OSU AED Policy & Outlook Fall 2007
Adapting Forests and Ecosystems to Climate Change
Brent SohngenDepartment of Agricultural, Environmental &
Development Economics,University Fellow, [email protected]
Outline of Presentation
• Key forest ecosystem impacts: General ecological results from IPCC and recent IUFRO assessment of impacts on forested ecosystems.
• Literature on economic impacts and adaptation
• Recent assessment of climate change impacts on forests and timber markets
OSU AED Policy & Outlook Fall 2007
Key Ecosystem Impacts• Productivity changes (IPCC, 2007)
– CO2 fertilization.– Warming in colder climates.– Precipitation gains where water is limited.
• Some current evidence that historical climate change and CO2 change have increased productivity to date (Boisvenue and Running, 2006).
• Potential limits to productivity gains: Net impacts– Species composition, age structure, seasonal and daily precipitation and temperature patterns, etc.
– Drying and forest fire effects
Key Ecosystem Impacts• Net effects ultimately weigh down gains: Ecosystems turn from carbon sink to source within the next several decades.
IPCC (2007) WG 2, Chapter 4, Figure 4.2IPCC (2007) WG 2, Chapter 4, Figure 4.2
OSU AED Policy & Outlook Fall 2007
Key Ecosystem Impacts
• Appreciable changes in ecosystems from 2000 to 2100.
• Green‐Blue: +
• Red‐Yellow: ‐
A2A2
B1B1
IPCC (2007), WG 2, Ch. 4, Figure 4.3IPCC (2007), WG 2, Ch. 4, Figure 4.3
Summary: Timber market results to date
Region Output Producer Returns2000–2050 2050–2100
North America -4% to +10% +12 to +16% DecreasesEurope -4% to +5% +2 to +13% DecreasesRussia +2 to +6% +7 to +18% DecreasesSouth America +10 to +20% +20 to +50% IncreasesAus./New Zealand -3 to +12% -10 to +30% Decr. & Incr.Africa +5 to +14% +17 to +31% IncreasesChina +10 to +11% +26 to +29% IncreasesSE Asia +4 to +10% +14 to +30% Increases
Alig et al. (2002), Irland et al. (2001), Joyce et al. (1995, 2001), Perez-Garcia et al. (1997, 2002), Sohngen et al. (2001), Sohngen and Mendelsohn (1998, 1999), Sohngen and Sedjo (2005); 2 Karjalainen et al. (2003), Nabuurs et al. (2002), Perez-Garcia et al. (2002), Sohngen et al. (2001) ; Lelyakin et al. (1997),
Adaptation of Forests and People to Climate Change. 2009. Alexander Buck, Pia Katila and Risto Seppälä. (eds.). IUFRO World Series Volume 22. Helsinki. 224 p.
OSU AED Policy & Outlook Fall 2007
The role of human adaptation in current timber market studies.
• Stock Management is important– Huge stock of forests now available to manage. – Climate change favors faster growing species and quickens the adoption of faster growing plantation species.
– Favors expansion of output in sub‐tropical regions
• Fires and other losses can have important implications for output.– Example from North America
• Adaptation through harvesting and planting can reduce impacts on the carbon system in managed ecosystems.
Age m3/ha/yr $/ha
US Southern Pine 30 4.8 $3,180
S. China mixed 50 1.8 $771
Canada Boreal SW 70 1.6 $288
Russia Boreal SW 100 1.0 $58
South Amer. Eucalypt 10 7.0 $8,453
Oceania SW 30 13.5 $7,937
The role of human adaptation in current timber market studies.
• Stock Management is important– Huge stock of forests now available to manage. – Climate change favors faster growing species and quickens the adoption of faster growing plantation species.
– Favors expansion of output in sub‐tropical regions
• Fires and other losses can have important implications for output.– Example from North America
• Adaptation through harvesting and planting can reduce impacts on the carbon system in managed ecosystems.
OSU AED Policy & Outlook Fall 2007
Production Impacts in North AmericaAnnual change relative to baseline
Sohngen et al. (2001)
-80
-60
-40
-20
0
20
40
60
80
100
120
1995 2005 2015 2025 2035 2045 2055 2065 2075 2085 2095
Year
Mill
ion
m3
per y
ear
PNW
South
N Conifer
Temp Dec
Mountain
-150
-100
-50
0
50
100
150
1995 2005 2015 2025 2035 2045 2055 2065 2075 2085 2095
Year
Mill
ion
m3
per y
ear
PNW
SouthN Conifer
Temp DecMountain
Regeneration Scenarios
Dieback Scenarios
Note: Currently 4.2 Mha/yr of forest area burns in NA (2.5 CA, 1.7 US).
Dieback implies approximately additional 1.6 Mha/yr (38%) burn, with largest increases in northern areas.
The role of human adaptation in current timber market studies.
• Stock Management is important– Huge stock of forests now available to manage. – Climate change favors faster growing species and quickens the adoption of faster growing plantation species.
– Favors expansion of output in sub‐tropical regions
• Fires and other losses can have important implications for output.– Example from North America
• Adaptation in markets can limit carbon emissions in managed ecosystems.
OSU AED Policy & Outlook Fall 2007
Adaptation in managed ecosystems• Adaptation through harvesting and replanting substantially reduce the
losses that would otherwise occur if natural systems adapted on their own.
• Results below are for the US only
DOLY&TEM; UKMO
-800
-600
-400
-200
0
200
400
600
800
1000
1995 2015 2035 2055 2075 2095 2115 2135Year
Tg Carbon per Decade
Natural Model
Human Response
-1400
-1200
-1000
-800
-600
-400
-200
0
200
400
600
1995 2015 2035 2055 2075 2095 2115 2135Year
Tg Carbon Per Decade
Natural Model
Human Response
MAPSS & B-BGC; UKMO
Sohngen et al., 1998
State of Adaptation Analysis
Adaptation of Forests and People to Climate Change. 2009. Alexander Buck, Pia Katila and Risto Seppälä. (eds.). IUFRO World Series Volume 22. Helsinki. 224 p.
OSU AED Policy & Outlook Fall 2007
Current analysis
• Climate Change: • A2, A1b scenarios• CSIRO, Hadley, MIROC models
• Ecological Analysis: DGVM• MC1 model (MAPPS and Century Model)
• Economic Analysis:• Global Land Use Model (Sohngen and Mendelsohn, 2007)
• Results rely on how DGVM results are used. CSIROCSIRO
MIROCMIROC
HADHAD
A2A2 A1BA1B B1B1
Change in tmax2070-2099 vs 1961-1990
OSU AED Policy & Outlook Fall 2007
A2A2 A1BA1B B1B1
% Change in precip.2070-2099 vs 1961-1990
MIROCMIROC
HADHAD
CSIROCSIRO
CSIRO A1b Example
OSU AED Policy & Outlook Fall 2007
CSIRO Example CSIRO Example
OSU AED Policy & Outlook Fall 2007
Indicators of Ecosystem Change: Proportional change in Aboveground C relative to 2005.Across climate scenarios (A2+ A1b + 3 climate models)
2040-2060 2080-2100
Indicators of Ecosystem Change:2040-2060 2080-2100
NPPNPP
%%BurnedBurnedEach Each YearYear
OSU AED Policy & Outlook Fall 2007
Indicators of Ecosystem Change: Potential forest area relative to 2005.
20402040--20602060 20802080--21002100
Approach to Economic Analysis
• Yield changes captured as:
• Stock losses captured as
• Model 1: Yield change is proportional to the change in aboveground C.
• Model 2: Stock losses due to burned area and yield change is proportional to NPP.
∑=
=A
atattA VV
1,,
&δ
( ) tatatatta ghXX ,1,,1,1 1 =++ +−−= γ
Two methods to account for ecosystem changes in timber management
OSU AED Policy & Outlook Fall 2007
Adaptations Incorporated
Model 2:
• Manage existing stock by• changing rotations• Salvage
• Replant new species if growing and economic conditions warrant
• Manage future stock by • Changing rotations• Changing management & investments
Model 1:
• Manage existing stock by
• changing rotations.
• Replant new species if growing and economic conditions warrant.
• Manage future stock by
• Changing rotations
• Changing management & investments
Global Forest Outputs: Million m3 per year.
Model 1: Yield shift only;
Proportional to aboveground C
Model 2:Yield shift proportional to NPPDisturbance due to burning
OSU AED Policy & Outlook Fall 2007
Regional Output ChangeProportional Change compared to Baseline
2040-2060 2080-2100
Mod. 1
Mod. 2
Regional Output ChangeProportional Change compared to Baseline
2040-2060 2080-2100
Mod. 1
Mod. 2
OSU AED Policy & Outlook Fall 2007
Canada Forest Outputs: Proportional Change relative to Baseline
Model 1: Yield shift only;
Proportional to aboveground C
Model 2:Yield shift proportional to NPPDisturbance due to burning
Why the difference from Model 1 to 2 for Canada?
• Model 1: Aboveground C declines from the beginning.
Model 2: Forest burning builds over time
OSU AED Policy & Outlook Fall 2007
Regional Output ChangeProportional Change compared to Baseline
2040-2060 2080-2100
Mod. 1
Mod. 2
US Forest Outputs: Proportional Change relative to Baseline
Model 1: Yield shift only;
Proportional to aboveground C
Model 2:Yield shift proportional to NPPDisturbance due to burning
OSU AED Policy & Outlook Fall 2007
Why the difference from Model 1 to 2 for the US?
• Model 1: Aboveground C increases from the beginning.
Model 2: Forest burning high to begin with
US/Canada Comparison
Canada
• Model 1: Output falls due to yield reductions over time
• Model 2: Output rises due to moderate dieback and increase in yield due to NPP change.
US
• Model 1: Output rises due to yield increases over time.
• Model 2: Output falls due to relatively high dieback and limited increase in yield due to NPP change.
OSU AED Policy & Outlook Fall 2007
Conclusions• Current results:
– Wide range of adaptation responses in managed forests: rotation age, species selection, management intensity, etc.
– Supply expands +5% to +30% over the century.– Adaptation can limit C emissions from disturbance.
• More recent results: – Supply response more limited than earlier studies: ‐5% to + 10%
• Still have a ways to go in assessing ecosystem impacts using economic/behavioral models.
• Not surprisingly, how ecological impacts are modeled affects market outcomes.