TinkerBird Project – 2024

Role of song and plumage colouration in species interaction

TinkerBird Project – 2024

Kube Yini

Introduction

Many animals use visual and auditory signals as a means of communication (Garcia & Favaro, 2017; Osorio & Vorobyev, 2008; Rosenthal & Ryan, 2000; Slabbekoorn & Smith, 2002). Visual and auditory signals such as song and plumage function in species recognition, territory defence, mate attraction, sexual selection and assessment of individual fitness (Delhey et al., 2017; Krebs et al., 1978; McGraw & Hill, 2000; Schmidt et al., 2007). The importance of these forms of communication goes beyond individual interactions, they also contribute to shaping the structure and dynamics of the entire community of animals(Badyaev et al., 2002; Mason et al., 2014; Santos et al., 2011). Many studies use these traits as models to understand their influence in shaping animal communication and enhancement in maintaining species boundaries (Kirschel et al., 2009; Mason & Bowie, 2020).

Birds may communicate through different vocal signals that include songs – often longer and more complex sounds used in territory defence and mate choice and calls – typically shorter simple sounds used in close proximity to receivers for contact purposes, alarm, and mobbing (Catchpole & Slater, 2003; Mason et al., 2017). Most passerines learn their song from the parental forms. However, some groups of birds such as non-oscines develop their song innately (ten Cate, 2021).

Visual signals, on the other hand, such as carotenoid-based plumage, serve multiple functions in sexual selection, particularly in species where visual cues are essential for mate recognition and attraction (Isaksson et al., 2007; Pike, 2017; Simons et al., 2014). In many birds, males exhibit elaborate and vibrant plumage which facilitates mate choice. Birds acquire carotenoids from their diet and displayed their colourful pigments in feathers and skin (Hill et al., 2023; Saks, McGraw, et al., 2003). These colours serve as an indicator of health (Saks, Ots, et al., 2003) and genetic quality (Lopes et al., 2016), as only birds with access to abundant, carotenoid-rich foods and good overall health can develop bright and vivid hues(Blount et al., 2003; Lozano, 1994). It is dietary yellow carotenoids that can be deposited onto integuments, but some taxa can convert these dietary carotenoids to ketocarotenoids through the action of an enzyme encoded by the gene CYP2J19 (Kirschel et al., 2020; Mundy et al., 2016; Twyman et al., 2018). Given that carotenoid-based colouration could potentially convey information about individual quality, it has often been assumed that yellow through red carotenoid-based colours might have a signalling function (Delhey et al., 2023; Hill, 2002; Willink & Wu, 2023).

Both bird song and feather colours derived from carotenoid pigments are strongly influenced by their environment(Brumm & Naguib, 2009; Parker, 1996; Slabbekoorn & Smith, 2002; Weaver et al., 2018). For instance, the vibrancy of colours is directly influenced by the availability of food resources, particularly those high in carotenoids (Y. Zhang et al., 2017). Therefore, a habitat with abundant nutrients creates an avenue for developing vivid colours, enhancing their attractiveness and signalling effect (Cooney et al., 2022). Similarly, bird song is hypothesised to be shaped by the environment. One mechanism widely supported is the acoustic adaptation hypothesis – which posits that acoustic signals in densely vegetated habitats tend to be more tonal and at lower frequencies, with a slower pace and simpler structure while in open habitats birds are expected to sing at higher frequencies (Ey & Fischer, 2009; Giuseppe Boncoraglio & Saino, 2007). In addition to the environment, bird song is predicted to be shaped by morphology (Derryberry et al., 2018). Specifically, larger birds have longer vocal tracts and therefore produce songs of lower frequency while birds with short vocal tracts produce higher frequency vocalisations (Hoese et al., 2000; Nowicki, 1987; Podos et al., 1995; Riede et al., 2006; Suthers & Zollinger, 2004). Avian body size and proportions are predicted to vary along an environmental gradient in accordance with Bergmann’s rule, one of the ecogeographic principles. This rule suggests that homeothermic animals within species and related species tend to have larger body sizes in colder climates at higher latitudes, while smaller body sizes are typical in warmer climates (Ashton, 2002; Salewski & Watt, 2017; Sebastianelli et al., 2022; Teplitsky et al., 2008). Nevertheless, studies performed to examine the effects of ecological factors on song and plumage colouration have produced equivocal results (Ballentine, 2006; Derryberry et al., 2018; Mikula et al., 2021).

Red-fronted and yellow-fronted tinkerbird are African barbets (Family: Lybiidae, Order: Piciformes) (Short & Horne, 2001) in which primarily males emit simple stereotypic songs that are genetically based (Kirschel et al., 2009; Sebastianelli et al., 2024) and differ in two vocal characters, frequency and pace (Nwankwo et al., 2018; Sebastianelli et al., 2024). Their range spans sub-Saharan Africa with red-fronted tinkerbird having a disjunct distribution in East and Southern Africa (Short & Horne, 2001). In Southern Africa, the two species have been documented to hybridise where their ranges overlap in north-eastern Eswatin i(Nwankwo et al., 2019) . Their common names reflect the carotenoid-based forecrown colour which is due to the deposition of dietary carotenoids and converted carotenoids respectively (Kirschel et al., 2020). The extent of hybridisation between them provides an opportunity to investigate the factors that may mediate a pattern of asymmetric assortative mating found between the two species(Sebastianelli et al., 2024). To what extent do song and carotenoid-based plumage character differences play in mediating reproductive isolation between the two species? Is there a preference for the fast-pace song of red-fronted tinkerbird over the slow-pace song of yellow-fronted tinkerbird? Or is there a preference for the red forecrown of red-fronted tinkerbird which is a result of conversion of dietary carotenoids to keto-carotenoids? If there is preference of either fast song and or the red colour, we expect that this could drive assortative mating among the two species, thus, promoting introgression of the alleles of red-fronted tinkerbird into yellow-fronted tinkerbird.

Also, to what extent do geographic and ecological factors shape song variation in yellow-fronted and red-fronted tinkerbird? Since the two species occupy different habitats, yellow-fronted tinkerbird mostly occurs in drier woodland savannah, while red-fronted tinkerbird prefers coastal and evergreen habitats (Short & Horne, 2001). Similarly, much plumage colouration can vary across landscapes, and it remains unclear to which extent, in addition to genetics, variation in carotenoid-based colour variation is governed by variations of carotenoid-rich diet across tinkerbirds environments.

To answer these questions, we conducted two independent studies. We investigated relative recognition of red-fronted and yellow-fronted tinkerbird song and plumage by conducting playback experiments with models. We included taxidermic mounts of the two species and 3D models (painted to resemble each species). Secondly, we examined the effect of genetic and ecological factors on song frequency to infer potential relationships between these factors and song frequency in red-fronted and yellow-fronted tinkerbird. Here, we report on the findings from the two independent studies and discuss further studies we aim to conduct going forward.

Figure 1: Samples and recordings of ringed and colour-ringed birds in Kube Yini. Also, shown are the playback experiments that were conducted.

INDEPENDENT PROJECTS

The role of song and plumage colouration on tinkerbird interactions at a local scale

We investigated whether yellow-fronted (YFT) or red-fronted tinkerbird (RFT) responded more strongly to conspecific or heterospecific song by conducting playback experiments in allopatric populations of YFT and RFT or the contact zone of the two species. Further, we tested whether the responding bird interacted more with the red-fronted (model with red forecrown) or yellow-fronted (model with yellow forecrown) tinkerbird model. We used alternatives of 3D printed models which avoid potential issues of pseudoreplication, but also taxidermic mount models that provided the added opportunity to replicate the air sac inflation of tinkerbirds when they sing. Something we achieved by manually inflating a balloon inserted inside the model and pumping it in synchrony with song playback. Both types of models were mounted onto servo motors that rotated the models side to side at a tempo equivalent to natural movement of tinkerbirds when they sing. The experiments were performed by first locating a singing bird in its territory. We then mounted the models on two tripods 3m apart on either side of the speaker – mounted on a tripod placed between the two models. We also mounted on a tripod a Canon EOSR7 camera for documenting the behaviour of the focal bird alongside audio recording and observer notes. For each experiment we alternated playing a synthetic playback of YFT and RFT for 10 minutes. After 10 minutes had elapsed, we observed the focal bird’s behaviour for a further minute. During the playback we recorded the a) latency to respond – the time taken for the focal bird to approach the playback experiment arena, b) time spent within 20 m of the experiment arena, and c) the closest approach distance within 20m of the experiment.

We first tested for differences in responses to the two songs in sympatry and allopatry, by fitting generalised linear mixed models using the glmmTMB package in R. We used closest approach distance, latency to approach and time spent within 20m, as a response in the models and playback (YFT or RFT), population (as a categorical factor consisting of RFT allopatry, YFT allopatry and the contact zone), and the interaction between them as fixed factors. Location nested in the subject territory of the focal bird and experiment number was entered as a random effect.

To test our second hypothesis that RFT will selectively respond strongly to its own song and interact only with RFT 3D model or RFT taxidermy, we fitted GLMMs using the closest distance to the model or speaker as a response variable, where the distance from either model or speaker is no more than 1.5m. The interaction of the focal bird and playback was entered as fixed effects. Model 1 (RFT model) and model 2 (YFT model) were entered as crossed random factors. We also tested the time spent with the speaker or the model by fitting a model that included the interaction of the focal bird and playback as fixed effects and model 1 and model 2 and location were entered as crossed random effect.

Samples and recordings obtained from Kube Yini

Since 2015, we have collected 29 blood and feather samples of red-fronted tinkerbird from Kube Yini Private Game Reserve. These include nine samples collected in December 2023. Apart from the samples, we recorded 23 songs, 12 of which from colour-ringed birds. Some samples have been sequenced using ddRAD sequencing and some with whole genome resequencing.

Results

Playback experiments

We conducted playback experiments in South Africa and Eswatini from 28^th November 2023 to 19^th December 2023. We conducted a total of 56 playback experiments, 26 of these were conducted in the allopatric population of yellow-fronted tinkerbird, 18 were conducted in the allopatric population of red-fronted tinkerbird (Kube Yini) and 13 experiments were performed in the contact zone

Figure 2: Playback experiment set-up. Two tripods with models mounted on top and one tripod in the middle where the speaker was placed.

Fig 3: Boxplots showing the response of red-fronted and yellow-fronted tinkerbird to conspecific and heterospecific song in the contact zone and allopatric populations.

Response to song of yellow-fronted and red-fronted tinkerbird

In the models where latency to approach was used as a response variable, there were no differences in responses to the two playbacks or among the populations, but there was a significant interaction effect, in that responses to the two playbacks in the contact zone were significantly different to in the allopatric YFT population (Figure 3a, Table 1). The significant interaction suggests that when we use YFT playback, responses are significantly slower in the contact zone compared with RFT playback, than they are to the two species in the YFT allopatry population. For the closest distance, the models shows that there is a significant interaction in that hybrids responded significantly less to YFT song than to RFT song compared to how RFT responds to YFT song (Figure 3b, Table 1).

Figure 3: Plots showing the model effects of playback and population on a) latency to approach and b) closest distance within 20m of the experiment set-up. Also depicted is the effect of the interaction between population and playback used.

Table 1: Generalised linear mixed model output showing the effects of focal bird and population as well their interaction on the response (measured as latency to approach and closest distance within 20m) of red-fronted (RFT) and yellow-fronted tinkerbird (YFT). P-values in bold indicate statistically significant effects.

	Latency to approach				Closest distance (within 20m)
Predictors	Estimates	SE	z-value	p-value	Estimates	SE	z-value	p-value
Intercept	4.88	0.19	25.24	<0.001	0.57	0.38	1.50	0.135
Playback [YFT]	0.41	0.31	1.30	0.194	0.33	0.62	0.53	0.597
Population [RFT allopatry]	0.40	0.27	1.47	0.141	0.17	0.53	0.32	0.748
Population [YFT allopatry]	0.22	0.29	0.76	0.446	0.21	0.57	0.37	0.715
Playback [YFT] × Population [RFT allopatry]	-0.60	0.43	-1.38	0.167	-0.67	0.85	-0.78	0.433
Playback [YFT] × Population [YFT allopatry]	-0.86	0.41	-2.07	0.038	-0.64	0.81	-0.80	0.426

Role of song and plumage on tinkerbirds interaction

Model (plumage) versus speaker (song)

RFT approaches the model more to RFT playback and the speaker more to YFT playback (Table 2, Figure 4a-b). YFT does not differentiate between model and speaker in response to either playback, but there is an interaction effect in that the difference between closest distances between RFT and YFT playbacks differs significantly from that difference in RFT. YFT also spends significantly less time within 1.5m of the experimental setup than RFT does.

Red-fronted model versus yellow-fronted model

Red-fronted tinkerbird differentiates between the two playbacks and responds significantly differently to them, approaching the red-fronted model to RFT playback and the yellow-fronted model to YFT playback (Table 2, Figure 4c-d). Yellow-fronted tinkerbird did not respond differently to the two models in response to either playback.

Figure 4: The role of song on the interaction of red-fronted and yellow-fronted tinkerbird measured as the a) time spent or b) closest distance to the speaker or the models used. The plots show the effects of the interaction between playback and the focal bird on the response of the focal bird. Also shown is the role of plumage colouration on the interaction of the two species measured as the c) time spent or d) closest distance by the bird to either the red-fronted or yellow-fronted tinkerbird model. Negative values of time spent, and closest distance depict time spent and closest distance to yellow-fronted tinkerbird while positive depict time spent and closest distance to red-fronted tinkerbird.

Table 2: GLMM outputs showing the effect of focal bird and playback on the closest distance to the speaker or model and the time spent with either the speaker or model of a bird responding during the experiment. For closest approach, negative effect represents closer to the speaker, positive closer to the model. For time spent, negative effects represent more time with the speaker, positive more with the model. P-values in bold indicate statistically significant effects.

	Closest distance				Time spent
Predictors	Estimates	SE	z-value	p-value	Estimates	SE	z-value	p-value
Intercept	0.04	0.35	0.13	0.900	24.75	17.11	1.45	0.148
Focal bird [YFT]	1.08	0.60	1.79	0.073	-68.75	29.63	-2.32	0.020
Playback [YFT]	1.55	0.53	2.91	0.004	-46.75	26.14	-1.79	0.074
Focal bird [YFT] × Playback [YFT]	-1.79	0.77	-2.34	0.019	57.49	37.61	1.53	0.126

Table 3: GLMM outputs showing the effect of focal bird and playback on the closest distance and time spent closer to the red-fronted or yellow-fronted tinkerbird model during the experiment. Positive values indicate closer to / more time with YFT model and negative to RFT model. P-values in bold indicate statistically significant effects.

	Closest distance				Time spent
Predictors	Estimates	SE	z-value	p-value	Estimates	SE	z-value	p-value
Intercept	-0.64	0.64	-1.00	0.315	7.38	28.51	0.26	0.796
Focal bird [YFT]	0.25	0.92	0.27	0.788	-50.88	49.38	-1.03	0.303
Playback [YFT]	1.33	0.63	2.13	0.033	5.96	43.54	0.14	0.891
Focal bird [YFT] × Playback [YFT]	-1.09	0.90	-1.20	0.230	48.67	62.67	0.78	0.437

Discussion

In this study we aimed to determine the extent to which yellow-fronted and red-fronted tinkerbirds distinguish between the plumage coloration and song of each species. The results of our experiments suggest red-fronted tinkerbird song elicits stronger responses in the contact zone, the only location where both species can be heard side by side, and RFT responds more appropriately to its own song, approaching the red-fronted model more closely than to either the speaker or the yellow-fronted model, while yellow-fronted tinkerbirds did not respond differently to the visual stimuli in either playback treatment.

Our findings also suggests that red-fronted tinkerbirds better distinguish the characteristics of yellow-fronted and red-fronted tinkerbird songs. According to Sebastianelli et al. (2024), female red-fronted tinkerbirds prefer to mate assortatively with male red-fronted tinkerbirds, whereas female yellow-fronted tinkerbirds females mate with red-fronted, yellow-fronted and even hybrid males (Sebastianelli et al., 2024).

Analysis of experiments using red-fronted and yellow-fronted models showed that red-fronted tinkerbirds interacted closely with the red forecrown model when exposed to red-fronted tinkerbird song. When exposed to yellow-fronted tinkerbird stimuli, red-fronted tinkerbirds interacted more with the yellow forecrown model, while the yellow-fronted tinkerbirds did not differentiate between the yellow or red forecrown, regardless of the playback used. The differentiated interaction of red-fronted tinkerbirds with the models suggests that song plays a major role in the interaction of the two species.

Environmental factors, body mass, and ancestry shape song frequency in red-fronted and yellow-fronted tinkerbird

In the second study, we examined the effect of ecological and genetic factors on song frequency in red-fronted and yellow-fronted tinkerbird. We utilized 384 song recordings of 165 ringed birds obtained during fieldwork (2015-2023) across a contact zone in Southern Africa between yellow-fronted and red-fronted tinkerbird (Fig. 5) and tested two long-standing hypotheses of factors influencing song frequency in animals: the acoustic adaptation hypothesis (AAH) and the morphological constraint hypothesis (MCH). The AAH suggests that the acoustic signals of animals, especially those used for communication, are adapted to the transmission properties of the habitat in which they are produced and received in. Therefore, environmental factors such as vegetation density and structure influence the evolution of these signals to enhance their effectiveness in a given habitat. As such, birds in forests might use lower-pitched, slower songs that penetrate foliage better, while birds in open areas might use higher-pitched, faster songs less affected by wind and open space. Because red-fronted tinkerbird occupies coastal forests and more riparian-like habitats while yellow-fronted tinkerbird occurs in more open dry woodland savanna, we expected red-fronted tinkerbird to sing at lower frequencies than yellow-fronted tinkerbird. The MCH, instead posits that in birds, body size and beak size and shape necessarily constrain song frequency characteristics.

Figure 5: a) Distribution of red-fronted and yellow-fronted tinkerbird across sub-Saharan Africa. Insert showing the range of the two species in Southern Africa and where they overlap in Eswatini. b) Spectrograms of red-fronted, hybrid and yellow-fronted tinkerbird song.

In previous work we performed at a continental scale, we showed that body size differs along an elevational and latitudinal scale, with larger birds at higher elevation and latitude, and that patterns of song frequency mirrored this finding, with lower frequencies higher up and closer to the poles (Sebastianelli et al., 2022). We also found some limited support for an effect of the AAH at that scale. The study was based on independent data sets, however, and no study has investigated the direct and indirect effects of the environment on song frequency, including via its effect on morphology. Furthermore, our work across the tinkerbird contact zone in Southern Africa has identified the genetic basis of vocal rhythm. It is not known to what extent song frequency may have a genetic basis as well. Here, we aimed to determine the relative effects of the environment and ancestry on song frequency, their direct effects and their indirect effects via their contribution to morphological differences in body size and beak size.

Environmental variables

To factor in environmental conditions at the capture locations, we obtained data on enhanced vegetation index (EVI), leaf area index (LAI), and percentage tree cover (VCF) from the Moderate-Resolution Imaging Spectroradiometer (MODIS) for each year and location where the birds were recorded. EVI quantifies vegetation greenness, LAI describes plant canopy structure, and VCF represents the percentage of tree cover. Additionally, data on bioclimatic variables including mean annual temperature (BIO1), annual rainfall (BIO12), and humidity were sourced from WorldClim, an online database providing climatic data from 1970 to 2000 (Fick & Hijmans, 2017). The environmental data were extracted in ArcGIS using the “extract values to points” function in the statistical analysis tool.

Inferring genomic ancestry

We collected samples from individuals and then extracted DNA for whole genome resequencing and ddRAD seq. We then used fastStructure (Raj, Stephens, and Pritchard 2014) (Purcell et al. 2007)to infer ancestry, setting the number of populations to K=2 (Nwankwo et al., 2019).

Effects of environmental and genetic factors on song

We used a structural equation model (SEM) (Grace, 2006) to investigate the direct and indirect effects of ancestry, morphology (beak size and body size), and environmental factors (EVI, LAI, VCF, absolute latitude, rainfall, temperature, humidity) on song frequency in yellow-fronted and red-fronted tinkerbirds. We used the SEM to partition the correlations between the exogenous explanatory variables and endogenous response variables into direct and indirect effects using a path analysis (Grace 2006).

We first log-transformed the song frequency because it is better represented on a logarithmic scale (Sebastianelli et al., 2022). Then, we fitted a model to infer the direct effects of LAI, EVI, VCF, temperature, rainfall, humidity, ancestry, body mass, and beak size. In the same model, we also looked at the indirect effects of ancestry, rainfall, temperature, and humidity on body mass, and whether body size would affect song frequency. We also considered the possibility that the size of the beak could constrain song frequency (sensu (Podos, 2001)), so we fitted a model to examine the indirect effects of mass, temperature, EVI, LAI, and VCF on beak size. We assessed the fit of our model using a χ2 goodness-of-fit test of the difference between the observed data and hypothesized model, the root mean square error of approximation (RMSEA) and the comparative fit index (CFI). A satisfactory model fit was indicated by: (1) a non-significant χ2 goodness-of-fit test (α=0.05), (2) CFI > 0.9, and (3) lower 90% confidence intervals (CIs) of RMSEA < 0.05 (X. Zhang & Savalei, 2016).

Figure 6: Scatter plots showing the relationship between song frequency (in log scale) and a) body mass, b) ancestry and c) humidity.

Results

There were both direct and indirect effects of the environmental and genetic factors on tinkerbird song frequency (Fig 6). Body size, beak size, and environmental factors all contributed to variation in song frequency. In particular, there was a striking negative effect of body mass on song frequency (β= -0.015, P(>|z|) = < 0.001) (Table 4), suggesting that larger birds sing at lower frequencies. Furthermore, there was a statistically significant negative effect of ancestry on song frequency (β= -0.035, P(>|z|) = 0.002) (Table 4), indicating that birds with more red-fronted tinkerbird ancestry sing at lower frequencies, but there was no relationship between ancestry and body mass (β= -0.326, P(>|z|) = 0.161) (Table 4).

There was also a negative direct effect of EVI, a measure of greenness, on song frequency (β= -0.008, P(>|z|) = 0.006) (Table 4), suggesting that birds in open habitats sing at higher frequencies compared to those in densely forested areas. There was a positive relationship between song frequency and humidity (β= 0.041, P(>|z|) = 0.003) (Table 4), indicating that birds sing at higher frequencies in more humid areas. Conversely, temperature had a negative statistically significant effect on body mass (β= -0.015, P(>|z|) = 0.002) (Table 4). There were no direct effects of LAI, VCF, temperature, rainfall, and beak size on song frequency, but with the effect of temperature on body mass, an indirect effect of temperature on song frequency was revealed (Table 4). Body mass also positively affected beak length, but beak length did not affect song frequency.

Table 4: Direct and indirect effects of ecological and genetic factors on song frequency in yellow-fronted and red-fronted tinkerbird. Statistical significant effects are shown with bold p-values.

Regressions:	Predictor variables
		Estimate	SE	z-value	P(>\|z\|)
Song frequency~
	EVI	-0.008	0.003	-2.762	0.006
	LAI	0.004	0.003	1.421	0.155
	VCF	-0.004	0.004	-1.006	0.314
	Body mass	-0.015	0.004	-4.117	<0.001
	Ancestry	-0.035	0.011	-3.092	0.002
	Temperature	-0.007	0.012	-0.529	0.597
	Humidity	0.041	0.014	3.006	0.003
	Rainfall	<0.001	0.005	0.07	0.944
	beak length	0.005	0.003	1.387	0.165
beak length~
	Body mass	0.19	0.076	2.494	0.013
	Temperature	-0.102	0.078	-1.307	0.191
	EVI	0.048	0.065	0.745	0.456
	VCF	-0.043	0.07	-0.611	0.541
	LAI	-0.094	0.064	-1.464	0.143
body mass~
	Ancestry	-0.326	0.232	-1.403	0.161
	Rainfall	0.093	0.107	0.869	0.385
	Temperature	-0.832	0.265	-3.141	0.002
	Humidity	0.406	0.297	1.364	0.173

Discussion

Our study provides insight into how the frequency yellow-fronted and red-fronted tinkerbird song is influenced by temperature, via the effect of temperature on body size in line with Bergmann’s rule. We also observed a significant positive impact of humidity on song frequency. The opposing effects of temperature and humidity on song frequency indicate that the song of tinkerbirds is strongly influenced by environmental factors. For instance, tinkerbirds living in warmer climates tend to have smaller body sizes, resulting in higher frequency songs, in accordance with Bergmann’s rule.

Additionally, we identified a negative relationship between ancestry and song frequency. This indicates that birds with a higher proportion of red-fronted tinkerbird ancestry sing at lower frequencies. Furthermore, the significant negative association between song frequency and EVI (a measure of greenness) suggests that birds found in dense coastal forests, such as the red-fronted tinkerbird, sing at lower frequencies. This aligns with the acoustic adaptation hypothesis, suggesting that for effective propagation of vocal signals, birds should adjust their songs based on their environment. Therefore, our results demonstrate that, for optimal song transmission, red-fronted tinkerbirds in coastal forests should sing at lower frequencies, contrary to yellow-fronted tinkerbirds in open habitats.

References

Ashton, K. G. (2002). Patterns of within‐species body size variation of birds: strong evidence for Bergmann’s rule. Global Ecology and Biogeography: A Journal of Macroecology, 11 (6), 505–523. https://doi.org/10.1046/j.1466-822x.2002.00313.x

Badyaev, A. V., Hill, G. E., & Weckwort, B. V. (2002). Species divergence in sexually selected traits: increase in song elaboration is related to decrease in plumage ornamentation in finches. Evolution; International Journal of Organic Evolution, 56 (2), 412–419. https://doi.org/10.1111/j.0014-3820.2002.tb01350.x

Ballentine, B. (2006). Morphological adaptation influences the evolution of a mating signal. Evolution; International Journal of Organic Evolution, 60 (9), 1936–1944. https://doi.org/10.1111/j.0014-3820.2006.tb00536.x

Blount, J. D., Metcalfe, N. B., Birkhead, T. R., & Surai, P. F. (2003). Carotenoid modulation of immune function and sexual attractiveness in zebra finches. Science, 300(5616), 125–127. https://doi.org/10.1126/science.1082142

Brumm, H., & Naguib, M. (2009). Chapter 1 Environmental Acoustics and the Evolution of Bird Song. In Advances in the Study of Behavior (Vol. 40, pp. 1–33). Academic Press. https://doi.org/10.1016/S0065-3454(09)40001-9

Catchpole, C. K., & Slater, P. J. B. (2003). Bird Song: Biological Themes and Variations. Cambridge University Press. https://play.google.com/store/books/details?id=sB24pLg4gywC

Cooney, C. R., He, Y., Varley, Z. K., Nouri, L. O., Moody, C. J. A., Jardine, M. D., Liker, A., Székely, T., & Thomas, G. H. (2022). Latitudinal gradients in avian colourfulness. Nature Ecology & Evolution, 6 (5), 622–629. https://doi.org/10.1038/s41559-022-01714-1

Delhey, K., Szecsenyi, B., Nakagawa, S., & Peters, A. (2017). Conspicuous plumage colours are highly variable. Proceedings. Biological Sciences / The Royal Society, 284(1847). https://doi.org/10.1098/rspb.2016.2593

Delhey, K., Valcu, M., Dale, J., & Kempenaers, B. (2023). The evolution of carotenoid-based plumage colours in passerine birds. The Journal of Animal Ecology, 92 (1), 66–77. https://doi.org/10.1111/1365-2656.13791

Derryberry, E. P., Seddon, N., Derryberry, G. E., Claramunt, S., Seeholzer, G. F., Brumfield, R. T., & Tobias, J. A. (2018). Ecological drivers of song evolution in birds: Disentangling the effects of habitat and morphology. Ecology and Evolution, 8 (3), 1890–1905. https://doi.org/10.1002/ece3.3760

Ey, E., & Fischer, J. (2009). THE ‘Acoustic Adaptation Hypothesis’—A Review of the Evidence from Birds, Anurans and Mammals. Bioacoustics, 19 (1-2), 21–48. https://doi.org/10.1080/09524622.2009.9753613