Abstract
Records of European long-distance migrant birds wintering in the Western Palaearctic have increased recently, possibly in response to milder climatic conditions due to climate change. The spotted flycatcher (Muscicapa striata), traditionally a trans-Saharan migrant, has recently been observed wintering north of the Sahara Desert. This study examines the winter status of this species in the Western Palaearctic and identifies climatically favourable yet unoccupied wintering areas. We compiled wintering records from 2000 to 2024 using the citizen science platform eBird. We then modelled the distribution of these data across Europe and the Middle East using 19 climatic and two topographic variables. Our results provide the first evidence of a regular wintering pattern of the species in the Western Palaearctic over the last decade, with over 80% of the records located on the Iberian Peninsula and the Balearic Islands. This aligns with positive temperature anomalies in Europe between 2014 and 2023. We identified four climatic variables as significant drivers, with the diurnal temperature range and the temperature of the coldest quarter within the study area being the strongest predictors. Favourable wintering areas encompass much of the Mediterranean basin and large Mediterranean islands. Using fuzzy logic, we produced ensemble models of these variables under different hypotheses regarding their relative importance, tested against independent North African wintering records. The most supported hypothesis indicates that while favourable values in the four variables are important, they are not essential, meaning that favourable values in some variables can compensate for less favourable values in others. This species may serve as a model for other trans-Saharan migrants now wintering in the Western Palaearctic. If temperatures continue to increase, more individuals are expected to shift their wintering strategy from trans-Saharan to pre-Saharan migrants, leading to an increase in the populations regularly wintering along Mediterranean coasts and islands.
Similar content being viewed by others
Introduction
Bird migration is one of the greatest wildlife spectacles, producing massive global changes in bird distributions twice yearly. It is an evolutionary adaptation to find areas with sufficient resource availability during the non-breeding season to ultimately increase survival and breeding competition1. Migration strategies (e.g. direction, time, distance, and stopovers), the onset of migration and physiological adaptations have a genetic basis2,3,4,5,6in addition to the natural learning process of juveniles from experienced individuals7. However, environmental changes can affect the manifestation of this genetic base, allowing a species to modify its migratory behaviour according to the current conditions8,9. This flexibility allows a variable response that ultimately benefits the persistence of the population if environmental conditions change in a short time10,11. Individuals have two types of short-term response mechanisms to the environment: the microevolutionary response to natural selection, which is a ‘non-labile’ trait that remains stable throughout an individual’s lifetime8,12,13,14,15and phenotypic plasticity, which is mainly ‘labile’ and can change repeatedly to fit the new environmental conditions10,11,16,17. These changes are currently influencing the distribution range of some species18,19,20,21as well as the migratory behaviour of others, since individuals are adapting their phenology in response to environmental shifts, such as climate-driven changes22,23,24,25.
Migration is the part of the annual cycle with the highest mortality rate, especially for long-distance migrants which often have to cross large ecological barriers26,27. A shorter migration distance could provide benefits to individuals, such as higher survival28; earlier arrival at their breeding grounds, allowing them to acquire breeding territories earlier and take full advantage of the available food resources29,30,31; and improved fitness for the next breeding season32. Therefore, if long-distance migrants establish wintering sites closer to their breeding grounds, they could become short-distance migrants.
Over the past few decades, temperatures have increased, and winters have become milder in Europe and Africa33. If favourable conditions for long-distance migratory birds persist during autumn and winter, some species could delay or even forgo migration and remain in the Western Palaearctic22,34. An increasing number of records show that long-distance bird migrants breeding in Europe have been wintering in the Iberian Peninsula over the last decades35such as the white stork (Ciconia ciconia)36the barn swallow (Hirundo rustica)37and the Eurasian reed warbler (Acrocephalus scirpaceus)25,38. Other examples include the Eurasian hoopoe (Upupa epops), the Eurasian wryneck (Jynx torquilla), and the yellow wagtail (Motacilla flava)37. These observations indicate that in recent decades, Southern Europe, particularly the south of the Iberian Peninsula, has offered new climatic conditions and wintering areas sufficiently suitable to promote a reduction in migration distance39. In addition, these temperature changes facilitate a northward expansion of African birds which have colonised Southern Europe during the past few decades18,19,40,41,42,43,44.
In this study, we focus on the spotted flycatcher (Muscicapa striata), a Palaearctic long-distance migrant that breeds in Northwest Africa and across Europe into Central Asia, and winters south of the Sahara Desert45. In recent years, the international citizen science platform eBird (https://ebird.org/home) has shown some records of the species wintering north of the Sahara, which are currently considered anecdotal cases. While the number of wintering records has been increasing, no systematic study has provided a detailed overview of the current situation of the species in the Western Palaearctic during winter, which may be relevant in the current context of climate change. Therefore, this study primarily aimed to examine the status of the spotted flycatcher during winter in the Western Palaearctic. For this purpose, we used eBird data for Europe, North Africa, and the Middle East from 2000 to 2024 to obtain information about its spatial and temporal distribution during winter. Our assumption was based on the increasing temperatures over the last two decades33. We also hypothesised that this species has even more favourable wintering areas in the Western Palaearctic than it currently occupies, which could lead to an increase in wintering populations in this region if climatic conditions remain favourable. To demonstrate this, we derived the characteristics of the spotted flycatcher’s current wintering areas in the Western Palaearctic and identified similar climatic regions that could become future wintering sites based on the short-term predictions from the distribution models18. To this end, we formulated four hypotheses. Hypothesis 1 (H1): all variables included in the model are relevant for the wintering of the species, but none are limiting; Hypothesis 2 (H2): all variables included in the model are limiting for the wintering of the species; Hypothesis 3 (H3): temperature and precipitation are generally limiting for the wintering of the species; Hypothesis 4 (H4): the winter temperature, along with the combination of the other variables included in the model, limit the wintering of the species. Temperature conditions influence food availability for insectivorous species during winter, mainly the abundance of flying insects46,47. Therefore, in H4, we have considered the temperature of the coldest quarter within the study area as one of the limiting variables. This research could be considered as an example for analogous studies of other insectivorous trans-Saharan migrants that are currently changing their phenology and wintering grounds.
Methods
Study area
The study area comprised the land area from 15° W to 65° E and from 20° N to 70° N, covering the Western Palaearctic and encompassing Europe, the Middle East, and North Africa. This area was divided using a hexagonal grid, with each hexagon representing an operational geographic unit (OGU) spanning an area of 7,774 km2. This biogeographical area had a total of 2,759 OGUs (Fig. 1). The Western Palaearctic is a region with heterogeneous physiography and climate, including desert, Mediterranean, Atlantic, tundra, and boreal climates48,49,50.
The study area divided into operational geographic units (OGUs). Map projection: GCS_WGS_1984 (ArcGIS 10.4.1).
The selected area allows for the effects of climate change on the wintering of trans-Saharan birds, such as the spotted flycatcher, to be studied, since any record of the species during the wintering period in Europe, the Middle East or North Africa is unusual.
The study species
The spotted flycatcher is a migratory insectivorous passerine, and is the most common and widespread flycatcher in the Western Palaearctic51. It breeds from mid-May to mid-August across Europe into Central Asia, and from April to July in Northwest Africa51. All populations traditionally winter in sub-Saharan Africa; however, over the last two decades, records from eBird have shown some individuals wintering north of the Sahara. Spring migration from the wintering grounds in sub-Saharan Africa to the Western Palaearctic begins in late February and continues until June, while autumn migration to the south starts in mid-August and ends in mid-November, with peak activity occurring in September and October45. The species effectively inhabits any open woodland or timbered area with raised perches providing an open view. In breeding areas, it prefers well-spaced mature trees and has, therefore, adapted well to avenues, parks, gardens, orchards, and other artificial habitats. It also occupies many types of deciduous or coniferous woodland and is found in trees along streams, rivers, and the edges of standing water. It occupies similar habitats in African non-breeding quarters45.
Species distribution data
We obtained wintering records of the species from the eBird platform from 1 December 2000 to 31 January 2024. We considered records between 1 December and 31 January as wintering to avoid late post-breeding and early pre-breeding migrants, respectively. All records were verified using pictures of each observation. When multiple records indicated the same or very close locations on similar dates within the same wintering period, it was assumed to be the same individual, and only one record was retained.
Predictor variables used for distribution modelling and climate data
A set of 21 environmental variables related to topography (n = 2) and climate (n = 19) between 1981 and 2010 were used to model the wintering distribution of the species in the Western Palaearctic (Table 1). We included topography along with climate in the modelling approach to avoid conflating their distinct effects and mistakenly attributing them solely to climate52,53.
These variables were downloaded in raster format with an approximately 1 km2 pixel resolution. The values of these variables at each OGU were obtained by averaging the 1 km2 pixel values within each OGU using the zonal statistic as table function in ArcGIS version 10.4.154, providing an average value for the surrounding environment in which each presence (and absence) was recorded.
Data sources: (1) US Geological Survey (1996)55; (2) Elaborated from the digital elevation model using the altitude variable55 with the geographic information system ArcGIS Desktop 10.4.1; (3) Chelsa Climate (http://chelsa-climate.org). The term ‘quarter’ refers to a three-month period (i.e. trimester) in this context.
The average estimated surface temperature change in Europe according to 1991–2020 reference levels (°C) for each year from 1979 to 2023 was downloaded from Copernicus Climate Change Service (C3S, https://climate.copernicus.eu). This estimated temperature change was then averaged for the periods 1979–2000, 2001–2013, and 2014–2023.
Modelling procedure
We used logistic regression analysis to identify potentially favourable areas where spotted flycatchers might appear during winter but are not currently detected. Logistic regression is a supervised machine learning algorithm that allows the use of environmental variables, such as topographic and climatic factors (Table 1), to explain the binary dependent variable (presence/absence)56. Understanding the environmental conditions that explain the current winter distribution of the spotted flycatcher could help identify other areas with similar environmental characteristics that might also support individuals during this time of the year.
We initially calculated Spearman’s correlation coefficients between variables to reduce multicollinearity among them57,58. For each pair of variables with r > 0.8, only the variable that showed the most significant relationship with the species’ wintering distribution was retained, according to the statistical significance of Rao’s score test59. The type II errors for the null hypotheses were controlled using the score test (p < 0.05), while the type I errors for the alternative hypotheses were managed using the false discovery rate (q < 0.05)60,61,62. Due to the high number of variables, we controlled for type I errors by implementing statistical corrections to minimise the likelihood of false positives, ensuring our findings are more robust and reliable. On the subset of variables that passed the two previous filters, we tested the response of the winter distribution of the species to each explanatory variable in Europe and the Middle East. This response can be monotonic or unimodal, in both cases with possible plateaus of no additional response of the species to the environment (Table S1). A monotonic response implies that the relationship of the explanatory variable with the wintering presence of the species is either positive or negative. In contrast, a unimodal response implies that after an optimum value in the environmental gradient is reached, the initially positive linear relationship becomes negative. To this end, binary stepwise logistic regression models were performed separately for each explanatory variable, using their original (x) and quadratic (x2)forms. A response was considered unimodal when the best logistic regression model included the original form as positive and the quadratic form as negative (x, −x2), as this was the only biologically plausible unimodal response. For all variables that showed this type of unimodal response, the result of the unimodal model (y or logit function) was used as an explanatory variable in the final modelling process63,64. In all other cases, the original form of the variable was used. Then, we performed an ensemble forecasting with the remaining subset of variables that passed the above filters using a multivariate stepwise logistic regression to obtain the probability of wintering in each OGU. This process started with a null model containing no predictor variables that produced a constant probability of wintering for each OGU equal to the prevalence (the proportion of OGUs where species has been confirmed to winter relative to the total number of OGUs). In the first step, the procedure selected the variable with the most significant relationship to the winter distribution of the species, according to the significance of the Rao’s score test59. In the following steps, it added the variable most significantly related to the residuals not explained by the previous step, until no additional variable significantly increased the predictive capacity of the model65. Thus, the environmental variables included in the multivariate model, an ensemble model, were a parsimonious representation of all the effects attributable to the set of variables analysed in the stepwise procedure. Finally, we obtained a significant combination of predictor variables (y or logit), whose coefficients were estimated using a machine learning algorithm based on a gradient ascent of likelihood66,67. As this was unlikely for highly correlated variables, this stepwise procedure also contributed to reduce multicollinearity among explanatory variables. The relative weight of each variable in the final model was assessed using the Wald test68 and the remaining multicollinearity between them by calculating the Variance Inflation Factor (VIF). The different steps followed to select the predictor variables are summarised in the supplementary material (see Fig. S1).We identified the OGUs where the species was confirmed to winter in the Western Palaearctic and then used the wintering occurrences in Europe and the Middle East to model the distribution of the species in this region, as North Africa is under-sampled and, by excluding it, we reduced the number of false absences included in the models. We also excluded Gibraltar from the modelling process, as its OGU was mostly located on the African continent. We then checked that there were no problems in extrapolating the model to North Africa by confirming that the mathematical domain of the function of the unimodal variables that entered the model was within the logical domain in the training area (i.e. Europe and the Middle East; Figs. S2–S5). We then extrapolated the model to North Africa. The logistic regression analysis provides a probabilistic output which details the probability of finding a spotted flycatcher in a given area based on its environmental characteristics. Since the prevalence of records may be underestimated, especially in regions with less citizen science activity or lower participation, it is essential to account for this bias. To address this, we transformed the probability outputs into favourability outputs using the favourability function69:
where n1 and n0 represent the number of wintering presences and absences in the study area, respectively. Favourability (F) values range from 0 (unfavourable environmental conditions) to 1 (highly favourable environmental conditions), with a value of 0.5 suggesting that the likelihood of the species’ presence is equal to its prevalence in the study area. Therefore, favourability refers to the extent to which environmental conditions favour the presence of the spotted flycatcher during the winter. Our modelling approach conforms to modelling protocols proposed by Sillero et al. (2021)70. All modelling processes were run with the IBM SPSS Statistics 25 software package, and maps were created using ArcMap software (ArcGIS version 10.4.1; https://desktop.arcgis.com/es/arcmap/).
Hypotheses testing through models’ assessment and validation
The favourability model derived from the multivariate logistic regression probability values was used to test H1. To test H2, we determined the fuzzy intersection between the favourability outputs of the univariate models of each variable included in the multivariate model. The degree of membership of each OGU to the fuzzy intersection was defined as the minimum value of favourability among the univariate models in each OGU19,71. To test H3, we first calculated the fuzzy union of the favourability outputs from the univariate models separately for temperature-related and precipitation-related variables. The degree of membership of each OGU to the fuzzy union was defined as the maximum value of favourability among the univariate models in each OGU19,71. We then calculated the fuzzy intersection between the favourability outputs of the temperature and precipitation models. To test H4, we initially computed the fuzzy union of the favourability outputs from the univariate models for all variables except the variable mean daily air temperatures of the coldest quarter. We then calculated the fuzzy intersection between the favourability of this temperature variable and the fuzzy union of all other variables. We also calculated the mean favourability in Europe and the Middle East, and separately in North Africa, from these models.
The resulting univariate and ensemble distribution models, as well as their corresponding hypotheses, were assessed and compared according to their discrimination and classification capacity in Europe and the Middle East, and when extrapolated to North Africa. The discrimination capacity was evaluated using the area under the receiver operating characteristic curve (AUC)72. The classification capacity, using F = 0.5 as the classification threshold, was assessed through the following measures: sensitivity (conditional probability of OGUs with reported wintering presences classified as favourable), specificity (conditional probability of OGUs with no reported wintering presences classified as unfavourable), correct classification rate (CCR; conditional probability of correctly classified OGUs), the over-prediction rate (OPR; the proportion of OGUs with no reported wintering presences in an area with F > 0.5), and the under-prediction rate (UPR; the proportion of OGUs with reported wintering presences in an area with F < 0.5). These measures are widely used, with values ranging from 0 to 173,74. We also used Cohen’s kappa index75whose values range from − 1 to + 1, to measure the degree to which the favourability of OGUs with reported or no reported wintering presences in the dataset was greater or less than 0.5, respectively.
The wintering occurrences in North Africa and Gibraltar, which were not included in the modelling process given that their OGUs were mostly or entirely located on the African continent, were used to validate the results of the different distribution models and their corresponding hypotheses. To this end, the discrimination capacity of the ensemble models in North Africa was assessed using the AUC and the classification capacity using the sensitivity and specificity.
Variation partitioning analysis
Interactions between factors often produce an overlaid spatial effect due to collinearity76,77. To illustrate this effect, we grouped the variables included in the distribution model according to H1 into two main factors: temperature and precipitation. We then employed a variation partitioning procedure (for a more detailed process, see Chamorro et al., 201925 and Muñoz et al., 2005 78) to determine the extent to which the variation in favourability of the final model was explained by the pure effects of each factor (i.e. unaffected by collinearity between them; Borcard et al., 199276) and the proportion attributable to their interaction65,77.
Results
Our eBird database search yielded 42 wintering records of the species in the Western Palaearctic from 1 December 2000 to 31 January 2024. They were located at the western and eastern extremities of the Mediterranean basin, mainly along the European shore, with some scattered records in North Africa (Fig. 2). We identified wintering records of the species in Peninsular Spain (n = 27), Portugal (n = 5), Balearic Islands (n = 3), Turkey (n = 2), Morocco (n = 1), Algeria (n = 1), Israel (n = 1), Egypt (n = 1), and Kuwait (n = 1). More than 80% of the records were detected on the Iberian Peninsula and the Balearic Islands (Fig. 2B). The first wintering bird of the species in the entire Western Palaearctic was detected on 12 January 2004, and the second on 22 December 2007, both in Southwestern Turkey (Fig. 2A). The species was not observed again until the winter of 2014–2015, when two individuals were detected: one in the south of the Iberian Peninsula and the other in the Balearic Islands (Fig. 2B). These were the first wintering records of the spotted flycatcher in Europe. Since the winter of 2014–2015, the species has been recorded every winter in Europe, except for the winter of 2017–2018, when there were no records across the entire Western Palaearctic (Fig. 2A, B). The number of records per wintering season has increased in recent years, starting from the winter of 2019–2020, primarily in the southern half of the Iberian Peninsula (Fig. 2B).
A: Wintering records of the spotted flycatcher in the Western Palaearctic from 1 December 2000 to 31 January 2024, with each colour representing the wintering season for the record. Wintering seasons not represented indicate that there were no records of the species. B: Detailed view of the wintering records in the Iberian Peninsula and the Balearic Islands, where more than 80% of the records were detected. Map projection: GCS_WGS_1984 (ArcGIS 10.4.1).
From the records described above, we identified 26 OGUs in the Western Palaearctic where the species was confirmed to winter (Fig. 3). We used the 22 OGUs located in Europe and the Middle East to model the distribution of the species specifically in this region (Fig. 3, black outlined hexagons). Europe and the Middle East have a total of 1887 OGUs, so that the number of absences considered to perform the model was 1865.
Operational geographic units (OGUs) corresponding to the wintering presence of the spotted flycatcher (red hexagons) based on the records shown in Fig. 2A. We used the wintering presences in Europe and the Middle East to model the distribution of the species only in this region (black outlined hexagons) and then extrapolated the model to North Africa (orange outlined hexagons). Map projection: GCS_WGS_1984 (ArcGIS 10.4.1).
Wintering records of the species in Europe were concentrated in the period 2014–2023, coinciding with the average surface temperature on the continent higher than 1991–2020 baseline (Fig. 4). No records of the species were detected in Europe until the winter of 2014–2015, during which the average surface temperature on the continent was lower in 1979–2000 and in 2001–2013 almost equal to the reference level.
Average estimated surface temperature change in Europe (°C) relative to 1991–2020 reference level for three periods during 1979–2023 (bars), (Copernicus Climate Change Service: https://climate.copernicus.eu), and the number of wintering records of the spotted flycatcher in Europe (black line). The average temperature for the period 1991–2020 was 9.2 °C.
The wintering distribution model for the spotted flycatcher included four climate-related variables: three related to temperature and one to precipitation (Table 2). Precipitation and temperature seasonalities were the first and second variables added in the stepwise procedure respectively, but they had less weight in the model according to the Wald test. The mean diurnal air temperature range and mean daily air temperatures of the coldest quarter were the third and final variables added in the stepwise procedure, respectively. They were the most significant predictors of the wintering distribution of the species in Europe and the Middle East, as they had the greatest weight in the model according to the Wald test. All four variables included in the model exhibited a unimodal response, meaning they positively affected the wintering distribution of the species only within a specific range of values (Figs.S2–S5). Therefore, precipitation seasonality, temperature seasonality, mean diurnal air temperature range, and mean daily air temperatures of the coldest quarter positively affected the wintering distribution of the spotted flycatcher within the following ranges using an F threshold of 0.5: 40–95 mm/month for precipitation seasonality (Fig. S2), 3–7 °C for temperature seasonality (Fig. S3), 7.5–12.5 °C for mean diurnal air temperature range (Fig. S4), and 5–15 °C for mean daily air temperatures of the coldest quarter (Fig. S5).
β is the coefficient in the logit function, S.E. is the standard error of each coefficient, Wald is the Wald statistic value (representing the relative importance of the variable in the model), p is the significance of each coefficient according to the Wald test, and VIF is the variance inflation factor (used to quantify collinearity between variables in the model).
The favourability values for the wintering of the spotted flycatcher in each OGU of the Western Palaearctic according to each unimodal variable included in the model are shown in the supplementary material (Figs. S6–S9), together with the assessment of these univariate models (Table S2). The favourability models according to H1 and H2 had the best assessment based on their discrimination and classification capacities, compared to those according to H3 and H4 (compare Table 3 and S3). Therefore, we decided to only present the favourability models according to H1 and H2 in the manuscript. The results of the models according to H3 and H4 are provided in the supplementary material (Fig. S10).
The resulting climatic favourability model according to H1 shows high favourability values for the wintering of the species over most of the Mediterranean basin, both on the European and African shores, except for Northern Italy and Greece (Fig. 5A). Areas with suitable climatic conditions for wintering of the spotted flycatcher include most of the Iberian Peninsula, Southern France, the southern half of Italy, Southern Greece, the west coast of Turkey, the Mediterranean coast of the Middle East, and the entire North African coast. Large Mediterranean islands such as Sardinia, Sicily, Crete, and Cyprus also have highly favourable climatic conditions for the species to winter. While the species has not yet been recorded in most places around the Mediterranean basin, in the last two winters (2022–2023 and 2023–2024), it has been recorded in new areas (Fig. 2). The climatic favourability model according to H2 is more restrictive and only shows intermediate-high favourability values in most of the Mediterranean coastal areas, especially in North Africa. It also shows intermediate-high favourability values in the southwestern quadrant of the Iberian Peninsula and the large Mediterranean islands such as Sardinia, Sicily, Crete, and Cyprus (Fig. 5B). The mean favourability of the models according to H1 and H2 was 0.112 and 0.044 in Europe and the Middle East and 0.371 and 0.143 in North Africa, respectively.
Climatic favourability (F) values for the wintering of the spotted flycatcher in each operational geographic unit of the Western Palaearctic according to A: hypothesis 1 (H1: all variables included in the model are relevant for the wintering of the species, but none are limiting) and B: hypothesis 2 (H2: all variables included in the model are limiting for the wintering of the species). Blue stars represent confirmed wintering records of the species not included in the modelling process. Map projection: GCS_WGS_1984 (ArcGIS 10.4.1).
To validate the resulting models, we used recently confirmed wintering records of the species not included in the modelling process (Fig. 5, blue stars). According to H1, all the records were located in areas with high or intermediate-high climatic favourability (FMorocco = 0.911, FAlgeria = 0.682, FGibraltar = 0.577) except for the record located on the Sinai Peninsula (Egypt), which was in an area with low favourability (FEgypt = 0.021). However, according to H2, all records were located in areas with low climatic favourability (FGibraltar = 0.070, FAlgeria = 0.063, FEgypt = 0.019) except for the record located in Morocco, which was in an area with intermediate-low favourability (FMorocco = 0.334). An assessment of the discrimination and classification capacities of the ensemble models in North Africa to compare the different hypotheses can be found in the supplementary material, where H1 obtained the best assessment (Table S4).
The resulting models according to H1 and H2 had high discrimination (AUC > 0.91) and classification (sensitivity, specificity, and CCR > 0.82 and a positive Cohen’s kappa except for the model according to H2, which showed a lower sensitivity) capacities in Europe and the Middle East and when extrapolated to North Africa (Table 3). However, the model according to H1 showed higher AUC and sensitivity than the model according to H2, both in Europe and the Middle East and its extrapolation to North Africa. In contrast, the model according to H2 showed higher specificity, CCR, and Cohen’s kappa than the model according to H1, both in Europe and the Middle East and its extrapolation to North Africa. The UPR were very low in all cases (< 0.01), whereas the OPR was substantial in all cases, especially in the models extrapolated to North Africa, with > 90% of the predicted favourable wintering OGUs unoccupied.
The pure effect of temperature in the climatic model according to H1 explained 76.6% of the variation. In contrast, the pure effect of precipitation explained only 1.7% of the variation, and the joint effect of both factors explained 21.7% of the variation (Fig. 6). These results show that temperature has the greatest influence on the winter distribution of the spotted flycatcher in the Western Palaearctic.
Variation partitioning of the climatic favourability model according to H1 using temperature and precipitation factors. The temperature factor includes the unimodal variables temperature seasonality, mean diurnal air temperature range, and daily air temperatures of the coldest quarter. The precipitation factor includes the unimodal variable precipitation seasonality. Values within the circles represent the percentage of variation explained by the indicated factors and their interaction.
Discussion
The scientific community has increasingly focused on the effects of climate change on animal distribution patterns, a subject extensively studied since the early twenty-first century19,22,53,79,80,81,82,83. While several studies have explored the influence of climate change on bird migration—primarily examining long-term shifts in spring arrival dates, autumn departures, and duration of stays22,23,24,84,85—there is limited information available on the responses to climate change during the winter period25. In this study, we provide the first evidence of regular wintering of the spotted flycatcher in the Western Palaearctic. Although traditionally all populations winter in sub-Saharan Africa45our findings indicate that wintering in the Western Palaearctic has become a relatively common phenomenon, particularly in recent years. Previous explanations for trans-Saharan migrants wintering in the Western Palaearctic have often focused on less experienced or juvenile individuals and sick or disoriented individuals. However, this does not seem to apply to the spotted flycatcher. Except for the winter of 2017–2018, the species has been regularly observed since the winter of 2014–2015, and the number of records has increased over the last two winters (2022–2023 and 2023–2024), suggesting a consistent wintering pattern.
The winter survival of insectivorous species like the spotted flycatcher is closely linked to the availability of flying insects, which is influenced by mild temperatures46,47. Additionally, some insectivorous birds switch to a more frugivorous diet during winter86. While the spotted flycatcher can consume small fruits from genera such as Berberis, Rhamnus, Cornus, Sorbus, Lonicera, Prunus, Morus, Rubus, and Trema45this option seems unlikely as most of these genera do not bear fruit in winter. Therefore, the growing number of records in recent years may be related to warmer temperatures in the Western Palaearctic, leading to milder winters33. The species primarily winters in coastal areas and large valleys in the Western Palaearctic, where milder temperatures may facilitate the availability of insects and other invertebrates. If these favourable conditions persist, the wintering population size could continue to increase in the coming years. Other factors may also contribute to the wintering of the species in the Western Palaearctic, such as the expansion of green spaces, including parks, urban forests, and grasslands, particularly in Europe87. Conversely, the growing number of wintering records might also be linked to more observers using citizen science platforms like eBird88,89,90. However, it is important to note that rare bird observations were documented before these platforms existed, although the information often took longer to become publicly available37,91.
As a nocturnal migrant, the spotted flycatcher follows a broad-fronted migration that covers much of the Mediterranean region92. Nonetheless, most wintering records are concentrated on the Iberian Peninsula, which serves as a critical stopover for European spotted flycatchers during their post-breeding migration93. The availability of resources93combined with mild autumns and winters22may lead the species to stay through the winter if environmental conditions allow, primarily due to food availability. This view is supported by the significant number of records from the Iberian Peninsula, accounting for three-quarters of the records in the entire Western Palaearctic. In addition to these records, our distribution models further support the idea that the Western Palaearctic is emerging as a new wintering area for this long-distance migratory insectivore. Our models identify favourable wintering areas across much of the Mediterranean basin, including both European and African shores and large Mediterranean islands such as Sardinia, Sicily, Crete, and Cyprus. While our models exhibited an OPR of > 78% in Europe and the Middle East, this does not necessarily indicate that they are erroneous74,94. Instead, it suggests they have identified favourable conditions in regions where the species has not yet been detected during winter.
When extrapolating the models to North Africa, the OPR exceeded 90%, suggesting that even more favourable but unoccupied (or undetected) wintering areas exist in North Africa compared to Europe and the Middle East. Indeed, the mean favourability was higher in North Africa than in Europe and the Middle East with both models, despite the lower number of wintering records. Notably, while the number of citizen science observers in African countries is steadily increasing, the overall number of participants remains significantly higher in Europe. Therefore, many areas predicted to be favourable in North Africa could indeed be occupied by the species during winter. Consequently, further research in North Africa is necessary, along with the promotion of citizen science, to better understand the role of this region in the wintering of the species and potentially other trans-Saharan migrants.
The model according to H1 was more optimistic, whereas the model according to H2 was more restrictive, requiring high favourability across all four variables in each OGU for the overall model favourability to be high. We propose that the current wintering situation of the spotted flycatcher in the Western Palaearctic likely falls somewhere between these two models. While both models exhibited high discrimination and classification capacities, the actual situation may be closer to the model according to H1, which is supported by the validation of all four wintering records. Three of these records were located in areas with high or intermediate-high climatic favourability values. While the record from the Sinai Peninsula (Egypt) was in an area with low favourability, most of the OGU in which it was located is in Saudi Arabia, and the part of the OGU in the Sinai Peninsula likely has similar favourability to the surrounding high-favourability area (Fig. 5). Furthermore, H1 obtained the best assessment among the different ensemble models according to their discrimination and classification capacities in North Africa (Table S4).
Based on these considerations, we predict a potential increase in the wintering populations of the species, mainly in the coastal areas and island systems of the Mediterranean, if climatic conditions in the Western Palaearctic remain consistent over the coming winters33. Precipitation and temperature seasonalities were key predictors in determining the broader winter distribution of the species, being the first and second variables included in the model, respectively. The first variable excluded Northern and Central Europe due to low precipitation seasonality and some regions of western and eastern North Africa due to high precipitation seasonality (Figs. S2 and S6). The second variable excluded continental areas of the Western Palaearctic because of high temperature seasonality, focusing instead on the coastal areas within the study area (Figs. S3 and S7).
The mean diurnal air temperature range had the greatest explanatory power in the model (Table 2), identifying areas with a mean diurnal air temperature of between 7.5 °C and 12.5 °C, broadly corresponding to Mediterranean regions (Figs. S4). The mean daily air temperature of the coldest quarter was the second most influential variable in the model, adding significant detail as the last variable included. This variable identified areas with mild but not excessively warm winters in the Western Palaearctic (Figs. S5 and S9).
In summary, the regions characterised by favourable climatic conditions for the wintering of the spotted flycatcher are predominantly located along the coastal areas of the Mediterranean basin, where winters are typically mild. Temperature has emerged as the primary environmental factor influencing the wintering distribution of the species, as shown by the variation partitioning analysis (Fig. 6), indicating that it plays a far more significant role than precipitation. The severity of winters in the Western Palaearctic has decreased in recent decades, particularly in Southern Europe and Northern Africa, due to ongoing climate change33,95. These newly established warmer conditions throughout the year, including winter, have led to an increase in the abundance of flying insects, such as mosquitoes96enhancing prey availability for spotted flycatchers. With temperature being a key driver of insect abundance and activity, warmer winters directly benefit the species by providing a more consistent food supply.
The species will likely establish a stable wintering population in the Western Palaearctic if it can access sufficient resources during the winter months, potentially providing these individuals with advantages over long-distance migrant populations as winters continue to moderate. Food availability allows the species to maintain optimal body condition throughout the winter, reducing energy expenditure in less severe climatic environments1,32. Milder temperatures also reduce the thermoregulatory demands on the species, lowering their overall energy costs. Furthermore, their proximity to breeding sites in the following spring enables them to select superior breeding locations compared to long-distance migrants31,97. Additionally, lower migration-related mortality is anticipated, which could result in higher reproductive success28,32,98. However, since temperature is the primary factor driving the current winter distribution, any rapid fluctuations or extreme cold events could significantly disrupt these newly established wintering populations, leading to greater mortality and posing a significant threat to their existence99,100.
Given that the primary purpose of bird migration is to reach environments with optimal conditions at specific times101the ability to adapt migratory behaviour has likely been crucial since the onset of animal migration on a planet subjected to environmental change102. The spotted flycatcher’s ability to adjust its migratory behaviour, largely in response to rising temperatures, suggests that this flexibility may be a critical factor in population regulation. Ultimately, the diversity of migratory strategies across different geographical regions, and even among species within the same area, serves as a survival strategy that may have significant implications for population regulation and viability39,103. Whether staying on breeding grounds throughout the winter or shortening migration distances improves winter survival or breeding success in spotted flycatchers remains unclear. Consequently, ongoing monitoring is essential to understand and track the temporal evolution of this long-distance migrant, which is now wintering north of the Sahara.
The spotted flycatcher may serve as a model for other trans-Saharan migrants that have recently begun wintering in the Western Palaearctic. If temperature increases persist, more long-distance migrants are expected to shorten their migration routes, leading to larger populations regularly wintering in the Western Palaearctic, particularly in coastal areas and island systems within the Mediterranean. This trend has already been observed in species such as the Eurasian reed warbler25,38. Our findings could guide targeted sampling and monitoring efforts by prioritising areas identified as highly favourable by the distribution models, thereby improving the efficiency of winter monitoring. While further research is necessary, this study significantly advances our understanding of the impact of climate change on the phenology and winter distribution of the species.
Data availability
The data that support the findings of this study are available in figshare at https://figshare.com/s/f1c19e6e3181a2ead02b, DOI: 10.6084/m9.figshare.27053167. These data were derived from the platform eBird available in the public domain: https://ebird.org.
References
Newton, I. The migration ecology of birds (2nd edition) (Academic Press, 2008).
Berthold, P. & Querner, U. Genetic basis of migratory behavior in European warblers. Sci. 212, 77–79 (1981).
Berthold, P. Evolutionary aspects of migratory behavior in European warblers. J. Evol. Biol. 1, 195–209 (1988).
Berthold, P. Genetic control of migratory behaviour in birds. Trends Ecol. Evol. 6, 254–257 (1991).
Berthold, P. & Helbig, A. J. The genetics of bird migration: stimulus, timing, and direction. Ibis 134, 35–40 (1992).
Gwinner, E. Circadian and circannual programmes in avian migration. J. Exp. Biol. 199, 39–48 (1996).
Flack, A., Nagy, M., Fiedler, W., Couzin, I. D. & Wikelski, M. From local collective behavior to global migratory patterns in white storks. Sci. 360, 911–914 (2018).
Berthold, P., Helbig, A. J., Mohr, G. & Querner, U. Rapid microevolution of migratory behaviour in a wild bird species. Nature 360, 668–670 (1992).
Ogonowski, M. S. & Conway, C. J. Migratory decisions in birds: extent of genetic versus environmental control. Oecologia 161, 199–207 (2009).
Charmantier, A. & Gienapp, P. Climate change and timing of avian breeding and migration: evolutionary versus plastic changes. Evol. Appl. 7, 15–28 (2014).
Horton, K. G., Morris, S. R., Van Doren, B. M. & Covino, K. M. Six decades of North American bird banding records reveal plasticity in migration phenology. J. Anim. Ecol. 92, 738–750 (2023).
Berthold, P. & Querner, U. Microevolutionary aspects of bird migration based on experimental results. Isr. J. Zool. 41, 377–385 (1995).
Pulido, F., Berthold, P., Mohr, G. & Querner, U. Heritability of the timing of autumn migration in a natural bird population. Proc. R Soc. Lond. Ser. B Biol. Sci. 268, 953–959 (2001).
Møller, A. P., Rubolini, D. & Saino, N. Morphological constraints on changing avian migration phenology. J. Evol. Biol. 30, 1177–1184 (2017).
Helm, B., Van Doren, B. M., Hoffmann, D. & Hoffmann, U. Evolutionary response to climate change in migratory pied flycatchers. Curr. Biol. 29, 3714–3719 (2019).
Charmantier, A. et al. Adaptive phenotypic plasticity in response to climate change in a wild bird population. Sci. (80-). 320, 800–803 (2008).
Conklin, J. R., Lisovski, S. & Battley, P. F. Advancement in long-distance bird migration through individual plasticity in departure. Nat. Commun. 12, 4780 (2021).
López-Ramírez, S., Chamorro, D., Real, R. & Muñoz, A. R. Southern Europe is becoming climatically favourable for African birds: anticipating the establishment of a new species. Front. Zool. 20, 18 (2023).
López-Ramírez, S., Márquez, A. L., Real, R. & Muñoz, A. R. Evaluating the expansion of African species into Europe driven by climate change. Divers. Distrib. 30, e13840 (2024).
López-Ramírez, S., Real, R. & Muñoz, A. R. The northern wheatear is reducing its distribution in its southernmost European range and moving to higher altitudes. J. Avian Biol. 2024, e03217 (2024).
Stiels, D., Bastian, H. V., Bastian, A., Schidelko, K. & Engler, J. O. An iconic messenger of climate change? Predicting the range dynamics of the European Bee-eater (Merops apiaster). J. Ornithol. 162, 631–644 (2021).
Gordo, O. & Sanz, J. J. Climate change and bird phenology: A long-term study in the Iberian Peninsula. Glob Chang. Biol. 12, 1993–2004 (2006).
Bussière, E. M. S., Underhill, L. G. & Altwegg, R. Patterns of bird migration phenology in South Africa suggest Northern hemisphere climate as the most consistent driver of change. Glob Chang. Biol. 21, 2179–2190 (2015).
Potvin, D. A., Välimäki, K. & Lehikoinen, A. Differences in shifts of wintering and breeding ranges lead to changing migration distances in European birds. J. Avian Biol. 47, 619–628 (2016).
Chamorro, D., Nieto, I., Real, R. & Muñoz, A. R. Wintering areas on the move in the face of warmer winters. Ornis Fenn. 96, 41–54 (2019).
Sergio, F. et al. When and where mortality occurs throughout the annual cycle changes with age in a migratory bird: individual vs population implications. Sci. Rep. 9, 17352 (2019).
Newton, I. Migration mortality in birds. Ibis 167, 106–123 (2025).
Sillett, T. S. & Holmes, R. T. Variation in survivorship of a migratory songbird throughout its annual cycle. J. Anim. Ecol. 71, 296–308 (2002).
Kokko, H. Competition for early arrival in migratory birds. J. Anim. Ecol. 68, 940–950 (1999).
Coppack, T., Pulido, F., Czisch, M., Auer, D. P. & Berthold, P. Photoperiodic response may facilitate adaptation to climatic change in long-distance migratory birds. Proc. R Soc. Lond. Ser. B Biol. Sci. 270, S43–S46 (2003).
Podhrázský, M. et al. Central European Greylag Geese Anser anser show a shortening of migration distance and earlier spring arrival over 60 years. Ibis 159, 352–365 (2017).
Alves, J. A. et al. Costs, benefits, and fitness consequences of different migratory strategies. Ecology 94, 11–17 (2013).
IPCC. Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. (2022). https://doi.org/10.1017/9781009325844
Martín, B. et al. Are Western European ospreys (Pandion haliaetus) shortening their migration distances? Evidence from trends of the wintering population in the Iberian Peninsula. Eur. J. Wildl. Res. 65, 72 (2019).
Morganti, M. & Pulido, F. Invernada de Aves Migradoras Transaharianas en España in Atlas de Aves en Invierno en España 2007–2010 (ed. SEO/BirdLife) 59–64 (Ministerio de Agricultura, Alimentación y Medio Ambiente-SEO/BirdLife, 2012).
Marchamalo, J. Cigüeña Blanca (Ciconia ciconia) in Atlas de las Aves Invernantes de Madrid 1999–2001 (eds. del Moral, J. C., Molina, B., de la Puente, J. & Pérez-Tris, J.) 84–85 (SEO-Monticola y Comunidad de Madrid, 2002).
SEO/BirdLife. Atlas de Aves en Invierno en España 2007–2010 (Ministerio de Agricultura, Alimentación y Medio Ambiente-SEO/BirdLife, 2012).
Nieto, I., Chamorro, D., Palomo, L. J., Real, R. & Muñoz, A. R. Is the Eurasian Reed Warbler Acrocephalus Scirpaceus a regular wintering species in the Iberian Peninsula? Ringing data say yes. Acta Ornithol. 53, 61–68 (2018).
Rotics, S. et al. Wintering in Europe instead of Africa enhances juvenile survival in a long-distance migrant. Anim. Behav. 126, 79–88 (2017).
Ramírez, J., Simón, M., Solís, S., Pérez, C. & García, E. Vencejo Moro Apus affinis Observaciones de Aves Raras en España. Ardeola 49, 161 (2002).
Elorriaga, J. & Muñoz, A. R. First breeding record of North African Long-legged Buzzard Buteo rufinus cirtensis in continental Europe. Br. Birds. 103, 399–401 (2010).
Cabrera, M. Corredor Sahariano Cursorius cursor in III Atlas de las Aves en Época de Reproducción en España (eds. Molina, B. et al.) https://atlasaves.seo.org/ave/corredor-sahariano/ (SEO/BirdLife, 2022).
Navarrete, J. Bulbul Naranjero Pycnonotus barbatus in III Atlas de las Aves en Época de Reproducción en España (eds. Molina, B. et al.) https://atlasaves.seo.org/ave/bulbul-naranjero/ (SEO/BirdLife, 2022).
Muñoz, A. R., Ramírez, J. & Real, R. A critically endangered African Vulture starts breeding in europe: escaping from extinction or entering a genetic amalgamation trap? Ardeola 71, 131–135 (2024).
Taylor, B. Spotted Flycatcher (Muscicapa striata) in Birds of the World (eds. del Hoyo, J., Elliott, A., Sargatal, J., Christie, D. & de Juana, E.) https://doi.org/10.2173/bow.spofly1.01 (Cornell Lab of Ornithology, 2020).
Ratte, H. T. Temperature and insect development in Environmental Physiology and Biochemistry of Insects (ed. Hoffmann, K. H.) 33–66. https://doi.org/10.1007/978-3-642-70020-0_2 (Springer, 1984).
Matthäus, L., Kugelschafter, K. & Fietz, J. Influence of ambient temperature on the phenology of the greater mouse-eared bat (Myotis myotis). Ecol. Evol. 13, e10081 (2023).
Font, I. Climatología de España y Portugal (Universidad de Salamanca, 2000).
Mill, H. R. The International Geography (D. Appleton and Company, 1902).
Udvardy, M. D. F. A Classification of the Biogeographical Provinces of the World (International Union for Conservation of Nature (IUCN) and Natural Resources Morges, 1975).
Cramp, S. & Simmons, K. E. L. Handbook of the Birds of Europe, the Middle East and North Africa. The Birds of the Western Palearctic. Vol III. Waders To Gulls (Oxford University Press, 1983).
Márquez, A. L., Real, R., Olivero, J. & Estrada, A. Combining climate with other influential factors for modelling the impact of climate change on species distribution. Clim. Change. 108, 135–157 (2011).
Muñoz, A. R., Márquez, A. L. & Real, R. Updating known distribution models for forecasting climate change impact on endangered species. PLoS One. 8, 65462 (2013).
ESRI. ArcGIS [software GIS]. Version 10.4.1. at (2016). https://www.esri.com/en-us/arcgis/products/arcgis-desktop/resources
US Geological Survey. GTOPO30. Land Processes Distributed Active Archive Center (LPDAAC), EROS Data Center, Washington D.C., (1996). available from https://lta.cr.usgs.gov/GTOPO30
Hosmer, D. W. & Lemeshow, S. Goodness of fit tests for the multiple logistic regression model. Commun. Stat. - Theory Methods. 9, 1043–4069 (1980).
Zanolla, M. et al. Assessing global range expansion in a cryptic species complex: insights from the red seaweed genus Asparagopsis (Florideophyceae). J. Phycol. 54, 12–24 (2018).
López-Ramírez, S. & Muñoz, A. R. Climate change drives range contraction and shapes species distribution in an alpine passerine: caution required when comparing atlas data. Glob Ecol. Conserv. 60, e03620 (2025).
Radhakrishna Rao, C. Large sample tests of statistical hypotheses concerning several parameters with applications to problems of Estimation. Math. Proc. Camb. Philos. Soc. 44, 50–57 (1948).
Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. R Stat. Soc. Ser. B. 57, 289–300 (1995).
Benjamini, Y. & Yekutieli, D. The control of the false discovery rate in multiple testing under dependency. Ann. Stat. 29, 1165–1188 (2001).
García-Carrasco, J. M., Muñoz, A. R., Olivero, J., Segura, M. & Real, R. An African West Nile virus risk map for travellers and clinicians. Travel Med. Infect. Dis. 52, 102529 (2023).
López-Ramírez, S., Márquez, A. L., Real, R. & Muñoz, A. R. Evaluating the expansion of African species into Europe driven by climate change. Divers. Distrib. 00, e13840 (2024).
García-Carrasco, J. M. et al. Present and future situation of West Nile virus in the Afro-Palaearctic pathogeographic system. Arch. Anesthesiol Crit. Care. 4, 527–534 (2024).
Legendre, P. & Legendre, L. Numerical Ecology (Elsevier, 1998).
Basha, S. M., Rajput, D. S. & Vandhan, V. Impact of gradient ascent and boosting algorithm in classification. Int. J. Intell. Eng. Syst. 11, 41–49 (2018).
Glasmachers, T. & Igel, C. Maximum likelihood model selection for 1-norm soft margin svms with multiple parameters. IEEE Trans. Pattern Anal. Mach. Intell. 32, 1522–1528 (2010).
Wald, A. Tests of statistical hypotheses concerning several parameters when the number of observations is large. Trans. Am. Math. Soc. 54, 426–482 (1943).
Real, R., Barbosa, A. M. & Vargas, J. M. Obtaining environmental favourability functions from logistic regression. Environ. Ecol. Stat. 13, 237–245 (2006).
Sillero, N. et al. Want to model a species niche? A step-by-step guideline on correlative ecological niche modelling. Ecol. Modell. 456, 109671 (2021).
Estrada, A. & Real, R. A stepwise assessment of parsimony and fuzzy entropy in species distribution modelling. Entropy 23, 1014 (2021).
Lobo, J. M., Jiménez-Valverde, A. & Real, R. AUC: a misleading measure of the performance of predictive distribution models. Glob Ecol. Biogeogr. 17, 145–151 (2008).
Fielding, A. H. & Bell, J. F. A review of methods for the assessment of prediction errors in conservation presence/absence models. Environ. Conserv. 24, 38–49 (1997).
Barbosa, A. M., Real, R., Muñoz, A. R. & Brown, J. A. New measures for assessing model equilibrium and prediction mismatch in species distribution models. Divers. Distrib. 19, 1333–1338 (2013).
Cohen, J. A coefficient of agreement for nominal scales. Educ. Psychol. Meas. 20, 37–46 (1960).
Borcard, D., Legendre, P. & Drapeau, P. Partialling out the spatial component of ecological variation. Ecology 73, 1045–1055 (1992).
Legendre, P. Spatial autocorrelation: Trouble or new paradigm? Ecology 74, 1659–1673 (1993).
Muñoz, A. R., Real, R., Barbosa, A. M. & Vargas, J. M. Modelling the distribution of Bonelli’s eagle in Spain: implications for conservation planning. Divers. Distrib. 11, 477–486 (2005).
Žalakevičius, M. Bird numbers, population state, and distribution areas in the Eastern Baltic region in the context of the impact of global climate change. Acta Zool. Litu. 11, 141–162 (2001).
Péron, G., Henry, P., Provost, P., Dehorter, O. & Julliard, R. Climate changes and post-nuptial migration strategy by two reedbed passerines. Clim. Res. 35, 147–157 (2007).
Massimino, D., Johnston, A. & Pearce-Higgins, J. W. The geographical range of British birds expands during 15 years of warming. Bird Study 62, 523–534 (2015).
Bay, R. A. et al. Genomic signals of selection predict climate-driven population declines in a migratory bird. Sci. 359, 83–86 (2018).
Blois, J. L., Zarnetske, P. L., Fitzpatrick, M. C. & Finnegan, S. Climate change and the past, present, and future of biotic interactions. Sci. 341, 499–504 (2013).
Both, C. & Visser, M. E. Adjustment to climate change is constrained by arrival date in a long-distance migrant bird. Nature 411, 296–298 (2001).
Lehikoinen, E. & Sparks, T. H. Changes in migration in Effects of Climate Change on Birds (eds. Møller, A. P., Fiedler, W. & Berthold, P.) 89–112 (Oxford University Press, 2010).
Arizaga, J. et al. Wintering survival of insect-eating passerines in Southern Europe. Bird Study 59, 37–42 (2012).
Leng, S., Sun, R., Yang, X. & Chen, L. Global inequities in population exposure to urban greenspaces increased amidst tree and nontree vegetation cover expansion. Commun. Earth Environ. 4, 464 (2023).
Sullivan, B. L. et al. The eBird enterprise: an integrated approach to development and application of citizen science. Biol. Conserv. 169, 31–40 (2014).
Amano, T., Lamming, J. D. L. & Sutherland, W. J. Spatial gaps in global biodiversity information and the role of citizen science. Bioscience 66, 393–400 (2016).
Pocock, M. J. O., Tweddle, J. C., Savage, J., Robinson, L. D. & Roy, H. E. The diversity and evolution of ecological and environmental citizen science. PLoS One 12, e0172579 (2017).
Comité de Rarezas de SEO/BirdLife. Observaciones de Aves Raras en España (Septiembre 2022 – Agosto 2023). Ardeola 71, 183–194 (2024).
Thorup, K. & Rabøl, J. The orientation system and migration pattern of long-distance migrants: conflict between model predictions and observed patterns. J. Avian Biol. 32, 111–119 (2001).
Hernández, Á. Summer-autumn feeding ecology of Pied Flycatchers Ficedula hypolueca and Spotted Flycatchers Muscicapa striata: The importance of frugivory in a stopover area in north-west Iberia. Bird. Conserv. Int. 19, 224–238 (2009).
Muñoz, A. R. & Real, R. Assessing the potential range expansion of the exotic monk parakeet in Spain. Divers. Distrib. 12, 656–665 (2006).
Brunet, M. et al. Temporal and spatial temperature variability and change over Spain during 1850–2005. J. Geophys. Res. Atmos. 112, 1–28 (2007).
Roiz, D., Ruiz, S., Soriguer, R. & Figuerola, J. Climatic effects on mosquito abundance in Mediterranean wetlands. Parasit. Vectors. 7, 333 (2014).
Forstmeier, W. Benefits of early arrival at breeding grounds vary between males. J. Anim. Ecol. 71, 1–9 (2002).
Lok, T., Veldhoen, L., Overdijk, O., Tinbergen, J. M. & Piersma, T. An age-dependent fitness cost of migration? Old trans‐Saharan migrating spoonbills breed later than those staying in Europe, and late breeders have lower recruitment. J. Anim. Ecol. 86, 998–1009 (2017).
Senar, J. C. & Copete, J. L. Mediterranean house sparrows (Passer domesticus) are not used to freezing temperatures: an analysis of survival rates. J. Appl. Stat. 22, 1069–1074 (1995).
Maclean, I. M. D. et al. Climate change causes rapid changes in the distribution and site abundance of birds in winter. Glob Chang. Biol. 14, 2489–2500 (2008).
Berthold, P. Bird Migration: A General Survey (Oxford University Press, 2001).
Fiedler, W. Recent changes in migratory behaviour of birds: A compilation of field observations and ringing data in Avian Migration (eds. Berthold, P., Gwinner, E. & Sonnenschein, E.) 21–38. https://doi.org/10.1007/978-3-662-05957-9_2 (Springer, 2003).
Lok, T., Overdijk, O. & Piersma, T. The cost of migration: spoonbills suffer higher mortality during trans-Saharan spring migrations only. Biol. Lett. 11, 20140944 (2015).
Acknowledgements
We would like to thank the eBird community and all contributors to this international citizen science platform.
Funding
This study was funded by Scientific Research Projects in the National Parks Network, 3092/2023 (Ministry for Ecological Transition and Demographic Challenge). S.L.-R. is a PhD student supported by a scholarship from the Spanish Ministry of Education, Culture, and Sport (FPU20/01360).
Author information
Authors and Affiliations
Contributions
Conceived the idea and conceptualised the study – A.-R.M. Performed the experiments (collected data, conducted the research) – A.-R.M. and R.R. Wrote the paper (or substantially edited the paper) – S.L.-R., J.-M.G.-C. and A.-R.M. Analyzed the data – S.L.-R. and J.-M.G.-C. Contributed substantial materials, resources, or funding - A.-R.M. All authors critically reviewed the manuscript and approved the final version.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
López-Ramírez, S., García-Carrasco, JM., Real, R. et al. Climate-driven shifts in trans-Saharan bird migration illustrated by the spotted flycatcher (Muscicapa striata). Sci Rep 15, 35201 (2025). https://doi.org/10.1038/s41598-025-19138-9
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41598-025-19138-9
