Transcript Habitat Variability of Anolis Lizards in the Caribbean and

Habitat Variability of Anolis Lizards in the Caribbean
and the Spatial and Ecological Relationships of
Anolis cristatellus on Puerto Rico
A. cristatellus
David Ullman
May 6, 2004
ENVE 424
The Wonderful Anolis




Over 200 species of Anolis lizards in United
States, Mexico, Central and South America, and
the Caribbean
124 known species in the Caribbean alone
Genus well known
Ideal group of species for evolution studies
- large amounts of data
- island species
- limited gene flow
What is an ecomorph?

Definition of an ecomorph:
“species with the same structural
habitat/niche, similar in morphology and
behavior, but not necessarily closes
phyletically.” (Williams, 1972)
 Microhabitat has profound
impact on the morphology of
Anolis
 What is the effect of large scale
habitat differences on species
diversity and morphology?
Picture taken from: Williams, E.E. 1983. Ecomorphs, faunas, island size,
and diverse end points in island radiations of Anolis. In: Lizard Ecology:
Studies of a Model Organism (R.B. Huey, E.R. Pianka, and T.W. Schoener,
eds), pp. 326-370. Harvard University Press, Cambridge, USA.
Part I: Habitat Variability and Species Diversity

Data collected to measure habitat variability in Land cover/vegetation,
surface temperature, annual precipitation, and elevation
 elevation, mean annual precipitation, and mean annual temperature for
these analyses was obtained from the WorldClim database at the
University of California (30 sec. Resolution, ESRI format)
http://biogeo.berkeley.edu/worldclim/worldclim.htm
 Land cover/vegetation data has been obtained from the Global
Vegetation Monitoring Unit (1 km2 resolution, ESRI format)
www.gvm.jrc.it/glc2000/ProductGLC2000.htm).
Measuring Habitat Variability (method)






Habitat Data added to ArcMap
Masks created to outline each of the islands in the Caribbean
“Raster Calculator” used to cut out temperature, precipitation, and
elevation data for each island
This data used to calculate standard deviation as a measure of
variability for each habitat data on each island
For land cover/vegetation variability, number of vegetation types counted
for each island
Each habitat variability measurement plotted against the log of the
number of species on each island as a measure of species diversity (log
transform to normalize data for parametric statistics).
Habitat Variability and Species Diversity (Results)
Habitat variability does have an
affect on species diversity

Land cover highly correlated with
species diversity (t = 6.934, P =
.00006), see right.

Elevation moderately correlated
with species diversity (t = 2.772,
P = .022)

Temperature moderately
correlated with species diversity
(t = 3.001, P = .015)

Precipitation NOT correlated with
species diversity (t = 1.153, P =
.279)
Land Cover Variability vs Species Diversity
Log (number of anolis
species)

2
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0
5
10
15
20
Number of Land Cover types
25
Part II: Morphological variability in A. cristatellus

Previous research shows
importance of microhabitat
variability on morphology

Habitat variability is
important in species
diversity

Look at specific species on
one island to see if broad
intra-island habitat
variability has an effect on
morphology
Anolis cristatellus on
Puerto Rico

Morphological variability in A. cristatellus (methods)








Morphological data from 448 museum specimens
Geographic location assigned to each of the 448 specimens
based on nominal data
Temperature, Elevation, and Precipitation data recorded for each
location
Female specimens filtered out due to sexual dimorphism
Filtering out specimens in same locations by averaging data
Effect of body size removed
15 morphological measurements?  Principal Component
Analysis (PCA)  condense to 3 principal components. These 3
principal components account for 82.2 % of the variance in the
data
Each principal component plotted against temperature, elevation,
and precipitation
Spatial relationships in Morphology of A. cristatellus
(Methods)






Moran’s I calculation of spatial autocorrelation
Using Rooks Case v0.9.6 (Mike Sawada,
University of Ottawa, 1998)
Irregular lattice
20 lags
10,000 m (10 km) lag distance
Correlogram generated
Morphological variability in A. cristatellus
(results)
 No correlation between morphology and any of the habitat conditions
 Temperature (correlations):
PCA1 (R2 = .0166, P > .05)
PCA2 (R2 = .0332, P > .05)
PCA3 (R2 = .0007, P > .05)
 Precipitation:
PCA1 (R2 = .0004, P > .05)
PCA2 (R2 = .008, P > .05)
PCA3 (R2 = .0012, P > .05)
 Elevation
PCA1 (R2 = .0222, P > .05)
PCA2 (R2 = .0178, P > .05)
PCA3 (R2 = .0037, P > .05)
Spatial Autocorrelation?

Moran’s I correlograms do not show spatial autocorrelation:
Moran's I PCA 1
Moran's I PCA 3
Moran's I PCA2
0.6
1.2
1
1
0.4
0.8
0.5
0.2
50000
100000
150000
-0.4
200000
250000
0
0
50000
100000
150000
-0.5
200000
250000
Moran's I
0
-0.2
Moran's I
Moran's I
0.6
0
0.4
0.2
0
-0.2
-0.6
0
50000
100000
150000
-0.4
-1
-0.8
-0.6
-1
-0.8
-1.5
Lag (m)
Lag (m)
Lag (m)
200000
250000
Kriging PCA1
Kriging PCA2
Kriging PCA3
Conclusions




Habitat variability influences species diversity
Habitat variability has no effect on morphology of A.
cristatellus
No spatial relationship in morphology
Future work:
More sampling of A. cristatellus
 Apply analyses to other species on Puerto Rico and
other islands
 Factor in temporal scale to reflect changes in
morphology or climate over time
