The Divergence of Male Reproductive Strategy as the Cause of Nomadism in Wood Warbler

doi:10.21203/rs.3.rs-4771979/v1

Download PDF

Research Article

The Divergence of Male Reproductive Strategy as the Cause of Nomadism in Wood Warbler

https://doi.org/10.21203/rs.3.rs-4771979/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The Wood Warbler Phylloscopus sibilatrix is a nomadic migratory songbird. Males often change territories during breeding season and every year displaying low nest-site fidelity. However, the benefits of frequent change between territories remain unclear. During our research, playback experiments were implemented to test whether male settlement or departure are affected by quality of conspecifics nearby or female presence. High or low rate song broadcasts were used to imitate the presence of males with ‘good’ and ‘poor’ quality, arriving males were subsequently banded with colour rings to track their further movements between plots. The results showed that between 52 and 66,7% of males left the plot during the first week after arrival (labelled as ‘floaters’) but after leaving they did not try to settle on the other plots, where high or low rate songs were broadcasted. We tested the 'hidden lek' hypothesis by tracking male appearances near local nests. The male abundance increased significantly with the start of egg-laying compared to the 10-days prior to it and decreased during the next 10 days after incubation start, assuming that males left the territory after copulation with females. The study suggests that to increase chances of leaving offspring, some floaters move through the territory of other males and copulate with their females rather than settle nearby.

сonspecific attraction

song playback

nomadism

hidden lek hypothesis

settlement decision

Wood Warbler

The reasons driving nomadism of Wood Warbler are currently widely discussed. In our field experiment, we have found no evidence that wood warblers attempt to gain a competitive advantage by settling nearby ‘poor’ or ‘good quality’ males, but an increase in male abundance on a plot was predicted by the nest presence at the egg-laying stage. The results suggest that males have two reproductive strategies: conservative and opportunistic. Conservative males typically display nest-site fidelity and bond with females, while opportunistic males regularly change their territory shortly after copulation with a local female. Opportunistic males do not have a permanent territory, so they never return to the place where they were banded, resulting in low nest-site fidelity and nomadic behaviour. We thus provide unique evidence of divergence in the reproductive strategies of typically “socially monogamous” species leading to partial nomadism.

Singing behaviour is a crucial element of the social life of songbirds, aimed at attracting a mate and defending the territory from rivals (Collins 2004; Catchpole and Slate 2008). Nonetheless, singing may also attract conspecifics to settle nearby and, as a result, lead to the increase in the local number of birds. The phenomenon is broadly referred to as ‘conspecific attraction’ (Stamps 1988; Ahlering and Faaborg 2006 etc.).

Conspecific attraction has been demonstrated experimentally across various invertebrates and vertebrates taxa (Buxton et al. 2020), especially among migratory birds (Valente et al. 2021). Assuming that dispersal is aimed at the rational use of resources scattered throughout the range, aggregation of animals should increase intraspecific competition. Therefore, сonspecific attraction should gain benefits that exceed the agonistic costs in order to be adaptive (Seppänen et al. 2007). According to the ‘location cue’ hypothesis, the conspecifics’ presence increases the detectability of suitable habitat and reduces search costs (Stamps 2001, Luepold et al. 2023). On the other hand, birds can also gather social information by observing behaviour and settlement decisions made by other individuals to reduce costs of habitat evaluation (Danchin et al. 2004; Nocera et al. 2010; Szymkowiak 2013). The habitat choice copying may be beneficial for dispersing individuals who lack personal experience and have relatively little time to explore a certain area.

Both hypotheses have been tested on the Wood Warbler Phylloscopus sibilatrix, a long-distance migratory passerine bird that breeds in Europe and the western part of Central Asia. Using song playback to simulate the conspecific presence during spring migration increased the number of wood warblers settling in an area and accelerated the settlement of unoccupied territory (Szymkowiak et al. 2016; Grendelmeier et al. 2016; Luepold et al. 2023). However, field experiments show that during habitat evaluation males did not prioritise social information over personal information and did not settle on the low quality sites, regardless of the conspecific song broadcast (Luepold et al. 2023).

A bird song encodes information about individual quality: health (Lampe and Espmark 1994; Garamszegi et al. 2004), aggressiveness (Van Duyse et al. 2002; Brumm and Ritschard 2011), reproductive success (Nemeth et al. 2012; Cramer 2013), stage of the reproductive cycle (Temrin 1986), social status (Spencer et al. 2004) and etc. In order to benefit from settling near conspecifics and reduce agonistic costs, birds can evaluate the song performance of neighbours before settlement (Szymkowiak et al. 2016; Kelly and Ward 2017; Morinay et al. 2020).

In the Wood Warbler, the high song rate is an aggressive signal in male-male interactions (Szymkowiak and Kuczynski 2017), which is also positively correlated with pairing success with females (Szymkowiak et al. 2016). Thus, broadcasting the song at a high or low rate (6 or 2 trills per minute) simulated the presence of ‘good’ and ‘poor’ quality males. The playback experiments have shown that males preferred to settle on song plots with ‘poor’ quality individuals (Szymkowiak et al. 2016). Most of the males that appeared on plots with ‘good’ quality conspecifics did not settle there, so those birds who left the territory were referred to as ‘floaters’. It has been suggested (Szymkowiak et al. 2016) that floaters avoid the costs of high competition with ‘good’ quality males, so we can assume that they eventually settle on other plots.

However, data on the movements of ringed birds so far provide very limited information about the fate of floaters (Herremans 1993; Lapshin 2009; Grendelmeier et al. 2016; Goretskaia et al. 2024), because most of them are hard to locate again. Frequent movements of males within a breeding season are very typical for wood warblers (Herremans 1993; Luepold 2023). New males can appear at the same site up to 6 times and it is often not obvious why the previous site-owner left (Lapshin 2009). Due to low nest-site fidelity and interannual fluctuations in bird local abundance, the wood warbler is described as a nomadic species (Wesołowski et al. 2009; Teitelbaum and Mueller 2019 etc).

There are several hypotheses, according to which birds derive social benefits by settling nearby conspecifics (Kramer et al. 2009; Hammer et al. 2023). The ‘hidden lek’ hypothesis assumes that aggregations of territorial birds facilitate mate attraction and increase their chances for extra-pair copulation (Wagner 1998; Fletcher and Miller 2006). To test this hypothesis a song playback was used to attract wood warblers and then assess nesting parameters at the plots. Despite the territory occupancy being higher on song plots compared to control ones, the number of pairs and the nest survival rates did not differ significantly between plots and only three extra-pair offspring were found (Grendelmeier et al. 2016). Male mating success also slightly decreased as the number and proximity of neighbouring males increased (Luepold et al. 2023).

Nevertheless, some studies show that up to 25% of all nestlings could be extra-pair offspring and could be found in 41% of nests (Goretskaia et al. 2024). Despite the fact that blood samples for kinship estimation were taken from males living nearby, only one genetic father of extra-pair nestlings was found. Thus, it appears that some of the males left the area after the copulation with females. These males were assumed to be polyterritorial and to belong to other pairs of wood warblers outside the study area (Goretskaia et al. 2024).

However, the hypothesis put forward in this study suggests that wood warblers could implement two different reproductive strategies: conservative and opportunistic. A conservative male (resident) looks for a nest site and a mate, whereas an opportunistic one settles next to conservative males, copulates with their females and then leaves this territory to find another female. According to this hypothesis, some floaters pursue the opportunistic strategy and move throughout the breeding season in search of an opportunity for extra-pair copulation, rather than a place to nest.

To test this hypothesis, we tried to answer the following questions:

1) If some males do not settle on plots with ‘good’ quality conspecifics due to the high rate of intraspecific competition, could they be found on plots with ‘poor’ quality conspecifics instead?

2) Do floaters appear during the egg-laying period of the local females and disappear after the start of incubation?

3) Were the floaters polyterritorial males arriving from other plots?

Study site and broadcast manipulation

The Wood Warbler Phylloscopus sibilatrix is a trans-Saharan migratory songbird, arriving in the northwestern woodlands of Russia from late April onwards. Female arrival starts approximately two weeks later than male and breeding begins on the 20th of May (Sokolov et al. 1996; Lapshin 2009). New males appear and sing on site up until the middle of July, and due to nest predation females try to renest throughout the breeding season.

The experiments for this study were conducted in 2021–2023 at Nizhne-Svirsky State Nature Reserve, situated in Leningrad Region of Russia near Lake Ladoga (60°34′58″ N, 33°00′24″ E). The northern periphery of the Wood Warbler breeding range is located within the Reserve. Despite the breeding density being relatively low in the north and fluctuating from 0 to 60–70 pairs per km² (Lapshin 2009), the main migration route runs to the southeast along the Ladoga coast. This means that migratory birds fly over there in spring and may respond to song playback. Apart from the experiments, data from the Ladoga Ornithological Station, located at the study area, were also used to track movements of the ringed wood warblers.

Although wood warblers prefer mature broad-leaved woods (Simms 1985), which are few in the north, they also settle in closed-canopy coniferous and mixed forests (Lapshin 2009). The study territory was identified as suitable for wood warbler habitat based on a pilot survey conducted in previous years. We carried out the song playback experiments in the following habitats independently (Table 1):

1. A mature pine forest with sparse undergrowth, low-level shrubbery and cover with moss (PF)

2. A mature mixed deciduous-spruce forest with sparse undergrowth and thin herb layer (DSF)

The treatment plots (9 ha each) were randomly assigned and set up linearly due to the woodlands being located in a similar linear way along the Ladoga shore and marshes. The most distant point of a plot was no further than 300 m from the playback station and the minimal distance between plot centers was 600 m.

The song playback began in the period between April 28 and May 2, at least one week prior to the arrival of the first wood warblers in each year. The song was broadcasted automatically every day until the last nestlings fledged in early August. Since the wood warbler is a nocturnal migrant, the song playback began 1–1,5 hours before sunrise and continued until 8–9 pm each day with a 5-minute silence period after 15 minutes of singing. Additional 3–4 hours of silence periods were added in the afternoon and 6 hours at night. Playback stations consisted of one audio speaker (frequency range 100–20000 Hz, > 80 dB) with loaded playback records, a portable USB power bank and a time relay. Each station was set up on a tree at a height of 1.5 m and was hidden in a waterproof plastic case of camouflage colour. We changed power banks every 2–4 days to ensure that broadcasts continued uninterrupted throughout the experiment period.

The Wood Warbler has two distinct song types: a high-frequency trill song and a lower-frequency whistle song. The trill song is used by all males during breeding season, while the whistle song is performed predominantly by unmated males (Temrin 1986). Trill singing rate is important for female attraction and interactions between unmated males (Szymkowiak and Kuczynski 2017). Based on the suggestions of previous studies (Szymkowiak et al. 2016), songs with a high rate (6 trills per min & 1 whistle song per min) and low rate (2 trills per min & 1 whistle song per min) were used to imitate the presence of ‘good’ or ‘poor’ quality males.

The experiments were conducted in 2021 and 2023 years in different habitats (PF and DSF respectively) to test the repeatability of results and to rule out effects of patch quality heterogeneity that may have been overlooked. The song playback experiment consisted of three treatment groups (Table 1):

1) control plots with no audio playback (2021, n = 5 plots; 2023, n = 6);

2) experiment plots with a high rate broadcasted song to simulate the presence of ‘good’ quality males (GQ treatment: 2021, n = 4 plots; 2023, n = 5);

3) experiment plots with a low rate broadcasted song to simulate the presence of ‘poor’ quality males (PQ treatment: 2021, n = 4 plots, 2023, n = 6).

The presence of one (2023) or two males 30 m apart (2021) was simulated on each of the song plots (Table 1). In 2021, song recordings from two different males (M1 and M2) were both used on each experiment plot. One plot broadcasted low-rate M1 and M2 songs (PQ treatment), while the other plot replayed the high-rate versions оf M1 and M2 songs (GQ treatment). The assumption was that wood warblers would be able to prioritise one experimental plot over the others based only on the singing rate of the songs varied across plots, but other characteristics of song playback could not affect settlement decisions. In 2023 it was decided to change the experimental design of the previous study and broadcast one of eight unique wood warbler songs (M1-M8) on each song plot in order to avoid pseudoreplication. Each record was broadcasted at one plot with a high or low rate. The songs were either recorded by a Sony ICD-SX712 voice recorder in 2021 from local males or downloaded from the xeno-canto.org website (http://www.xeno-canto.org/). Then the Audacity® Cross-Platform Sound Editor 3.0.2 was used to process song records.

Although after settlement the birds habituated to the song broadcast, the records were not changed during the experiment, as the effect was only intended to attract new males to settle but not to affect residents. The observations showed that replacing the records or replaying the song after several days of the silence period led to aggressive responses of neighbour wood warblers to the new 'intruder'.

In 2022 in the PF wood warbler songs were broadcasted on the plots but they did not imitate male of GQ and PQ quality (Table 1). Bird counts and banding results from the 2022 (DSF) and 2023 (PF) were used to determine a site preference by wood warblers and return rates in subsequent years.

Another crucial aspect that had to be considered was the presence of other species of birds, such as the Common Chiffchaff Phylloscopus collybia and the Eurasian blackcap Sylvia atricapilla. According to research, the common chiffchaff has been shown to attract wood warblers to settle nearby (Szymkowial et. al. 2017). For this reason, in 2022 and 2023 we also assessed whether there is a correlation between the numbers of common chiffchaffs and wood warblers on the plot. The Eurasian Blackcap, on the contrary, is known to repel wood warblers from settling nearby (Szymkowiak et. al. 2017). However, there were only a few of them present on our study plots. The number of common chiffchaffs and eurasian blackcaps was calculated simultaneously with other bird counts.

Bird data

Singing wood warblers that stayed on the plot for 8 days and more were considered ‘residents’. Males that appeared on the site temporarily (no more than 7 days) and then flew away were counted separately as ‘floaters’. In 2023 the goal was to capture as many residents and floaters as possible and individually mark them with aluminium and colour rings, in 2021–2022 the focus was mostly on ringing resident males (Table 2). In order to catch wood warbler males, we used mist-nets and song audio playback. The number of wood warblers was counted on each plot at least once every three days in the early morning hours in suitable weather conditions (no heavy rain, no storm wind). If an unringed male could not be detected during two morning counts, he was considered to have left the territory. The number of days spent on the plots was calculated for each male.

The movements of ringed males on the plots and neighbouring territory (total area 5,4 km²) were tracked throughout the breeding season. We measured the distance from each male to the playback station and from each male to the nearest pairs of birds using handheld GPS devices. At least 47 visits were made to each plot per breeding season (from late April to early August) for observing males and looking for females and nests. Locations of nests were mapped using handheld GPS devices as well.

We were also able to determine the date when the first egg was laid for 71 nests during 2021–2023 breeding seasons. For birds whose nest had been depredated before it was found (in 19 cases), we counted a 10-day period of egg-laying from the moment the female was registered, since in the north wood warbler females begin to build a nest soon after pairing with a male (Lapshin 2009). For the nests found, we calculated the egg-laying periods by estimating the age of the nestlings; in the case when nestlings were of different ages, the first-egg-laid date was calculated from the oldest nestling and the last-egg-laid date - from the youngest nestling in the nest. The period of the female readiness for copulation was determined according to the following scheme:

Сopulation period = Nest detection date - Age of nestlings − 13-day incubation period (incubation begins on the day the last egg is laid or earlier) - Number of eggs (females lay 1 egg per day) − 3 days for nest construction + 10 days

We added 3 extra days for nest construction, implying that the female should have already been on the nest-site for at least 3 days to build a nest. The female is noticeable during the nest-building period, which leads to the assumption that males seeking an opportunity for extra-pair copulation may find her in advance and wait until she is ready to copulate. After the last egg is laid, females no longer copulate with males.

We counted males living within a 300 m radius of each nest during the 10-day period before the egg-laying period, during the egg-laying (10 days) and after the end of copulation (10 days after). It was not possible to record data blindly because our study involved focal animals in the field, where the location of birds with a unique set of colour rings were analysed.

Table 1

Description of the song playback experiments in 2021–2023. N - number of plots; GQ, PQ - plots where the presence of ’good quality’ or ‘poor quality’ males was simulated; PF - pine forest, DSF - deciduous-spruce forest
Year	Experimental plots, N		Control plots, N	Number of males simulated on plots	Number of unique song records broadcasted	Treatment, study area
Year	High rate song, GQ	Low rate song, PQ	No playback	Number of males simulated on plots	Number of unique song records broadcasted	Treatment, study area
2021	4	4	5	2	2	Song playback, PF
2022	-	-	13	-	-	Song playback, PF
2023	-	-	13	-	-	Control, PF
2022	-	-	14	-	-	Control, DSF
2023	5	6	6	1	8	Song playback, DSF

Data analysis

To evaluate plot settlement in response to the GQ and PQ treatments (2021, 2023) we tested for differences in male abundance across treatments with a generalised linear model (GLM) fitted with a Poisson distribution. Models included treatment, as well as the number of residents per plot in the control year (2023 for PF area and 2022 for DSF, Table 1), current and pre-treatment years (as a measure of the plot quality for wood warblers habitat) and their interaction as fixed effects. The response variables were the total number of males, total number of residents and floaters on the plots during the breeding season. For the first model, the experimental treatment was only an effect (Table 2, model 1–2). The next models included the number of residents per plot in the control year (no audio playback), current year (only for floaters abundance) and pre-treatment year and its interaction with treatment as covariates (Table 2, model 3). In 2022 and 2023, the models were extended by adding the common chiffchaff abundance per plot and its interaction with other predictors as covariates (Table 2, model 3). The optimal model was chosen based on the Akaike information criterion optimised for small datasets (MacQuarrie and Tsai 1998; Burnham et al. 2011) (Table 3, model 1–4).

We fitted a generalised linear mix model (GLMM) to analyse the differences in male abundance depending on the presence or absence of an egg-laying female. For the first model, male abundance during a 10-day period within 300 m of each nest was the response variable. The predictors were 10-day periods before the first eggs were laid, during the egg-laying, and after the start of incubation (Table 2, model 4). Since there were from 0 to 5 nests per plot in different years, plot name was added as a random effect. The best-fitting model was selected based on the lowest AICc of models with different random effects, such as year, plot name, treatment at the plot, and nested random effects between its.

The male abundance changes as the breeding season progresses. We divided each month in 3 decades and numbered them consecutively from May 1 to July 30. Then for another GLM, the response variable was male abundance within 300 m of each nest and predictors were the decades numbered from 2nd to 8th, for the period from May 10 to July 20 during which birds were counted (Table 2, model 5). The next model included the 10-day periods before, during and after egg-laying, as well as the decade numbers as fixed effects and plot name as random effect. To further examine whether variation in male abundance on each plot is better explained by the course of the season or the nest presence, we compared these three models with a χ2 goodness-of-fit test (Table 3, model 5).

We used GLM to assess whether the nest presence at egg-laying stage can be predicted from changes in the male abundance within 300 m of the nest. The ‘nest presence’ and ‘nest absence’ was a binomial dependent variable. The differences in male abundance during 10 days before the first eggs were laid (‘Before EL’), during the 10 day period of egg-laying (‘During EL’) and 10 days after the start of incubation (‘After EL’) were chosen as fixed effects (‘During EL - Before EL’; ‘During EL - After EL’). To create dataset with ‘nest absence’, the model included the difference in the number of males per plot during the period of fixed dates: from 22 to 31 May (3rd decade), from 1 to 10 June (4rd decade) and from 11 to 20 June (5rd decade) (‘4rd-3rd decades’; ‘4rd-5rd decades’). The time frame was not related to the beginning and end of the egg-laying of the local nests but was chosen due to the most active period of bird arrival and breeding (Table 2, model 6).

To estimate the difference in wing and tail lengths between residents and floaters, we apply a GLM with a Gaussian distribution. For all models we used the ‘DHARMa’ package (version 0.4.6) as a simulation-based approach to create readily interpretable scaled residuals to check if the distributional assumptions of the fitted models are correct (Hartig 2022). The statistical software R 4.3.1 (R Core Team 2014) and RStudio-2023.12.1–402 were used for statistical analysis and data visualisation (Arel-Bundock 2022). For GLMMs we used the package ‘lme4’ (Bates et al. 2014). Adobe Photoshop CS3 3.2.4.9 was used to add additional explanation to the figures.

Table 2

Summaries of GLMs and GLMMs assessing the relationship between response variables and predictors. CI – confidence intervals, n – the numbers of observations; EL – egg-laying period, GQ, PQ – plots with simulated presence of ’good quality’ or ‘poor quality’ males. Statistically significant values are in bold
Model (year, area)	Response Variable	Predictors	Estimates	CI	P value	n
1 (2021, PF)	All male abundance	Control plots (Intercept)	6.00	3.60–8.40	< 0.001	13
		Treatment21 [GQ]	4.00	0.40–7.60	0.029
		Treatment21 [PQ]	2.50	-1.10–6.10	0.174
	Resident abundance	Control plots (Intercept)	3.00	1.13–4.87	0.002
		Treatment21 [GQ]	1.25	-1.55–4.05	0.382
		Treatment21 [PQ]	1.25	-1.55–4.05	0.382
	Floater abundance	Control plots (Intercept)	3.00	1.35–4.65	< 0.001
		Treatment21 [GQ]	2.75	0.27–5.23	0.030
		Treatment21 [PQ]	1.25	-1.23–3.73	0.323
2 (2023, DSF)	All male abundance	Control plots (Intercept)	4.17	1.98–6.35	< 0.001	17
		Treatment [GQ]	0.03	-3.21–3.28	0.984
		Treatment [PQ]	0.67	-2.42–3.76	0.673
	Resident abundance	Control plots (Intercept)	1.67	0.38–2.95	0.011
		Treatment [GQ]	-0.27	-2.17–1.64	0.784
		Treatment [PQ]	-0.17	-1.98–1.65	0.857
	Floater abundance	Control plots (Intercept)	2.50	1.12–3.88	< 0.001
		Treatment [GQ]	0.30	-1.75–2.35	0.774
		Treatment [PQ]	0.83	-1.12–2.78	0.403
3a	All male abundance	(Intercept)	1.27	2.75–4.54	< 0.001	43
		No. of residents in control year	0.26	1.17–1.43	< 0.001
		(Intercept)	1.27	2.83–4.45	< 0.001
		No. of residents in pre-treatment year	0.11	1.05–1.19	< 0.001
		(Intercept)	1.53	3.69–5.73	< 0.001	26
		No. of common chiffchaffs abundance	0.11	0.95–1.30	0.157	26
3b	Resident abundance	(Intercept)	-0.01	0.63–1.48	0.945	43
		No. of residents in control year	0.46	1.35–1.85	< 0.001	43
		(Intercept)	0.83	0.29–1.37	0.004	43
		No. of residents in pre-treatment year	0.54	0.36–0.71	< 0.001	43
		(Intercept)	0.69	1.42–2.72	< 0.001	26
		No. of common chiffchaffs	0.25	1.04–1.55	0.015	26
3c	Floater abundance	(Intercept)	1.00	1.98–3.67	< 0.001	43
		No. of residents in control year	0.11	0.97–1.27	0.121
		(Intercept)	0.92	1.86–3.34	< 0.001
		No. of residents in pre-treatment year	0.03	0.94–1.13	0.527
		(Intercept)	0.99	1.91–3.25	< 0.001	56
		No. of residents in current year	-0.05	0.99–1.16	0.065	56
		(Intercept)	0.99	1.97–3.57	< 0.001	26
		No. of common chiffchaffs	-0.05	0.73–1.1	0.669	26
4	Male abundance around the nest	During EL (Intercept)	2.52	2.15–2.90	< 0.001	270
		Before EL	-0.42	-0.79 – -0.06	0.024
		After EL	-1.16	-1.52 – -0.79	< 0.001
5		(Intercept)	1.56	3.73–6.02	< 0.001	270
5		Decades	-0.18	0.79–0.88	< 0.001	270
6	Nest presence or absence	(Intercept)	0.53	0.44– 0.62	< 0.001	146
		Change in male abundance Before EL	0.11	0.05–0.17	0.001
		Change in male abundance After EL	0.06	0.01–0.11	0.015

Table 3

Comparison of models describing the effect of predictors on the number of wood warblers, number of residents and floaters. K – number of parameters; AICc – Akaike’s information criterion adjusted for small sample size; ΔAICc = AICci - minAICc; wi – model weight; χ2 - values of Chi-squared; GQ|PQ – plots with simulated presence of ’good quality’ or ‘poor quality’ males; PF – pine forest, DSF – deciduous-spruce forest; the control year - year without audio playback, the pre-treatment year - previous year with or without audio playback. Given the data, the model with the lowest AICc was considered the best. The best-fitting models are in italics
No. of residents/plot
Models 1 (2021, PF)	K	AIC c	Δ AIC c	w i
Treatment (GQ\|PQ\|Control)	3	59.83	7.81	0.02
No. of residents in the control year	2	52.02	0.00	0.95
Treatment X No. of residents in the control year	4	58.75	6.73	0.03
Models 2 (2023, DSM)	K	AIC c	Δ AIC c	w i
Treatment (GQ\|PQ\|Control)	5	62.65	5.37	0.06
No. of residents in the pre-treatment (control) year	3	57.27	0.00	0.94
Treatment X No. of residents in the pre-treatment (control) year	4	72.35	15.08	0.00
Models 3 (2023, PF, no audio playback)	K	AIC c	Δ AIC c	w i
No. of residents in the pre-treatment year	3	35.90	0.00	0.89
The common chiffchaff abundance	3	46.70	10.79	0.00
The common chiffchaff abundance X No. of residents in the pre-treatment year	4	40.12	4.21	0.11
No. of floaters/plot¹
Model 4 (2021, PF)²	K	AIC c	Δ AIC c	w i
Treatment (GQ\|PQ\|Control)	2	58.18	0.00	0.39
No. of residents in the current year	2	61.06	2.88	0.09
No. of residents in the control year	2	59.44	1.26	0.21
Treatment X No. of residents	3	61.65	3.47	0.07
Treatment X No. of residents in the control year	3	60.00	1.82	0.16
No. of all male/plot during 10 days
Model 5 (2021–2023)	df value	AIC	χ2	P value
Decades (for the period from May 10 to July 20)		950.15		0
Each 10-day period: before, during and after egg-laying. Name plot as random effect	2	947.73	24.408	< 0.001
Each 10-day periods: before, during and after egg-laying + Decades Name plot as random effect	1	905.62	26.122	< 0.001
¹ The models of 2023 were not included in the table because no significant correlation was found between the response variable and the predictors
²Only models whose AICc weight is more than 0.05 are included

Annual return rate and bird monitoring

For our study, it was important to distinguish residents from floaters, and polyterritorial males from new arrivals, therefore, we banded all residents annually. In 2021–2022, we caught and ringed mostly resident males on 13 plots located in a pine forest: in 2021–23 males, in 2022–20 males. In 2023, we ringed 79 males (residents and floaters) out of 132 registered males on 30 plots located in a pine forest and a mixed deciduous-spruce forest (Table 4). However, there was no significant difference in wing (mean = 76.42, 95%CI: -0.96–0.47, t=-0.669, p = 0.50) and tail length (mean = 50.89, 95%CI: -1.05–0.63, t=-0.496, p = 0.62) between residents and floaters.

It is necessary to note that we did not have enough time to catch some males in 2023, as 29 unringed males (22% out of all detected males) left the sites after 1–2 days of singing. Most of those males arrived simultaneously during a short period between May 10 and May 30. Another 15 floaters and residents could not be caught for various reasons.

The situation with females was as follows: in 2021–2023 the presence of 104 females (2021–29; 2022–22; 2023–53) was registered and 71 nests, undepredated at early stages, were found (Table 4). We also ringed 11 females that were caught near the nests.

The percentage of floaters singing on plots for less than an 8-day period varied from 52 to 66.7% in 2021–2023 (Table 4). In 2023, 42 out of 79 ringed males left the plots soon after arrival, but only two males were located again. They left one control plot and relocated to another control plot. These males formed pairs with females and bred at the new location. The other ringed floaters were no longer found on the plots and in neighbouring areas, as none of them were caught at the Ladoga ornithological station at the study area. Each year we registered the presence of one to three polyterritorial males, whose second territory was located at a distance of up to 300 m from the first one. A male was considered to be polyterritorial if his second territory was more than 100 m from the first one and he pretended to be unmated by singing a whistle song and longer trills (Temrin 1986).

In 2022, only one male out of 32 wood warblers (23 males and 9 females) ringed in 2021 returned to the previous nest-site. From the 20 males and one female that were ringed in 2022 none (0%) returned in 2023. As of June 20, 2024, not a single male banded in 2023 had returned to the study area. In 2021–2024, annual return rates of adult wood warblers amounted to 0.8%, resident males – 1.3% (one out of 79 residents).

The arrival of birds at the study area in 2021–2023 began no earlier than May 2. Most of the birds arrived in the period from May 10 to May 31, but new males continued to appear until the middle of July. The first eggs were laid on May 22, the last – on July 13. The plot occupancy was 100% in 2021–2023 in the PF and 76.5% in 2023 in the DSF. The mean number of males per plot was 8.0, 5.8 and 4.4 or 52.0, 37.5 and 29.3 males per km² of suitable habitat in 2021, 2022 and 2023 respectively. The dynamics showed a decrease in mean numbers of residents each year − 24.5, 18.0 and 12.0 males per km² in 2021, 2022 and 2023 respectively.

Table 4

Number of registered and banded wood warblers (counted as residents and floaters), and nests detected during experiments in 2021–2023. N - number of males, GQ|PQ - simulation of the presence of ’good quality’ or ‘poor quality’ males
Years	Treatment	Number of residents		Number of floaters		Total number of males		Total number of nests found	Type of habitat
Years	Treatment	N	%	N	%	Total	Ringed	Total number of nests found	Type of habitat
2021	PQ\|GQ	49	47,1	55	52,9	104	24	15	Pine forest
2022	Song playback	36	48,0	39	52,0	75	22	15
2023	No	27	47,4	30	52,6	57	36	20
2022	No	-	-	-	-	21	No	-	Deciduous-spruce forest
2023	PQ\|GQ	25	33,0	50	66,7	75	43	21	Deciduous-spruce forest

Settlement response to song playback

The effect of simulated presence of GQ and PQ conspecifics on the number of wood warblers was analysed for 2021 and 2023. In 2021 the number of males significantly increased on the plots with GQ males simulation compared to control plots (no audio playback) (Table 2, model 1; Fig. 1a). Apart from that, the results showed that the total number of floaters was also significantly higher on the GQ treatment plots; the number of residents was not affected (Fig. 1b and 1c). Unfortunately, this result was not repeated in a similar experiment in 2023 – there was no disparity in the number of floaters between the control and the GQ treatment plots. Both years also demonstrated no significant differences in the male abundance between the control and the PQ treatment plots (Table 2, model 1 and 2; Fig. 1).

Moreover, no substantial contrast in the number of residents between plot treatments was seen in 2021 and 2023 (Table 2, model 1 and 2; Fig. 1b). The total number of males per plot was positively correlated with the number of residents in the control (no audio playback) and pre-treatment years (Table 2, model 3a), as well as the number of residents in the current year was positively correlated with the number of residents in the control and pre-treatment years (Table 2, model 3b). In addition, we found a positive effect of the number of the common chiffchaff per plot on the resident abundance per plot in the current year (Table 2, model 3b). However, the top supported model that predicted the resident abundance per plot included only the resident abundance in the control or pre-treatment years as predictor (Table 3, model 1, 2 and 3).

The resident abundance per plot in the current, pre-treatment and control years showed no effect on the number of floaters. Furthermore, there was also no correlation between the number of floaters and the common chiffchaff abundance (Table 2, model 3c). The top supported model that predicted the floater abundance included the treatment only for 2021, but this model explained less than 40% of variations (Table 3, model 4). Other models had little empirical support.

The male abundance was significantly greater during the 10-day period of the egg-laying stage of local nest rather than during the 10-day periods prior to and after it (Table 2, model 4; Fig. 2). There was also an overall negative effect of decade number (from 3rd to 9th, for the period from May 22 to July 30) on the total number of males, showing that the number of male per plot decreased as breeding season progressed (Table 2, model 5). However, the best-supported model showed that both predictors combined explained variation in male abundance in each plot better than both predictors individually (Table 3, model 5).

We simulated another model to test whether it was possible to predict the start of egg-laying in each nest by the difference in the number of males before and during egg-laying, as well as during and after it. The male abundance during ‘nest absence’ was taken randomly on the plot between the 3rd, 4th and 5th decades. Model predictions were statistically significant for both periods pre- and post-egg-laying change in male abundance (Table 2, model 6).

The Wood Warbler is known as a trans-Saharan migrant and at the same time as a nomadic species, since males do not return to the same place every year and often change territory during the breeding season, thus displaying low nest-site fidelity. However, the benefits of frequent change between territories are still widely discussed (Herremans 1993; Wesołowski et al. 2009; Szymkowiak and Kuczynski 2015). In field experiments we have shown that one of the reasons behind the tendency of males to leave the plot lies in their extra-pair mating behaviour rather than in the attempt to find a better nest-site.

In this study, males were divided into two groups, depending on their behaviour on a site:

1) Residents - males who settle on site for a long period of time;

2) Floaters - males who settle on site but leave after several days of singing.

Our initial hypothesis was that the emergence and departure of floaters is a result of their behaviour, in which males evaluate the environment during settlement and then leave the site if they find it unsuitable. To determine which males arrived and which left, we banded all residents in 2021 and 2022, as well as most residents and floaters in 2023.

As argued, wood warblers avoid high intraspecific competition by escaping settlement near highly competitive conspecifics. Previous experiments demonstrated the preference of males to settle on plots with ‘poor quality’ (PQ) male song broadcast and, on the contrary, to often leave plots with ‘good quality’ (GQ) male song broadcast (Szymkowiak et al. 2016). However, in our study, neither PQ nor GQ treatment plots were preferred by males for settlement, since resident abundance did not differ between plots (Fig. 1b).

The lack of evidence that song broadcasts increase the number of residents is not in stark contrast to the findings of other studies. Clustering of wood warbler territories is more evident at large scales, whereas at local scales it may be masked by bird territoriality (Broughton et al. 2020). Furthermore, settlement is a hierarchical decision-making process based on personal (nonsocial) information and social information obtained from observation of other conspecifics and their interaction with the environment (Danchin et al. 2004). The Wood Warbler does not prioritise social over nonsocial information during settlement and does not settle on sites of low quality regardless of the conspecific song broadcasted there (Luepold et al. 2023). In addition, some factors, such as the predator presence or rodent abundance, may also negatively influence bird settlement decisions (Szymkowiak and Kuczynski 2015; Szymkowiak and Thomson 2019). As a result, plots with higher numbers of residents could be of better quality and, therefore viewed by wood warblers as favourable for settlement due to some characteristics that researchers may underestimate. In our study, the resident abundance per plot was positively correlated with the number of residents in the previous and subsequent years, which allows us to assume that wood warblers preferred to settle on the same plots every year regardless of song broadcast.

On the other hand, the absence of difference in resident abundance between plots might derive from the relatively low population densities in the north, where our study was conducted (Lapshin 2009). However, the mean number of males ranged from 8.0 to 4.4 male per plot in 2021 and 2023 respectively and did not differ much from what was shown in other song playback experiments (Szymkowiak et al. 2016; Grendelmeier et al. 2016; Luepold et al. 2023). In addition, in the north wood warblers are less aggressive in establishing territorial boundaries (Matantseva et al. 2015), thus there may be no benefit for males to preferentially settle near PQ males and the resident abundance may not differ significantly between plots.

Given the limited favourable time for breeding in the north, we assumed that an adequate strategy for males is to settle and begin attracting females as soon as possible. Despite our expectations, between 52 and 66.7% of males left the plot within the first seven days after arrival. Based on the suggestion of previous study (Szymkowiak et al. 2016), a hypothesis was put forward that after leaving floaters could settle on song plots with PQ conspecifics. Consequently, the focus shifted towards finding these males again after their relocation. In 2023, 42 out of 79 males left the plot after we banded them, but only two (4.5%) wood warblers resettled on other plots at a distance of 1.5-2 km. Those males relocated from control plots to other control plots and bred there. During three years of research, there were no encounters of any other ringed floaters, despite the fact that we regularly visited all plots and adjacent territories (5.4 km²) throughout the entire breeding season. None of the ringed individuals were caught at the Ladoga Ornithological Station either. It is important to point out that the study areas were limited to Ladoga Lake and swamps, so males likely dispersed mainly along the coast and rivers, where the Ladoga ornithological station and our study area were located. The reason for why these males did not try to settle on PQ treatment plots seemed unclear.

More than half of the male population remains unmated and continues to arrive and leave within the season. According to other studies (Herremans 1993; Szymkowiak et al. 2016; Grendelmeier et al. 2016; Luepold et al. 2023), this phenomenon is observed even in the central part of the Wood Warbler's range. However, little is known about where these males came from. Another study has confirmed that most of the extra-pair offspring found in the nests were not related to neighbouring males and only in one case the extra-pair father was identified (Goretskaia et al. 2024). There was another suggestion that some males sing in secondary territories far up to 1.4 km from the primary territory and may participate in copulations with females from other pairs, and then return to their first mate in the nest (Goretskaia et al. 2024). However, our bird banding data showed that the newly arrived floaters were not polyterritorial males from a neighbouring territory, since most of them had not been ringed yet or found again after banding. In this study, a few males had a second territory and they always returned to the first site, where their females incubated the eggs. Other studies (Herremans 1993; Norman 1994) documenting the within-season movements of unmated males have also found only a few wood warblers, who moved to a new territory up to 14–32 km away from the previous location.

We tried to determine whether the number of floaters depends on the number of residents in the current or previous years, but did not find any significant correlation, except for treatment in 2021, when the floater abundance was more substantial on the GQ treatment plots. The experiment was conducted again in 2023 in another area but the results were not repeated. That brought up the assumption that using the same song recordings across plots may have resulted in pseudoreplication in 2021, therefore creating biased results, whereas in 2023 song recordings from different males were used. However, the phenomenon of attracting floaters to GQ treatment plots was in agreement with the results obtained in the central part of the Wood Warbler range (Szymkowiak et al. 2016). On the other hand, in 2023, only 76.5% of the plots in the DSF were occupied by wood warblers, despite the song playback, meaning that if there were no residents, then there were either few floaters or none at all. Hence the plots we selected might not be equally attractive to wood warbler as we mentioned above.

There is some experimental evidence suggesting that the presence of the common chiffchaff can also attract wood warblers and influence them to settle nearby (Szymkowiak et al. 2017). However, our results demonstrated that the common chiffchaff abundance was positively correlated with the number of residents and not with the number of floaters or the total number of males per plot. There is hardly any benefit for wood warblers to gather post-breeding social cues in the current year to make settlement decisions for the following year, since only 0.8% of ringed adult males returned next year and the return rate of ringed nestling is even lower (Sokolov et al. 1996; Lapshin 2009). It is conceivable that wood warblers may collect post-breeding cues during juvenile and autumn migration, far away from the current breeding site, but we do not have enough data to confirm this.

Clustering of male territories is another common feature of the Wood Warbler even though not all males in a cluster succeed in attracting females (Herremans 1993; Norman 1994; Grendelmeier et al. 2016; Luepold 2023). However, the ‘hidden lek’ hypothesis postulates that aggregations of territorial birds may as well increase their chances for extra-pair copulation (Wagner 1998; Fletcher and Miller 2006). Since each female is ready to be fertilised only during short periods of time, males should gather around the nest during egg-laying.

Despite the new clutch appearing asynchronously throughout the breeding season – from 22 May to 13 July (53 days), the fluctuations in male abundance were not caused only by the arrival of new birds from wintering grounds or the decreasing number of birds as the season progressed. Within each plot, nonlinear dynamics are observed: the number of males increases during the egg-laying in a local nest, and then decreases after the incubation start (Fig. 2), which can occur several times on the plot depending on the number of nests. However, the degree of the fluctuations decreases towards the end of the season. Both of these phenomena describe the local population dynamics within a season better than all factors independently (Table 3, model 5). Moreover, a greater increase in male abundance on a plot was predicted by the nest presence at the egg-laying stage (Table 2, model 5). During our field work, we could detect the female arrival by the appearance of new males in the aggregation, sometimes even before registering her presence ourselves. Furthermore, we observed that most unpaired males in aggregations leave the area immediately after the end of the copulation period.

So far, our results support the previous findings (Luepold 2023) that plot quality is a better predictor of resident abundance than song playback. Consequently, according to our hypothesis, the presence of paired males should also attract floaters to the plots. In this study, most of the residents have paired with female eventually. However, no correlation was found between residence and floater abundance in the current year. This was not surprising since not all floaters implement an opportunistic strategy. The difference between ‘residents’ and ‘floaters’ is not quite the same as between ‘conservative’ and ‘opportunistic’ males, as some males may be classified as floaters if they found the plot unsuitable, died or left the site too early due to nest depredation.

According to the ‘hidden lek’ hypothesis, some floaters do not reside on a permanent site but move through the territories of other males in order to copulate with as many females as possible. This leads to the question of why floaters may be attracted to GQ treatment plots with a high rate song broadcast. As it has been previously shown (Szymkowiak et al. 2016), males with a higher song rate are more likely to pair with a female. It is also known that a certain phrase of a song and its song rate are significantly associated with testosterone levels in wood warblers, suggesting that some males may sing more often but invest less time in guarding the female (Belokon et al. 2020). Thus, floaters are more likely to copulate with females who paired with a higher song rate singing males. However, this assumption requires further research.

The reasons of nomadism

The Wood Warbler is described as a nomadic species due to large interannual fluctuations in the bird’s local abundance and low nest-site fidelity (Norman 1994; Wesołowski et al. 2009; Teitelbaum and Mueller 2019). For some reasons most adult wood warblers do not return to the previous breeding site after wintering. Annual return rates of ringed adult birds varies greatly depending on the region and reaches a maximum of 28% in Western Europe (Norman 1994), while in Central and Eastern Europe only a few birds were found or recaptured again (Herremans 1993; Sokolov et al. 1996; Lapshin 2009). We strongly doubt that a high mortality rate could be the cause, since the species did not increase reproduction success to compensate for the loss of individuals (Wesołowski and Maziarz 2009; Maag et al. 2022), and it does not show any tendency towards extinction except in western territories (Vickery et al. 2014). Apart from that, the wood warblers exhibit plasticity in feeding behaviour (Herremans 1993; Maziarz and Wesołowski 2010; Mallord et al. 2017), so the distribution of food resources is not the cause of nomadic patterns either.

Local numbers of birds may fluctuate because wood warblers avoid settling in areas with a large number of rodents and predators that depredate their ground nests (Szymkowiak and Kuczynski 2015; Szymkowiak and Thomson 2019). The distribution of rodents may vary annually due to many factors, so wood warblers may adjust to these fluctuations, and thus display low site fidelity. Nevertheless, there is no large-scale increase in reproductive success (Wesołowski and Maziarz 2009; Lapshin 2020; Maag et al. 2022), so it is not obvious what benefit they receive compared to other ground-nesting species with high site fidelity. The number of wood warblers is negatively correlated with the abundance of bank vole Clethrionomys glareolus and yellow-necked mouse Apodemus flavicollis (Wesołowski et al. 2009; Gerber 2011), but there is no evidence that birds deliberately avoid rodents or may perceive their abundance using acoustic signals (Stelbrink et al. 2019). Whether they can use visual (Gerber 2011) or olfactory cues to estimate the local number of rodents is unknown.

In accordance with our data, we put forward the hypothesis that males have two reproductive strategies: conservative and opportunistic. Conservative males typically display nest-site fidelity and bond with females, while opportunistic males show low nest-site fidelity, regularly changing their territory shortly after copulation with a paired female. The opportunistic strategy aims to leave as many extra-pair offspring in local nests as possible and, as a result, it drives nomadic behaviour of males as well. This is in agreement with other studies that have shown that within-season movements of wood warblers are related to mate searching (Luepold et al. 2024), and that most fathers of extra-pair young are hard to identify because they have left the area (Goretskaia et al. 2024).

In 2021–2023, the population remained unstable during the breeding season, as between 52 and 66.7% of all detected males left the territory. Based on our assumption, at least part of the floaters implement opportunistic strategies, therefore, they do not pair with females during the breeding season and do not help rear offspring. Then, if the male does not have a permanent nest-site to attract a female, he does not need to return to the place where the researcher banded him. This behaviour partly explains why we observed so few ringed males returning in the following years.

For the Wood Warbler, which is considered to be a social monogamous species, the number of males adopting opportunistic strategies may depend on the following territorial conditions. Firstly, on territories where males greatly outnumber females. Secondly, on territories where birds arrive and start breeding in asynchrony. In central areas of the range the breeding period is relatively long and periods for the mate search is also longer than on the northern periphery of the range, where more males are forced to be involved in extra-pair mating rather than attracting a mate. So, it appears that in the most optimal habitats the return rate increases to 28% (Norman 1994), while in the north the return rate of males remains critically low (0.8% or less).

Although the mortality rate of females is usually higher than that of males in birds (Xirocostas et al. 2020; Payevsky 2021), the reasons for the greater predominance of wood warbler males over females compared to other passerine species are not entirely clarified. An additional explanation for the higher female mortality may be winter habitat segregation among the sexes in the Wood Warbler (Hobson et al. 2014). On the other hand, the constant movements of males should also distort the image of the actual sex ratio in the local population.

At the end of our study we were left with some interesting questions for further investigation: do males switch between strategies during the season or between years? Is there a genetic difference between males with different strategies? Our result also showed that there was no difference in the length of the wing and tail between residents and floaters, as well as in similar results obtained in another study, when paired and unpaired males were compared (Herremans 1993). It would be curious to determine if first-year males are more likely to try to attract a female, while older males likely switch to an opportunistic strategy. This could be the factor that reduces the return rate of successfully nesting birds in following years as well. The main difficulty we face is that we cannot determine yet, where the floaters fly away to within the breeding season and next year. Apart from that, another problem lies in determining the age of wood warblers, which occurs due to the complete moult during the winter season. The number of extra-pair offspring also appears to vary greatly between studies (Grendelmeier et al. 2016; Goretskaia et al. 2024), which may be due to the use of different markers to determine kinship estimation in a given species. Thus, these outcomes point to the need for further investigation of the extra-pair involvement of floaters in the reproduction of the young and their contribution to the genetic diversity of the population.

Acknowledgments We thank D.N. Tolstov, A.S. Simonov and M.V. Matanseva for help in preparing the equipment for the experiments and assistance in the statistical processing. We are grateful to T. A. Rymkevich, V.A. Ryzhenkova, D.A. Starikov and D.A. Vasiliev for consultations and for assistance in the field. We are also very grateful to Yulia Novikova for valuable notes and suggestions that helped to improve the manuscript.

Funding The study was supported by a grant from the Russian Science Foundation № 23-24-00415, https://rscf.ru/project/23-24-00415/

Data availability The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files.

Ethical approval. This field work is licensed under a Northwestern Interregional Department of the Russian Natural Surveillance. Permission No. 013 was received for the use of flora and fauna objects located in specially protected natural areas of federal significance, which permits capturing, banding and then releasing the birds into their natural habitat. In this study, the wood warblers were released immediately after measurements and resumed singing activities soon thereafter. All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in animal studies were in accordance with the ethical standards of the institution or practice in which the training is provided were held.

Author contributions Conceptualization: KA (equal), LN (equal); Data curation: KA (lead), LN (supporting); Formal analysis: KA (lead), LN (supporting); Funding acquisition: LN (equal), KA (equal); Investigation: KA; Methodology: KA (equal), LN (equal); Supervision: LN; Visualization: KA; Writing – original draft: KA; Writing – review and editing: KA

Conflict of interest The authors declare that they have no conflict of interest.

Ahlering MA, Faaborg J (2006) Avian habitat management meets conspecific attraction: If you build it, will they come? The Auk 123:301–312. https://doi.org/10.1093/auk/123.2.301
Arel-Bundock V (2022) modelsummary: data and model summaries in R. J. Stat. Softw. 103: 1–23 https://doi.org/10.18637/jss.v103.i01
Bates D, Maechler M, Bolker B, Walker S (2014) lme4: Linear mixedeffects models using Eigen and S4. R package version 1:1–23, http://CRAN.R-project.org/package=lme4
Belokon’ MM, Belokon’ YuS, Goretskaya MYa (2020) Extra-pair offspring, song structure and the level of testosterone in the wood warbler in central Russia. In: Ornithological research in the countries of Northern Eurasia. Beloruskaya navuka, Minsk, pp. 65–66
Broughton RK, Bubnicki JW, Maziarz M. (2020) Multi-scale settlement patterns of a migratory songbird in a european primaeval forest. Behavioral Ecology and Sociobiology 74:128. https://doi.org/10.1007/s00265-020-02906-0
Brumm H, Ritschard M (2011) Song amplitude affects territorial aggression of male receivers in chaffinches. Behavioral Ecology 22:310–316. https://doi.org/10.1093/beheco/ars012
Burnham KP, Anderson DR, Huyvaert KP (2011) Aic model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons. Behavioral ecology and sociobiology 65:23–35. https://doi.org/10.1007/s00265-010-1084-z
Buxton VL, Enos JK, Sperry JH, Ward MP (2020) A review of conspecific attraction for habitat selection across taxa. Ecology and evolution 10:12690–12699. https://doi.org/10.1002/ece3.6922
Catchpole CK, Slater PJ (2003) Bird song: biological themes and variations. Cambridge university press
Collins S (2004) Vocal fighting and flirting: the functions of birdsong. Nature’s music: the science of birdsong, pp. 39–79. https://doi.org/10.1016/B978-012473070-0/50005-0
Cramer ER (2013) Physically challenging song traits, male quality, and reproductive success in house wrens. PLoS One 8:e59208. https://doi.org/10.1371/journal.pone.0059208
Danchin E, Giraldeau L-A, Valone TJ, Wagner RH (2004). Public information: from nosy neighbors to cultural evolution. Science 305:487–491. https://doi.org/10.1126/science.1098254
Fletcher Jr RJ, Miller CW (2006) On the evolution of hidden leks and the implications for reproductive and habitat selection behaviours. Animal Behaviour 71:1247–1251. https://doi.org/10.1016/j.anbehav.2005.11.007
Garamszegi LZ, Møller AP, T¨or¨ok J, Michl G, P´eczely P, Richard M (2004) Immune challenge mediates vocal communication in a passerine bird: an experiment. Behavioral Ecology 15:148–157. https://doi.org/10.1093/beheco/arg108
Gerber, M (2011). Territory choice of the Wood Warbler Phylloscopus sibilatrix in Switzerland in relation to habitat structure and rodent density. Unpubl MSc thesis, Zurich University
Goretskaia M, Belokon Y, Belokon M (2024) Extra-pair paternity in the Wood Warbler (Phylloscopus sibilatrix) in Central Russia. Biology Bulletin 51:152–158. https://doi.org/10.31857/S1026347024010096
Goretskaia M, Beme I (2021) Influence of testosterone on different aspects of bird behavior and physiology. Biology Bulletin 48:1323–133 https://doi.org/10.31857/S1026347024010096
Grendelmeier A, Arlettaz R, Olano-Marin J, Pasinelli G (2016) Experimentally provided conspecific cues boost bird territory density but not breeding performance. Behavioral Ecology 28:174–185. https://doi.org/10.1093/beheco/arw144
Hammer TL, Bize P, Gineste B, Robin J-P, Groscolas R, Viblanc VA (2023) Disentangling the “many-eyes”, “dilution effect”, “selfish herd”, and “distracted prey” hypotheses in shaping alert and flight initiation distance in a colonial seabird. Behavioural Processes 210:104919. https://doi.org/10.1016/j.beproc.2023.104919
Hartig F (2022) DHARMa: residual diagnostics for hierarchical (multi-level/ mixed) regression models. R package version 0.4.6. http://florianhartig. github.io/DHARMa/
Herremans, M. (1993). Clustering of territories in the Wood Warbler Phylloscopus sibilatrix. Bird Study 40:12–23. https://doi.org/10.1080/00063659309477124
Hobson KA, Van Wilgenburg SL, Wesoowski T, Maziarz M, Bijlsma RG, Grendelmeier A, Mallord JW (2014) A multiisotope (δ 2 H, δ 13 C, δ 15 N) approach to establishing migratory connectivity in palearctic-afrotropical migrants: an example using Wood Warblers Phylloscopus sibilatrix. Acta Ornithologica 49:59–71. https://doi.org/10.3161/000164514X682896
Kelly JK, Ward MP (2017) Do songbirds attend to song categories when selecting breeding habitat? A case study with a wood warbler. Behaviour 154:1123–1144. https://doi.org/10.1163/1568539X-00003461
Kramer AM, Dennis B, Liebhold AM, Drake JM (2009) The evidence for allee effects. Population Ecology 51:341–354. https://doi.org/10.1007/s10144-009-0152-6
Lampe HM, Espmark YO (1994) Song structure reflects male quality in pied flycatchers, Ficedula hypoleuca. Animal Behaviour 47:869–876. https://doi.org/10.1163/156853900502358
Lapshin N (2009) Specific features of annual cycles in long-distance migrant birds living at the range boundary: the example of the wood warbler, Phylloscopus sibilatrix, in the taiga zone of northwestern Russia. Russian journal of ecology 40:199–205. https://doi.org/10.1134/S1067413609030096
Luepold SB, Kokko H, Grendelmeier A, Pasinelli G (2023) Habitat detection, habitat choice copying or mating benefits: What drives conspecific attraction in a nomadic songbird? Journal of Animal Ecology 92:195–206. https://doi.org/10.1111/1365-2656.13844
Luepold SB, Korner-Nievergelt F, Züst Z, Pasinelli G (2024) It’s about her: Male within-season movements are related to mate searching in a songbird. The American Naturalist 203:562–575. https://doi.org/10.1086/729424
Maag N, Burgess MD, Maziarz M, Lüpold S, Mallord JW, Broughton RK, Cristinacce A, Arlettaz R, Carlotti S, Castello J et al (2022) Reproductive success of the wood warbler Phylloscopus sibilatrix varies across Europe. Journal of Avian Biology 2022:e03033. https://doi.org/10.1111/jav.03033
Mallord JW, Orsman CJ, Cristinacce A, Stowe TJ, Charman EC, Gregory RD (2017) Diet flexibility in a declining long distance migrant may allow it to escape the consequences of phenological mismatch with its caterpillar food supply. Ibis 159:76–90. https://doi.org/10.1111/ibi.12437
Matantseva MV, Lapshin NV, Simonov SA (2015) Ethological aspects of biodiversity within and between Phylloscopus species: behavioral variation among birds from the centre and periphery of breeding ranges. Acta Biologica Universitatis Daugavpiliensis 15:273–283
Maziarz M, Wesołowski T (2010) Timing of breeding and nestling diet of Wood Warbler Phylloscopus sibilatrix in relation to changing food supply. Bird Study 57:540–552. https://doi.org/10.1080/00063657.2010.512954
McQuarrie AD, Tsai C-L (1998) Regression and time series model selection. World Scientific. https://doi.org/10.2307/2669491
Morinay J, Forsman JT, Doligez B (2020) Heterospecific song quality as social information for settlement decisions: an experimental approach in a wild bird. Animal behaviour 161:103–113. https://doi.org/10.1101/748103
Nemeth E, Kempenaers B, Matessi G, Brumm H (2012) Rock sparrow song reflects male age and reproductive success. PLOS ONE 7: e43259. https://doi.org/10.1371/journal.pone.0043259
Nocera JJ, Betts MG (2010) The role of social information in avian habitat selection. The Condor 112:222–224. https://doi.org/10.1525/cond.2010.100005
Norman D (1994) The return rate of adult male Wood Warblers Phylloscopus sibilatrix to a peripheral breeding area. Ringing & Migration 15:79–83. https://doi.org/10.1080/03078698.1994.9674077
Payevsky V (2021) Sex ratio and sex-specific survival in avian populations: A review. Biology Bulletin Reviews 11:317–327. https://doi.org/10.1080/03078698.1994.9674077
R Core Team (2014) R: a language and environment for statistical computing. In: R Foundation for Statistical Computing, Vienna, Austria http://www.R-project.org
Seppänen J-T, Forsman JT, Mönkkönen M, Thomson RL (2007) Social information use is a process across time, space, and ecology, reaching heterospecifics. Ecology 88:1622–1633. https://doi.org/10.1890/06-1757.1
Simms E (1985) British Warblers. Collins, London
Spencer K, Buchanan K, Goldsmith A, Catchpole C (2004). Developmental stress, social rank and song complexity in the European starling (Sturnus vulgaris). Proceedings of the Royal Society of London. Series B: Biological Sciences 271:S121–S123. https://doi.org/10.1098/rsbl.2003.0122
Stamps J (2001) Habitat selection by dispersers: integrating proximate and ultimate approaches. In: Clobert J, Danchin E, Dhondt AA, Nichols JD (ed): Dispersal, pp 110–122. https://doi.org/10.1093/oso/9780198506607.003.0018
Stamps JA (1988) Conspecific attraction and aggregation in territorial species. The American Naturalist 131:329–347. https://doi.org/10.1086/284793
Stelbrink P, Grendelmeier A, Schabo D, Arlettaz R, Hillig F, Pasinelli G (2019) Does acoustically simulated predation risk affect settlement and reproduction of a migratory passerine? Ethology 125:535–547. https://doi.org/10.1111/eth.12879
Sokolov LV, Payevsky VA, Markovets MJ, Morozov Y (1996) The comparative analysis of the levels of phylopatry, postfledging dispersion and population changes in willow warbler (Phylloscopus trochilus) and Wood Warbler (Ph. sibilatrix) on Courish Spit (the BalticSea). Zoologichesky zhurnal 75: 565–578
Szymkowiak J (2013) Facing uncertainty: how small songbirds acquire and use social information in habitat selection process? Springer Science Reviews 1:115–131. https://doi.org/10.1007/s40362-013-0012-9
Szymkowiak J, Kuczyński L (2015) Avoiding predators in a fluctuating environment: responses of the wood warbler to pulsed resources. Behavioral Ecology 26:601–608. https://doi.org/10.1093/beheco/aru237
Szymkowiak J, Kuczyński L (2017) Song rate as a signal of male aggressiveness during territorial contests in the wood warbler. Journal of Avian Biology 48: 275–283. https://doi.org/10.1111/jav.00969
Szymkowiak J, Thomson RL (2019) Nest predator avoidance during habitat selection of a songbird varies with mast peaks and troughs. Behavioral ecology and sociobiology 73:1–11. https://doi.org/10.1007/s00265-019-2702-z
Szymkowiak J, Thomson RL, Kuczyński L (2016) Wood warblers copy settlement decisions of poor quality conspecifics: support for the tradeoff between the benefit of social information use and competition avoidance. Oikos 125:1561–1569. https://doi.org/10.1111/oik.03052
Szymkowiak J, Thomson RL, Kuczyński L (2017) Interspecific social information use in habitat selection decisions among migrant songbirds. Behavioral Ecology 28:767–775. https://doi.org/10.1093/beheco/arx029
Teitelbaum CS, Mueller T (2019) Beyond migration: causes and consequences of nomadic animal movements. Trends in ecology & evolution 34:569–581. https://doi.org/10.1016/j.tree.2019.02.005
Temrin H (1986) Singing behaviour in relation to polyterritorial polygyny in the Wood Warbler (Phylloscopus sibilatrix). Animal Behaviour 34:146–152. https://doi.org/10.1080/00063659309477124
Valente JJ, LeGrande-Rolls CL, Rivers JW, Tucker AM, Fischer RA, Betts MG (2021) Conspecific attraction for conservation and management of terrestrial breeding birds: Current knowledge and future research directions. The Condor 123:1-15. https://doi.org/10.1093/ornithapp/duab007
Van Duyse E, Pinxten R, Eens M (2002). Effects of testosterone on song, aggression, and nestling feeding behavior in male great tits, Parus major. Hormones and Behavior 41:178–186. https://doi.org/10.1006/hbeh.2001.1747
Vickery JA, Ewing SR, Smith KW, Pain DJ, Bairlein F, Skorpilov´a J, Gregory RD (2014) The decline of Afro-Palaearctic migrants and an assessment of potential causes. Ibis 156:1–22. https://doi.org/10.1111/ibi.12118
Wagner RH (1998) Hidden leks: sexual selection and the clustering of avian territories. Ornithological Monographs 4: 123–145
Wesołowski T, Maziarz M (2009) Changes in breeding phenology and performance of Wood Warblers Phylloscopus sibilatrix in a primeval forest: a thirty-year perspective. Acta Ornithologica 44:69–80. https://doi.org/10.3161/000164509X464902
Wesołowski T, Rowiński P, Maziarz M (2009) Wood Warbler Phylloscopus sibilatrix: a nomadic insectivore in search of safe breeding grounds? Bird Study 56:26–33. https://doi.org/10.1080/00063650802681540
Xirocostas ZA, Everingham SE, Moles AT (2020) The sex with the reduced sex chromosome dies earlier: a comparison across the tree of life. Biology letters 16:20190867. https://doi.org/10.1098/rsbl.2019.0867

SupplementarymaterialBEASD2400235.xlsx

Download PDF

Version 1

posted

You are reading this latest preprint version

The Divergence of Male Reproductive Strategy as the Cause of Nomadism in Wood Warbler

Status:

Version 1

Abstract

Figures

Significance Statement

Introduction

Methods

Study site and broadcast manipulation

Bird data

Data analysis

Results

Annual return rate and bird monitoring

Settlement response to song playback

Discussion

The reasons of nomadism

Declarations

References

Supplementary Files

Status:

Version 1