Title: Complex Interactions Shaping Aspen Dynamics in the Greater Yellowstone Ecosystem
1Complex Interactions Shaping Aspen Dynamics in
the Greater Yellowstone Ecosystem
- Brown, K., A.J. Hansen, R.E. Keane, L.J.
Graumlich. 2006. Complex interactions shaping
aspen dynamics in the Greater Yellowstone
Ecosystem. Landscape Ecology. Landscape Ecology
21933951
2Widespread Aspen Decline
- Aspen loss documented across Rocky Mountain
ecosystems - Fire exclusion
- Conifer encroachment
- Increased elk herbivory
- Climatic fluctuations
3Regional-scale Studies Dispute Decline
- Regional-scale studies in Colorado actually show
increasing/stable aspen - -Kulakowski et al. In Press
- -Kaye et al. 2003
- -Manier Laven 2002
- -Suzuki et al. 1999
- Several small-scale studies in GYE show decline
- Lack a regional perspective for aspen in the GYE
4Aspen in the Greater Yellowstone Ecosystem
- Small scale studies in areas with high elk
densities show decline - Jackson Hole valley
- YNP Northern Range
- Need for regional perspective
- 45 loss in Gravelly range
- 4,000 ha loss in YNP
- 75 loss in Centennials
- 4 loss in Northern Range
- Base-line needed to understand
- Fire
- Herbivory
- Competition
make a series of maps to bring in -plain study
area outline w/ hillshade, precip, dday, lith
5What are the Influences on Aspen?
Fire exclusion, herbivory, competition
6Objectives
- Distribution
- Does aspen occupy specific biophysical settings?
- Landscape Change
- Is aspen loss associated with particular
biophysical settings?
- Growth Rates
- How do aspen growth rates (ANPP) vary along
biophysical gradients?
7Characterizing the Biophysical Environment
Climate Regime
Weather
?Tave ?TDay ?TNight ?Tsoil ---PPT ?DDay ?Dsr ?Dss
?Srad.tg ?Srad.fg ? Ppfd
?Tmax ? Tmin ? PPT ? VPD ? Rh ? SRAD
WXFIRE
Ecosystem
Polygon
? Elevation ? Aspect ? Slope ? Soil Depth ?
Sand ? Silt ? Clay ?TopoShading ? LAI ? site
type (pft)
? AET ? PET ? PSI ? PSI.MAX ? EVAP ? TRANS ?
VMC ? GSWS ? OUTFLOW ? SNOW
8Objectives
- Distribution
- Does aspen occupy specific biophysical settings?
- Landscape Change
- Is aspen loss associated with particular
biophysical settings?
- Growth Rates
- How do aspen growth rates (ANPP) vary along
biophysical gradients?
9Aspen Distribution
- Compiled digital maps of aspen distribution
- (78 overall accuracy)
- Aspen occupy 1.4 of the mapped land area in GYE
- Aspen much more prevalent in the southern GYE
10Aspen Distribution
11Aspen Distribution
- Built a classification and regression tree (CART)
model - Used the model to predict aspen presence for the
validation datasets - Accuracy assessment of the model
12Aspen Distribution CART Analyses
13Aspen Distribution CART Analyses
- Aspen occupies a narrow range of available
biophysical settings - high radiation
- intermediate high PET
- intermediate snowfall, temperature
- Model performed well at classifying aspen where
it was present (high producers accuracy) - Model over-predicts the occurrence of aspen (low
users accuracy) - Aspen does not occur in all biophysically-favorabl
e locations
14Aspen Distribution
15Aspen Distribution
16Aspen Distribution
17Aspen Distribution
18Aspen Distribution
19Aspen Distribution -- Conclusions
- Aspen is rare in the GYE 1.4 of mapped land
area - Much more aspen in southern GYE seems to be
related to higher GS radiation - Fire?
- Aspen restricted to a narrow range of biophysical
settings - high radiation, intermediate snowfall,
temperature, PET - but not found in all suitable biophysical
settings - Other factors may restrict current range
- fire
- establishment events
20Objectives
- Distribution
- Does aspen occupy specific biophysical settings?
- Landscape Change
- Is aspen loss associated with particular
biophysical settings?
- Growth Rates
- How do aspen growth rates (ANPP) vary along
biophysical gradients?
21Landscape Change
- Air photo interpretation from early (1955 1963)
and current (1992 2001) time periods - ANOVA, regressions
22Landscape Change
23How Do Aspen Loss, Gain, Stable Plots Differ?
24Landscape Change
- Used AICc to select variables for a final
regression model
25Landscape Change
- Selected Model
- Aspen Change
- -4.93 Snowfall
- -6.13 Conifer Change
- -9.22 (Intercept)
AICc 1414 AICc Weight 0.14 R2 0.18
26Where is Aspen Declining?
- Moisture
- Low summer precipitation and high winter
precip./snowfall - Temperature
- Cooler, shorter growing season
- Light
- Higher radiation
- Water Availability/Flow
- More arid sites with higher evapotranspiration,
run-off, and lower water retention -
27Landscape Change - Conclusions
- 34 of plots declined in aspen cover
- Averaged 10 decline overall
- Areas that gain aspen cover warmer and wetter
during the growing season than loss areas - Loss areas drier during the growing season but
higher annual precip. and snowfall - Lower radiation levels associated with aspen gain
- Other factors?
- insects/disease outbreaks defoliation
- fire suppression mature stems senesce w/ no
replacement
28Objectives
- Distribution
- Does aspen occupy specific biophysical settings?
- Landscape Change
- Is aspen loss associated with particular
biophysical settings?
- Growth Rates
- How do aspen growth rates (ANPP) vary along
biophysical gradients?
29Aspen Growth Rates
- Used aspen ANPP as an index of aspen performance
- increment cores for growth rates
- density and size of aspen to estimate biomass
- Density and size class data for trees and shrubs
used to estimate biomass of competing vegetation - Soil characteristics
- soil moisture
- NH4
- NO3-
30Aspen Growth Rates
- Used AICc to select variables for the final
regression model
31Aspen Growth Rates
Model selection results for ANPPN 107 sites
- Selected Model
-
- Aspen ANPP 2255 GS Precip. X Ann. Min.
Temp - 1902 Clay
- 1829 Annual Minimum Temperature
- -970 Conifer Biomass
- 378 GS Precipitation
- 4536 (Intercept)
- AICc 2130.37
- AICc Weight 0.71
-
- R2 0.37
-
32Growth is not Maximized in Current Distribution
33Growth is not Maximized in Current Distribution
34Aspen Growth Rates - Conclusions
- Growth rates are fastest where summer precip. is
high, temperatures are warmer, and on clay soils.
High conifer biomass is negatively associated
with aspen growth. - High ANPP at the tail edge of aspens
distribution - Competitive exclusion from favorable sites
- Possibly more susceptible to herbivory
- 37 of the variance explained with biophysical,
other influences - Competition
- Defoliation - insects
35Conclusions
36Conclusions
- Aspen is rare in the GYE, 1.4 of land area
- Aspen does not occupy all biophysically favorable
settings - Decline is less widespread than local-scale
studies suggest. - Loss of aspen cover appears to be localized to
particular environmental condiditons. - Possible constriction of aspen distribution
- fire
- herbivory
- competition
- Aspen may not currently occupy full range of
abiotic tolerances and performance may not be
optimized in current locations - Aspen in the GYE likely more susceptible to
herbivory, competition, fire exclusion as a
result
37Acknowledgements
- Committee
- Andrew Hansen
- Lisa Graumlich
- Robert Keane
- National Park Service
- Roy Renkin
- USGS-BRD
- Don Despain
- Bridger-Teton National Forest
- Tim Kaminski
- Field crew/Technicians
- Steve Wilcox
- Chris Brown
- Casey Sensesqua
- Funding
- Fire Sciences Lab, US Forest Service
- National Park Service
- University of Wyoming-NPS Co-op Unit
- Bridger-Teton National Forest