Acrocephalus 2.J (128-129): 1-2, 2006 Nekatere spremembe v rubrikah Some changes in the journal’s contents Obsežni {tevilki, ki jo imate pred seboj, bi rad pristavil zgolj nekoliko suhoparen uvodnik. Po dolgih posvetih in pripravah sem se odlo~il izpolniti napoved s konca prej{njega letnika: uvedbo nekaterih novih rubrik. Uvodnik je namenjen njihovi kratki predstavitvi. Pregledni ~lanki so pomembna skupina ~lankov, namenjenih povzemanju klju~nih ugotovitev na dolo~enem podro~ju ornitologije. Mi{ljena so predvsem nova, neobdelana in hitro razvijajo~a se podro~ja. Citirane naj bi bile vse klju~ne reference. ^lanki lahko vsebujejo tudi dodatne izvirne podatke, obravnavali pa jih bomo enako kot druge znanstvene ~lanke, kar vklju~uje dvojno recenzijo (poleg uredni{ke). Stali{~e je poseben tip ~lanka, ki je namenjen novim, predvsem zanimivim, provokativnim in v~asih kontroverznim idejam. Avtorji naj se pred oddajo rokopisov zato posvetujejo z urednikom o primernosti teme. Tak{ni prispevki lahko vklju~ujejo tudi vnovi~no predelavo starih podatkov, a na druga~en na~in. Ker gre za hipoteze, bodo stali{~a opremljena z eno neobvezno recenzijo, druga recenzija pa bo uredni{ka. Kratki prispevki bodo sodili v najbolj obsežno novo rubriko. Gre za serije podatkov, favnisti~ne popise, zanimive preliminarne podatke ter posamezne podatke velike vrednosti, pri katerih avtorji napi{ejo dodaten komentar in vsaj deloma svoje podatke primerjajo z objavljeno literaturo. V pri~ujo~i {tevilki vam predstavljam prvih osem kratkih prispevkov, nekatere smo predelali kar neposredno iz odli~no napisanih notic. Uredni{ka obdelava bo vklju~evala eno recenzijo. Znanstveni ~lanki ne bodo ve~ deljeni na dolge in kratke, saj je mnogokrat tak{na delitev umetna. Vse rubrike pa bodo opcijske, glede na kakovost in koli~ino prejetega materiala. Do naslednje {tevilke bomo tudi ustrezno prenovili navodila za avtorje. Vsekakor pa upam, da boste uživali pri branju raznolikih in zanimivih ~lankov. Primož Kmecl I Uvodnik / Editorial I would like to add just a short and laconic editorial to the issue you are now holding in your hands. After long consultations, I decided to fulfil my promise made at the end of the last volume: to introduce new sections in our journal. The aim of this editorial is their short presentation. Review papers constitute an important group of articles that summarize some of the key findings in a certain field of ornithology, particularly in new, fast developing fields not presented as yet. All key references should be cited. These papers can include original data and will be treated as normal research papers which includes editorial and double peer review. Point-of-view section contains new, interesting, provocative and sometimes controversial hypotheses. The authors should consult the editor prior to the manuscript submission, whether a certain topic is appropriate for publication. Such papers can also include new data manipulation, although in a different way. One non-binding peer review and an editorial revision will be provided for. Short communications will be part of the most extensive new section of the journal. Series of data, faunistic studies, interesting preliminary data and high value single data will be included in it. Commentaries and references are required. This issue presents eight short communications, some of which have been rewritten from high quality short notes submitted to the journal. One peer review will be provided. Research papers will no longer be divided into long and short papers, considering that this separation is sometimes rather artificial. All sections will be optional depending on the material. We will also amend the instructions to authors in the next issue. In any case I do hope that you will enjoy yourselves while reading highly interesting and diverse papers. Primož Kmecl 2 Acrocephalus 2.J (128-129): 3-12, 2006 Field observations of the Sichuan Wood Owl Strix uralensis davidi in western China Terenska opazovanja se~uanske koza~e Strix uralensis davidi v zahodni Kitajski Wolfgang Scherzinger1 & Yun Fang2 1 Bavarian Forest National Park, Guntherstraße 8, D-94568 St. Oswald, Germany, e-mail: drscherzinger@gmx.de 2 Zoological Institute, Chinese Academy of Science, 19 Zhongguancun, Beijing 100080, PR of China, e-mail: fangyun@vip.sina.com In 1995, 1997, and 1999 at least one pair of the rare Sichuan Wood Owl Strix uralensis davidi has been observed in the nature reserve “Lianhuashan”, Gansu Province (Kangle County / Central China). During 13 encounters, one photo, the territorial song, and a rising scale (“nest-site demonstration”) of the male and “begging” notes of the female were recorded. Supported by a conservation programme sponsored by British Petrol (BP), several nest boxes were mounted in suitable habitat in 2002 - 2003, which enabled close observation of breeding success of this rare species in 2005 for the very first time. In 2006 we obtained recordings of 3 territorial males, 2 females and 2 freshly hatched young ones. Description of the species is based on field observations and compared to characteristics of the European subspecies of Ural Owl (Strix u. liturata and Strix u. macroura). Due to the small remaining patches of mountain forests, the habitat of this owl is extremely threatened by fragmentation. With its 47 km2 of natural woodland, the reserve forms a distributional relict of great importance, which will hopefully allow the survival of this nearly unknown owl. Key words: Sichuan Wood Owl, Strix uralensis davidi, montane forest, fragmentation of habitat, preservation, central China, eastern Tibet Klju~ne besede: se~uanska koza~a, Strix uralensis davidi, gorski gozd, fragmentacija habitata, varstvo narave, osrednja Kitajska, vzhodni Tibet 1. Introduction The Sichuan Wood Owl (in Chinese language = “Sichuan liu xiao”, MacKinnon & Phillipps 2000) was first described by Sharpe (1875) as Syrnium davidi, named after the French missionary Father Armand David (like a large number of plants, birds and mammals in China) who discovered this large owl during his first expedition to China in 1866. The majority of descriptions of the species in the literature are traceable to single observations and collections from the late 19th and early 20th centuries. Because of the obvious resemblance to the Ural Owl of Eurasia, initial taxonomic estimations classified the Sichuan Wood Owl as a subspecies of Strix uralensis (Stresemann 1923) or even a “local phase” (Grossman & Hamlet 1964). Although since then only a few incidental observations and a few specimens in museums have become known, nowadays lists emphasize the status of this rare owl as a separate species, mostly based on its absolute isolation for millions of years (König et al. 1999, Clements 2003). As information about distribution, ecology, reproduction, voice, and behaviour of the Sichuan Wood Owl are completely lacking, a final decision requires new field observations and molecular analysis. As a first approach, a comparison of the territorial songs of Sichuan Wood Owl (recorded in Lianhuashan nature reserve) and Ural Owl (tape recordings from Bavarian Forest National Park) showed a clear similarity. Since, in field experiments, the Sichuan Wood Owl fully recognized the play back of Ural Owl’s song, implying conspecificity, this was a crucial fact for my suggestion to rank the Sichuan Wood Owl 3 W. Scherzinger & Y. Fang: Field observations of the Sichuan Wood Owl Strix uralensis davidi in western China as a subspecies of Ural Owl (Scherzinger 2005). Given the sparse records of the Sichuan Wood Owl, maps of its distribution give a rather rough picture – primeval forests in the high mountains of Sichuan, Sikiang and Qinghai. This owl species is not listed in the extensive catalogue of vertebrate fauna of Gansu (Liu 1995), as the owl’s occurrence in this province was only detected on our first visit in 1995 (Sun et al. 2001, Scherzinger 2005). Clearly, the remote area of distribution is limited to coniferous forests at high elevations in the mountains of western China and eastern Tibet, therefore interpreted as a true relict of the last ice ages and the pre-glacial distribution of woodland (Voous 1962; Figure 1). In this paper we not only present recent observations and the first record of successful breeding of this rare owl, but also point to the new opportunities for field work in the formerly closed areas of Chinese mountains. Figure 1: Whereas the range of the Eurasian Ural Owl Strix uralensis extends over the entire palaearctic forest belt, the Sichuan Wood Owl Strix uralensis davidi occurs only in a strictly isolated area of montane forests in western China and eastern Tibet (map of distribution, from Voous 1962) Slika 1: Areal koza~e Strix uralensis se razteza ~ez celoten gozdni pas Palearktika, se~uanska koza~a Strix uralensis davidi pa se pojavlja le na popolnoma izoliranem obmo~ju gorskih gozdov v zahodni Kitajski in vzhodnem Tibetu (povzeto po Voous 1962) 2. Study area and methods In the years 1995, 1997 and 1999 we succeeded in making field observations, photographic and tape recordings of at least one breeding pair of the Sichuan Wood Owl, during stays from April-May, May-June, and June-July in the “Lianhuashan Nature Reserve”. Though preliminary, these observations can help to improve the description and to document aspects of the behaviour of this rare forest-dwelling owl. Stimulated by the offer of nearly 40 large nest boxes, one successful brood was observed by Y. Fang in 2005. This event was monitored by automatic video-camera in the nest box during the full nestling time. A 5-week stay in 2006, from the end of May to the beginning of July, resulted in records of 3 territorial males, 2 females, and two freshly fledged young ones. We have now obtained a long series of photos by high-speed digital camera. 2.1. Study Area “Lianhuashan Nature Reserve” (coordinates 34°56´-58´N / 103°44´- 48´E) was founded at the eastern edge of the mountainous Qinghai-Tibetan Plateau in Gansu Province (western China), south of the provincial capital city Lanzhou. Covering an area of approximately 120 km2, the reserve includes coniferous forest (dominated by fir and spruce trees; mainly Abies fargesii and Picea asperata) on shaded slopes, with dry slopes on the southern exposition as well as treeless alpine meadows. The portion of non-fragmented woodland is approximately 47 km2, the tree line being more than 3000 m a.s.l. (Klaus et al. 1996). The reserve is named after a conspicuous rock summit, nearly 3500 m in height, which is visited by several thousand Buddhist pilgrims every year. Due to the sacredness of this precious landscape, large coniferous forests have persisted in natural stands. But due to heavy logging in the 1970’s, stands of old growth forest only remain on extremely steep slopes, on exposed rocky ridges, and in inaccessible canyons (Figure 2; Sun et al. 2007, in press). Figure 2: Natural forest near the tree line at 3000 m a.s.l. in the “Lianhuashan Nature Reserve”, Gansu Province, China; The island character of the forest relict in the reserve is intensified due to extensive clear cuts in surrounding landscapes and overgrazing of vegetation (photo: W. Scherzinger) Slika 2: Naravni gozd blizu drevesne meje na 3000 m n.v. v “Lianhuashan Nature Reserve”, provinca Gansu, Kitajska; Oto{ki zna~aj reliktnega gozda je poudarjen zaradi intenzivnih golosekov v okolici in intenzivne pa{e (foto: W. Scherzinger) 4 Acrocephalus 2.J (128-129): 3-12, 2006 All the field records of Sichuan Wood Owl were collected in the main valley of “Lianhuashan Nature Reserve”, which stretches from a saddle at the base of the “Three Sisters Mountain” in the west (approximately 2900 m a.s.l., with dense coniferous old growth forest on a steep slope) to a swampy area along a small creek to the east, with dense coniferous stands. 2.2. Field observations As the main purpose of the stay in the nature reserve was field research on endemic woodland grouse (Chinese Grouse Bonasa sewerzowi; see Klaus et al. 1996), we detected the occurrence of Sichuan Wood Owl fortuitously. Whenever possible, we took the opportunity to collect at least rough information about habitat use and distribution, and about behaviour and vocalization of this very special bird. Oral imitation of territorial song and the nest-site-demonstration note (both well known notes from Strix uralensis liturata / macroura, the European subspecies) were used to stimulate vocalization of the free ranging owls. Due to heavy timber harvesting in the 1970’s, before Lianhuashan became a strict reserve, even natural forest stands lack the very old trees that could offer large raptor’s nests or hollow trunks as suitable nesting sites for the big owls. Therefore the birds could only use rocky precipices for nesting. In view of previous experience of preservation management for the Ural Owl in the Bavarian Forest National Park Figure 3: Due to the lack of natural tree cavities, simple nest boxes could stimulate Sichuan Wood Owls Strix uralensis davidi to breed in an artificial nesting site (photo: Y. Fang) Slika 3: Zaradi pomanjkanja naravnih dupel, lahko gnezdilnice spodbudijo se~uansko koza~o Strix uralensis davidi h gnezdenju (foto: Y. Fang) (Scherzinger 1996 & 2006), we decided to mount big nest boxes for the Sichuan Wood Owl. This should not only optimise opportunities for suitable breeding sites, but also increase the chance, for the first time, to observe these owls when breeding and rearing. With the helpful support of a conservation programme of British Petrol Company (BP), about 20 specific nest boxes were mounted in 2002 - 2003 (Figure 3). 3. Results 3.1. Description of the Sichuan Wood Owl Successful field observations were made mostly in the evening twilight, but on 13 Apr 1995, we discovered a male owl during daylight. Under field conditions in the dim coniferous stand, the plumage looked pale greyish-brown, with distinct longitudinal streaks on the breast and under parts; nape, mantle and backside were darker than the front, the white dots on the scapular feathers were prominent; the roundly framed facial disc was distinct, with a dull-yellow beak and narrow almond-shaped eyes (colour appendix - Figure 1). These characteristics are clearly shown in the close-up photo of the female breeding in one of the nest boxes in 2005 (colour appendix - Figure 2). The unicoloured central tail feathers are also specific. 3.2. Field observations On calm evenings the owl’s hooting could be heard over the whole of the main valley of Lianhuashan reserve, stretching in an east-west direction for about 1,5 km. Based on the observation points, the home range of the pair of owls measures a minimum of 200 ha. During our field-stays in 1995, 1997 and 1999 we obtained a total of 6 visual records of male and 4 of female; photo documents of a male were obtained in 1995 and 2006 (colour appendix - Figure 1). 14 acoustical records of the male were registered (territorial song and initial strophes of this note, nestsite-demonstration, and aggressive calls; Figure 4), and 3 of the female (begging notes). We regularly heard duets from the couple, with territorial song and “nestsite-demonstration” by the male and begging notes by the female. Twice we observed the male delivering prey to its female (Table 3; see appendix). In addition, Jia Chen-xi, a colleague of the Chinese research team, took a close-up photo of a male in summer 2003, and Fang Yun attracted a male by playback of tape recordings in May 2004. In 2005, one nest box was occupied by a pair of Sichuan Wood Owls. Egg laying started between April 5 W. Scherzinger & Y. Fang: Field observations of the Sichuan Wood Owl Strix uralensis davidi in western China Figure 4: Sonographic comparison of territorial song of male Ural Owl Strix uralensis (left; record from captive birds in Bavarian Forest National Park) and Sichuan Wood Owl Strix uralensis davidi (right; record from Lianhuashan nature reserve) Slika 4: Sonografska primerjava obmo~nega petja samca koza~e Strix uralensis (levo; posnetek osebkov v ujetni{tvu iz nacionalnega parka Bavarski gozd) in se~uanske koza~e Strix uralensis davidi (desno; posnetek iz naravnega rezervata Lianhuashan) 25 - 27; the two fertile eggs measured 41.2 x 49.1 mm and 40.8 x 50.1 mm. Brooding started with the second egg at least, and the eggs hatched about 30 May. Nestlings remained in the nest box for 33 - 35 days. One sound fledgling left the nesting site between 3 and 5 Jul (colour appendix - Figure 1). This is the Figure 5: Due to its size and diurnal activity the Gansu Pica Ochotona cansus is a favoured prey of the big wood dwelling owls (photo: W. Scherzinger) Slika 5: Zaradi primerne velikosti in dnevne aktivnosti je žvižga~ vrste Ochotona cansus priljubljen plen velikih v gozdu žive~ih sov (foto: W. Scherzinger) first breeding record of this rare species under human observation. Favoured prey items include the Gansu Picka Ochotona cansus (Figure 5), but systematic records on hunting are lacking. During an additional stay in 2006 we recorded 3 singing males and 2 females, and also detected 2 freshly fledged young ones in front of a nearly vertical rock-precipice, where they may have hatched in a cavity. Roosting at top of a spruce tree during daylight one of the fledglings got attacked by a Carrion Crow Corvus corone corone, while the adult male displayed conspicuously by singing and flying around (Figures 6 & 7). Among the birds in the reserve we noted 3 further species of owls, the Little Owl Athene noctua impasta near the village, the Eagle Owl Bubo bubo tibetanus, and the Tengmalm’s Owl in its local subspecies Aegolius funereus beickianus. With the support of suitable nest boxes, a minimum of 5 - 7 successfully breeding pairs of the latter species have been confirmed to date. Further play-back experiments with the songs of Collared Pygmy Owl Glaucidium brodiei and Tawny Owl Strix aluco nivicola, which could also live in this forest area, were not successful. 6 Acrocephalus 2.J (128-129): 3-12, 2006 Figure 6: As soon as the Carrion Crow Corvus corone corone detected the helpless young at treetop, it approached in low-level attacks (photo: Y. Fang) Slika 6: Takoj, ko je ~rna vrana Corvus corone corone zaznala nemo~ne mladi~e se~uanske koza~e Strix uralensis davidi v drevesnem vrhu, je za~ela s previdnimi napadi (foto: Y. Fang) Figure 7: In striking contrast to the Ural Owl Strix uralensis the male Sichuan Wood Owl Strix uralensis davidi did not attack intruders, but showed a conspicuous distraction display, singing exposed and flying hyper-actively nearby (photo: Y. Fang) Slika 7: Kot nasprotje koza~i Strix uralensis, samec se~uanske koza~e Strix uralensis davidi ne napade vsiljivcev ampak za~ne z odvra~alnim razkazovanjem, pri ~emer poje popolnoma odkrit ter hiperaktivno leti v bližini (foto: Y. Fang) 4. Discussion 4.1. Comparison of Sichuan Wood Owl and Ural Owl With respect to body size and specific characters, Sichuan Wood Owl strongly resembles the Eurasian Ural Owl, which also shows coarse longitudinal stripes on the breast and under parts, and almond-shaped, blackish-brown eyes in a distinct facial disc. However, the Chinese species / subspecies has a much darker appearance, caused by the blackish-brown plumage on its scapulars. The central tail feathers are unicoloured dark brownish-black and without streaks (colour appendix - Figure 2; also photos of a mounted specimen, in Scherzinger 2005). Recorded sizes are: body length 540 – 580 mm, wing length 371 – 372 mm, and tail length 266 – 290 mm. Eck & Busse (1973) measured an index of tail to wing length 71.5 – 77.8 (compared to 70.1 in the smallest subspecies of Ural Owl Strix u. hondoensis from Japan, and 83.6 in the largest subspecies Strix u. macroura from the Carpathian Mountains; Table 1). There is no information about the body mass of Sichuan Wood Owl. At first impressions the characteristics of the male’s territorial song appear to be quite similar to those of the Ural Owl. However, the whole strophe is much shorter (0.42 – 1.41 s in Sichuan Wood Owl; 2.24 s in Ural Owl), and the frequency is significantly lower (basal amplitude maximal 356 – 420 Hz in Sichuan Wood Owl; 420 Hz in Ural Owl; Scherzinger 1980 & 2005). The difference is not so clear in the “nestsite-demonstration”. But clearly the owls themselves did not recognise this difference, when stimulated by imitations of Ural Owl’s call notes, and displayed their full territorial performance. As species specific patterns of vocalization usually function as the strongest characteristics for reproductive isolation (König 1994), this result strongly suggests that separation of these both types of owls has not yet reached species level. We therefore favour the original classification of Sichuan Wood Owl as a subspecies of Eurasian Ural Owl, until clarified by molecular analysis (Stresemann 1923, Voous 1962, Burton 1973, Wolters 1975, Glutz & Bauer 1980, Scherzinger 2005; Table 2). Although one successful brood was raised in one of the nest boxes in the reserve, our knowledge about reproductive biology is still only preliminary. Duration of brooding and nestling time corresponds with the breeding phenology of European subspecies of Ural Owl (Glutz & Bauer 1980). Nevertheless, further observations are necessary, so monitoring of vocalization, behaviour, ontogenetic development, breeding biology, and nutrition was carried out in 2006. 4.2. Survival of Sichuan Wood Owl – a function of forest management The Sichuan Wood Owl has always been classified as extremely rare (Weigold, in Stresemann 1923, Holt et al. 1999). The isolated distribution, as well as the apparent rarity, of the Sichuan Wood Owl may suffice 7 W. Scherzinger & Y. Fang: Field observations of the Sichuan Wood Owl Strix uralensis davidi in western China Table 1: Weights and measurements of different subspecies of Strix uralensis (after Momiyama 1928, Mo{ansky 1958, Eck & Busse 1973, Glutz & Bauer 1980, Mikkola 1983, Pietiäinen 1988, König et al. 1999); Sichuan Wood Owl Strix uralensis davidi belongs to the medium sized subspecies of east Asia Tabela 1: Teža in izmere podvrst Strix uralensis (po Momiyama 1928, Mo{ansky 1958, Eck & Busse 1973, Glutz & Bauer 1980, Mikkola 1983, Pietiäinen 1988, König et al. 1999); Se~uanska koza~a spada med srednje velike podvrste Strix uralensis vzhodne Azije Subspecies Weight [max]/ Tot. length/ Wing/ Tail/ Teža [maks] (g) Dolžina (mm) Krilo (mm) Rep (mm) Strix u. fuscescens Strix u.hondoensis Strix u. japonica Strix u. nikolskii (daurica) Strix u. yenissensis Strix u. davidi Strix u. uralensis 310 - 332 Index of tail-to-wing length/ Indeks dolžine rep - krilo Range Mean N 232 ? 657 - 950 Strix u. liturata 720 - 871 [1320] Strix u.macroura (carpathica) 503 - 950 [1307] 295 - 347 223 - 244 259 - 313 201 - 235 310 - 355 328 - 370 580 - 590 371 - 372 266 - 290 500 - 620 340 - 380 338 - 396 354 - 415 270 - 315 70.1 - 74.4 72.88 4 71.4 - 78,8 75.43 4 (71.5) - 77.8 74.65 2 77.1 77.10 1 77.4 - 78.2 77.87 3 80.6 - 83.6 82.10 3 to list this population in the category “vulnerable” (Collar et al. 1994), but more fieldwork is required to clarify this status. In China this owl is listed in category II of “nationally threatened” species (Zheng & Wang 1998). The availability of suitable habitats is strictly limited because of natural fragmentation of old growth coniferous forests, caused by the diverse morphology of the high mountains. In addition, human impact in the 1970’s decreased the owl’s chance of survival by extensive cutting and grazing in the natural mountain forests in large parts of the Gansu and Sichuan provinces, especially in the eastern Qinghai-Tibetan-Plateau (N. Liu, pers. comm.). Nowadays there is hope because, in the new politics in China, great effort is being made to maintain the woodland areas and encourage reforestation. Recently China’s Ministry for land use and environment decreed a logging ban for nearly all the forests in mountainous areas, in some areas even for bamboo cutting. Furthermore, several reserves have been designated, some with a status comparable to that of national parks. The first results of mapping the actual distribution of forest stands in the “Lianhuashan Nature Reserve”, based on satellite data, can be used to plan reforestation, and to create stepping stones and corridors between the remnant forest patches (Klaus et al. 2001). In addition the new supply of suitable nest boxes will offer adequate nesting sites for the Sichuan Wood Owl, even in younger forest stands. We estimate a maximum density of 4 territories of Sichuan Wood Owl in Lianhuashan reserve (woodland area about 47 km2; = 0.85 pairs / 10 km2). Further potential habitats exist at a distance of 30 km, and in the extensive forest alongside the mountain ridge at the border between the provinces of Gansu und Sichuan (Y.-H. Sun, pers.comm.). The owl was also observed several times in the famous Jiuzhaigou Reserve (Hobcroft, pers.comm., C. Jia, pers.comm.). Unfortunately these data are not sufficient for even a rough estimate of total population size. Acknowledgements: The fieldstudiesin“Lianhuashan Nature Reserve” were supported by the German Association for Research (DFG, project KL 962/3-3) and the National Natural Science Foundation of China. The support of the “British Petrol conservation programme”, which enabled a suitable supply of nesting sites, was a key factor in observing the first successful brood. We express our special gratitude for hospitality at the administrative station of “Lianhuashan” reserve, and especially to Prof. Dr. Sun Y.-H. from the Zoological Institute of the Chinese Academy of Sciences, Beijing. His helpfulness made fieldwork possible in forest habitats up to elevations of about 3,000 m. We are grateful to Dr. H.-W. Helb, from the University of Kaiserslautern, for drawing the sonograms. Last, but not least, we thank Dr. A. Vrezec and Prof. Dr. H. Pietiäinen for their helpful review. 8 Acrocephalus 2.J (128-129): 3-12, 2006 Table 2: Suggestions in the literature about the taxonomic ranking of Sichuan Wood Owl as subspecies (Strix uralensis davidi) or distinct species (Strix davidi). Although there are no new findings, there has been a shift from “isolated race” to “valid species” in recent years, mainly in view of the very long isolation of the area of distribution. Tabela 2: Pregled taksonomske razvrstitve se~uanske koza~e kot podvrste (Strix uralensis davidi) ali vrste (Strix davidi). Kljub pomanjkanju novih podatkov je opazen premik od oznake “izolirana podvrsta” k “veljavni vrsti”, predvsem zaradi zelo dolge izoliranosti areala. Strix uralensis davidi Strix davidi Reference Insufficient knowledge/ Nezadostno poznana No splitting/ Ne lo~uje Subspecies/ Podvrsta Distinct species/ Razlo~na vrsta Holt et al. (1999) Dickinson (2003) more study needed more study needed may be ? Grossman & Hamlet (1964) Zheng & Wang (1998) Stresemann (1923) Voous (1962) Burton (1973) Eck & Busse (1973) Wolters (1975) Cheng (1987) Voous & Cameron (1988) Duncan (2003) Sibley & Monroe (2003) Sharpe (1875) Peterson (1999) König et al. (1999) MacKinnon & Phillipps (2000) Sun, Y.-H. (2000; pers. comm.) Howard & Moore (2003) Clements (2003) Nicolson (2004; pers. comm.) “Zoonomen”1 “ITIS”2 “Avibase”3 “Animal Diversity Web”4 “Global Owl Project”5 del Hoyo et al. (1999) dark phase isolated population heavily pigmented isolated totally isolated isolated W China X X well differentiated like Strix u. from Japan new species X absolutely isolated sometimes as race endemic endemic species valid subspecies probably Nr. 2698 valid species (uncertain) X valid spec, Nr. 555434 endemic species X X Strix davidi as superspecies of Strix uralensis Remarks / opombe: 1http://www.zoonomen.net/ 2http://www.itis.usda.gov/servlet/ 3http://www.bsc-eoc.org/avibase/ 4http://www.animaldiversity:ummz.umich.edu/site/accounts/classification.html 5http://www.globalowlproject.com 9 W. Scherzinger & Y. Fang: Field observations of the Sichuan Wood Owl Strix uralensis davidi in western China 5. Povzetek V letih 1995, 1997 in 1999 je bil opazovan v naravnem rezervatu Lianhuashan, v provinci Gansu, osrednja Kitajska, najmanj en par redke se~uanske koza~e Strix uralensis davidi. Med 13 sre~anji sta avtorja posnela fotografije in teritorialno petje ter ogla{anje. Namestila sta ve~ gnezdilnic v primernem habitatu, v letih 2002 - 2003, ki so omogo~ile bližnje opazovanje gnezditve v letu 2005, prvi~ za to podvrsto. V letu 2006 sta posnela ogla{anje 3 teritorialnih samcev, 2 samic in 2 pravkar speljanih mladi~ev. V diskusiji primerjata podvrsto z evropsko podvrsto koza~e (Strix u. liturata in Strix u. macroura). Habitat te sove je izjemno ogrožen zaradi fragmentacije. V rezervatu je 47 km2 ohranjenih gozdov, ki bodo morda omogo~ali preživetje te malo znane sove. 6. References Burton, J. (1973): Owls of the world. – Eurobook, Peter Lowe publ., Hertford, England. Clements, J. (2003): Birds of the world: a checklist, 5th edition. – Ibis Publishing Comp., Vista, California. Collar, N., Crosby, M. & Stattersfield, A. (1994): Birds to watch 2: The world list of threatened birds. – BirdLife International, Cambridge. Eck, S. & Busse, H. (1973): Eulen. Die rezenten und fossilen Formen – Aves, Strigidae. – Neue Brehm Bücherei 469. Glutz v. Blotzheim, U. & Bauer, K. (1980): Strix uralensis Pallas 1771 – Habichtskauz (Uralkauz). pp. 611–629 In: Handbuch der Vögel Mitteleuropas. – Aula, Wiesbaden. Grossman, M.-L. & Hamlet, J. (1964): Birds of prey of the world. – Bonanza Books, New York. Holt, D., Berkley, R., Deppe, C., Enriquez Rocha, P. , Olsen, P. , Petersen, J., Rangel Salazar, J., Segars, K. & Wood, K., (1999): Species accounts of Strigidae. pp. 153–242 In: Hoyo et al.: Handbook of the birds of the world. Vol. 5: Barn owls to Hummingbirds. – Lynx Edition, Barcelona. Klaus, S., Scherzinger, W. & Sun, Y.-H. (1996): Ökologie und Verhalten des Chinahaselhuhns Bonasa sewerzowi. – Ornithol. Beob. 93: 343–365. Klaus, S., Selsam, P. , Sun, Y.-H. & Fang, Y. (2001): Analyse von Satellitenbildern zum Schutz bedrohter Arten. Fallbeispiel Chinahaselhuhn (Bonasa sewerzowi). – Naturschutz u. Landschaftsplanung 33: 281–285. König, C. (1994): Lautäußerungen als interspezifische Isolationsmechanismen bei Eulen der Gattung Otus (Aves: Strigidae) aus dem südlichen Südamerika. – Stuttgarter Beitr. Naturkunde/Ser. A 511: 35. König, C., Weick, F. & Becking, J.-H. (1999): Owls – a guide to the owls of the world. – Pica Press, Sussex. Liu, N. (1995): Vertebrate fauna of Gansu. – Univ. Lanzhou, China. IC MacKinnon, J. & Phillipps, K. (2000): A field guide to the birds of China. – Oxford Univ. Press., Oxford. Mikkola, H. (1983): Owls of Europe. – Poyser, Calton. Mo{ansky, A. (1958): Beitrag zur Kenntnis der systematischen Stellung Karpatischer Habichtskäuze (Strix uralensis). – Sylvia 15: 55–66. Momiyama, T. (1928): New and known forms of the Ural Owl (Strix uralensis) from south eastern Siberia, Manchuria, Korea, Sakhalin and Japan. – Auk 45: 177–185. Pietiäinen, H. (1988): Breeding season quality, age and the effect of experience on the reproductive success of the Ural owl (Strix uralensis). – Auk 105: 316–324. Scherzinger, W. (1980): Zur Ethologie der Fortpflanzung und Jugendentwicklung des Habichtskauzes (Strix uralensis) mit Vergleichen zum Waldkauz (Strix aluco). – Bonn. Zool. Monogr. 15. Scherzinger, W. (1996): Walddynamik und Biotopansprüche des Habichtskauzes (Strix uralensis). – Abh. Zool.-Bot. Ges. Österreich 29: 5–16. Scherzinger, W. (2005): Remarks on Sichuan Wood Owl Strix uralensis davidi from observations in south-west China. – Bull. B. O. C. 125: 275–286. Sharpe, R.B. (1875): Contributions to a history of the Accipitres. Notes on birds of prey in the museum at the Jardin des Plantes and the Collection of Mons. A. Bouvier. – Ibis Vol. 5 (3rd Ser.): Syrnium davidi: 256. Stresemann, E. (1923): Zoologische Ergebnisse der Walter Stötznerschen Expeditionen nach Szetschwan, Osttibet und Tschili. 2/12: Striges bis Ralli. – Abhandl. Ber. Mus. Tierk. Völkerkunde, Dresden 16: 58–70. Sun, Y.-H., Scherzinger, W., Liu, N.-F., Klaus, S. & Fang, Y. (2001): New distribution areas of the Sichuan Wood Owl in Gansu. – Zoologica Sinica 47/4: 473–475. Voous, K.-H. (1962): Die Vogelwelt Europas und ihre Verbreitung. Ein tiergeographischer Atlas über die Lebensweise aller in Europa brütenden Vögel. – Parey, Hamburg, Berlin. Wolters, H. (1975): Die Vogelarten der Erde. Eine systematische Liste mit Verbreitungsangaben sowie deutschen und englischen Namen, vol. 1. – Parey, Hamburg, Berlin. Zheng, G. & Wang, Q. (1998): Aves. In: Wang, S. (ed.): China red data book of endangered animals. – Science Press, Beijing, Hong Kong, New York. Arrived / Prispelo: 31.1.2006 Accepted / Sprejeto: 5.10.2006 ACROCEPHALUS 2.J (128-H9): 3-12, 200Ć APPENDIX / PRILOGA Table 3: Diary of field observations of Sichuan Wood Owl Strix uralensis davidi Tabela 3: Dnevnik terenskih opazovanj se~uanske koza~e Strix uralensis davidi Time / ^as Habitat Date/ Datum 10 Apr 1995 19.45 – 21.00 coniferous forest, snow cover, calm 2700 m a.s.l. wind, clear sky, waxing moon, +2°C Weather / Vreme Description of observation / Opis opazovanja 11 Apr 1995 16.40 coniferous forest overcast, 0°C 19.30 – 20.25 mixed forest, slope 12 Apr 1995 19.19 – 20.04 creek in coniferous forest 13 Apr 1995 12.30 – 17.25 dense spruce tree calm snowing, stand –3°C 19.30 shrubby conifer, at hillside 14 Apr 1995 19.45 – 20.05 conifers in valley full moon, 0°C, calm wind male owl hoots spontaneously; I try to stimulate the owl by imitating Ural Owl’s territorial song: male intensifies its vocalization; dives above my head, and lands in a high spruce tree, singing variable strophes of territorial song; also a rising scale (similar to “nest-site-demonstration” of Ural Owl) male owl calls spontaneously male owl utters territorial song, and long sequences of “nest-site-demonstration”; female owl answers from a rocky area in a steep canyon, with hoarse “begging calls”, and approaches the male; hasty duet (territorial song by the male, “begging calls” by female; male breaks into full song and “nest-site-demonstration” with high intensity; female returns suddenly to rocky slope) male utters single songs, flies to an open area on a dry slope; utters “nest-site-demonstration” male utters single strophes of territorial song; Chinese Nutcracker Nucifraga caryocatactes macella performs mobbing against the owl; the owl tries to escape with short flights; vocal activity in large intervals; I stimulate the owl by imitation of territorial song: male owl lands about 5 m directly above me, in a dense canopy of spruce male handing over a piece of prey to the female, at top of an exposed coniferous tree male starts activity with territorial song; female softly sails past me and lands in the crown of a spruce tree – well visible against the sky; male lands beside its partner, while singing; a long duet follows (territorial song by the male, “begging” by the female); both owls sitting close to each other, turn their face towards each other and lean forward, handing over some piece of prey with the beak; female presses the fetched mouse against the sitting branch with one leg, tears some pieces with its beak and swallows the entire remainder; after this it rubs its beak against the branch; male departs; female continues “begging”, flies to a rocky slope II W. Scherzinger & Y. Fang: Field observations of the Sichuan Wood Owl Strix uralensis davidi in western China continuation of Table 3 / nadaljevanje tabele 3 Time / ^as Habitat Date/ Datum 2 May 1997 20.40 mixed forest, slope Weather / Vreme Description of observation / Opis opazovanja 4 May 1997 20.15 – 20.30 mixed forest, rocky slope free of snow, up to 18°C during daytime, +2°C in evening rain during night, 0°C 6 May 1997 07.35 mixed forest, slope full sunshine 19.20 mixed forest, slope 7 May 1997 20.08 – 20.22 coniferous forest 1 Jul 1999 20.45 – 21.15 rocky canyon, dry mostly rain, thunderstorm, 13°C I stimulate vocalization by imitating territorial song: male answers from long distance, but does not approach I stimulate vocalization by imitating territorial song: male answers immediately with territorial song, and flies near to me; At least long series of “nest-site-demonstration” spontaneous territorial song of male male starts vocal evening activity with “nest-sitedemonstration” male starts singing at an inaccessible steep slope; female answers and a duet follows; I stimulate vocalization by imitating territorial song: male approaches and lands on an exposed branch (about 40 m distance); female follows (both birds clearly visible against the sky); female departs into darkness I stimulate vocalization by imitating territorial song: male utters the first syllables of territorial song; later the female utters a strident “begging” call; parts of territorial song are heard in a duet (probably two males) - followed by fierce aggressive calls (similar to territorial demarcation of Ural Owl in autumn) 12 Acrocephalus 2.J (128-129): 13-20, 2006 Population development, nest site selection and conservation measures for White Stork Ciconia ciconia along the lower Tami{ River (Vojvodina, N Serbia) Bela {torklja Ciconia ciconia ob spodnjem toku reke Tami{ (Vojvodina, S Srbija) – populacijski trendi, izbira gnezdi{~ in varstveni ukrepi zanjo Marko Tucakov Marka Ore{kovi}a 9, 25275 Ba~ki Breg, Serbia, e-mail: mtucakov@eunet.yu Number of breeding pairs, their spatial distribution, selection of nest sites and the breeding success of White Stork Ciconia ciconia was studied in 2004 in 20 villages adjacent to the lower Tami{ valley. 322 pairs, which occupied their nests for at least four weeks (HPa) during the first half of the breeding season, were found, 307 of which were HPm: pairs with fledged young. The number of breeding pairs in the 1957 - 2004 period grew in the area, but population in the entire Vojvodina fluctuated. The study area is the most important breeding area for White Stork in the province (30.7% of all breeding pairs) and the country: 27.4% of national population breeds there. One of the most important reasons for the high population density are very suitable feeding conditions. Most of the pairs with fledged young (HPm) had 3 chicks per pair (40.1%), followed by pairs with two chicks (32.5%), four chicks (15.2%), one chick (11.4%) and five chicks (2.4%). The majority of nests were situated on buildings (53.7%) and electric pylons (41.9%). There is a marked change in the breeding habits compared with those in the 1980’s: straw and hay bales have been almost completely abandoned as nest sites since then. Out of the entire number of HPa, 58 (18%) are situated in seven villages adjacent to the first river sector, 213 (66%) in nine villages along the second sector (where the river has wide floodplain) and 51 (16%) in four villages along the third river sector. Key words: White Stork, Ciconia ciconia, population development, Tami{, Serbia, Vojvodina Klju~ne besede: bela {torklja, Ciconia ciconia, razvoj populacije, Tami{, Srbija, Vojvodina 1. Introduction At the national level, White Stork Ciconia ciconia census in Serbia was carried out in 1996. A breeding population of 872 pairs was counted, 93% of them in Vojvodina, indicating the region as the most important for breeding of this species (Pelle 1996). Surveys of breeding pairs of White Stork in Vojvodina have long tradition. Censuses in the province were carried out in 1957 (Szlivka 1959), 1974 (Garovnikov 1977), 1979 (Garovnikov 1980-81) and 2000 (Gergelj et al. 2000). The parameters which were subject of census routine included only total number of active nests (breeding pairs) in particular villages and the nest site selection. Breeding success was surveyed only during the 2000 census (Gergelj et al. 2000). As the last census carried out in 2000 confirmed that one quarter of all pairs breed in villages situated on the edge of the lower Tami{ valley (Gergelj et al. 2000), this region has been subject of survey in 2004. The aim was to determine the number of breeding pairs along the lower Tami{, breeding success and nest site selection. The results of this census are presented and discussed in this paper. I3 M. Tucakov: Population development, nest site selection and conservation measures for White Stork Ciconia ciconia along the lower Tami{ River (Vojvodina, N Serbia) 2. Study area and methods 2.1. Study area The study area was part of the Tami{ River valley, which is situated in Serbia (between 40o50’ N 45o28’ E and 20o23’ N 20o58’ E) in central and southwest Banat (Lazi} 1996). Out of 359 km of this transboundary river, 118 pass through Serbia, between the village of Ja{a Tomi} and the river mouth at Pan~evo (Figure 1). Fluvial erosion and very pronounced fluctuations of the water level, as well as meandering, are the main river’s characteristics. Extremely high water levels are usually recorded in April, extremely low in October (Tomi} 1989). The river regulations started in 1728 and were completed in 1977, when part of the river stretch was included into the Danube-Tisa-Danube hydro system (Tomi} 1989). However, despite the fact that the greater part of the stretch is currently canalized (from the state border downstream to Boto{ (the first sector in the text below) and from Opovo to the river mouth (the third sector in the text below), the river is free-flowing between Boto{ and Opovo (the second sector in the text below) with intensive meandering, preserved river branches (near Farkaždin and Baranda), oxbows (near ^enta), floodplain meadows (near Toma{evac, Uzdin, Farkaždin, ~enta and Boto{) and alluvial forests. Three sectors are different from the hydrological point of view. The first is 33 km long, the second 42 km, the third 43 km (Lazi} 1996). The widest river valley, regularly flooded, follows the boundaries of the second sector. Its widest parts are situated between ^enta and Baranda (10.6 km) and between Orlovat and Uzdin -9.3 km. Five large fishponds are situated in the valley: Sutjeska (900 ha) near Sutjeska, Sveti Nikola (400 ha) at Neuzina, Uzdin (430 ha) near Uzdin, Baranda (1005 ha) between Baranda, Sakule and Opovo, and ^enta (120 ha) near ^enta (Bugar~i} 1999). 2.2. Methods The census was carried out during the breeding season in 2004, from late May to mid August, in particular between 4 Jul and 18 Jul, in the phase of breeding cycle when chicks are visible from the ground, as recommended by the International White Stork Census methodology (Schulz 1999). This allowed simultaneous survey of nest site selection and breeding success. The following data were recorded: nest occupancy, nest site selection (with the following categories: building, electric pylon, tree, straw/hay bales, other) and number of fledged chicks. Only nests occupied by a pair for at least four weeks during the first half of the breeding season are considered to be occupied (Schulz & Thomsen 1999). All nests situated on man-made buildings (houses, churches, observation towers and local power stations) were taken as single category, having in mind that further separation of this category was in many cases impossible. Besides settlements, all other potential breeding areas were surveyed. While counting StDBiol, I took in account surface data for the river valley given by Lazi} (1996). Figure 1: The study area of the lower Tami{ River Slika 1: Raziskovano obmo~je spodnjega toka reke Tami{ 14 ACROCEPHALUS 2.J (128-H9): I3-2O, 200Ć Table 1: Number of breeding pairs of White Stork Ciconia ciconia in villages along the lower Tamiš River in 2004 Tabela 1: Število gnezdečih parov bele štorklje Ciconia ciconia v vaseh vzdolž spodnjega toka reke Tamiš v letu 2004 Naselje HPa HPm Nesting sites / Gnezdi{~a Settlement / Building/ Zgradba Electric pylon/ El. drog 6 Trees/ Drevesa 0 Hay bales/ Bale sena Other niches/ Ostale ni{e Sakule 31 29 24 0 1 Uzdin 30 28 23 4 1 0 2 Boto{ 30 28 0 30 0 0 0 ^enta 28 28 18 9 1 0 0 Idvor 26 25 20 3 2 0 1 Baranda 26 25 24 2 0 0 0 Orlovat 21 17 1 19 0 0 1 Neuzina 21 19 1 20 0 0 0 Opovo 17 17 14 3 0 0 0 Jabuka 13 13 12 0 1 0 0 Boka 12 12 4 8 0 0 0 Toma{evac n 11 1 9 0 0 1 Sefkerin 11 11 11 0 0 0 0 Glogonj 10 10 9 1 0 0 0 Farkaždin 10 9 6 2 0 2 0 Sutjeska 9 9 2 7 0 0 0 Se~anj 7 7 3 4 0 0 0 Ban. Despotovac 4 4 0 4 0 0 0 Ja{a Tomi} 3 3 0 2 0 1 0 Šurjan 2 2 0 2 0 0 0 Total 322 307 173 135 5 3 6 Abbreviations of the breeding parameters follow the methodology of International White Stork Census (Schulz & Thomsen 1999; Table 5 in the appendix). White Stork nests situated on wires of power pylons, as well as birds themselves, can cause a short-circuit, if they complete electric circuit between live and ground wire. Having results of nests site selection from 2004, in order to minimize the conflict and following the best practice in protection of nests situated on the wires (e. g. Perrenou et al. 1996, Mu`ini} 1999), the most problematic nests in the study area placed on overhead wires on electric pylons in 2004 were supported by erection of platforms on top of the pylons before the start of the breeding season in 2005 (between 9 and 30 March). The metal platform was designed in order to create space (70 cm) between the nest and the wires. During the erection process, old nests were taken down from the wires, after which initial layer of branches was fixed at the bottom of the platform on the ground. Then platform was erected and fixed to the pylon top. Nest acceptance was checked between 29 May and 10 Jul 2005. 3. Results In the study area, 360 nests were counted, 322 of which were occupied. Occupied nests were found only in settlements: there were no nests outside them. Nests were found in each of 20 villages bordering the river floodplain (Table 1). Out of all 322 breeding pairs, 58 (18%) were situated in seven villages adjacent to the first river sector, 213 (66%) in nine villages along the second sector, and 51 (16%) in four villages along the third river sector. Breeding success of all breeding pairs (JZa) was 2.58, and 2.60 of all breeding pairs that raised chicks (JZm). Most of the pairs with fledged young (HPm) had 3 chicks per pair (40.1%), followed by pairs with two chicks (32.5%), four chicks (15.2%), one chick (11.4%) and five chicks (2.4%; Table 2). There is no 15 M. Tucakov: Population development, nest site selection and conservation measures for White Stork Ciconia ciconia along the lower Tami{ River (Vojvodina, N Serbia) 3 100% m 90% » 80% S 70% V 60% « 50% L 40% "S 30% § 20% '€ 10% | 0% a. 1957 1974 1979 2000 2004 Year / Leto [H other niches ¦ straw or hey bales ¦ tree S electric pylon ¦ building Figure 2: Changes in White Stork Ciconia ciconia nest site selection in the lower Tami{ valley Slika 2: Spremembe v izbiri {torkljinega gnezdi{~a bele {torklje Ciconia ciconia v dolini spodnjega toka reke Tami{ correlation between the pairs’ breeding success within different villages (density groups; Figure 3). The vast majority of nests were situated on buildings (53.7%) and electric pylons (41.9%; Table 1). “Biological” potential density (StDBiol) was 26.9 pairs / 100 km2. The majority of nests on the pylons were situated on the non-isolated wires. Out of 135 these nests in the study area (Table 1), 37 nest-isolation platforms were erected before the beginning of the breeding season, in March 2005. 29 of them (78.3%) were accepted already in that year (Table 4). 4. Discussion The number of breeding pairs in the 1957 - 2004 period grew (Table 3). Despite the fact that in some censuses the coverage of all villages was not complete in the entire valley, the trend is very indicative. However, population in the entire Vojvodina fluctuated in the same period (Szlivka 1959, Čarovnikov 1980-81, Čarovnikov 1977, Pelle 1996). The breeding success (Table 2) was lower than in 2000, when JZm in the same area was 3.38, similar to the whole province where JZm was 3.14 (Gergelj et al. 2000). A possible reason for this is the difference in the level of precipitation between the two years. The year 2000 was the droughtiest in Serbia since the very beginning of weather surveys in Serbia (Republic 16 Hydrometeorological Service of Serbia 2001), while the territory of Vojvodina was extremely wet in 2004 (Republic Hydrometeorological Service of Serbia 2005). Similarly high reproductive success of the White Stork in 2000 was recorded in Switzerland (Boetticher-Streim 1991). It has been proved that the breeding success is in negative correlation with the level of precipitation (Bert & Lorenzi 1999). Absence in correlation of breeding success with density of breeding pairs (Figure 3) opposes the one in the Sava River valley in Croatia, where White Storks breeding in high densities have had higher breeding success, indicating that food resources were evenly distributed and almost unlimited (Schneider-Jacoby 1993). Table 2: Breeding success of White Stork Ciconia ciconia at villages in the lower Tami{ valley in 2004 Tabela 2: Gnezditveni uspeh bele {torklje Ciconia ciconia v vaseh vzdolž spodnjega toka reke Tami{ v letu 2004 Parameter Value/ Vrednost H 360 HPa 322 HPm 307 HB1 16 HB2 6 HPo 16 HPx 33 HPm1 28 HPm2 94 HPm3 116 HPm4 44 HPm5 7 JZG 798 JZa 2.48 JZm 2.60 It is estimated that recently 1000 - 1100 pairs bred in Vojvodina, and 1100 - 1250 in the entire Serbia (Puzovi} et al. 2003), which makes the Tami{ River valley the most important breeding area for White Stork in the province (30.7% of breeding pairs breed there) and the country: 27.4% of the national population breed there. One of the most important reason for such a high population density are very suitable feeding conditions, which are considerably ACROCEPHALUS 2.J (128-H9): I3-2O, 200Ć 4 3.5 3 2.5 2 1.5 1 0.5 0 t' 5 10 15 20 25 30 35 Hpa Figure 3: Correlation between the number of occupied nests of White Stork Ciconia ciconia per village and the breeding success in the lower Tami{ valley (Spearman ? = -0.049, df = 18, NS, two-tailed test, H0 is not rejected) Slika 3: Korelacija med {tevilom zasedenih gnezd na vas in gnezditveni uspeh bele {torklje Ciconia ciconia v dolini spodnjega toka reke Tami{ (Spearman ? = -0.049, df = 18, NS, dvorepi test, H0 ni zavrnjena) more favourable then in other areas in Serbia where regional censuses are conducted (e.g. Ra{ajski 1988, Kuli} 2004). The highest concentration of breeding pairs along the second Tami{ sector is due to the very favourable local feeding conditions in this wide inundation stretch, particularly in the preserved, extensive, temporarily flooded meadow and pastures that follow the river along both banks in this stretch. Findings of Schneider-Jacoby (1993), Eichelmann (1999) and [tumberger & Velevski (2001) also prove that these are optimal foraging habitats for the White Storks and, when covering large areas, they support high breeding densities. These habitats are almost completely missing along the first and the third sectors (own data). It has already been proven that foraging areas during the breeding season in favourable areas Table 3: Development of breeding population of White Stork Ciconia ciconia in the lower Tami{ valley between 1957 and 2004 Tabela 3: Razvoj gnezde~e populacije bele {torklje Ciconia ciconia vzdolž spodnjega toka reke Tami{ med letoma 1957 in 2004 Census year/ Leto popisa HPa Source / Vir 1957 58 SZLIVKA I959 1974 178 Čarovnikov 1977 1979 221 Čarovnikov 1980-81 2000 25O Gergelj et al. 2000 2004 322 this paper are situated in immediate vicinity of nests (OŻgo & Bogucki 1999). All nest sites that White Storks use traditionally in Vojvodina are used in the Tami{ valley as well. However, there is a marked change in the breeding habits compared with those in the 1980s (starting from 1979, as Ra{ajski (1988) describes the situation for S Banat): straw and hay bales have been almost completely abandoned as nest sites since then (Figure 2). The most probable cause for this are changed practices of straw and hay conservation (in stables, under roofs), as well as evident absence of these nesting places in recent years (own data). The same trend was proven for Hungary, but sharp increase of electric pole usage for White Stork breeding started at least one decade earlier then in the Tami{ valley and the entire Vojvodina (Lovászi 1999). Use of trees as nest sites in the study area decreased between 1957 and 2004 as well (Figure 4), although reasons for that are not clear. One of the suggested reasons can be absence of old trees suitable for breeding (A. Žuljevi}, pers. comm.), which happened, for example, in Slovenia (Denac 2001). Numerous concentrated tree-breeding White Stork pairs are site-specific in Vojvodina (Kanjo 2000). Table 4: Results of the White Stork Ciconia ciconia nest-protection program in the lower Tami{ valley in 2005 Tabela 4: Rezultati programa za za{~ito {torkljinih Ciconia ciconia gnezd v dolini spodnjega toka reke Tami{ v letu 2005 Village/ Erected/ Accepted/ Vas Postavljeno Sprejeto Orlovat 10 6 Neuzina 10 9 Botos 8 7 Sutjeska 4 3 Tomasevac 3 2 Sakule 2 2 Acknowledgements: I dedicate this article to my friend David Reeder who is in love in Vojvodinian floodplains and their wildlife and who believes even more than I do that they can and will be sufficiently protected. This study was financed by the Rufford Small Grants for Nature Conservation of The Rufford Maurice Laing Foundation. Many thanks also to Elektrovojvodina Public Enterprise for their professional cooperation in White Stork conservation, to Robert MacCurrach who helped me during survey. Damijan Denac and Slobodan Puzovi} gave useful comments to the manuscript. 17 M. Tucakov: Population development, nest site selection and conservation measures for White Stork Ciconia ciconia along the lower Tami{ River (Vojvodina, N Serbia) 5. Povzetek Leta 2004 je avtor prispevka v 20 vaseh, meje~ih na spodnji tok reke Tami{ v Vojvodini, ugotavljal {tevilo gnezde~ih parov, izbiro gnezdi{~ in gnezditveni uspeh bele {torklje Ciconia ciconia. Zabeleženih je bilo 322 parov, ki so v prvi polovici gnezditvenega obdobja zasedali gnezda najmanj {tiri tedne (HPa), 307 izmed katerih so bili pari s speljanimi mladi~i (HPm). V obdobju 1957 - 2004 je {tevilo gnezde~ih parov v tem obmo~ju naraslo, medtem ko je populacija bele {torklje v celotni Vojvodini nihala. Preu~evano obmo~je je najpomembnej{e gnezditveno obmo~je za belo {torkljo tako v Vojvodini (30,7% vseh parov) kot v celotni državi (27,4% srbske populacije). Eden izmed najpomembnej{ih razlogov za visoko populacijsko gostoto bele {torklje v tem obmo~ju so nadvse ugodne prehranjevalne razmere. Ve~ina parov s speljanimi mladi~i (HPm) je imela po tri mladi~e (40,1%), njim pa so sledili pari s po dvema (32,5%), {tirimi (15,2%), enim (11,4%) in petimi mladi~i (2,4%). Ve~ina gnezd je bila spletenih na stavbah (53,7%) in elektri~nih drogovih (41,9%). Sicer pa so se v precej{nji meri spremenile gnezditvene navade bele {torklje v primerjavi s tistimi v osemdesetih letih prej{njega stoletja: odtlej so bile kot gnezdi{~a skoraj povsem opu{~ene slamnate bale. Od celotnega {tevila parov s speljanimi mladi~i jih 58 (18%) živi v sedmih vaseh, meje~ih na prvi re~ni sektor, 213 (66%) v devetih vaseh vzdolž drugega sektorja (z veliko poplavno ravnico), 51 (16%) pa v {tirih vaseh vzdolž tretjega re~nega sektorja. 6. References Bert, E. & Lorenzi, M.C. (1999): The influence of weather conditions on the reproductive success of the White Stork (Ciconia ciconia) in Piedmont/Italy. pp. 437– 442 In: Schulz, H. (ed.): Weissstorch im Aufwind? – White Storks on the up? Proceedings of International Symposium on the White Stork, Hamburg. – NABU, Bonn. Boetticher-Streim, W. (1991): Der Weissstorch in der Schweiz. – Schweizerische Gesellschaft fur den Weissstorch, Altreu. Bugar~i}, P. (1999): Artificial lakes of Vojvodina – geographical aspects and problems. – PhD thesis, University of Novi Sad, Faculty of Sciences, Institute of Geography, Novi Sad. Denac, D. (2001): Breeding biology, phenology and distribution of White Stork Ciconia ciconia in Slovenia. – Acrocephalus 22 (106–107): 89–103. Garovnikov, B. (1977): White Stork census in Vojvodina in 1974. – Arhiv biolo{kih nauka 29 (1–2): 89–95. Garovnikov, B. (1980-81): Brojnost belih roda (Ciconia ciconia L.) u Vojvodini na osnovu popisa gnezda u 1974. i 1979. godini. – Priroda Vojvodine 6–7: 47–57. 18 Gergelj, J., Puzovi}, S., Ra{ajski, J., Balog, I., Luka~, S., Žuljevi}, A., Tucakov, M., Matovi}, ^., Stojni}, N. & Kova~evi}, B. (2000): Bela roda (Ciconia ciconia) u Vojvodini 2000. godine – populacija i distribucija. – Ciconia 9: 32–44. Kanjo, B. (2000): Gnež|enje bele rode (Ciconia ciconia) u Svilojevu u periodu 1990 – 1999. – Ciconia 9: 187. Kuli}, S. (2004): Promene brojnosti i rasporeda parova bele rode Ciconia ciconia u Leskova~koj kotlini (jugoisto~na Srbija). – Ciconia 13: 114–121. Lazi}, L. (1996): Tami{ river watereconomy problems of Potamisje in Yugoslavia. – PhD Thesis, University of Novi Sad, Faculty of Natural Sciences and Mathematics, Institute of Geography, Novi Sad. Lovászi, P. (1999): Conservation of the White Stork in Hungary. pp. 203–211, In: Schulz, H. (ed.): Weissstorch im Aufwind? – White Storks on the up? Proceedings of International Symposium on the White Stork, Hamburg. – NABU, Bonn. Mu‘ini}, J. (1999): A frame for White Stork nest. – Israel Journal of Zoology 45: 497–499. Ożgo, M. & Bogucki, Z. (1999): Home range and intersexual differences in the foraging habitat use of a White Stork (Ciconia ciconia) breeding pair. pp. 481–492 In: Schulz, H. (ed.): Weissstorch im Aufwind? – White Storks on the up? Proceedings of International Symposium on the White Stork, Hamburg. – NABU, Bonn. Pelle, Z. (1999): Status and biology of White Stork in Yugoslavia. pp. 219–221 In: Schulz, H. (ed.): Weissstorch im Aufwind? – White Storks on the up? Proceedings of International Symposium on the White Stork, Hamburg. – NABU, Bonn. Perennou, C., Sadoul, N., Pineau, O., Johnson, A. & Hafner, H. (1996): Management of nest sites for colonial waterbirds. Tour de Valat, Le Sambuc. Puzovi}, S., Simi}, D., Savelji}, D., Gergelj, J., Tucakov, M., Stojni}, N., Hulo, I., Vizi, O., [}iban, M., Ruži}, M., Vu~anovi}, M. & Jovanovi}, T. (2003): Birds of Serbia and Montenegro – breeding population estimates and trends: 1990 – 2002. – Ciconia 12: 35–120. Ra{ajski, J. (1988): Brojnost gnezde}ih parova belih roda (Ciconia ciconia) sa prate}im pojavama gnež|enja u južnom Banatu za period 1976 – 1985. – Larus 40: 111–123. Republic Hydrometeorological Service of Serbia (2005): Osnovne klimatske karakteristike na teritoriji Srbije u periodu januar-decembar 2004. godine. – Republic Hydrometeorological Service of Serbia. , (Downloaded: 10 Sep 2006) Republic Hydrometeorological Service of Serbia (2001): Padavinski re`im u Srbiji 1961 - 1990. – Republic Hydrometeorological Service of Serbia. , (Downloaded: 10 Sep 2006) Schneider-Jacoby, M. (1993): Vögel als Indikatoren für das ökologische Potential der Saveauen und Möglichkeiten für deren Erhaltung. – PhD Thesis, Biologischen Fakultät, Universität Konstanz. ACROCEPHALUS 2.J (128-H9): I3-2O, 200Ć Schulz, H. & Thomsen, K-M. (1999): Abbreviations, Glossary. pp. 25–26 In: Schulz, H. (ed.): Weissstorch im Aufwind? – White Storks on the up? Proceedings of International Symposium on the White Stork, Hamburg. – NABU, Bonn. Szlivka, L. (1959): Results of the White Stork census in the Vojvodina (Yugoslavia). – Aquila 66: 262–266. Tomi}, P. (1989): Reka Tami{ i njeni vodoprivredni problemi. – Zbornik radova Instituta za geografiju 16: 20–26. Arrived / Prispelo: 6.6.2006 Accepted / Sprejeto: 5.10.2006 I9 M. Tucakov: Population development, nest site selection and conservation measures for White Stork Ciconia ciconia along the lower Tami{ River (Vojvodina, N Serbia) APPENDIX / DODATEK Table 5: Abbreviations of White Stork Ciconia ciconia breeding parameters used in the text (after Schulz & Thomsen 1999) Tabela 5: Okraj{ave za gnezdilne parametre bele {torklje Ciconia ciconia, uporabljene v tekstu (po Schulz & Thomsen 1999) H HPa HPm HPmx HPo HPx HB1 HB2 JZG JZa JZm StDBiol pair which has occupied a nest for at least four weeks during the first half of the breeding season pair with fledged young pair with x fledged young pair without fledged young which has occupied a nest for at least four weeks during the first half of the breeding season pair with unknown breeding success which has occupied a nest for at least four weeks during the first half of the breeding season single bird visiting the nest, no binds to the nest two birds (pair) visiting the nest, no binds to the nest total number of fledged young in a defined area per year breeding success, average number of fledged young per pair related to all HPa of a defined area breeding success, average number of fledged young per pair related to all HPm of a defined area “biological” population density, number of pairs (HPa) per 100 km2 of potential feeding habitat nest 2C Acrocephalus 2.J (128-129): 21-3 5, 2006 Hidden leks in the Yellow-browed Warbler Phylloscopus inornatus? -Investigations from the Khan Khentey reserve (Mongolia) Prikriti leki pri mu{ji listnici Phylloscopus inornatus? – raziskava iz rezervata Khan Khentey (Mongolija) Peter Hans Wilhelm Biedermann Department of Behavioural Ecology, Institute of Zoology, University of Bern, Wohlenstrasse 50A, CH-3032 Hinterkappelen, Switzerland, e-mail: peterbiederm@students.unibe.ch The breeding biology of the Yellow-browed Warbler Phylloscopus inornatus is largely unknown. Recently the species was found to breed in clusters and some hypotheses on the function of aggregated breeding sites have been proposed. To obtain new insights, two clusters of Yellow-browed Warblers were observed from May to July 2003 in the Mongolian Khentey. Mist-netting, nest searching, behavioural observations and habitat analysis were conducted. In contrast to the large breeding clusters found in the main area of the species distribution in Siberia, pairs in the Khentey are breeding in single groups of a maximum of 3 nesting pairs. At this southern border of its breeding range the species has problems with immense predation pressure and unequal sex-ratios (approx. 5:1 males to females). As a consequence, four breeding pairs had at least ten nests, which were all predated before young were hatched. Nests are built solely by females and one of the females replaced its nest three times following predation of eggs. There was great fluctuation of territorial males in the clusters, which sometimes contained more than 20 males. Breeding pairs were formed after peaks in the numbers of territorial males. Due to its small breeding groups or clusters, the Khentey has great advantages for understanding and testing the process of cluster formation in this species. Locations of breeding territories were independent of habitat parameters, but centred in the middle of unsuccessfully established territories. Thus, this small dataset strongly supports the hidden lek hypothesis as the main explanation of clustering, saying that males with territories in the centre of a territory-congregation are sexually the most attractive for females. Yellow-browed Warblers could be one of the few and best examples for the correctness of this hypothesis. New data on biometry, breeding biology and behaviour in this species are discussed. Key words: Phylloscopus inornatus, Yellow-browed Warbler, hidden lek, cluster breeding, sex ratio, predation Klju~ne besede: Phylloscopus inornatus, mu{ja listnica, skriti lek, rasti{~e, gnezditev v skupkih, razmerje spolov, predacija 1. Introduction In central Europe, the Yellow-browed Warbler Phylloscopus inornatus is widely known from accidental sightings. In the past there were three subspecies combined under the species Phylloscopus inornatus: The Yellow-browed Warbler (formerly known as Phylloscopus inornatus inornatus) and the two subspecies of the Hume’s / Buff-browed Warbler (formerly known as Phylloscopus inornatus humei and P. i. mandelleii). Because of morphological, ecological and behavioural criteria the Yellow-browed Warbler and the Hume’s / Buff-browed Warbler are now seen as two species, and the second is now called Hume’s Warbler Phylloscopus humei (Glutz v. Blotzheim & Bauer 1991, Svensson 1992). Breeding sites are relatively rare in the European part of the Ural-mountains, but fairly common east of it. The species is, without large geographic differences, 21 P.H.W. Biedermann: Hidden leks in the Yellow-browed Warbler Phylloscopus inornatus? - Investigations from the Khan Khentey reserve (Mongolia) distributed in the taiga-forests of Siberia between the Urals and the Ochotsk-sea (Glutz v. Blotzheim & Bauer 1991). Recently breeding attempts of Yellow-browed Warblers were reported from the Mongolian Khentey (A. Barkow unpublished data, Wichmann 2001, Wichmann & Pokrovskaya 2004), although this is outside the formerly accepted breeding range (Chabry 1989). This area of forest-steppes in northern Mongolia constitutes the southern border of the species’ breeding range. The breeding biology of the Yellow-browed Warbler and the closely related Hume’s Warbler is largely unknown (compare Price & Jamdar 1991 for P. (i.) humeii). Despite its abundance in some areas, only some parameters of nesting sites and eggs are described in the literature (Witherby et al. 1943, Worobjew 1963, Glutz v. Blotzheim & Bauer 1991). “Cluster breeding” in birds is a quite unusual form of breeding, not found in many species. Breeding pairs are not randomly distributed over a suitable habitat, but clustered in groups. In contrast to colonial breeding, each pair still has its own defended territory. Most species that congregate in clusters for breeding do so for common defence against predators and/or follow an uneven distribution of food. Recent studies showed that the form of the mating system can also lead to cluster breeding (Herremans 1993, Danchin & Wagner 1997, Wagner 1997). Breeding in clusters for Phylloscopus inornatus is briefly discussed in the literature (Bourski & Forstmeier 2000), but its function and frequency are largely unknown or speculative. 2. Study area Bordering the Russian Federation in the north and the People’s Republic of China in the east, south and west, Mongolia is a landlocked country which covers an area of 1.56 million km2. It extends 1236 km from north to south and 2405 km from east to west, and is the seventh largest country in Asia. The Khan Khentey Strictly Protected Area (KKSPA), situated in the northeast of Mongolia, was founded in 1992 (Figure 1). This huge uninhabited area, stretching from the Russian border to the northeast of Ulaanbaatar, lies between 48° and 49°N and from 107° to 110°E. It covers 1.2 million ha and is the fourth largest protected area in Mongolia (Myagmarsuren 2000). Compared with the protected boreal ecosystems in Europe, it is as large as the complete protected forests of Fennoscandia (Von Velsen-Zerweck 2002). Cooperation between Göttingen University and the National University of Mongolia led, in 1998, to the research station “Khonin Nuga” being established by the side of the Eröö river in the western buffer zone of the KKSPA (Figure 1). The study area at Khonin Nuga (49o04’N, 107o24’E) lies at an altitude of 1000 m (+/-50 m) a.s.l. and covers 1.8 km2. Figure 1: The location of Khan Khentey Strictly Protected Area and the research station Khonin Nuga (after MNE & WWF 1994). Slika 1: Lokacija rezervata Khan Khentey in raziskovalne postaje Khonin Nuga (povzeto po MNE & WWF 1994). In Khonin Nuga, Yellow-browed Warblers inhabit Betula fusca shrubs, situated between riverine meadow pastures and Asian White Birch Betula platyphylla / Siberian Larch Larix sibirica forests on hillsides at the base of the northern mountainside. These shrub areas can be divided into a central region of vegetation of relatively homogenous height (approximately 3 metres) with only a few taller trees of Asian White Birch and small numbers of Siberian Larch, that contrasts with the hillside belt of shrubs that includes a lot of umbrella-trees (Larix sibirica) and young growth of larch, Siberian Spruce Picea obovata and willows (Salix sp.). In 2003, I found two areas with a large number of clumped territories (clusters) of Yellow-browed Warblers (see Figure 2) in Khonin Nuga. One of these clusters (“Patch 1”) was also colonized the year before (A. Barkow unpublished data), the other (“Patch 2”) was new. Patch 1 ("2.7 ha) lies in the south-western part of the area in the smooth transition zone between Betula fusca shrubs and the hillside forest (see Figure 22 ACROCEPHALUS 2.J (128-H9): 21-3 J, 200Ć Figure 2: The location of patch 1 (photo at the top), situated north of hillside forests, and patch 2 (photo at the bottom) in the centre of the valley (photo: P.H.W. Biedermann). Slika 2: Lokacija obmo~ja 1 (zgoraj), severno od gozdov na pobo~ju in območja 2 (spodaj) na sredi doline (foto: P.H.W. Biedermann). 23 P.H.W. Biedermann: Hidden leks in the Yellow-browed Warbler Phylloscopus inornatus? - Investigations from the Khan Khentey reserve (Mongolia) 2). Patch 2 ("0.9 ha) lies in the centre of the Betula fusca shrub, bordered by rows of larch and birch aged about 30 years (see Figure 2). Nearly the whole study area was burnt down 30 years ago, which explains the high density of shrubs instead of forest. In central Siberia, Yellow-browed Warblers prefer younger regrown vegetation in burnt habitats (Glutz v. Blotzheim & Bauer 1991, Forstmeier et al. 2001). 3. Methods In 2003 the whole study area (1.8 km2) was covered by two censuses a day, between 10 May and 15 May to look for Yellow-browed Warblers; then, until 3 Jul, the area was surveyed once a day. When the first individuals arrived on 12 May, I started to catch birds full-time with mist nets until 21 May. After that mist netting was continued for only half a day until 7 Jun. Catching in the morning between 8.00 and 11.00 h was particularly efficient. More birds were caught when the weather was bad (around 80% of ringed birds when the cloud density was more than 50%). All birds were marked by colour rings, for recognition later in the field and released next to the net where they were caught. Yellow-browed Warblers cannot be sexed by plumage. Most birds were thus sexed by wing-length (males: > 66 mm; females: < 64 mm) (for techniques and more information see Svenson 1992), the remainder by their behaviour in the field. In addition to wing-length, the length of the 8 primary (1st primary next to 1st secondary) and body-weight were measured. Additionally, I surveyed the area (in particular both patches mentioned above) for ringed and non-ringed individuals. Each Yellow-browed Warbler was documented, including its location and behaviour (particularly territorial, feeding, mating and nesting behaviour). Nesting territories were estimated by mapping singing posts and territorial behaviour. A male was considered to “own” a tree if it was observed singing in it. Two males were never recorded singing in the same tree, except at territory borders. From this observation it was possible to distinguish between two types of territories. The first was the centre of the territory where all the singing posts, territorial behaviour and nests were recorded – this area was defended against other males. The second territory was the whole area in which males were not observed every day (especially when there were a lot of territorial males around) and where they usually did not sing. Each time non-ringed birds were observed, I replaced the nets on the following day in an attempt to catch them. 24 When females started to build nests I tried to locate the nest sites. Each nest was checked every morning around 7.00 h for eggs. During the period of nestbuilding it is surprisingly easy to locate females. At this time of season shrubs are foliating and females often contact the male with calls after leaving the nest (for similar observations on P. (i.) humeii see Price & Jamdar 1991). Behaviour patterns, like territorial behaviour between males, nest-building behaviour of females, anti-predator and mating behaviour were observed (see Results). At the end of the breeding season all nest sites were analysed for material used, direction of nest entrance and habitat. At the end of June, when only two pairs were left in the area, an attempt was made to obtain footprints of terrestrial nest predators with a sand-filled hole on a path near one nest. Around another nest, six small-mammal traps were placed at a distance of 5 metres. Habitat parameters of “long-time territories” were compared with those of the area bordering the patch. At randomly chosen points (3 x 3 m) the mean height and diameter of Betula fusca shrubs and the number of wood-trunks were measured. “Long-time territories” are defined as areas where birds stayed more than two days (N = 8). The numbers of small / large Asian White Birches, Siberian Larches and other trees in plots of 50 x 50 m within and outside territories with nests were recorded. 4. Results 4.1. Catching phenology During 18 catching days I ringed 29 individuals (24 males, 5 females), with one to nine mist nets (9143.75 net-hours x m2 or 0.00317 caught birds / net-hour x m2). The first territorial males arrived in the study area on 15 May. The first female was caught on 18 May. Ringing was most successful during the first three days of the study when more than one third of the total catch were ringed (10 males, 1 female). After that only three birds could be caught until the 1 Jun, when again ten males and one female were ringed (Figure 3). The sex ratio of those caught by mist-netting was 24 males to 3 females. The two other females were caught in front of their nests. Most individually ringed specimens were not seen again after the first capture (17 males, 2 females). All mated individuals remained in the study area for more than ten days (Figure 4). ACROCEPHALUS 2.J (128-H9): 21-3 J, 200Ć 30 25 20 :15 10 5 o-<5 pooooo .o-o-oo-ooo 30 25 20 15 10 5 15.05 18.05 21.05 24.05 27.05 30.05 02.06 05.06 08.06 Date / Datum Figure 3: Numbers of Yellow-browed Warblers Phylloscopus Inomatus caught during the survey. Open circles and dotted line denotes cumulative total ringed, filled circles number ringed each day and triangles females ringed each day. Slika 3: Število mušjih listnic Phylloscopus inomatus ujetih med raziskavo. Odprti krožci in pikčasta črta pomenijo skupno število obročkanih ptic, polni krožci število obročkanih ptic po dnevih, trikotniki pa število obročkanih samic po dnevih. The sample size of measured females was too small to do statistical tests. 4.3. Bird counts The first two singing males were recorded on 15 May (patch 1). This date is about one week later than the year before (A. Barkow unpublished data), probably due to bad weather with snow at higher altitudes. Patch 2 was first settled on 22 May. After that four to five flocks (with 8 to 22 males) were observed that migrated through the area (see Figure 6), as well as many territorial males which established short-term territories for one to ten days in the two small patches (2.7 + 0.9 ha). The number of singing males fluctuated greatly. The greatest numbers recorded were 7.4 (patch 1) and 13.3 (patch 2) territorial males per hectare (Figure 6). Most of the males were observed in patch 1 (max. 20 males; in contrast to max. 12 males in patch 2). The high density of territorial males did not result in high breeding densities, because of the almost complete absence of females. Although a lot of the 4.2. Biometrics The biometry of captured birds was very similar to the sparse data already published. Five females, with wing length 52 – 55 mm (mean 53.4 mm, SD 1.0) and weight 5.5 – 6.5 g (5.9 g, SD 0.4), and 24 males, with wing length 55 – 60 mm (mean 57.5 mm, SD 1.2); weight 5.5 – 7.0 g (mean 6.4 g, SD 0.4). Glutz v. Blotzheim & Bauer (1991) give a mean wing length of 54.0 mm and a mean weight of 6.0 g for females (N = 47), and a mean wing length of 57.0 mm and a mean weight of 6.0 g for males (N = 75) (data from Hopei / NE China for April / May and August / October). Males that remained in the study area for more than one day (N = 7) were slightly smaller (mean wing length = 57.1 mm, SD 1.77) in contrast to those that were not sighted again (57.6 mm, SD 1.0, N = 17). No significant correlation was observed between wing length and body weight of the two groups (t-test: wing length (t) = 0.92, P > 0.05; Mann-Whitney Rank Sum test: T(weight) = 92; P > 0.05) (Figure 5). The three mated males had the largest wing length (mean = 58.3 mm) but the smallest body mass (mean = 6.2 g compared to 6.4 g); the sample size was too small to test for statistical significance. It appeared that larger and heavier males arrived earlier in the year (see Figure 5), but the correlation was weak (Spearman rank order correlation: P (wing length) = 0.10; P (weight) = 0.58). D 0 days ¦ 1-10 days S 11-20 days ¦ 21-30 days ¦ 31-40 days S 41+days Figure 4: Number (proportion) of birds and how long they stayed in the study area. Most birds left the area soon after ringing. Slika 4: Število (delež) ptic glede na dolžino postanka na obmo~ju raziskave. Ve~ina ptic je obmo~je zapustila kmalu po obročkanju. 25 P.H.W. Biedermann: Hidden leks in the Yellow-browed Warbler Phylloscopus inornatus? - Investigations from the Khan Khentey reserve (Mongolia) 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1 Date of ringing / Datum obročkanja Figure 5: The dependence of wing length of ringed males on ringing date (15 May 2003 to 1 Jun 2003). R2 is the variance of the regression line (= square of residuals). Diamonds (¦) represent N for the particular day. Individuals at the beginning of the season were not significantly larger than individuals at the end of the season. Slika 5: Odvisnost dolžine kril od datuma obro~kanja (15.5.2003 – 1.6.2003). R2 pomeni varianco regresijske premice. Diamanti (¦) predstavljajo {tevilo osebkov. Osebki na za~etku sezone niso bili signifikantno ve~ji od osebkov ob koncu sezone gnezdenja. males were ringed (see above) there were never more than three ringed males in the study area. On days with large numbers of singing males I usually caught new females. The last flock of males left the area on 20 Jun. The last bird was observed on 4 Jul. 4.4. Breeding biology By recording colour-ringed males it was possible to see which individuals established “real” (long-time) territories. 7 males (5 in Patch 1 and 2 in Patch 2 – 29.2% of all males) stayed for longer than one day in the area. Other males, which obviously did not mate, left the area within 10 days of being first sighted. 3 males mated (2 in patch 1, 1 in patch 2). Only one of the three ringed females was seen again after ringing. It was mated with one of the males in Patch 1. Tw o additional females were ringed in front of their nests, by non standard mist-netting. In total there were 4 breeding pairs (3 in Patch 1, 1 in Patch 2) in the area – one of the males was bigyn with 2 females. The 4 pairs observed during the study built at least 9 or 10 nests, which were all predated during the egg laying period. The first nest-building female was seen on 23 May (Figure 7). The last nest was left, probably after predation of eggs, on 4 Jul (by the same female). During that time four females laid a total of 15 - 16 eggs in at least 10 different nests (1st female: 4 nests, 2nd: 1, 3rd: 3, 4th: 2). Most nests were predated after the first egg was laid (N = 6), one nest with two, one with three and one with four eggs. After predation, nests were abandoned and the females started to built new nests one to ten days later (only once did a female lay an egg in an already predated nest). Nest building took females a minimum of 3 - 7 days, before laying the first egg. Eggs were laid on consecutive days. During the day, laying females were never seen at the nest or in its vicinity. One of the females probably never laid eggs – it was the second female of a male – and abandoned the completed nest (and the area) after five days of building. 4.5. Nest analysis Nine nests were collected after the birds left the area and analyzed for dimensions, nesting materials, i6 ACROCEPHALUS 2.J (128-H9): 21-3 J, 200Ć 25 n 20 15 10 OOOOOC15 cm), moss, short dry grass (< 5 cm), old leaves, dry horsetail Equisetum sp., bark (Salix sp.) and rotten wood (Betula platyphylla). The inner nests contained only three components: needles (Larix sibirica), animal hairs (horse, wild pig, deer etc.) and short dry grass (<15 cm). With the exception of animal hairs all materials of inner nests were found up to 10 m from nests. Nests were solely built by females. There was a tendency to mate with males that owned territories in the centre of a patch, i.e. in the centre of the other males’ territories. Mating and nest building always followed days with peak numbers of territorial males (see also Figure 6). During nest building the female called frequently – in 8 out of 14 approaches to the nest site with new nest material (mean of 3.9 calls). In 2 out of 15 departures (13.3%) she called again (mean of 1.5 calls). Thus, to locate new nests one should look out for calling birds, usually females, in occupied territories. All nine nests were found among Betula fusca shrubs (1.8 – 2 m height) in the transition zone between dense and more open shrubs, with an average distance of one to three metres between bushes. Half the nests were built in holes at the roots or under old branches directly at the base of Betula fusca bushes (one at the base of a Sibiran Larch) or not more than half a metre away. The other half were built in tussocks. In both cases only the entrance could be seen, the rest being covered by branches and mainly old grass (Figure 9). This camouflage was not constructed by the bird; the nest was simply built under it. Normally nests were directly on the ground. Tw o were 10 cm and 20 cm above the ground (in these cases the nesting site was very wet). Nests were always covered by overhanging branches. In most cases there was a landmark, like 27 P.H.W. Biedermann: Hidden leks in the Yellow-browed Warbler Phylloscopus inornatus? - Investigations from the Khan Khentey reserve (Mongolia) 5 i i f 3 .3 u u Nestl o -\