Radiation in Yellow Wagtails Motacilla flava has resulted in a species-swarm of about 18 taxa which are distinguishable by the often characteristic summer nuptial plumages of adult males (Alström & Mild 2003). Yet, in spite of distinctive plumage differences, the taxonomic status of all taxa is as yet far from resolved. Some authors recognize only M. flava as a polymorphous species (e.g. Glutz von Blotzheim & Bauer 1985, Tyler 2004), whereas others elevate 11 taxa to species status based on differences in morphology of adult males (Sibley & Monroe 1990, Sangster et al. 1999).

Recent molecular studies indicate a likely polyphyletic origin of M. flava. Based on sequences of mitochondrial DNA Voelker (2002) and Pavlova et al. (2003) grouped all taxa into three clades which, as they state, should be considered species. Ödeen & Björklund's (2003) study on mitochondrial and nuclear DNA, however, merged the two eastern clades but supported the western clade of Voelker (2002) and Pavlova et al. (2003), effectively implying two separate species within the M. flava complex. Obviously, Yellow Wagtail taxonomy is in a state of flux and as Tyler (2004, p.695) states “a great deal of additional research is required, both in the field and in the laboratory, before the relationships (…) of the taxa are understood.”

Figure 1.

Map showing the study area along rivers Ob and Irtysh, with towns Khanty-Mansiysk and Surgut, Russia.

If all western taxa belong to one monophyletic species (Voelker 2002, Ödeen & Björklund 2003, Pavlova et al. 2003), hybridization should occur between sympatric taxa. Though hybridization is often reported and hybridization zones exist (Glutz von Blotzheim & Bauer 1985), assigning individuals to hybrid forms is not always straightforward due to phenotypic variability within populations of a given taxon. In populations of dark-headed birds, for instance, individuals with clear eyebrows are reported though no neighbouring taxon with eyebrows occurs, which makes hybridization as a cause unlikely (Glutz von Blotzheim & Bauer 1985). Therefore, the regularity with which hybridization occurs might be difficult to ascertain. Some taxa (iberiae, thunbergi, feldegg, beema and lutea) are not reported to interbreed extensively when occurring in sympatry (Glutz von Blotzheim & Bauer 1985). For instance, the South European iberiae and cinereocapilla are geographically each other's neighbours but not known to interbreed and their wintering areas are widely separated as well (Alström & Mild 2003). Additionally, feldegg interbreeds only to a limited extent with beema with which it does occur in sympatry (Glutz von Blotzheim & Bauer 1985).

Since some taxa within the western species (Voelker 2002, Ödeen & Björklund 2003, Pavlova et al. 2003) are less prone to hybridization than others, there seems to be incongruence between contemporary molecular insights and what is actually happening on the breeding grounds. General as their habitat preferences may seem, Tyler (2004) specifies those habitats as rather diverse “from damp meadows, marshes (…), bogs to damp steppe and grassy tundra.” These subtle yet distinct differences in vegetation and structure could provide an opportunity for segregation by breeding habitat requirements. Indeed, some authors describe varying habitat preferences between neighbouring taxa (Dement'ev & Gladkov 1954, Bakhtadze & Kazakov 1985, Glutz von Blotzheim & Bauer 1985, Boyko 1998a, 1998b, Alström & Mild 2003). It is clear that besides molecular studies, more field studies are needed to assist in clarifying taxonomy and habitat requirements of different M. flava taxa.

Two allegedly interbreeding taxa are thunbergi and beema (Alström & Mild 2003), which geographically occur in little studied sympatry, along the borders of their summer ranges in European Russia and Western Siberia (Dement'ev & Gladkov 1954). By visiting their West-Siberian environs of sympatry we determined whether both taxa share their breeding habitat. We aimed to determine whether breeding habitat could be an ecological factor for sub-specific segregation in this part of the breeding range of thunbergi and beema.

Methods

Between 2002 and 2012 we made opportunistic observations during ornithological expeditions in the Khanty-Mansi Autonomous Okrug (Western Siberia, Russia), where we visited extensive Sphagnum bogs (n = 12 locations; Appendix 1) and the vast floodplains of the rivers Irtysh, Ob and tributaries (n = 7 locations; Appendix 1). All locations were in an area of approximately 300 km by 200 km, roughly between towns Khanty Mansyisk and Surgut (Figure 1). This area is allocated in a broad zone of sympatry of both taxa, which in Western Siberia is roughly between 55°–60°N (Red'kin 2001, Ryabitsev 2008). Since thunbergi and beema are long-distance migrants (Glutz von Blotzheim & Bauer 1985) we are keen to distinguish occasional migrants from breeding birds. Migrants can turn up in any damp habitat, a potential pitfall when studying the birds' breeding habitat. To minimize confusion we decided to only use observations of feeding or strongly alarming parents, indicating the presence of (fledged) juveniles. Upon finding birds meeting our criteria we established their sub-specific identity. Fortunately, males of both taxa are easy to separate in the field: thunbergi males have a distinctive all dark head and a yellow throat, whereas beema males have a light, bluegrey head, a strong supercilium and a white throat (Alström & Mild 2003). There is controversy over the taxonomic status of dark-headed birds east of the Ural Mountains. These birds are widely recognized as belonging to the taxon plexa (Tyler 2004) but the validity of the taxon has been a subject of debate. Alström & Mild (2003) merged plexa with thunbergi since plexa specimens appeared indistinguishable from thunbergi, based on plumage and morphological characteristics. Therefore, and because thunbergi takes nomenclatural precedence over plexa (Tyler 2004), we choose to assign our dark-headed birds to thunbergi, though we are aware that they might refer to plexa since our site is close to collecting sites of plexa (Pavlova et al. (2003).

Figure 2.

Floodplain and bog are separated by 1.5–2 km of forest near Mukhrino Field Station, where thunbergi and beema occur on either side of the forest ecotone (photo Rick Ruijs).

Females are much more difficult to separate, and to avoid faulty identification we focussed on adult males only. Next, we described vegetation composition of breeding habitat and carefully investigated the site for breeding birds.

Results and discussion

We found indications of differential habitat preferences in thunbergi (n = 12 locations, minimum of n = 97 birds; Appendix 1) and beema (n = 7 locations, minimum of n = 31 birds; Appendix 1). Thunbergi was only encountered as a breeding bird on Sphagnum bogs where we never saw breeding beema. The latter occurred as a breeding bird on floodplains of the river Irtysh and here beema was the predominant taxon breeding, despite careful searching.

As one would expect, floodplains and bogs differed tremendously in vegetation composition. Sphagnum bogs consisted of a mesh of ridges and hollows. On ridges dwarfshrubs Ledum palustre, Betula nana and Rubus chamaemorus grew within lush moss vegetation of Sphagnum, Dicranum and Polytrichum. The occasional oppressed Pinus silvestris grew as the only tree species on the breeding grounds of thunbergi. Lichens were only scantly encountered, and rarely one encountered Eriophorum and Carex sedges. In the moister hollows Carex, Eriophorum, Menyanthes trifoliata (and rarely Comarum palustre) grew. Moss cover consisted of Sphagnum, Dicranum and Polytrichum.

In contrast, the open, grassy floodplains were dominated by Poaceae (Agrostis gigantea, Calamagrostis purpurea and Phalaroides arundinacea), Carex spp. (C. aquatilis, C. acuta, C. nigra and C. gracilis) and occasional Salix trees.

Our observations are in concordance with Dement'ev & Gladkov (1954), who found that thunbergi generally occurs in peat bogs and beema on meadows and river valleys. Additionally, Boyko (1998a, 1998b) writes that “thunbergi generally [inhabits] Sphagnum bogs, and beema [is] mostly found in meadows, lowland bogs and near human habitation” in the Konda river lowlands (roughly at 60°42.2′N, 69°40.1′E, adjacent to the western borders of our study area). Given that flava and beema are closely related (Dement'ev & Gladkov 1954, Alström & Mild 2003), it is of interest to note that these results match a study in central Sweden, where thunbergi also breeds on bogs and flava in the adjacent meadows (Bylin 1974 in Glutz von Blotzheim & Bauer 1985).

In spite of our limited sample-size, segregation by strongly differing habitats appeared rather remarkable. As an example, near Shukhrino Field Station (60°53. 2′N, 68°41.4′E), floodplain meadows and Sphagnumbogs were separated by only 1.5–2 km of forest but here too, solely thunbergi occurred on the bogs and beema on the floodplains (Figure 2). However, in 2011 we observed individuals of both taxa along the lower reaches of the river Vynga (right tributary of the river Ob) in floodplain sedge meadows. Birds of both taxa were observed either alarming or feeding nestlings, indicating that both were breeding locally. The rarity of this observation is corroborated by remarks of Boyko (1998a), who states that “mixed pairs of both subspecies are not regular anywhere, though occasionally beema nests near thunbergi on Sphagnum bogs both in the Konda and Surgut regions.” Alström & Mild (2003) however, report thunbergi and beema to intergrade in the northern Kirgiz steppes, but they do not provide details on the observed frequency of hybridization, or breeding habitat.

We conclude tentatively that thunbergi and beema appear to be largely segregated by breeding habitat in the area of study and this conclusion is corroborated by results of studies in other areas (Dement'ev & Gladkov 1954, Glutz von Blotzheim & Bauer 1985, Boyko 1998a, 1998b). However, given the still rather limited dataset and size of study area it remains to be seen whether this ecological segregation reflects the general situation along the vast area of sympatry of thunbergi and beema. If so, taxonomic conclusions based on breeding ecology are perhaps inconsistent with recent molecular findings (Voelker 2002, Ödeen & Björklund 2003, Pavlova et al. 2003).