Breeding birds of shelterbelts near Sombor (NW Serbia) Gnezdilke drevoredov pri Somboru (SZ Srbija)
Thomas Oliver Mero
Milana Rakica 20, RS-25000 Sombor, Serbia, e-mail: thomas.oliver.mero@gmail.com
In 2008, breeding birds were studied in the shelterbelts surrounded by arable land NW of Sombor (Vojvodina, NW Serbia). Seven 500 m long sections of wooded shelterbelts were selected and surveyed for birds six times between May and July. Sampling plots differed in tree species composition and dominance, in height of trees, and presence and composition of shrub and herb layers. 22 breeding species were found with a mean density of 9.86 ± 1.47 pairs/km of shelterbelt. The mean number of species per shelterbelt was 8.3 ± 1.2. Species richness ranged from 4 (shelterbelt G) to 14 (shelterbelt D) species. Similarity, measured by the Sorensen index, showed strong qualitative similarity (over 70%) between the compared communities in only 19% of all possible pairs of shelterbelts. The most abundant breeding species were Golden Oriole Oriolus oriolus and Greenfinch Carduelis chloris, with overall linear densities of 2.9 and 2.6 pairs/km, respectively. Overall linear densities of a further four species (Red-backed Shrike Lanius collurio, Syrian Woodpecker Dendrocopos syriacus, Whitethroat Sylvia communis and Cuckoo Cuculus canorus) exceeded 1 pair/ km. The densities of breeding birds are compared to those obtained during other studies in Europe and elsewhere.
Key words: shelterbelts, agricultural landscape, breeding birds, linear density, Sombor, Vojvodina, Serbia
Klju~ne besede: drevoredi, kmetijska krajina, gnezdilke, linearna gostota, Sombor, Vojvodina, Srbija
1. Introduction
Shelterbelts are man-made habitats consisting mainly of rows of trees and shrub. They are excellent habitats for a variety of birds and other animals (Schroeder et al. 1992). A study in the United States showed that, of 108 bird species using shelterbelts in agricultural areas, 29 benefited substantially from their existence, 37 moderately and 42 only a little (Johnson & Beck 1988). The number of species recorded in old, diverse small wooded areas in farmland in Poland (avenues, shelterbelts and clumps of trees) was 68, but only 18 in young shelterbelts (Kujawa 1997 & 2004). Recently, 100 species were reported to be breeding in small wooded areas of this kind in Poland (Tryjanowski et al. 2009).
The density of breeding birds in shelterbelts and other wooded areas surrounded by arable land varies.
In South Dakota (United States), the mean density of birds in multi-row windbreaks was 82.6 individuals/ha, but only 48.8 in single-row windbreaks (Emmerich & Vohs 1982). In Minnesota (United States), the mean density of breeding birds in shelterbelts was 88.5 nests/ ha, with a range from 28.8 to 186.4 nests/ha (Yahner 1982). Research carried out in parks and wooded areas surrounded by arable land in Poland yielded breeding densities of 165.2 and 282 pairs/10 ha, respectively (Kujawa 1992).
The protection of European avifauna and wider environment depends on ecologically oriented land use. Intensive farming is recognized as a factor that affects very importantly the breeding birds of Europe (Tucker et al. 1994, Tucker & Evans 1997). Wooded areas surrounded by arable land are very important for the protection of biodiversity in agricultural landscapes (Ryszkowszki et al. 1999), and efforts
Table 1: Physical characteristics of the shelterbelts studied Tabela 1: Fizične značilnosti preučevanih drevoredov
Shelterbelt/ Drevored	Road next to the shelterbelt/ Cesta ob drevoredu	No. of tree rows/ St. nizov dreves	Orientation/ Usmerjenost	Width/ Sirina (m)	No. of tree and shrub species/ St. vrst dreves in grmov	Crops alongside the shelterbelt/ Vrsta pridelka vzdolž drevoreda
A	Dirt/ makadam	4	west — east/ zahod — vzhod	14	3	corn / koruza
B	Asphalt/ asfalt	4	west — east/ zahod — vzhod	14	2	wheat, sugar beet/ pšenica, sladkorna pesa
C	Dirt/ makadam	4	west — east/ zahod — vzhod	14	2	soya bean, sugar beet/ soja, sladkorna pesa
D	Asphalt/ asfalt	4	west — east/ zahod — vzhod	14	3	forest belt, sugar beet/ pas gozda, sladkorna pesa
E	Dirt/ makadam	4	west — east/ zahod — vzhod	15	4	corn, sugar beet/ koruza, sladkorna pesa
F	Asphalt/ asfalt	4	west — east/ zahod — vzhod	16	5	sugar beet / sladkorna pesa
G	Asphalt/ asfalt	3	south — north/ sever — jug	11	2	corn / koruza
......."' E..
F
B
G
A
Figure 1: Distribution of the 500 m long shelterbelt sampling plots in the study area (A-G; marked with lines)
Slika 1: Razporeditev 500 m dolgih vzorčnih ploskev drevoredov na območju raziskave (A-G; označeno z linijami)
should thus be made to conserve and renew them to preserve the birds. Wooded areas surrounded by arable land are characterized by higher species richness and higher population densities of birds in the breeding communities than other elements of agricultural landscape in the lowlands of Europe (Bezzel 1982).
The aim of this study was to investigate the breeding bird communities of shelterbelts near Sombor in NW
Serbia. The part of the northern province of Vojvodina we studied is a typical agricultural area with a dense shelterbelt network and many wooded areas surrounded by arable land. Here we present the first findings on the abundance and density of breeding birds and the similarity of bird communities in shelterbelts in Serbia.
2. Study area and methods
2.1.	Study area
The research was conducted in the shelterbelts NW of Sombor (UTM CR47 & 48, NW Serbia). Seven shelterbelts were selected from the whole shelterbelt network in the area bounded by the settlements of Gakovo, Bezdan and Sombor. The 500 m long sampling plots were selected according to their characteristics (Figure 1, Table 1).
The age of these shelterbelts is about 20 years. There were no gaps between individual trees greater than 10 m. Sampling plots differed in tree species composition and their dominance, height of trees and presence and composition of shrub and herb layers (Table 2).
2.2.	Methods
All sampling plots were surveyed six times between 26 May and 19 Jul 2008 in the early morning,
Table 2: Description of vegetation characteristics of the shelterbelts studied Tabela 2: Opis značilnosti vegetacije preučevanih drevoredov
Trees I Drevesa
Shrubs I Grmi
Herbs I Zeli
A
ShelterbeltI CoverI
Drevored Pokrovnost
(%)
Species and
1 HeightI dominance1I
,r .	Višina
Vrste in
dominanca1
90
80
90
80
90
lila (3), 8-12
Robinia pseudacacia (4)
lila (2), 8-13 Robinia pseudacacia (4)
Morus alba (2),
Robinia pseudacacia (5)
Morus alba (1), Ulmuspumila (+), Robinia pseudacacia (5)
Populus alba (4), Robinia pseudacacia (3)
Populus alba (4), Robinia pseudacacia (3)
Gleditschia triacanthos (3), Robinia pseudacacia (4)
7-10
6-10
8-18
7-20
7-8
Cover/	Height/ Dominant
Pokrovnost Species / Vrste Višina	species/
(%)	(m) Dominantne vrste
30
Rosa canina, Rubus caesius
0.5-2
Sambucus nigra,
Robinia pseudacacia,
Rosa canina
Sambucus nigra,
Robinia pseudacacia,
Rosa canina, Cornus sanguinea
1-3
1-3
Bromus sterilis, Calamagrostis sp., Elymus sp., Arrhenatherum sp. Bromus sterilis
Bromus sterilis, Elymus sp., Arrhenatherum sp. Bromus sterilis
Bromus sterilis, Calamagrostis sp.
Bromus sterilis, Arrhenatherum sp.
Bromus sterilis, Arrhenatherum sp.
< 10
B
C
D
E
20
F
G
1 dominance classes I dominancni razredi: (+) 0.1-1%, (1) 1.1-5%, (2) 5.1-25%, (3) 25.1-50%, (4) 50.1-75%, (5) 75.1-100% (Braun-Blanquet 1994)
approximately once every ten days. The survey of a single plot lasted for about 20-40 min, depending on the number of species. Attempts were made to observe and record every breeding bird. If no nest or nesting behaviour was observed, a localised bird was interpreted as a breeding territory only when male territorial behaviour (singing) was registered at least twice during the study period. Individuals not meeting the above criteria were excluded from the study. The percentage cover of a given vegetation layer
was estimated by the basic method used for general coenological surveying (Braun-Blanquet 1994).
Since the breeding birds' territories were distributed along the shelterbelt, their density is expressed as a number of pairs per km of shelterbelt (herein referred to as linear density). Breeding density was analysed and presented based on total data from all shelterbelts.
Statistical analysis was performed using the software Biostat 2008 (AnalystSoft 2008). The similarity of bird communities between all possible pairs of
Table 3: Number of pairs, overall linear density and dominance of breeding birds in the shelterbelts (total length = 3.5 km) Tabela 3: Število parov, skupna linearna gostota in dominance gnezdilk drevoredov (skupna dolžina = 3,5 km)
Shelterbelt - No. of pairs/ Drevored — št. parov
Total / Skupaj
Species / Vrsta	A	B	C	D	E	F	G	No. of pairs/ St. parov	Linear density/ Linearna gostota (p/km)	Dominance/ Dominanca (%)
Carduelis chloris	1	4	2	2			1	10	2.9	14.5
Oriolus oriolus	1	1	1	2	2	2		9	2.6	13.0
Lanius collurio			1	1	2	1	1	6	1.7	8.7
Dendrocopos syriacus	1	1	1	1	1	1		6	1.7	8.7
Sylvia communis		1	2	2		1		6	1.7	8.7
Cuculus canorus	1		1	1		1		4	1.1	5.8
Sylvia atricapilla	1	1				1		3	0.9	4.3
Carduelis carduelis		1				1	1	3	0.9	4.3
Luscinia megarhynchos				1	1		1	3	0.9	4.3
Falco tinnunculus	1				1			2	0.6	2.9
Corvus cornix	1		1					2	0.6	2.9
Pica pica				1		1		2	0.6	2.9
Picus viridis				1	1			2	0.6	2.9
Parus major				1	1			2	0.6	2.9
Fringilla coelebs	1			1				2	0.6	2.9
Streptopelia decaocto				1				1	0.3	1.4
Passer montanus				1				1	0.3	1.4
Sturnus vulgaris					1			1	0.3	1.4
Muscicapa striata			1					1	0.3	1.4
Turdus merula				1				1	0.3	1.4
Coccothraustes coccothraustes						1		1	0.3	1.4
Motacilla alba			1					1	0.3	1.4
Total / Skupaj (22 species / vrst)	8	9	11	17	10	10	4	69	19.7	100.0
the studied shelterbelts, Cs, was calculated by the Sorensen index (Podani 1997, Majer 2004):
cs =
2c a + b
100
(a = the number of species in one community; b = the number of species in the other community; c = the number of species common to both communities)
The diversity of each shelterbelt was calculated by the Shannon-Wiener (H) and Simpson diversity indexes (D) (Colinvaux 1973):
//=-£«,.log«,-
m	N(N-l)
(S = the number of species in the sample, N = the total number of individuals in the sample; n. = the number of individuals of every species in the sample)
3. Results
69 breeding pairs of 22 bird species were found in the seven shelterbelt sampling plots. The most abundant breeding species were Greenfinch Carduelis chloris and Golden Oriole Oriolus oriolus, with total linear densities of 2.9 and 2.6 pairs/km, respectively. The total linear densities of a further four species exceeded 1 pair/km. Seven species bred in densities lower than 0.5 pairs/km, while other species bred in densities between 0.5 and 1 pair/km (Table 3).
The number of breeding bird species per shelterbelt varied greatly — from four (shelterbelt G)
15
1
» 12 >3 ^ 9
IB J «
Ï6 (A
*3 o
m
b c d e f Shelterbelt / Drevored
25 s a «8 — 20 Sg
15 "t I
_ in the occupied shelterbelts ranged from 2.0 to 4.0 30 J ^ pairs and 2.0 to 8.0 pairs/km, respectively. — *» ™ In most cases, the diversity of the breeding bird species in the shelterbelts was similar. The highest value was obtained for shelterbelt D and the lowest values for shelterbelts B and G (Table 4).
52.4% of the Serensen index values were over 50%, indicating a high qualitative similarity between the breeding bird communities in different shelterbelts, while of all compared communities only 19% indicated a strong (over 70%) similarity (Table 5).
00
S M
V Q>
Figure 2: Number of species (bars) and overall linear density (dots) of breeding birds in the shelterbelts
Slika 2: Stevilo vrst (stolpci) in skupna linearna gostota (tocke) gnezdilk v drevoredih
to 14 (shelterbelt D) (Figure 2). The mean number of species per shelterbelt (± SE) was 8.3 ± 1.2, and the mean density of breeding pairs per shelterbelt reached 9.86 ± 1.47 pairs/km. Golden Oriole was present in all shelterbelts except G. Greenfinch breeding pairs were absent from shelterbelts E and F. Their linear densities
4. Discussion
22 breeding bird species were found in shelterbelts during this study. This finding is in accordance with research in lowland Poland that revealed 18 species in young shelterbelts (Kujawa 2004), 50 in parks and 44 in wooded areas surrounded by arable land (Kujawa 1992). Another study from Poland demonstrates that, between 1964 and 1994, the number of breeding species of small mid-field woodlots increased from 44 in the 1960s to 51 in the 1990s (Kujawa 2002). In
Table 4: The diversity of breeding bird species at the shelterbelts studied, calculated by the Shannon-Wiener and Simpson diversity indexes
Tabela 4: Vrstna diverziteta gnezdilk preučevanih drevoredov, izračunana na podlagi Shannon-Wienerjevega in Simpsonovega indeksa
Shelterbelt / Drevored	A	B
Shannon-Wiener (H)	2.08	1.58
Simpson (1 - D)	0.88	0.74
C	D	E	F	G
2.15	2.59	2.03	2.16	1.39
0.88	0.92	0.86	0.88	0.75
Table 5: Similarities of breeding bird communities between all possible pairs of shelterbelts studied, calculated by the S0rensen index (Cs)
Tabela 5: Vrstna identičnost združb gnezdilk med vsemi možnimi pari preučevanih drevoredov, izračunana na podlagi S0rensenovega indeksa (Cs)
Compared shelterbelts/ Primerjan par drevoredov
CS value/
vrednost (%)
Compared shelterbelts/ Primerjan par drevoredov
CS value/
vrednost (%)
Compared shelterbelts/ Primerjan par drevoredov
CS value/
vrednost (%)
A vs. B	57.1	B vs. D	30.0	C vs. G	46.2
A vs. C	58.8	B vs. E	57.1	D vs. E	45.5
A vs. D	45.5	B vs. F	80.0	D vs. F	43.5
A vs. E	62.5	B vs. G	40.0	D vs. G	33.3
A vs. F	70.6	C vs. D	52.1	E vs. F	35.3
A vs. G	33.3	C vs. E	70.6	E vs. G	50.0
B vs. C	66.7	C vs. F	77.8	F vs. G	46.2
considering the smaller numbers in the present study, it should be pointed out that early breeding birds like the Great Tit Parus major, Starling Sturnus vulgaris or Hooded Crow Corvus cornix are probably under-represented in this study owing to the late start of the fieldwork. However, because of the relatively young age of these shelterbelts, the probability of finding significantly more pairs of these early breeding species is low.
The reason why shelterbelt G had markedly smaller number of bird species probably lies in its younger tree stand and different composition of tree species. Shelterbelt G contained Honeylocust Gleditschia triacanthos, a tree species disliked by most breeding species because of its very bushy and dense growth with long thorns. However, a habitat like this is a good nesting place for the Red-backed Shrike Lanius collurio, which requires large bush or shrub vegetation growing in warm and dry areas for nesting (Fuisz et al. 1998), with high insect richness (Jakober & Stauber 1987). Indeed, this was one of only four species registered in shelterbelt G. On the other hand, shelterbelt D adjoined a larger wooded area with older trees, which was probably the main factor, not only for the much higer number of bird species, but also for the presence of several hole-nesting birds species that were generally sparsely distributed in the studied shelterbelts.
The linear density of the most abundant species in our study (Greenfinch with 2.9 pairs/km) was similar to that of Corn Bunting Miliaria calandra in young shelterbelts in Poland (2.78 pairs/km), the most numerous species in the study of Kujawa (2004). A comparison of these results with those from mature shelterbelts revealed much higher densities in the latter, where the density of the most abundant species was 22.8 pairs/10 ha, while in parks the mean density of the most abundant species, the Starling, was 33 pairs/10 ha (Kujawa 1992 & 1997). In Slovakian shelterbelts, the density of Nightingale Luscinia megarhynchos varied between 2.44 and 3.55 pairs/ha (Nemethova 2007).
The generally high qualitative similarity of the compared communities confirmed that the breeding species composition variability between individual shelterbelts was low. However, by comparing the results of this study to others, we concluded that the similarities between the compared communities were lower. Indeed, the high similarity of the communities might be the result of a fairly similar vegetation structure in our samples (plant species and vegetation layers in shelterbelts). In Poland, a study that compared wood belts with tree clumps resulted in a qualitative similarity of 79%, whereas the comparison of wood
belts with avenues and tree clumps with avenues yielded the similarities of 70% and 59%, respectively (Kujawa 1997). In yet another study, Kujawa (1992) gives the similarity of 70% with the breeding communities of parks and systems of wooded areas compared.
The linear densities of the two most abundant species, Golden Oriole and Greenfinch, are generally higher than in studies from Poland (Kujawa 1992, 1997). In the case of Golden Oriole, this can be explained by different latitudes of the two countries' study areas. Populations of the Golden Oriole show an increase in densities towards the southern and south-eastern regions of Europe (Wassmann 1997).
Many studies conclude that shelterbelts and other types of wooded areas situated in large expanses of arable land are important habitats for a number of breeding birds. On the other hand, shelterbelts and similar plantations could also negatively affect the birds of open country (e.g. Avery 1989, Pierce et al. 2001). In the Great Plains (United States), bird species exhibiting significant population increases were primarily associated with human-made structures and woody vegetation, while those exhibiting significant decline were primarily grassland and wetland breeding birds (Igl & Johnson 1997). Similar trends have been observed in common birds throughout Europe (e.g. pecbms 2007). In this study, the relatively high densities of several woodland birds in shelterbelts indicate the importance of small wooded areas surrounded by large areas of unsuitable habitat as refuges for at least some woodland breeding birds.
Acknowledgements: Many thanks to Jelena Rakovic and Marko Tucakov for checking the manuscript. I am also grateful to Szabolcs Lengyel (HU) for the final check of the manuscript.
5. Povzetek
Leta 2008 je avtor članka preučeval gnezdilke v drevoredih, obdanih z orno zemljo, SZ od Sombora (Vojvodina, SZ Srbija). Za šestkratno popisovanje med majem in julijem je izbral sedem 500 m dolgih odsekov drevoredov. Vzorčne ploskve so se razlikovale po sestavi drevesnih vrst in njihovi dominanci, višini dreves ter prisotnosti in sestavi grmovnega in zeliščnega sloja. Zabeležil je 22 gnezdečih ptičjih vrst s povprečno gostoto 9,86 ± 1,47 para/km drevoreda. Povprečno število vrst na drevored je bilo 8,3 ± 1,2, medtem ko se je število vrst gibalo med 4 (drevored G) in 14 (drevored 14) vrstami. Vrstna identičnost, izmerjena s Sorensonovim indeksom, je le v 19% vseh možnih
parov preučevanih drevoredov pokazala veliko vrstno podobnost (prek 70%) med primerjanimi združbami gnezdilk. Najbolj številčni vrsti sta bili kobilar Oriolus oriolus in zelenec Carduelis chloris s skupno linearno gostoto 2,9 oz. 2,6 para/km. Skupna linearna gostota naslednjih štirih vrst (rjavi srakoper Lanius collurio, sirijski detel Dendrocopos syriacus, rjava penica Sylvia communis in kukavica Cuculus canorus) je presegala 1 par/km. Gostoto gnezdilk avtor primerja z gostotami, ugotovljenimi v drugih študijah v Evropi in drugod.
6. References
AnalystSoft (2008): Biostat. Version 5.2.5.0. - Analystsoft
Inc., Vancouver. Avery, M.I. (1989): Effects of Upland Afforestation on Some Birds of the Adjacent Moorlands. - Journal of Applied Ecology 26 (3): 957-966. Bezzel, E. (1982): Vögel in der Kulturlandschaft. - Ulmer, Stuttgart.
Braun-Blanquet, J. (1994): Pflanzensoziologie, Grundzüge der Vegetationskunde. 3. Auflage - Springer Verlag, Wien. Colinvaux, P. (1973): Introduction to ecology. - John Wiley
& Sons, New York. Emmerich, J.M. & Vohs, P.A. (1982): Comparative use of four woodland habitats by birds. - Journal of Wildlife
Management 46: 43-49. Fuisz, T.I., Mosket, C. & Park, J.Y. (1998): Nest site selection and habitat use in Red-backed Shrike Lanius collurio in Hungary. pp. 30-33 In: Yosef, R. & Lohrer, F.E. (eds.): Shrikes of the World - II: Conservation Implementation. Proceedings of the Second International Shrike Symposium. - International Birdwatching Center in Eilat, Eilat.
Igl, L.D. & Johnson, D.H. (1997) Changes in breeding bird populations in North Dakota: 1967 to 1992/93. -Auk 114 (1): 74-92. Jakober, H. & Stauber, W. (1987): Habitatansprüche des Neuntöters (Lanius collurio) und Massnahmen für seinen Schutz. - Beihefte zu den Veröffentlichungen für Naturschutz und Landschaftspflege in BadenWürttemberg 48: 25-53. Johnson, R.J. & Beck, M.M. (1988): Influences ofshelterbelts on wildlife management and biology. - Agriculture, Ecosystems & Environment 22/23: 301-335. Kujawa, K. (1992): [Role of manorial parks for preservation of avifauna richness in agricultural landscape]. pp. 137-150 In: Balazy, S., Ryszkowski, L. (eds.): [Primary production, animal resources and leaching of organic matter in an agricultural landscape]. - ZBSRiL PAN, Poznan. (in Polish) Kujawa, K. (1997): Relationships between the structure of mid-field woods and their breeding bird communities. -Acta Ornithologica 32 (2): 175-184.
Kujawa, K. (2002): Population density and species composition changes for breeding bird species in farmland woodlots in western Poland between 1964 and 1994. - Agriculture, Ecosystem & Environment 91: 261-271.
Kujawa, K. (2004): Importance of young shelterbelts for breeding avifauna in agricultural landscape (Turew area, West Poland). - Polish Journal of Ecology 52 (4): 433-443.
Majer, J. (2004): [Introduction to ecology]. - Dialog Campus Kiado, Budapest-Pecs. (in Hungarian)
Nemethova, D. (2007): Density-dependent habitat selection in Nightingale (Luscinia megarhynchos C. L. Brehm) in selected windbreaks of SW Slovakia. — Ekologia Bratislava 26: 313-321.
PECBMS (2007): State of Europe's Common Birds 2007. -CSO / RSPB, Prague.
Pierce, R.A., Farrand, D.T. & Kurtz, W.B. (2001): Projecting the bird community response resulting from the adoption of shelterbelt agroforestry practices in Eastern Nebraska. - Agroforestry Systems 53: 333-350.
Podani, J. (1997): [Introduction to the analysis of multivariate biological data]. - Scientia Kiado, Budapest. (in Hungarian)
Ryszkowski, L., Karg, J. & Kujawa, K. (1999): [Preserving and managing biodiversity in an agricultural landscape]. pp. 59-80 In: Ryszkowszki, L. & Balazy, S (eds.): [Conditions of preserving biodiversity and landscape diversity]. - ZBSRiL PAN, Poznan. (in Polish)
Schroeder, R.L., Cable, T.T. & Haire, S.L. (1992): Wildlife species richness in shelterbelts: Test of a habitat model. -
Wildlife Society Bulletin 20: 264-273.
Tryjanowski, P., Ku^niak, S., Kujawa, K., & Jerzak, L. (2009): [Ecology of farmland birds]. - Bogucki Wydawnictwo Naukowe, Poznan. (in Polish)
Tucker, G.M., Heath, M.F., Tomialojc, L. & Grimmett, R.F.A. (1994): Birds in Europe: Their Conservation Status. - BirdLife Conservation Series No. 3. BirdLife International, Cambridge.
Tucker, G.M. & Evans, M.I. (1997): Habitat for birds in Europe: A Conservation Strategy for the Wider Environment. BirdLife Conservation Series No. 6. -BirdLife International, Cambridge.
Wassmann, R. (1997): Golden Oriole Oriolus oriolus. pp. 658-659 In: Hagemeijer, W.J.M. & Blair M.J. (eds.): The EBCC Atlas of European Breeding Birds. Their Distribution and Abundance. - T & A D Poyser, London.
Yahner, R.R. (1982): Avian nest densities and nest-site selection in farmstead shelterbelts. - Wilson Bulletin 94: 156-175.
Arrived / Prispelo: 28.5.2009
Accepted / Sprejeto: 15.10.2010