Transcript Stephen Matthews(6 MB, Updated: Dec

REGIONAL IMPACTS OF CLIMATE CHANGE ON
FORESTS AND BIRD COMMUNITIES
Stephen Matthews1,2, Louis Iverson2,
Anantha Prasad2, Matthew Peters2
1
School of Environment and Natural Resources
Ohio State University
2USFS Northern Research Station
Today's topic




Climate is changing and species are responding
Conservation and management strategies that
ignore these change will likely fail (Lawler et al. 2009)
How can we begin to integrate climate change
impacts into complicated realm of conservation
Must address in stages and build our understanding
Building broad-scale models is a logical start
 Working with managers to translate results to “on the
ground realities”

Land temp
Eleven indicators
that the climate
is changing
Sea temp
Sea level
Snow cover
State of the climate 2009, NOAA
Vertebrates: phenological responses to
a changing climate
• Amphibians
• Calling 10 -13 days
earlier for half of species
from ~1912 to 1990
Gibbs and Breisch 2001
• Birds
• Early arrival
Beaumont et al. 2006
• Hatching earlier
Both and Visser 2005, GCB, Fig 4
Such shifts can lead to
mismatches in timing
• Reviewed in Root et al. 2005
Wood frog = 13 days
Wildlife distribution shifts linked to
climate change

Distributional shifts

Mammals:
 Southern
species increase
and northern species
decline, southern flying
squirrel N 225 km since
1880 (Myers et al. 2009, GCB, Fig 3)

Birds:
 winter
range northern
boundary shift 1.48 km/yr
Sorte and Thompson 2007
 breeding
range also show
northward trend 2.35 km/yr
Hitch and Leberg, 2006
Future effects:
Highly tied to the level of CO2 emissions
??
Rising Temperatures in NE. US
12
o
te m p e ra tu re c h a n g e ( F )
(annual average)
10
8
Higher: 6.5-12.5oF
o b s e rva tio n s
h ig h e r e m is s io n s
lo w e r e m is s io n s
Lower:3.5-6.5oF
6
Large difference
between High and Low
emissions!
4
2
0
-2
-4
1900
1950
2000
2050
2100
Expected growing season changes
(for northern Wisconsin)
HAD Hi
+8 C (14F)
PCM Lo
+2 C
Growing season temperature higher and not much change
in precipitation = more physiological stress on biota
Challenges of modeling species
impacts of climate change

Future climate uncertainty


GCM variations
Human-produced levels of CO2 uncertain
Bottom line: we need to incorporate different
 Species likely to respond individually
approaches to quantify and “whittle away” at
 Biology not that well-known for many species
uncertainties
to develop ecologically informed
projections
Model validation
far into
nottopossible
– modeling
is afuture
key tool
do this
DISTRIB model
Species habitat prediction
• Tree
Data
abundance
• Bird abundance
• Climate
• Environment
• Forest density
• Species traits
SHIFT model
Species colonization
probabilities
ModFacs
• Biological factors
• Disturbance factors
• Model uncertainty
Iverson et al. 2011,
Ecosystems
Tree &
Bird
Atlases
DISTRIB
+
SHIFT
Potential
migration
by 2100
DECISION
SUPPORT
FRAMEWORK
Scoring system
for tree species
Multi-stage modelling scheme
Current and
future species
management
• Management
guidelines
• Implications & tools
The role of climate in shaping
vertebrate distributions
Root 1988
Currie 1991
But there is also a strong habitat
component for most species
Therefore we use climate and individual tree
species to build our 147 bird models
Important to note when interpreting
these models!

The models are predicting potential
suitable habitat by year 2100 – not
where the species will be.

The DISTRIB model does not
account for biotic interactions, other
human or natural disturbances.
Climate Change
Tree & Bird Atlas
http://www.nrs.fs.fed.us/atlas
Examples of species with projected habitat increases
Prothonotary
Warbler
Low
Emissions scenarios
High
?
Brown-headed
Nuthatch
http://www.nrs.fs.fed.us/atlas
Examples of species with projected habitat decreases
Black-throated
Blue Warbler
Low
Emissions scenarios
High
?
Black-capped
Chickadee
http://www.nrs.fs.fed.us/atlas
Species
General trends of all 147 species
across the eastern US
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Incidence
change
(Ratio)
>2
1.2 - 2
Hadhi
AVGhi
AVGlo
PCMlo
0.9 - 1.1
0.5 - 0.9
< 0.5
Mean Center Potential
Movement
Km (sd)
PCMlo 109 (64.3)
Avglo 142 (88.9)
Avghi 210 (139.5)
Matthews et al. 2011 Ecography
Hadhi 212 (149.9)
The Chicago ecosystem assessment provides one example of
these data being summarized regionally to focus the results
(Hellmann et al. 2010, J. Great Lakes Res.)
Habitat Changes:
Under high emissions
• 130 species
changes >=10%
• With 76 declining
• With 54 increasing
Under low emission
• 116 species
changes >= 10%
• 64 declining
• 52 increasing
Summer tanager
Are these
data being
used??
The Goal: Identify
strategies and
approaches to climate
change adaptation and
mitigation
Bridge the gap between
scales of prediction
 management activities
on National Forests
 interactions with the
greater community

Swanston et al. 2011
Sugar Maple
Low
?
High
Modifying factors




Many other factors (biological and disturbance)
come in to play to determine more likely outcomes
We rate biological (n=9) and disturbance (n=12)
characteristics for positive or negative impacts
Goal was to evaluate more realistic outcomes at
regional and local levels
The results from the multi-criteria framework can
be applied to the results present today
Red Maple:
• Projected habitat
declines
• Characteristics suggest
high adaptability
Black Oak:
• Projected habitat
increases
• Positive ModFac profile
suggests it may be
able to persist in harsh
areas
Balsam Fir:
• Projected habitat
declines
• Negative ModFac
• All metrics suggest it
will likely face severe
limits in eastern US
Matthews et al. 2011, For. Ecol. Manag.
Potential Changes for Tree Species
The “Model Forest” Project


Evaluated 73 species from the region
Put in to 8 classes of impacts









Class 1: extirpated (1 species)
Class 2: large decrease (12 species)
Class 3: small decrease (6 species)
Class 4: no change (6 species)
Class 5: small increase (4 species)
Class 6: large increase (17 species)
Class 7: new entry-high and low emissions (11 species)
Class 8: new entry-high emissions (16 species)
Score each species for modification factors to help
managers interpret potential impacts and suggest
adaptation strategies
Class 1
Class 2
Class 3
(Losers)
Class 4
(Stayers)
Class 6
Class 5
(Gainers)
(New Migrants)
Class 8
Class 7
Large Decreasers (Cl. 2)
Better
Worse
Overall habitat change for the 7 major
species groups in N Wisconsin by
GCM/emission
Current
Habitat
species
habitat total
balsam fir
4.8
aspen
16.8
paper birch
6
jack pine
3.1
white and red pine
4.7
all oak
10.5
northern hardwood
17
% Change in Habitat
PCMlo
(mild)
-3.6
-8.2
-3.3
-0.8
0
6.8
-3.1
Hadhi
(harsh)
-4.4
-14.3
-5.4
-1.5
-2.9
11.3
-10
Ecosystem Vulnerabilities
The potential changes in species composition may thus lead to
a number of ecosystem vulnerabilities:
 Lowland hardwood forests (presently dominated by black
ash) will be disrupted from drying and especially the emerald
ash borer, and probably converted to red maple
 Lowland conifer forests (e.g., balsam fir), may be stressed
more by dry late summers, disrupting that entire ecosystem
 Several ecosystems with species that have been recently
declining (e.g., hemlock, paper birch, white spruce) will likely
continue to decline
 Vegetation changes will have significant effects on wildlife
Swanston et al. 2011
A few final thoughts…

Modeling potential responses of 134 tree and 147 birds
species using multi-stage methodology


Continues to provide new knowledge of species distributions and
potential for change
Carry the research results forward to ensure that it is
management relevant and assumptions are communicated



What species may be players: lists of species to evaluate?
How might species get there?
Do the species life history characteristics compliment or
contradict the habitat changes?

Must enter into an adaptive management framework moving
forward: this includes monitoring to provide feedback loops.

We are confronted with the challenge of understanding the
response of ecological systems to changing landscapes.
More than ever we need to consider how local management
decisions fit into the distributions of target species.
Thank you!
Web site for most data
presented today:




Climate change atlases
Species-environment data for
147 birds and 134 trees
Pdfs of related papers
www.nrs.fs.fed.us/atlas

Acknowledgements
•
•
•
Thanks to USDA FS Northern Global Change Program for support
US Forest Service Northern Research Station
Ohio State University