Acrocephalus 113 - 114.qxd 28.3.2003 8:52 Page 1^^ Acrocephalus 23 (113-114): 115 – 121, 2002 The structure of habitat used by Hazel Grouse Bonasa bonasia during winter Struktura zimskega habitata gozdnega jereba Bonasa bonasia Hubert Zeiler1 , Monika Breuss1 , Mark wöss1 & Veronika Szinovatz1 Department of Wildlife Biology and Game Management, University of Agricultural Sciences Vienna, Peter Jordan Str. 76, 1190 Vienna, Austria Steirische Landesjägerschaft, Schwimmschulkai 88, 8010 Graz, Austria, e-mail: hubert.zeiler@jagd-stmk.at A study was carried out over three winter seasons (1995-1998) to evaluate the most important habitat parameters for Hazel Grouse Bonasa bonasia on the forest stand scale. The study area was situated in the Southern Limestone Alps in Austria. It was divided into two parts, reflecting two different forest management strategies: (1) SIMPLE, an age class system with clear-cutting, and (2) MULTI, multi-layered stands with selection felling. An analysis of the roosting sites and habitat use was conducted, as well as a dropping investigation and diet analysis. Monotonous even-aged forest stands, as well as multi-layered old forests, can provide Hazel Grouse habitats but, compared to the latter, monotonous forests are risky habitats because the suitability can depend on only one factor. The habitats in the age class forest are limited in time, whereas multi-layered forests offer habitats for a full rotation period. Key words: Hazel Grouse, Bonasa bonasia, winter habitat, habitat parameters, winter ecology, Austria Klju~ne besede: gozdni jereb, Bonasa bonasia, zimski habitat, parametri habitata, zimska ekologija, Avstrija 1. Introduction Hazel Grouse has already been mentioned by Valentinitsch (1892). Similarly, Bergmann et al. Negative effects of forestry on tetraonids are well (1996) concluded that the structure of the habitat documented and discussed in Central Europe within multi-layered forests is perhaps the most (Swenson 1993, Klaus 1994, Tucker & Evans 1997). important factor of all habitat requirements for Hazel Multi-layered forests with high plant diversity have, in Grouse. Moreover, Swenson (1991) detected a the past, been typical of small farm forests in Austria. relationship between forest structure and the survival They have usually been managed by single tree rate of Hazel Grouse. Predation rate was the highest in selection felling or group cutting, with natural dense young forest stands, whereas the safest forest type regeneration. 46% (3.8 mill ha) of Austria is covered by consisted of two layers and was about 100 years old. forest; 53% of that area belongs to small forest owners The aim of this study is to describe key indicators with forest properties up to a size of 200 ha. Another for the type of vegetation structure used of by Hazel common type of forest is the even aged forest with Grouse and to analyse the effects of different forest clearcutting and afforestation. Clearcuts are limited in management strategies. size to two hectares, while most of them are smaller. According to Swenson & Angelstam (1993) Hazel 2. Study area and methods Grouse Bonasa bonasia inhabits the early secondary successional stages as well as old growth forest. One 2.1. Study area common habitat denominator is dense cover from the ground up to two meters in height (Swenson 1993). For our study we selected an area in the southern The importance of a specific habitat structure for Limestone Alps, at the centre of the distribution zone 115 Acrocephalus 113 - 114.qxd 28.3.2003 8:52 Page 1^^ H. Zeiler et al: The structure of habitat used by Hazel Grouse Bonasa bonasia during winter Figure 1: Situation of the study area in the Southern Limestone Alps, Carinthia, Austria Slika 1: Lega obmocja raziskave v Južnih apneniških Alpah na avstrijskem Koroškem of the Hazel Grouse where its occurrence is high. It is situated in the province of Carinthia in Austria at an elevation of 635 m a.s.l. (Figure 1) and has forest management regimes of selection felling as well as clearcutting. The area is characterised by a climate with alpine and illyrican influence (mean precipitation 1244 mm per year, and a mean temperature of 7.7°C). The specific climate and the limestone bedrock have resulted in a high vegetation diversity, with 38 different tree and shrub species and nine dwarf shrub species. Detailed habitat analysis was carried out on the core study area, which was limited to 41.75 ha. 2.2. Methods The study area was separated into two parts (SIMPLE and MULTI) concerning their different type of forest management. MULTI was characterised by a multi-layered forest structure resulting in a very high plant diversity. The size of the area was 20.5 ha, of which 19 ha was covered by forest. Scotch Pine Pinus sylvestris, Norway Spruce Picea abies, Beech Fagus sylvatica and Larch Larix decidua dominate in the canopy. SIMPLE was characterised by a poor plant diversity and a monotonous vertical structure. The size of the area was 21.25 ha with 17.5 ha forested dominated by Norway Spruce (Figure 2). Roosting site and habitat analyses were carried out in each of the two parts of the study area, to compare the ecological effects of the different forest management strategies. The two parts (SIMPLE and MULTI) were compared with respect to their habitat structure, vegetation and plant species composition. u 6 For the analysis of the roosting sites, we used 32 sample plots, each of 5 m radius. These plots were compared with 54 randomly chosen control points of the grid map. In each plot we surveyed the height of every individual tree and shrub, if DBH (= Diameter Breast Height at a height of 1.3 m) was more than 2 cm. The proportion of canopy closure was measured by Leaf Area Index (LAI) with the help of a Canopy Analyser (LAI 2000). LAI was measured as a ratio of green leaf area and base area. To survey the habitat structure, the study area was described by grid system mapping of 162 0.25 ha plots. The corners of the squares were marked in the field by coloured ribbons or poles. A detailed habitat analysis was conducted in each of the squares separately for each layer (canopy, middle and shrub layer). The following parameters have been surveyed: the average diameter of trees at a height of 1.3 m (DBH), the average height of the trees, type of mixture, phase of age (not the years, but the stage of succession of a tree are important), percentage of different tree species, cover in ten-percent steps and the percentage of gaps within a square. Within the shrub layer (up to 2.5 m in height) we surveyed visibility with a cover board, canopy cover of the shrub layer in ten-percent steps, small spruce groups in thickets and ground vegetation. The intensity of use of each square by Hazel Grouse was determined by collecting droppings over three winters (1995-1998). The collection was done systematically in all squares during five days after a period of four weeks without snow fall. Each site with five or more droppings was recorded. Roosting and feeding trees have been marked in the field by coloured ribbons or poles. To determine the winter feeding spectrum of Hazel Grouse the collected droppings were used (Breuss 1999). For the statistical analysis we chose t-test and discriminant analysis (SPSS W6 PC-Version). 3. Results Over three winter periods, 67 roosting trees and 37 feeding trees were marked. 16 roosting trees (24%) were used more than once during a winter season and over several years. Especially in a dense tree regeneration phase, the same tree was used several times by Hazel Grouse. The analysis showed that a preferred roosting site is characterised by a large number of Norway Spruce and Grey Alder Alnus incarta. All the roosting trees (n = 32) were Norway Spruce, with an average DBH of 13.5 cm and an average height of 9.3 m. Roosting sites exhibited significantly higher shelter (LAI = 2.72, Acrocephalus 113 - 114.qxd 28.3.2003 8:52 Page l¥f ACROCEPHALUS 23 (113-II4): II5 — 121, 2002 SIMPLE MULTI Picea rosting tree Om 10 m 20 m 30 m 40 m 50 m Figure 2: Cross sections of two different types of Hazel Grouse Bonasa bonasia habitat, SIMPLE and MULTI. Note the different vertical forest structure. In SIMPLE we additionally show the difference in use of thinned and unthinned areas by Hazel Grouse in the winter season 1997/98. Slika 2: Prerez dveh razli~nih tipov habitata gozdnega jereba Bonasa bonasia: enostavnega (SIMPLE) in ve~plastnega (MULTI). Razlika je v vertikalni strukturi obeh tipov gozda. V enostavnem je dodatno prikazana razlika v uporabi razred~enega in nerazred~enega obmo~ja gozdnega jereba pozimi 1997/1998. II"7 94 51 Acrocephalus 113 - 114.qxd 28.3.2003 8:52 PagelilK H. Zeiler et al.: The structure of habitat used by Hazel Grouse Bonasa bonasia during winter SD = 0.568) than control points (LAI = 1.77, SD = 1.040; t = 4.55, p < 0.0001, n = 69). A comparison of the proportion of coniferous and deciduous tree and shrub species at the control points and roosting sites provided the following result: at the control points, we registered an average proportion of 73.1% coniferous species and, at the roosting sites, 79.8% (Table 1). A more detailed observation of the deciduous species stressed the importance of shrub species. At the control Table 1: Frequency of roosting sites of Hazel Grouse Bonasa bonasia in coniferous and deciduous species Tabela 1: Primerjava preno~i{~ gozdnega jereba Bonasa bonasia glede na iglasti in listopadni gozd Coniferous Deciduous 73.1% 26.9% Available Tree species Shrub species 46.9 % 53.1 % 79.8% 20.2% Used Tree species Shrub species 30.7% 69.3% points we registered an average proportion of 53.1% shrub species and, at the roosting sites, 69.3% (Table 1). For the SIMPLE forest age class system four factors explained the difference between used and unused squares (Table 2). Canopy cover in the upper layer was determined as the most important variable for the differentiation of squares with and without records of Hazel Grouse, followed by the cover of Hazel Corylus avellana in the shrub layer. Small spruce groups in thickets were a significant parameter for explaining the use of a grid-square in both areas. The importance of cover was emphasized additionally by the frequency of squares with pole stands composed of 100% Norway Spruce in the canopy layer and with Hazel in the shrub layer, being used 13.4 times more frequently than those with a smaller Norway Spruce percentage (n = 61). Stands with only Scotch Pine in the upper layer and without undergrowth were strictly avoided by the birds (Table 2). Within the multi-layered forest (MULTI), ten variables regarding the vegetation characteristics could be detected as important factors for differences between squares with and without records of Hazel Grouse (Table 3). Range of sight in the shrub layer was identified as the most important variable followed by small Norway Spruce groups in thickets and Grey Alder in the upper layer as well as in the under growth. Monotonous stands with Scotch Pine in the canopy 118 Table 2: Four habitat parameters accounting for the use of squares (n = 162) by Hazel Grouse Bonasa bonasia in SIMPLE (even aged forest) in the case of forest age class system (Wilks' Lambda = 0.392, JC2 = 41.71, p = 0.273). Tabela 2: [tirje parametri habitata, ki pojasnjujejo izbor kvadratov (n = 162) pri gozdnem jerebu Bonasa bonasia v enostavnem (SIMPLE) gozdu enake starosti (Wilks' Lambda = 0.392, Z2 = 41.71, p = 0.273). Habitat parameter F p Canopy cover in the canopy layer 6.661 0.012 Hazelnut in the shrub layer 5.599 0.021 Small Norway Spruce groups in thickets 4.881 0.031 100% Scotch Pine in the canopy layer without undergrowth 9.451 0.003 layer without undergrowth (negative parameter), and the number of tree species in the upper storey followed in importance. Further characteristics were the percentage of deciduous trees, percentage of cover in the shrub layer, the forest age class in the upper layer and the percentage of Larch in the upper storey (Table 3). The habitat structure of the study area was documented by cross sections of 50 m length in stands of both SIMPLE and MULTI (Figure 2). The Table 3: Habitat parameters accounting for the use of grid squares by Hazel Grouse Bonasa bonasia within the multi-layered forest (MULTI) (Wilks' Lambda = 0.549, X2 = 31.47, p = 0.443). Tabela 3: Parametri habitata, ki pojasnjujejo izbor kvadratov pri gozdnem jerebu Bonasa bonasia v mre`i znotraj ve~plastnega (MULTI) gozda (Wilks' Lambda = 0.549, JC2 = 31.47, p = 0.443). Habitat parameter Range of sight in the shrub layer Small Norway Spruce groups in thickets Alder in the canopy layer Alder in the shrub layer 100% Scotch Pine in the canopy layer without undergrowth Number of tree species in the canopy layer 6.287 Percentage of deciduous trees Percentage of cover in the shrub layer Trees of high age in the canopy layer Percentage of Larch in the canopy layer 12.158 0.001 9.724 0.003 7.700 0.007 7.534 0.008 7.563 0.008 6.287 0.015 6.059 0.016 5.546 0.021 4.219 0.044 3.959 0.051 F p 51 Acrocephalus 113 - 114.qxd 28.3.2003 8:52 Pagel^K ACROCEPHALUS 23 (113-II4): lij — 121, 2002 vertical projection of the intensively used, even-aged one winter season and in different following winter forest in SIMPLE shows a clear dominance of spruce seasons, in contrast to the results of Thompson & in the upper layer. The position of the cross section Fritzell (1988). The main reason for this difference was chosen in such a way that half the strip was can be the forest type. We explain the repeated use of thinned and the other half unthinned. By recording roosting trees by the very good habitat quality in our the positions of droppings on the 50 m section, a study area and therefore small winter territories. Small strong preference for the unthinned part with Norway winter territories lead to repeated use of “ideal” Spruce and Hazel could be detected. The vertical roosting sites. The small size of winter territories is projection of the cross section in MULTI shows at emphasized by telemetry results (Zeiler 1998). least three different layers and a much higher species Roosting sites were common in the forest layer at richness in the vegetation than in SIMPLE. The high about 2 – 4 m in height. Shrubs offer better cover canopy layer was dominated by light demanding than broad-leafed tree species at that height. Based on species like Larch and Scotch Pine. Norway Spruce the results in Table 3, the demand for more than twice and Beech formed the middle layer. Shrubs as well as as much shrub than tree species in the deciduous class Norway Spruce regeneration dominated in the shrub stresses the importance of structure at roosting sites. layer. Coniferous trees are preferred because of cover and In the multi-layered forest the potential food thermal benefits. Although the use of roosting trees supply was more diverse. The results of the diet was common in our study area the birds also roosted analysis showed that in MULTI stands Hazel Grouse in snow caves on forest openings or even on mainly used buds, twigs and catkins of shrubs (ten unploughed forest roads if the snow depth allowed it species), deciduous trees (nine species), dwarf shrubs (Andreev 1977). (four species) and coniferous trees (four species) – Within the forest age class system (SIMPLE), cover altogether 27 different species. Additionally, leaves of of the canopy layer and a cover of Hazel in the shrub plants of the ground layer were used as well. In the layer was the most important for the use of a grid-age-class forest, the plant spectrum was less diverse square (Figure 2). In particular, a high percentage of and included 13 different species (two shrub species, Norway Spruce in pole forest stands influenced the four deciduous tree, three dwarf shrub and four habitat use in a positive way. Dense young groups of coniferous tree species). There, Hazel Grouse used Norway Spruce stands in thickets were important for mainly buds and catkins of Hazel, Alder, Aspen the habitat use in both parts of the study area. This is Populus tremula and the leaves of Wood Sorrel Oxalis in agreement with the results of Swenson (1991) who acetosella. detected a positive relationship between the amount Roosting trees were located on sites with the of Norway Spruce stands in the habitat and the densest shrub layer. A large number of Norway Spruce survival rate of birds, assuming enough food was and Grey Alder at the roosting site indicates an available. While in the forest age class system shelter availability of cover and food. Within the Norway was important for the use of a square, within the Spruce groups, most of the roosting sites were found multi-layered forest the number of tree species in the in trees up to a height of 9 m. Temperature upper layer as well as the percentage of Larch and high measurements which we took on roosting sites in age trees were essential. These three parameters are dense Norway Spruce cones and in neighbouring responsible for the structure of the canopy. Forests mature stands without undergrowth showed with a high percentage of light demanding species, differences of 1.5°C at average temperatures of about like Larch and Scotch Pine, in the upper layer have a –3°C. loose canopy structure and therefore more light can reach the ground layer. In Larch forest, a dense and 4. Discussion good structured undergrowth offers better cover and food for Hazel Grouse. There the tree regeneration is Besides cover, dense Norway Spruce groups offer able to grow in dense clusters preferred for night thermal benefits (Swenson & Olson 1991). roosting sites by Hazel Grouse (Figure 2). Therefore, Thompson & Fritzell (1988) created a model which two opposing parameters can explain habitat use in indicates that coniferous roosts are the best alternative the different age classes: dense cover in young stands for reducing heat loss in Hazel Grouse. Repeated use versus light canopy layer in old stands. Dense cover in of roosting trees is documented for Capercaillie Tetrao even aged pole stands can be associated with early urogallus (Fuschlberger 1956) but not for Hazel successional stages which are generally identified as Grouse. Roosting trees were used several times during typical Hazel Grouse habitat (Valentinitsch 1892, 119 Acrocephalus 113 - 114.qxd 28.3.2003 8:52 Pagel2^K H. Zeiler et al.: The structure of habitat used by Hazel Grouse Bonasa bonasia during winter Lieser 1993, Swenson 1993). On the other hand, a one factor, such as a single main winter food plant. fragmented canopy with gaps in between corresponds The habitats in the age class forest are limited in time to late successional stages of gap dynamics. whereas multi-layered forests offer habitat over the The density of the shrub layer, as well as the full time of the rotation period common in age class percentage of cover in this layer, were main important forest systems (100 – 120 years). Our results, as well factors for habitat use in the whole study area. The as those of Swenson (1991), suggest that the most denser the shrub layer the higher was the use by Hazel secure forest type is multi-layered old forest with Grouse. Squares with even aged monotonous old- selection felling as the management strategy. This type growth Scotch Pine stands without undergrowth, and of forest includes all successional stages in the same squares with very poor forest structure were avoided area, as is typical for primeval forest. Multi-layered by Hazel Grouse during winter. forests support high habitat quality as well as low risk The results of our analysis also indicate that shrubs in terms of contemporary forestry. with a high stem diameter have more stronger branches and a more plentiful supply of buds. These Acknowledgements: We thank the Zentralstelle der shrubs offer better cover and birds can climb more Österreichischen Landesjagdverbände for financial easily on them. A high stem diameter in shrubs thus support and the Department of Wildlife Biology and results in greater availability of both food and cover. Game Management, University of Agricultural The overall number of different tree and shrub Sciences Vienna for facilitating this research. species was not important for the use of a square. However, the results of the diet analysis (Breuss 1999) 6. Povzetek showed that the birds in the even-aged forest, with fewer species, used about half the number of species Avtorji prispevka so v treh zaporednih zimah (1995- for feeding than in the multi-layered forest with a 1998) ocenili najpomembnej{e parametre habitata high species diversity. Al the same, MULTI squares gozdnega jereba Bonasa bonasia v Ju`nih apneni{kih with a high percentage of Grey Ader (Table 3) in the Alpah v Avstriji glede na tipe gozdnih sestojev. Glede upper as well as in the undergrowth, were significantly na razli~ni strategiji gospodarjenja z gozdom je bilo preferred by the birds. The results show that Hazel preu~evano obmo~je razdeljeno v dva dela: (1) Grouse is able to colonise comparatively monotonous enostavni (SIMPLE) - sistem starostnega razreda z stands. However, in with the context of contemporary jasami in (2) ve~plastni gozd (MULTI) - ve~plastni forestry, this could be very risky, because thinning in sestoji s selektivno se~njo. Avtorji so opravili analizo these stands normally results in instantaneous habitat po~ivali{~, izbora habitata, iztrebkov in prehrane. loss for Hazel Grouse. Monotoni gozdni sestoji enakih starosti kot tudi Our results show that only four out of 12 ve~plastni stari gozdovi lahko zagotavljajo bivali{~e significant habitat parameters have a strong gozdnemu jerebu, le da so monotoni gozdovi v connection to food. Flexibility in the use of the primerjavi z ve~plastnimi gozdovi tvegano bivali{~e available food plants can be seen. Depending on their za to ptico, saj je ustreznost habitata lahko odvisna `e occurrence, different food plants can play an od enega samega dejavnika. Bivali{~a v enostavnem important role in winter feeding ecology. In MULTI, tipu gozda so ~asovno omejena, medtem ko Hazel Grouse used a wider spectrum than in ve~plastni gozdovi zagotavljajo bivali{~e v celotnem SIMPLE, explained by a higher vegetation diversity in ciklusu. the former (Breuss 1999). The remaining eight parameters account for the importance of habitat 7. References structure for Hazel Grouse. 5. Conclusions Andreev, S. (1977): Temperature Conditions in Snow Cavities of Hazel Grouse (Tetrastes bonasia kolymensis). Soviet Journal of Ecology 8 (5): 454-455. Bergmann, H.H., S. Klaus, F. Mueller, W. Hazel Grouse is often seen as a typical species for Scherzinger, J.E Swenson & J. Wiesner (1996): Die young successional forest stages. From our results we Haselhuehner. 4. Auflage. Die Neue Brehmbücherei, can conclude that monotonous even-aged forest Band 77. (in German) 1 ¦ ¦ ¦ • ¦ j rjr -j Breuss, M. (1999): Untersuchungen zum winterlich en stands, as well as multi-layered old forests, can provide ,, , , , TT „ :r ,„ , . N . . ri- r^ 1 JNahr ungsspektrum des Haselhuhns (Bonasa bonasia) i n Hazel Grouse with a winter habitat. Compared to den Gailtaler Kalkalpen (Losungsanalysen). Diploma multi-layered forests, monotonous forests are risky thesis at the University of Vienna, Vienna. (in German habitats because their suitability can depend on only with English summary) 120 Acrocephalus 113 - 114.qxd 28.3.2003 8:52 Page 12-:^ ACROCEPHALUS 23 (113-II4): lij — 121, 2002 Fuschlberger, H. (1956): Das Hahnenbuch. C. Mayer Verlag, München. Klaus, S. (1994): To survive or to become extinct: small populations of tetraonids in Central Europe. 137-156 In: Remmert, H. (ed.): Minimum animal populations. Springer Verlag. Lieser, M. (1993): Untersuchung der Lebensraumansprüche des Haselhuhns (Bonasa bonasia, L. 1758) im Schwarzwald im Hinblick auf Maßnahmen zur Arterhaltung. Inaugural Dissertation an der Forstwissenschaftlichen Fakultät der Albert-Ludwigs-Universität Freiburg i.Br., Freiburg. (in German) Swenson, J.E. (1991): Social organization of hazel grouse and ecological factors influencing it. Diss. Univ. of Alberta, Edmonton. Swenson, J.E. (1993): Habitat requirements of Hazel Grouse. 6th International Grouse Symposium Udine, Italy: 155-159. Swenson, J.E. & B. Olsson (1991): Hazel Grouse night roost site preferences when snow-roosting is not possible in winter. Ornis Scandinavica 22: 284-286. Swenson, J.E. & P. Angelstam (1993): Habitat separation by sympatric forest grouse in Fennoscandia in relation to boreal forest succession. Can. J. Zool. 71: 1303-1310. Thompson, F.R. & E.K. Fritzel (1988): Ruffed Grouse winter roost site preference and influence on energy demands. J. Wild. Manage. 52: 454-460. Tucker, G.M. & M.I. Evans (1997): Habitats for birds in Europe: a conservation strategy for the wider environment. BirdLife International, BirdLife Conservation Series 6., Cambridge, U.K. Valentinitsch, F. (1892): Das Haselhuhn (Tetrao bonasia) -dessen Naturgeschichte und Jagd. Eine ornithologische und jagdliche Monographie. Adolph W. Künast. k.u.k. Hofbuchhändler Wien. (in German) Zeiler, H. (1998): Die Bedeutung der Habitatstruktur in der Winterökologie des Haselhuhns. Endbericht “Haselhuhnprojekt Dellach/Drau“ mit Ableitungen für eine haselhuhn-freundliche forstliche Bewirtschaftung. Department of Wildlife Biology and Game Management, University of Agricultural Sciences Vienna, Vienna. (in German) Arrived / Prispelo: 7.11.2002 Accepted / Sprejeto: 6.3.2003 121