DIGITALNA ARHIVA ŠUMARSKOG LISTA
prilagođeno pretraživanje po punom tekstu



ŠUMARSKI LIST 9-10/2011 str. 55     <-- 55 -->        PDF

T. Kirin, J. Kralj, D. Ćiković, Z. Dolenec: HABITAT SELECTIONAND SIMILARITY OF THE FOREST ...Šumarski list br. 9–10, CXXXV (2011), 467-475
Bird Community Sampling –Metode istraživanja ornitofaune


The study was carried out during breed ing seasons
2006 and 2007. Standard point count method was used
(Bibby et al. 1992),with 10 minutes counting period.
Two counting bands were used:inner – with the diameter
of 50 m and outerclose to the infinity.The research
wascarriedon49 points on Medvednica and 52 on
Žumberak – Samoborsko gorje. Counting points were
situated inside the forest, at least 500 m apart. Every
point was visited three times during the breeding season:
inApril, May and June.Visits started after the sunrise
and lasted up to three hours, covering the period of
the highest bird activity. Singing males were con sidered
as representing breeding territories. For quantitative
analysis, only birds recorded in the inner band
wereused. Songbird species with largebreedingterri


tories (as Jay – Garrulus glandariusand Raven–
Corvus corax) were excluded from the analyses.


For detailedanalyses of bird communities, species
were grouped according to their breeding and foraging
ecology. Regarding the nest site, birds were divided
into four groups: i) canopy nesting species,ii) species
nesting in the shrub layer, iii) hole-nesting species and
iv) ground nesting species. Regarding the layer where
birds feed they were divided into five groups: i) canopy
feeding species, ii) species feeding in shrub layer, iii)
bark gleaning species, iv) ground feeding species and
v) aerial feeders (Table 1). Species recorded with only
one specimen during the study were excluded from
analyses of ecological groups.


Habitat Sampling –Metode istraživanja staništa


At each counting point, habitat mapping was carried
out by the circular plot method (JamesandShugart
1970, Cyrand Oelke1976,Bibbyetal. 1992).
Plot size was 0.04 ha.The tree species and tree diameter
(DBH) were recorded for each tree inside the plot.
Tree diameter was measured with the calibrated ruler
and is given in eight classes:A7.5–15 cm, B 15–23 cm,
C 23–38 cm, D 38–53 cm, E 53–68 cm, F 68–84 cm,
G 84–101 cm, H > 101 cm.Tree height was not measured.
Basal area was calcu lated for trees in each diameter
class, accord ing to Cyrand Oelke(1976). The
average tree basal area was calculated by dividing the
total basal area with the total number of trees on the plot
and was used as indication of the stand maturity (Bibby
et al. 1992). For further analyses, trees from groupAand
B were pooled together as “small trees”, C, D and E – as
“medium sized trees” and F, G and H – as “large trees”.


The shrub density was recorded along two transects
of outstretched armlength across the circular plot, each
equals to approximately 0.008 ha.The percentages of
ground cover and cano py cover were calculated basing
on 20 readings made through a sighting tube with cross


Data Analyses–


Shannon-Weiner (H’) index was used for calculating
diversity of communities (Odum1971).Sorensen
index was used for comparison of similarity in structural
characteristics of forests and bird communities
between two study areascommunities (Odum1971).


Shapiro-Willks Wtest showed that variables were
not normally distributed. Therefore, non-parametric


threads taped across one end of a tube. Detailed floris


tic structure of the shrub and ground layers was not


studied, only the dominant species were noted.


We didn’t attempt to determine the forest community
for every counting point. Instead, the proportion of tree
basal area per species was used to classify studied points
into five forest types (Delahayeand Vandevyvre
2008) (beech, oak, coniferous, mixed deciduous and
mixed coniferous forests). Counting points with more
than 70% of total basal area belonging to the beech
(Fagus sylvaticaL.) and those with more than 50 % belonging
to the oak (Quercussp.) were classified as beech
and oak stands, respectively. If more than 70% of total
basal area referred to coniferous trees of any species (fir,
spruce, pine and larch),counting point was classified as
coniferous stand. Other points were classified as mixed
stands, either deciduous or coniferous, depending on
presence of coniferous trees. Habitat sampling methods
and classification of forest types differ from standarised
methodology used in forestry. Applied methods thus
were not comparable with methods used in systematic
forest inventory in Croatia.


Analiza podataka


tests (Chi-square, Kruskal–Wallis and Kendal Tau)
were applied.All statistical analyses were performed
using Ecological Methodology (Krebs2003) and
STATISTICA v.7.0(StatSoft 2004) software.


RESULTS – Rezultati
During this study, 27 songbird species weretypes, except in the beech stands, were higher in
recorded in the forests of Medvednica and32in Žum-Medvednica. Contrary, Shannon – Wiener index of diberak
– Samoborsko gorje, with27species present inversity of bird communities in almost all forest types
both Parks (Table 1). Densities of birds in all forestwas higher in Žumberak – Samoborsko gorje (Fig 2).