Conservation value of forest plantations for bird communities in western Kenya


Farwig, N.; Sajita, N.; Boehning-Gaese, K.

Forest Ecology and Management 255(11): 3885-3892

2008


Tree plantations of native and exotic species are frequently used to compensate for forest loss in the tropics. However, these plantations may support lower species diversity and different communities than natural forest. We therefore investigated bird communities in stands of natural forest, different types of tree plantations and secondary forest in Kakamega Forest, western Kenya. We compared birds differing in habitat specialisation, i.e. forest specialists, generalists, and visitors. We recorded significant differences in mean species richness and number of individuals among the different forest types. Stands of natural forest and plantations of indigenous tree species comprised more species and individuals than plantations of exotic tree species and secondary forest. This was caused by a significant decline of forest specialists and generalists from natural forest and indigenous plantations to exotic plantations and secondary forest. Species composition of the bird communities did not differ between natural forest stands and plantations of a mixture of indigenous tree species, but clearly changed between natural forest and plantations of single tree species. These findings demonstrate that natural forest areas are needed for the conservation of forest bird diversity, but that plantations with a mixture of indigenous tree species can have similarly high conservation value.

Forest
Ecology
and
Management
255
(2008)
3885-3892
Contents
lists
available
at
ScienceDirect
Forest
Ecology
and
Management
Forest
Ecology
and
Management
itti Ittf
1
ELSEVIER
journal
homepage:
www.elsevier.com/locate/foreco
Conservation
value
of
forest
plantations
for
bird
communities
in
western
Kenya
Nina
Farwig
a
'
b
'* ,
Nixon
Sajita
C
,
Katrin
Bohning-Gaese
a
'
b
Institute
for
Zoology,
Department
of
Ecology,
Johannes
Gutenberg-University
of
Mainz,
Becherweg
13,
55128
Mainz,
Germany
b
Department
of
Ornithology,
National
Museums
of
Kenya,
Nairobi,
Kenya
`P.O.
Box
143,
Kakamega,
Kenya
ARTICLE
INFO
ABSTRACT
Article
history:
Received
11
January
2008
Received
in
revised
form
17
March
2008
Accepted
20
March
2008
Keywords:
Forest
avifauna
Tropical
rainforest
Africa
Kakamega
Forest
Tree
plantations
of
native
and
exotic
species
are
frequently
used
to
compensate
for
forest
loss
in
the
tropics.
However,
these
plantations
may
support
lower
species
diversity
and
different
communities
than
natural
forest.
We
therefore
investigated
bird
communities
in
stands
of
natural
forest,
different
types
of
tree
plantations
and
secondary
forest
in
Kakamega
Forest,
western
Kenya.
We
compared
birds
differing
in
habitat
specialisation,
i.e.
forest
specialists,
generalists,
and
visitors.
We
recorded
significant
differences
in
mean
species
richness
and
number
of
individuals
among
the
different
forest
types.
Stands
of
natural
forest
and
plantations
of
indigenous
tree
species
comprised
more
species
and
individuals
than
plantations
of
exotic
tree
species
and
secondary
forest.
This
was
caused
by
a
significant
decline
of
forest
specialists
and
generalists
from
natural
forest
and
indigenous
plantations
to
exotic
plantations
and
secondary
forest.
Species
composition
of
the
bird
communities
did
not
differ
between
natural
forest
stands
and
plantations
of
a
mixture
of
indigenous
tree
species,
but
clearly
changed
between
natural
forest
and
plantations
of
single
tree
species.
These
findings
demonstrate
that
natural
forest
areas
are
needed
for
the
conservation
of
forest
bird
diversity,
but
that
plantations
with
a
mixture
of
indigenous
tree
species
can
have
similarly
high
conservation
value.
2008
Elsevier
B.V.
All
rights
reserved.
1.
Introduction
Global
forest
destruction
has
increased
dramatically
in
the
last
few
decades
(Mayaux
et
al.,
2005;
Lewis,
2006)
with
only
half
of
the
world's
natural
forests
still
being
intact
(UNEP,
2001).
This
loss
of
natural
forest
is
most
profound
in
the
tropics
where
annually
0.8%
of
the
remaining
forests
are
cleared
and
converted
into
cultivated
land
(FAO,
2001;
Wright,
2005).
Growing
human
populations
and
accelerating
demand
for
forest
resources
such
as
wood
resulted
in
increasing
areas
of
degraded
forests
and
forest
plantations
(Petit
et
al.,
1999;
FAO,
2001;
Peres
et
al.,
2006).
One
consequence
of
this
rapid
conversion
of
native
habitat
is
the
loss
of
biodiversity
(Myers
et
al.,
2000).
Yet,
not
all
species
are
equally
vulnerable
to
habitat
transformation.
Increasing
specialisation
of
species
such
as
dependency
on
natural
forest
habitat
is
often
coupled
with
increasing
likelihood
of
extinction
(Raman,
2001;
*
Corresponding
author.
Present
address:
Zoological
Institute,
Community
Ecology,
University
of
Bern,
Baltzerstr.
6,
CH-3012
Bern,
Switzerland.
Tel.:
+41
31
6314539;
fax:
+41
31
6314888.
E-mail
address:
(N.
Farwig).
0378-1127/S
see
front
matter
©
2008
Elsevier
B.V.
All
rights
reserved.
doi:10.1016/goreco.2008.03.042
Sekercioglu,
2002;
Woltmann,
2003;
Sekercioglu
et
al.,
2004;
Sodhi
et
al.,
2004).
Thus,
the
species
diversity
and
composition
of
communities
in
different
forest
management
types
such
as
plantations
and
secondary
forests
are
of
particular
interest
for
conservation
strategies.
Recently,
a
number
of
studies
have
examined
the
conservation
value
of
different
forest
management
types
such
as
planted
or
naturally
regenerating
forests
for
bird
communities.
Several
studies
have
shown
that
these
different
forest
management
types
often
retained
a
large
proportion
of
forest
birds
(e.g.
Greenberg
et
al.,
1997b;
Sekercioglu,
2002;
Peh
et
al.,
2006;
Zurita
et
al.,
2006;
Barlow
et
al.,
2007).
In
Uganda,
forest
birds
were
best
represented
in
scarcely
logged
native
forests
while
exotic
conifer
plantations
retained
the
smallest
number
of
forest
birds
(Sekercioglu,
2002).
Also
Barlow
et
al.
(2007)
recorded
highest
species
richness
of
birds
in
indigenous
primary
forest
and
the
least
species-rich
assemblage
in
Eucalyptus
plantations.
Secondary
forests
harboured
an
inter-
mediate
number
of
bird
species
(Barlow
et
al.,
2007).
Yet,
most
studies
compared
only
few
different
forest
management
types
(e.g.
Sekercioglu,
2002;
Barlow
et
al.,
2007).
Thus,
forests
that
comprise
a
mixture
of
different
forest
management
strategies
such
as
plantations
with
a
mixture
of
tree
species,
monocultures
as
well
as
3886
N.
Farwig
et
al./
Forest
Ecology
and
Management
255
(2008)
3885-3892
secondary
forest
stands
are
important
for
further
investigations
of
biodiversity
conservation.
Despite
the
influence
of
forest
management
type
on
forest
bird
diversity
the
embedding
in
the
surrounding
landscape
might
influence
the
bird
composition.
For
instance,
the
proportion
of
forest
birds
was
particularly
high
in
structurally
rich
landscapes
containing
large
areas
of
natural
forest
(Greenberg
et
al.,
1997b;
Peh
et
al.,
2006).
Thereby,
the
distance
to
the
nearest
natural
forest
habitat
was
of
major
importance
for
the
maintenance
of
forest
bird
populations
in
managed
forest
stands
(Estrada
et
al.,
1997;
Greenberg
et
al.,
1997b;
Peh
et
al.,
2006).
Managed
plantations
in
close
proximity
to
natural
forest
comprised
more
forest
birds
than
those
located
farther
away
from
natural
habitat
(e.g.
Estrada
et
al.,
1997).
Thus,
highly
structured
landscapes
containing
natural
habitat
as
well
as
managed
forest
types
may
be
suitable
for
sustainable
forest
management
(Curran
et
al.,
2004;
Barlow
et
al.,
2007).
High
biodiversity
has
not
only
been
associated
with
the
complexity
of
landscapes
but
also
the
vegetation
structure
within
forest
types.
Several
studies
highlighted
the
importance
of
tree
cover
in
tropical
forest
management
types
for
conservation
of
forest
bird
communities
(Thiollay,
1995;
Lindell
and
Smith,
2003;
Waltert
et
al.,
2005).
Even
though
homogeneous
tree
plantations
with
a
dense
canopy
may
still
support
some
forest
species
(Thiollay,
1995;
Greenberg
et
al.,
1997a;
Peh
et
al.,
2006),
vegetation
heterogeneity
has
been
shown
to
increase
the
number
of
niches
and
consequently
bird
species
richness
(Roth,
1976;
Greenberg
et
al.,
1997a;
Sekercioglu,
2002;
Peh
et
al.,
2006).
Overall,
high
variation
in
forest
structure
was
of
particular
importance
for
forest
specialists
that
heavily
rely
on
primary
forest
habitat
(Sekercioglu,
2002;
Peh
et
al.,
2005).
In
this
study
we
quantified
bird
assemblages
in
five
different
forest
management
types
within
Kakamega
Forest,
western
Kenya.
Existing
studies
compared
only
a
few
different
forest
management
types
such
as
natural
forest,
exotic
and/or
indigenous
plantations
(e.g.
Sekercioglu,
2002;
Zurita
et
al.,
2006).
Moreover,
the
different
forest
management
types
were
predominantly
separated
by
cultivated
land
(e.g.
Waltert
et
al.,
2005;
Peh
et
al.,
2006)
and
not
located
within
one
forest.
We
compared
species
richness,
abundance
and
composition
of
bird
communities
in
stands
of
natural
forest,
in
plantations
with
a
mixture
of
indigenous
tree
species,
in
monocultures
of
an
indigenous
tree,
in
monocultures
of
an
exotic
tree
and
in
secondary
forest.
Moreover,
we
looked
specifically
whether
species
richness
and
abundance
of
avian
groups
with
different
habitat
requirements
(i.e.
forest
specialists,
generalists
or
visitors)
varied
among
these
forest
management
types.
Finally,
we
quantified
the
number
of
species
and
individuals
of
the
different
IUCN
red
list
categories
(IUCN,
2006)
for
each
forest
management
type.
2.
Material
and
methods
2.1.
Study
area
The
study
took
place
between
March
2005
and
March
2006
in
Kakamega
Forest,
western
Kenya
(00°08'-00°22'N,
34°46'-
34°57'E),
located
approximately
50
km
north-east
of
Lake
Victoria.
Kakamega
Forest
is
Kenya's
only
remaining
mid-altitude
tropical
rainforest
with
an
elevation
varying
between
1500
and
1700
m
and
is
considered
to
be
the
easternmost
remnant
of
the
lowland
Congo
basin
rainforests
of
Central
Africa
(Kokwaro,
1988).
Annual
precipitation
in
Kakamega
Forest
averages
1960
mm
(Forest
Department
records
at
Isecheno
Forest
Station
1982-2005)
and
is
highly
seasonal
with
a
rainy
season
from
April
to
November
and
a
short
dry
season
from
December
to
March.
Monthly
temperature
ranges
from
11
to
29
°C
(Kokwaro,
1988).
The
forest's
species
composition
indicates
a
transitional
position
between
the
lowland
Congolian
rainforests
and
the
Afro-montane
forests
east
of
the
Rift
valley
(White,
1983;
Kokwaro,
1988).
Kakamega
Forest
is
an
Important
Bird
Area
with
367
bird
species
including
two
globally
threatened
species
(Chapin's
Flycatcher
Muscicapa
lendu
and
Turner's
Eremomela
Eremomela
turneri
[BirdLife
International,
2006]),
and
15
regionally
threatened
species
(Bennun
and
Njoroge,
1999).
The
194
forest-dependent
species
include
40
of
Kenya's
43
Guinea—Congo
Forests
biome
species,
and
33
of
Kenya's
70
Afrotropical
Highlands
biome
species.
At
least
46
species
occurring
in
Kakamega
Forest
are
probably
found
nowhere
else
in
Kenya
(e.g.
Grey
Parrot
Psittacus
erithacus
[Bennun
and
Njoroge,
1999]).
The
natural
forest
cover
of
Kakamega
has
been
reduced
from
23,785
ha
in
1933
to
13,990
ha
in
the
1990s
(Blackett,
1994).
Forestry
operations
converted
large
areas
of
natural
forest
into
different
forest
management
types.
Kakamega
Forest
consists
of
a
mosaic
of
natural
forest
stands,
plantations
with
a
mixture
of
indigenous
tree
species,
monocultures,
secondary
forest
stands
as
well
as
some
open
areas.
Presently,
the
remaining
natural
forest
and
mixed
plantations
of
a
variety
of
indigenous
tree
species
cover
an
area
of
8457
ha
(Lung,
2004).
Monocultures
of
single
indigenous
species
(e.g.
Maesopsis
eminii),
monocultures
of
single
exotic
tree
species
(e.g.
Bischoffia
javanica)
and
secondary
forest
areas
comprise
a
total
area
of
4878
ha
(Lung,
2004).
The
different
forest
types
have
been
mapped
by
means
of
satellite
data
in
combination
with
ground
truth
information
and
stored
in
a
Geographic
Information
System
(GIS
[Lung,
2004;
Lung
and
Schaab,
2006]).
Kakamega
Forest
is
surrounded
by
agricultural
land,
belonging
to
the
most
densely
settled
parts
of
Kenya
(600
people/km
2
).
This
farmland
has
a
very
heterogeneous
structure
and
is
dominated
by
cultivation
of
sugar
cane,
maize,
tea,
fruits,
vegetables,
scattered
patches
of
natural
vegetation
and
isolated
trees
(KIFCON,
1994).
We
sampled
birds
in
five
forest
types,
i.e.
natural
forest,
plantations
of
a
mixture
of
indigenous
tree
species
(in
the
following
mixed
indigenous),
monocultures
of
the
indigenous
tree
M.
eminii
(in
the
following
indigenous
monoculture),
monocultures
of
the
exotic
tree
B.
javanica
(in
the
following
exotic
monoculture)
and
in
secondary
forest.
Natural
forest
stands
are
under
protection
of
the
Kenya
Wildlife
Service
and
show
only
low
levels
of
selective
logging
or
other
types
of
human
disturbance
(Bleher
et
al.,
2006).
These
areas
harbour
a
mean
number
of
40
different
tree
species
per
hectare
(N.F.,
unpublished
data).
Mixed
indigenous
plantations
have
been
planted
by
the
Kenya
Forest
Service
(KFS)
in
the
1940s
and
are
still
under
their
control
showing
high
levels
of
selective
logging
(Bleher
et
al.,
2006;
Mitchell,
personal
communication).
Here,
a
mean
number
of
30
different
tree
species
are
found
per
hectare
(N.F.,
unpublished
data).
Indigenous
and
exotic
monocultures
have
been
planted
by
KFS
in
the
1960s
(Mitchell,
personal
communication).
Secondary
forests
comprise
a
mean
number
of
19
trees
per
hectare
and
derive
from
the
1980s
(N.F.,
unpublished
data;
Mitchell,
personal
communication).
2.2.
Bird
sampling
To
quantify
the
bird
community
we
performed
point
counts
in
the
five
different
forest
types.
In
each
forest
type,
we
established
three
1-ha
plots.
Distances
between
study
plots
were
at
least
500
m.
The
position
of
the
15
plots
was
recorded
using
a
Global
Positioning
System
(Garmin
60).
We
censused
birds
in
each
of
the
15
plots
at
nine
point
count
locations
once
per
month
for
a
period
of
13
months.
The
1-ha
plots
were
dissected
by
five
marked
transects
of
100
m
length
with
neighbouring
transects
separated
from
each
other
by
20
m.
Three
point
count
locations
were
placed
N.
Farwig
et
al./
Forest
Ecology
and
Management
255
(2008)3885-3892
3887
along
the
first
transect
with
each
point
separated
from
the
other
by
40
m
(i.e.
at
10,
50,
and
90
m
distance
from
the
start
of
the
transect),
three
points
were
placed
along
the
third
and
three
along
the
fifth
transect,
resulting
in
nine
point
count
locations
per
plot
separated
by
a
minimum
of
40
m.
Point
counts
were
conducted
in
the
early
morning
(07:00-08:30
h)
recording
all
birds
heard
and
seen
within
a
radius
of
20
m
from
the
point
count
location
for
a
period
of
10
min.
All
point
counts
were
conducted
by
N.S.
Species
determination
followed
Zimmerman
et
al.
(1999).
The
study
design
with
three
1-ha
plots
per
forest
type
and
9
point
locations
per
plot
was
chosen
because
it
could
be
fitted
best
in
the
spatial
distribution
of
forest
stands,
with
a
minimum
distance
of
500
m
among
plots
allowing
for
a
maximum
of
spatial
independence.
Nine
point
count
locations
per
plot
were
chosen
to
increase
reliability
of
bird
data
per
plot
and
hence
forest
type.
For
analyses
we
pooled
for
each
monthly
census
all
bird
species
and
individuals
over
the
nine
point
count
locations
per
plot.
We
then
calculated
the
mean
number
of
species
and
the
mean
number
of
individuals
(log-transformed)
over
the
13
censuses
per
plot.
In
addition,
bird
species
were
classified
according
to
their
forest
dependency
as
either
(1)
forest
specialists,
for
those
bird
species
that
live
and
breed
in
the
interior
of
closed-canopy
or
little-
disturbed
forests;
(2)
forest
generalists
for
those
bird
species
that
can
occur
and
breed
in
both
undisturbed
and
disturbed
forest;
or
(3)
forest
visitors
for
those
bird
species
which
are
recorded
in
forests
but
are
more
common
in
non-forest
habitats
following
Bennun
et
al.
(1996).
We
then
calculated
the
mean
number
of
species
and
individuals
per
plot
separately
for
the
different
habitat
dependency
groups.
Further,
we
classified
the
species
according
to
IUCN
red
list
as
either
threatened
species
including
(1)
Critically
Endangered,
(2)
Endangered
and
(3)
Vulnerable
or
Lower
Risk
species
comprising
the
categories
(4)
Conservation
Dependent,
(5
)
Near
Threatened
and
(6)
Least
Concern
(IUCN,
2006).
After
that
we
quantified
the
number
of
species
and
individuals
for
the
different
threat
categories
for
each
forest
management
type.
2.3.
Distance
to natural
forest
and
vegetation
structure
For
each
of
the
151-ha
plots
we
determined
the
distance
from
the
middle
of
the
plot
to
the
nearest
natural
forest
stand
using
GIS
(ArcGIS
9.1).
To
determine
whether
bird
communities
were
associated
with
vegetation
heterogeneity
we
quantified
for
each
plot
the
vertical
foliage
height
diversity
at
the
nine
point
count
locations.
We
estimated
the
percentage
of
vegetation
cover,
again
within
a
circle
with
radius
20
m
around
the
point
count
location,
in
0,
1,
2,
4,
8,
16
and
32
m
height
to
the
nearest
5%.
We
then
calculated
vertical
foliage
height
diversity
using
the
Shannon
Wiener
diversity
index
with
low
values
indicating
that
all
vegetation
cover
was
in
one
or
a
few
vegetation
layers
and
high
values
for
an
even
distribution
of
vegetation
cover
across
the
seven
vegetation
layers
(Bibby
et
al.,
2000).
2.4.
Statistical
analyses
We
tested
for
effects
of
forest
type
on
the
mean
number
of
species
and
the
mean
number
of
individuals
(log-transformed)
per
plot.
We
also
tested
for
effects
on
mean
relative
species
richness
by
using
the
mean
number
of
species
as
response
variable
and
including
in
the
model
the
mean
number
of
individuals
(log-
transformed)
as
a
covariate
to
control
for
the
fact
that
species
richness usually
increases
with
the
number
of
sampled
individuals.
We
used
the
programme
EstimateS
(Colwell,
2005)
to
estimate
total
species
richness
per
plot.
We
used one
abundance-based
and
several
incidence-based
estimators
(Sekercioglu,
2002;
Peh
et
al.,
2006;
Zurita
et
al.,
2006;
Barlow
et
al.,
2007;
O'Dea
and
Whittaker,
2007).
Yet,
all
estimators
showed
results
similar
to
the
mean
number
of
bird
species.
We
therefore
report
only
the
results
on
mean
species
richness,
mean
number
of
individuals
and
mean
relative
species
richness.
We
used
ANOVA
and
ANCOVA
including
distance
to
nearest
natural
forest
and
vegetation
heterogeneity
as
covariates
in
the
model.
We
stepwise
excluded
first-interaction
terms
and
then
covariates
if
they
were
not
significant,
starting
with
the
least
significant.
Differences
among
forest
types
in
mean
species
richness,
number
of
individuals
and
relative
species
richness
were
tested
with
multiple
pairwise
comparisons
using
Tukey's
HSD
test
which
controls
the
group-wise
Type
I
error
rate
(Quinn
and
Keough,
2002).
We
ran
the
same
analyses
separately
for
forest
specialists,
forest
generalists,
and
forest
visitors.
The
composition
of
the
bird
communities
was
compared
among
the
different
forest
types
using
Principal
Component
Analysis
(PCA).
To
prevent
distortions
we
removed
rare
species,
present
in
two
or
fewer
plots
(15
species),
retaining
46
species
in
the
analysis.
Table
'1
ANOVAs
and
ANCOVAs
testing
the
effect
of
forest
type
on
mean
number
of
species,
mean
number
of
individuals
(log-transformed)
and
mean
relative
number
of
species
per
plot
for
all
bird
species
as
well
as
for
forest
specialists,
generalists
and
visitors
Source
d.f.
Mean
#
species
Mean
log
#
individuals
Mean
relative
#
species
R
2
F
P
R
2
F
P
R
2
F
P
All
species
Whole
model
4/5
0.79
9.48
0.0020
0.88
17.78
0.00020
0.79
6.83
0.0068
Forest
type
4
9.48
0.0020
17.78
0.00020
1.14
0.40
Mean
log
#
individuals
-/1
0.00
0.96
Forest
specialists
Whole
model
4/5
0.91
24.066
<0.0001
0.92
27.87
<0.0001
0.98
7453
<0.0001
Forest
type
4
24.066
<0.0001
27.87
<0.0001
9.56
0.0027
log
#
individuals
-/1
26.91
0.00060
Forest
generalists
Whole
model
4/5
0.68
538
0.014
0.88
19.20
0.00010
0.68
3.88
0.038
Forest
type
4
538
0.014
19.20
0.00010
1.24
036
log
#
individuals
-/1
0.0121
0.91
Forest
visitors
Whole
model
4/5
0.96
66.43
<0.0001
0.97
89.03
<0.0001
0.98
7453
<0.0001
Forest
type
4
66.43
<0.0001
89.03
<0.0001
0.82
0.55
log
#
individuals
-/1
4.84
0.06
Mean
log
#
individuals
was
included
as
covariate
in
the
analyses
of
relative
species
richness.
Given
are
model/error
d.f.-,
R
2
-,
F-
and
p-values.
Error
d.f.s
are
10/9.
Significant
results
(p
<
0.05)
are
given
in
bold.
mean
#
o
f
sp
ec
ies
(a)
30
25
20
15
10
5
0
a
a
lab
NatFor
a
a
ab
Mixlnd
a
a
ab
T
All
birds
I
I
Forest
specialists
Forest
generalists
I
Forest
visitors
b
a
a
ab
a
-
a
b
3888
N.
Farwig
et
al./
Forest
Ecology
and
Management
255
(2008)
3885-3892
The
PCA
was
based
on
the
correlation
matrix.
A
PCA
allows
comparing
the
similarity
in
the
composition
of
the
bird
commu-
nities
among
the
forest
types.
The
bird
species
with
the
highest
positive
and
negative
loadings
on
the
PC
axes
are
the
species
whose
abundance
most
distinctively
differs
among
the
forest
types.
All
analyses
were
done
with
STATISTICA
7.1
(Statsoft
Inc.,
2005).
3.
Results
We
detected
a
total
of
61
bird
species
and
6738
bird
individuals
across
all
plots
and
censuses
(Appendix
A).
Among
the
species,
25
were
forest
specialists,
25
forest
generalists,
and
nine
forest
visitors
(two
species
could
not
be
classified).
Almost
50%
of
all
detected
individuals
were
forest
specialists,
37%
were
forest
generalists
and
13%
were
forest
visitors.
Turner's
Eremomela
(2
individuals
detected
in
indigenous
monoculture
plots)
is
the
only
globally
threatened
species
that
was
observed
during
the
study
period
(Appendix
A).
All
other
60
species
belong
to
the
category
Least
Concern.
We
recorded
significant
differences
in
mean
bird
species
richness,
number
of
individuals
and
relative
species
richness
among
the
five
forest
types
(Table
1).
Multiple
pairwise
comparisons
showed
significantly
higher
numbers
of
species
and
individuals
in
natural
forest,
mixed
indigenous
and
indigenous
monoculture
plots
than
in
exotic
monoculture
and
secondary
forest
plots
(Fig.
1).
Plots
in
mixed
indigenous
plantations
and
indigenous
monocultures
had
significantly
higher
relative
species
richness
than
plots
in
exotic
monocultures
and
secondary
forest
(Fig.
1).
This
pattern
was
caused
by
significantly
more
forest
specialist
species
and
individuals
in
natural
forest,
mixed
indigenous
and
indigenous
monoculture
plots
compared
to
plots
in
exotic
monocultures
and
secondary
forest
(Table
1,
Fig.
1).
The
pattern
was
similar
for
relative
species
richness
but
less
pronounced.
Also,
forest
generalist
species
and
individuals
differed
significantly
among
forest
types
(Table
1).
Most
species
and
individuals
were
recorded
in
indigenous
monocultures
differing
significantly
from
all
other
forest
types
(Fig.
1).
The
analyses
for
forest
visitor
species
and
individuals
also
showed
significant
differences
among
forest
types
(Table
1).
Here,
the
number
of
species
and
individuals
was
significantly
lower
in
natural
forest
and
mixed
indigenous
plots
than
in
indigenous
and
exotic
monocultures
and
secondary
forest
plots.
Neither
the
covariate
distance
to
nearest
natural
forest
nor
the
covariate
vegetation
heterogeneity
was
significant
in
any
of
the
analyses.
Thus,
both
covariates
were
excluded
from
all
models.
The
PCA
clearly
differentiated
the
composition
of
the
bird
communities
among
the
15
plots
by
forest
type.
The
first
ordination
axis
(explaining
a
variance
of
38.7%)
separated
natural
forest,
mixed
indigenous
and
indigenous
monoculture
plots
from
exotic
monoculture
and
secondary
forest
plots
(Fig.
2).
Natural
forest
and
mixed
indigenous
plantations
had
the
greatest
negative
loadings
and
secondary
forest
plots
the
greatest
positive loadings.
Indigenous
and
exotic
monocultures
were
located
in
between
these
extremes
(Fig.
2).
The
three
bird
species
with
the
greatest
negative
loadings
were
all
forest
specialists
whereas
two
of
the
bird
species
with
the
greatest
positive
loadings
were
forest
visitors
and
one
a
forest
specialist
(Fig.
2).
The
second
axis
of
the
ordination
(explaining
10.9%
of
the
variance)
separated
in
particular
the
natural
forest
and
mixed
indigenous
plots
from
the
homogenous
indigenous
monocultures
(Fig.
2).
Mixed
indigenous
plots
were
scattered
among
the
natural
forest
plots
suggesting
similar
species
compositions.
Here,
two
of
the
species
with
the
greatest
positive
loadings
were
forest
generalists
and
one
a
forest
specialist
whereas
all
three
species
with
the
greatest
negative
loadings
were
forest
generalists
(Fig.
2).
All
birds
O
Forest
specialists
a
Forest
generalists
Forest
visitors
b
b
T
T
as
bb
7
Ic
Ind
Mono
ExotMono
SecFor
forest
type
(b)
1.8
1.6
To
1.4
1.2
T2
1.0
o
0.8
ar
ai
E
co
0.4
0.2
0.0
NatFor
MixInd
IndMono
ExotMono
SecFor
forest
type
All
birds
O
Forest
specialists
Forest
generalists
I
Forest
visitors
b
T
ab
lc
Modnd
IndMono
ExotMono
forest
type
Fig.
1.
(a)
Mean
number
of
species
(b)
mean
number
of
individuals
(log-
transformed)
and
(c)
mean
relative
number
of
species
of
all
birds,
of
forest
specialists,
generalists
and
visitors
recorded
during
point
counts
in
five
different
forest
types,
i.e.
natural
forest
(NatFor),
mixed
indigenous
plantation
(MixInd),
indigenous
monoculture
(IndMono),
exotic
monoculture
(ExotMono)
and
secondary
forest
(SecFor).
Shown
are
Ism
(+1S.D.),
different
letters
indicate
significant
differences
among
forest
types.
b
0.6
m
ean
r
d.
#
o
f
bir
d
sp
eci
es
CO
1.
11
-
I-
I
Tab
a
ri
NatFor
5-
0
aab
1
n
b
c
b
SecFor
2
-
PC
2
(
10.
9%)
-2
-
N.
Farwig
et
al./
Forest
Ecology
and
Management
255
(2008)3885-3892
3889
4
-
Honeyguide
Greenbul
FS
(0.236)
Ross's
Turaco
FG
(0.227)
Crested
Guineafowl
FG
(0.227)
SF1
SF2
SF3
EMo3
EMo1
EMo2
IMol
Black
and
White
Casqued
Hornbill
FG
(-0.356)
Black-collared
Apalis
FG
(-0.354)
-4
-
Bocage's
Bushshrike
FG
(-0.275)
MI3
MI1
NF3
IMo3
IMo2
0
-4
-2
Ugandan
Woodland
Warbler
FS
(-0.227)
Olive-green
Camaroptera
FS
(-0.216)
Red-tailed
Bristlebill
FS
(-0.210)
2
4
6
PC1
(38.7%)
Grey-backed
Camaroptera
FV
(0.227)
Common
Wattle-Eye
FV
(0.219)
Grey-winged
Robin
FS
(0.188)
Fig.
2.
Bird
community
composition
in
five
forest
types.
i.e.
natural
forest
(NF),
mixed
indigenous
plantation
(MI),
indigenous
monoculture
(IMo),
exotic
monoculture
(EMo),
and
secondary
forest
(SF)
along
two
PC
axes.
The
characteristic
bird
species
and
their
forest
dependency
(forest
specialists
FS,
forest
generalists
FG,
forest
visitors
FV)
are
indicated
on
each
axis
with
their
loadings
in
brackets.
4.
Discussion
The
mean
number
of
bird
species
and
individuals
showed
significant
differences
among
the
five
forest
types.
Natural
forest
and
indigenous
plantations
had
more
species
and
individuals
than
exotic
plantations
and
secondary
forest.
This
was
caused
mainly
by
a
decline
of
forest
specialists
and
generalists
from
natural
forest
and
indigenous
plantations
towards
exotic
plantations
and
secondary
forest;
the
decline
was
only
partly
compensated
by
a
slight
increase
of
forest
visitors.
Accordingly,
a
PCA
clearly
separated
the
bird
communities
in
the
forest
types
from
one
another.
The
overall
number
of
detected
bird
species
was
with
61
species
relatively
low
compared
to
the
367
bird
species
that
have
been
recorded
for
Kakamega
Forest
(Bennun
and
Njoroge,
1999).
Yet,
a
large
number
of
the
367
species
is
either
restricted
to
specific
habitats
such
as
wetlands
or
grasslands
that
were
not
sampled
in
this
study,
or
are
present
in
Kakamega
Forest
only
occasionally
such
as
vagrants.
We
recorded
more
than
25%
of
the
forest-
dependent
species
representing
most
of
the
common
forest
birds
occurring
in
Kakamega
Forest.
Our
results
further
demonstrate
that
by
recording
the
abundant
species,
we
already
detected
significant
differences
among
the
studied
forest
types.
As
a
consequence
we
expect
that
additional
sampling
focussing
on
threatened
and
near-threatened
species
would
result
in
even
stronger
differences
in
bird
communities
among
the
forest
types.
So
far,
the
differences
in
bird
communities
among
the
five
forest
types
were
mainly
caused
by
changes
in
the
mean
number
of
bird
individuals.
Patterns
in
mean
species
richness
showed
similar
but
less
pronounced
results
and
mean
relative
species
richness
was
least
different
among
the
forest
types
stressing
the
fact
that
species
richness
usually
increases
with
the
number
of
sampled
individuals.
Our
findings
are
in
line
with
previous
studies
that
show
lower
species
richness
and
abundance
in
managed
forest
areas
than
in
natural
forests
(Sekercioglu,
2002;
Waltert
et
al.,
2005;
Peh
et
al.,
2005,
2006;
Zurita
et
al.,
2006;
O'Dea
and
Whittaker,
2007).
In
concordance
with
previous
studies
lower
species
richness
and
abundance
were
predominately
caused
by
the
decline
of
forest-
dependent
birds
such
as
forest
specialists
and
generalists
in
non-
natural
forests
(Sekercioglu,
2002;
Waltert
et
al.,
2005;
Peh
et
al.,
2005,
2006;
Zurita
et
al.,
2006).
In
our
study,
this
decrease
of
forest
specialists
and
generalists
in
managed
areas
was
partly
compen-
sated
by
an
increase
in
forest
visitors.
Similar
findings
have
been
demonstrated
in
managed
forest
areas
elsewhere
(e.g.
Estrada
et
al.,
1997;
Sekercioglu,
2002;
Castelletta
et
al.,
2005;
Waltert
et
al.,
2005).
This
becomes
especially
obvious
in
a
PCA
in
which
the
first
axis
connected
natural
forest
plots,
mixed
indigenous
plots
and
indigenous
monocultures
with
forest
specialist
species
while
exotic
monocultures
and
secondary
forest
plots
were
characterised
by
forest
visitor
species.
The
second
axis
separated
the
natural
forest
plots
and
mixed
indigenous
plots
that
were
connected
by
specific
forest
generalist
and
forest
specialist
species
from
the
indigenous
monocultures
that
were
characterised
by
a
different
set
of
forest
generalist
species.
Thus,
even
though
some
managed
forest
areas
may
comprise
high
numbers
of
birds;
species
composition
differed
considerably
among
natural
forest
and
monocultures.
Further,
bird
assemblages
in
little
and
highly
disturbed
natural
forest
plots
as
recorded
in
a
previous
study
within
Kakamega
Forest
contained
about
the
same
number
of
species,
of
forest
specialist,
forest
generalist
and
forest
visitor
species
as
the
natural
forest
plots
in
the
present
study
(Farwig
et
al.,
2006;
Bleher,
unpublished
data).
Also,
the
species
composition
of
these
differently
disturbed
natural
forest
plots
was
similar
to
the
one
in
natural
forests
and
mixed
indigenous
plots
of
this
study.
Yet,
even
though
bird
assemblages
were
very
similar
in
the
differently
disturbed
natural
forest
areas,
modification
of
ecological
processes
have
been
recorded
(Farwig
et
al.,
2006,
in
press-a;
Kirika
et
al.,
2008,
in
press).
For
instance,
seed
dispersal
of
fig
trees
and
overall
seedling
establishment
were
diminished
in
heavily
compared
to
little
disturbed
sites
(Farwig
et
al.,
in
press-a;
Kirika
et
al.,
in
press).
Thus,
it
can
be
expected
that
the
changes
in
bird
community
composition
recorded
among
the
different
forest
management
types
result
in
alterations
of
valuable
ecological
processes
and
functions
(Balvanera
et
al.,
2001;
Sekercioglu,
2006a,b;
Tscharntke
et
al.,
in
press).
In
our
study,
distance
to
the
nearest
natural
forest
had
no
effect
on
birds.
This
might
be
due
to
a
number
of
reasons.
First,
before
the
1930s
the
complete
area
was
covered
by
natural
forest
and
forest
conversion
has
only
taken
place
within
the
last
70-80
years
(Blackett,
1994;
Lung,
2004;
Farwig
et
al.,
in
press-b).
This
former
connectivity
might
explain
the
presence
of
forest
specialist
species
in
the
different
forest
types
(Castelletta
et
al.,
2005)
that
are
presumably
still
present
within
their
former
ranges.
Second,
3890
N.
Farwig
et
al./
Forest
Ecology
and
Management
255
(2008)
3885-3892
distances
to
natural
forests
were
in
general
very
small
(range
0.4-
1.7
km)
suggesting
that
nearby
natural
forest
areas
might
have
served
as
source
habitat.
Even
forest
plantations
at
larger
distances
(-10
km)
from
primary
forests
comprise
a
variety
of
forest
bird
species
indicating
that
the
natural
habitat
might
have
served
as
a
source
area
(Castelletta
et
al.,
2005;
Peh
et
al.,
2006).
Third,
the
habitat
separating
our
study
plots
consisted
of
a
mosaic
of
different
forest
stands
which
might
facilitate
movement
even
of
forest-
dependent
species
between
the
studied
forest
management
types.
In
a
number
of
studies
vegetation
structure
in
managed
forests
differed
strongly
from
natural
forests
explaining
differences
in
bird
assemblages
(e.g.
Sekercioglu,
2002;
Sodhi
et
al.,
2004;
Peh
et
al.,
2006;
Zurita
et
al.,
2006).
Structural
differences
were
particularly
evident
in
lower
canopy
closure
or
vegetation
heterogeneity
(Sekercioglu,
2002;
Peh
et
al.,
2006).
Yet,
our
analyses
showed
no
significant
effect
of
vegetation
heterogeneity
on
birds.
Even
though
vegetation
heterogeneity
differed
sig-
nificantly
among
the
forest
types
(ANOVA:
F4
,
10
=
14.08,
p
=
0.0004),
the
categorical
predictor
forest
type
seems
to
predict
the
bird
communities
on
the
plots
more
distinctively
than
our
index
of
vegetation
heterogeneity.
As
our
predictors
distance
to
natural
forest
and
vegetation
heterogeneity
did
not
explain
the
differences
in
species
assem-
blages
among
the
five
different
forest
types
other
factors
seem
to
be
important
(Fig.
2).
For
instance,
our
index
of
vegetation
heterogeneity
might
not
be
appropriate
to
distinguish
subtle
differences
in
microhabitats
such
as
dead
trees
or
the
composition
of
the
leaf
litter
among
the
management
types
that
might
be
important
for
understory
insectivores
(Canaday,
1996;
Ford
et
al.,
2001;
Sekercioglu
et
al.,
2002).
Also,
the
diversity
and
composition
of
the
tree
communities
varied
among
the
studied
forest
types.
This
might
lead
to
differences
in
overall
food
availability.
Forest
modification
has
been
shown
to
lead
to
impoverished
resources
for
understory
insectivores
(Zanette
et
al.,
2000;
Ford
et
al.,
2001)
but
also
fruit
or
nectar
availability
might
differ
among
the
studied
forest
types.
Thus,
more
detailed
studies
are
needed
to
detect
important
microhabitat
structures
or
other
factors
that
determine
the
differences
in
species
composition
between
natural
and
managed
forests.
The
results
of
our
study
have
implications
for
forest
manage-
ment.
Species
composition
of
mixed
indigenous
plantations
was
Appendix
A
very
similar
to
the
one
of
natural
forests
(Fig.
2).
Mixed
indigenous
plantations
were
planted
already
around
the
early
1940s
and
comprise
a
high
number
of
different
tree
species.
This
could
explain
the
high
resemblance
to
natural
forests
suggesting
a
high
potential
of
mixed
indigenous
plantations
for
forest
restoration.
Also
on
satellite
images,
natural
forest
cannot
be
distinguished
from
mixed
indigenous
plantations
(Lung,
2004).
Monocultures
of
the
indigenous
species
M.
eminii
retained
a
much
higher
proportion
of
forest
birds
including
the
only
globally
threatened
species
than
monocultures
of
the
exotic
species
B.
javanica
even
though
both
forest
types
have
been
planted
in
the
same
period.
Thus,
in
our
study
the
indigenous
monoculture
seems
to
provide
better
habitat
for
forest
birds
than
the
exotic
monoculture.
However,
our
results
have
to
be
interpreted
cautiously
as
we
only
compared
monocultures
of
two
different
tree
species.
Secondary
forests
comprised
the
least
number
of
bird
species
and
individuals.
This
might
be
due
to
the
fact
that
these
areas
date
only
from
the
1970s
and
consequently
comprise
only
a
low
number
of
different
tree
species
and
hence
habitat.
To
conclude,
our
results
showed
significantly
higher
species
richness
and
bird
numbers
in
natural
forest
than
in
tree
plantations
of
single
tree
species.
In
addition,
species
composition
in
these
tree
plantations
differed
markedly
from
natural
forest
with
forest
visitors
partly
replacing
forest
specialists.
Thus,
the
maintenance
of
large
natural
forest
areas
is
needed
to
conserve
the
diversity
of
forest
birds.
Nevertheless,
plantations
with
a
mixture
of
indigen-
ous
tree
species
may
be
useful
to
compensate
for
forest
loss,
as
their
bird
species
richness
and
community
composition
closely
resembled
natural
forests.
Thus,
mixed
indigenous
plantation
maintaining
a
high
conservation
value
for
birds
might
be
a
useful
tool
for
conservation
management.
Acknowledgements
We
thank
the
Kenya
Wildlife
Service
and
the
Forestry
Department
for
their
permission
to
work
in
Kakamega
Forest.
Field
work
benefited
from
the
help
of
B.B.
Chituyi.
We
thank
B.
Bleher
for
useful
input,
D.G.
Berens
and
one
anonymous
reviewer
for
comments
on
the
manuscript.
Financial
support
was
provided
by
the
BMBF
(BIOTA
East
Africa
01LC0405).
The
field
work
complied
with
the
current
laws
of
Kenya.
List
of
61
bird
species,
their
forest
dependency,
their
IUCN
threat
category
and
their
total
number
of
observations
during
point
counts
in
stands
of
natural
forest
(NatFor),
mixed
indigenous
forest
plantations
(MimInd),
monocultures
of
the
indigenous
tree
Maesopsis
eminii
(IndMono),
monocultures
of
the
exotic
tree
Bischoffia
javanica
(ExotMono)
and
secondary
forest
(SecFor).
Forest
dependency:
forest
specialist
(FS),
forest
generalist
(FG),
forest
visitor
(FV);
IUCN
status:
Least
Concern
(LC),
endangered
(EN).
Common
name-scientific
name
Forest
dependency
IUCN
status
NatFor
Mixind
IndMono
ExotMono
SecFor
African
Broadbill-Smithornis
capensis
meinertzhageni
FS
LC
3
17
1
0
0
African
Emerald
Cuckoo-Chrysococcyx
cupreus
FG
LC
0
3
7
4
5
African
Paradise
Flycatcher—Terpsiphone
viridis
FV
LC
12
2
4
25
11
Bar-tailed
Trogon-Apaloderma
vittatum
FS
LC
1 1
0 0
0
Black
and
White
Casqued
Hornbill—Bycanistes
subcylindricus
FG
LC
40
21
86
24
11
Black-billed
Weaver-Ploceus
melanogaster
stephanophorus
FS
LC
2
0
0 0
0
Black-collared
Apalis—Apalis
p.
pulchra
FG
LC
1
1
103
2
0
Black-faced
Rufous
Warbler-Bathmocercus
rufus
vulpinus
FS
LC
128
152
167
7
12
Blue-headed
Bee-eater-Merops
m.
muelleri
FS
LC
14
7
8
0
0
Blue-shouldered
Robin-Chat-Cossypha
cyanocampter
bartteloti
FG
LC
56
38
27
4
5
Bocages'
Bush
Shrike-Malaconotus
bocagei
jacksoni
FG
LC
1
1
4
5
6
Brown
llladopsis-Illadopsis
fulvescens
ugandae
FS
LC
51
71
52
57
69
Brown-eared
Woodpecker-Campethera
c.
caroli
FG
LC
0
1
0 0
0
Cabanis
Greenbul-Phyllastrephus
cabanisi
FS
LC
68
43
88
40
24
Chestnut
Wattle-eye-Dyaphorophyia
c.
castanea
FS
LC
3
0
2
0
0
Common
Bulbul-Pycnonotus
barbatus
FV
LC
28
35
70
123
92
Common
Drongo-Dicrurus
a.
adsimilis
LC
25
17
21
0
0
N.
Farwig
et
aL/
Forest
Ecology
and
Management
255
(2008)3885-3892
Appendix
A
(Continued)
3891
Common
name-scientific
name
Forest
dependency
IUCN
status
NatFor
Mixind
IndMono
ExotMono
SecFor
Common
Wattle-eye-Platysteira
cyanea
nyansae
FV
LC
2
0
12
23
55
Crested
Guineafowl-Guttera
pucherani
FG
LC
10
3
0
0
1
Crowned
Hornbill-Tockus
alboterminatus
geloensis
FV
LC
0 0
0
0
2
Dark-backed
Weaver-Ploceus
bicolor
FG
LC
31
41
39
23
Equatorial
Akalat-Sheppardia
p.
polioptera
FS
LC
58
57
27
10
0
Great
Blue
Turaco-Corythaeola
cristata
FG
LC
3
4
Great
Sparrowhawk-Accipiter
m.
melanoleucus
FG
LC
2
0 0
Green-headed
Sunbird-Nectarinia
verticalis
viridisplendens
FG
LC
0
0
0
0
0
1
Grey-backed
Camaroptera-Camaroptera
brachyura
FV
LC
0
108
161
Grey-throated
Barbet-Gymnobucco
bonapartei
FG
LC
16
36
25
19
3
Grey-winged
Robin-Sheppardia
p.
polioptera
FS
LC
0 0
2
7
29
Honeyguide
Greenbul-Baeopogon
i.
indicator
FS
LC
11
23
7
12
6
Jamesons'
Wattle-eye-Dyaphorophyia
jamesoni
FS
LC
0 0
2
0 0
Joyful
Greenbul-Chlorocichla
leatissima
FG
LC
30
55
42
21
23
Klaas'
Cuckoo-Chrysococcyx
klaas
FV
LC
0
0
1
2
Liihder's
Bush
Shrike-Laniarius
L
luehderi
FV
LC
0 0
39
12
20
Olive
Sunbird-Nectarinia
olivacea
changamwensis
FS
LC
42
50
38
29
24
Olive
Thrush-Turdus
olivaceus
abyssinicus
FG
LC
0 0
0
0
2
Olive-green
Camaroptera-Camaroptera
chloronota
toroensis
FS
LC
140
133
112
20
14
Red-chested
Cuckoo-Cuculus
s.
solitarius
FG
LC
4
8
3
3
14
Red-headed
Bluebill-Spermophaga
r.
ruficapilla
FG
LC
0 0
1
2
0
Red-tailed
Bristlebill-Bleda
syndactyla
woosnami
FS
LC
87
92
77
54
27
Ross's
Turaco-Musophaga
rossae
FG
LC
2
1
5
Scaly
Francolin-Francolinus
squamatus
FG
LC
0
0
0
0
0 0
Scaly-breasted
Illadopsis-Illadopsis
albipectus
FS
LC
60
35
40
11
Shelley's
Greenbul-Andropadus
masukuensis
kakamegae
FS
LC
0 0
1
0
Slender-billed
Greenbul-Andropadus
gracilirostris
FS
LC
0
0
0 0
Snowy-headed
Robin-Chat-Cossypha
niveicapilla
melanota
FG
LC
0 0
0
0
0
Stuhlmann's
Starling-Poeoptera
stuhlmanni
FS
LC
0 0
0
13
Tambourine
Dove-Turtur
tympanistria
FG
LC
21
20
15
27
34
Toro
Olive
Greenbul-Phyllastrephus
hypochloris
FS
LC
4
0
2
2
113
Turner's
Eremomela-Eremomela
turneri
FS
EN
0 0
2
0 0
Uganda
Woodland
Warbler-Phylloscopus
budongoensis
FS
LC
140
143
104
38
11
Western
Black-headed
Oriole-Oriolus
brachyrhynchus
laetior
FG
LC
32
28
9
2
White-browed
Scrub
Robin-Cercotrichas
leucophrys
LC
0
0
4
0
White-chinned
Prinia-Prinia
leucopogon
reichenowi
FG
LC
0
0
3
5
White-headed
Saw-wing-Psalidoprocne
a.
albiceps
FV
LC
0
0 0
0
1
White-headed
Wood
Hoopoe-Phoeniculus
bollei
jacksoni
FS
LC
0
0
0 0
White-tailed
Ant
Thrush-Neocossyphus
poensis
praepectoralis
FS
LC
51
55
32
35
34
Yellow
White-eye-Zosterops
senegalensis
FV
LC
4
11
8
Yellow-billed
Barbet-Trachylaemus
purpuratus
elgonensis
FG
LC
0
0
3
1
0
Yellow-rumped
Tinkerbird-Pogoniulus
bilineatus
FG
LC
90
81
91
98
81
Yellow-spotted
Barbet-Buccanodon
d.
duchaillui
FS
LC
24
15
13
8
2
Yellow-whiskered
Greenbul-Andropadus
L
latirostris
FG
LC
211
225
216
172
79
References
Balvanera,
P.,
Daily,
G.C.,
Ehrlich,
P.R.,
Ricketts,
P.H.,
Bailey,
S.A.,
Kark,
S.,
Kremen,
C.,
Pereira,
H.,
2001.
Conserving
biodiversity
and
ecosystem
services.
Science
291,
2047-12047.
Barlow,
J.,
Gardner,
TA,
Araujo,
I.S.,
Avila-Pires,
T.C.,
Bonaldo,
A.B.,
Costa,
J.E.,
Esposito,
M.C.,
Ferreira,
LV.,
Hawes,
J.,
Hernandez,
M.I.M.,
Hoogmoed,
M.S.,
Leite,
R.N.,
Lo-Man-Hung,
N.F.,
Malcom,
J.R.,
Martins,
M.B.,
Mestre,
LA.M.,
Miranda-Santos,
R.,
Nunes-Gutjahr,
A.L,
Overal,
W.L,
Peters,
S.L,
Ribeiro-Junior,
MA.,
da
Silva,
M.N.F.,
da
Silva
Motta,
C.,
Peres,
CA,
2007.
Quantifying
the
biodiversity
value
of
tropical
primary,
secondary,
and
plantation
forests.
Proc.
Natl.
Acad.
Sci.
U.S.A.
104,
18555-18560.
Bennun,
LC.,
Dranzoa,
C.,
Pomeroy,
D.,
1996.
The
forest
birds
of
Kenya
and
Uganda.
J.
East.
Mr.
Nat.
Hist.
85,
23-48.
Bennun,
L,
Njoroge,
P.,
1999.
Important
Bird
Areas
in
Kenya.
NatureKenya,
Nairobi.
Bibby,
CJ.,
Burgess,
N.D.,
Hill,
D.A.,
Mustoe,
S.H.,
2000.
Bird
Census
Techniques.
Academic
Press,
London.
BirdLife
International,
2006.
Species
factsheets.
Available
online
at
http://www.bir
-
dlife.org.
Blackett,
H.L,
1994.
Forest
Inventory
Report
No.
3
Kakamega.
Forest
Dept./KIFCON,
Nairobi.
Bleher,
B.,
Uster,
D.,
Bergsdorf,
T.,
2006.
Assessment
of
threat
status
and
manage-
ment
effectiveness
in
Kakamega
forest,
Kenya.
Biodivers.
Conserv.
15,
1159-
1177.
Castelletta,
M.,
Thiollay,
J.M.,
Sodhi,
N.S.,
2005.
The
effects
of
extreme
forest
fragmentation
on
the
bird
community
of
Singapore
Island.
Biol.
Conserv.
121,
135-155.
Canaday,
C.,
1996.
Loss
of
insectivorous
birds
along
a
gradient
of
human
impact
in
Amazonia.
Biol.
Conserv.
77,
63-77.
Colwell,
R.K.,
2005.
EstimateS:
statistical
estimation
of
species
richness
and
shared
species
from
samples.
Version
7.5.
Available
online
at
http://purl.ocic.org/
estimates.
Curran,
LM.,
Trigg,
S.N.,
McDonald,
A.K.,
Astiani,
D.,
Hardiono,
Y.M.,
Siregar,
P.,
Caniago,
I.,
Kasischke,
E.,
2004.
Lowland
forest
loss
in
protected
areas
of
Indonesian
Borneo.
Science
303,
1000-1003.
Estrada,
A.,
Coates-Estrada,
R.,
Merrit,
D.A.,
1997.
Anthropogenic
landscape
change
and
avian
diversity
at
Los
Tuxtlas,
Mexico.
Biodivers.
Conserv.
6,
19-43.
FAO,
2001.
State
of
the
World's
Forests
2001.
Food
and
Agriculture
Organisation
of
the
United
Nations,
Rome.
Farwig,
N.,
Bohning-Gaese,
K.,
Bleher,
B.,
2006.
Enhanced
seed
dispersal
of
Prunus
africana
in
fragmented
and
disturbed
forests?
Oecologia
147,
238-252.
Farwig,
N.,
Sajita,
N.,
Schaab,
G.,
BOhning-Gaese,
K.
Human
impact
diminishes
seedling
species
richness
in
Kakamega
Forest,
Kenya.
Basic
Appl.
Ecol.,
in
press.
Farwig,
N.,
Braun,
C.,
BOhning-Gaese,
K.
Human
disturbance
reduces
genetic
diver-
sity
of
an
endangered
tropical
tree,
Prunus
africana
(Rosaceae).
Conserv.
Genet.,
in
press.
Ford,
HA.,
Barrett,
G.W.,
Saunders,
D.A.,
Recher,
H.F.,
2001.
Why
have
birds
in
the
woodlands
of
Southern
Australia
declined?
Biol.
Conserv.
97,
71-88.
Greenberg,
R.,
Bichier,
P.,
Angon,
A.C.,
Reitsma,
R.,
1997a.
Bird
populations
in
shade
and
sun
coffee
plantations
in
central
Guatemala.
Conserv.
Biol.
11,
448-459.
Greenberg,
R.,
Bichier,
P.,
Sterling,
J.,
1997b.
Bird
populations
in
rustic
and
planted
shade
coffee
plantations
of
eastern
Chiapas,
Mexico.
Biotropica
29,
501-514.
IUCN,
2006.
2006
IUCN
Red
List
of
Threatened
Species,
http://www.iucnredlist.org/.
KIFCON,
1994.
Kakamega
Guide.
The
Official
Guide.
Kenya
Indigenous
Forest
Conservation
Programme,
Nairobi.
Kirika,
J.M.,
Bleher,
B.,
Bohning-Gaese,
B.,
Chira,
R.,
Farwig,
N.
Fragmentation
and
local
disturbance
of
forests
reduce
frugivore
diversity
and
fruit
removal
in
Ficus
thonningii.
Basic
Appl.
EcoL,
in
press.
Kirilca,
J.M.,
Farwig,
N.,
BOhning-Gaese,
K.,
2008.
Local
disturbance
of
tropical
forests
diminishes
frugivores
and
seed
removal
of
a
small-seeded
Afrotropical
tree.
Conserv.
Biol.
22,
318-328.
Kokwaro,
J.0.,
1988.
Conservation
status
of
the
Kakamega
Forest
in
Kenya.
The
Eastern
most
relic
of
the
equatorial
rainforest
of
Africa.
Monogr.
Syst.
Bot.
Mo.
Bot.
Gard.
25,
471-489.
3892
N.
Farwig
et
al./
Forest
Ecology
and
Management
255
(2008)
3885-3892
Lewis,
S.L.,
2006.
Tropical
forests
and
the
changing
earth
system.
Philos.
Trans.
R.
Soc.
B
361,
195-210.
Lindell,
C.,
Smith,
M.,
2003.
Nesting
bird
species
in
sun
coffee,
pasture,
and
understory
forest
in
southern
Costa
Rica.
Biodivers.
Conserv.
12,
423-440.
Lung,
T.,
2004.
Landbedeckungsanderungen
im
Gebiet
"Kakamega
Forest
und
assoziierte
Waldgebiete"
(Westkenia).
Mulitspektrale
Klassifikation
von
Land-
sat-Satellitenbilddaten
und
Auswertung
mittels
Methoden
im
Raster-GIS.
Karlsruher
Geowissenschaftliche
Schriften,
Schaab
G
(Ed.),
Karlsruhe.
Lung,
T.,
Schaab,
G.,
2006.
Assessing
fragmentation
and
disturbance
of
west
Kenyan
rainforests
by
means
of
remotely
sensed
time
series
data
and
landscape
metrics.
Mr.
J.
Ecology
44,
491-506.
Mayaux,
P.,
Holmgren,
P.,
Achard,
F.,
Eva,
H.,
Stibig,
HJ.,
Banthomme,
A.,
2005.
Tropical
forest
cover
change
in
the
1990s
and
options
for
future
monitoring.
Philos.
Trans.
R.
Soc.
B
360,
373-384.
Myers,
N.,
Mittermeier,
R.A.,
Mittermeier,
C.G.,
da
Fonseca,
GAB.,
Kent,
J.,
2000.
Biodiversity
hotspots
for
conservation
priorities.
Science
403,
853-858.
O'Dea,
N.,
Whittaker,
R.J.,
2007.
How
resilient
are
Andean
montane
forest
bird
communities
to
habitat
degradation?
Biodivers.
Conserv.
16,
1131-1159.
Peh,
K.S.H.,
de
Jong,
J.,
Sodhi,
N.S.,
Lim,
&LH.,
Yap,
CAM.,
2005.
Lowland
rainforest
avifauna
and
human
disturbance:
persistence
of
primary
forest
birds
in
selec-
tively
logged
forests
and
mixed-rural
habitats
of
southern
Peninsular
Malaysia.
Biol.
Conserv.
123,
489-505.
Peh,
K.S.H.,
Sodhi,
N.S.,
de
Jong,
J.,
Sekercioglu,
C.H.,
Yap,
CAM.,
Lim,
&LH.,
2006.
Conservation
value
of
degraded
habitats
for
forest
birds
in
southern
Peninsular
Malaysia.
Divers.
Distrib.
12,
572-581.
Peres,
CA.,
Barlow,
J.,
Laurance,
F.W.,
2006.
Detecting
anthropogenic
disturbance
in
tropical
forests.
Trends
Ecol.
Evol.
21,
227-229.
Petit,
L.J.,
Petit,
D.R.,
Christian,
D.G.,
Powell,
H.D.,
1999.
Bird
communities
of
natural
and
modified
habitats
in
Panama.
Ecography
22,
292-304.
Quinn,
G.P.,
Keough,
KJ.,
2002.
Experimental
Design
and
Data
Analysis
for
Biol-
ogists.
Cambridge
University
Press,
Cambridge.
Raman,
T.R.S.,
2001.
Effect
of
slash-and-burn
shifting
cultivation
on
rainforest
birds
in
Mizoram,
north-east
India.
Conserv.
Biol.
15,
685-698.
Roth,
R.R.,
1976.
Spatial
heterogeneity
and
bird
species
diversity.
Ecology
57,
773-
782.
Sekercioglu,
C.H.,
2002.
Effects
of
forestry
practices
on
vegetation
structure
and
bird
community
in
Kibale
National
Park,
Uganda.
Biol.
Conserv.
107,
229-240.
Sekercioglu,
C.H.,
Ehrlich,
P.R.,
Daily,
G.C.,
Aygen,
D.,
Goehring,
D.,
Sandi,
R.F.,
2002.
Disappearance
of
insectivorous
birds
from
tropical
forest
fragments.
Proc.
Natl.
Acad.
Sci.
U.S.A.
99,
263-267.
Sekercioglu,
C.H.,
Ehrlich,
P.R.,
Daily,
G.C.,
2004.
Ecosystem
consequences
of
bird
declines.
Proc.
Natl.
Acad.
Sci.
U.S.A.
101,
18042-18047.
Sekercioglu,
C.H.,
2006a.
Increasing
awareness
of
avian
ecological
function.
Trends
Ecol.
Evol.
21,
464-471.
Sekercioglu,
C.H.,
2006b.
Ecological
significance
of
bird
populations.
Handbook
of
the
Birds
of
the
World,
vol.
11.
pp.
15-51.
Sodhi,
N.S.,
Clow,
LH.,
Bazzaz,
F.A.,
2004.
Avian
extinctions
from
tropical
and
subtropical
forests.
Ann.
Rev.
Ecol.
Evol.
S
35,
323-345.
Statsoft,
Inc.,
2005.
STATISTICA
Mr
Windows
[Software-System
Mr
Datenanalyse]
Version
7.1.
http://www.statsoft.com/.
Thiollay,
J.M.,
1995.
The
role
of
traditional
agroforests
in
the
conservation
of
rain-
forest
bird
diversity
in
Sumatra.
Conserv.
Biol.
9,
335-353.
Tscharntke,
T.,
Sekercioglu,
C.H.,
Dietsch,
T.V.,
Sodhi,
N.S.,
Hoehn,
P.,
Tylianakis,
J.M.
Landscape
constraints
on
functional
diversity
of
birds
and
insects
in
tropical
agroecosystems.
Ecology,
in
press.
UNEP,
2001.
Assessment
of
the
status
of
the
world's
remaining
closed
forests.
United
Nations
Environmental
Programme,
Nairobi,
Kenya.
Waltert,
M.,
Bobo,
S.K.,
Saing,
N.M.,
Fermon,
H.,
MOhlenberg,
M.,
2005.
From
forest
to
farmland:
habitat
effects
on
Afrotropical
forest
bird
diversity.
Ecol.
Appl.
15,
1351-1366.
White,
F.,
1983.
The
Vegetation
of
Africa.
Natural
Resources
Research
20.
UNESCO,
Paris,
France.
Woltmann,
S.,
2003.
Bird
community
responses
to
disturbance
in
a
forestry
con-
cession
in
lowland
Bolivia.
Biodivers.
Conserv.
12,
1921-1936.
Wright,
SJ.,
2005.
Tropical
forests
in
a
changing
environment.
Trends
Ecol.
Evol.
20,
553-560.
Zanette,
L.,
Doyle,
P.,
Tremont,
S.M.,
2000.
Food
shortage
in
small
fragments:
evidence
from
an
area-sensitive
passerine.
Ecology
81,
1654-1666.
Zimmerman,
D.A.,
Turner,
D.A.,
Pearson,
DJ.,
1999.
Birds
of
Kenya
and
Northern
Tanzania.
Field
Guide
Edition.
Princeton
University
Press,
New
Jersey.
Zurita,
GA,
Rey,
N.,
Varela,
D.M.,
Villagra,
M.,
Bellocq,
M.I.,
2006.
Conversion
of
the
Atlantic
Forest
into
native
and
exotic
tree
plantations:
effects
on
bird
commu-
nities
from
the
local
and
regional
perspectives.
Forest
Ecol.
Manage.
235,
164-
173.