Transcript No Slide Title

Forest birds in landscape mosaics:
theory and empirical evidence
Cristián F. Estades
Theory of Island Biogeography
(MacArthur & Wilson 1963, 1967)
extinction
rate
colonization
Importance of:
Island size
Isolation
N*
Number of species
Metapopulations
(Levins 1969, 1970
Gilpin & Hanski 1991)
dp
= mp(1-p) - ep
dt
pˆ = 1 - e/m
Ideal Free Distribution
Suitability
(Fretwell and Lucas 1970)
G
P
Number of competitors
Density dependent
habitat selection
Source - sink dynamics
(Pulliam 1988)
Sources (R > M, E > I)
subsidize
Sinks (R<M, E < I)
Birds in habitat islands - agricultural landscapes
Birds in Illinois
forest fragments
(Blake & Karr 1984)
Species richness
30
20
10
p < 0.001
1
10
100
500
Forest fragment area (ha)
1.0
Probability
.
.
.. .
.
.
.. .
40
Incidence function
of Red-eyed Vireo
(Robbins et al. 1989)
0.8
0.6
0.4
0.2
0.0
0.3
3.2
32
320 3200
Forest fragment area (ha)
Bird species richness (spp/plot)
Birds in forest fragments in Southern Chile
20
18
16
14
12
10
8
6
4
2
0
20
18
16
14
12
10
8
6
4
p = 0.001
2
0
0.1
1
10
100
Forest fragment area (ha)
Matrix: agricultural fields
(Willson et al. 1994)
1000
p = 0.07
0.1
1
10
100
Forest fragment area (ha)
Matrix: pine plantations
(Estades & Temple 1999)
1000
Extended foraging area
d
d
d
d
d: maximum distance of foraging trip
Mosaic approach
(Wiens 1995)
General theoretical framework (but is it really workable?)
Landscape Mosaics are too Idiosyncratic
Mosaic approach
Habitat mosaic
Quality
Suitability
Fitness
etc...
Resource mosaic
Cover
Nest sites
Food
1. The effect of breeding-habitat patch size
on bird population density
Correlation between breeding-habitat
patch size and bird population density
A
B
0
-
C
+
A
?
B
B
B
?
C
0
Spatial covariance between food and nest sites
+
Individual based model:
Goal: Explore the effect of relative distribution of food and
nest sites on the relationship between breeding habitat patch
size and bird density.
80 x 80
Breeding
habitat
Density
Simulation model: results
.
.
.
...
Density
Log (area)
Log (area)
Effect of flight distance
General model
Relationship between breeding-habitat
patch size and bird population density
+
+ Edge effects
+ Negative microclimate
effects
0
+ Competitor or
predator release
-
0
Spatial covariance between food and nest sites
+
Conclusions
The effect of breeding-habitat patch size on bird population
density depends on the relative location of food and nesting
sites in the landscape. Management of the matrix may
affect the suitability of fragmented landscapes for breeding
birds
The effect of additional foraging resources in the matrix on
birds in fragmented habitat-patches depends on the species’
maximum flight distance and perceived foraging risk.
2. Spatial dynamics of bird communities
in a forest landscape mosaic
Problem
The distributions of most forest bird species in the
fragmented landscape of the Maule region are not
adequately predicted by the size and/or isolation of forest
patches (Estades and Temple, 1999).
Hypotheses
The distribution of bird species in a landscape can be
modeled by looking at the distribution of food and nest sites.
The distribution of birds in the landscape changes over time
in accordance to the distribution of the limiting resources.
During the breeding season the best predictor of the
distribution of birds is the product of the local abundance of
nest sites and food resources and during the non breeding
season the best predictor is the abundance of food
resources.
Study area
Maule Region, Chile
Pine plantations 80%
Native forests
10%
Open areas
7%
Others
3%
10,000 ha
Dominant species: Nothofagus dombeyi
Dominant species: Nothofagus glauca
Dominant species: Pinus radiata
Studied species
Tufted Tit-tyrant
Open-cup nester
Understory
Des Murs’Wiretail
Open-cup nester
Understory
White-crested Elaenia
Open-cup nester
Understory and canopy
Thorn-tailed Rayadito
Cavity nester
Fire-eyed Diucon
Open-cup nester
Understory and canopy
White throated Treerunner
Cavity nester
Sampling
120 point count stations
(variable radius point counts with correction for detectability)
Field seasons:
Winter (June) 1999
Spring (Oct-Nov) 1999
Summer (February) 2000
Winter (June) 2000
Spring (Oct-Nov) 2000
Summer (February) 2001
Model
Breeding season
Non-breeding
Bird abundance = f(Nest sites * Food abundance)
Bird abundance = f(Food abundance)
Generalized linear model, negative binomial distribution
Nest sites
Food abundance
.
. . .. . . . . ..
. .....
.
.
..
.
. .
0
50
100
200m
Diet
Analysis of droppings
Food abundance
Foliage shaking
Food abundance
Aristotelia chilensis
Fruit sampling
Nest sites
Abundance of cavities
Density of understory
Effect of nest site abundance and arthropod biomass
on the abundance of cavity nesters
Species and Season
Thorn-tailed
Rayadito
Winter 1999
Spring 1999
Summer 2000
Winter 2000
Spring 2000
Summer 2001
White-throated
Treerunner
Winter 1999
Spring 1999
Summer 2000
Winter 2000
Spring 2000
Summer 2001
Cavities
0-50 m 0-100m 0-200m
Arthropods
0-50 m 0-100m 0-200m
Cavities * Arthropods
0-50 m 0-100m 0-200m
+
.
+++
+++
+++
+++
++
+++
+++
+++
+++
+++
.
+++
+++
+++
+++
+++
++
+
+
+
+
+
.
+
+
+++
.
+++
++
+++
+++
+++
+++
++
+
. : p<0.1, +: p<0.05, ++: p<0.01, +++: p<0.001. Bold: lowest AIC.
+++
+++
+++
+++
+++
++
+++
+++
++
+++
+++
.
+++
+++
+++
+++
+++
+++
++
+
+
+
.
+++
+
+++
++
+++
+++
Effect of nest site abundance and arthropod biomass
on the abundance of open-cup nesters
Species and Season
Tufted Tit-tyrant
Winter 1999
Spring 1999
Summer 2000
Winter 2000
Spring 2000
Summer 2001
Understory
0-50 m 0-100m 0-200m
+
Des Murs' Wiretail
Winter 1999
Spring 1999
.
Summer 2000
.
Winter 2000
Spring 2000
+++
Summer 2001
++
+
+++
+
+++
Arthropods
0-50 m 0-100m 0-200m
Understory * Arthropods
0-50 m 0-100m 0-200m
++
+
.
.
+
+++
. : p<0.1, +: p<0.05, ++: p<0.01, +++: p<0.001. Bold: lowest AIC.
+++
+++
+
+++
Effect of food abundance on the abundance of White-crested Elaenias
Season
Spring 1999
Summer 2000
Spring 2000
Summer 2001
Arthropods
0-50 m 0-100m 0-200m
+
+
Fruits
0-50 m 0-100m 0-200m
Fruits * Arthropods
0-50 m 0-100m 0-200m
+
+
+
. : p<0.1, +: p<0.05, ++: p<0.01, +++: p<0.001. Bold: lowest AIC.
Conclusions
Most studied birds changed their distributions in the
landscape between seasons and many of these changes
matched the changes in the distribution of key resources.
Although factors such as competition, predation and
parasitism may influence the distribution of birds in a
landscape mosaic, the “resource mosaic”model represents a
parsimonious approach to understanding the distribution of
birds in a patchy landscape where the matrix surrounding
patches of preferred habitat is not completely useless
3. Nest success of the Thorn-tailed Rayadito
in a forest landscape mosaic
Problem
Thorn-tailed Rayaditos in the Maule region have higher
densities in smaller forest fragments because these birds
can forage in the surrounding pine plantations (Estades and
Temple, 1999). But is the breeding success of birds that
include pine plantations in their home ranges equal to the
breeding success of birds that forage entirely in native
forest?
Hypothesis
The breeding success of Thorn-tailed Rayaditos is
positively associated with the amount of native vegetation
around the nest because of the effect of reduced food
concentration and a potentially higher level of predation in
the artificial forests
Upland forest
native vegetation:
100%
Riparian forest
Pine plantations
25.5%
0%
50m
1.34
1.13
1.0 relative arthropod ab.
Nest success of Thorn-tailed Rayaditos
Most failures
were due to
predation
No evidence of
an effect of food
density
Thylamis elegans
House Wren
Rodent
Nest predators
Thylamis elegans
Phylodrias chammisonis
Predators of fledglings and adults
Accipiter chilensis
Glaucidium nanum
Competitors
Picoides lignarius
Troglodytes aedon
Liolaemus tenuis
Pygarrichas albogularis
Conclusions
Concentration of food resources did not limit breeding
performance of Thorn-tailed Rayaditos
Small riparian forest fragments surrounded by pine
plantations not only harbor higher rayadito densities than
large upland forest patches (Estades and Temple, 1999),
but they also provide a safer place to breed for this species.
Differences in breeding success between habitat types
were probably due to differences in concentration of
predators and competitors and average nest height.
Ongoing work
Nest success and nesting
density of open-cup nesters
in different parts of the
landscape
Movements of individual
birds in the landscape
Acknowledgements
•Fondecyt (Chile) grants 1990786 and 7990027
•Zoological Society of Milwaukee County
•Association of Field Ornithologists
•Dept. Wildlife Ecology. University of Wisconsin - Madison
•School of Forest Sciences. Universidad de Chile (Santiago)
•Committee: Stanley Temple, Nancy Mathews, Christine Ribic,
Tony Ives, Tim Moermond
•All the field assistants
•Friends here and there
•Paula and Josefa