Research Article |
Corresponding author: Piotr Zduniak ( kudlaty@amu.edu.pl ) Academic editor: George Sangster
© 2015 Piotr Zduniak, Reuven Yosef, Keith J. Bensusan, Charles E. Perez, Piotr Tryjanowski.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Zduniak P, Yosef R, Bensusan KJ, Perez CE, Tryjanowski P (2015) Biometry and phenology of two sibling Phylloscopus warblers on their circum-Mediterranean migrations. ZooKeys 530: 113-127. https://doi.org/10.3897/zookeys.530.5955
|
The Mediterranean Sea is known as an ecological barrier for numerous migratory birds flying from European breeding grounds to African wintering sites. Birds generally avoid migration over open sea and fly over land. In the Mediterranean Basin, few land bridges or bottlenecks for migratory birds exist. The narrowest are at the western and eastern extremes: the Strait of Gibraltar and Israel. Comparative studies between these locations are extremely rare to date. Therefore, in order to elucidate the differences between the two flyways, we compared data collected simultaneously for two sister leaf warbler species, the Bonelli’s Warbler complex, Phylloscopus bonelli and Ph. orientalis, at ringing stations in the western Mediterranean Basin Gibraltar, and the eastern Eilat, Israel. Data on biometrics and passage dates of individuals trapped at Gibraltar and Eilat were used, and it was found that mean arrival date of Western Bonelli’s Warblers at Gibraltar was 15 days later than Eastern Bonelli’s Warblers at Eilat. Furthermore, Western Bonelli’s Warblers had shorter wings than Eastern Bonelli’s Warblers. On the other hand, birds in Eilat were in poorer body condition than individuals in Gibraltar. The comparison between geographically distant stop-over sites contributes to furthering our understanding of the development of migration strategies across ecological barriers in sibling species. Our study showed that populations that breed in southwestern Europe migrate through Gibraltar and winter in West Africa are able to accomplish migration in comparatively good body condition. This is in contrast to those that winter in East Africa, migrate through Israel and have to endure the combined challenge of crossing the Sahel, Sahara and Sinai deserts before reaching their breeding grounds across southeast Europe and southwest Asia. Hence, the discrepancies described between the western and the eastern flyway suggest that individuals in the west, in general, migrate shorter distances, have a physiologically less demanding crossing of the North African deserts and appear to stage before their crossing the Strait of Gibraltar, a privilege unavailable to the migrants of the eastern flyway.
geographical barrier, Mediterranean, migration, Phylloscopus bonelli , Phylloscopus orientalis
Bird populations migrate annually from breeding sites to wintering localities and back. This phenomenon influences not only their proximate behavior and ecology, but also their ultimate evolutionary processes (
Intra- and interspecific comparisons of the two bottlenecks are of interest because they are located at extreme ends of the Mediterranean Basin and birds use them to avoid long sea crossings. However, the two sites are ecologically distinct: Gibraltar is located on a mountain range with well-developed vegetation and has a hinterland with typically Mediterranean vegetation whilst Eilat is situated in the midst of a complex of deserts (Sahara to the south, Arabian to the east, Sinai to the west, Negev to the north). Differences in habitat around stop-over sites may favor different migratory strategies, especially phenology and time of stop-over use (
Leaf warblers (Phylloscopus spp.) are considered model species in bird migration studies because of their wide spectrum of ecological and migratory traits. Nevertheless European studies are mostly limited to Willow Warbler (Phylloscopus trochilus Linnaeus, 1758) and Chiffchaff (Phylloscopus collybita Vieillot, 1817) (e.g.,
In this paper, we analyzed two sibling leaf warbler species: the Western and Eastern Bonelli’s Warbler (Phylloscopus bonelli aVieillot, 1819 and Ph. orientalis Brehm, 1855). These species are biogeographically and ecologically separated and were recently split as full species, having previously been classified as subspecies of Phylloscopus bonelli (
In order to avoid migration over large expanses of desert and then water, both species have to migrate over the Sahara and then circumnavigate the Mediterranean basin, crossing at Gibraltar in the west and Israel in the east. Because information in many handbooks and ringers manuals is, to date, strongly limited for both species and not provided to the species level (given as races/subspecies), basic morphometric and phenology studies are considered to be of importance (cf.
Owing to the fact that bird migration in the Mediterranean Basin is North-South-North, we collated data simultaneously at the two outermost sites (east and west), located ca. 3820 km from each other.
The Rock of Gibraltar is located at the eastern end of the Strait of Gibraltar (36°07'N; 5°21'W), where the Mediterranean meets the Atlantic. It lies ca. 21 km north of North Africa. Habitat in Gibraltar is typically Mediterranean and the vegetation around the ringing station at Jews’ Gate consists of dense maquis that is dominated by fruit-bearing shrubs (
Eilat (29°33'N; 34°57'E) is located at the southernmost end of Israel and at the northern tip of the Gulf of Aqaba. Eilat provides a land bridge between three continents, Africa, Asia and Europe, and is located along a set of flyways used by birds wintering in sub-Saharan Africa and breeding in Eurasia. For migratory birds, Eilat is located before (late summer and autumn, or post-nuptial migration) or after (spring, or pre-nuptial migration) the Sahel, Sahara and Sinai Desert crossings, at the edge of almost 2000 km of continuous desert. To the northeast lies the Syrian Desert and to the east the Arabian Desert (
Data analyzed were collected during the same period of 19 years, from 1992 to 2010. All trapping at Gibraltar was undertaken with mist-nets which were operated for ca. 6 hrs daily during the spring (March-May) and autumn (September-November) migration seasons. At Eilat, birds were trapped with mist nets for ca. 6 hr/d in the years 1992–1999. After 2000, birds were trapped using eight Helgoland/Rybachy traps, located along the trapping lanes of the mist-nets in the boundaries of the Eilat Bird Sanctuary and operated for ca. 6 hrs daily during the migration seasons. At both ringing stations, the trapping effort did not differ between spring and autumn in individual years and did not change throughout each season.
All trapped birds were ringed with the standard aluminum rings and biometrics were recorded. Flattened maximum wing chord was measured to the nearest millimeter and body mass determined using Pesola and/or digital scales to the nearest 0.1 g. Although adult Phylloscopus warblers have slightly longer wings than juveniles, as is the case in most passerines, we did not attempt to include age of birds in the analyses because leaf warblers are notoriously difficult to age during the pre-nuptial migration, following a complete molt in the wintering areas (
During the 19 years of study, 1101 Western Bonelli’s Warblers and 1706 Eastern Bonelli’s Warblers were ringed. Both species were recorded mainly during the pre-nuptial passage (98.1% and 99.2% of all ringed individuals of Western and Eastern Bonelli’s Warblers, respectively). Hence, owing to the sparse post-nuptial migration data we analyzed only data collected during the pre-nuptial migration. Furthermore, for the pre-nuptial period we chose to include in the analyses only those seasons for which at least ten individuals were recorded (18 springs in Gibraltar with insufficient data for 1999 and 16 springs in Eilat with insufficient data for 1994, 1997 and 2004). The average sample size per season was 59.6 birds (95% CL: 38.5–80.7, N = 18) for Gibraltar and 104.6 birds (95% CL: 58.0–151.3, N = 16) for Eilat. The total number of birds included in the in-depth analysis was 1073 for Gibraltar and 1674 for Eilat.
We checked for possible differences between species in migration phenology according to Julian days and the biometric parameters wing chord length and body mass. In these analyses only data from the first captures were used. Moreover, to examine possible differences between stopover sites, which have habitat types that probably differ in quality and importance for migrants, we also calculated and compared body condition index (body mass divided by wing length; e.g.,
Because timing of migration and biometric parameters (wing chord length, body mass and body condition index) compared between species at the two study sites could be influenced by year, we controlled for year as a factor in the analyses. Comparisons between species were performed using Main Effects ANOVA with species and year as factors. Furthermore, we checked for patterns in changes of biometrics during spring migration. In the analyses we used a standardized time of migration for each study site calculated by subtracting the median Julian date of catching time for each spring season from each catching date in that season. Full biometric data were not always available for all individuals ringed and this has resulted in different sample sizes between individual analyses.
Standard statistical methods were used to describe and analyze the data (
In general, migration period and duration of spring passage were similar for both species. Both in Gibraltar and Eilat the passage began in early March and lasted to mid May. However, the main passage period of Western Bonelli’s Warbler occurred from the third decade of March till the second decade of May with the peak at the second decade of April (Fig.
Western Bonelli’s Warblers trapped in Gibraltar had shorter wings (two-way ANOVA, F1,2718 = 352.73, P < 0.001; Table
Mean values with 95% confidence of limits in parentheses, ranges and sample size for wing chord length, body mass and body condition index of Western and Eastern Bonelli’s Warbler recorded in Gibraltar and Eilat, Israel, respectively.
Variable | Western Bonelli’s Warbler | Eastern Bonelli’s Warbler |
---|---|---|
wing chord length (mm) | 63.6 (63.4−63.8) range: 56−73, N = 1072 | 66.0 (65.9−66.2) range: 49−77, N = 1666 |
body mass (g) | 7.2 (7.2−7.3) range: 4.8−10.5, N = 1025 | 7.0 (6.9−7.0) range: 4.6−11.0, N = 1646 |
body condition index | 0.114 (0.113−0.115) range: 0.075−0.161, N = 1025 | 0.106 (0.105−0.106) range: 0.073−0.167, N = 1644 |
A significant decrease of the wing length during the spring passage was recorded both in the case of the Western as well as the Eastern Bonelli’s Warbler (Pearson correlation, r = -0.28, N = 1072, P < 0.001; r = -0.29, N = 1666, P < 0.001, respectively). However the body condition of migrants decreased over the passage only in the case of Eastern Bonelli’s Warbler (r = -0.15, N = 1644, P <0.001) and no such pattern was observed in the case of Western Bonelli’s Warbler (r = -0.03, N = 1025, P = 0.323).
Throughout the study period we retrapped 17 (1.6%) Western and 339 (20.2%) Eastern Bonelli’s Warblers. The probability of Western and Eastern Bonelli’s Warblers recapture during the pre-nuptial period differed significantly between sites (Mann-Whitney U test, Z = 3.37, P < 0.001). The mean probability of recapture was 0.009 (CL: 0.020–0.017, N = 18) at Gibraltar and 0.159 (CL: 0.090–0.228, N = 16) at Eilat.
This study documents differences in migration phenology and energetic requirements between the Western and the Eastern Bonelli’s Warblers at the two extremes of the Mediterranean Sea: Gibraltar in the west and Eilat, Israel, in the east. The Eastern Bonelli’s Warbler peaked in the third decade of March, whereas the Western Bonelli’s Warbler peaked in the second decade of April; the mean arrival date of Western Bonelli’s Warbler at Gibraltar was 15 days later than that of Eastern Bonelli’s Warbler at Eilat. Western Bonelli’s Warblers trapped in Gibraltar had shorter wings and were in significantly better condition than Eastern Bonelli’s Warblers at Eilat, but the body condition of migrants only decreased over the passage period in Eastern Bonelli’s Warbler. Furthermore, a decrease of wing length was found during the spring passage in both species. Also, the mean probability of recapture was greater at Eilat.
Gibraltar and Israel (especially Eilat in the south) are located along very important flyways for birds migrating between Eurasia and Africa. Comparisons of data collected at such ringing stations can shed new light on bird migration strategies after having crossed the Sahara Desert and around the Mediterranean Sea. However, although ringing operations have been conducted at these two migratory hotspots simultaneously since 1992, a study comparing their respective data has never been undertaken. To our knowledge and in spite of numerous advances in migration research over the past decades (
A large disparity in numbers was recorded between the spring and autumn migration seasons at both sites. Almost all migrants from both species were recorded during the pre-nuptial passage (98.1% and 99.2% of all ringed individuals of Western Bonelli’s Warbler at Gibraltar and Eastern Bonelli’s Warbler at Eilat, respectively). A similar disparity has also been observed in some other species of long distance migrant passerines studied at Eilat (e.g.,
In the case of both ringing sites, the very low number of birds recorded during the post-nuptial migration could be the result of the majority of Western and Eastern Bonelli’s Warbler populations having their stopovers further to the north, allowing them to overfly Gibraltar and Eilat on their way south (e.g.,
It is also possible that both sibling-species engage in loop migration, wherein the majority of the population flies to the wintering grounds along a different flyway, or that both species migrate on a broad front on their way to Africa by flying straight across the Mediterranean Sea (
The morphometric data for the two taxa offered in the published literature are similar to this study. The mean wing length and body mass of Eastern Bonelli’s Warbler on a Greek island was 67.1 mm and 6.9 g, respectively (
At Eilat, Eastern Bonelli’s Warblers peaked in the third decade of March. In contrast, for Eastern Bonelli’s Warblers on Antikythira, Greece, the median date of spring passage was 4 April (1st, 3rd quartiles: 2 April, 14 April), with the earliest bird trapped on 21 March 2008 and the latest on 8 May 2010 (
Western Bonelli’s Warbler migration peaked in the second decade of April at Gibraltar. This is similar to
It is of further interest that within the above mentioned similarity in migration period, the mean arrival date of Western Bonelli’s Warblers at Gibraltar was on average 15 days later than that of Eastern Bonelli’s Warblers at Eilat. One might argue that the position of Eilat at ca. 730 km further south than Gibraltar might play a part in this, although the discrepancy in the number of days is too great for the small difference in distance involved between the two sites. Differences may also be related to conditions in African wintering grounds, which can influence migratory patterns and phenology in particular, in a similar way to weather at stopover and breeding grounds (
We recommend future studies of similar focus to expand the scope by evaluating stop over sites on the Mediterranean coastline prior to their crossing the sea to Europe. This could explain why
Another point of interest in our data that relates to the idea forwarded that the two species may be fanning out of Africa to their breeding grounds in Europe, while the more easterly choose to stop over at Eilat, is the fact that we found a significant decrease in wing length during the spring passage in both study species. It has been established that longer distance migrants have relatively longer wings than conspecifics with shorter wings that migrate correspondingly shorter distances (for Eilat cf.
It is of interest that body condition of migrants only decreased over the passage period in the case of Eastern Bonelli’s Warbler and no that such pattern was evident in Western Bonelli’s Warbler. There may be several reasons for this loss of body mass. Older, more experienced birds are known to migrate more efficiently than inexperienced juveniles (cf.
In conclusion, our results not only expand existing knowledge of leaf warbler Phylloscopus migration patterns (e.g.,
We dedicate this study to all of those who have devoted their time and effort in ringing at both locations over the past decades. Furthermore, we thank Ian Henshaw and two anonymous referees for constructive comments on the manuscript. The data analyses were financially supported by the Polish Ministry of Science and Higher Education (grant to P.Z. No. IP2011063771).