The effect of isolation,fragmentation, and population bottlenecks on song structure of a Hawaiian honeycreeper

Little is known about how important social behaviors such as song vary within and among populations for any of the endemic Hawaiian honeycreepers. Habitat loss and non‐native diseases (e.g., avian malaria) have resulted in isolation and fragmentation of Hawaiian honeycreepers within primarily high elevation forests. In this study, we examined how isolation of Hawai'i ‘amakihi (Chlorodrepanis virens) populations within a fragmented landscape influences acoustic variability in song. In the last decade, small, isolated populations of disease tolerant ‘amakihi have been found within low elevation forests, allowing us to record ‘amakihi songs across a large elevational gradient (10–1800 m) that parallels disease susceptibility on Hawai'i island. To understand underlying differences among populations, we examined the role of geographic distance, elevation, and habitat structure on acoustic characteristics of ‘amakihi songs. We found that the acoustic characteristics of ‘amakihi songs and song‐type repertoires varied most strongly across an elevational gradient. Differences in ‘amakihi song types were primarily driven by less complex songs (e.g., fewer frequency changes, shorter songs) of individuals recorded at low elevation sites compared to mid and high elevation populations. The reduced complexity of ‘amakihi songs at low elevation sites is most likely shaped by the effects of habitat fragmentation and a disease‐driven population bottleneck associated with avian malaria, and maintained through isolation, localized song learning and sharing, and cultural drift. These results highlight how a non‐native disease through its influence on population demographics may have also indirectly played a role in shaping the acoustic characteristics of a species.     

Unsupervised classification to improve the quality of a bird song recording dataset

Open audio databases such as Xeno-Canto are widely used to build datasets to explore bird song repertoire or to train models for automatic bird sound classification by deep learning algorithms. However, such databases suffer from the fact that bird sounds are weakly labelled: a species name is attributed to each audio recording without timestamps that provide the temporal localization of the bird song of interest. Manual annotations can solve this issue, but they are time consuming, expert-dependent, and cannot run on large datasets. Another solution consists in using a labelling function that automatically segments audio recordings before assigning a label to each segmented audio sample. Although labelling functions were introduced to expedite strong label assignment, their classification performance remains mostly unknown. To address this issue and reduce label noise (wrong label assignment) in large bird song datasets, we introduce a data-centric novel labelling function composed of three successive steps: 1) time-frequency sound unit segmentation, 2) feature computation for each sound unit, and 3) classification of each sound unit as bird song or noise with either an unsupervised DBSCAN algorithm or the supervised BirdNET neural network. The labelling function was optimized, validated, and tested on the songs of 44 West-Palearctic common bird species. We first showed that the segmentation of bird songs alone aggregated from 10% to 83% of label noise depending on the species. We also demonstrated that our labelling function was able to significantly reduce the initial label noise present in the dataset by up to a factor of three. Finally, we discuss different opportunities to design suitable labelling functions to build high-quality animal vocalizations with minimum expert annotation effort.     

Anthropogenic noise decreases urban songbird diversity and may contribute to homogenization

More humans reside in urban areas than at any other time in history. Protected urban green spaces and transportation greenbelts support many species, but diversity in these areas is generally lower than in undeveloped landscapes. Habitat degradation and fragmentation contribute to lowered diversity and urban homogenization, but less is known about the role of anthropogenic noise. Songbirds are especially vulnerable to anthropogenic noise because they rely on acoustic signals for communication. Recent studies suggest that anthropogenic noise reduces the density and reproductive success of some bird species, but that species which vocalize at frequencies above those of anthropogenic noise are more likely to inhabit noisy areas. We hypothesize that anthropogenic noise is contributing to declines in urban diversity by reducing the abundance of select species in noisy areas, and that species with low‐frequency songs are those most likely to be affected. To examine this relationship, we calculated the noise‐associated change in overall species richness and in abundance for seven common songbird species. After accounting for variance due to vegetative differences, species richness and the abundance of three of seven species were reduced in noisier locations. Acoustic analysis revealed that minimum song frequency was highly predictive of a species' response to noise, with lower minimum song frequencies incurring greater noise‐associated reduction in abundance. These results suggest that anthropogenic noise affects some species independently of vegetative conditions, exacerbating the exclusion of some songbird species in otherwise suitable habitat. Minimum song frequency may provide a useful metric to predict how particular species will be affected by noise. In sum, mitigation of noise may enhance habitat suitability for many songbird species, especially for species with songs that include low‐frequency elements.     

Song characteristics track bill morphology along a gradient of urbanization in house finches (<Emphasis Type="Italic">Haemorhous mexicanus</Emphasis>)

Introduction

Urbanization can considerably impact animal ecology, evolution, and behavior. Among the new conditions that animals experience in cities is anthropogenic noise, which can limit the sound space available for animals to communicate using acoustic signals. Some urban bird species increase their song frequencies so that they can be heard above low-frequency background city noise. However, the ability to make such song modifications may be constrained by several morphological factors, including bill gape, size, and shape, thereby limiting the degree to which certain species can vocally adapt to urban settings. We examined the relationship between song characteristics and bill morphology in a species (the house finch, Haemorhous mexicanus) where both vocal performance and bill size are known to differ between city and rural animals.

Results

We found that bills were longer and narrower in more disturbed, urban areas. We observed an increase in minimum song frequency of urban birds, and we also found that the upper frequency limit of songs decreased in direct relation to bill morphology.

Conclusions

These findings are consistent with the hypothesis that birds with longer beaks and therefore longer vocal tracts sing songs with lower maximum frequencies because longer tubes have lower-frequency resonances. Thus, for the first time, we reveal dual constraints (one biotic, one abiotic) on the song frequency range of urban animals. Urban foraging pressures may additionally interact with the acoustic environment to shape bill traits and vocal performance.

     

White‐crowned sparrow males show immediate flexibility in song amplitude but not in song minimum frequency in response to changes in noise levels in the field

The soundscape acts as a selective agent on organisms that use acoustic signals to communicate. A number of studies document variation in structure, amplitude, or timing of signal production in correspondence with environmental noise levels thus supporting the hypothesis that organisms are changing their signaling behaviors to avoid masking. The time scale at which organisms respond is of particular interest. Signal structure may evolve across generations through processes such as cultural or genetic transmission. Individuals may also change their behavior during development (ontogenetic change) or in real time (i.e., immediate flexibility). These are not mutually exclusive mechanisms, and all must be investigated to understand how organisms respond to selection pressures from the soundscape. Previous work on white‐crowned sparrows (Zonotrichia leucophrys) found that males holding territories in louder areas tend to sing higher frequency songs and that both noise levels and song frequency have increased over time (30 years) in urban areas. These previous findings suggest that songs are changing across generations; however, it is not known if this species also exhibits immediate flexibility. Here, we conducted an exploratory, observational study to ask whether males change the minimum frequency of their song in response to immediate changes in noise levels. We also ask whether males sing louder, as increased minimum frequency may be physiologically linked to producing sound at higher amplitudes, in response to immediate changes in environmental noise. We found that territorial males adjust song amplitude but not minimum frequency in response to changes in environmental noise levels. Our results suggest that males do not show immediate flexibility in song minimum frequency, although experimental manipulations are needed to test this hypothesis further. Our work highlights the need to investigate multiple mechanisms of adaptive response to soundscapes.     

Metabolic and respiratory costs of increasing song amplitude in zebra finches

Bird song is a widely used model in the study of animal communication and sexual selection, and several song features have been shown to reflect the quality of the singer. Recent studies have demonstrated that song amplitude may be an honest signal of current condition in males and that females prefer high amplitude songs. In addition, birds raise the amplitude of their songs to communicate in noisy environments. Although it is generally assumed that louder song should be more costly to produce, there has been little empirical evidence to support this assumption. We tested the assumption by measuring oxygen consumption and respiratory patterns in adult male zebra finches (Taeniopygia guttata) singing at different amplitudes in different background noise conditions. As background noise levels increased, birds significantly increased the sound pressure level of their songs. We found that louder songs required significantly greater subsyringeal air sac pressure than quieter songs. Though increased pressure is probably achieved by increasing respiratory muscle activity, these increases did not correlate with measurable increases in oxygen consumption. In addition, we found that oxygen consumption increased in higher background noise, independent of singing behaviour. This observation supports previous research in mammals showing that high levels of environmental noise can induce physiological stress responses. While our study did not find that increasing vocal amplitude increased metabolic costs, further research is needed to determine whether there are other non-metabolic costs of singing louder or costs associated with chronic noise exposure.     

Rufous‐capped Warblers Basileuterus rufifrons show seasonal,temporal and annual variation in song use

In the majority of songbird species, males have repertoires of multiple song types used for mate attraction and territory defence. The wood‐warblers (family Parulidae) are a diverse family of songbirds in which males of many migratory species use different song types or patterns of song delivery (known as ‘singing modes’) depending on context. The vocal behaviour of most tropical resident warblers remains undescribed, although these species differ ecologically and behaviourally from migratory species, and may therefore differ in their vocal behaviour. We test whether male Rufous‐capped Warblers Basileuterus rufifrons use distinct singing modes by examining song structure and context‐dependent variation in their songs. We recorded multiple song bouts from 50 male Warblers in a Costa Rican population over 3 years to describe seasonal, diel and annual variation in song structure and vocal behaviour. We found that Rufous‐capped Warbler songs are complex, with many syllable types shared both within and between males’ repertoires. Males varied their song output depending on context: they sang long songs at a high rate at dawn and during the breeding season, and shortened songs in the presence of a vocalizing female mate. Unlike many migratory species, Rufous‐capped Warblers do not appear to have different singing modes; they did not change the song variants used or the pattern of song delivery according to time of day, season or female vocal activity. Our research provides the first detailed vocal analysis of any Basileuterus warbler species, and enhances our understanding of the evolution of repertoire specialization in tropical resident songbirds.     

Multiple signaling functions of song in a polymorphic species with alternative reproductive strategies

Vocal traits can be sexually selected to reflect male quality, but may also evolve to serve additional signaling functions. We used a long‐term dataset to examine the signaling potential of song in dimorphic white‐throated sparrows (Zonotrichia albicollis). We investigated whether song conveys multifaceted information about the vocalizing individual, including fitness, species identity, individual identity, and morph. We also evaluated whether song traits correlate differently with fitness in the two morphs, as the more promiscuous strategy of white, relative to tan, morph males might impose stronger sexual selection. Males with high song rates achieved higher lifetime reproductive success, and this pattern was driven by white morph males. In addition, males that sang songs with many notes survived longer, but this pattern was less robust. Thus, song traits reflect differences in fitness and may more strongly affect fitness in the white morph. Song frequency was unrelated to fitness, body size, or morph, but was individual specific and could signal individual identity. Songs of the two morphs displayed similar frequency ratios and bandwidths. However, tan morph males sang songs with longer first notes, fewer notes, and higher variability. Thus, song could be used in morph discrimination. Variation in frequency ratios between notes was low and could function in conspecific recognition, but pitch change dynamics did differ between four different song types observed. Our results support a multiple messages model for white‐throated sparrow song, in which different song traits communicate discrete information about the vocalizing individual.     

Effects of fragmentation and hybridization on geographical patterns of song variation in the endangered Yellow Cardinal Gubernatrix cristata

The song learning process may lead to small‐scale geographical variation in vocalizations of oscine birds. This variation can be further enhanced by the effects of habitat fragmentation or hybridization. Populations of the endangered Yellow Cardinal Gubernatrix cristata are now patchily distributed in the southern South American thorny shrubland forests and are small as a consequence of the pressure exerted by increased habitat transformation and illegal trade. We study the Yellow Cardinal's vocalizations throughout its distribution and assess how habitat characteristics and fragmentation, as well as hybridization with the Common Diuca Diuca diuca, have affected song patterns of different populations. We expected to find song differences among populations and songs to be more similar to those of the Common Diuca in areas where hybridization occurs. Multivariate analyses revealed significant differences in song between the four populations studied and confirmed that songs are more similar to those of Common Diuca where hybridization takes place. These results, in conjunction with genetic studies, can help to establish management units that preserve genetic and cultural variation in this endangered species.     

A Potential Metric of the Attractiveness of Bird Song to Humans

The prevalence of passerines (mostly oscines, or songbirds) in international bird trade suggests that the possession or production of a song that is attractive or desirable to people may contribute to the likelihood of a species being traded. Testing this is difficult because we lack a general and readily available metric that quantifies attractiveness of bird song to humans. We propose and validate such a metric, based on the number of sound files lodged for a species on the Xeno‐Canto website ( www.xeno-canto.org ). Our hypothesis is that species with more attractive songs are likely to be recorded more often, and so be represented more often in this online bird sound resource, all else being equal. Using a sample of North American and European passerines, we show that song repertoire size and geographic range size are consistently related to the number of recordings on Xeno‐Canto. We use these results to derive a metric (the residuals of a model of the number of recordings in Xeno‐Canto as a function of geographic range size) that may identify songs that are attractive to humans. Bird species whose songs are known to have inspired classical music, including several well known for their songs (e.g. common nightingale, European blackbird), have higher values of the metric than those that have not been referenced in classical music. The metric may help explain which bird species are present in trade, and so contribute to studies of invasion and conservation biology.     

Differentiation of Mating Vocalizations in Birds: Acoustic Features in Mainland and Island Populations and Evidence of Habitat-Dependent Selection on Songs

Local environments can act as selective agents on some characteristics of birds’ songs, whereas other song traits may not reflect local genetic adaptation. Geographic variation in songs of two Australian bird species (red‐capped robins Petroica goodenovii, western gerygones Gerygone fusca) was studied to examine one component of the ‘habitat‐dependent selection’ hypothesis. This hypothesis suggests that: (1) the detailed spectral characteristics of male songs are an evolved response to local habitat conditions affecting signal propagation and detection and (2) parallel evolution of other fitness traits sets up the potential for assortative mating by female choice. To examine the first part of the hypothesis, I made comparisons among widespread mainland populations and an island population using two levels of analysis: a typological analysis of song morphology (phonology: notes, syllables, syntax, temporal pattern, repertoires) and a spectral analysis of acoustic characteristics of songs (mean frequency, Wiener entropy, frequency modulation) using an automated procedure of feature extraction (Sound Analysis Pro). Spectral analysis was also used to extract values of the same acoustic features from the background sound environment of each recorded population. The typological analysis revealed no differences among mainland populations of either species, but large differences between mainland songs and those on the island. In contrast, the spectral analysis revealed acoustic divergence among populations, both mainland and island. For both species, Wiener entropy of songs correlated negatively with that of the ambient sound environment, consistent with predictions of the habitat‐dependent selection hypothesis of environmental selection on signal design.     

Exposure to noise pollution across North American passerines supports the noise filter hypothesis

The noise filter hypothesis predicts that species using higher sound frequencies should be more tolerant of noise pollution, because anthropogenic noise is more intense at low frequencies. Recent work analysed continental‐scale data on anthropogenic noise across the USA and found that passerine species inhabiting more noise‐polluted areas do not have higher peak song frequency but have more complex songs. However, this metric of song complexity is of ambiguous interpretation, because it can indicate either diverse syllables or a larger frequency bandwidth. In the latter case, the finding would support the noise filter hypothesis, because larger frequency bandwidths mean that more sound energy spreads to frequencies that are less masked by anthropogenic noise. We reanalysed how passerine song predicts exposure to noise using a more thorough dataset of acoustic song measurements, and showed that it is large frequency bandwidths, rather than diverse syllables, that predict the exposure of species to noise pollution. Given that larger bandwidths often encompass higher maximum frequencies, which are less masked by anthropogenic noise, our result suggests that tolerance to noise pollution might depend mostly on having the high‐frequency parts of song little masked by noise, thus preventing acoustic communication from going entirely unnoticed at long distances.     

When do trade‐offs occur? The roles of energy constraints and trait flexibility in bushcricket populations

In many animal species, the expression of sexually selected traits is negatively correlated with traits associated with survival such as immune function, a relationship termed a ‘trade‐off’. But an alternative in which sexually selected traits are positively correlated with survival traits is also widespread. We propose that the nature of intertrait relationships is largely determined by overall energy expenditure, energy availability and trait flexibility, with trade‐offs expected when individuals are subject to energy constraints. We tested this hypothesis in Ephippiger diurnus, a European bushcricket in which males are distinguished by two prominent sexually selected traits, acoustic calls and a large spermatophore transferred to the female at mating, and where immune function may be critical in survival. Ephippiger diurnus are distributed as small, isolated populations that are differentiated genetically and behaviourally. We analysed songs, spermatophores and the immune function in male individuals from eight populations spanning a range of song types. As predicted, we only found trade‐offs in those populations that expended more energy on song and were less flexible in their ability to adjust that expenditure. Ultimately, energy constraints and resulting trade‐offs may limit the evolution of song exaggeration in E. diurnus populations broadcasting long calls comprised of multiple ‘syllables’.     

Experimental evidence that distinct song phrases in the Grey‐cheeked Fulvetta Alcippe morrisonia permit species and local dialect recognition

Birds with small song repertoires have a limited number of song types which may serve multiple functions. The Grey‐cheeked Fulvetta Alcippe morrisonia is a non‐migratory species of lowland forests in Taiwan. Its song consists of two distinct phrases: a whistled phrase and a harmonic one. Each individual usually sings only one type of whistled phrase, and the geographical patterns of songs can be distinguished by the motif of this phrase. We proposed a dual adaptation hypothesis for the functions of these two phrases. Playback experiments including six sound stimuli (familiar whistled phrases, familiar harmonic phrases familiar complete songs, foreign whistled phrases, foreign harmonic phrases and foreign complete songs) were conducted at 12 sites in the Shoushan Nature Park of Kaohsiung, Taiwan. Grey‐cheeked Fulvettas came closer to the playback speaker and increased counter‐singing more in response to the familiar whistled phrase (local dialect) than to a foreign whistled phrase (distant dialect). However, birds responded equally to the playback of harmonic stimuli from familiar and foreign sources. We suggest that in this species, whistled phrases are used for local recognition and harmonic phrases are used for species recognition in short‐range communication.     

Male and female rufous‐and‐white wrens do not match song types with same‐sex rivals during simulated territorial intrusions

In birds with song repertoires, song‐type matching occurs when an individual responds to another individual's song by producing the same song type. Song‐type matching has been described in multiple bird species and a growing body of evidence suggests that song‐type matching may serve as a conventional signal of aggression, particularly in male birds in the temperate zone. Few studies have investigated song‐type matching in tropical birds or female birds, in spite of the fact that avian biodiversity is highest in the tropics, that female song is widespread in the tropics, and that female song is the ancestral state among songbirds. In this study of rufous‐and‐white wrens Thryophilus rufalbus, a resident neotropical songbird where both sexes sing, we presented territorial males and females with playback that simulated a territorial rival producing shared and unshared songs. In response, both males and females sang matched song types at levels statistically equal to levels expected by chance. Furthermore, males and females exhibited similar levels of aggression and similar vocal behaviours in response to playback of both shared and unshared songs. These results indicate that rufous‐and‐white wrens do not use song‐type matching in territorial conflicts as a conventional signal of aggression. We discuss alternative hypotheses for the function of song‐type sharing in tropical birds. In particular, we point out that shared songs may play an important role in intra‐pair communication, especially for birds where males and females combine their songs in vocal duets, and this may supersede the function of song‐type matching in some tropical birds.     

Experimental Evidence for Evolved Tolerance to Avian Malaria in a Wild Population of Low Elevation Hawai‘i ‘Amakihi (Hemignathus virens)

Introduced vector-borne diseases, particularly avian malaria (Plasmodium relictum) and avian pox virus (Avipoxvirus spp.), continue to play significant roles in the decline and extinction of native forest birds in the Hawaiian Islands. Hawaiian honeycreepers are particularly susceptible to avian malaria and have survived into this century largely because of persistence of high elevation refugia on Kaua‘i, Maui, and Hawai‘i Islands, where transmission is limited by cool temperatures. The long term stability of these refugia is increasingly threatened by warming trends associated with global climate change. Since cost effective and practical methods of vector control in many of these remote, rugged areas are lacking, adaptation through processes of natural selection may be the best long-term hope for recovery of many of these species. We document emergence of tolerance rather than resistance to avian malaria in a recent, rapidly expanding low elevation population of Hawai‘i ‘Amakihi (Hemignathus virens) on the island of Hawai‘i. Experimentally infected low elevation birds had lower mortality, lower reticulocyte counts during recovery from acute infection, lower weight loss, and no declines in food consumption relative to experimentally infected high elevation Hawai‘i ‘Amakihi in spite of similar intensities of infection. Emergence of this population provides an exceptional opportunity for determining physiological mechanisms and genetic markers associated with malaria tolerance that can be used to evaluate whether other, more threatened species have the capacity to adapt to this disease.     

Changes in Performance of Shared and Unshared Songs Within and Between Years in the White‐crowned Sparrow

Song of passerine birds is one of the few animal signals that is learned and that improves with practice. Vocal practice is crucial early in life to perfect a song imitation, but it also occurs throughout life and may continue to improve aspects of song performance. Differences in song performance among males that share song types, that is sing structurally similar songs may be particularly salient to receivers. We here test the hypothesis that aspects of song performance improve in a songbird species that deletes song types from its repertoire early in the first breeding season to share their final single song type with territorial neighbours. Over 3 yrs, we recorded songs in a population of Puget Sound white‐crowned sparrows Zonotrichia leucophrys pugetensis and measured percentage peak performance and consistency thereof in all of the song types in each male's repertoire. We found that within the first year on territory, percentage peak performance was higher in shared than unshared songs but did not change from first to second recording. Contrary to the hypothesis that song performance improves with age, song performance declined from the first to the second year. Our results support the hypothesis that high‐performance singers share songs. We did not find support for song performance improving within or between years, like it does in some other songbird species.     

Larval habitat for the avian malaria vector Culex quinquefasciatus (Diptera: Culicidae) in altered mid‐elevation mesic‐dry forests in Hawai'i

Effective management of avian malaria (Plasmodium relictum) in Hawai'i's endemic honeycreepers (Drepanidinae) requires the identification and subsequent reduction or treatment of larval habitat for the mosquito vector, Culex quinquefasciatus (Diptera: Culicidae). We conducted ground surveys, treehole surveys, and helicopter aerial surveys from 2001–2003 to identify all potential larval mosquito habitat within two 100+ ha mesic‐dry forest study sites in Hawai'i Volcanoes National Park, Hawai'i; ‘Ainahou Ranch and Mauna Loa Strip Road. At ‘Ainahou Ranch, anthropogenic sites (43%) were more likely to contain mosquitoes than naturally occurring (8%) sites. Larvae of Cx. quinquefasciatus were predominately found in anthropogenic sites while Aedes albopictus larvae occurred less frequently in both anthropogenic sites and naturally‐occurring sites. Additionally, moderate‐size (≈20–22,000 liters) anthropogenic potential larval habitat had >50% probability of mosquito presence compared to larger‐ and smaller‐volume habitat (<50%). Less than 20% of trees surveyed at ‘Ainahou Ranch had treeholes and few mosquito larvae were detected. Aerial surveys at ‘Ainahou Ranch detected 56% (95% CI: 42–68%) of the potential larval habitat identified in ground surveys. At Mauna Loa Strip Road, Cx. quinquefasciatus larvae were only found in the rock holes of small intermittent stream drainages that made up 20% (5 of 25) of the total potential larval habitat. The volume of the potential larval habitat did not influence the probability of mosquito occurrence at Mauna Loa Strip Road. Our results suggest that Cx. quinquefasciatus abundance, and subsequently avian malaria, may be controlled by larval habitat reduction in the mesic‐dry landscapes of Hawai'i where anthropogenic sources predominate.     

Soft Song in Song Sparrows: Acoustic Structure and Implications for Signal Function

In many species of songbirds, males sometimes produce songs at distinctly lower amplitude than in normal singing. Depending on the species, these 'soft songs' may be sung in the context of female courtship, male–male aggression, or both. In song sparrows, males produce soft songs during aggressive interactions with other males, and the amount of soft song produced is the only singing behavior that can be used to reliably predict a subsequent attack by the singer. Although soft song is clearly an important signal in this species, little is known about the acoustic structure of soft song or about how that structure compares to the structure of normal 'broadcast song'. We recorded a large sample of soft songs and broadcast songs from 10 male song sparrows, and measured song amplitudes in the field while controlling the subject's distance to a calibrated microphone. We show that song sparrow males produce songs over a wide range of amplitudes, with soft songs in the range of 55–77 dB sound pressure level and broadcast songs in the range of 78–85 dB. We present evidence for two types of soft song: 'crystallized' soft songs that are broadcast repertoire song types sung at low amplitude, and 'warbled' soft songs that are not found in the broadcast repertoire. Although highly variable, warbled soft songs produced by individual birds could be grouped into song types based on spectrographic similarity. To our knowledge, a distinct repertoire of soft song types has not been previously reported for any songbird.