Title: Remote Sensing and Avian Biodiversity Patterns in the United States
1Remote Sensing and Avian Biodiversity Patterns
in the United States
- Volker C. Radeloff1, Anna M. Pidgeon1, Curtis H.
Flather2, Patrick Culbert1, Veronique St-Louis1,
and Murray K. Clayton3 - 1Department of Forest and Wildlife Ecology,
University of Wisconsin-Madison, Wisconsin - 2U.S.Forest Service Rocky Mountain Research
Station, Fort Collins, Colorado - 3Department of Statistics, University of
Wisconsin-Madison, Madison, Wisconsin
2Introduction
- We know generallywhat affects biodiversity
- MacArthurs big three
- Habitat loss and fragmentation
- Human threats
- History
-
The machinery controlling species diversity
after MacArthur (1972)
3Introduction
- Less clear are relative importance, interactions,
and regional variability of these factors - Basic science question
- How can we explain observed spatial patterns of
biodiversity? - Applied science needs
- Landscape level biodiversity maps
4Questions
- Can remote sensing measures of habitat structure
predict avian biodiversity patterns? - Can measures of human threats to habitat predict
biodiversity?
5Questions
- Can remote sensing measures of habitat structure
predict avian biodiversity patterns? - Can measures of human threats to habitat predict
biodiversity?
6Questions
- And how do relationshipsdiffer among
- Species Guilds, and
- Ecoregions?
7Habitat Structure
- Local measures of habitat structure (e.g.,
foliage height diversity) are among the strongest
predictors of biodiversity - More structure means more ecological niches, and
generally higher biodiversity - The challenge is to measure vegetation structure
with remote sensing
8Habitat Structure New Mexico
Mesquite Sandsage
600 m
600 m
St-Louis et al. 2006. Remote Sensing of
Environment
9Habitat Structure New Mexico
High texture Low texture
- Many texture measures available
- Can image texture capture fine- scale
habitat structure?
10Habitat Structure New Mexico
R2 0.50 p lt 0.001
11Habitat Structure New Mexico
NIR SWIR NDVI
R2 34
R2 66
R2 73
- Image texture captures fine-scale habitat
structure in semi-deserts - Good predictor of species richness
St-Louis et al. 2008. Ecography
12Habitat Structure Wisconsin
13Habitat Structure Wisconsin
3x3 Window Standard Deviation TM4
14Habitat Structure Wisconsin
R2 0.31 p lt 0.001
- Multivariate model R2 0.56
- BUT relationship is negative
15Habitat Structure Wisconsin
3x3 Window Standard Deviation TM4
16Habitat Structure
- Phenology affects texture measures
- Both problem and opportunity
17Habitat Structure
- Satellite image texture is a good measure of
fine-scale habitat structure - The scale of texture is between point
measurements and landscape indices - Texture captures structure within a land cover
class - Texture predicts avian biodiversity well
18Human Threats
- Habitat loss
- Habitat fragmentation
- Habitat modification
- These threats may interact
19Human Threats
20National Land Cover Database
21(No Transcript)
22(No Transcript)
23Human Threats
Northern Wisconsin
Madison, Wisconsin
24Human Threats
Proportion () of possible occurrences in models
Pidgeon et al 2007. Ecological Applications
25Human Threats
- Habitat loss is the major predictor
- Habitat fragmentation is important
- Habitat modification (i.e., housing development)
just as important - Threats interact multivariate models predict
best
26Conclusions
- Can remote sensing measures of habitat structure
predict avian biodiversity patterns? - YES!
- Can measures of human threats to habitat predict
biodiversity? - YES!
- Our ability to explain, and thus to predict avian
biodiversity patterns is increasing
27Thank You!