Content uploaded by Vassiliki I Kati
Author content
All content in this area was uploaded by Vassiliki I Kati on Jun 03, 2019
Content may be subject to copyright.
Journal for Nature Conservation 17 (2009) 47—59
Ecological management of a Mediterranean
mountainous reserve (Pindos National Park,
Greece) using the bird community as an indicator
Vassiliki Kati
a,
, Panayotis Dimopoulos
a
, Haritakis Papaioannou
a,b
,
Kostas Poirazidis
c
a
Department of Environmental and Natural Resources Management, University of Ioannina,
Seferi 2, 30100 Agrinio, Greece
b
Center of Biological and Cultural Diversity, 44004 Papingo, Greece
c
Department of Forestry, Environment and Natural Resources,
Democritus University of Thrace, 68200 Orestiada, Greece
Received 23 April 2008; accepted 5 December 2008
KEYWORDS
Ecological structure;
Forest management;
Indicators;
NATURA 2000;
Passerines;
Protected area
Summary
We used the community of passerines and woodpeckers as a target group for the
conservation management of Pindos National Park (NW Greece). We conducted
bird point counts twice during springtime in 72 plots that represented the
main vegetation types (16 sites). We recorded 56 species (14 of conservation
concern-SPEC). The montane grasslands were the most important habitats in
terms of species of conservation concern, whereas the agricultural mosaics were the
most species-rich habitats. The mixed pine-beech woods were significantly
richer than the pinewoods, whereas pinewoods and broad-leaved woods did not
differ significantly between them. The bird diversity was significantly correlated
with the number of tree layers, the vertical structural complexity and the maximum
height of trees. The presence of grassland, forest and agricultural habitat type, as
well as the altitude and the vegetation structural complexity were the main
environmental parameters determining species composition (Canonical Correspon-
dence Analysis). We identified a set of 17 typical species (IndVal analysis) to be used
in the monitoring scheme of the Park, which were characteristic of the main bird
habitat types distinguished by Ward’s hierarchical clustering. Conservation measures
should involve maintenance of the traditional agricultural practices, montane
ARTICLE IN PRESS
www.elsevier.de/jnc
1617-1381/$ - see front matter &2008 Elsevier GmbH. All rights reserved.
doi:10.1016/j.jnc.2008.12.002
Corresponding author. Tel.: +30 26510 70993; fax: +30 2641033716.
E-mail addresses: vkati@cc.uoi.gr,kikikati@hotmail.com (V. Kati).
grasslands, old growth woods, as well as the vertical vegetation complexity and high
trees in forest stands.
&2008 Elsevier GmbH. All rights reserved.
Introduction
Birds exhibit a diverse range of ecological
functions, and benefit humans by providing a wide
range of ecosystem services (Sekercioglu, 2006).
However, birds also constitute one of the most
vulnerable groups of organisms, with many species
currently undergoing significant global population
declines, leading to detrimental consequences
for total biodiversity and ecosystem function
(Sekercioglu et al., 2004).
Because birds are, especially in Europe, among
the best-studied organisms, they are often used to
evaluate the impact of human activities (BirdLife
International, 2004;Heath et al., 2000). On
account of their ubiquitousness, terrestrial passer-
ine communities in particular have been used as
indicators of environmental change (Bani et al.,
2005;Gregory et al., 2004), or as ecological
substitutes for other, more difficult to quantify
groups of organisms (Howard et al., 1998;Kati
et al., 2004a;Lombard, 1995;Prendergast et al.,
1993). Although their use as indicator taxa has been
sometimes criticised for its potentially low extra-
polative value to other groups of organisms
(Lindenmayer et al., 2006;Simberloff, 1998), small
terrestrial birds are considered as an important
component of biodiversity and have been inte-
grated into studies evaluating its conservation
(Dobson et al., 1997;Kati et al., 2004b;Lawton
et al., 1998;Thiollay, 2002;Vessby et al., 2002).
Thus, small terrestrial birds have often been used
to assess and guide ongoing forest management
practices (e.g. Donald et al., 1998;Gil-Tena et al.,
2007;Johnson & Freedman, 2002;King & DeGraaf,
2000;Laiolo et al., 2003;Lance & Phinney, 2001;
Muller et al., 2007;Sekercioglu, 2002) or current
agricultural practices (e.g. Chamberlain et al.,
2000;Donald et al., 2001;Pain & Pienkowski,
1997). In this applied context, the European Union
has recently adopted, for the first time, an
ecological index that considers the conservation
status of common farmland birds (Farmland Bird
Index) as a formal indicator of sustainable devel-
opment in Europe (structural-sustainable develop-
ment indicator). A prerequisite step before using
small terrestrial birds as an indicator group is to
obtain a deeper understanding of the underlying
patterns in diversity and the ecological structure of
this community.
In the present study we conducted an ecological
study, considering the land bird community (passerine
and woodpecker species) as a target group, in order
to provide a guideline for the ecological management
of a remote area, surrounding the mountainous
Pindos (or Valia Calda) National Park in NW Greece.
The Park was created in 1966 and is considered to be
one of the most important protected areas for the
maintenance of mountainous biodiversity and eco-
system integrity at national scale. However, existing
knowledge on the biological diversity of the area
remains fragmentary (Mertzanis, 1999;Tsounisetal.,
1985). The Park is under the administrative authority
of a recently created (2003) Management Body, which
is responsible for the ecological management of a
much larger region that includes eight protected
areas of the NATURA 2000 network. This authority
together with the local Forest Service is responsible
for regulating human activities and maintaining the
ecological value of Pindos National Park (NP).
However, this target remains elusive, due to a lack
of expertise and of a science-based management plan
for the NP. Thus a great need exists to conduct
conservation-oriented research that can be trans-
lated directly into practical management proposals.
The present research attempts to fill this gap in
relation to the bird community. Specifically, our
objectives are: (a) to assess the relative importance
of different habitat types for avian conservation; (b)
to test whether tree composition, vertical structural
complexity and tree height affect bird community in
forest stands; (c) to analyse the ecological structure
of bird communities and to detect the main environ-
mental factors regulating species diversity and
distribution; and, (d) to identify species that are
characteristic for the main bird habitat types
distinguished. On the applied level, we attempt to
translate these findings into practical monitoring and
management schemes for the NP. Our findings could
provide a management guideline for other national
parks throughout the Greek mountains and in the
Mediterranean area.
Methods
Study area and site selection
The study area is located in north-western
Greece (391540N, 211070E) and covers a surface of
ARTICLE IN PRESS
V. Kati et al.48
16,000 ha. The National Park covers 6868 ha and
consists of two sites of the European NATURA
2000 network. The strictly protected core area
(GR1310002) covers 3294 ha, and has been declared
a ‘‘biogenetic reserve’’ by the European Council;
human activities such as woodcutting, grazing,
hunting, and access by car are not allowed. The
buffer zone (GR1310003) covers 3574 ha and human
activities are controlled (Figure 1). The region is
extremely mountainous with elevations ranging
from 900 to 2177 m asl. The climate is montane
and varies depending on elevation and aspect;
annual rainfall ranges between 1000 and 1800 mm
whereas mean monthly temperatures vary between
0.9 and 21.4 1C(Trakolis et al., 1996). The
dominant vegetation type is black pine forest
(Pinus nigra) reaching up to 1700 m, whereas
common beech forest (Fagus sylvatica) covers the
northern slopes up to 1800 m. Sub-alpine grasslands
extend above 1800 m to peaks often dotted with
Balkan pines (Pinus leucodermis). The region is
sparsely populated and human activities mainly
include logging and livestock breeding, both regu-
lated by the local Forest Service, as well as
agriculture of very limited scope.
We selected 16 sites representative of the main
habitat types of the study area on the basis of the
standard European habitat typology (European
Commission, 2003)(Table 1). Sampling covered
ten forest sites containing pinewoods (P), beech
ARTICLE IN PRESS
Figure. 1. Habitat types and sampling sites in the broader area of Pindos National Park, Greece.
Bird communities and reserve management 49
ARTICLE IN PRESS
Table 1. Description of the sites sampled and bird diversity. Habitat codes refer to the Appendix I of the Directive 92/43/EEC (European Commission 2003).
Site code Habitat type Site description Bird diversity of sites
Habitat code Area
(ha)
ALT
(m)
NL 1/D MH
(m)
Land
use
Zone SHMean
S
Mean
WS
Points
P1 Sub-Mediterranean pine forests with endemic
black pines (Pinus nigra)
9530 20 1519 2.00 0.39 12–20 w 1, 2 11 3.17 4.83 6.33 5
P2 9530 20 1442 2.67 0.40 20–35 –1 20 3.18 9.50 12.00 5
P3 9530 20 1130 4.00 0.39 20–25 –3 20 3.15 9.60 11.80 5
P4 9530 20 1452 2.20 0.39 16–24 –1 16 3.19 8.80 11.20 5
P5 Mediterranean pine forests with endemic
Mesogean pines (Pinus leucodermis)
9540 15 1572 3.00 0.38 18–22 –1 7 3.16 7.00 7.00 3
B1 Asperulo-Fagetum beech forest (Fagus sylvatica) 9130 20 1419 2.86 0.26 16–25 –1 17 2.93 6.14 6.71 5
B2 9130 20 1558 2.29 0.31 12–17 w 2 15 2.85 6.14 9.00 5
PB1 Mixed pine-beech woods 9530 9130 20 1356 3.50 0.39 20–35 w –20 3.2 9.00 13.50 5
PB2 20 1479 3.20 0.38 20–35 w –19 3.21 8.60 10.80 5
R Constantly flowing Mediterranean rivers with Salix
and Populus alba
3280 20 964 2.80 0.39 15–22 w –20 3.2 7.20 8.40 5
G1 Semi-natural dry grasslands and scrubland facies
on calcareous substrates
6210 20 1714 2.17 0.35 g 2 18 3.1 8.33 13.17 5
G2 Endemic oro-Mediterranean heaths with gorse 4090 20 1733 2.17 0.36 w, g –18 3.18 5.33 9.33 5
G3 4090 10 1412 2.00 0.38 g 2 7 3.18 5.00 11.00 2
G4 4090 10 1461 2.00 0.39 g 1 15 3.18 9.00 17.50 2
A1 Rural mosaics 1020 20 986 2.20 0.28 g –19 2.57 7.80 9.20 5
A2 1020 20 1221 2.80 0.29 g, a –23 2.89 10.00 14.40 5
ALT: altitude, NL: mean number of vegetation layers, 1/D: mean Simpson reciprocal index of vertical structure, MH: maximum height of upper tree layer, S: species richness, H: Shannon
diversity index, MeanWS: mean weighted species richness, 1: NP core area, 2: NP buffer zone, 3: sacred grove a: agricultural use, g: grazing, w: woodcutting.
V. Kati et al.50
woods (B), mixed woods (PB) and riverine vegeta-
tion (R), as well as four grassland sites (G) and two
agricultural ones (A) (Table 1). Most of the sites
sampled were located inside areas protected by the
NATURA 2000 network, and one site (P3) was
strictly protected as a sacred grove.
Bird sampling
We surveyed the bird communities at 72 point
locations, recording all passerine and woodpecker
species seen or heard within a circle of a 100 m
radius of each point. Points were located at a
distance of 200 m between them and at least 100 m
from site edge, in order to avoid double counting
and edge effects. All sites (20 ha) were represented
by five points with the exception of three smaller
sites (10–15 ha) that were covered with 2–3 points.
To cover both sedentary species and late-arriving
migrants, we sampled every site twice, once in late
spring (30 April–30 May 2003) and once in early
summer (5 June–5 July 2003). In our bird count
methodology (see Bibby et al., 1992;Blondel et al.,
1970), a singing, territorial male represented a
breeding pair and was therefore counted as two
individuals. Every call different from the male
breeding song was counted as one individual. Each
count lasted for 10 min and took place between
30 min before and four hours after sunrise.
Measurement of environmental variables
We located a smaller quadrate (50 m 50 m) at
the centre of the each one of the 72 bird survey
points where we sampled the four main vegetation
layers: (1) upper tree layer (47 m); (2) lower tree
layer (4.1–7 m); (3) upper shrub layer (2.1–4 m); (4)
lower shrub layer (0.5–2m) (Kent & Coker, 1994).
We visually determined the percentage cover
(relative area occupied by the vertical projection
of all aerial parts of woody plants as a percentage
of the surface area of the sample plot) for the
separate layers (van der Maarel, 2005), using the
following vegetation cover classes (Ku¨chler, 1988)
1¼1–5%, 2 ¼6–25%, 3 ¼26–50%, 4 ¼51–75%,
5¼76–100%. For each quadrate we recorded the
altitude (ALT),the overall vegetation cover (COV),
the maximum height of the upper tree layer (MH)in
forest habitats, the number of vegetation layers
(NL) and the reciprocal of Simpson’s diversity index
(1/D) as a measure of vertical vegetation complex-
ity (Magurran, 2004). We also noted the general
habitat category the station referred to (agricul-
ture, forest, and grassland).
Data analysis
We estimated the overall number of species of
the study area using the software EstimateS
(Colwell 2006) with the option of 1000 randomisa-
tions. We opted for non-parametric species esti-
mators, which consider mainly the presence of rare
species, recorded by one or two individuals
(Chao 1), or present in one or two point counts
(Chao 2) (Magurran, 2004). Species diversity was
estimated in terms of species richness (S),
weighted species richness (WS), and Shannon–Wei-
ner index (H)(Magurran, 2004), using the maximum
abundance recorded over the two seasons. We also
calculated their mean values per sampling point
(MeanS, Mean WS) and we compared them for
the pine, broad-leaved and mixed woods using the
Kruskal–Wallis non-parametric test and then the
Mann–Whitney tests with Bonferroni correction,
with the help of SPSS software (Field, 2005).
Weighted species richness is the species richness
of the site, with each species having a different
weight based on its conservation status (SPEC
category) (BirdLife International, 2004). We as-
signed a standard weight (w¼1) to the species of
favourable conservation status (SPEC 4 category or
non-SPEC category). We assigned double weight
(w¼2) to the species of SPEC 3 category, which are
not concentrated in Europe but have an unfavour-
able conservation status. Finally we assigned a
quadruple weight (w¼4) to the species in the SPEC
2 category, which are concentrated in Europe and
have an unfavourable conservation status and to
the species listed in Annex I of the European
Directive 79/409 EU, which includes all bird species
subject to special conservation measures in
Europe. We also calculated the bird diversity
indices (S, WS, H) for each point count and
tested their correlation with the five quantitative
environmental variables sampled using Spearman
correlation coefficients (SPSS, vers 15.).
To assess which environmental factors shape bird
community composition, we used the Canonical
Correspondence Analysis (CCA) option from the
ordination program CANOCO (ter Braak & Smilauer,
2002), because the length of the gradients of
Detrended Correspondence Analysis (DCA) indi-
cated a heterogeneous dataset (Leps & Smilauer,
2005). This method extracts the major gradients in
the data that are accounted for by the measured
variables. The position of a species in the resulting
plot indicates the characteristics of the ecological
optimum for this taxon; its abundance will de-
crease with increasing distance from this point.
Species encountered only once were not included
in the analysis and species abundances were log
ARTICLE IN PRESS
Bird communities and reserve management 51
transformed. A forward selection procedure using a
Monte-Carlo permutation test with 1000 iterations
was used to select and present in the CCA diagram
only the significant (po0.05) environmental vari-
ables that explained variation in community struc-
ture. The multi-collinearity of environmental
variables was tested using the inflation factor.
We used Ward’s clustering method to group
hierarchically the point counts into clusters,
expressing the different habitat types of bird
community. We also used the indicator value
procedure (IndVal) (Dufreˆne & Legendre, 1997)to
identify the typical species characterising each of
the clusters. IndVal is a percentage that ranges
between 0 and 100 and takes its maximum value
when the species is present exclusively in all sites
of a single cluster. All calculations were carried out
using IndVal software (Dufreˆne, 1999). A species
is considered to be a ‘‘symmetrical indicator’’
(IndVal450%) for one cluster, when it is present
in 470% of the sites of the cluster and when 470%
of its individuals occur in the cluster. A random
reallocation procedure (1000 iterations) of sites
among site groups was used to test IndVal sig-
nificance (alpha ¼0.05).
Results
Point count sampling resulted in 56 species (14 of
conservation concern), whereas the overall bird
species number recorded inside and outside point
counts was 62 species, including six wood-
peckers (Piciformes) and 56 passerine species
(Passeriformes) (Appendix I). The ornithological
importance of the study area is high, given that
we recorded six and ten species of unfavourable
conservation status concentrated in Europe (SPEC 2)
or not (SPEC 3), respectively (Appendix I). The
overall estimated species richness using non-para-
metric estimators was between 56 (Chao 1 estima-
tor) and 66 species (Chao 2 estimator). We have
therefore sampled exhaustively the bird commu-
nity of the area.
Bird diversity
The most important site for bird conservation, on
the basis of the criterion of mean weighted species
richness (MeanWS) was the montane grassland G4
(Table 1). Although not very species-rich (15
species), it held seven out of the 16 species of
conservation concern (SPEC 2: Carduelis cannabi-
na, Emberiza hortulana, Lullula arborea; SPEC 3:
Anthus campestris, Lanius collurio, Monticola
saxatilis, Oenanthe oenanthe). The second most
important site, which was also the richest in
species number was a traditionally cultivated site
(A2), including six out of the 16 species of
conservation concern (C. cannabina, E. hortulana,
L. collurio, Milaria calandra, Muscicapa striata,
Passer domesticus). The next sites in rank was a
mixed pine-beech forest (PB1), a montane grass-
land (G1), a black pine forest located in the core
area of the NP that has not been harvested for at
least two centuries and is accessible exclusively on
foot (P2), and a sacred grove of black pines that has
also not been harvested for at least two centuries
(P3) (Table 1). The species richness and the
weighted species richness were significantly higher
in the mixed pine-beech woods than in the pine-
woods (Kruskall Wallis test p¼o0.05, Mann–Whitney
post-hoc tests po0.0160), whereas pinewoods
and broad-leaved woods did not differ significantly
between them.
Community structure and typical species
Bird species richness in each point count (S)
decreased significantly (po0.01) with increasing
altitude (Table 2). Bird species richness together
with the weighted species richness in each point
count (WS) increased significantly (po0.01) with
the number of vegetation layers (NL) and with the
maximum height of upper tree layer (MH) in the
forested sites. Finally, the Shannon diversity index
for birds in each point count (H) was significantly
(po0.01) correlated with the vertical structural
complexity as expressed by the Simpson reciprocal
index 1/D(Table 2).
The CCA model was significant (po0.001) and
revealed clearly that the presence of grassland,
forest or agricultural habitat type, as well as
altitude and vegetation structural complexity
ARTICLE IN PRESS
Table 2. Spearman correlation coefficients between
bird diversity indices of point counts and environmental
parameters.
Diversity indices Environmental parameters
NL 1/DMH ALT
S0.385
–0.541
–0.305
WS 0.219
–0.790
–
H–0.637
–
ALT: altitude, NL: number of vegetation layers, 1/D: Simpson
reciprocal index of vertical structure, MH: maximum height of
upper tree layer, S: species richness, WS: weighted species
richness, H: Shannon diversity index.
Po0.01.
Po0.05.
V. Kati et al.52
(1/D) affected significantly bird species assem-
blages (Figure 2). The first CCA axis (11.4% of data
variability and 49% of species-environment rela-
tion) also explained the first hierarchical division in
Ward’s dendrogram; the latter reflected a gradient
from forested sites at lower altitudes towards
grasslands at higher altitudes (Figure 3). The first
ordination axis of species dataset predicts the bird
community in the right part of the CCA diagram
(montane grassland community) and is calculated
as: [0.89 Grassland +0.61 Altitude +0.14 1/D0.29
Agriculture], if we consider the statistically
significant environmental variables without
multicollinearity in the CCA model (Figure 2).
Three species (Emberiza citrinella, L. arborea,
O. oenanthe) were typical species of the above
community (Figure 3). The second CCA axis (5.6% of
species data variability and 24% of species-environ-
ment relation) explained the second division in
Ward’s dendrogram; the latter reflected a gradient
from tall forests towards sites of lower trees and
agricultural character (Figure 3). The respective
equation of the second axis of species dataset that
predicts the bird community in the upper part of
the CCA diagram (agriculture and low trees) is the
following: [0.68 Agriculture +0.18 Grassland 0.55
1/D0.4 Altitude]. A group of eight species was
ARTICLE IN PRESS
Figure 3. Hierarchical clustering of sites produced by Ward’s method and typical species with significant indicator
values (in parenthesis) that are greater than 50% for each cluster.
1.0
-
1.0
-0.6 1.
0
Agriculture
Mcal Ecirl
Tirogl
Acaud Cchl
Mstr
Pmaj
Tmer
Sattr
Erub
Cbra Rign
Par er
Forest
1/D
Altitude
Grassland
Sser Arriv
Ecitr
Lularb
Phochr
Ooen
Prmod
Ccard
Lcol
Ehort
Ccoc
Figure 2. Triplot of bird species, significant environ-
mental variables and sampling sites after Canonical
Correspondence Analysis. Only species with a fit greater
than 30% are shown.
Bird communities and reserve management 53
identified as typical for the above community,
increasing in abundance along the second axis
(Aegithaulus caudatus, Troglodytes troglodytes,
Parus major, Parus caerulaeus, M. striata, Turdus
merula, Coccothraustes coccothraustes, Emberiza
cirlus)(Figure 3). On the other hand, the relative
abundance of two typical forest species (Parus ater,
Regulus ignicapilla),decreased along this axis.
Furthermore, two typical species for pinewoods
were identified (Dendrocopos major,Parus
cristatus)¸whereas no typical species existed
for beech woods. Finally, we found that some
species were generalists, being common through-
out the study area and presenting their highest
indicator value for all the sites sampled (i.e.
Fringilla coelebs, Phylloscopus collybita, Erithacus
rubecula),whereas two forest species were general
indicators for all sites of forest character (Sylvia
atricapilla, Certhia brachydactyla)(Figure 3).
Discussion
Community structure
The presence of agricultural land, grasslands or
forests, as well as altitude and vegetation vertical
complexity were the main statistically significant
environmental gradients that influence the distri-
bution of bird community in the broader area of
Pindos National Park. The bird community was
distributed along a gradient from forested habitats
at lower altitudes towards montane grasslands at
higher altitudes and along a gradient from forest
stands with high trees and greater vertical struc-
tural complexity towards lower trees and culti-
vated land. These results are in agreement with
other studies in the Mediterranean area (e.g. Dı´az,
2006;Kati & Sekercioglu, 2006;Prodon & Lebreton,
1981). Birds distinguish three general bird habitat
types in the study area: grasslands; shrubby
habitats of agricultural land or riverine vegetation;
and forest habitats of pinewoods or beech woods,
thus perceiving the environment at a relatively
coarse scale.
Typical species
Our study offers a valuable conservation tool for
the management and monitoring plan of the study
area, as it reveals a set of 17 typical species of the
different habitat types, occurring almost exclu-
sively in the relevant habitats and with great
abundances. Some of the species recorded were
generalists having a broad ecological niche and
distribution range in the study area with no
indicator value. Several species were encountered
with increased abundances in the montane grass-
lands and three of them were revealed as grassland
indicators (Figures 2 and 3). In the same context,
eight bird species were typical of the agricul-
tural sites and the bushy habitats of riverine
vegetation, whereas another six bird species
were typical for the forest conditions (Figures. 2
and 3). If we compare our results with the typical
species found in a Mediterranean reserve at
low altitude (Dadia NP, less than 600 m), we find
some similarities, concerning the generalist species
(e.g. F. coelebs, E. rubecula) and the indicator
species of forest conditions (e.g. S. atricapilla,
C. brachydactyla, typical for broad-leaved woods).
However, we found that L. arborea is typical of
montane grasslands in Pindos NP but it is also
typical for lowland heaths in Dadia NP in eastern
Greece (Kati & Sekercioglu 2006). Special attention
should be paid to the typical species that have an
unfavourable conservation status in Europe at this
time (SPEC 2, 3). For our study area, these were:
L. arborea and O. oenanthe (typical for montane
grasslands); M. striata (typical farmland species);
and P. cristatus (typical pine forest species). All
indicator species as well as the species with
unfavourable conservation status should be inte-
grated into the future monitoring program of the
NP, in order to evaluate their own conservation
status as well as their habitat condition, on a long-
term basis.
Bird diversity
Bird species richness was inversely correlated
with altitude in our study area, but the weighted
species richness, which also considers the conser-
vation status of the species, was not. This well
known pattern of decreased species richness with
increasing altitude is explained by the theory of
island biogeography and has been shown for bird
communities in a variety of climatic zones (Kattan
& Franco, 2004;Prodon et al., 2002). However, we
found that the bird community of montane grass-
lands, which is a quite widespread habitat type in
the Mediterranean mountains, comprises fewer but
important species with unfavourable conservation
status in Europe (SPEC 2, 3) and therefore should be
considered as a primary habitat for conservation in
the NP.
Another important site for bird diversity con-
servation in the study area was the agricultural
site A2. It was a rural mosaic, situated in the
forested zone, alternating small cultivated plots
ARTICLE IN PRESS
V. Kati et al.54
with pastures and orchards, separated by natural
vegetation of hedges and tree lines. Site A1 was
less species-rich, where former agricultural plots
were not cultivated any more but were periodically
grazed by livestock (Table 1). These rural mosaics
were the rarest habitat in our study area, because
small-scale farming using traditional agricultural
practices has become financially unprofitable over
the last decades and has therefore led to land
abandonment in the broader area of Pindos NP,
as in other mountainous areas in Europe (Gellrich &
Zimmermann, 2007;MacDonald et al., 2000).
Farmland bird diversity is subject to two major
threats in Europe: agricultural intensification
mostly in the northern lowlands (Chamberlain
et al., 2000;Donald et al., 2001;Kati & Sekercioglu,
2006;Pain & Pienkowski, 1997); and agro-pastoral
land abandonment mostly in the Mediterranean
mountains. The latter results in forest encroachment,
landscape heterogeneity decline, and agricultural
habitat loss negatively affecting farmland bird
diversity (Farina, 1997;Laiolo et al., 2004;Preiss
et al., 1997;Suarez-Seoane et al., 2002). Our results
provide additional evidence supporting the need to
enhance the agri-environmental measures of the
European Common Agricultural Policy (CAP) against
land abandonment in less favoured mountainous
areas, so as to maintain the traditional agricultural
practices and the associated, biologically rich
agricultural mosaics.
We found that the mixed pine-beech woods were
richer and more important for bird conservation
rather than the pinewoods. However, the mixed
woods did not form a separate cluster in hierarch-
ical clustering analysis and did not hold a specia-
lised avifauna (no indicator species for mixed
woods). Mixed stands can be either more species-
rich or not of unmixed stands, but they are known
to hold an intermediate bird species community
between those of broadleaved and coniferous
stands (Archaux & Bakkaus, 2007;Dı´az, 2006;
Donald et al., 1998;Willson & Comet, 1996).
Besides, we found no statistically significant differ-
ence when comparing the species richness (and the
weighted species richness) of the unmixed
natural pine and broad-leaved woods. We question
therefore the general admission that broad-leaved
stands hold richer bird communities than pure
coniferous stands, when natural native stands
are compared. This relationship could be attributed
to the reduced ecological value of introduced
conifer plantations compared to the natural native
character of broad-leaved woods (Archaux &
Bakkaus, 2007). These results may have a good
extrapolative value to other Mediterranean natural
forest ecosystems, since the forest habitat types
sampled in Pindos NP, such as the pinewoods
(P. nigra) and the beech woods (F. sylvatica)
are encountered in other mountains in the
Mediterranean (Italy, Corsica, Spain, Portugal,
France) with the exception of the Balkan
pinewoods, encountered only in Greece and
southern Italy.
Bird diversity was strongly correlated with
vertical structural vegetation complexity (NL,
1/D). Vertical stratification in forest vegetation
has a positive influence over forest avifauna, as the
presence of shrub layers offers more ecological
niches, foraging opportunities and breeding re-
sources especially for undergrowth-dependent bird
species (e.g. Camprodon & Brotons, 2006;Dı´az,
2006;James & Wamer, 1982;Prodon & Lebreton,
1981;Wesolowski, 2007). This relationship was
detected for the first time in Greek island
ecosystems (Watson, 1964), but no other study
has been conducted to quantify it in different
Greek forest ecosystems since then.
Although tree height was not a statistically
significant environmental parameter to predict bird
species in our CCA model, we found that the
presence of high trees was positively correlated
with the number of species (S) and also with the
number of important bird species (WS) in forest
habitats. We also found that two tall old growth
pinewoods that have not been harvested for
centuries (P2, P3) were richer than those lower
pinewoods (P1, P4, P5). Forest bird species richness
increases with tree height (Donald et al., 1998;
Helle & Monkkonen, 1990). If we assume that high
tree stands are also more mature, we could
attribute this positive relationship to the fact that
older forest stands provide more tree holes, as well
as a higher amount of dead wood as breeding and
feeding habitat for a variety of specialist taxa (e.g.
Dı´az, 2006;Gil-Tena et al., 2007;Hobson & Bayne,
2000;Keller et al., 2003;Laiolo et al., 2003;
Sekercioglu, 2002;Wesolowski, 2007).
Conservation management
In the applied context, our study proves the
importance and the need for maintenance of the
montane grasslands in their present status. Sec-
ondly, it suggests that the enhancement of tradi-
tional agricultural practices in the broader area
surrounding the Pindos NP could have beneficial
effects on local bird communities. Thirdly, we
emphasise the need for protection of the natural
remaining old growth forests in the core area.
Although timber extraction is not allowed in the
core area since the creation of the NP (1966),
ARTICLE IN PRESS
Bird communities and reserve management 55
logging of old growth pine trees for commercial
reasons seems to have occurred in 80s, facilitated
by the existing road network. Fourthly, our results
indicate that sustainable forestry practices need to
maintain the mixed character of forest stands
where they occur, the vertical structural complex-
ity as well as a number of high and mature trees in
managed forest stands. Finally, we argue that the
14 species of unfavourable conservation status
(SPEC 2 and 3) together with the typical species
found per bird habitat type should be integrated
into the future monitoring scheme of Pindos
National Park, as a tool for forest management
(Kati & Sekercioglu 2006;Mu¨ller 2005). The above
proposals can have an extrapolative value in other
mountainous reserves across the country and in the
Mediterranean mountains.
Acknowledgements
This research was partially funded by the
Hellenic Ministry of Environment and Public Works
in the frame of the Sustainable Development
Project (ETERPS funds) and from Pindos Perivallon-
tiki NGO. We are grateful to I. Leonardos from the
University of Ioannina for support, to J. Foufopou-
los and to S. Sfenthourakis for helpful comments on
the manuscript.
Appendix I
Inventory of all 62 bird species recorded in the
study area and weighted index (w) according to
their conservation status (SPEC category 2004).
ARTICLE IN PRESS
Table A1
Code Species wSPEC 2004 Code Species wSPEC 2004
Piciformes Phcol Phylloscopus collybita 1
Pvir Picus viridis 42 Rreg Regulus regulus 14
Dmart Dryocopus martius 4
a
Rign Regulus ignicapilla 14
Dmaj Dendrocopos major 1Mstr Muscicapa striata 23
Dsyr Dendrocopos syriacus 4
a
4Ppal Parus palustris 23
Dmed Dendrocopos medius 4
a
4Plug Parus lugubris 14
Dmin Dendrocopos minor 1Pmont Parus montanus 1
Passeriformes Pcrist Parus cristatus 42
—Galerida cristata 23 Pater Parus ater 1
Larb Lullula arborea 4
a
2Pcaer Parus caeruleus 14
—Alauda arvensis 23 Pmaj Parus major 1
—Hirundo daurica 1Acaud Aegithalos caudatus 1
Durb Delichon urbica 23 Seur Sitta europaea 1
Acamp Anthus campestris 4
a
3Cfam Certhia familiaris 1
Atriv Anthus trivialis 1Cbra Certhia brachydactyla 14
Mflav Motacilla flava 1Lcol Lanius collurio 4
a
3
—Motacilla cinerea 1Ggland Garrulus glandarius 1
Malb Motacilla alba 1Ccor Corvus corone 1
Ccincl Cinclus cinclus 1Ccorax Corvus corax 1
Ttrogl Troglodytes troglodytes 1Pdom Passer domesticus 23
Pmod Prunella modularis 14 Fcoel Fringilla coelebs 14
Erub Erithacus rubecula 14 Sser Serinus serinus 14
Lmeg Luscinia megarhynchos 14 Cchl Chloris chloris 14
Phocrh Phoenicurus ochrurus 1Ccard Carduelis carduelis 1
Ooen Oenanthe oenanthe 23 Ccan Carduelis cannabina 42
Montsax Monticola saxatilis 23 Lcurv Loxia curvirostra 1
Tmer Turdus merula 14 Pyrpyr Pyrrhula pyrrhyla 1
Tphil Turdus philomelos 14 Ccoc Coccothraustes coccothraustes 1
Tvisc Turdus viscivorus 14 Ecitr Emberiza citrinella 14
—Sylvia melanocephala 14 Ecirl Emberiza cirlus 14
—Sylvia curruca 1Ehort Emberiza hortulana 42
Satr Sylvia atricapilla 14 Mcal Milaria calandra 42
a
Species of Annex I of 79/409EU. SPEC 2: concentrated in Europe and with unfavourable conservation status, SPEC 3: not concentrated
in Europe but with unfavourable conservation status, SPEC 4: concentrated in Europe and with favourable conservation status.
V. Kati et al.56
Species codes are given only for the species
recorded during the point counts. See Table A1
References
Archaux, F., & Bakkaus, N. (2007). Relative impact of
stand structure, tree composition and climate on
mountain bird communities. Forest Ecology and
Management,247,72–79.
Bani, L., Massimino, D., Bottoni, L., & Massa, R. A.
(2005). Multiscale method for selecting indicator
species and priority conservation areas: A case study
for broadleaved forests in Lombardy, Italy. Conserva-
tion Biology,20, 512–526.
Bibby, C. J., Burgess, N. D., & Hill, D. (1992). Bird census
techniques. London: Academy Press.
BirdLife International (2004). Birds in Europe: Population
estimates, trends and conservation status. Cam-
bridge, UK: BirdLife International (BirdLife Conserva-
tion Series No 12).
Blondel, J., Ferry, F., & Frochot, B. (1970). La me´thode
des indices ponctuels d‘abondance (I.P.A.) ou des
releve´s d‘ avifauna par ‘‘stations d‘e´coute’’. Alauda,
38,55–71.
Camprodon, J., & Brotons, L. (2006). Effects of under-
growth clearing on the bird communities of the
Northwestern Mediterranean Coppice Holm oak for-
ests. Forest Ecology and Management,221,72–82.
Chamberlain, D. E., Fuller, R. J., Bunce, R. G. H., &
Duckworth, J. C. (2000). Changes in the abundance of
farmland birds in relation to the timing of agricultural
intensification in England and Wales. Journal of
Applied Ecology,37, 771–778.
Colwell, R. K. (2006). Estimate S: Statistical Estimation
of species richness and shared species from sample.
Version 8. User’s Guide and application /http://
viceroy.eeb.unconn.edu/EstimateSS.
Dı´az, L. (2006). Influences of forest type and forest
structure on bird communities in oak and pine wood-
lands in Spain. Forest Ecology and Management,223,
54–65.
Dobson, A. P., Rodrigue, J. P., Roberts, W. M., &
Wilcove, D. S. (1997). Geographic distribution of
endangered species in the United States. Science,
275, 550–553.
Donald, P. F., Fuller, R. J., Evans, A. D., & Gough, S. J.
(1998). Effects of forest management and grazing on
breeding bird communities in plantations of broad-
leaved and coniferous trees in western England.
Biological Conservation,85, 183–197.
Donald, P. F., Green, R. E., & Heath, M. F. (2001).
Agricultural intensification and the collapse of Eur-
ope’s farmland bird populations. Proceedings of the
Royal Society of London B,268,25–29.
Dufreˆne, M. (1999). IndVal or how to identify indicator
species of a sample typology./http://mrw.wallonie.
be/dgrne/sibw/outils/indval/home.htmlS.
Dufreˆne, M., & Legendre, P. (1997). Species assemblages
and indicator species: The need for a flexible
asymmetrical approach. Ecological Monographs,67,
345–366.
European Commission (2003). Interpretation manual of
European Union habitats (EUR25). Brussels: European
Commission.
Farina, A. (1997). Landscape structure and breeding bird
distribution in a sub-Mediterranean agro-ecosystem.
Landscape Ecology,12, 365–378.
Field, A. (2005). Discovering statistics using SPSS. UK:
SAGE publications.
Gellrich, M., & Zimmermann, N. E. (2007). Investigating
the regional-scale pattern of agricultural land aban-
donment in the Swiss mountains: A spatial statistical
modelling approach. Landscape and Urban Planning,
79,65–76.
Gil-Tena, A., Saura, S., & Brotons, L. (2007). Effects of
forest composition and structure on bird species
richness in a Mediterranean context: Implications for
forest ecosystem management. Forest Ecology and
Management,242, 470–476.
Gregory, R. D., Noble, D. G., & Custance, J. (2004). The
state of play of farmland birds: Population trends and
conservation status of lowland farmland birds in the
United Kingdom. Ibis,146,1–13.
Heath, M. E., Evans, M. I., Hoccom, D. G., Payne, A. J., &
Peet, N. B. (2000). Important bird areas in Europe:
Priority sites for conservation. Cambridge: BirdLife
International.
Helle, P., & Monkkonen, M. (1990). Forest succession and
bird communities: Theoretical aspects and practical
implications. In A. Keast (Ed.), Biogeography and
Ecology of Forest Bird Communities (pp. 299–318).
The Hague: SPB Academic Publishing.
Hobson, K., & Bayne, E. (2000). The effects of stand age
on avian communities in aspen dominated forests of
central Saskatchewan, Canada. Forest Ecology and
Management,136, 121–134.
Howard, P. C., Viskanic, P., Davenport, T. R. B.,
Kigenyi, F. W., Baltzer, M., Dickinson, C. J., et al.
(1998). Complementarity and the use of indicator
groups for reserve selection in Uganda. Nature,394,
275–472.
James, F. C., & Wamer, N. O. (1982). Relationships
between temperate forest bird communities and
vegetation structure. Ecology,63, 159–171.
Johnson, G. A. M., & Freedman, B. (2002). Breeding birds
in forestry plantations and natural forest in the
vicinity of Fundy National Park, New Brunswick.
Canadian Field-Naturalist,116, 475–487.
Kati, V., & Sekercioglu, C. H. (2006). Diversity, ecological
structure, and conservation of the landbird commu-
nity of Dadia reserve, Greece. Diversity & Distribu-
tion,12, 620–629.
Kati, V., Devillers, P., Dufreˆne, M., Legakis, A., Vokou, D.,
& Lebrun, Ph. (2004a). Testing the value of six
taxonomic groups as biodiversity indicators at a local
scale. Conservation Biology,18, 667–675.
Kati, V., Devillers, P., Dufreˆne, M., Legakis, A., Vokou, D.,
& Lebrun, Ph. (2004b). Hotspots, complementarity or
representativeness? Designing optimal small-scale
ARTICLE IN PRESS
Bird communities and reserve management 57
reserves for biodiversity conservation. Biological
Conservation,120, 471–480.
Kattan, G. H., & Franco, P. (2004). Bird diversity along
elevational gradients in the Andes of Colombia: Area
and mass effects. Global Ecology and Biogeography,
13, 451–458.
Keller, J. K., Richmond, M. E., & Smith, C. R. (2003). An
explanation of patterns of breeding bird species
richness and density following clearcutting in north-
eastern USA forests. Forest Ecology and Management,
541–564.
Kent, M., & Coker, P. (1994). Vegetation description
and analysis: A practical approach. Chirchester:
Wiley.
King, D. I., & DeGraaf, R. M. (2000). Bird species diversity
and nesting success in mature, clearcut and shelter-
wood forest in northern New Hampshire, USA. Forest
Ecology and Management,129, 227–235.
Ku¨chler, A. W. (1988). A physiognomic and structural
analysis of vegetation. In A. W. Ku¨chler, & I. S.
Zonneveld (Eds.), Vegetation Mapping. Handbook of
Vegetation Science (pp. 37–50). Dordrecht: Kluwer
Academic Publishers.
Laiolo, P., Caprio, E., & Rolando, A. (2003). Effects of
logging and non-native tree proliferation on the birds
overwintering in the upland forests of north-western
Italy. Forest Ecology and Management,179, 441–451.
Laiolo, P., Dondero, R., Ciliento, E., & Rolando, A.
(2004). Consequences of pastoral abandonment for
the structure and diversity of the alpine avifauna.
Journal of Applied Ecology,41, 294–304.
Lance, A. N., & Phinney, M. (2001). Bird responses to
partial retention timber harvesting in central interior
British Columbia. Forest Ecology and Management,
142, 267–280.
Lawton, J. H., Bignell, D. E., Bolton, B., Bloemers, G. F.,
Eggleston, P., Hammond, P. M., et al. (1998).
Biodiversity inventories, indicator taxa and effects of
habitat modification on tropical forest. Nature,391,
72–76.
Leps, J., & Smilauer, P. (2005). Multivariate analysis of
ecological data using CANOCO. UK: Cambridge Uni-
versity Press.
Lindenmayer, D. B., Franklin, J. F., & Fischer, J. (2006).
General management principles and a checklist of
strategies to guide forest biodiversity conservation.
Biological Conservation,131, 433–445.
Lombard, A. (1995). The problem with multi-species
conservation: Do hotspots, ideal reserves and existing
reserves coincide? South African Journal of Zoology,
30, 145–163.
MacDonald, D., Crabtree, J. R., Wiesinger, G., Dax, T.,
Stamou, N., Fleury, P., et al. (2000). Agricultural
abandonment in mountain areas of Europe: Environ-
mental consequences and policy response. Journal of
Environmental Management,59,47–69.
Magurran, A. E. (2004). Measuring biological diversity.
USA: Blackwell Publishing.
Mertzanis, G. (1999). Special Environmental study of
Pindos-Zagori. Life 96 NAT/GR/3222. Project ARCTOS:
Conservation of brown bear and its habitats in Greece
(phase b). Thessaloniki: ARCTOUROS.
Mu¨ller, J. (2005). Bird communities as indicators for
woodland structures in oak woods. Ornithologische
Beobachter,102,15–32.
Mu¨ller, J., Hothorn, T., & Pretzsch, H. (2007). Long-term
effects of logging intensity on structures, birds,
saproxylic beetles and wood-inhabiting fungi in stands
of European beech Fagus sylvatica L. Forest Ecology
and Management,242, 297–305.
Pain, D. J., & Pienkowski, M. W. (1997). Farming and
birds in Europe. The common agricultural policy and
its implications for bird conservation. London: Aca-
demic Press.
Preiss, E., Martin, J.-L., & Debussche, M. (1997). Rural
depopulation and recent landscape changes in a
Mediterranean region: Consequences to the breeding
avifauna. Landscape Ecology,12,51–61.
Prendergast, J. R., Quinn, R. M., Lawton, J. H.,
Eversham, B. C., & Gibbon, D. W. (1993). Rare species,
the coincidence of diversity hotsposts and conserva-
tion strategies. Nature,365, 335–337.
Prodon, R., & Lebreton, J-D. (1981). Breeding avifauna of
a Mediterranean succession: The holm oak and
cork oak series in the eastern Pyrenees, 1. Analysis
and modelling of the structure gradient. Oikos,37,
21–38.
Prodon, R., Thibault, J.-C., & Dejaifve, P.-A. (2002).
Expansion vs. compression of bird altitudinal
ranges on a Mediterranean island. Ecology,83,
1294–1306.
Sekercioglu, C. H. (2002). Effects of forestry practices on
vegetation structure and birds community of Kibale
National Park, Uganda. Biological Conservation,107,
229–240.
Sekercioglu, C. H. (2006). Increasing awareness of avian
ecological function. Trends in Ecology & Evolution,21,
464–471.
Sekercioglu, C. H., Daily, G. C., & Ehrlich, P. R. (2004).
Ecosystem consequences of bird declines. Proceedings
of the National Academy of Science USA,101,
18042–18047.
Simberloff, D. (1998). Flagships, umbrellas and key-
stones: A single species management passe in the
landscape era? Biological Conservation,83, 247–257.
Suarez-Seoane, S., Osborne, P. E., & Baudry, J. (2002).
Responses of birds of different biogeographic origins
and habitat requirements to agricultural land aban-
donment in northern Spain. Biological Conservation,
105, 333–344.
ter Braak, C. J. F. & Smilauer, P. (2002). CANOCO
Reference Manual and Canoco Draw for Windows
User’s Guide: Software for Canonical Community
Ordination (version 4.5). Ithaca, NY, USA: Microcom-
puter Power.
Thiollay, J.-M. (2002). Bird diversity and selection of
protected areas in a large neotropical forest tract.
Biodiversity and Conservation,11, 1377–1395.
Trakolis, D., Platis, P., Tsiontsis, A., Spanos, K., Meliadis,
I. & Malamidis, G. (1996). Management Plan of Pindos
ARTICLE IN PRESS
V. Kati et al.58
National Park. Thessaloniki, Greece: Forest Service of
Grevena, Ioannina, and Metsovo.
Tsounis, G., Malakou, M. & Ravasini, M. (1985).
Study of the fauna of Pindos National Park
(in greek). Athens: Hellenic Society for Nature
Protection.
van der Maarel, E. (2005). Vegetation ecology. Oxford:
Blackwell Publishing.
Vessby, K., Soderstrom, B., Glimskar, A., & Svensson, B.
(2002). Species-richness correlations of six different
taxa in Swedish seminatural grasslands. Conservation
Biology,16, 430–439.
Watson, G. E. (1964). Ecology and evolution of passerine
birds on the islands of the Aegean Sea. Ph.D. Thesis,
Yale University.
Wesolowski, T. (2007). Primeval conditions—what can we
learn from them? Ibis,149,64–77.
Willson, M. F., & Comet, T. A. (1996). Bird communities of
northern forests: Patterns of diversity and abundance.
Condor,98, 337–349.
ARTICLE IN PRESS
Bird communities and reserve management 59
