Community and Evolutionary Consequences of
Record Drought in the Southwest
Gery Allan – molecular systematics
Randy Bangert – biogeography
Samatha Chapman – nutrient cycling
Neil Cobb – insect ecology
Steve DiFazio – molecular ecology
Dylan Fischer – ecophysiology
Robert Foottit – molecular systematics
Alicyn Gitlin – restoration ecology
Steve Hart – ecosystem/soil ecology
Paul Heinrich – public outreach
Paul Keim – microbial/bacteria genetics
Karla Kennedy – resotration ecology
Zsuzsi Kovacs – mycorrhizal ecology
Rick Lindroth – chemical ecology
Jane Marks – aquatic ecology
Nashelly Meneses – ecological genetics
Brad Potts – population genetics
Jen Schweitzer – ecosystems
Steve Shuster – theoretical genetics
Chris Stulz – population ecology
Randy Swaty – conservation biology
Amy Whipple – ecological genetics
Gina Wimp – community ecology
Scott Woolbright - molecular genetics
Joe Bailey – community ecology
Brad Blake – restoration biology
Aimee Classen – nutrient cycling
Ron Deckert – endophyte ecology
Eck Doerry – bio-informatics
Kevin Floate – insect ecology
Catherine Gehring – microbial ecology
Allen Haden – aquatic ecology
Kristin Haskins – mycorrhizal ecology
Barbara Honchak – ecological genetics
Art Keith – insect community ecology
George Koch – ecophysiology
Carri LeRoy – aquatic ecology
Eric Lonsdorf – genetic modeling
Greg Martinsen – ecological genetics
Becky Mueller – plant ecology
Brian Rehill – chemical ecology
Crescent Scudder – plant demography
Adrian Stone – insect communities
Richard Turek – statistics
Talbot Trotter – dendrochronology
Tom Whitham – ecology
Stuart Wooley – phytochemistry
Tong-Ming Yin – quantitative genetics
Key Issues
1. Climate change alters community structure and
negatively affects biodiversity.
2. Drought impacts on dominant and
keystone species are most important
because they are community drivers.
3. Some vegetation types are far more
likely to be affected than others.
4. Drought is likely to be an evolutionary event.
5. There are likely to be major surprises with
drought events that cannot be predicted.
2200 years of climate history argues major switches
in selection pressures from wet to dry periods that
could affect biodiversity and rapid evolution.
Grissino-Meyer 1996
Expansion of junipers (Juniperus monosperma)
into grasslands
1899
1977
Enchanted Mesa, New Mexico
Allen et al. 1998
Pinyon mortality
North side of the San Francisco Peaks, AZ
Photo mosaic by Paul Heinrich taken December 2003
Photo by Tom Whitham
Sept 15, 2004 - North of
San Francisco Peaks, AZ
Photo by Tom Whitham
Sept 15, 2004 - North of
San Francisco Peaks, AZ
2000
75
UT
km2
2003
CO
2001
332 km2
2002
1,377 km2
km
2003
10,406 km2
Total
12,191 km2
AZ
NM
Pinyon + Ponderosa Pine Mortality - Compiled by the National Forest Health Monitoring Program
Slope Function
Aspect Function
Elevation Function
= Low Risk
= Moderate Risk
= High Risk
= Known High
Pinyon Mortality
Sites
Model correctly intersects
known high pinyon mortality
sites with 75% accuracy.
~10% of error is from low
resolution DEM
~15% of error is from
undocumented high pinyon
mortality sites + unknown error.
Trotter et al. 2005
Lightning strikes twice in the same spot.
Fig. 1
Stand-level % Pinyon Mortality in 2002
80
r2 = 0.504
F = 9.15
p = 0.01
60
40
20
0
0
20
40
60
Stand-level % Pinyon Mortality in 1996
Mueller et al. 2005
80
Larger pinyons are far more likely to die
than smaller ones.
Fig. 2
120
100
r2 = 0.712
F = 69.4
p < 0.001
% Mortality
80
60
40
20
0
0
10
20
Size Class (1.0 cm)
Mueller et al. 2005
30
Mean Arthropod Richness / Tree
On Trees Growing Under Low Stress
14
B
High Stress
Low Stress
B
12
10
8
A
A
B
6
4
A
2
Mean Arthropod Abundance / Tree
0 1B: Overall Mean Arthopod Abundance Is Highest
Figure
Predators
Parasites
OnHerbivores
Trees Growing
Under Low
Stress
High Stress
Low Stress
600
B
40
20
A
In an arthropod
community of
266 species,
species richness
is 2 X higher and
abundance is
12 X higher in
low stress sites
than high stress
sites.
B
A
A
B
Trotter et al. 2005
0
Herbivores Predators
Parasites
Figure 2: Canopy Arthropod Community Composition Differs
Significantly
Between Trees
Growing InCommunities
High and Low
Stress Changes
Arthropod
Stress Environments
0.6
High Stress
Sites
0.4
Y Axis
0.2
High Stress 1
High Stress 2
High Stress 3
Low Stress 1
Low Stress 2
Low Stress 3
0.0
-0.2
-0.4
-0.6
-0.6
Low Stress
Sites
-0.4
-0.2
0.0
X Axis
0.2
0.4
0.6
Trotter et al. 2005
Watering
experiment
conducted by
Crescent Scudder
to test for
community
release.
Photo taken May 23, 2003
Mean Athropod Abundance on
Arthropod
Non-watered
vs Abundance
Watered Pinyons
40
p=.028
30
Watered Pinyons
Non-watered Pinyons
20
10
0
Watered Control
Mean Arthropod Richness on
Non-watered vs Watered Pinyons
20
Species Richness
p=.031
Supplemental
watering
increases
arthropod
abundance
and species
richness.
15
10
Scudder unpub. data
5
0
Watered Control
Watered and control trees support different arthropod
communities (Scudder et al. 2005).
2003 Community Composition of Watered and Control Trees
Axis 2
Watered Trees
Control Trees
Watered
Control
Axis 1
Ectomycorrhizae on fine root of pinyon pine
Environmental Stress Shifts Ectomycorrhizal Community
O'Neill Crater RFLPs
R = 0.652
P < 0.00001
Low Mortality
Sites
High Mortality
Sites
Swaty et al. 2004 Ecology
Low Mortal
High Morta
Tree Rings Predict Arthropod Species Richness
30% Richness vs. Ave 97-01
40
Species
Richness
per Tree
Y Data
30
20
10
0
0
500
1000
1500
2000
2500
X Data
Average Tree Ring Width 1997-2001
Ave 97-01 vs 30% Richness
Plot 1 Regr
Adrian Stone unpub. data
3000
Moth Resistant Pinyon
These phenotypes
are genetically based
and have extended
phenotypes that
have community
consequences.
Susceptible Pinyon
Photo by Tom Whitham
Switch in performance - Moth resistant trees
3X more likely to die during 2002 drought than moth
susceptible trees.
80
% mortality
70
60
50
40
30
20
10
0
Stulz et al. unpub. data
Moth
Resistant
Pinyons
Moth
Susceptible
Pinyons
Scales associated with reversals
•
•
•
•
Number of species
Time
Spatial scales
Any mix of above
Interactions Increase With the Addition of Factors
% of Significant Interaction Terms
100
Observed
Total Observed Significant Interactions
Observed
ObservedReversals
Reversals
80
60
40
20
0
0
1
2
3
4
Number of Statistical Factors (Species, Time, Space)
# of Statistical Factors
Bailey & Whitham 2005
The extended phenotypes of moth resistant and susceptible
trees affect a diverse community of about 1000 species.
Whitham et al. 2003 Ecology
Drought impacts most
dominant plants and their
dependent communities.
Photos by Tom Whitham & Alicyn Gitlin
Trees like
cottonwoods should be
of special concern due
to low coverage.
Gitlin et al. unpub. data
Landscape coverage (%) Population mortality (%)
Mortality of
dominant plants
at 20 randomly
selected sites for
each species
within a 80km
radius of Flagstaff.
60
50
P < 0.0001
a
40
30
b
b
20
10
b
b
c
b
60
0
50
40
30
20
10
0
(0.16)
Juni. Manz. Asp. Pond. Cotton. Piny.
Dominant plant species
Summary
1. Through its effects on community drivers (e.g.,
dominant trees and keystone herbivores),
drought alters community structure and
negatively affects biodiversity.
2. Some dominant vegetation types are far more
likely to be affected than others.
3. Switches or unexpected outcomes are likely (e.g.,
insect resistant trees are more likely to die,
which in turn results in a major community
shift).
4. Extreme drought is a bottleneck event that is also
likely to be an evolutionary event.
Management & Research Issues
1. In the absence of long-term community-level studies, fundamental
errors in interpretation are likely due to the high probability of
switches.
2. Need to minimize human impacts that exacerbate the effects of
drought (e.g., water diversions that make a 100 year drought into
a millennium-level drought).
3. Need special emphasis on rare habitat types that are especially
sensitive to drought (e.g., riparian habitat and springs).
4. Marginal or edge habitats that suffer chronic stress can be
barometers of change and may be crucial to preserve as sources of
extreme genotypes that may be best adapted to changing
environments.
5. Restoration with local genotypes may not best strategy if climate
change continues (e.g., drought adapted genotypes from lower
elevations should be included in a diverse genetic seeding
program).
6. Need to better interface climate change models with community
change models.
DIREnet (Drought Impacts on Regional Ecosystems Network) coordination and synthesis of ecological research on drought
effects and the potential role of global climate change.
http://www.mpcer.nau.edu/direnet/