Blackwell Science, LtdOxford, UK ERE Ecological Research 0912-38142002 Ecological Society of Japan 173June 2002 493 Bird strategies and habitat changes S. Tworek 10.1046/j.0912-3814.2002.00493.x Original Article339359BEES SGML

Ecological Research (2002) 17, 339–359

Different bird strategies and their responses to habitat changes in an agricultural landscape STANISL¢ AW TWOREK*

Institute of Nature Conservation, Polish Academy of Sciences, Al. Mickiewicza 33, 31-120 Kraków, Poland Breeding bird communities were investigated in habitat islands in the agricultural landscape of southern Poland in 1995–1998. Using the following species-specific characteristics of birds – type and location of nest, type of food, place and way of foraging, type of migration, clutch size, incubation period, fledging period and number of broods per year – five typical bird strategies were delimited: ‘terrestrial’, ‘predators’, ‘hole-nesters’, ‘vagrants’ and ‘arboreal’. Stepwise multiple regression and canonical correlation analyses were used to examine relationships between ecological parameters for bird strategies and habitat variables describing sample plots. Bird densities, domination, species abundance and turnover rates depended on the area and structure of a habitat island and to a lesser extent on isolation and surrounding land use. Terrestrial birds tend to occupy large, open habitats, often wetlands, predators also need extensive areas of forest for nesting and open habitats for foraging, hole-nesters demand areas with old tree stands or other places where they can hide their nests, many close to human settlements, vagrants prefer dampish habitats with rich herb and shrub layers and arboreal birds appear to be characteristic of forest edges. Results from this study show that responses of birds to habitat changes differ depending on their strategies. Some life styles benefit from habitat fragmentation, while for others it is a principal threat. The methodology used in this study can be a model for species that share distributional, ecological or life-history features and may enable more effective conservation of bird species. Key words: agricultural landscape; breeding birds; bird strategies; conservation implications; habitat changes; habitat islands.

INTRODUCTION The concept of habitat islands, habitat patches surrounded and isolated from similar habitats by other, often contrasting, types of ecosystems, has an important place in landscape ecology. Because of their limited size and isolation, these habitats scattered in a sea of ecologically different environments resemble islands in the ocean. That is why many authors explaining phenomena occurring in these ecosystems refer to MacArthur and Wilson’s (1967) equilibrium theory of island biogeography. It is now accepted that in island conditions two main factors, the size of the island and the degree of its isolation, influence the natural richness. These two factors determine many secondary hab*Email: [email protected] Received 2 April 2001. Accepted 10 October 2001.

itat characteristics, for example, the proportion of ecotone zone in relation to the island area, microclimatic changes, greater susceptibility to disturbance, different species composition and intensification of synanthropic processes. These problems have been discussed, particularly in the context of a minimum viable population (Lovejoy & Oren 1981; Hanski et al. 1996). With the development of landscape ecology, the concept of metapopulation has been used in studies on the functioning of habitat islands (Levins 1970; Hanski & Gilpin 1997). It stresses that the existence of species in the fragmented environment is the result of local extinction and recolonization in particular habitat patches and dispersal. Ensuring the connectivity between particular subpopulations and binding them into one system is of key importance. Due to the different surroundings of terrestrial habitat islands and different methods of origin

340

S. Tworek

many other factors, in addition to the size of an island and its situation in relation to other islands, were taken into account in studies on these habitats. These factors include the shape, age and biotope diversity of a site, the degree of its synanthropization, habitat similarity between an island and its surroundings and microclimatic changes. All these factors can significantly influence the development of specific biocenoses governing population relationships. Hypotheses constructed to test these relationships are based on many sources. In addition to studies examining the theory of island biogeography and the theory of metapopulation, these sources include studies examining patch dynamics theory (Pickett & White 1985), mathematical models on dispersal and spatially distributed populations (Kareiva 1990) and field studies in fragmented habitats. However, there are still very few landscape-level experimental studies, especially on vertebrates. Agricultural landscapes remain under particularly intense human pressure. The chance of survival and persistence of many species in this environment is closely connected with the presence of natural habitat enclaves or with adequate adaptation to change. Attempts at examining the importance of habitat islands for birds were undertaken with reference to both the area and isolation of an island (Opdam et al. 1985; Blake & Karr 1987) and different elements of its spatial structure (McGarigal & McComb 1995). The authors formulated either general conclusions for bird populations, depending on certain habitat characteristics (Rolstad 1991; Kujawa 1997), or detailed recommendations for particular species (Hinsley et al. 1995a). Simultaneously, some authors indicated that certain species respond positively to habitat changes, which result in the growth of landscape fragmentation, even when it concerned forests (Haila et al. 1994; Hagan et al. 1996). As biological similarity, irrespective of taxonomic affiliation, is related to similar habitat adaptations and their ecological consequences (e.g. techniques of foraging or avoiding predators), the question of whether groups of species showing similar features of evolution respond to environmental changes in a similar way arises. Clustering of bird species, based on selected features, is a technique often used by scientists studying changes in the avifauna. However, the goals and

methods of these studies can differ. As a result, different criteria for groupings are adopted. Many authors refer to ecological groups or ‘guilds’ distinguished on the basis of different features, most often, food habits, migration status, habitat preferences, foraging strategy or nesting habits (Tomial¢ oj´c et al. 1984; Blake & Karr 1987; Böhning-Gaese & Bauer 1996; Jokimäki & Huhta 1996; Bentley & Catterall 1997). In recent years, another approach has been presented. Multidimensional statistical methods have been used to analyze the relationship between the occurrence of birds and the vegetation structure of fragmented forests (Bersier & Meyer 1994), plantations (Diaz et al. 1998) or habitat characteristics (Petersen 1998) and the authors have formulated conclusions for the groups of species responding in a similar way. Most of these studies were focused on single characteristics, but life history traits do not occur separately. Some are correlated positively with each other, others negatively. The number of possible combinations is large and all combinations of features are subject to natural selection and form the adaptive strategy of a species (Hansen & Urban 1992; Stearns 1992). Taking previous studies into account, I have attempted to distinguish bird strategies on the basis of a combination of many features and then to identify factors determining the species composition of bird communities and the parameters of the occurrence of birds representing different strategies. The overall aims of this study were to examine the response of birds with different strategies to environmental changes connected with habitat fragmentation, to identify threats connected with this phenomenon and to formulate practical recommendations for conservation. To deal with the problem of correlated features, I decided to determine the strategies of the birds by identifying groups of species with similar demographic characteristics (e.g. reproduction) and similar adaptations to environmental factors.

METHODS Study area The study area (approx. 44 km2) lies north-west of Kraków in southern Poland (50 °06′−50°08′N,

Bird strategies and habitat changes 19°45′−19°55′E) and has an agricultural character (Fig. 1). The largest area is occupied by arable fields, fresh pastures and meadows that because of intense drainage, fertilization and other agrotechnical measures replaced the former fertile wet meadows. Water courses, rivers, ox-bow lakes, streams and drainage ditches, are important elements of the landscape and are associated with remnants of natural forest communities. Among wet meadows there are scattered small patches of alder wood showing strongly disturbed species composition due to intense drainage. Along the largest rivers fragments of riverine forests and fens overgrown to a varying degree with shrubs and trees have survived. There are also small patches of osiers scattered among non-utilized wet meadows and fens; they have developed in place of former reed beds and cut-down alder wood in marshy areas. They also occur along streams and older drainage ditches. Among the mown meadows there are only single clumps of osiers. Reed beds,

40 35 30

Cover (%)

25 20 15 10 5 0 1

2

3

4

5 6 7 8 Land cover types

9

10

11 12

Fig. 1. Proportion of land cover types in the study area according to PHARE CORINE land cover data: (1), non-irrigated arable land; (2), complex cultivation patterns; (3), pastures; (4), mixed forest; (5), discontinuous urban fabric; (6), broad-leaved forest; (7), coniferous forest; (8), land principally occupied by agriculture with significant areas of natural character; (9), green urban areas; (10), sport and leisure facilities; (11), water courses; (12), industrial or commercial units.

341

sedge beds and poorer wet meadows also represent wet habitats in different stages of degradation. Multispecies deciduous forests dominated by beech cover the southern part of the study area. In addition, in some places, mostly at the border between arable fields and buildings, there are orchards, rural parks and vegetable gardens. The abandoned fields undergo quick transformation into ruderal vegetation communities. Some areas, particularly those situated near Kraków, have recently been afforested. Over the whole study area, 38 sample plots were delimited with an area range of 0.2–52 ha (mean ± SD = 14.0 ± 13.9). The plots included habitats different from the surrounding landscape, so their borders were mostly natural and easy to identify in the field.

Habitat and landscape variables Habitat features of the plots were described using the variables listed in Table 1. In the case of small sites, their size and perimeter were measured directly in the field, while for larger sites geodetic maps (scale 1 : 10 000) were used. The shape of the sites was determined (according to Hinsley et al. 1995a), using the Pm/Pc index, where Pm was the measured perimeter and Pc was the perimeter of a circular plot of the same area. Thus, the more linear the shape of the measured plot, the greater the proportion of its edge to the whole area and the higher the coefficient describing its shape. The identification of the habitats occurring in each sample plot and an estimation of their number was based on 16 categories: deciduous forest, coniferous forest, mixed forest, dense brushwood, thin brushwood, old undergrowth, young undergrowth, alley of trees, orchard, meadow, reed bed, fen, cultivated area, pond, stream or drainage ditch, building(s) and waste land. For forest habitats a stage of stand development was determined using the tree-age index. The tree age within each plot was assessed independently by two workers of the local forest inspectorate. I have distinguished the following categories of stands: 0 = no stand, 1 = 1–20-year-old stand, 2 = 21–50-year-old stand, 3 = over 50-year-old stand. I determined the proportion of a given stand category to the whole area by marking each category on the maps. The age index was calculated using the formula:

342

S. Tworek

Table 1 Variables describing the area and structure, the degree of isolation and the surrounding land use that was used in multiple regression analyses and/or canonical correlation analyses Variable Area and structure 1. Plot area (ha) 2. Perimeter (m) 3. Shape 4. Age of tree stand 5. Cover of tree stand older than 50 years (%) 6. Cover of agricultural land (%) 7. Density of canopy layer 8. Density of shrub layer 9. Density of herb layer 10. Cover of wetland (%) 11. Number of habitats Isolation and surrounding land use 12. Distance to the nearest similar plot (km) 13. Distance to the nearest similar plot larger than investigated (km) 14. Perimeter adjoined by wood (%) 15. Perimeter adjoined by grass (%) 16. Perimeter adjoined by crops (%) 17. Perimeter adjoined by buildings and yards (%)

Index = Â (category ¥ proportion) The index calculated in this way varied from 0 (no stand) to 3 (100% of over 50-year-old stand). Another variable was used to describe the proportion of over 50-year-old stand to the whole of the stand in a given sample plot. The remaining proportion variables, that is, the percentage cover of agricultural land (arable fields, hay-meadows, pastures) and the percentage cover of wetland (including rivers, streams, small ponds and other water bodies) were calculated in relation to the whole area. Once per year in June, during the period of full development of vegetation, I determined a cover of the canopy, shrub and herb layers. The term ‘herb layer’ refers also to the above-ground layer of herb vegetation beyond the forest areas. The cover was scored on an arbitrary scale of 0–2, where 0 = lack of vegetation in a given layer, 1 = partial cover (sparse vegetation) and 2 = full cover (closed tree canopy or dense vegetation in the shrub or herb layers). I also noted the proportion of the area in a given category to the whole sample plot area.

Abbreviation

Mean ± SD

Range

AREA PERIMET SHAPE TREEAGE OLDTREES AGRI CANOPY SHRUB HERB WATER HABITATS

13.97 ± 13.99 1937.1 ± 1366.9 2.06 ± 1.77 1.23 ± 0.85 15.85 ± 19.44 27.70 ± 38.17 0.71 ± 0.71 0.85 ± 0.65 1.57 ± 0.40 10.35 ± 14.27 6.8 ± 3.03

0.2–52.0 240–7700 1.14–9.82 0–2.75 0–70 0–100 0–2.0 0–1.9 0.3–2.0 0–50 1–14

NEAREST LARGER

0.23 ± 0.33 0.72 ± 0.92

0–1.55 0–3.40

S_WOOD S_GRASS S_CROPS S_FARMS

13.35 ± 20.49 26.65 ± 26.34 37.30 ± 33.21 18.20 ± 22.72

0–95 0–100 0–100 0–75

As was the case for the age stand index, the cover index was calculated using the formula:

Index = Â (category ¥ proportion). The minimum and maximum values for each layer were 0 and 2, respectively. Two variables were used as a measure of isolation. For all sample sites I measured a distance to the nearest similar plot and to the nearest similar plot larger than the investigated plot, directly in the field or from maps scaled at 1 : 10 000 or 1 : 25 000. To ascertain plot similarity I compared the proportions of main habitats between plots based on the 16 categories distinguished and then each plot was classified into ‘woody’, ‘scrubby’, ‘ecotone’ and ‘field-meadow’ categories. The plots belonging to one category were considered to be similar. Surrounding land-use was assessed annually and was expressed as the percentage of the perimeter of each sample plot adjoining the four most frequent categories of land use: forests, meadows, arable fields and buildings. As in the case of other variables, estimates for small sites were made

Bird strategies and habitat changes Table 2 Factor loadings obtained in factor analysis for variables characterizing vegetation structure and surrounding land use Variable

Factor 1

Factor 2

Factor 3

TREEAGE OLDTREES AGRI CANOPY SHRUB HERB WATER S_WOOD S_GRASS S_CROPS S_FARMS Variation explained Percent of total Cumulative (%)

0.9439 0.6010 −0.8962 0.8269 0.7891 −0.0459 0.2983 −0.3058 0.1015 0.2145 −0.1119 3.61 32.9 32.9

−0. 1907 −0.1382 −0.0950 −0.3652 0.1777 0.6544 0.7145 −0.4625 0.7498 −0.3007 −0.0315 2.04 18.5 51.4

0.1031 0.4616 0.0297 0.1342 −0.1357 −0.2089 −0.0823 0.3821 0.3735 −0.8978 0.5856 1.75 15.9 67.3

directly in the field and for larger sites maps scaled at 1 : 10 000 were used. To reduce the number of correlated variables connected with the vegetation structure of the habitats and the surroundings of sample plots and to determine the character of relationships between variables, I used a factor analysis based on principal components. The analysis referred to 11 variables including the age of the stand, proportion of over 50-year-old stand, vegetation density in the canopy, undergrowth and herb layers, proportions of agricultural land and wetland and variables characterizing the surroundings. Correlations of these variables with the factors distinguished as factor loadings are listed in Table 2. The other variables that did not show a strong correlation were neglected in order to arrive at reliable estimates of factor loadings (cases to variables ratio; Stevens 1986). Factor scores for particular sample areas were calculated from the first three principal components extracted during the analysis. Under the heading ‘factors’ they were used as variables in the regression analysis.

Bird censusing Birds were censused in 1995–1998 basing on a territory mapping method (Bibby et al. 1992). Each year before censuses started I prepared

343

detailed plans of sample sites including landmarks such as water courses, roads, ditches, electric lines, vegetation elements and buildings. Thanks to a dense network of reference points, I did not need to use stakes as marking points, which might have been used by breeding birds for singing and observation and may have attracted hunting raptors. In the two largest forest sites, where natural landmarks were scarce, I marked trees by painting them. Each breeding season sample sites were censused on 7–11 occasions. Censuses started in the early morning (4.00–6.00 AM) and were continued until the singing activity of birds decreased markedly (usually between 11.00 and 12.00 AM). When atmospheric conditions in the morning were unfavourable (rain, strong wind, fog) censuses started in the afternoon (4.00–5.00 PM) and were continued until dawn. Each sample site was censused in the evening at least once in May or June. The survey period started, depending on atmospheric conditions in particular years, between 25 March and 10 April and ended at 10–20 July. All observations were noted on the prepared sketches of the sites with the orientation points marked. On similar and adjacent small sites I tried to find as many nests as possible to determine which site particular pairs or territories should be ascribed to. If I had no direct evidence of breeding, I accepted that a territory was occupied if I had at least three records of a singing male, or a pair of birds, or other behavior indicating territory occupation. When registering singing males of the most numerous species, I paid particular attention to simultaneous records. The number of pheasants (Phasianus colchicus) was estimated on the basis of sounding males. In the case of the cuckoo (Cuculus canorus) I took into account only the courtship behavior of a pair of birds.

Data analyses Each year the field maps were converted into species maps and I estimated the following parameters of bird communities: number of breeding pairs (N), their domination (N%), density of pairs (N/ha) number of species (S) and turnover rate (T). Results from one breeding season per plot have been treated in further analyses as independent cases. Following Diamond (1969), I calculated turnover rate using the formula:

344

S. Tworek

T = (E + C) (S1 + S2) ¥100% where E and C are numbers of species that disappeared from a plot (extinction) and appeared in a plot (colonization) between seasons 1 and 2, and S1 and S2 are numbers of all species breeding in a plot in seasons 1 and 2. To analyze the strategies of birds I described all species according to their preferences with regard to nine species-specific characters. The variables that were distinguished and their categories are placed in Appendix I. The species were described in a binary way, using 0–1 code: if the category distinguished referred to a given species, it was ascribed symbol ‘1’, if not it was ascribed symbol ‘0’. Altogether 43 categories of variables were used. In this way the ecological preferences and adaptations of birds were defined. Data concerning the variables were taken, for the most part, from the literature (Cramp & Simmons 1993; Hagemeijer & Blair 1997; Snow & Perrins 1998; Glutz von Blotzheim & Bauer 1966–1997). This information was compiled with my own field observations. To distinguish the groups of species representing similar strategies I used a cluster analysis. Using Ward’s linkage rule method I tested the heterogeneity of the dataset describing species according to the categories mentioned and I formulated a hypothesis that a division into five groups of birds reflected their different responses to environmental changes resulting from habitat fragmentation. Next, I grouped species using a Kmeans method, maximizing the distances between groups when calculating initial cluster centers. The groups of species distinguished in this way were treated as bird strategies. To evaluate the effect of plot size and habitat components on the density, domination, number of species and the species turnover rate in the distinguished bird strategies I used the stepwise multiple regression in its general form:

Y = B1X1 + B2X 2 + ... + Bk X k where Y is the dependent (criterion) variable, X1, X2,…Xk are independent (predictor) variables and B1, B2,…Bk are partial regression coefficients. Independent variables included, the size, perimeter and shape of a plot, the number of habitats,

principal components from the factor analysis and two variables describing the degree of site isolation. The first three principal components reach values permitting interpretation and use in the multiple regression analysis. Altogether these components account for almost 70% of the variation in the analyzed variables (Table 2). Factor 1, which accounts for almost 33% of the total variation, describes the vegetation structure on a plot in a gradient from agricultural land to forest; sites harbouring mature forests with well-developed undergrowth are characterized by the least proportion of agricultural land. Factor 2 combines the presence of wetland with the dense herb layer, which occurs most frequently in sites that were remnants of former extensive wet meadows limited now to small fragments due to drainage work. A further 16% of the variation is explained by factor 3, which best describes the surroundings of sites characterized by a negative value of the factor loading for a variable describing the surroundings of arable fields in conjunction with positive loadings of the remaining surroundings variables. Canonical correlation analyses were used to generalize the results of the multiple regression analyses and to evaluate the relationships between the bird strategies and the environment. To make the interpretation of the relationships in the canonical correlation models easier, I did not use factors from the factor analysis as variables, but rather raw variables that accounted for most variation in the regression models. Dependent variables included components characterizing the distinguished bird strategies (density, domination, number of species) and independent variables included components describing the sites and their structure (size of a plot, its periphery and shape, age of the stand, proportion of old stand, covers of the canopy, undergrowth and herb layers and shares of agricultural land and wetland). Interpreting results of the regression models, when the domination strategy was analyzed, I added the independent variable describing the number of habitat types on a plot. In the case of density of pairs, this variable was found irrelevant, whereas for the number of species, the relationship is fairly obvious: the number of species increases with an increase in the number of habitats. Thus, in these cases leaving the variable ‘number of hab-

Bird strategies and habitat changes itats’ out contributes to the clarification of the relationships revealed. Results for the two sets of variables are shown on tree clustering diagrams using the unweighted pair-group centroid method. The figures are drawn using correlations of variables with canonical roots following Jongman et al. (1987). For statistical calculations and figures I used STATISTICA for Windows. A variable describing the plot perimeter was log-transformed (lg) to linearize the relationships with dependent variables in the multiple regression and to normalize distribution in the canonical correlation analysis. This transformation was not necessary for any of the other variables.

RESULTS Bird strategies The cluster analysis divided the bird species into five groups, representing different life styles of birds, which are also referred to as strategies (Appendix I). The first strategy comprised 12 species representing 5 orders: Anseriformes, Galliformes, Gruiformes, Charadriiformes and Passeriformes. Of all groups, these species are the least connected with forests. They build open nests on the ground, migrate mostly to southwest Europe, have the largest number of eggs in a clutch and their diet has the largest proportion of plant food. Species belonging to the second strategy also built open nests, but placed them high above the ground. All of these species, although to a different degree, feed on vertebrates; they are mostly sedentary, raise one brood per year with the smallest number of eggs in a clutch and have the longest incubation and fledging periods. This group consists of 20 species belonging to 7 orders (Ciconiiformes, Falconiformes, Accipitriformes, Charadriiformes, Strigiformes, Coraciiformes, Passeriformes). They nest mostly in forests and woodlots and forage in open areas, often very distant from their nests. The third strategy comprises species placing nests in holes, or hiding them in other ways; they forage in trees and shrubs, mostly on insects and

345

they are either sedentary or migrate long distances. This group consists of 29 species representing 4 orders: Piciformes, Columbiformes, Apodiformes and Passeriformes. Some of these species are considered as characteristic of the forest interior or of the mature forest succession stage. Although many of these species are forest species they are also numerous in the neighborhood of human settlements. Some species show a preference for wooded habitats in urbanized areas, others are strictly connected with human settlements. Species grouped into the fourth strategy place their nests near the ground. They are insectivorous and most often forage in the herb or shrub layers; they are mostly tropical migrants and have the shortest incubation and fledging periods. This was the most numerous group and made up over 50% of the individuals in the avian communities in the study area. This was also the most uniform group, of 28 species all but the cuckoo belong to Passeriformes. The common feature of this group is a tendency toward long migrations. The substantial majority of these birds are tropical migrants. In contradiction to other strategies, forest habitats are not as important for this group. Much more important are wetland habitats with shrubby vegetation and patches of tall-forb vegetation. In these habitats the domination of these strategists reached more than 70%. The last strategy includes species belonging to the orders Columbiformes and Passeriformes. The characteristic feature of this group is its dependence on forest habitat; species build nests on trees and shrubs and often forage among trees. They undertake short migrations and raise the most broods of all strategies in the breeding season. Among the species belonging to this group insectivores are in minority and most species willingly take plant food. For simplicity I have named these five strategies: ‘terrestrial’, ‘predators’, ‘hole-nesters’, ‘vagrants’ and ‘arboreal’ bearing in mind that the characteristics of some species do not always entirely match the strategy name. The highest average number of breeding species in the study area was found for vagrants. In addition, this group was the most numerous group with the highest densities, lowest turnover rate and the highest variation in the density and num-

346

S. Tworek

ber of species. The terrestrial group was characterized by the highest turnover rate of species and the highest variation in numerical domination, while predators were characterized by the least average number of species, least numerical domination and density and the least variation in these parameters. With respect to domination, number of species and density of individuals, the arboreal strategy yielded precedence only to the vagrants strategy, and with respect to species turnover rate only to the terrestrial strategy. Hole-nesters reach mean values in the majority of the measured parameters. However, the difference between the maximum and mean number of species was highest in this group (Fig. 2).

Bird–habitat relationships

12

(a)

Number of species/plot

Number of pairs/ha

Depending on the strategy, the tested set of independent variables accounted for 30–85% of the variation in bird density (Table 3), 57–94% of the variation in strategy domination (Table 4), 77– 94% of the variation in the number of species (Table 5), and 42–71% of the variation in the species turnover rate (Table 6). The regression equa-

tions for each particular strategy differed from each other. Among terrestrial, in contrast to the other strategies, domination and number of species increased with the diminishing edge effect and the increasing proportion of agricultural land in relation to wooded area (factor 1). In wet sites, with a well-developed herb-layer (factor 2), the density, domination and number of species classified as terrestrial increased, while the species turnover rate decreased. In addition, the number of species increased when arable fields (factor 3) were a dominating element in the surroundings and the distance to a larger similar plot increased. The significance of factor 1 for the density, domination and number of species indicates that a close relationship exists between predator distribution and forests/coppices. The dependence of density on the shape of a site suggests a positive edge effect for this strategy. However, the number of species decreased with an increase in the proportion of wet areas with well-developed herb vegetation (factor 2). The influence or presence of humans and the absence of stands, particularly old stands (factor 3), exerts a similar effect on species abundance. The most pronounced adverse consequences of the edge

8 4 0

20

(b)

16 12 8 4 0

–4 T

P

H-N

V

A

T

(c)

50 Turnover (%)

Domination (%)

70 50 30 10

P

H-N

V

A

P

H-N

V

A

(d)

40 30 20 10

–10 T

P

H-N

V

T

A Bird strategies

Fig. 2. Results of (a) density of pairs (mean ± SD), (b) number of species (mean and range), (c) numerical domination (mean ± SD) and (d) turnover rate (mean ± SE) for five strategies of birds (T = terrestrial’, P = ‘predators’, HN = ‘hole-nesters’, V = ‘vagrants’, A = ‘arboreal’).

Bird strategies and habitat changes Table 3 Regression equations for densities of pairs in different strategies against variables characterizing habitat and landscape

Equation

Reg. coefficient (B)

‘Terrestrial’(R2 = 0.30) FACTOR 2 0.39 PERIMET 0.12 2 ‘Predators’(R = 0.46) SHAPE 0.19 FACTOR 1 0.24 ‘Hole-nesters’(R2 = 0.59) FACTOR 1 0.87 PERIMET 0.36 FACTOR 3 0.37 SHAPE −0.29 AREA −0.03 FACTOR 2 0.30 NEAREST −0.86 ‘Vagrants’(R2 = 0.85) SHAPE 2.37 FACTOR 2 2.24 FACTOR 1 2.03 PERIMET 0.51 ‘Arboreal’(R2 = 0.80) SHAPE 0.37 FACTOR 1 1.71 PERIMET 0.54 AREA −0.07 NEAREST −1.59 FACTOR 2 0.65

Standard error (SE)

P-value

0.12 0.04

0.0013 0.0069

0.02 0.06

<0.0001 0.0003

0.17 0.06 0.13 0.08 0.01 0.13 0.40

<0.0001 <0.0001 0.0042 0.0008 0.0169 0.0213 0.0377

0.26 0.41 0.53 0.20

<0.0001 <0.0001 0.0003 0.0154

0.14 0.29 0.11 0.02 0.65 0.22

0.0093 <0.0001 <0.0001 0.0012 0.0162 0.0043

effect were visible for hole-nesters. In this strategy, with an increase in the proportion of edge to interior of a site, the density of species decreased and the species turnover rate increased. Although these birds are clearly connected with stands (factor 1), their dependence on factor 3 indicates that their occurrence may also increase in other habitats, especially synanthropic ones, on the condition that old trees are present. A combination of wet habitats with well-developed herb-layer vegetation surrounded by meadows (factor 2) exerted the most adverse effect on the number of species and domination of this strategy, which indicated a close connection of this group with forests. A short distance to the nearest similar site had a positive

347

Table 4 Regression equations for the dominance of the strategies against variables characterizing habitat and landscape

Equation

Reg. coefficient (B)

‘Terrestrial’(R2 = 0.81) FACTOR 1 −7.05 PERIMET 5.25 HABITATS −3.33 SHAPE −2.61 FACTOR 2 2.52 2 ‘Predators’(R = 0.57) PERIMET 0.45 FACTOR 1 1.66 2 ‘Hole-nesters’(R = 0.78) FACTOR 3 4.70 FACTOR 1 4.03 PERIMET 1.43 FACTOR 2 −2.07 AREA −0.17 2 ‘Vagrants’(R = 0.94) PERIMET 6.60 FACTOR 2 6.50 FACTOR 3 −3.52 LARGER −4.34 ‘Arboreal’(R2 = 0.89) HABITATS 0.97 AREA −0.39 PERIMET 2.90 FACTOR 1 7.50 FACTOR 2 −3.20 NEAREST 8.22

Standard error (SE)

P-value

1.61 0.62 0.57 0.78 1.23

<0.0001 <0.0001 <0.0001 0.0011 0.0438

0.07 0.37

<0.0001 <0.0001

0.63 0.90 0.32 0.68 0.06

<0.0001 <0.0001 <0.0001 0.0029 0.0093

0.51 1.39 1.47 1.93

<0.0001 <0.0001 0.0189 0.0270

0.49 0.11 0.39 1.46 1.03 3.50

0.0488 0.0007 <0.0001 <0.0001 0.0025 0.0210

effect on densities in this group. Wetland habitats with rich lower-layer vegetation, surrounded by meadows (factor 2), have a significant effect on the density and domination of the vagrants strategy. However, isolation from other similar sites results in an increase in the number of species and a decrease in the turnover rate of this group. As in the arboreal strategy, the species turnover rate increased in vagrants with increasing homogeneity of a biotope. The most important difference between vagrants and arboreal species concerns their dependence on factor 2. Arboreal species, similar to predators and hole-nesters, are much closer connected with forests (factor 1) than vagrants strategists and, in addition, domination

348

S. Tworek

Table 5 Regression equations for species numbers in different strategies against variables characterizing habitat and landscape

Equation

Reg. coefficient (B)

‘Terrestrial’(R2 = 0.86) AREA 0.05 FACTOR 2 0.60 FACTOR 3 −0.36 FACTOR 1 −0.44 SHAPE −0.24 PERIMET 0.13 LARGER 0.32 ‘Predators’(R2 = 0.82) HABITATS 0.10 FACTOR 1 0.74 AREA 0.04 FACTOR 3 −0.19 FACTOR 2 −0.16 2 ‘Hole-nesters’(R = 0.77) HABITATS 0.21 FACTOR 1 1.77 FACTOR 2 −0.90 AREA 0.08 FACTOR 3 0.46 2 ‘Vagrants’(R = 0.94) HABITATS 0.88 AREA 0.13 NEAREST 2.87 FACTOR 1 0.67 ‘Arboreal’(R2 = 0.91) HABITATS 0.57 FACTOR 1 1.74 FACTOR 2 −0.72 AREA 0.07

Standard error (SE)

P-value

0.01 0.12 0.11 0.15 0.07 0.06 0.16

<0.0001 <0.0001 0.0027 0.0049 0.0021 0.0277 0.0469

0.02 0.10 0.008 0.08 0.08

<0.0001 <0.0001 <0.0001 0.0207 0.0457

0.05 0.27 0.20 0.02 0.20

<0.0001 <0.0001 <0.0001 <0.0004 0.0251

0.10 0.02 0.75 0.31

<0.0001 <0.0001 0.0002 0.0343

0.07 0.29 0.21 0.02

<0.0001 <0.0001 0.0010 0.0054

of this group increased with increasing isolation from other similar habitats. In canonical correlation analysis of the bird densities for particular strategies and variables describing sample sites, three canonical elements are statistically significant (P < 0.05). Variables describing the sites account for 42.5%, 17.5% and 11.5% of the variation in bird densities between the strategies. The site size and perimeter, presence of agricultural land and the herb-layer vegetation exert the greatest effect on the density of terrestrial species. The shape of a site, cover of the under-

Table 6 Regression equations for species turnover rates in different strategies against variables characterizing habitat and landscape

Equation

Reg. coefficient (B)

‘Terrestrial’(R2 = 0.57) PERIMET 5.49 AREA −0.96 FACTOR 2 −10.09 SHAPE 5.43 2 ‘Predators’(R = 0.61) PERIMET 10.12 AREA −1.31 FACTOR 3 15.31 2 ‘Hole-nesters’(R = 0.42) PERIMET 11.72 HABITATS −5.69 SHAPE −10.06 2 ‘Vagrants’(R = 0.68) PERIMET 7.85 AREA −0.58 NEAREST −23.26 HABITATS −3.68 FACTOR 3 4.02 LARGER 6.12 FACTOR 1 5.87 FACTOR 2 4.47 ‘Arboreal’(R2 = 0.71) PERIMET 13.86 HABITATS −5.99 FACTOR 3 8.24 AREA −0.78 SHAPE −6.07 NEAREST −22.56 FACTOR 2 7.49

Standard error (SE)

P-value

1.37 0.32 4.09 2.29

<0.0001 0.0029 0.0152 0.0194

1.89 0.04 4.10

<0.0001 0.0016 0.0004

2.86 2.22 4.84

0.0001 0.0124 0.0409

0.90 0.18 6.48 0.87 1.80 2.65 2.33 1.87

<0.0001 0.0013 0.0005 <0.0001 0.0276 0.0230 0.0133 0.0183

1.20 0.98 2.71 0.22 1.60 8.41 2.83

<0.0001 <0.0001 0.0030 0.0005 0.0003 0.0085 0.0093

growth layer and habitat humidity have an important influence on the density of vagrants, and to a smaller degree on predators and arboreal densities, while the presence of a stand, particularly with old trees, accounts for variation in the density of arboreal species (Fig. 3). In the canonical correlation analysis of domination of strategies and habitat variables, three canonical elements are statistically significant (P < 0.05). Variables describing the sites account for 45.2%, 22.5% and 15.5%, respectively, of the variation in domination of particular strategies.

16

12

12

Linkage distance

16

8

4

8

4

16

12

8

4

OLDTREES

PERIMET HABITATS SHRUB Arboreal TREEAGE CANOPY Hole-nesters

AREA Predators WATER HERB SHAPE Vagrants

AGRI

Terrestrial

0

Fig. 4. Tree diagram showing the relationship between variables describing habitats and landscape and domination of the five strategies of birds (‘terrestrial’, ‘predators’, ‘hole-nesters’, ‘vagrants’, ‘arboreal’) obtained for results of the correlation of variables with canonical roots. See Table 1 for an explanation of the variable abbreviations.

The site size and the proportion of agricultural land explain most variation in the domination of terrestrial species, and the site perimeter and shape, habitat humidity and the herb-layer vegetation are important for vagrants. Domination of

HERB

WATER PERIMET

AGRI

Terrestrial AREA Vagrants

Arboreal

OLDTREES

SHRUB Hole-nesters

CANOPY

TREEAGE Predators

SHAPE

HERB

AREA

PERIMET

SHAPE

WATER Terrestrial AGRI

SHRUB Vagrants

Arboreal

Predators

CANOPY

TREEAGE

Fig. 3. Tree diagram showing the relationship between variables describing habitats and landscape and densities of pairs of the five strategies of birds (‘terrestrial’, ‘predators’, ‘hole-nesters’, ‘vagrants’, ‘arboreal’) obtained for results of the correlation of variables with canonical roots. See Table 1 for explanation of the variable abbreviations.

Linkage distance

349

0 OLDTREES

0

Hole-nesters

Linkage distance

Bird strategies and habitat changes

Fig. 5. Tree diagram showing the relationship between variables describing habitats and landscape and the number of species of five strategies of birds (‘terrestrial’, ‘predators’, ‘hole-nesters’, ‘vagrants’, ‘arboreal’) obtained for results of the correlation of variables with canonical roots. See Table 1 for an explanation of the variable abbreviations.

hole-nesters shows the strongest relationship with the stand, especially an old stand, and domination of arboreal species with habitat diversity and the undergrowth-layer vegetation. The domination of predators had the weakest relationship with the tested set of independent variables. However, site size and the proportion of agricultural land had the strongest effect on predator variation and was much less important for terrestrial species (Fig. 4). In the canonical correlation analysis of species abundance in particular strategies and habitat variables, four canonical elements are statistically significant (P < 0.05). Variables describing the sites account for 53.9%, 25.4%, 7.3% and 6.8%, respectively, of the variation in the number of species in particular strategies. Size of the site and, to a lesser degree, the proportion of agricultural land exerts the strongest influence on the variation in the number of terrestrial species. The vegetation of canopy and undergrowth layers best explains the variation in the number of species in the predator, hole-nester and arboreal strategies, the presence of old stands is most important for hole-nesters and the undergrowth vegetation for predators. The number of species in the vagrant strategy is connected with habitat humidity, periphery of a site and herb-layer vegetation (Fig. 5).

350

S. Tworek

DISCUSSION Effects of habitat changes Bird densities, domination, species occurrence and turnover rate depend mainly on area and structure variables, but also on isolation and surrounding land use. These results are similar to many previous studies (Opdam et al. 1985; Blake & Karr 1987; Hinsley et al. 1995b; McGarigal & McComb 1995; Fernandez-Juricic 2000). However, the results obtained for the vagrants and arboreal strategies do not confirm the general opinion of nature conservators that habitat fragmentation produces numerous adverse effects (Merriam 1988; Saunders et al. 1991; Haila et al. 1993). As a result of human expansion, natural and seminatural ecosystems undergo progressive degradation. Their fragmentation because of environmental changes has major consequences and this factor has been called the principal threat to most species in the temperate zone (Wilcove et al. 1986), and even the single greatest threat to biological diversity (Noss 1991). In contrast, some scientists point to the positive effects of this process, accentuating a beneficial influence of habitat fragmentation on certain bird species and communities (Hagan et al. 1996; Hawrot & Niemi 1996; Petersen 1998). It is worth examining the nature of these relationships to explore the causes of these divergent opinions. A cause–effect mechanism triggered by habitat fragmentation that eventually results in a decrease in species diversity, together with all its consequences, quickly won popularity. This became a predominant view despite empirical evidence that after fragmentation birds were rare. Most of the evidence for this view was provided by studies on forest fragmentation in agricultural landscapes, which may not apply to other conditions (Andrén 1994; Diaz et al. 1998). Therefore, the perception of fragmentation effects is based on theoretical assumptions and predictions, particularly those connected with island biogeography, rather than on the results of empirical studies. The usefulness of the equilibrium theory of island biogeography in terrestrial conditions has been questioned many times (Gilbert 1980; Merriam 1988; Soberon 1992; Haila et al. 1993), but it has still dominated thinking about the effects of habitat fragmenta-

tion. In accordance with its guiding principles, most authors fixed their attention on measuring the selected variables (e.g. size of an area, degree of its isolation, number of species) while neglecting other important questions. The results obtained in this paper show that threats resulting from habitat fragmentation do not affect all birds to the same degree. The growth of area patchiness appears to have a positive effect on species belonging to vagrant and arboreal strategies, which is reflected in the values of the parameters measured in this study. In contrast, a negative effect is clearly revealed for the terrestrial and holenester strategies. In particular, the negative consequences of habitat fragmentation affect species living in the forest interior and species living in open wet biotopes. Large habitat extents protect conservative species that have difficulty adapting to environmental changes. Fragmentation, however, increases the attractiveness of an area for species that can adapt quickly to changes. These species are able to use ecological niches abandoned by conservative species. It is important to note that the results only permit determination of the ‘capacity’ of particular habitats for species representing different life styles, while factors commonly attributed to habitats with a high proportion of edge to interior are not taken into account. An increased pressure of predation and parasitism in these habitats may significantly affect the breeding success of birds (Rolstad 1991; Paton 1994). Scattered populations connected by dispersion of individuals between subpopulations form the basis of Levins’ (1970) model of metapopulation. When the rate of dispersion is high subpopulations may function as single populations (Harrison 1991). Therefore, the species representing holenester and arboreal strategies (i.e. connected with forest and having small dispersion abilities) should have the most clear metapopulation structure in the study area. However, the mode in which individuals’ move within the metapopulation depends not only on the dispersion abilities of the species, but also on the quality of the isolated habitat patches and their surroundings (Hinsley et al. 1995b). In this way, the relatively low determination coefficient of the regression equation of the turnover rate of hole-nesters may be explained by the absence of ‘links’ between isolated fragments of habitats and, thus, the significance of random

Bird strategies and habitat changes processes increases. Ecological corridors, which in the case of hole-nesters can be alleys of old trees, can fulfil the role of links. In this paper the linearity of an area was measured using values of the ‘shape’ variable. The results of the regression of the measured parameters for bird strategies indicate that predator, vagrant and arboreal strategists commonly use ecological corridors. In the case of predators, this result is mainly because of the domination of the magpie (Pica pica), which tends to breed in tree and shrub lines and avoid clearly larger forests. However, kestrels (Falco tinnunculus), crows (Corvus corone) and great gray shrikes (Lanius excubitor) often placed their nests in tree alleys. In addition, the buzzard (Buteo buteo) was most numerous in narrow tree lines and the kingfisher (Alcedo atthis) was dominate along the longest river. Habitats in ecological corridors are also occupied by other species from this group, for example, white stork (Ciconia ciconia) occupy single trees, alleys and electric poles, and hobby (Falco subbuteo), long-eared owl (Asio otus), little owl (Athene noctua) and rook (Corvus frugilegus) occupy tree belts. Thus, the dependences recorded (Tables 3,4) are not only the effect of the magpie. Vagrants are typical of the ecotone zone, except for a few species, and occupy various habitats in the ecological corridors. Similar results (Tables 3–6) were obtained for arboreal strategies. However, as a group with more specific habitat preferences, they occur only in corridors connecting forests. Terrestrial and hole-nester strategists avoid corridor habitats. In the case of terrestrial species, this probably results from their conservatism and increased pressure from predators. These are species of open habitats, so the neighborhood of other biotopes (e.g. forests) is connected with an additional risk of contact with hunting raptors. In the case of hole-nesters, the relationship obtained is caused by a different reason, that is, a scarcity of nesting places (holes) because old tree lines are much rarer than tree lines in earlier succession stages. The number of hole-nesters grows in these habitats when nesting boxes are placed there.

Implications for conservation The results of this study clearly show that groups of species can profit or can respond negatively to different types of habitats within a mosaic agricul-

351

tural landscape. Attempts to identify and interpret the importance of different habitat factors in determining the occurrence of breeding birds revealed that responses differ depending on the strategy adopted by the species. This means that there is a need to maintain a diversity of habitats to maximize bird biodiversity. The simplest way of increasing biodiversity in agricultural landscapes is by creating tree and shrub lines, alleys, green belts, woodlots and hedgerows. The importance of these habitats for birds has already been established (Parish et al. 1994, 1995; Sparks et al. 1996; Hinsley & Bellamy 2000). However, the creation and protection of meadows, pastures and wetlands is at least of equal importance (Petersen 1998) and is a question neglected by the majority of authors focusing their attention on the negative consequences of forest fragmentation, despite species nesting in open agricultural area often showing a greater decrease in numbers than forest species (Tucker & Heath 1994; Hagemeijer & Blair 1997). This finding is confirmed by the classification of Species of European Conservation Concern (SPEC) by Tucker and Heath (1994). Half of the species ascribed to terrestrial strategy have an unfavorable conservation status, while the least threatened are species belonging to the arboreal strategy, which are mostly species of forest margins: 95% of them have a favorable conservation status. Agricultural intensification affects most of the declining species. This threat, however, covers a range of specific threats including irrigation of dry grasslands and cereals and their subsequent conversion to dense, fast-growing heavily fertilized and pesticidetreated crops, lowering of the water table on grasslands followed by re-seeding fertilizer applications and high stocking levels, loss of hay meadows often due to their conversion to intensive silage crops and intensification of arable farmland through high inorganic fertilizer inputs, crop specialization and increasing field size. All these phenomena influence species of terrestrial and vagrants strategies and are considered to be the main threats to these species. Forest and tree loss (clear-cutting, unmanaged cutting, loss of trees from orchards, farmland copses, hedgerows, etc.) is reflected in the declining population trends of species representing the hole-nester strategy. Hunting is probably a significant contributory cause of decline in some other species of terrestrial

352

S. Tworek

strategy (game species), whereas persecution, which is still occurring, is the threat to the predator strategy species. Additional, and relatively recent, threats include egg-collecting and the taking of eggs and young birds for commercial purposes. Other common threats include wetland drainage for terrestrial and vagrant strategy species, disturbance by humans for terrestrial and predator strategy species, afforestation for terrestrial strategists and effects of pesticide toxicity for predator strategists. Among the positive factors, species protection and hunting protection are of particular importance for terrestrial and predator strategists, establishing of nesting boxes is important for holenester strategists, increasing the patchiness of agricultural land stimulates populations of species belonging to the vagrants group, while the introduction of different plantings (urban, rural, midfield, riverine) results in an increase in the numbers of arboreal strategists. Many species increase in numbers due to intrapopulation factors connected, most often, with changes in behavior or ecological niche or with a specific adaptation. Most of these species appeared in anthropogenic habitats and as a result the threat of their extinction diminished. The absence of endangered and critically endangered species in the study area results from the suboptimal character of the majority of sample sites surrounded by agricultural land. However, comparisons of bird communities in sites differing in a habitat character indicate the importance for birds of fragments not used for agricultural purposes, which from economic point of view are wasteland, but their positive effects on the occurrence of birds have been shown previously (Berg & Pärt 1994; Kurlavicius 1995; Tworek 1998). The results of this paper reveal that the importance of these habitats varies depending on the birds’ strategy. For example, woodlots or tree lines among fields positively influence species placing their nests on trees and shrubs. Simultaneously, if their network is too dense this habitat becomes less attractive for terrestrial strategists (see Figs 3–5). Thus, habitat changes in an agricultural landscape may influence birds positively or negatively and differences in bird responses can be attributed to different life styles or strategies based on both demographic characteristics and adaptations to environmental factors. In nature conservation, as

in most ecological studies, there is a tendency to polarize thinking, that is, theory on the one hand and practice on the other. In landscape ecology, a need for a complex approach to the functioning of biocenotic units in landscape, and not only of single species, must be emphasized (Opdam et al. 1994). It is important to note that general theories do not apply in many situations. At the same time it is impossible to undertake detailed studies on every species in every situation to establish stable scientific principles, which in future could be used in nature conservation and management of natural resources in any situation (Wiens 1994). The identification of indicative species with different sensitivities to environmental changes, depending on the type of ecosystem, may be a solution. Later on, these species could be used to generate different models and the management of other species, similar in respect to occurrence, ecological features or life histories may be able to be modelled after them (Collins et al. 1993; Wiens et al. 1993). In the present paper, I propose an alternative approach to the method of identification of indicative species. I tested whether similarity in ecological features and life histories reflects the distribution of birds and the parameters of their occurrence. The results obtained show that this approach can have a wider use in landscape ecology and can contribute to the development of a theory of patchiness that has a specific and limited range of application. However, I do not encourage the immediate transformation of these ideas, based on many theoretical assumptions and often limited empirical information, into management principles. The scientific principles of any nature conservation activity should be first verified in practice and more than once. This procedure enables a better understanding of the changes occurring in physiocenoses and whole landscape units.

ACKNOWLEDGEMENTS I am especially grateful to Zbigniew Gl¢owaci´nski and Mal¢gorzata Makomaska-Juchiewicz for their many helpful comments on the manuscript. I also wish to thank the farmers and landowners who have allowed me to work on their land and the Forest Inspectorate in Zabierzów for access to woodland and for help in assessing tree ages. In

Bird strategies and habitat changes 1996–1998 the study was supported by the State Committee for Scientific Research (No. 6P04G00910).

REFERENCES A NDRÉN H. (1994) Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat: a review. Oikos 71: 355–366. B ENTLEY J. M. & C ATTERALL C. P. (1997) The use of bushland, corridors and linear remnants by birds in southeastern Queensland, Australia. Conservation Biology 11: 1173–1189. B ERG A. & P ÄRT T. (1994) Abundance of breeding farmland birds on arable and set-aside at forest edges. Ecography 17: 147–152. B ERSIER L-F. & M EYER D. R. (1994) Bird assemblages in mosaic forests: the relative importance of vegetation structure and floristic composition along the successional gradient. Acta Oecologica 15: 561–576. B IBBY C. J., B URGESS N. D. & H ILL D. A. (1992) Bird Census Techniques. Academic Press, London. B LAKE J. G. & K ARR J. R. (1987) Breeding birds of isolated woodlots: area and habitat relationships. Ecology 68: 1724–1734. B ÖHNING -G AESE K. & B AUER H. G. (1996) Changes in species abundance, distribution, and diversity in Central European bird community. Conservation Biology 10: 175–187. C OLLINS S. L., G LENN S. M. & R OBERTS D. W. (1993) The hierarchical continuum concept. Journal of Vegetation Science 4: 149–156. C RAMP S. & S IMMONS K. E. L. (1993) Handbook of the birds of Europe, the Middle East and North Africa. The Birds of the Western Palearctic. Oxford University Press, Oxford. D IAMOND J. M. (1969) Avifaunal equilibria and species turnover rates on the Channel Islands of California. Proceedings of the National Academy of Sciences of the United States of America 69: 3199– 3203. D IAZ M., C ARBONELL C., S ANTOS T. & T ELLERIA J. L. (1998) Breeding bird communities in pine plantations of the Spanish plateaux: biogeography, landscape and vegetation effects. Journal of Applied Ecology 35: 562–574. F ERNANDEZ -J URICIC E. (2000) Bird community composition patterns in urban parks of Madrid: The role of age, size and isolation. Ecological Research 15: 373–383.

353

G ILBERT F. S. (1980) The equilibrium theory of island biogeography: fact or fiction? Journal of Biogeography 7: 209–235. G LUTZ VON B LOTZHEIM U. N. & B AUER K. (1966–1997) Handbuch der Vögel Mitteleuropas, Vol. 1–14. Akademische Verlagsgesellschaft, AULA-Verlag, Frankfurt. H AGAN J. M., V AN DER H AEGEN W. M. & M C K INLEY P. S. (1996) The early development of forest fragmentation effects on birds. Conservation Biology 10: 188–202. H AGEMEIJER E. J. M. & B LAIR M. J. (1997) The EBCC Atlas of European Breeding Birds: Their Distribution and Abundance. T. & A.D. Poyser, London. H AILA Y., H ANSKI I. K., N IEMELA J., P UNTILLA P., R AIVIO S. & T UKIA H. (1994) Forestry and the boreal fauna: matching management with natural forest dynamic. Annales Zoologici Fennici 31: 187–202. H AILA Y., S AUNDERS D. A. & H OBBS R. J. (1993) What do we presently understand about ecosystem fragmentation?. In: Nature Conservation 3: Reconstruction of Fragmented Ecosystems (eds R. J. Hobbs, P. R. Ehrlich & D. A. Saunders) pp. 45– 55. Surrey Beatty & Sons, Chipping Norton. H ANSEN A. J. & U RBAN D. L. (1992) Avian response to landscape pattern: the role of species’ life histories. Landscape Ecology 7: 163–180. H ANSKI I. & G ILPIN M. (1997) Metapopulation Biology: Ecology, Genetics and Evolution. Academic Press, London. H ANSKI I., M OILANEN A. & G YLLENBERG M. (1996) Minimum viable population size. American Naturalist 147: 527–541. H ARRISON S. (1991) Local extinction in a metapopulation context: an empirical evaluation. Biological Journal of the Linnean Society 42: 73–88. H AWROT R. Y. & N IEMI G. J. (1996) Effects of edge type and patch shape on avian communities in a mixed conifer-hardwood forest. Auk 113: 586–598. H INSLEY S. A. & B ELLAMY P. E. (2000) The influence of hedge structure, management and landscape context on the value of hedgerows to birds: a review. Journal of Environmental Management 60: 33–49. H INSLEY S. A., B ELLAMY P. E. & N EWTON I. (1995b) Bird species turnover and stochastic extinction in woodland fragments. Ecography 18: 41–50. H INSLEY S. A., B ELLAMY P. E., N EWTON I. & S PARKS T. H. (1995a) Habitat and landscape factors influencing the presence of individual

354

S. Tworek M AC A RTHUR R. H. & W ILSON E. O. (1967) The Theory of Island Biogeography. Princeton University Press, Princeton. M C G ARIGAL K. & M C C OMB W. C. (1995) Relationships between landscape structure and breeding birds in the Oregon Coast Range. Ecological Monographs 65: 235–260. M ERRIAM G. (1988) Landscape dynamics in farmland. Trends in Ecology and Evolution 3: 16– 20. N OSS R. F. (1991) Landscape connectivity: different functions at different scales. In: Landscape Linkages and Biodiversity (ed. W. E. Hudson) pp. 27–39. Island Press, Washington. O PDAM P., F OPPEN R., R EIJNEN R. & S CHOTMAN A. (1994) The landscape ecological approach in bird conservation: integrating the metapopulation concept into spatial planning. Ibis 137: 139– 146. O PDAM P., R IJSDIJK G. & H USTINGS F. (1985) Bird communities in small woods in an agricultural landscape: effects of area and isolation. Biological Conservation 34: 333–352. P ARISH T., L AKHANI K. H. & S PARKS T. H. (1994) Modelling the relationship between bird population variables and hedgerow, and other field margin attributes. I. Species richness of winter, summer and breeding birds. Journal of Applied Ecology 31: 764–775.

breeding bird species in woodland fragments. Journal of Avian Biology 26: 94–104. J OKIMÄKI J. & H UHTA E. (1996) Effects of landscape matrix and habitat structure on a bird community in northern Finland: a multi-scale approach. Ornis Fennica 73: 97–113. J ONGMAN R. H. G., T ER B RAAK C. J. F. & V AN T ONGEREN O. F. R. (1987) Data Analysis in Community and Landscape Ecology. Pudoc, Wageningen. K AREIVA P. (1990) Population dynamics in spatially complex environments: theory and data. Philosophical Transactions of the Royal Society of London. Series B 330: 175–190. K UJAWA K. (1997) Relationships between the structure of mid-field woods and their breeding bird communities. Acta Ornithologica 32: 175– 184. K URLAVICIUS P. (1995) Birds of Forest Islands in South-East Baltic Region. Baltic ECO, Vilnius. L EVINS R. (1970) Extinctions. In: Some Mathematical Questions in Biology. Lectures on Mathematics in the Life Sciences. Amer. Math. Soc. Vol. 2. pp. 77–107. Providence, Rhode Island. L OVEJOY T. E. & O REN D. C. (1981) The minimum critical size of ecosystem. Forest island dynamics in man-dominated landscapes. In: Ecological Studies 41 (eds D. M. Sharp & R. L. Burgess) pp. 7–12.

APPENDIX I Categorisation of variables describing bird species used for determining their strategies Categories: type of nest: (1), open uncovered; (2), open covered or semi-open; (3), enclosed; (4), hidden location of nest; (5), on the ground; (6), in vegetation up to 1.5 m high; (7), in a hole or den; (8), on trees or shrubs above 1.5 m high; (9), at the water; (10), on anthropogenic elements; place and way of foraging: (11), in water; (12), on the ground; (13), in the undergrowth zone; (14), in the trunk and branches zone; (15), in the canopy zone; (16), actively in flight; type of food: (17), green parts of plants; (18), fruits; (19), seeds; (20), invertebrates except for insects; (21), insects; (22), vertebrates; migration status: (23), sedentary; (24), nomadic; (25), SPECIES

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Accipiter gentilis* Accipiter nisus Acrocephalus palustris Acrocephalus schoenobaenus Acrocephalus scirpaceus Aegithalos caudatus* Alauda arvensis Alcedo atthis Anas platyrhynchos

1 1 1 0 1 0 0 0 1

0 0 1 1 1 0 1 0 1

0 0 0 0 0 1 0 0 0

0 0 0 0 0 0 0 1 0

0 0 0 1 0 0 1 0 1

0 0 1 1 1 1 0 0 0

0 0 0 0 0 0 0 0 0

1 1 0 0 0 1 0 0 1

0 0 0 0 1 0 0 1 1

0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 1 1

1 1 0 0 0 0 1 0 1

0 0 1 1 1 1 0 0 0

1 1 0 0 0 1 0 0 0

0 1 0 0 0 1 0 0 0

1 1 0 1 0 0 0 1 0

0 0 0 0 0 0 1 0 1

0 0 0 0 1 0 1 0 1

0 0 0 0 0 0 1 0 1

0 0 1 0 1 0 0 0 1

Bird strategies and habitat changes

355

Breeding birds in field boundaries in an agricultural landscape. Agriculture Ecosystems and Environment 60: 1–8. S TEARNS S. C. (1992) The Evolution of Life Histories. Oxford University Press, Oxford. S TEVENS J. (1986) Applied Multivariate Statistics for the Social Sciences. Erlbaum, Hillsdale. T OMIAL¢ OJ c´ L., W ESOL¢ OWSKI T. & W ALANKIEWICZ W. (1984) Breeding bird community of primeval temperate forest (Bialowieza National Park, Poland). Acta Ornithologica 20: 241–310. T UCKER G. M. & H EATH M. F. (1994) Birds in Europe: their conservation status. BirdLife International, Cambridge. T WOREK S. (1998) The importance of habitat diversity for the breeding avifauna of agricultural land in the Rudawa River valley near Kraków. Chron´ my Przyrode¸ Ojczysta¸ 54: 39–52. W IENS J. A. (1994) Habitat fragmentation: island vs landscape perspectives on bird conservation. Ibis 137: 97–104. W IENS J. A., S TENSETH N. C., V AN H ORNE B. & I MS R. A. (1993) Ecological mechanisms and landscape ecology. Oikos 66: 369–380. W ILCOVE D. S., M C L ELLAN C. H. & D OBSON A. P. (1986) Habitat fragmentation in the temperate zone. In: Conservation Biology. The Science of Scarcity and Diversity (ed. M. E. Soul) pp. 237–256. Sinauer Associates, Massachusetts.

P ARISH T., L AKHANI K. H. & S PARKS T. H. (1995) Modelling the relationship between bird population variables and hedgerow, and other field margin attributes. II. Abundance of individual species and of groups of similar species. Journal of Applied Ecology 32: 362–371. P ATON P. W. C. (1994) The effect of edge on avian nest success: how strong is the evidence? Conservation Biology 8: 17–26. P ETERSEN B. S. (1998) The distribution of Danish farmland birds in relation to habitat characteristics. Ornis Fennica 75: 105–118. P ICKETT S. T. A. & W HITE P. S. (1985) The Ecology of Natural Disturbances and Patch Dynamics. Academic Press, Orlando. R OLSTAD J. (1991) Consequences of forest fragmentation for the dynamics of bird populations: conceptual issues and the evidence. Biological Journal of the Linnean Society 42: 149–163. S AUNDERS D. A., H OBBS R. J. & M ARGULES C. R. (1991) Biological consequences of ecosystem fragmentation: a review. Conservation Biology 5: 18–29. S NOW D. W. & P ERRINS C. M. (1998) The Birds of the Western Palearctic. Concise Edition. Oxford University Press, Oxford. S OBERON J. M. (1992) Island biogeography and conservation practice. Conservation Biology 6: 161. S PARKS T. H., P ARISH T. & H INSLEY S. A. (1996)

European migrant; (26), Tropical migrant; number of broods per year: (27), one brood; (28), two broods; (29), more than two broods; clutch size: (30), 1–2 eggs; (31), 3–5 eggs; (32), 6–8 eggs; (33), at least 9 eggs; incubation period: (34), up to 12 days; (35), 13–16 days; (36), 17–21 days; (37), 22–29 days; (38), at least 30 days; fledging period: (39), up to 12 days; (40), 13–16 days; (41), 17–22 days; (42), 23–30 days; (43), longer than 30 days. (S), strategy: (T), ‘terrestrial’, (P), ‘predators’, (HN), ‘hole-nesters’, (V), ‘vagrants’, (A), ‘arboreal’. 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

S

0 0 1 1 1 1 1 1 1

P P V V V A T P T

1 1 0 0 0 0 0 1 0

1 1 0 0 0 0 0 0 0

0 1 0 0 0 1 0 1 1

0 0 0 0 0 0 1 1 1

0 0 1 1 1 0 0 0 0

1 1 1 0 1 0 0 1 1

0 0 0 1 1 1 1 1 0

0 0 0 0 0 0 1 0 0

1 0 0 0 0 0 0 1 0

1 1 1 1 1 0 1 1 0

0 1 0 1 0 1 0 0 0

0 0 0 0 0 1 0 0 1

0 0 1 0 1 1 1 0 0

0 0 1 1 0 1 0 0 0

0 0 0 0 0 0 0 1 0

0 0 0 0 0 0 0 0 1

1 1 0 0 0 0 0 0 0

0 0 1 0 1 0 0 0 0

0 0 0 1 0 1 0 0 0

0 0 0 0 0 1 1 1 0

0 1 0 0 0 0 0 0 0

1 1 0 0 0 0 0 0 1

356

S. Tworek

APPENDIX I Continued SPECIES

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Anthus pratensis Anthus trivialis Apus apus* Asio otus* Athene noctua* Buteo buteo Carduelis cannabina Carduelis carduelis Carduelis chloris Carpodacus erythrinus Certhia brachydactyla Certhia familiaris Ciconia ciconia Ciconia nigra* Circus aeruginosus Coccothraustes coccothraustes Columba oenas Columba palumbus Corvus corax Corvus corone Corvus frugilegus* Corvus monedula* Coturnix coturnix Crex crex Cuculus canorus Delichon urbica* Dendrocopos major Dendrocopos medius Dendrocopos minor Dendrocopos syriacus Dryocopus martius Emberiza citrinella Emberiza schoeniclus Erithacus rubecula Falco subbuteo Falco tinnunculus Ficedula albicollis Ficedula hypoleuca Fringilla coelebs Gallinago gallinago Gallinula ch;oropus Garrulus glandarius Hippolais icterina Hirundo rustica* Jynx torquilla Lanius collurio Lanius excubitor Larus ridibundus* Locustella fluviatilis Locustella naevia Luscinia luscinia

0 0 0 1 0 1 1 1 1 1 0 0 1 1 1 1 0 1 1 1 1 0 0 0 1 0 0 0 0 0 0 1 0 1 1 1 0 0 1 1 0 1 1 0 0 1 1 1 0 0 1

1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 1 1 1 0 1 0 0 0 0 1 0 1 0 0 1 0 0 1 1 1

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 1 0 1 0 0 0 0 0 1 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 1 1 1 1 0 0 0 0 0 1 1 0 0 0 0 0 1 1 0 0 0 0 0 0

1 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 1 1

0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 1

0 0 1 0 1 0 0 0 0 0 1 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 1 0 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0

0 0 0 1 0 1 1 1 1 0 0 0 1 1 0 1 0 1 1 1 1 0 0 0 1 0 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 1 1 0 0 1 1 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 0 0

0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0

1 1 0 1 1 1 1 0 1 1 0 0 1 1 1 0 1 1 1 1 1 1 1 1 1 0 0 0 0 1 0 1 1 1 0 1 1 1 1 1 1 1 0 0 1 1 1 1 0 1 1

0 0 0 0 1 0 1 1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 0 0 1 1 0 0 1 1 0 1 1 1

0 1 0 1 1 1 0 0 0 0 1 1 0 0 0 0 1 1 1 1 0 0 0 0 1 0 1 1 1 1 1 0 0 0 0 0 1 1 1 0 0 1 0 0 1 1 1 0 0 0 0

0 1 0 1 0 0 1 1 1 1 0 0 0 0 0 1 1 1 0 0 0 0 0 0 1 0 1 1 1 1 0 0 0 1 0 0 1 1 1 0 0 1 1 0 0 1 1 0 0 0 1

1 0 1 1 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 1 0 1 1 0 0 0 0

0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 1 0 0 1 0 1 1 0 0 0 0 0 1 0 1 1 0 0 0 0 0 1 0 0 1

0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 1 1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 1 0 0 0 0 0 0 1 1 0 1 0 0 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 1 1 0 0 0 0 0 1 0 0 1

Bird strategies and habitat changes

357

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 1 1 1 0 1 1 0 1 0 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

0 0 0 1 1 1 0 0 0 0 0 0 1 1 1 0 0 0 1 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0 0 1 1 1 0 0 0

0 0 0 1 1 0 1 1 1 0 1 1 0 0 0 1 0 0 1 1 1 1 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1 1 0 0 0

0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 1 0 0 0 1 1 1 0 0 0 0 0 0 1 0 0 1 1 0 0 1 0 0 1 0 0 1 0 0 0 0 1 1 0 0 0

1 0 0 0 0 1 1 0 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0

0 1 1 0 0 0 0 0 0 1 0 0 1 1 1 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 0 1 1 1 1 0 0 1 1 1

1 1 1 1 1 1 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 1 1 1 0 1 1 1 1 1 1 0 0 0 1 1 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1

1 1 0 1 0 0 1 1 1 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 1 1 0 0 0 0 0 0

0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0

0 0 1 1 0 1 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0

1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 0 0 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 0 1 1 1 0 1 1 1 1 1 1

0 1 0 1 0 0 1 1 1 0 1 1 0 0 1 0 0 0 1 1 0 1 0 0 1 1 1 1 1 1 1 0 0 1 0 1 1 1 0 0 1 1 0 1 1 1 1 0 1 1 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0

1 1 0 0 0 0 1 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 0 0 1 0 1 0 0 0 1 0 1 0 0 0 0 1 1

1 1 0 0 0 0 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 0 0 0 0 1 1 1 0 1 1 1 1 1 0 0 1 1 1 0 0 0 1 1 1 1 1 0 1 1 1

0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 1 1 0 0 0 0

0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0

0 0 0 0 1 1 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 0 0 0 0 1 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1

1 1 0 0 0 0 1 1 1 1 1 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 1 1 1 0 0 0 1 0 0 1 1 0 1 0 0

0 0 0 0 0 0 0 1 1 0 1 1 0 0 0 0 1 1 0 0 0 0 1 0 1 1 1 1 1 0 0 0 0 1 0 0 1 1 0 1 0 1 0 1 1 0 1 0 0 0 0

0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 1 1 0 0 0 1 1 1 0 1 1 0 0 0 1 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0

0 0 1 1 1 1 0 0 0 0 0 0 1 1 1 0 0 1 1 1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0

S V V HN P P P A A A V HN HN T T P A HN A P P P P T T V HN HN HN HN HN HN A A V P P HN HN A T T P V HN HN V P P V V V

358

S. Tworek

APPENDIX I Continued SPECIES

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Luscinia megarhynchos Motacilla alba Motacilla flava Muscicapa striata Oenanthe oenanthe Oriolus oriolus Parus ater Parus caeruleus Parus major Parus montanus Parus palustris Passer domesticus Passer montanus Perdix perdix Phasianus colchicus Phoenicurus ochruros Phoenicurus phoenicurus Phylloscopus collybita Phylloscopus sibilatrix Phylloscopus trochilus Pica pica Picus canus Picus viridis Porzana porzana Prunella modularis Pyrrhula pyrrhula Regulus regulus Remiz pendulinus Riparia riparia Saxicola rubetra Saxicola torquata Scolopax rusticola Serinus serinus Sitta europaea Streptopelia decaocto Streptopelia turtur Strix aluco Sturnus vulgaris Sylvia atricapilla Sylvia borin Sylvia communis Sylvia curruca Sylvia nisoria Tringa totanus Troglodytes troglodytes Turdus merula Turdus philomelos Turdus pilaris Vanellus vanellus

1 0 0 0 0 1 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 1 0 1 1 1 0 1 1 1 1 1 1 0 1 1 1 1

* non-breeding species

1 1 1 1 1 0 0 0 0 1 1 0 1 1 1 1 1 0 0 0 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 0 1 1 0 0

0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0

0 0 0 0 1 0 1 1 1 1 1 1 1 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 1 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0

1 1 1 0 1 0 0 0 0 0 0 0 0 1 1 0 0 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 1

1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 1 0 0 0 1 1 0 0 0 0 0 0 0 1 1 1 1 1 0 1 0 1 0 0

0 0 0 1 1 0 1 1 1 1 1 1 1 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 0 1 1 0 0 0 0 1 0 1 1 1 0 1 1 0 0 0 0 1 1 1 1 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 1 0 1 1 0 0 1 1 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0

0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0

1 1 1 0 1 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 1 1 1 0 1 1 1 1 1 0 0 0 0 1 1 0 1 1 1 1

1 0 0 0 1 1 0 0 0 1 1 1 0 0 0 0 0 1 0 0 1 0 0 0 1 1 0 0 0 1 1 0 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 1 0

0 0 0 1 0 0 1 1 1 1 1 1 1 0 0 1 1 0 1 0 1 1 1 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 0 0 0 0 0 0 1 1 1 0

1 0 0 0 0 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 0 0 0 1 1 1 1 0 0 0 0 1 0 1 1 0 1 1 1 0 1 1 0 0 1 1 1 0

1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0

0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0

1 0 0 0 1 1 0 0 0 0 0 1 1 1 1 0 0 0 0 0 1 0 0 1 0 1 0 0 0 0 1 1 1 0 1 1 0 1 0 0 0 0 0 0 0 1 1 1 0

0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 1 0 0 1 0 1 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 0 1 0 0 0 0 0 0 0 0 1 1 1 1 0 1 0 1 0 0 1 1 0 0 0 0 1 1 1 0 0 0 0 1 1 0 0 0 0 1 1 0 1 1 1 1

Bird strategies and habitat changes

359

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 0 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 0 0 0 0 0 1 1 0 0 0

0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0

0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 1 0 1 1 0 0 1 1 1 0 0 0 0 1 1 0 0 0 0 1 1 1 1 1 1

1 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 1 0 0 0 1 1 1 1 0 0 0 0 0 0

1 0 1 1 1 1 0 1 0 1 1 0 0 1 1 0 1 0 1 1 1 1 1 1 0 0 0 1 1 1 0 1 0 1 0 0 1 1 0 1 1 1 1 1 0 0 0 1 1

0 1 1 1 1 0 1 1 1 0 1 1 1 0 0 1 1 1 1 0 0 0 0 0 1 1 1 1 1 1 1 0 1 0 0 1 0 1 1 1 1 1 0 0 1 1 1 1 0

0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 0 0 0 0 0 1 1 0 0 1 0 1 1 1 1 0 1 0 1 1 0 0 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 1 1 1 1

0 1 1 1 1 0 1 0 1 1 1 0 1 0 0 1 1 1 1 1 1 1 1 0 1 0 0 1 1 1 1 0 0 1 0 0 0 1 1 0 1 1 0 0 1 0 0 1 0

0 0 0 0 0 0 1 1 1 1 1 0 0 1 1 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 0 0 0 1 0 0 1 1 0 0 0 1 0 1 1 0 0 0 0 1 1 0 0 1 1 0 0 1 0 0 0 0 1 1 1 1 1 1 0 0 1 1 1 0

1 1 1 1 1 0 1 1 1 1 1 1 1 0 0 1 1 1 1 1 0 1 0 0 1 1 1 1 1 1 1 0 1 1 1 1 0 1 0 0 0 1 1 0 1 1 1 1 0

0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 1 1 0 0 1 0 0 0 0 1 0 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 1

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 0 1 0 0 0 0 0 0 1 0 0 1 1 0 0 1 1 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0 0 1 1 1 1 1 0 0 1 1 1 0

0 1 1 1 1 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 0 0 1 0 0 0 1 1 1 1 0 1 0 0 0 1 0 0 1 0 0 1 1 1 1 0

0 0 0 0 1 1 1 1 1 1 1 1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 1 1 0 1 1 0 1 0 0 0 0 0 0 1 1 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 1

S V V V HN V V HN HN HN HN HN A HN T T HN HN V V V P HN HN T A A A HN HN V V T A HN A A P HN V V V V V T A A A A T