Acrocephalus 28 (134): 95-104, 2007 Study on the habitat selection by birds in mature and over-mature Macedonian Pine Pinus peuce forests in Pirin National Park (SW Bulgaria) Ptice in njihov izbor habitata v optimalni in terminalni fazi odraslega gozda molike Pinus peuce v Narodnem parku Pirin (JZ Bolgarija) Stoyan Ch. Nikolov Central Laboratory of General Ecology, Bulgarian Academy of Sciences, 2 Gagarin Str., BG–1113 Sofa, Bulgaria, e–mail: nikolov100yan@abv.bg Habitat selection by birds in mature (60–100 years old) and over-mature (>120 years old) Macedonian Pine Pinus peuce forests was studied over 3 years (2004–2006) in Pirin National Park (SW Bulgaria). Overall, 185 study plots were randomly located and at each study plot habitat characteristics were described within a radius of 25 m. Birds were counted twice per year (June–July) within a radius of 50 m using the distance-sampling point-count technique. There was almost no difference in bird diversity between the studied forest age classes. Five species (Wren Troglodytes troglodytes, Chiffchaff Phylloscopus collybita, Coal Tit Parus ater, Nuthatch Sitta europaea and Treecreeper Certhia familiaris) showed preferences to over-mature Macedonian Pine forests and they had higher breeding densities in over-mature than in mature forests. Only one species (Mistle Thrush Turdus viscivorus) was found to prefer mature forests. The important criteria in habitat selection of these 6 species were analyzed, using a forward stepwise Multiple Regression Analysis. From the species that preferred over-mature forest, 2 species were trunk and bark-feeders and 4 species were hollow nesters. The paper stresses the importance of the amount of dead wood and decaying trees for forest bird communities and suggests that removal of dead wood from old forests, especially those in protected areas, should be minimized. Key words: forest birds, habitat selection, distance sampling, point transects, Macedonian Pine, Pinus peuce, Pirin National Park Klju~ne besede: gozdne ptice, izbor habitata, to~kovni transekti, molika, Pinus peuce, Narodni park Pirin 1. Introduction characteristics have a greater infuence on biological diversity, particularly in Mediterranean forests In the context of global reduction and degradation (Gil-Tena et al. 2007). Therefore, better understanding of natural habitats, many forests are undervalued and of relationships between forest birds and their habitat clear-cut. The protected areas have an essential role in protected areas is crucial to the implementation to play in countering this process. Historically, they of sustainable, ecosystem based forest management were only concerned with protection, but now there (Kirk & Hobson 2001). It is well known that avian is also a need to focus on conservation, sustainable use communities differ greatly in early succession forests and ecological restoration (Bennett 2003). However, and in older forest stands (Helle & Mφnkkφnen there is still a lack of knowledge on which forest 1990, Conner & Dickson 1997, Kirk & Hobson 95 S. Ch. Nikolov: Study on the habitat selection by birds in mature and over-mature Macedonian Pine Pinus peuce forests in Pirin National Park (SW Bulgaria) 2001). While the early succession bird species, majority of which are feld and shrub-nesting species, decline in abundance as a response to vegetation growth, bird species associated with older forests age classes begin to increase in number and abundance (Dickson et al. 1993). But still little is known on the differences between bird communities in mature and over-mature forest age classes (Davis et al. 1999, Pearce & Venier 2005). In this paper we present the results from a short-term study on habitat selection by birds on local scale in mature and over-mature Macedonian Pine Pinus peuce forests in Bulgaria. 2. Studied habitat and area The Macedonian Pine is a tertiary relict belonging to primeval quasi-boreal mesophyte vegetation, distributed endemically in Serbia, Montenegro, Macedonia, Albania, Greece and Bulgaria (Tutin et al. 1993). In Bulgaria, it can be found mainly in Pirin Figure 1: Location of Pirin National Park in Bulgaria and localities of studied areas within the park: (1) Bela Reka valley; (2) Bunderishka River valley; (3) area of Shiligarnika; (4) Demjanishka River valley; (5) Bezbojka River valley; (6) Demirkapyiska River valley; (7) Bashmandrevska River valley; (8) Kelyova River valley; and (9) Sinanishka River valley. Slika 1: Lokacija narodnega parka Pirin v Bolgariji in lokacije raziskovanih obmo~ij znotraj parka: (1) dolina Bele reke; (2) dolina reke Bunderishka; (3) obmo~je Shiligarnika; (4) dolina reke Demjanishka; (5) dolina reke Bezbojka; (6) dolina reke Demirkapyiska; (7) dolina reke Bashmandrevska; (8) dolina reke Kelyova; (9) dolina reke Sinanishka. and Rila mountains (covering totally 12,000 ha), where it normally grows between 1700 and 2000 m a.s.l. (exceptionally 1200-2200 m a.s.l.) on rocky acid soils of silicate grounds (Iordanov 1963, Bondev 1991). It comprises up to (approximately) 30 m of comparatively light forest stands (Velchev 2002). This study was carried out in 9 localities within Pirin National Park (Figure 1), south-western Bulgaria (41°40' N, 23°30' E), which is recognized as a UNESCO World Heritage Site, Important Bird Area (code EG055) and Natura 2000 site (code BG0002056). It holds a signifcant part (42%) of the Macedonian Pine forest in Bulgaria, representing 5,416 ha (23.4% of the forest cover in the park) – 95% of which is of native origin (Petrov 2003). The study area has a moderate continental climate and falls into a mountainous climatic sub region with average annual temperatures of 2–3oC and annual amplitude about 17oC. Summers are cool (mean temperature 3–5oC) and winters are cold (mean January temperatures from –2oC to –5oC) with snow cover (60–160 cm deep) presence for 120–160 days annually (Koleva 2003). 3. Methods 3.1. Bird sampling design Field work was carried out in June and July for a total of 59 days, over three successive years (2004–2006). Overall, 185 study plots were located randomly using ArcMap 8.3 software (ESRI 2000) at 1,700–2,230 m elevation in two age classes of Macedonian Pine forests: mature, 60–100 years old (n = 95) and over-mature, >120 years old (n = 90). Minimal centre distance of adjacent study plots was 180 m (Bibby et al. 1998). The exact location of study plots was established in the feld by means of Global Positioning System (GPS) Garmin 60CS. Bird sampling was carried out using the distance-sampling point-count technique (Bibby et al. 1992, Buckland et al. 1993). At each study plot, birds seen and / or heard were counted within a radius of 50 m. A breeding pair was chosen for a counting unit, as described in Nikolov & Spasov (2005). Birds in fight were not counted. The counts were made between 6.00 h and 11.00 h local time. There was a setting down period of 2 min and the duration of a count was 5 min. Two counts per point transect were made each year, with a 10–40 day (on average 21) interval between the two successive visits. Most of the fundamental assumptions in distance-sampling methods (Bibby et al. 1992, Buckland et al. 1993) were met. All observations were carried out by the <)6 Acrocephalus 28 (134): 95-104, 2007 same person (author of this paper), in the same daytime period, in similar weather conditions and within similar habitat types. This minimizes the interactions of detection probability with the observer, habitat, weather and day-time. Although the bird counts were carried out in mornings during the breeding season, when the vocal activity of birds is higher, we should take into consideration that in forested habitats some of the birds situated directly overhead the observer could be missed (Bδchler & Liechti 2007). As the assumption g(0) = 1 cannot be guaranteed in all counts, only the relative bird density is discussed further in the paper. The bird sampling method used in this study does not consider nocturnal birds (Owls and Nightjars) and was not applicable to species with very low abundance in the studied habitat (e.g. raptors, grouse, and some species of Woodpeckers). 3.2. Habitat sampling design At each study plot, several physical characteristics of habitat and structural parameters of vegetation were measured. Elevation and exposition were measured using the GPS Garmin 60CS at the centre. Tree measurements were made for the closest 10 trees (with diameter at breast height >0.2 m) to the centre of the study plot. Tree diameter at breast height (DBH) and a distance of the 10th tree to the centre of study plot was measured by means of callipers and tape respectively. Foregoing measured parameters were used for the calculation of 2 derived parameters: density of trees (DT) and index of basal area of trees (IBAT). The rest of habitat parameters were estimated visually within a radius of 25 m from the centre of the plot: cover of rocks and piles of stones; vegetation layers; tree canopy; number of dead trees (with diameter at breast height >0.2 m) and stumps (higher than 0.5 m). The averages of the collected vegetation physiognomy data was used to represent a study plot in the statistical analysis. The studied forests were with up to 10% mixture of other than Macedonian Pine tree species: Norway Spruce Picea abies (L.) Karst., Scots Pinus silvestris L. and Bosnian Pines Pinus heldreichii Christ, Silver Fir Abies alba Mill., and Rowan Sorbus aucuparia L. Within these forests, 6 foliage profles were distinguished: (1) Ground level (0.1–0.4 m high), composed of Bilberry Vaccinium myrtillus L., herbs and Wild Geranium Geranium spp.; (2) Low level (0.4–1 m high), composed of Hellebore Veratrum spp., thistles and ferns; (3) Small shrubs (0.4–1 m high), composed of Siberian Juniper Juniperus sibirica Burgsd., Balkan Broom Chamaecytisus absinthioides (Janka) Kuzm. and Raspberry Rubus idaeus L.; (4) Mid-level (1.6–4 m high), composed of Mountain Dwarf Pine Pinus mugo Turra., and saplings; (5) Sub-canopy level (4–7 m high), composed of short trees; (6) Canopy level, composed of tall trees (9–30 m high). The habitat sampling method applied was based on estimation rather than actual measurement, but it is believed that such semi-quantitative methods have suffcient accuracy and effciency for characterizing vegetation profles (Braun-Blanquet 1932, Moore et al. 1970, Mueller-Dombois & Ellenberg 1974, Barbour et al. 1980) and are effective enough in analyzing avian-habitat relationships (Sabo 1980, Moskαt & Waliczky 1992). 3.3. Data analysis Species diversity was described using the reciprocal Simpson’s index N2 (Hill 1973), as species diversity calculated in this manner is highly dependent upon the most abundant species (Krebs 1999) who are the focus of this study. Habitat preferences of birds were tested using the G-test (Manly et al. 1993, Krebs 1999), which is an improved version of the Utilization-availability analysis (Neu et al. 1974). The technique comprises the calculation of the Selection Index (Cock 1978), which was used for estimating whether the species occurs more or less frequently than expected in each of the studied habitat categories. The signifcance of the Selection Index was analysed using the chi-square test (Krebs 1999) with the null hypothesis that species have no preferences to the studied habitat types. The Standardized Selection Index (Manly et al. 1993) was calculated for comparison between the habitat preferences of different species. Because of the constraint of Dixon & Massey (1969) that no more than 20% of all categories should contain less than 5 expected observations, only species with more than 10 observations were included in the analysis (n = 22 species). The sample sizes of all species used in the analysis were suffciently large because the conservative rule of Hayes & Winkler (1970) that np and n(1 - p) ? 5, where n is the number of observed birds and p is the proportion of observed birds in the studied habitat category, was respected. The relative bird densities were computed using Distance 5.0 Release 2 software (Thomas et al. 2006). Conventional distance sampling analysis was applied, considering that the study was limited to only one habitat where the detection probability was similar, and consequently was solely a function of distance from the observation point. The model of uniform key function with simple polynomial series expansion 97 S. Ch. Nikolov: Study on the habitat selection by birds in mature and over-mature Macedonian Pine Pinus peuce forests in Pirin National Park (SW Bulgaria) was selected on the basis of a minimum value selection (z = 5.84, P < 0.001), but trees were higher in over-of Akaike’s Information Criterion (Akaike 1973). mature forests (z = 2.92, P < 0.01). Although the Coal The study plots were grouped in clusters taking Tit abundance was negatively correlated with the tree into account the Central Limit Theorem, with the density and positively correlated with the IBAT (Table aim to achieve normal distribution of data (Fowler & 5), the main factor related to the species preference to Cohen 1992). A single cluster was composed by study over-mature forests was DT. It was higher in mature plots in forests with similar age, pattern and exposition forests (z = 4.48, P < 0.01), whereas the difference in and it was represented by average values of study IBAT was not signifcant between the studied forest plots’ measured/estimated parameters. Altogether, 48 age classes. According to the MRA, the important clusters (24 in mature and 24 in over-mature forests) criteria in habitat selection for Wren, Nuthatch and were made. Data were tested for normality using the Treecreeper was the amount of dead wood (Table Shapiro-Wilk’s Test (Shapiro & Wilk 1965). Bird data 5). The number of stumps, dead and fallen trees was were additionally square root transformed to achieve almost twice higher (z = 5.18, P < 0.001) in over-a normal distribution (Fowler & Cohen 1992). mature (mean = 5.78 ± 1.84) than in mature (mean Habitat characteristics were compared by a z test = 3.14 ± 1.69) Macedonian pine forests. Additional (Fowler & Cohen 1992). Forward stepwise Multiple factors for Nuthatch and Wren were the tree height Regression Analysis (MRA) was performed for bird- and the rocks and stone piles cover respectively. The habitat relationships (Jongman et al. 1997). Analyses latter was present to a greater extent in over-mature were computed with Statistica 7.0 software (StatSoft forests (z = 1.90, P < 0.05). The abundance of Mistle 2004). Averages and their standard deviations are Thrush was positively correlated with the cover of reported. ground level vegetation profle and DT and negatively correlated with the canopy cover, but only the last two 4. Results factors played a role in the habitat selection of species, since there was no signifcant difference in the cover There was almost no difference in bird diversity of ground level vegetation profle between mature and between the studied forest age classes (in over-mature over-mature Macedonian Pine forests (Table 5; z = N2 = 10.09 and in mature forests N2 = 10.45). Five 1.14, P > 0.05). species were found in only one of the studied habitat types. Four among them (Firecrest Regulus ignicapillus, 5. Discussion Redstart Phoenicurus phoenicurus, Gray Wagtail Motacilla cinerea and Black Woodpecker Dryocopus The species found to prefer over-mature forest in martius) were with very low abundance in the studied Pirin National Park are more or less related to the old habitat (n < 3 observations). Therefore the registration forest throughout their range in Palearctic (Cramp of these species in only one forest age class may be 1998). So far, we have had no quantitative studies on due to fortuity. The Greenfnch Carduelis chloris (n habitat selection of forest Passerines in Bulgaria, hence = 7 observations) was found in mature forests only a comparison of the presented results with previous and it is very probable that the species prefers this works at the national level is impossible. forest age class. According to the G-test, 6 species Although the Wren inhabits different types of showed preferences to one of the studied age classes of habitats in its Holarctic range, there are several factors Macedonian pine forests in Pirin National Park (Tables that play an important role in habitat selection of 1 & 2). Only one species (Mistle Thrush Turdus species: the humidity, the presence of dense vegetation viscivorus) showed preference to mature Macedonian at the ground level and the vertical structures as Pine forests, while the rest fve species (Wren Troglodytes stumps, stones and fallen branches (Hagemeijer troglodytes, Chiffchaff Phylloscopus collybita, Coal Tit & Blair 1997, Iankov in press). The results from Parus ater, Nuthatch Sitta europaea and Treecreeper the present study confrm the importance of the Certhia familiaris) preferred over-mature forests and forementioned factors in habitat selection of Wren. therefore had higher breeding densities in this habitat Probably this relation is due to the availability of more type (Table 3). In the forementioned fve species, sites for nesting and refuge (Cramp 1998). So far in the degrees of preference to over-mature forests were Bulgaria, Wren has been found to prefer older forests similar (Table 4). Chiffchaff was negatively correlated in Scots Pine woodlands (Nankinov 1997). Higher with the canopy cover and to a lesser extent with density of Wren in old forests was registered also in the tree height (Table 5). The canopy was closer in Britain (Fuller 1995). mature than in over-mature Macedonian Pine forests 98 Acrocephalus 28 (134): 95-IO4, 2OO7 Table 1. Selection indices of birds in mature (60–100 years) and over-mature (above 120 years) Macedonian Pine Pinus peuce forests during the breeding seasons of 2004–2006 in Pirin National Park, Bulgaria. Abbreviations: n – number of breeding pairs observed; o? & om – observed proportions of birds in over-mature and mature forests respectively; p? & pm – expected proportions of birds in over-mature and mature forests respectively; w? & wm – selection indices of birds in overmature and mature forests respectively; ns – P > 0.05. Tabela 1: Selekcijski indeksi ptic v optimalni (60–100 let) oz. terminalni fazi (nad 120 let) odraslih gozdov molike Pinus peuce v gnezditvenem obdobju 2004–2006 v bolgarskem Narodnem parku Pirin. Okraj{ave: n – {tevilo opa`enih gnezde~ih parov; o? & om – zabele`eni dele`i ptic v terminalni oz. optimalni fazi; p? & pm – pri~akovani dele`i ptic v terminalni oz. optimalni fazi; w? & wm – selekcijski indeksi ptic v terminalni oz. optimalni fazi; ns – P > 0.05. Species / Vrsta n Proportion / Dele` Selection index/ Selekcijski indeks x2 0C 0M wc wm P (p = 0.5) 0.53 (p = 0.5) 0.47 0.18 (df = 1) Dendrocopos major 48 1.06 0.94 m Anthus trivialis 16 0.50 0.50 1.00 1.00 0.00 m Troglodytes troglodytes** 124 0.60 0.40 1.19 0.81 4.60 I20 II.O 1.24-1.90 60-100 I.I 13.0 0.87-1.46 Turdus viscivorus >I20 0.9 37.1 0.4I-I.75 60-100 1.4 28.9 0.81-2.52 Phylloscopus collybita >I20 2.7 8.4 2.29-3.18 60-100 2.0 10.8 1.59-2.43 Parus ater >I20 7.8 11.8 6.18-9.82 60-100 5.8 13.9 4.40-7.60 Sitta europaea >I20 2.5 21.3 1.67-3.84 60-100 1.0 13.6 0.82-1.39 Certhia familiaris >I20 2.7 20.6 1.78-3.99 60-100 1.6 27.0 0.95-2.74 among the dominant species in these habitats (Hagemeijer & Blair 1997). All this makes it diffcult to explain its preference to over-mature Macedonian Pine forests. Moreover, the species is not highly dependent on the availability of hollows, as it often breeds in holes in the ground, between roots and stones (Cramp 1998, Hagemeijer & Blair 1997). The Coal Tit is considered socially subordinate to the other Parus species, but it tends to dominate in the Macedonian Pine forests (Nikolov 2007) and there is little probability for the species preference to over-mature forests to be related to inter species competition. The correlation between the abundance of Coal Tit and the habitat characteristics DT and IBAT could be related to the food supply. Similar results are obtained from other areas in Europe, where the species was found to prefer old coniferous forests (Fuller 1995). In Western Europe, Nuthatch is closely attached to over-mature, and even decaying deciduous trees (Cramp 1998). With the present study, the same relation was proved for coniferous forests. Treecreeper was also predominantly found in old trees (Jansson & Andrιn 2003, Ellermaa 2005) and it was proposed as indicator species for the status of forest communities (Jansson & Andrιn 2003, IOC Acrocephalus 28 (134): 95-104, 2007 Table 4: Standardized selection indices of breeding bird species with preference to mature or over-mature Macedonian Pine Pinus peuce forests during the breeding seasons of 2004–2006 in Pirin National Park, Bulgaria. Abbreviations: B? & B? – standardized selection indices of birds in over-mature and mature forests respectively. Tabela 4: Standardizirani selekcijski indeksi vrst ptic gnezdilk, ki so raje izbirale optimalno oz. terminalno fazo odraslih gozdov molike Pinus peuce v gnezditvenem obdobju 2004–2006 v bolgarskem Narodnem parku Pirin. Okraj{ave: B? & B? – standardizirani selekcijski indeksi ptic v terminalni oz. optimalni fazi odraslih gozdov. Species / Vrsta Troglodytes troglodytes Turdus viscivorus Phylloscopus collybita Parus ater Sitta europaea Certhia familiaris Standardized selection index / Standardizirani selekcijski indeks B0 Bu 0.60 0.40 0.36 0.64 0.58 0.42 0.55 0.45 0.61 0.39 0.60 0.40 Suorsa et al. 2005). The preference of species to the over-mature forest communities were compared in old forests is because of its feeding strategy to look the present study. The lack of preference to any of the for small invertebrates in trees covered in loose bark studied Macedonian Pine forest types probably means providing many crevices (Cramp 1998). According that both of the studied forest age classes satisfy the to Gibb (1954), the Treecreeper feeds more often habitat requirements of these species. on live than dead trees, but the amount of the dead Two of the species preferring over-mature forest in wood within the forest is crucial for the survival rate Pirin National Park were trunk and bark-feeders and of the species during the winter. Both Nuthatch and four of them were hollow nesters. It is known that the Treecreeper, being secondary hollow nesters, beneft of density of hole-nesting and trunk and bark-feeding higher amount of old trees in over-mature forests as birds increases with forest maturation (Moss 1978, well in their nest-site selection. Helle & Mφnkkφnen 1990, Pearce & Venier 2005), It is known that eight amongst the species in relation to the greater structural complexity of the inhabiting the Macedonian Pine forests prefer old bark and greater availability of holes and crevices in the forests, but this was not confrmed during the present old trees (Smith et al. 1985, Schieck et al. 1995). Apart study. Four of them (Capercaillie Tetrao urogallus, from birds (Fuller 1995), the old and dead trees are Tawny Owl Strix aluco, Tengmalm’s Owl Aegolius home to many other animals and storage for moisture funereus and Black Woodpecker) were not included in and nutrients. Therefore the retention of snags, limb, the analysis due to the small sample sizes. The other logs and decaying old trees is an essential component four species (Great Spotted Woodpecker Dendrocopos of any wildlife conservation or management plan major, Mistle Thrush, Willow Tit Parus montanus (Ranius & Fahrig 2006). Nevertheless, the practice to and Crested Tit Parus cristatus) had sample sizes large consider snags and logs as signs of “unkempt” forests enough to obtain reliable inferences. In Europe, the and as waste materials, which should be discarded, Great Spotted Woodpecker has higher density in continue. For the 1993–2000 period, a total of 93,914 older forests, but this is not a rule (Gorman 2004). m3 of wood was obtained in Pirin National Park. This The Mistle Thrush prefers open canopy and old forest amount of cut wood exceeds the permitted limit from communities, but this was proved only for deciduous the management plan with 6.5% and what is more, forests (Hagemeijer & Blair 1997). A main factor in 80% of obtained wood consisted of dead and decaying the habitat selection by Willow and Crested Tits is the trees (Petrov, unpubl.). Taking into consideration that amount of dead wood, where they dig their hollows the removal of dead wood from the managed forests is (Baker 1991, Hagemeijer & Blair 1997) and that is one of the main factors for the decrease of woodland why these species prefer old forests (Fuller 1995). The birds (Hanski & Walsh 2004), we strongly suggest results from the present study did not show a preference that more attention should be paid to minimizing the of forementioned species to mature or over-mature dead wood and decaying trees clearing from old forest Macedonian Pine forests, but this does not mean and especially from the protected areas. that they do not prefer the old forest communities. In most of the other studies, comparison was made Acknowledgements: This study was initiated by the between young and old forests, whereas mature and Central Laboratory of General Ecology / Bulgarian 101 S. Ch. Nikolov: Study on the habitat selection by birds in mature and over-mature Macedonian Pine Pinus peuce forests in Pirin National Park (SW Bulgaria) Table 5: Results from the forward stepwise Multiple Regression Analysis (computed by STATISTICA 7.0), testing correlations between abundances of bird species preferring mature or over-mature Macedonian Pine Pinus peuce forests, and habitat parameters. Only signifcant betas are shown, * P < 0.05, ** P < 0.01, *** P < 0.001. Tabela 5: Rezultati multiple regresijske analize, metoda forward stepwise (izra~unani s programom STATISTICA 7.0) za testiranje korelacij med abundanco vrst ptic, ki so bolj naklonjene do optimalnih oz. terminalnih faz odraslih gozdov molike Pinus peuce, in habitatnimi parametri. Prikazani so samo statisti~no zna~ilni faktorji ί, * P < 0.05, ** P < 0.01, *** P < 0.001. Dep. Var. / Odv. spr. Ind. Var./ Neodv. spr. ί R R F df T. troglodytes ADW ROCK 0.45 0.33 0.42 0.18 3.11* 3, 43 T. viscivorus L1 DT CC 0.56 0.61 -0.39 0.57 0.33 4.07** 5, 42 P. collybita CC TH -0.43 -0.37 0.61 0.37 8.53*** 3, 43 P. ater DT IBAT -0.64 0.32 0.48 0.23 4.31** 3, 43 S. europaea ADW TH 0.46 0.39 0.56 0.31 4.68** 4, 42 C. familiaris ADW 0.42 0.45 0.21 5.74** 2, 44 Remarks / Opombe: * Dep. Var. – Dependent variables (abundances of species); Ind. Var. - Independent variables (habitat parameters): ADW – amount of dead wood, ROCK - cover of rocks and piles of stones, L1 – cover of the ground level foliage profle, DT – density of trees, CC – canopy cover, TH – tree height, IBAT – index of basal areas of trees. ** Odv. spr. – Odvisne spremenljivke (abundanca vrst); Neodv. spr. – Neodvisne spremenljivke (habitatni parametri): ADW – koli~ina odmrlega lesa, ROCK – pokrovnost skal in kamnitih skladov, L1 – profl talnega listnega pokrova, DT – gostota drevja, CC – pokrovnost kro{enj, TH – vi{ina dreves, IBAT – indeks bazalnih predelov drevja. Academy of Sciences, fnanced by the Bulgarian Biodiversity Foundation and supported by the Pirin National Park Central Offce. 6. Povzetek Avtor {tudije je v obdobju treh let (2004–2006) preu~eval, kako si ptice izbirajo habitat v optimalni (60–100 let) in terminalni fazi (>120 let) odraslega gozda molike Pinus peuce v Narodnem parku Pirin (JZ Bolgarija). Na skupaj 185 naklju~no izbranih ploskvah s polmerom 25 m je preu~eval in opisal zna~ilnosti habitatov. Ptice je s to~kovnim popisom (r = 50 m) {tel dvakrat na leto (junija in julija) in ugotovil, da med preu~evanimi starostnimi razredi gozdov skoraj ni razlik v {tevilu pti~jih vrst. Pet vrst, in sicer str`ek Troglodytes troglodytes, vrbji kova~ek Phylloscopus collybita, meni{~ek Parus ater, brglez Sitta europaea in dolgoprsti plezal~ek Certhia familiaris, je bilo bolj naklonjenih terminalni fazi gozda molike (v kateri je bila zabele`ena tudi vi{ja gostota gnezdenja) kot optimalni. Slednji je bila bolj naklonjena le ena vrsta, in sicer carar Turdus viscivorus. Avtor je z uporabo multiple regresije raz~lenil pomembne kriterije za izbor habitata, kar zadeva teh {est vrst. Pri dveh vrstah, ki so bile bolj naklonjene terminalni fazi, je {lo za ptice, ki se prehranjujejo na deblih in lubju, pri {tirih pa za duplarje. Avtor poudarja pomen koli~ine odmrlega lesa in trohne~ega drevja za pti~je zdru`be in predlaga, da je treba minimizirati odstranitev odmrlega lesa iz vseh gozdov, posebno tistih v za{~itenih obmo~jih. 7. References Akaike, H. (1973): Information theory and an extension of the maximum likelihood principle. pp. 267–281 In: Petran, B. & Csaaki, F. (eds.): International Symposium on Information Theory, 2nd edn. – Akadeemiai Kiadi, Budapest, Hungary. Bδchler, E. & Liechti, F. (2007): On the importance of g(0) for estimating bird population densities with standard distance-sampling: implications from a telemetry study and a literature review. – Ibis 149: 693–700. Baker, H. (1991): Habitat use by the Crested Tit Parus cristatus in Scottish pinewoods. – University of Dundee, Dundee. 102 Acrocephalus 28 (134): 95-104, 2007 Barbour, M., Burk, J. & Pitts, W. (1980): Terrestrial plant ecology. – Menlo Park, Benjamin Cummings, California. Bennett, A. (2003): Linkages in the Landscape: The Role of Corridors and Connectivity in Wildlife Conservation. – IUCN, Gland, Switzerland and Cambridge, UK. Bibby, C., Burgess, N. & Hill, D. (1992): Bird Census Techniques. – Academic Press, London. Bibby, C., Jones, M. & Marsden, S. (eds.) (1998): Expedition Field Techniques, Bird Surveys. – Expedition Advisory Centre, London. Bondev, I. (1991): Vegetation of Bulgaria. – Sofa University Press, Sofa. (in Bulgarian) Braun-Blanquet, J. (1932): Plant sociology: study on plant communities. – McGraw-Hill, New York. Buckland, S., Anderson, D., Burnham, K. & Laake, J. (1993): Distance Sampling: Estimating abundance of biological populations. – Chapman & Hall, London. Cock, M. (1978): The assessment of preference. – Journal of Animal Ecology 47: 805–816. Conner, R. & Dickson, J. (1997): Relationships between bird communities and forest age, structure, species composition and fragmentation in the West gulf coastal plain. – Texas Journal of Science 49 (3): 123–138. Cramp, S. (ed.) (1998). The complete birds of the western Palearctic on CD-ROM. – Oxford University Press. Davis, L., Waterhouse, M. & Armleder, H. (1999): A comparison of the breeding bird communities in seral stages of the Engelmann Spruce-sub-alpine Fir zone in East Central British Columbia. – Ministry of Forest Research Program, British Columbia. Dickson, J., Conner, R. & Williamson, J. (1993): Neotropical migratory bird communities in a developing pine plantation. – Proc. Ann. Conf. Southeast. Assoc. Fish & Game Agencies 47: 439–446. Dixon, W. & Massey, F. (1969): Introduction to statistical analysis. – McGraw-Hill, New York. Ellermaa, M. (2005): Breeding densities of common breeding species in managed mixed and moist forests in Pδrnumaa, Estonia. – Hirundo 18: 58–66. ESRI 2000 (2000): ArcMap 8.3. – Environmental Systems Research Institute, USA. Fowler, J. & Cohen, L. (1992): Statistics for ornithologists. 2nd edn. – BTO, London. Fuller, R. (1995): Bird life of woodland and forest. – Cambridge University Press, Cambridge. Gibb, J. (1954): Feeding ecology of tits, with notes on Treecreeper and Goldcrests. – Ibis 96: 513–543. Gil-Tena, A., Saura, S. & Brotons, L. (2007): Effects of forest composition and structure on bird species richness in a Mediterranean context: Implications for forest ecosystem management. – Forest Ecology and Management 242 (2/3): 470–476. Gorman, G. (2004): Woodpeckers of Europe. – D & N Publishing, Berkshire. Hagemeijer, E. & Blair, J. (eds.) (1997): The EBCC Atlas of European breeding birds: their distribution and abundance. – T. & A. D. Poyser, London. Hanski, I. & Walsh, M. (2004): How much, how to? Practical tools for forest conservation. – BirdLife European Forest Task Force, Helsinki. Hayes, W. & Winkler, R. (1970): Statistics: probability, inference and decision. Vol. I. – Holt, Rinehart & Winston, New York. Helle, P. & Mφnkkφnen, M. (1990): Forest succession and bird communities: theoretical aspects and practical implications. pp. 299–318 In: Keast, A (ed.): Biogeography and ecology of forest bird communities. – SPB Academic Publ., Hague. Hill, M. (1973): Diversity and evenness: a unifying notation and its consequences. – Ecology 54: 427–432. Iankov, P. (ed.) (in press): Atlas of the breeding birds in Bulgaria. – Bulgarian Society for the Protection of Birds, Sofa. Iordanov, D. (1963): Flora of National Republic Bulgaria. Vol 1. – Bulgarian Acad. Sci., Sofa. (in Bulgarian) Jansson, G. & Andrιn, H. (2003): Habitat composition and bird diversity in managed boreal forests. – Scandinavian Journal of Forest Research 18: 225–236. Jongman, R., Ter Braak, C. & Va n Tongeren, O. (1997): Data analysis in community and landscape ecology. – Cambridge University Press, Cambridge. Kirk, D.A. & Hobson, K.A. (2001): Bird-habitat relationships in jack pine boreal forests. – Forest Ecology and Management 147: 217–243. Koleva, E. 2003. Climate of the Pirin national park. pp. 24–26 In: National park Pirin management plan 2004–2013. – Ministry of Environment and Waters & Bulgarian Biodiversity Foundation, Sofa. (in Bulgarian) Krebs, C. (1999): Ecological Methodology. 2nd edn. – Benjamin Cummings, Menlo Park, California. Manly, B., McDonald, L. & Thomas, D. (1993): Resource Selection by Animals: Statistical Design and Analysis for Field Studies. – Chapman & Hall, London. Moore, J., Fitzsimmons, P. , Lambe, E. & White, J. (1970): A comparison and evaluation of some phytosociological techniques. – Vegetatio 20: 1–20. Moskαt, C. & Waliczky, Z. (1992): Bird-vegetation relationships along ecological gradients: ordination and plexus analysis. – Ornis Hungarica 2: 45–60. Moss, D. (1978): Diversity of woodland song-bird populations. – Journal of Animal Ecology 47: 521–527. Mueller-Dombois, D. & Elienberg, H. (1974): Aims and methods of vegetation ecology. – John Wiley, New York. Nankinov, D. (1997): Species composition of the ornithofauna in the Scots Pine ecosystems. – Forestry Science 3: 84–95. (in Bulgarian) Neu, C., Byers, C. & Peek. J. (1974): A technique for analysis of utilization-availability data. – Journal of Wildlife Management 38 (3): 541–545. Nikolov, S. (2007): Density and community structure of breeding birds in Macedonian pine Pinus peuce forests in Bulgaria. – Avocetta 31: 49–56. Nikolov, S. & Spasov, S. (2005): Frequency, density and number of some breeding birds in south part of Kresna Gorge (SW Bulgaria). – Acrocephalus 26 (124): 23–31. Pearce, J. & Venier, L. (2005) Boreal bird community response to jack pine forest succession. – Forest Ecology and Management 217 (1): 19–36. IO3 S. Ch. Nikolov: Study on the habitat selection by birds in mature and over-mature Macedonian Pine Pinus peuce forests in Pirin National Park (SW Bulgaria) Petrov, I. (2003): Vegetation cover of the Pirin National Park. pp. 41–46 In: National park Pirin management plan 2004–2013. – Ministry of Environment and Waters & Bulgarian Biodiversity Foundation, Sofa. (in Bulgarian) Ranius, T. & Fahrig, L. (2006): Targets for maintenance of dead wood for biodiversity conservation based on extinction thresholds. – Scandinavian Journal of Forest Research 21: 201–208. Sabo, S.R. (1980): Niche and habitat relationships in subalpine bird communities of the White Mountains of New Hampshire. – Ecological Monographs 50: 214–259. Schieck, J., Nietfeld, M. & Stelfox, J. (1995): Differences in bird species richness and abundance among three successional stages of aspen-dominated boreal forests. – Canadian Journal of Zoology 73: 1417–1432. Shapiro, S. & Wilk, M. (1965): An analysis of variance test for normality (complete samples). – Biometrika 52 (3/4): 591–611. Smith, K., Averis, B. & Martin, B. (1985): The breeding bird community of oak plantations in the forest of Dean, southern England. – Acta Oecologica 8: 209–217. StatSoft (2004): STATISTICA 7.0 Data Analysis Software. – StatSoft, Inc. – [http://www.statsoft.com/]. Suorsa, P., Huhta, E., Jδntti, A., Nikula, A., Helle, H., Kuitunen, M., Koivunen, V. & Hakkarainen, H. (2005): Thresholds in selection of breeding habitat by the Eurasian treecreeper (Certhia familiaris). – Biological Conservation 121: 443–452. Thomas, L., Laake, J., Strindberg, S., Marques, F., Buckland, S., Borchers, D., Anderson, D., Burnham, K., Hedley, S., Pollard, J., Bishop, J. & Marques, T. (2006): Distance 5.0. Release 2. Research Unit for Wildlife Population Assessment. – University of St. Andrews, UK. – [http://www.ruwpa.st-and.ac.uk/ distance/ ]. Tiainen, J., Vickholm, M., PaKKALA, T, Piiroinen, J. & Virolainen, E. (1983): The habitat and spatial relations of breeding Phylloscopus warblers and the goldcrest Regulus regulus in southern Finland. – Annales Zoologici Fennici 20: 1–12. Tutin, T, Burges, N, Chater, A., Edmondson, J., Heywood, V., Moore, D., Valentine, D., Walters, S. & Webb, D. (eds.) (1993): Flora Europaea. Vol. I. -Cambridge Univ. Press, Cambridge. Velchev, V. (2002): Vegetation types. pp. 324–336 In: Kopralev, I., Jordanova, M. & Mladenov, C. (eds.): Geography of Bulgaria. – ForCom, Sofa. (in Bulgarian) Arrived / Prispelo: 15.6.2007 Accepted / Sprejeto: 18.3.2008 104