Song and call identification of the first Green Warbler for the Netherlands

On Monday 24 June 2019 around 12:30, I returned to my home in the Netherlands after a pleasant morning of fieldwork near Lochem, Gelderland. While my son Julian – just home from school – was eating a sandwich, I was checking websites for interesting bird reports, like I do every day. Most notable was a Greenish Warbler Phylloscopus trochiloides discovered on this hot summer day by Francien Domenie in an urban park at Vlaardingen, Zuid-Holland. Casually, I played a sound-recording of the bird made by Karel Hoogteyling and then it all happened… My heart skipped a beat and I got a lump in my throat. The calls did not sound right for Greenish Warbler and the song was odd too! I had studied calls of Green Warbler P nitidus before and in my memory these ones sounded identical. The only problem: Green Warbler is a mega rarity in north-western Europe with, thus far, zero records in the Netherlands. I got a bit shaky…

I listened to the sound-recording a few more times and decided to contact Karel. I asked him if he had been able to see the bird and notice anything strange about it. When I told him that I suspected it to be a Green Warbler, I think Karel got a bit shaky too… It was time to send out a Dutch Bird Alert to inform the rest of the birding community before trying to see the bird myself. With high hopes I travelled to Vlaardingen and, apparently, I was not the only one. A total of c 40 birders searched for the bird until darkness fell, but unfortunately it was nowhere to be found. It turned out that just eight local birders had seen it that morning, before it disappeared around 11:00.

Description

The bird had been behaving like a typical Phylloscopus warbler, nervously moving and singing high up in the trees. It was much easier to hear than to see. Initially, the bird was identified as a Greenish Warbler based on its song. Features of its plumage seemed to justify this; several observers noted for example a prominent supercilium and a wing bar (formed by pale tips to the greater coverts). Two observers also saw a yellowish tinge on the breast, an important feature for Green Warbler. These sightings were only brief and not enough to let the alarm bells ring.

The bird was not photographed; three poor photographs posted online seem to show a Common Chiffchaff P collybita, of which a few individuals were present in the same area.

Identification

My identification as a Green Warbler is based on the sound-recordings of song and call made that morning by Francien Domenie and Karel Hoogteyling. Plumage features support this, and the yellowish tinge on the breast can be seen as a confirmation.

Characteristic of Green Warbler are:

1)    A rather complex cheeluee call, which in sonagrams looks w-shaped. The difference from Greenish Warbler, a species that turns up annually in the Netherlands nowadays (P t viridanus), is quite easy to hear; viridanus has shorter, hurried, ‘up-down’ calls (chwee-oo, n-shaped sonagram).

2)    Fairly short song phrases, which are not preceded by an introductory note and are built up out of rather complex elements (many V- or W-shaped). The first element often returns within the same song phrase; however, prominent repetitions are lacking. Songs of viridanus normally start with a short introductory note that is followed by a fast series of mostly descending elements. Single elements are often repeated in series a few times and songs are broken up into sections, more or less as in Common Chaffinch Fringilla coelebs.

Greenish Warbler complex

The Greenish Warbler complex is a very interesting one and has been well studied, more than once before. It is often cited as an example of a ‘ring species’; the group has its roots in the Himalayas, from where populations spread both eastward and westward around the Tibetan Plateau. For an overview of the variation in song and call, Irwin (2000) and Irwin et al (2008) are very useful sources. Hereafter, I will discuss the various taxa and illustrate the most important differences with sonagrams; for a good comparison I have made them on the same scale as those of the Vlaardingen warbler above (0-9 kHz, 0-3.5 sec). The focus is on how to separate Green Warbler from the other taxa. Sound-recordings of all taxa within the Greenish Warbler complex can be found on the internet, even if there are only a few. With the help of several people (see acknowledgements) I was able to collect enough recordings to get a clear picture of Green Warbler and of the most important differences from the other taxa.

Plumbeitarsus

Calls of the Vlaardingen warbler could be confused with those of plumbeitarsus (Two-barred Warbler). Songs of this East-Asian taxon, however, are clearly different. Its song phrases are usually fairly long and melodious, often with a swelling and undulating character, and with song elements that succeed each other at a high pace. Another striking feature is that phrases are usually built up out of fixed combinations of elements (‘song units’). Often, these units are repeated in series a few times within a phrase. There are numerous variants, including phrases consisting for a large part of just one very long song unit. However, the previously mentioned characters are still conspicuous (see for example descriptions in Irwin (2000) and Kovylov et al (2012)).

Obscuratus

This taxon – ‘intermediate’ between plumbeitarsus and nominate trochiloides – has its breeding range in central China and has never been recorded in Europe. Its song is clearly different from the Vlaardingen warbler, consisting of a simple repetition of one or two rather complex song units. Obscuratus is quite similar to nominate trochiloides and ludlowi in this respect (Irwin 2000).

Nominate trochiloides

This taxon breeds in the central and eastern Himalayas and has never been recorded in Europe. Its songs, in particular, are very different from those of the Vlaardingen warbler. They normally repeat just one rather complex song unit, often at a high pace and within a small frequency range (Irwin 2000).

Ludlowi

This taxon – ‘intermediate’ between nominate trochiloides and viridanus – breeds in the western Himalayas and has never been recorded in Europe. Both calls and songs are clearly different from those of the Vlaardingen warbler. The use of repeated song units is typical also for this taxon (Irwin 2000).

Viridanus

Viridanus has a rather large breeding range, extending all the way to north-western Europe. Both calls and songs are clearly different from those of the Vlaardingen warbler. In the entire range of virididanus, the commonest call-type is characterised by a descending final part. This is confirmed by Irwin et al (2008) and by many sound-recordings on the internet.

Song of viridanus clearly differs from the taxa discussed above and from nitidus. Typical song phrases start with a short introductory note, followed by a fast series of mostly descending elements. Single elements are often repeated in series a few times and songs are broken up into distinct sections. Repetitions of song units within the same phrase are rare.

Nitidus

Nitidus is the only one that ticks all the boxes! Common calls of this taxon are w-shaped, with a rising final element, quite similar to calls of plumbeitarsus but on average more ‘compact’ and higher-pitched. Calls of the Vlaardingen warbler had a minimum frequency of c 3.0 kHz, a maximum frequency of c 8.4 kHz and a duration of c 0.25 sec. These values are in line with nitidus and also the structure of these calls was similar, as the sonagrams below show.

Songs of nitidus differ from those of all other taxa by the very few repetitions within their phrases. One or two elements may return within the same phrase, but series of the same elements or units are rare. Furthermore, many elements are V- or W-shaped. Because they resemble one another but are not identical, the sonagram seems to be almost ‘hand-written’. The stereo-typed repetitions in songs of the other taxa, on the other hand, make those sonagrams look like they were ‘printed’.

The variation in songs of plumbeitarsus, obscuratus, nominate trochiloides, ludlowi and viridanus is described extensively by Irwin (2000). This publication helped me enormously to understand the differences between nitidus and all the other taxa. However, Irwin (2000) and Irwin et al (2008) did not study nitidus song, so I listened to and analysed the available recordings myself, including ones Magnus and I made during a Sound Approach trip to eastern Turkey in 2002. At first sight, these recordings of nitidus songs show a series of elements, almost randomly ordered. Then taking a closer look at several phrases by the same bird, it becomes clear that these are also built up out of distinct units, i.e. fixed combinations of elements.

Normally, song phrases of nitidus consist of only one or two units. When there are two these are different. Because the units are often incomplete, the entire strophe often sounds more ‘improvised’. Single elements or incomplete units may return within the same phrase, but obvious repetitions are lacking. This explains why the presence of song units in nitidus is rather inconspicuous.

The use of song units by the Vlaardingen warbler is in line with other recordings of nitidus, as the figure below shows.

I have summarised the most important findings of my study of songs of all taxa in the Greenish Warbler complex in the table below. It is clear that in my opinion there can be no doubt that the Vlaardingen warbler really was a Green Warbler!

Acknowledgements

A number of people have contributed to this study. It all started with Francien Domenie, who discovered the bird. Karel Hoogteyling and Ben Gaxiola shared their field impressions of the bird with me. Magnus Robb, Per Alström, Pratap Singh and Trevor Price helped me in various ways and provided sound-recordings. Peter Boesman, Oscar Campbell, Francien Domenie, Karel Hoogteyling, Jens Kirkeby, Lars Lachmann, Jelmer Poelstra and Stanislas Wroza also kindly let me use their recordings. Furthermore, I have studied many recordings published on websites like Xeno-canto and Macaulay Library. And last but not least, I would like to thank The Sound Approach for giving me the opportunity to be part of a trip to eastern Turkey back in 2002 to study Green Warblers and other birds.

References

Irwin, D E 2000. Song variation in an avian ring species. Evolution 54: 998-1010.

Irwin, D E, Thimgan, M P & Irwin, J H 2008. Call divergence is correlated with geographic and genetic distance in greenish warblers (Phylloscopus trochiloides): a strong role for stochasticity in signal evolution? Journal of Evolutionary Biology 21: 435-448.

Kovylov, N S, Marova, I M & Ivanitskii, V V 2012. Variation of song and plumage in the Western (Phylloscopus trochiloides viridanus) and Eastern (Phylloscopus trochiloides plumbeitarsus) forms of the Greenish Warbler in a sympatry zone: is the hypothesis of ring speciation true? Biology Bulletin 91: 702-713.