Impacts of Climate Change on the Global Forest Sector

Abstract

The path and magnitude of future anthropogenic emissions of carbon dioxide will likely influence changes in climate that may impact the global forest sector. These responses in the global forest sector may have implications for international efforts to stabilize the atmospheric concentration of carbon dioxide. This study takes a step toward including the role of global forest sector in integrated assessments of the global carbon cycle by linking global models of climate dynamics, ecosystem processes and forest economics to assess the potential responses of the global forest sector to different levels of greenhouse gas emissions. We utilize three climate scenarios and two economic scenarios to represent a range of greenhouse gas emissions and economic behavior. At the end of the analysis period (2040), the potential responses in regional forest growing stock simulated by the global ecosystem model range from decreases and increases for the low emissions climate scenario to increases in all regions for the high emissions climate scenario. The changes in vegetation are used to adjust timber supply in the softwood and hardwood sectors of the economic model. In general, the global changes in welfare are positive, but small across all scenarios. At the regional level, the changes in welfare can be large and either negative or positive. Markets and trade in forest products play important roles in whether a region realizes any gains associated with climate change. In general, regions with the lowest wood fiber production cost are able to expand harvests. Trade in forest products leads to lower prices elsewhere. The low-cost regions expand market shares and force higher-cost regions to decrease their harvests. Trade produces different economic gains and losses across the globe even though, globally, economic welfare increases. The results of this study indicate that assumptions within alternative climate scenarios and about trade in forest products are important factors that strongly influence the effects of climate change on the global forest sector.

Full text

Page 1

IMPACTS OF CLIMATE CHANGE ON THE GLOBAL FOREST SECTOR
JOHN PEREZ-GARCIA1, LINDA A. JOYCE2, A. DAVID McGUIRE3and
XIANGMING XIAO4
1Center for International Trade in Forest Products, University of Washington, Seattle,
WA 98195-2100, U.S.A.
2Rocky Mountain Research Station, USDA Forest Service, Fort Collins, CO 80526-2098, U.S.A.
3U.S. Geological Survey, Alaska Cooperative Fish and Wildlife Research Unit, University of Alaska,
Fairbanks, AK 99775, U.S.A.
4Complex Systems Research Center, Institute for the Study of Earth, Oceans, and Space,
University of New Hampshire, Durham, NH 03824, U.S.A.
Abstract. The path and magnitude of future anthropogenic emissions of carbon dioxide will likely
influence changes in climate that may impact the global forest sector. These responses in the global
forest sector may have implications for international efforts to stabilize the atmospheric concen-
tration of carbon dioxide. This study takes a step toward including the role of global forest sector
in integrated assessments of the global carbon cycle by linking global models of climate dynamics,
ecosystem processes and forest economics to assess the potential responses of the global forest sector
to different levels of greenhouse gas emissions. We utilize three climate scenarios and two economic
scenarios to represent a range of greenhouse gas emissions and economic behavior. At the end of
the analysis period (2040), the potential responses in regional forest growing stock simulated by the
global ecosystem model range from decreases and increases for the low emissions climate scenario
to increases in all regions for the high emissions climate scenario. The changes in vegetation are
used to adjust timber supply in the softwood and hardwood sectors of the economic model. In
general, the global changes in welfare are positive, but small across all scenarios. At the regional
level, the changes in welfare can be large and either negative or positive. Markets and trade in forest
products play important roles in whether a region realizes any gains associated with climate change.
In general, regions with the lowest wood fiber production cost are able to expand harvests. Trade in
forest products leads to lower prices elsewhere. The low-cost regions expand market shares and force
higher-cost regions to decrease their harvests. Trade produces different economic gains and losses
across the globe even though, globally, economic welfare increases. The results of this study indicate
that assumptions within alternative climate scenarios and about trade in forest products are important
factors that strongly influence the effects of climate change on the global forest sector.
1. Introduction
The path and magnitude of future anthropogenic emissions of carbon dioxide and
other greenhouse gases will be influenced by growth in populations, economies,
and technology development, as well as policies to control or reduce fossil fuel
emissions (Wigley et al., 1996). The time path of atmospheric carbon dioxide
will likely influence changes in climate (IPCC, 1995). Future changes in climate
and atmospheric carbon dioxide are likely to affect the productivity of forests
Climatic Change 54: 439–461, 2002.
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.

Page 2

440
JOHN PEREZ-GARCIA ET AL.
(Kirschbaum et al., 1996; Mooney et al., 1999; Joyce et al., 2001) and the forest
sector (Brown et al., 1996; Solomon et al., 1996). A number of modeling studies
have applied equilibrium ecological responses from the doubled carbon dioxide
(CO2) equilibrium climate scenarios to transient forest sector models (Callaway et
al., 1994; Joyce et al., 1995; Sohngen et al., 1998; Sohngen and Mendelsohn, 1998;
Perez-Garcia et al., 1997; McCarl et al., 2000). These studies have shown that
changes in atmospheric CO2and climate influence the production, consumption
and international trade in timber products through effects on the growth of timber
growing stocks. Theforest sector responses tothese growtheffects werealso shown
to further influence timber stocks through the harvest response. In these studies the
climate was assumed to be in equilibrium with the doubled atmospheric concen-
tration of CO2and the ecological response was assumed to be in equilibrium with
the altered climate. The ecological response was also assumed to be increasing
at a constant rate over the time period needed to reach the doubled atmospheric
concentration of carbon dioxide. A recent study by Irland et al. (2001) examined
the nature of transient climate scenarios on the U.S.forest sector using the dynamic
forest sector model, Forest and Agricultural Sector Optimization Model (FASOM).
Unfortunately, the above studies did not consider the uncertainties in the transient
nature of changing carbon dioxide concentrations and climate on the global forests
or the global forest sector.
The global extent of forests changes over time with human use, natural succes-
sion and natural disturbances such as fire and disease outbreaks. Currently, forests
cover approximately 30% of the land surface, 4.1 billion hectares. Of this area,
about 11% is managed for goods and services with the extent of management
varying by regions across the globe (Brown et al., 1996, using Food and Agri-
cultural Organization (FAO) data). Consumption of industrial roundwood totaled
1.5 billion cubic meters in 1997 with an approximate value of $150 billion (FAO,
1999). Forest product values are several times larger as lumber, structural panels,
engineered wood products, and pulp all utilize this industrial raw material in their
manufacturing process. International trade in timber products is approximately
$80 billion per year, which comprises about 3% of the world’s trade (Laarman
and Sedjo, 1992, using FAO data). Globally, conifer species comprise nearly 70%
of the world’s timber harvest for industrial products (Brooks, 1993). In develop-
ing countries, the harvest is predominately nonconiferous species and the use is
primarily fuelwood. Analyses of the carbon stored in extant forests indicate that
forests currently offset significant amounts of fossil fuel emissions in individual
countries (Brownet al., 1996). Interactions between climate change and the world’s
forest sector have the potential to influence carbon storage of the world’s forests
(Sedjo et al., 1997). This may have important implications for international efforts
to control the concentration of atmospheric CO2. Thus, it is important to assess
how transient changes in atmospheric carbon dioxide and climate may influence
the world’s forest sector and timber trade into the future.

Page 3

IMPACTS OF CLIMATE CHANGE ON THE GLOBAL FOREST SECTOR
441
A previous analysis (Perez-Garcia et al., 1997) on the potential economic im-
pacts of climatic change on the global forest sector was based on equilibrium
responses of climate to a doubling of the atmospheric CO2as simulated by gen-
eral circulation models (GCMs). These GCM simulations from the 1980s did not
consider the effects of atmosphere-ocean coupling or the effects of atmospheric
aerosols. The ecological responses in the analysis were based on equilibrium re-
sponses of a terrestrial biogeochemical model that used the climatic scenarios
to simulate spatially explicit changes in net primary production (NPP) of global
forests at 0.5◦resolution (latitude by longitude). The equilibrium responses of NPP
were used to define a constant rate of change in growing stock between 1990 and
2040. These changes were used to simulate economic responses of the global forest
sector using an economic baseline constructed prior to the recent Asian recession
and the fall in consumption and production from the former Soviet republics.
State-of-the-art scenarios of projected climate are currently based on simula-
tions that consider how the time-dependent response of climate is influenced by
the effects of the coupling between the atmosphere and ocean and by the effects
of atmospheric aerosols (Mitchell et al., 1995). Similar advances have been made
with respect to ecological responses. For example, state-of-the-art responses of
large-scale carbon dynamics are based on the time-dependent simulations of ter-
restrial biogeochemical models that are driven by transient changes in climate and
atmospheric CO2(Kindermann et al., 1996; Xiao et al., 1998; Tian et al., 1998,
1999, 2000; McGuire et al., 2000, 2001; Schimel et al., 2000). These advances in
the simulation of climate and carbon dynamics provide the opportunity to simulta-
neously assess how both climate and terrestrial carbon dynamics may respond to
different levels of greenhouse gas emissions (Xiao et al., 1998). In this study, we
evaluate the potential economic responses of the global forest sector to different
scenarios of climate change produced from alternative levels of greenhouse gas
emissions. The economic baseline in the analysis considers the recent collapse of
the Asian economy, the fall in production and consumption of wood products in
Russia, and their influence on the global forest sector.
2. Methods
2.1.
APPROACH
In this study we conducted a set of simulations that represent the first step towards
integrating global ecological and economic models to simulate alternative futures
of the responses of the forest sector behavior to projected climate change. Our ap-
proach is to drive a global ecological model with different climate change scenarios
derived, in part, from a global economic model of greenhouse gas emissions, and
to use the outputs of the ecological model as inputs to a global model of the forest
economies. Time-dependent changes of CO2emissions, ocean heat diffusion and

Page 4

442
JOHN PEREZ-GARCIA ET AL.
aerosol effects are used as inputs to the Terrestrial Ecosystem Model (TEM; see
Xiao et al., 1998) for simulating the physiological responses of vegetation carbon
to projected climate change. The changes in vegetation carbon simulated by TEM
are then used to modify timber supply in the CINTRAFOR Global Trade Model
(CGTM; see Cardellichio et al., 1989) for forest products to simulate economic
changes in timber supply and the subsequent economic behavior of the forest
sector. While our approach in this study does not attempt to achieve full integra-
tion or consistency in assumptions among the models, it allows us to explore the
types of regional responses in the forest sector that might emerge and provides
us with insights for taking the next steps towards achieving more comprehensive
integration.
2.2.
MODELS USED IN THE ANALYSIS
2.2.1. The Terrestrial Ecosystem Model
The TEM is a process-based ecosystem model that makes monthly estimates of
important carbon and nitrogen fluxes and pools for terrestrial ecosystems across
the globe (Raich et al., 1991; Melillo et al., 1993; McGuire et al., 1992, 1993,
1995, 1997, 2000, 2001; Xiao et al., 1997, 1998). In this study, we used version
4.1 of TEM (Tian et al., 1999), which simulates the time-dependent response of
carbon pools in response to time-dependent inputs of climate and atmospheric CO2
(Melillo et al., 1996; Prinn et al., 1999; Tian et al., 1998, 1999, 2000; McGuire et
al., 2000; Clein et al., 2000; Schimel et al., 2000). Regional analyses with TEM
indicate that the model simulates the temporal changes in carbon dynamics associ-
ated with major regional climate events such as the Dust Bowl of the Central U.S.
in the 1930s (Tian et al., 1999), the El Niño Southern Oscillation (Tian et al., 1998,
1999, 2000; McGuire et al., 2001), and the North Atlantic Oscillation (McGuire et
al., 2000). We applied TEMglobally to examine the vegetation carbon responses of
terrestrial ecosystems to three transient scenarios of changes in atmospheric CO2
and climate from 1990 to 2100. The three scenarios were derived from projections
of the Integrated Global System Model (IGSM; see Prinn et al., 1999).
The application of TEM to a grid cell requires the use of data describing
monthly climate (precipitation, mean temperature, and mean cloudiness), soil tex-
ture (proportion of sand, silt, and clay), elevation, and vegetation types (Pan et al.,
1996). Soil texture and vegetation types are used to define the soil- and vegetation-
specific parameters for a grid cell. For extrapolating TEM globally, we used
spatially explicit data sets organized at a spatial resolution of 0.5◦longitude by 0.5◦
latitude. The natural vegetation classification has 18 upland vegetation types and
13 floodplain and wetland vegetation types (Melillo et al., 1993). The processing
of climate data from the IGSM projections for deriving the inputs required by TEM
are described by Xiao et al. (1998).

Page 5

IMPACTS OF CLIMATE CHANGE ON THE GLOBAL FOREST SECTOR
443
2.2.2. The CINTRAFOR Global Trade Model
The CGTMsimulates harvests and prices in log and product markets by calculating
economic welfare produced by alternative assumptions of forest growth (Kallio et
al., 1987; Cardellichio et al., 1989). In a previous study, we used the CGTM to sim-
ulate how the effects of equilibrium changes in climate on forest growth influence
the global forest sector (Perez-Garcia et al., 1997). In the present study, the CGTM
adjusts regional timber supply equations by using the time-dependent changes in
vegetation carbon stocks simulated by TEM that are associated with alternative
assumptions on carbon dioxide emissions, ocean heat diffusion and aerosol effects.
The model then computes the spatial equilibrium of the global forest economy by
maximizing economic welfare subject to processing capital and timber growing
stock constraints.
CGTM computes market equilibria for all regional forest products markets
considering constraints on processing capacity and available wood resources to de-
scribe the behavior of the global forest sector. The dynamic structure of the model
does not imply inter-temporal market equilibrium because the CGTM does not
compute the market solution for all periods simultaneously. Instead, the model cap-
tures regional details using 43 log-producing and 33 product-consuming regions
year to year adjusting forest products demand, forest inventory and production
capacity based on demand end-use factor sub-models, timber supply sub-models
of forest management and historical profitability respectively. The specification of
the economic model allows us to describe the effects of alternative climatic futures
on forest growth, timber supply, processing capacity, consumption of various forest
products, and trade in these forest products.
2.3.
EXPERIMENTAL DESIGN AND SIMULATION PROTOCOL
To assess the sensitivity of the forest sector to changes in forest growth asso-
ciated with changes in climate, we used a series of transient climate scenarios
and economic scenarios. We create an economic baseline and then impose one
of two economic scenarios with one of three transient climate scenarios. We re-
port changes as measured from the economic baseline. The three transient climate
scenarios integrate different assumptions about future CO2emissions, atmosphere-
ocean thermal interactions and aerosol effects as described below. The economic
scenarios place boundaries on behavioral assumptions of non-market mecha-
nisms employed in the forest sectors of centrally-planned or otherwise non-market
economies.
The three transient climate change scenarios from the IGSM sensitivity study
(Prinn et al., 1999) describe possible future climates for the period of 1977 to 2100
using a range of uncertainty in parameters that lead to low and high levels of green-
house gas emissions under the assumption that no emission control policies are
implemented. The projections of anthropogenic emissions of greenhouses gases as
derived from theEPPA(Emissions Production and Policy Analysis) model(Yang et