Content uploaded by Nicolas Martinez
Author content
All content in this area was uploaded by Nicolas Martinez on Jun 15, 2020
Content may be subject to copyright.
163
Geographical variation in Common
Redstart calls
Nicolas Martinez & Ralph Martin
Common Redstart Phoenicurus phoenicurus
occurs in two subspecies. Nominate P p
phoe nicurus (hereafter phoenicurus) breeds
across most of the species’ range, from Morocco,
Spain and Britain in the west to lake Baikal,
Russia, in the east. The other subspecies, Ehren-
berg’s Redstart P p samamisicus (hereafter sama-
misicus), has a south-easterly distribution. Its oc-
currence stretches from eastern Anatolia, Turkey,
the southern Caucasus and the Middle East to
south-western central Asia (Glutz von Blotzheim
& Bauer 1988, Clement & Rose 2015). However,
some authors include the whole of Turkey and
central Asia in the breeding distribution range of
samamisicus (Roselaar 1995, Kirwan et al 2008,
Shirihai & Svensson 2018). Birds in the Balkan,
Crimea and the northern Cau ca sus may show a
mix of features and some consider these to be in-
tergrades between the two subspecies (Glutz von
Blotzheim & Bauer 1988, Svensson 1992, Kirwan
et al 2008, Martinez 2010, Clement & Rose 2015);
more research is needed. Adult males of both sub-
species are identied by the presence (samamisi-
cus) or absence (phoenicurus) of a white wing
patch. Given good views, most rst-winter males
can be identied on other plumage characters,
too (Small 2009). Identication of females based
on plumage features is only tentative (Small 2009,
Clement & Rose 2015, Shirihai & Svensson 2018).
Addition ally, contact calls were thought to be a
reliable identication feature for some time
(Bergmann et al 2008, Small 2009, Ayé et al 2012,
Svensson et al 2015): phoenicurus was meant to
invariably utter a rising, slightly dissylabilic, huid,
whereas the analogous call of samamisicus was
described as a heed with constant frequency. It
was generally assumed that phoenicurus shows
the huid call across its whole range, until pheno-
typic phoenicurus with heed calls were observed
in the east of its breeding range (Ayé et al 2014).
In this study, we analysed calls of Common
Redstarts from most parts of its breeding range to
get more information about the distribution of the
different calls.
Methods
We collected sound recordings of Common Red-
starts (in total 287 individuals). The main source
for recordings was www.xeno-canto.org (record-
ings of 127 individuals). We considered all re-
cordings of calls available from the breeding pe-
riod (May-July) published on this website by the
end of July 2019. We added six birds of samamisi-
cus from the second half of April, which sang the
[Dutch Birding 42: 163-174, 2020]
FIGURE 1 Illustration of how measurements were taken from sonagrams for different call variants. Red crosses mark
position from which we noted time and frequency. Gradient (blue line) was calculated by dividing frequency differ-
ence between 1st and 2nd frequency (1st Freq/2nd Freq) point by length.
Length
Length
Length
Length
1st Freq 1st Freq
1st Freq
1st Freq
2nd Freq 2nd Freq
2nd Freq
2nd Freq
164
typical song of this subspecies (based on Ayé et al
2014) and therefore likely defended a territory
(southern birds arrive earlier at their breeding
sites; Glutz von Blotzheim & Bauer 1988). In or-
der to get a more representative collection from
specic regions, additional material (number of
individuals used given in parentheses) from May
to July as well as from birds where videos proved
that they were obviously breeding was included
from: www.youtube.com (59 individual), own and
privately shared recordings (38), www.facebook.
com (17), www.macaulaylibrary.org (13), several
ornitho-platforms (12; Austria, France, Germany,
Italy, Spain, Switzerland), www.tierstimmenarchiv.
de (7), www.observation.org (3), www.hbw.com/
ibc (1), www.tarsiger.com (2), www.vimeo.com
(2), www.mdahlem.net (1), www.birdsong.it (1)
and www.sounds.bl.uk (1). We determined the
call variant(s) for each recording by listening to
the recording and checking the sonagram. There-
fore, all recordings were converted to 44 kHz wav
les (16 bit). Sonagrams were made with Raven
(Bio acoustics Research Program 2014). We used
Kaiser-Window to visualize the calls in combina-
tion with a sample overlap of 95% and a DFT size
of 500 samples.
We categorised the calls qualitatively based on
previous described categories (huid and heed),
and if that was not possible, we made new cate-
gories. For a denition of the call variants see sec-
tion ‘Results’. Just one recording was considered if
several recordings of the same call variant at the
same site in the same year were available, unless
it was clear that different birds were involved.
Geographical variation in Common Redstart calls
FIGURE 2 Variation of different types of calls of Common Redstart Phoenicurus phoenicurus. 1 Typical example of
huid call in central Europe. Most of recorded huid calls strongly resemble this and following example. Male,
Gütersloh, Germany, 9 June 1987 (Klaus Conrads; TSA 0148); 2 female, Lombardia, Italy, 27 May 2016 (Francesco
Sottile; XC318770); 3 extraordinary steep example, Hautes-Alpes, France, 17 May 2011 (Peter Boesman; XC270190);
4 Staffordshire, England, 6 May 2018 (Dominic Garcia-Hall; XC414853); 5 female, Cheboksary, Russia, 22 June
2013 (Albert Lastukhin; XC139395); 6 especially birds from eastern distribution range showed very pronounced
difference in gradient between beginning and end of call. Korgalzhyn, Kazakhstan, 3 May 2013 (Thijs Fijen;
XC145071); 7 moreover, two of three birds from Mongolia had downward inected end. Tereldsh, Mongolia, 11 June
1979 (Dieter Wallschläger); 8 this and three following calls are examples of commonest call variant in Italy north to
Alps. Toscana, Italy, 15 June 2011 (Marco Dragonetti; XC118530); 9 Calabria, Italy, 3 July 2015 (Francesco Sottile;
XC254972); 10 Italy, May 2011 (Simona Inaudi; ornitho.it); 11 Ticino, Switzerland, 30 June 2011 (Tanya Harvey
Ciampi; Youtube); 12 Huelva, Andalucía, Spain, May (Alberto Plata Ortiz); 13 this and following recording show
extraordinary records of heed calls in terms of geography. Stromberg, Rheinland-Pfalz, Germany, 15 June 2015
(Frank Holzapfel; XC254143); 14 Dividal, Troms, Norway, 23 June 2011 (Stein Ø Nilsen; XC92981).
Germany
Germany
Italy
ItalyItalyItaly Switzerland Spain Norway
France Britain Russia Kazakhstan Mongolia
P p phoenicurus
P p phoenicurus?
huid calls
heed calls (gradient > 2 kHz/s
165
Geographical variation in Common Redstart calls
FIGURE 2 (continued) 15 male, Samtskhe-Javakheti, Georgia, 6 July 2013 (Jarmo Pirhonen; XC142822); 16 one out
of only three rising heed calls from breeding range of P p samamisicus, still quite ‘at’ in sonagram. Georgia, 31 May
2018 (Aslan Bolkvadze; Facebook); 17 Roodbarak, Mazandaran, Iran, 15 April 2017 (Patrik Åberg; XC405281);
18 this call was recorded in supposed intergradation zone. Male bird showing features of P p samamisicus, Thrace,
Greece, 16 May 1976 (Hans-Heiner Bergmann); 19 following three calls show constant heed calls from P p phoeni-
curus breeding ranges. Bird from call 19 is same bird as in plate 220. Teramo, Abruzzo, Italy, 28 May 2016 (Dimitri
Marrone; XC318906); 20 this bird switched regularly between these heed calls and typical huid calls. Astana,
Kazakhstan, 4 July 2015 (Youtube); 21 Ala Archa NP, Kyrgyzstan, 18 May 2015 (Ralph Martin; XC312877); 22 male,
Gijón, Asturias, Spain, 24 May 2011 (Youtube); 23 female, Lisboa, Portugal, 17 May 2009 (Magnus Robb);
24 Lugones, Asturias, Spain, breeding period 2014 (Vimeo); 25 note pronounced downward inection at end of call,
which is typical for vist variant type 2 and 3. Andalucía, Spain, 19 May 2015 (José Carlos Sires; XC274770);
26 Andalucía, Spain, 7 June 2015 (José Carlos Sires; XC279476); 27 Andalucía, Spain, 21 May 2016 (José Carlos
Sires; XC319577).
In a statistical analysis, we tested for sexual and
geographical differences in the calls. Therefore,
we took several measurements (gure 1) and cal-
culated the gradient (kHz/s) of the call by dividing
the difference between 1st and 2nd frequency
through the timespan. As calls within individuals
were relatively stable, we chose one representa-
tive call per individual for testing. We tted a lin-
ear model to the measurements of the huid calls
(101 sexed and 96 unsexed individuals) and an-
other one to the measurements of the heed calls
(23 sexed and 44 unsexed individuals), using the
‘lm’ function of the statistics package of R (R Core
Team 2016), to check for differences of the calls
between sexes as well as geographic differences.
We checked various diagnostics of model validity
and stability. The t of the models met the assump-
tions of normatily and homogeneity of residuals
(Quinn & Keough 2002, Field 2005). We analysed
the inuence of longitude, latitude, their interac-
tion and sex on the gradient of the call, 1st and
2nd frequency and call length.
Results
Call variants of Common Redstart
We analysed 287 individuals, covering most parts
of the species breeding range. The following call
variants were detected: rising huid, rising or con-
stant heed calls and three variants of a vist call
(gure 2). Huid and heed calls are well known but
the variability of heed calls was much higher than
previously known. For a few calls it was thus far
from straightforward to assign them to either vari-
P p samamisicus P p phoenicurus?
KazakhstanIranGeorgiaGeorgia Greece Italy Kyrgyzstan
SpainSpainPortugalSpain Spain Spain
P p phoenicurus
vist calls, variation 1
heed calls (gradient < 2 kHz/s
vist calls,
variation 2
vist calls,
variation 3
166
ant (huid or heed) and some had to be classied
as intermediate. We have retained the two vari-
ants, however, as the calls in previous publica-
tions were classied as such. Vist calls had not
been described before to our knowledge. Call
variant was stable within an individual, with the
exception of at least eight birds from Kazakhstan
and Russia switching between huid and heed
calls. All call variants were regularly combined
with ticking alarm calls and do not seem to differ
in this respect.
Rising huid call
This is the classic call of phoenicurus that central
and northern European birdwatchers are familiar
with and can be best described by huid. All but
three birds (n=170) used huid calls (98%, gure 3)
within the breeding area of phoenicurus in Central
Europe north of the Alps, Britain, Fennoscandia
and European Russia. The call is clearly uprising
(similar to the contact call of Willow Warbler
Phyllo scopus trochilus), with a mostly low starting
frequency. Gradient increases towards the end of
the call, which makes the call sound disyllabic. In
the linear model, sex had no signicant effect on
the measured values. Therefore, we ran another
model without ‘sex’ as a covariate to increase the
sample size by using the unsexed birds as well.
Latitude and longitude showed signicant effects
on gradient, 1st and 2nd frequency and call length
(appendix 1 and 2). Gradient increased from
south-west to the north-east while call length in-
creased from south-east to north-west (appendix
3). 1st frequency was highest in the south-east and
lowest in the north-east, while 2nd frequency was
lowest in the south-west and north-east and high-
est in the north-west and south-east.
Heed call
The heed calls form a line with a constant, rising
or slightly decreasing gradient when visualised in
a sonagram. Calls given at constant frequency
(reminiscent of the slightly higher pitched contact
calls of Collared Flycatcher Ficedula albicollis) are
Geographical variation in Common Redstart calls
FIGURE 3 Distribution of call variants across breeding range of Common Redstart Phoenicurus phoenicurus. Each
marker corresponds to an individual. If markers covered each other, we slightly shifted points to retain visibility. Map
sources: www.birdlife.com (breeding range) and Natural Earth (www.naturalearthdata.com, background, countries).
Subspecies ranges based on Shirihai & Svensson (2018); note, however, that some authors consider birds of Balkan
peninsula and western Turkey to be intergrades. Note that there are no recordings for one individual from Altai moun-
tains and one from lake Baikal, that were switching between huid and heed call variants (Raffael Ayé & Balduin
Fischer pers comm). As stated in text, recording quality of three birds from Spain labelled as vist type 1 does not allow
to decide if their calls are better assigned to vist or heed calls.
Call variants
huid (gradient > 10 kHz/s)
huid (gradient 6-10 kHz/s)
huid (gradient 2-6 kHz/s)
intermediate (rising heed/huid)
rising heed (gradient > 6 kHz/s)
rising heed (gradient 2-6 kHz/s)
constant heed (gradient < 2 kHz/s)
switcher (huid-heed)
vist, type 1
vist, type 2
vist, type 3
Subspecies breeding range
P p phoenicurus
P p samamisicus
167
known as the typical samamisicus call. We found
only this call type from the breeding range of sa-
mamisicus (n=20; two calls showed a gradient
>2.0 kHz/s, none exceeded 4.0 kHz/s), but also
on the Balkan and in Italy including adjacent re-
gions from Austria, southern France and Switzer-
land. Heed calls with a rising gradient are espe-
cially widespread in the southern Alps with a u-
ent transition to huid calls with a small gradient.
As a consequence, several calls had to be classi-
ed as intermediate (Alps, northern Balkan, gure
3). Rise in pitch of heed calls become roughly
audible when starting and ending frequency differ
by at least 0.3 kHz (but note that there are sub-
stantial differences between observers in this re-
spect). Heed calls were also detected in Kyrgyzstan
close to the south-eastern border of the phoe-
nicurus range, and in the Russian Far East and
Kazakhstan, where several birds combined this
call type with the huid call (‘switchers’ in
gure 3-4).
In the linear model, sex had no signicant effect
on the measured values, but sample size was very
small (n=23). However, as sex had no signicant
effect for the huid calls either, we ran another
model without ‘sex’ as a covariate to increase the
sample size by using the unsexed birds as well.
Latitude and longitude showed signicant effects
on gradient, 1st and 2nd frequency but not on call
length (appendix 4 and 5). Gradient increased
while 1st and 2nd frequency decreased from
south-east to north-west (appendix 6).
Variations of vist calls
We found highly differing calls in Iberia (plus one
call from adjacent south-western France). Here,
Common Redstart is patchily distributed and re-
stricted to mountainous regions, especially in the
south (see gure 3). Some birds showed interme-
diate calls between heed and huid but with a
rough or hoarse sounding in some, thanks to a
quickly modulated uprising call (vist variation 1).
Most south-western calls (vist variations 2 and 3)
had a prominent downward inection in the end
in common (calls reminded a Common Chiffchaff
Phylloscopus collybita with sweo calls), and some
showed even further uprising and decreasing ele-
ments (variation 3). These calls are reminiscent of
calls of Black Redstart P ochruros but are given at
lower frequencies (vist calls in Common Redstarts
start at frequencies between 3.0 and 3.8 kHz, the
highest point in the sonagram equalled 4.4-5.2
kHz. Contrarily, Black Redstart calls range rough-
ly from 4.5 up to 6.0 kHz at the highest point). A
total of three recordings from Spain (2) and
Portugal (1) with low recording quality sound sim-
ilar to heed calls (and maybe they are) but record-
ing quality is not sufcient to exclude further vari-
ants of vist calls.
Discussion
In this study, we analysed call variation of
Common Redstart of most parts of its distribution
range. Phoenicurus of the north-western popula-
tion uttered almost exclusively huid calls. Birds
from the samamisicus breeding range only used
heed calls and the majority of calls of samamisi-
cus showed almost no gradient. We found birds
giving heed calls on the Balkan (in line with the
hypothesis of an intergradation zone in this re-
gion; Matvejev & Vasi 1973, Clement & Rose
2015) and even in southern France and, more so,
on the Italian peninsula, where surprisingly few
birds with huid calls were recorded (just two huid
callers out of 17 birds from Italy south of Genova).
Calls of some of these heed calling birds were
identical to calls that were considered to be typi-
cal of samamisicus. In these regions, at least from
time to time, unusual birds with rather prominent
white wing patches seem to occur, and some were
Geographical variation in Common Redstart calls
FIGURE 4 ‘Switcher’, using huid and heed calls. Adult, female, same bird as in plate 221. Mikhaylovka, Irkutsk
Oblast, Russia, 19 June 2019 (Ralph Martin).
switcher
168
claimed to be samamisicus. According to Small
(2009), seven claims come from southern France,
but none was sufciently documented to exclude
any phoenicurus with extensive pale fringing on
secondaries and, subsequently, the French rarities
committee has not accepted them (Frédéric Jiguet
pers comm). There are two records of samamisicus
in Italy. The rst, a bird from Lam pedusa on 26
September 2007, mentioned in Ruggieri & Sighele
(2008), Small (2009) and Clement & Rose (2015),
should possibly be re-evaluated in light of new
knowledge (Andrea Corso pers comm; see also
rejected bird in the Netherlands, cf Wassink &
Ebels (2005)). The second has been caught at a
ringing station on the small Pontine island Vento-
tene on 2 May 2019 (www.ornitho.it/index.
php?m_id=54&backlink=skip&mid=289548,
Jacopo Bar chiesi pers comm). With only one or
two certain records from the region, it is reason-
able to assume that pure samamisicus are at least
genuinely rare in southern France and Italy and
the high proportion of heed callers here is unlike-
ly to be explained by samamisicus vagrants.
Further more, it is worth emphasizing that some of
the Italian birds giving heed calls (constant and
rising) were photographed or lmed, and none of
the males showed a prominent white wing panel
as would be expected in adult samamisicus. We
Geographical variation in Common Redstart calls
220 Common Redstart / Gekraagde Roodstaart Phoeni curus phoenicurus phoenicurus, second calender-year male,
Teramo, Abruzzo, Italy, 24 April 2016 (Dimitri Marrone). This breeding bird repeatedly gave constant heed calls
(cf gure 2, call 19). Note that it has lost its tail. Most P p samamisicus of same age show some whitish base to sec-
ondaries (Small 2009). 221 Common Redstart / Gekraagde Roodstaart Phoeni curus phoenicurus phoenicurus, adult
female, Mikhay lovka, Irkutsk Oblast, Russia, 20 June 2019 (Ralph Martin). Only female in our study that switched
between huid and heed calls was this male-like coloured bird (cf gure 4). 222 Common Redstart / Gekraagde
Roodstaart Phoeni curus phoenicurus phoenicurus, adult male, Sierre de las Nieves, Andalucía, Spain, 30 May 2016
(Ricky Owen). Bird from southern Iberia, whose breeding birds give highly distinct vist call variants 2 and 3.
169
thus exclude the possibility that these birds were
pure samamisicus. However, the increase of the
gradient of the heed calls to the north-west might
be a hint for introgression of samamisicus into
phoenicurus in the Balkan and Italy (if calls are
inherited in the Common Redstart). Hogner et al
(2012) analysed mtDNA in Common Redstart.
Unfortunately, their markers did not separate
phoenicurus and samamisicus but Ertan (2002)
used methods which showed differences between
both subspecies but also high gene ow between
them, supporting a large mixing zone. It is pos sible
that the corresponding locus or loci for call inherit-
ance can introgress independently from plumage
patterns. Typically, morphological characters are
used to distinguish between subspecies but here
they do not seem to be conclusive, as the line be-
tween phoenicurus and samamisicus is drawn very
differently by different authors (including the
Balkan as large contact zone in Clement & Rose
(2015) while Shirihai & Svensson (2018) dene
birds in western Turkey as intergrades) or to be ob-
scured by second calendar-year male samamisicus
with juvenile primaries that may have been some-
times misidentied as phoenicurus.
In eastern regions, where samamisicus and
phoe ni curus might get in contact on migration or
thanks to overshooting birds, heed calls occur as
well (in the Altai, Ayé et al 2014; Tian-Shan, lake
Baikal and north of the Caspian Sea, gure 3).
Surprisingly, birds able to switch between huid
and heed calls occur there. While in the begin-
ning, it seemed that the description of two such
birds was more unusual (Ayé et al 2014), we were
now able to nd a total of eight such ‘call switch-
ing’ birds (almost half of the birds recorded in the
Far East). Indeed such birds might be often over-
looked, as at least some birds seemed to switch
between the two calls only when they were very
agitated. With little agitation, however, just one
call variation was uttered and they would have
been classied as typical huid or heed callers dur-
ing a brief encounter. This leads us to conclude
that call switching from huid to heed occurs regu-
larly in Far Eastern populations. While we did not
nd any hint of such call switching in Western
Europe during our analyses, NM found one fe-
male in Switzerland shortly before publication of
the present article, that uttered few heed calls, but
huid calls were the norm (XC562589). Thus, call
switching may very rarely be present even in
Western Europe.
Another unexpected nding for such a com-
mon species in Europe was the existence of calls
in Iberia which are very different from the hitherto
described calls of Common Redstart. There might
be a geographical pattern of the different vist calls,
however, vist variation 2 and vist variation 3 calls
were found close to each other with very small
sample sizes and might be just a variant of the
same call. We also do not know what the transi-
tion between these vist calls looks like, nor if the
three huid calls and the two heed calls from Iberia
(gure 3) were just from very late migrants or if
these calls occur regularly here, too. Although vist
calls of these Iberian birds are quite different from
calls of all other phoenicurus populations, we no-
ticed that song does not differ in length and vari-
ability of the introductory part. Thus, we agree
with the ndings of Ayé et al (2014) that song in
phoenicurus seems to be very constant across the
whole breeding range.
To sum up, it would be interesting to obtain
more recordings of Common Redstart calls of
breeding birds from the Iberian Peninsula and,
even more, from the Maghreb as well as from the
south-eastern phoenicurus range in Kyrgystan and
eastern Kazakhstan and westernmost China and
from samamisicus birds from Tajikistan and Uzbe-
kistan.
Conclusions
Following our data, huid calls seem to be restrict-
ed to phoenicurus as we found no samamisicus
giving this call variant. Moreover, heed and espe-
cially rising heed calls seem to be common near
contact zones of both subspecies. Variation of the
calls of phoenicurus is larger than previously
thought. We found undescribed call variants in
Iberia and heed calls to be common in southern
Europe, especially in Italy, perhaps originating
from genetic introgression with samamisicus
(maybe independent of plumage features). Thus, a
heed calling Common Redstart can no longer be
considered as a valuable samamisicus candidate
or even an ‘Eastern Common Redstart’ anymore.
We recommend that in central and western
Europe, only very typical adult male birds should
be recognized as samamisicus (features given in
Small 2009). However, the documentation of less
typical birds would be helpful to better under-
stand the total variation and a thorough study of
plumage characters in suspected intergradation
zones of phoenicurus and samamisicus would be
much appreciated.
Acknowledgements
First of all, we would like to thank the people that made
their recordings, videos and photographs available,
either directly by sending them to us or by uploading
Geographical variation in Common Redstart calls
170
them on the internet: Patrik Åberg, Nils Agster, James
Aldridge, Georg Amman, Volker Arnold, Nils Austa,
Raffael Ayé, Marcin Bączkowski, Enzo Baratta, Jacopo
Barchiesi, Paolo Bello, Rob van Bemmelen, Hans-Heiner
Bergmann, Jack Berteau, Lorenzo Bianchini, Peter Boes-
man, Aslan Bolkvadze, Bernard Bousquet, Pierandrea
Brichetti, Joost van Bruggen, Oscar Campell, Florent
Cerruti, Tanya Harvey Ciampi, Stefano Conconi, Alan
Dalton, M Di Lullo, Marco Dragonetti, Bruno Durand,
Lars Edenius, João Esteves, Matthias Feuer senger, Balduin
Fischer, Stuart Fisher, Enric Fontcuberta, Franco
Francesco, Dominic Garcia-Hall, Maria Rita Gelso,
Galina Georgieva, Olivier Grangier, Leonardo Graziani,
Håkan Gustavsson, Louis A Hansen, H Heerklotz, Franck
Hollander, Frank Hollan der, Frank Holzapfel, Juha
Honkala, Gordana Ibis, Francisco Javier Calvo Iesmes,
Tatyana Ilyina, Simona Inaudi, Vadim Ivushkin, Sylvain
Jaskowiak, Philippe Jourde, Jarmila Kačírková, James
Kennerley, Eero Kiuru, Terje Kolaas, Klaus Konrads, Niels
Krabbe, Ludwik Kuć, Volodymyr Kucherenko, Ivanna L,
Guido La Rovere, Lars Lachmann, Frank Lambert,
Alessandro Landi, Albert Lastukhin, Tero Linjama, Mikael
Litsgård, Grzegorz Lorek, Donato Lorubio, Per Lundkvist,
Hans Lütgen, Vladimir Maiski, Miroslav Mareš, Dimitri
Marone, Michele Martelli, Jarek Matusiak, Bas van de
Meulengraaf, Cedric Mroczko, Rudolf Eberhard Neuber,
Vladimir Nikolenko, Stein Ø Nilsen, Rene Odeide,
Ireneusz Oleksik, Petr Omel chenko, Herman van Oos-
ten, Jos van Oost veen, Fede rica Pecchi, Sergey Petrov,
Gérard Picotin, Bram Piot, Jarmo Pirhonen, Alberto
Plata, Tommaso Renzulli, Stephan Risch, Magnus Robb,
Julien Rochefort, Johan Roeland, Leo Rouge, Joachim
Rupik, Vladimir S, Manolo Salas, Nikolay Sariev, Andrii
Simon, José Carlos Sires, Sergey Slyusarev, Marcin Soło-
wiej, Francesco Sottile, Giuseppe Speranza, Gui seppe
Speranza, Thomas Stal ling, John Storm, Karl-Birger
Strann, Jerzy Strzelecki, Piotr Szczypinski, Paolo Taranto,
Günter Tembrock, Gregorio Chaguaceda Tomás, José
Luis Valls, Juan Verhelst, Bram Vogels, Otto von Lexmond,
Tomasz Wałachowski, Dieter Wall schläger, D Wilczyns-
ka, Au brey John Wil liams, Peter Zeller, Adnan Zimić as
well as the facebook, youtube and vimeo users whose
real names we do not know: Abramis, Aladdin,
Andrea70Video, Ardisia1234, Birding Dnipro, Botschgo,
Brickegickel, cotuabellar, duboka mis ao, iSpiny,
Kozjanski park, Matteo, Maudoc, Moments in Latvia,
Msokolovru, Naturaleza Cantábrica, nedo_s_nature, pic
bytom, Ratel, Robin Hood, Roby, Sonnenburg, T. Rex,
ValGotraBaganza.it, Wardstone99, Красота и гармония
природы, Мир вокруг нас, Радуйся каждому дню!
Such analyses would not be possible whithout so many
recordists making their recordings available, and with
www.xeno-canto.org, www.tierstimmenarchiv.de, the
Macaulay Library at the Cornell Lab of Ornithology and
several ornitho platforms offering extremely valuable
platforms to do so. A list of all recordings used for the
article can be found at www.dutchbirding.nl/redstart_
calls.
Furthermore, we are grateful to photographers for
sharing their pictures, to Aurélien Martinez for help with
Russian speaking recordists and to the following persons
for valuable discussions on the topic: Raffael Ayé, Hans-
Heiner Bergmann, Andrea Corso, Egidio Fulco, Severin
Hauenstein, Dimitri Marrone, Magnus Robb, Fabian
Schneider and Manuel Schweizer. We thank Frederic
Jiguet from the French rarities committee for sharing in-
formation on samamisicus claims from France.
Samenvatting
GeoGrafische variatie in roep van GekraaGde rood-
staart Gekraagde Roodstaart Phoenicurus phoenicurus
heeft twee ondersoorten waarvan de nominaat P p
phoenicurus (hierna phoenicurus) als broedvogel voor-
komt in het grootste deel van het verspreidings gebied,
namelijk van Marokko, Spanje en Brittannië in het wes-
ten, tot het Baikalmeer, Rusland, in het oosten. De on-
dersoort P p samamisicus (hierna samamisicus) heeft een
meer zuidoostelijke verspreiding. Adulte mannetjes van
samamisicus zijn van phoenicurus te onderscheiden
door de aanwezigheid van een witte vleugelvlek. Ook
eerste-winter mannetjes zijn doorgaans goed te onder-
scheiden, maar een betrouwbare determinatie van
vrouwtjes is niet mogelijk. Van het verschil in contact-
roep tussen beide ondersoorten werd aangenomen dat
het een betrouwbaar kenmerk was. De contactroep van
phoenicurus kan omschreven worden als een stijgend
hoe-wied, terwijl de contact roep van samamisicus klinkt
als een op gelijke toonhoogte blijvend hied. In dit artikel
worden opnames van de contactroep uit vrijwel het ge-
hele verspreidings gebied van Gekraagde Rood staart ge-
analyseerd om meer inzicht te krijgen in de variatie en
verspreiding van de contactroep. In totaal werden hier-
voor opnames gebruikt van 287 roepende individuen uit
de broedperiode (mei-juli).
Naast de voor phoenicurus karakteristieke stijgende
hoe-wied-roep, werden ook meerdere varianten van de
hied-roep vastgesteld en een nog niet eerder beschreven
viest-roep. In het noordwestelijke deel van het versprei-
dingsgebied van phoenicurus werd bijna alleen de ken-
merkende hoe-wied-roep geregistreerd en in het ver-
spreidingsgebied van samamisicus alleen de hied-roep.
Bin nen het verspreidingsgebied van phoenicurus wer-
den op de Balkan zowel de klassieke hoe-wied-roep als
meerdere varianten van de hied-roep vastgesteld en ook
in Italië werd de hied-roep veel genoteerd en was veel
variatie in die roep aanwezig. De niet eerder geregis-
treerde viest-roep werd vastgesteld op het Iberisch
Schier eiland. Daarmee is de variatie in contactroep van
Gekraagde Roodstaart groter dan gedacht, met name in
het verspreidingsgebied van phoenicurus. De veelvuldig
vastgestelde hied-roep in Italië kan afkomstig zijn van
genetische introgressie met populaties van samamisicus.
Daarmee is dus een hied roepende Gekraag de Rood-
staart niet per denitie een kandidaat samamisicus als
dit niet in de eerste plaats wordt ondersteund door
veren kleedkenmerken.
References
Ayé, R, Martinez, N & Stalling, T 2014. The vocalisations
of ‘Ehrenberg’s Redstart’. Br Birds 107: 26-36.
Ayé, R, Schweizer, M & Roth, T 2012. Birds of Central
Asia – Kazakhstan, Turkmenistan, Uzbekistan, Kyrgyz-
Geographical variation in Common Redstart calls
171
stan, Tajikistan and Afghanistan. London.
Bergmann, H-H, Helb, H W & Baumann, S 2008. Die
Stimmen der Vögel Europas. Wiebelsheim.
Bioacoustics Research Program 2014. Raven Pro: Inter-
active sound analysis software (version 1.5). Ithaca.
Clement, P & Rose, C 2015. Robins and chats. London.
Ertan, K E 2002. Evolutionary biology of the genus Phoe-
nicurus. Phylogeography, natural hybridization and
population dynamics. PhD thesis, University Tübin-
gen. Marburg.
Field, A 2005. Discovering statistics using SPSS. Thou-
sand Oaks, California.
Glutz von Blotzheim, U N & Bauer, K M (editors) 1988.
Handbuch der Vögel Mitteleuropas 11/II. Wiesbaden.
Hogner, S, Laskemoen, T, Lifjeld, J T, Porkert, J, Kleven,
O, Albayrak, T, Kabasakal B & Johnson, A 2012. Deep
sympatric mitochondrial divergence without repro-
ductive isolation in the common redstart Phoenicurus
phoenicurus. Ecol Evol 2: 2974-2988.
Kirwan, G M, Boyla, K A, Castell, P, Demirci, B, Özen,
M, Welch, H & Marlow, T 2008. The birds of Turkey.
The distribution, taxonomy and breeding of Turkish
birds. London.
Martinez, N 2010. The Crimean Peninsula: a zone of
integradation of Common Redstart subspecies? Br
Birds 103: 405-406.
Matvejev, S D & Vasi, V F 1973. Catalogus Faunae Yugo-
slaviae. Ljubljana.
Quinn, G P & Keough, M J 2002. Experimental designs
and data analysis for biologists. Cambridge.
R Core Team 2016. R: A language and environment for
statistical computing. Vienna. www.R-project.org.
Roselaar, C S 1995. Songbirds of Turkey – an atlas of
biodiversity of Turkish passerine birds. Haarlem.
Ruggieri, L & Sighele, M 2008. Annuario 2007. Verona.
Shirihai, H & Svensson, L 2018. Handbook of Western
Palearctic birds 1 – Passerines: larks to Phylloscopus
warblers. London.
Small, B J 2009. From the Rarities Committee’s les. The
identication of male ‘Ehrenberg’s Redstart’, with
comments on British claims. Br Birds 102: 84-97.
Svensson, L 1992. Identication guide to European pas-
serines. Fourth edition. Stockholm.
Svensson, L, Grant, P J, Mullarney, K & Zetterström, D
2015. Collins bird guide. Second edition (revised).
London.
Wassink, A & Ebels, E B 2005. DB Actueel: Waarschijn-
lijke Oosterse Gekraagde Roodstaart op Texel. Dutch
Birding 27: 366-367.
Geographical variation in Common Redstart calls
Nicolas Martinez, Hintermann & Weber AG, Austrasse 2a, 4153 Reinach, Switzerland
(nicolas.martinez44@yahoo.de)
Ralph Martin, University of Freiburg, Tennenbacher Straße 4, 79106 Freiburg, Germany
(ralph.martin@wildlife.uni-freiburg.de)
APPENDIX 1 Results of linear models of huid calls, including sex as covariate, as described under Methods.
Signicant p-values indicated with asterisk (p < 0.05 = *, p < 0.01 = ** and p < 0.001 = ***).
Response estimate standard error t-value p-value
gradient intercept -2.016 3.675 -0.549 0.585
longitude 0.427 0.182 2.353 0.021*
latitude 0.204 0.071 2.863 0.005**
sexm -0.405 0.416 -0.975 0.332
interaction latitude:longitude -0.008 0.003 -2.203 0.030*
1st frequency intercept 3.190 0.321 9.919 0.000
longitude 0.040 0.016 2.515 0.014
latitude -0.006 0.006 -0.962 0.338
sexm 0.018 0.036 0.497 0.620
interaction latitude:longitude -0.001 0.001 -2.475 0.015*
2nd frequency intercept 2.676 0.370 7.228 0.001***
longitude 0.075 0.018 4.129 0.001***
latitude 0.029 0.007 4.007 0.001***
sexm 0.028 0.042 0.665 0.508
interaction latitude:longitude -0.001 0.001 -4.148 0.001***
call length intercept 108.635 25.716 4.224 0.001***
(interaction longitude -0.211 0.141 -1.498 0.138
longitude:latitude latitude 1.300 0.493 2.634 0.010**
not signicant and sexm 5.678 5.998 0.947 0.346
therefore excluded)
172
APPENDIX 2 Results of linear models of huid calls, excluding sex as covariate, as described under Methods.
Signicant p-values indicated with asterisk (p < 0.05 = *, p < 0.01 = ** and p < 0.001 = ***).
Response estimate standard error t-value p-value
gradient intercept 4.716 1.381 3.416 0.001***
(interaction longitude 0.023 0.006 3.856 0.001***
longitude:latitude latitude 0.073 0.027 2.698 0.001**
not signicant and
therefore excluded)
1st frequency intercept 3.442 0.166 20.754 0.001***
interaction latitude:longitude -0.001 0.000 -6.051 0.001***
2nd frequency intercept 3.254 0.223 14.571 0.001***
interaction latitude:longitude -0.001 0.000 -5.142 0.001***
call length intercept 101.202 17.926 5.646 0.001***
(interaction longitude -0.194 0.079 -2.473 0.014*
longitude:latitude latitude 1.473 0.348 4.231 0.001***
not signicant and
therefore excluded)
APPENDIX 3 Gradient, 1st frequency, 2nd frequency and call length of huid calls in dependency of longitude and
latitude. Covariate ‘sex’ not included in gure as it was not signicant.
Geographical variation in Common Redstart calls
173
Geographical variation in Common Redstart calls
APPENDIX 4 Results of linear models of heed calls, including sex as covariate, as described under Methods.
Signicant p-values indicated with asterisk (p < 0.05 = *, p < 0.01 = ** and p < 0.001 = ***).
Response estimate standard error t-value p-value
gradient intercept 6.164 6.467 0.953 0.353
(interaction longitude -0.009 0.022 -0.408 0.688
longitude:latitude latitude -0.059 0.156 -0.381 0.708
not signicant and sexm -1.521 0.950 -1.601 0.126
therefore excluded)
1st frequency intercept 1.092 1.337 0.817 0.424
sexm 0.269 0.131 2.055 0.055
interaction latitude:longitude -0.001 0.001 -2.447 0.025*
2nd frequency intercept 4.430 0.549 8.059 0.001***
(interaction longitude 0.001 0.002 0.175 0.863
longitude:latitude latitude -0.004 0.013 -0.298 0.769
not signicant and sexm 0.004 0.081 0.049 0.961
therefore excluded)
call length intercept 117.834 87.780 1.342 0.197
(interaction longitude -0.464 0.302 -1.534 0.144
longitude:latitude latitude 1.892 2.136 0.886 0.388
not signicant and sexm -12.895 12.605 -1.023 0.321
therefore excluded)
APPENDIX 5 Results of linear models of heed calls, excluding sex as covariate, as described under Methods.
Signicant p-values indicated with asterisk (p < 0.05 = *, p < 0.01 = ** and p < 0.001 = ***).
Response estimate standard error t-value p-value
gradient intercept -1.718 2.196 -0.782 0.437
(interaction longitude -0.044 0.009 -4.744 0.001***
longitude:latitude latitude 0.124 0.051 2.447 0.017*
not signicant and
therefore excluded)
1st frequency intercept 5.429 0.383 14.160 0.001***
(interaction longitude 0.010 0.002 6.252 0.001***
longitude:latitude latitude -0.041 0.009 -4.627 0.001***
not signicant and
therefore excluded)
2nd frequency intercept 5.348 0.252 21.212 0.001***
(interaction longitude 0.004 0.001 3.589 0.001***
longitude:latitude latitude -0.027 0.006 -4.682 0.001***
not signicant and
therefore excluded)
call length intercept 206.322 26.121 7.899 0.001***
(interaction longitude 0.028 0.109 0.254 0.801
longitude:latitude latitude -0.930 0.605 -1.536 0.130
not signicant and
therefore excluded)
174
APPENDIX 6 Gradient, 1st frequency, 2nd frequency and call length of heed calls in dependency of longitude and
latitude. Covariate ‘sex’ not included in gure as it was not signicant.
Geographical variation in Common Redstart calls
