agreed on the grouping of general song types but there was disagreement on the grouping of some aberrant or faintly recorded songs (e.g., one observer thought they fit in a common type and the other that they were distinct). However, after consultation, consensus was reached on the characterization of the rare and aberrant song types.
In order to determine whether Ovenbirds establish territories near neighbors that sing the same song type, we first needed to determine which birds were neighbors. Ovenbirds hold small territories (0.20–1.82 ha; diameter = 50.5–152.5 m) and once they are established, they remain relatively unchanged throughout the breeding season (Hann, 1937). It is therefore possible to track countersinging neighbors in the field based on location. Ovenbird neighbors were defined as birds that were recorded within 50 m of each other and/or were observed countersinging close to one another (n = 145). In cases where we recorded two birds >50 m apart and did not observe countersinging, we identified them as possible neighbors (n = 39) but did not include them in our analysis of potential microdialects. Each bird had on average 2.0 recorded neighbors (SE = 0.07; range 1–4) in our sample. This measure is likely an underestimate of the true number of neighbors given that some birds could be missed during our census.
To determine if song types were uniformly distributed at our local study site, we used a Chi-square test. We performed a bootstrap analysis to determine the probability of neighboring pairs (n = 143) sharing the same song type more often than expected by chance. It is important to note that most birds had multiple neighbors (e.g., bird 1 might be neighbors with bird 2 and 3) and so these pairs are not fully independent. We excluded neighbor pairs when one of the birds sang a hybrid song composed of two phrase types (n = 2 males; n = 2 neighbor pairs). We reasoned that, for simplicity's sake, it is best to exclude these two songs from the analysis as we would have to include neighbor pairs twice for this sample. We randomly sampled (with replacement) 143 pairs 1000 times from our total sample of song types of n = 156 songs (again excluding the two hybrid songs) to calculate confidence intervals of the expected number of neighbors sharing the same song type (Janes and Ryker, 2016). Statistical analyses were performed in R v4.1.1 (R Core Team, 2021) using the “boot” package (Canty and Ripley, 2021).
We viewed and analyzed all Ovenbird spectrogram recordings from the eBird database that were recorded in Canada and the United States from May to July 2021 (n = 1079). We chose to explore songs from eBird (as opposed to other media sharing platforms) because it is the largest community science data sharing platform that is vetted by automated filters and regional experts to ensure quality control (Sullivan et al., 2014). Out of this sample, we excluded all recordings that: 1) did not include songs (calls); 2) were duplicates (same song type from different recordings with the same geographic coordinates); 3) were unclear (had poor signal-to-noise ratios and so could not be assigned to a song type); and 4) fell within the study period but were located outside of the breeding range. Some recordings included more than one Ovenbird and all individuals from each recording with song types that were identifiable were included. After calls, duplicates, and unusable songs were excluded, our sample size was 512 songs. As we noted previously, the two authors separately visually inspected and assigned each recording to a song type, consulting with each other in order to reach a consensus when there was disagreement about song type assignment. For example, for the Canadian eBird recordings, the two observers agreed on 120 of 123 songs that fit into the 10 local song types. Most cases of disagreement were for recordings of low quality.
We used a Chi-square test to determine whether 1) the song types shared between our local site from the eBird dataset were uniformly distributed and 2) the non-shared song types were uniformly distributed. We used a Fisher's Exact test (Mehta and Patel, 1983) to compare the relative proportions of shared song types between the local and breeding range samples.
In order to determine whether song types were clustered to form dialects within the entire Ovenbird breeding range, we created maps of all the song types in QGIS (v.3.22.3) for visual inspection using the recording coordinates from our eBird sample. In addition, for the song types found in both local and breeding range samples with three or more recordings, we calculated the distance between each pair of locations using the distance function in MATLAB v. 9.13.0 (R2022b; The Mathworks, Natick, MA, USA) and from these distances calculated the maximum distance between locations for each song type. We then randomized song types across the recording locations 500 times and calculated the distance between all points. From this randomization, we calculated the 95% confidence interval for the maximum distance of these randomized positions for each song type and compared this to the observed distances.
Among the 158 males recorded in Sault Ste. Marie, we identified 10 distinct song types (Fig. 2), five of which were relatively common and sung by 147 of 156 males that each sang a single phrase type. Two males sang hybrid songs composed of two phrases that were also both of the 10 types (see examples of hybrid songs from these 2 males in Appendix Fig. S1). These hybrid songs were not included in our statistical analyses. The two males singing hybrid songs were separated by ~250 m, and so were not neighbors. Songs were not uniformly distributed within the local population. (χ2 = 153.9, df = 9, P < 0.0001; Fig. 3A). Of 143 pairs of neighbors, 34 (23.8%) sang the same song type (Fig. 1), which is at a level that is significantly higher than expected by chance (mean = 27.9 pairs, 95% CI high = 28.2, low = 27.6).
Among the 498 single-phrase songs sampled from eBird from across the Ovenbird breeding range (excluding 14 hybrid songs; see below), we found 7 of the 10 song types we characterized in our local population (n = 421). The distribution of the seven Sault Ste. Marie song types found elsewhere across the entire breeding range was not uniform (χ2 = 217.4, df = 6, P < 0.0001; Fig. 3B). In addition, the relative proportions of these seven song types differed between the Sault Ste. Marie site and the rest of North America (χ2 = 62.2, P = 0.0002; Fig. 3). We observed no visual evidence of clustering of any song types across the Ovenbird breeding range (Fig. 4; Appendix Figs. S2 and S3). In addition, the maximum distance between locations where song types were recorded were equally likely to be closer than the randomized confidence interval (7 song types) or further apart (6 song types) suggesting that the extent of song type distribution across the breeding range has no consistent pattern (Appendix Table S1).
We found 24 additional song types (n = 77 birds; Fig. 5) that were not recorded in our study population. The distribution of these additional songs also differed significantly from uniform (χ2 = 229.3, df = 23, P < 0.0001; Fig. 6). Of those 24 song types, only two were common and sung by 39 males, while the others were each recorded fewer than 4 times with an approximately random distribution across the breeding range (Appendix Fig. S3). In this larger sample, we found 14 males with hybrid songs composed of two different phrases (which were of the 34 types already mentioned above; Appendix Fig. S2), but, as in the local sample, we did not include these in our analyses.
We used two song datasets collected from a local population and from across the breeding range to systematically characterize the distribution of Ovenbird song types and extend early work by Lein (1981) from recordings in New England, USA. We found some evidence for song clustering in Ovenbird neighborhoods but no clear evidence of macro-dialects.
Our local study site was ideal for investigating song-type clustering at a “neighborhood” scale because the density of birds in the population was high and we are confident that we sampled most of the individuals in the area. We found some evidence that Ovenbirds share the same song type with their neighbors more often than expected by chance, suggesting that song neighborhoods exist in local Ovenbird populations. However, despite occurring more than expected by chance, the majority of neighborhoods were composed of individuals that all sang different song types (Fig. 1), and most pairs of neighbors did not share songs (~76%). Whether sharing song types with neighbors carries an advantage for Ovenbirds in either competitive or mate attraction contexts is currently unknown, though it has been shown to be an advantage in species like Song Sparrows (Wilson et al., 2000). In a foundational study, Weeden and Falls (1959) discovered that Ovenbird males react more aggressively to unfamiliar songs than to songs of their (familiar) neighbors. Whether sharing song types with neighbors or having different song types relates to neighbor recognition remains to be explored in Ovenbirds. It would also be interesting to explore whether countersinging interactions differ among neighbors that share and do not share songs and to compare the songs of second year males establishing a first breeding territory to their neighbors.
Ovenbirds were thought to have random song distribution patterns within local populations (Lein, 1981), similar to Nashville (Leiothlypis ruficapilla) and Kentucky Warblers (Tsipoura and Morton, 1988; Janes and Ryker, 2016). While our findings here do not support these previous assumptions, it may be the case that only birds with particular song-types form song clusters (McGregor and Krebs, 1982), or that song sharing varies in intensity between years (Liu and Kroodsma, 2006). Additionally, for single-song species it may not always be possible for new recruits to settle near individuals with the same song type if territory vacancies are random each year. In some species with fixed song learning, song type selection can be adjusted in the first breeding season (Baptista and Morton, 1988) but the time of crystallization in Ovenbirds is not known. As such, clusters may form sporadically if they are advantageous but may not be ubiquitous, particularly if songs crystallize prior to the first breeding season and adult annual survival and territory tenure are high.
While all Ovenbirds sang only one song type, we found variation in song types sung at both our local site and across the Ovenbird breeding range. In our collective analyses, we identified the same song types described by Lein in 1981 (Appendix Table S2), as well as additional types not previously described in the literature. We found that the song types at our local study site (1–10) and those across the breeding range (1–6 and 8) were not uniformly distributed, and the proportion of song types (1–6 and 8) between both samples differed significantly, suggesting that our local study site song type distribution is not representative of the broader breeding range. Chipping Sparrow songs that are found across their range also differ in relative proportions among regions (Searfoss et al., 2020). Whether other local populations feature other more common song type variations would be interesting to explore and may help elucidate how learning and territory establishment influence song type selection.
However, given that sampling was uneven, some areas in the breeding range could show a similar distribution if sampled more intensively. The additional song types we characterized (11–34) were also not uniformly distributed as only two song types (11 in particular) dominated the soundscape. Why some song types are more common than others generally and why the common song types vary among populations is not known but may result from mechanisms of cultural evolution that either influence certain song types to persist or dissipate over time (Baker and Gammon, 2008; Wright et al., 2008).
Over a 40-year period (visually comparing spectrograms in Lein (1981) to those from this study; Appendix Table S2), Ovenbird song types appear to have persisted. In Lein (1981; Figs. 1 and 2), we identified four of the song types recorded both at our local study site and elsewhere in the breeding range (Fig. 2 exemplars 1, 6, 8 and 4) and two song types (Fig. 5 exemplars 11 and 23) recorded elsewhere in the breeding range. Only 4 of the 27 exemplars from Lein (1981) did not match song types we identified. Song types 1, 4, 8 and 11 were all relatively common. Common songs may persist in the population if they provide some fitness advantage to individuals who sing those variants. In Black-capped Chickadees for instance, individuals who vocalized a common variant of the “gargle” call tended to weigh more than individuals who vocalized less-common variants of the call (Baker and Gammon, 2008). In Indigo Buntings (Passerina cyanea), the probability of a song type persisting was associated with the number of males who sang the variant the previous year (Payne et al., 1988). Of the 34 song types observed in the combined local and breeding range sample (n = 654), two dominated (types 4 and 8) and were sung by ~48% of birds (~24% each). Five other song types (types 1, 2, 3, 5 and 11) were sung by 42% of birds (range ~4–11% males per type). The remaining 27 rarer types were sung by a combined 10% of birds (range ~0.1–1.8% of males per type). The uncommon variants could perhaps disappear from the population over time if they signal poor quality in males and are therefore unfavorable to females (Nowicki et al., 2001). If choice of song type does not have a fitness benefit for Ovenbirds, then we would expect to see a random change over time in song types as new songs are introduced into a population via miss copying and improvisation rather than persisting over generations (e.g., Ince et al., 1980). We would also then expect that at another time point, there may be different rare variants than those currently found in the population. Novel song types could also increase in frequency over time if favoured by females (e.g., Ju et al., 2019; Williams et al., 2022). However, because many species have an innate template that governs song type learning (Marler, 1970) perhaps there are also a limited number of possible variants that could emerge from the Ovenbird species-specific song template.
Our use of citizen science data from eBird allowed us to analyze a large, randomly sampled dataset of Ovenbird song types across their breeding range. We found no evidence of song clustering anywhere throughout the Ovenbird breeding range, suggesting that Ovenbirds likely do not have song dialects. Most common song type variants we characterized at our local study site were found throughout the breeding range, although the majority of datapoints were clearly sampled from eastern North America with fewer recordings from the western and northern portion of the species’ range. The additional song types we characterized (11–34) also exhibit a random distribution across the Ovenbird breeding range. Song type distribution is random in nature and similar to the geographic distribution exhibited by Prothonotary (Protonotaria citrea), Kentucky, Nashville, and Connecticut Warblers (Bryan et al., 1987; Tsipoura and Morton, 1988; Janes and Ryker, 2016; Hannah et al., 2020). Ovenbirds are thus similar to some other warblers where high intraspecific variation within local populations translates to a mosaic of song types distributed across the breeding range. One explanation for the relative similarity of song types on a continental scale in Ovenbirds and other warblers could be song learning outside the breeding range. Otter et al. (2020) suggest that wintering range overlap in White-throated Sparrows (Zonotrichia albicollis) may contribute to cultural evolution of song. Many warblers, including Ovenbirds, also show variation in local breeding density in response to insect outbreaks (reviewed by Venier and Holmes, 2010), which could also promote conserved song types or result in spread of songs if dispersal distances are great or adults move between breeding seasons as insect defoliation changes habitat (e.g., Holmes et al., 2009). Chipping sparrows may move between breeding seasons, which potentially contributes to local syllable diversity (Searfoss et al., 2020).
Additionally, we found hybrid songs, composed of two distinct phrases, at both the local and breeding-range scale suggesting that further song complexity in Ovenbirds exists. Hybrid song types have been observed in White-crowned Sparrows (Bell et al., 2003), Northern Cardinals (Cardinalis cardinalis; Lemon, 1975), and Winter Wrens (Troglodytes hiemalis; Kroodsma, 1980), and are thought to arise when individuals learn songs from multiple tutors (Heinemann, 1981). We surmise that young males “test out” songs (i.e., overproduction) they learn from multiple tutors before converging onto one of the locally adapted song types (e.g., Liu and Kroodsma, 2006; Peters and Nowicki, 2017), which is an intriguing notion as it hints at the possibility of Ovenbirds having plasticity in song choice (e.g., Nelson and Marler, 1994; Thomas et al., 2021). Interestingly, all birds with hybrid songs appeared to always include both phrase types within the same song. Hybrid song types do suggest that Ovenbirds may learn from multiple tutors and select song types via selective attrition that in some cases may result in hybrid songs where two phrase types are retained within a song. Perhaps they are rare (~2% of males) because, like uncommon song types, they may carry no advantage for singers despite the fact that they could be shared with multiple neighbors that sing different song types. In our local sample, we did not record any neighbors for one of the males with a hybrid song and for the other, neither of his phrases matched the song of his two neighbors.
We have systematically characterized Ovenbird song types at both a local and breeding-range scale. We found a variety of song types not previously described in the literature and determined that Ovenbirds do not have macro-dialects but show some evidence for song clustering within a local neighborhood. Collectively, our findings provide additional insights into Ovenbird song, expanding our understanding of the behavioural ecology of this species. Future research would benefit from studying how song choice influences reproductive success in Ovenbirds, and whether or not atypical (uncommon/hybrids) song types persist in the population and become more common over time. Including more samples from the northern and western part of the Ovenbird breeding range, which were under-represented in our sample, would help to confirm the presence/absence of dialects. Furthermore, investigating whether birds with certain song-types form song clusters, whether neighbors that share or do not share song-types have different countersinging patterns, and/or whether song sharing varies between years would ultimately provide more insight into the Ovenbird song system. Finally, this study could be repeated at intervals to examine range-wide patterns over multiple years (e.g., as seen recently in White-throated Sparrow dialects; Otter et al., 2020).
JRF collected the field data while PMJ drafted the manuscript. All authors shared in designing the project, conducting the acoustic analyses, and editing and revising the manuscript. All authors read and approved the final manuscript.
All research involving birds was carried out in accordance with the Algoma University policy on ethical principles of animal care and use and was approved by the Algoma University Animal Care Committee (2018-JFovenbird-R002). All handling of birds was carried out with banding and scientific permits to JRF from the Canadian Wildlife Service (permit 10832).
We hereby confirm that we have no conflicts of interest, financial, personal, or otherwise that would influence this work.
We would like to acknowledge that this research was conducted in the Robinson-Huron Treaty territory and that the land on which we worked is the traditional territory of the Anishnaabeg, specifically the Garden River and Batchewana First Nations, as well as Métis People. We would like to thank the Sault Ste Marie Conservation Authority for letting us use their land for our study. We thank Alex Sarno, Kaitlyn Plastino, and Adam Gouge for assisting with fieldwork. We thank The Macaulay Library at the Cornell Lab of Ornithology for providing the eBird dataset to create Fig. 5 (see appendix). We thank Brandon Schamp for calculating the distance between locations and randomizing song types across locations. We thank three anonymous reviewers and the editor for comments that improved our manuscript.
Supplementary data to this article can be found online at https://doi.org/10.1016/j.avrs.2023.100096.
Other data are available from the senior author (JRF) on reasonable request.
The following song files used in creating Fig. 5 were obtained from the Macaulay Library at the Cornell Lab of Ornithology: ML339252091; ML337542121; ML340809061; ML337141771; ML397635971; ML343793861; ML343553061; ML417746241; ML417715751; ML343092941; ML335411451; ML344335931; ML340191571; ML351086101; ML347809091; ML335973521; ML417749021; ML332874041; ML355820881; ML342752771; ML374142421; ML333921811; ML342960011; ML355257631.
|
Bioacoustics Research Program. Raven Pro: Interactive sound analysis software (version 1.5). Ithaca, New York: Cornell Lab of Ornithology; 2011.
|
|
Cheng T-H. A synopsis of the avifauna of China. Beijing: Science Press; 1987.
|
|
Clement P, Rose C. Robins and chats. London: Christopher Helm; 2015.
|
|
Collar NJ. 2005. Family Turdidae (thrushes). In: J. del Hoyo et al., editors. Handbook of the birds of the world, vol. 10. Cuckoo-shrikes to Thrushes. Barcelona: Lynx Edicions. p. 514–807.
|
|
del Hoyo J, Collar NJ. HBW and BirdLife international illustrated checklist of the birds of the world. Volume 2: Passerines. Barcelona: Lynx Edicions; 2016.
|
|
Dickinson EC, Christidis L (eds). The Howard & Moore complete checklist of the birds of the world. Vol. 2 Passerines. Fourth edition. Eastbourne: Aves Press; 2014.
|
|
Eaton JA, van Balen B, Brickle NW, Rheindt FE. Birds of the Indonesian Archipelago. Greater Sundas and Wallacea. Barcelona: Lynx Edicions; 2016.
|
|
Rasmussen PC, Anderton JC. Birds of South Asia: the Ripley guide. Barcelona: Lynx Edicions; 2005.
|
| 1. | Min Zhao, J. Gordon Burleigh, Urban Olsson, et al. A near-complete and time-calibrated phylogeny of the Old World flycatchers, robins and chats (Aves, Muscicapidae). Molecular Phylogenetics and Evolution, 2023, 178: 107646. DOI:10.1016/j.ympev.2022.107646 |
| 2. | Chentao Wei, George Sangster, Urban Olsson, et al. Cryptic species in a colorful genus: Integrative taxonomy of the bush robins (Aves, Muscicapidae, Tarsiger) suggests two overlooked species. Molecular Phylogenetics and Evolution, 2022, 175: 107580. DOI:10.1016/j.ympev.2022.107580 |
| 3. | Per Alström, Pamela C Rasmussen, Canwei Xia, et al. Morphology, vocalizations, and mitochondrial DNA suggest that the Graceful Prinia is two species. Ornithology, 2021, 138(2) DOI:10.1093/ornithology/ukab014 |
| 4. | Joseph A Tobias, Paul F Donald, Rob W Martin, et al. Performance of a points-based scoring system for assessing species limits in birds. Ornithology, 2021, 138(2) DOI:10.1093/ornithology/ukab016 |
| 5. | Benjamin G. Freeman, Matthew W. Pennell. The latitudinal taxonomy gradient. Trends in Ecology & Evolution, 2021, 36(9): 778. DOI:10.1016/j.tree.2021.05.003 |
| 6. | Arya Y. Yue, Elize Y. X. Ng, James A. Eaton, et al. Species limits in the Elegant Pitta (Pitta elegans) complex from Wallacea based on bioacoustic and morphometric analysis. Avian Research, 2020, 11(1) DOI:10.1186/s40657-020-00227-4 |
| 7. | Subir B. Shakya, M. Irham, Matthew L. Brady, et al. Observations on the relationships of some Sundaic passerine taxa (Aves: Passeriformes) previously unavailable for molecular phylogenetic study. Journal of Ornithology, 2020, 161(3): 651. DOI:10.1007/s10336-020-01766-9 |
| 8. | Alexey Opaev, Yulia Kolesnikova. Lack of habitat segregation and no interspecific territoriality in three syntopic cryptic species of the golden‐spectacled warblers Phylloscopus ( Seicercus ) burkii complex. Journal of Avian Biology, 2019, 50(11) DOI:10.1111/jav.02307 |
| 9. | Chyi Yin Gwee, James A Eaton, Kritika M Garg, et al. Cryptic diversity in Cyornis (Aves: Muscicapidae) jungle-flycatchers flagged by simple bioacoustic approaches. Zoological Journal of the Linnean Society, 2019, 186(3): 725. DOI:10.1093/zoolinnean/zlz003 |
Figures(6) / Tables(4)
DownLoad:
Email Alerts
RSS Feeds