Historic, Archive Document

Do not assume content reflects current
scientific knowledge, policies, or practices.

ff^^ United States
aULI;) Department of
Agriculture

Forest Service

Climate Change and America's Forests

Linda A. Joyce, Michael A. Fosberg, and Joan M. Comanor

Rocky Mountain
Forest and Range
Experiment Station

Fort Collins,
Colorado 80526

General Technical
Report RM-187

4s

936692

Preface

The Forest and Rangeland Renewable Resources Planning Act of 1974
(RPA) , PL. 93-378, 88 Stat. 476, as amended, directed the Secretary
of Agriculture to prepare a Renewable Resources Assessment by De-
cember 31, 1975, with an update in 1979 and each tenth year there-
after. This Assessment is to include "an analysis of present and
anticipated uses, demand for, and supply of the renewable resources
of forest, range, and other associated lands with consideration of the
international resource situation, and an emphasis of pertinent sup-
ply, demand and price relationship trends" (Sec. 3. (a)).

The 1989 RPA Assessment is the third prepared in response to the
RPA legislation. It is composed of 13 documents, including this one.
The summary Assessment document presents an overview of analyses
of the present situation and the outlook for the land base, outdoor recre-
ation and wilderness, wildlife and fish, forest-range grazing, miner-
als, timber, and water. Complete analyses for each of these resources
are contained in seven supporting technical documents. There are also
technical documents presenting information on interactions among
the various resources, the basic assumptions for the Assessment, a
description of Forest Service programs, and the evolving use and
management of the nation's forests, grasslands, croplands, and relat-
ed resources.

The Forest Service has been carrying out resource analyses in the
United States for over a century. Congressional interest was first ex-
pressed in the Appropriations Act of August 15, 1976, which provided
$2,000 for the employment of an expert to study and report on forest
conditions. Between that time and 1974, Forest Service analysts pre-
pared a number of assessments of the timber resource situation inter-
mittently in response to emerging issues and perceived needs for better
resource information. The 1974 RPA legislation established a period-
ic reporting requirement and broadened the resource coverage from
timber to all renewable resources from forest and rangelands.

USDA Forest Service

General Technical Report RM-187

February 1990

Climate Change and America's Forests

Linda A. Joyce, Range Scientist
Rocky Mountain Forest and Range Experiment Station1

Michael A. Fosberg,
Forest Fire and Atmospheric Sciences Research Staff, USDA Forest Service

and

Joan M. Comanor, Assistant Director for Resource Management
Cooperative Forestry, USDA Forest Service

Acknowledgments

We thank Dr. R. A. Houghton, Dr. Chris Eager, Dr. Michael G. Ryan,
and Dr. Douglas Fox for useful reviews of the manuscript. Work on
this paper was completed while the senior author was a visiting scien-
tist at the Ecosystem Center, Marine Biological Lab, Woods Hole, MA,
and discussions with scientists at the Ecosystem Center were helpful.

1 Headquarters is in Fort Collins, in cooperation with Colorado State University.

Contents

Page

Introduction 1

The Greenhouse Effect 1

Scientific Bases for the Greenhouse Effect 1

Quantifying the Atmospheric and Ecological Responses 2

Modeling the Atmospheric Response 2

Coupling the Biosphere to the Geosphere 3

Modeling the Ecological Response 3

Forest Changes Under Climate Change 4

Eastern Forests 5

South and Southwest 5

West 5

Sensitivity of Forest Species Predictions to Uncertainties in the General

Circulation Models (GCM's) 7

Ecological Uncertainties in the Projections of Climate Change 7

The Future 9

References 9

Climate Change and America's Forests

Linda A. Joyce, Michael A. Fosberg, and Joan M. Comanor

Introduction

The timber projections in the Forest Service Assess-
ment assume a future in which the climate follows histor-
ical trends and in which changes in timber production
and land use are an outgrowth of these trends, not abrupt
discontinuities from the past (Darr in press, Haynes in
press). These assumptions may not be met if the earth's
climate changes rapidly.

Since the beginning of the industrial revolution, the
chemistry of the atmosphere has been altered by in-
creases in the concentrations of trace gases such as car-
bon dioxide and methane. These greenhouse gases trap
a portion of the earth's infrared radiation and warm the
planet. Further increases in the concentration of these
greenhouse gases are predicted to raise the atmospheric
temperature by 3 to 5°C within the next 100 years — a time
span comparable to the planting-to-harvest interval of
many commercial tree species. Major changes in the lo-
cation and abundance of North American tree species
were associated with a similar 5°C warming that occurred
over a period of 8,000 years between 15,000 and 7,000
years ago (Bernabo and Webb 1977). This interglacial tem-
perature change is of the same magnitude as the predict-
ed temperature rise, 5°C, but this currently predicted
temperature rise is projected to occur in less than 100
years, one one-hundredth of the interglacial time span.

Our perception of change is often associated with
seasonal to decadal regional weather changes, such as
the summer drought of 1988 or the hot, dry years in the
1980 's; and local to regional environmental changes, such
as the impacts of acid-rain or urban smog on vegetation.
As we begin to understand the earth system, we need to
consider long-term changes, such as those changes as-
sociated with global climate. There is great uncertainty
in the projections of climate change on local ecosystem
responses. However, we can say that these factors will
play a major role in abrupt changes in the landscape:
changes in precipitation and, to a lesser extent, temper-
ature will restrict the persistence of ecological systems;
and changes in disturbances, such as fire, insects, and
disease, will impose new and different stresses on
ecosystems. There is great need to determine the impact
of this potential climate change on North American
ecosystems and, in particular, our forest resources. Relia-
ble estimates of the magnitude and rate of climate change
are needed at many decision levels within society: in-
dividuals (e.g., ranchers and farmers), industry (e.g., fore-
stry), and governments (e.g., resource managers and
regulators). This document summarizes the current
research on the impacts of climate change on America's
forests.

The Greenhouse Effect

Scientific Bases for the Greenhouse Effect

The balance between the incoming solar energy and
the outgoing energy determines the earth's temperature.
A small change in either direction would result in a cool-
ing or warming of the earth. Approximately 43% of the
incoming energy is absorbed at the earth surface and this
energy warms the land and oceans. A portion of the
received energy warms the atmosphere directly through
heat transfer. A portion is also reradiated toward space
as long-wave infrared radiation. A small percentage of
this outgoing long-wave radiation is absorbed by certain
trace gases such as carbon dioxide, methane, and water
vapor, and this absorption further warms the atmosphere.
If the total incoming solar energy is balanced by the to-
tal energy returned to space, the temperature of the earth
would remain constant. The equilibrium temperature for
the earth is currently 15 °C.

Recent interest in the greenhouse effect is focused on
whether the equilibrium temperature has been disturbed
through increases of the greenhouse gases such as car-
bon dioxide, methane, nitrous oxides, and others (Han-
sen et al. 1987). Precise monitoring of the amount of
carbon dioxide in the atmosphere (Keeling 1984) has
shown a steady increase since 1958, the beginning of the
record at Mauna Loa Observatory (fig. 1). Comparing
these recent data from direct methods with measure-
ments of atmospheric carbon dioxide from indirect
methods, such as carbon dioxide in air trapped in per-
manent ice fields and the analysis of isotopic carbon ra-
tios in tree rings, indicates that before 1850, atmospheric
concentration of carbon dioxide was approximately
270 ± 10 parts per million (ppm) , as compared to the cur-
rent level of 350 ppm (Hoffman and Wells 1987). This
25% increase in atmospheric carbon dioxide since 1850
is important because atmospheric concentrations of car-
bon dioxide are strongly correlated with global temper-
ature. Ice core data extending back 160,000 years (fig. 2)
clearly demonstrate this correlation (Fifield 1988, Fried-
li et al. 1986). Increases in other greenhouse gases such
as methane have also been observed (Hoffman and Wells
1987). The increases in atmospheric carbon dioxide are
attributable to human activities such as the burning of
fossil fuels, deforestation, the burning of forests,
byproducts from agriculture such as methane from cattle
and rice production, and release of manufactured chem-
icals such as chlorofluorocarbons, and biotic sources
such as the decomposition in forests and carbon cycling
in oceans.

1

32S

320

316

310

19M I960 1

(.«) .04 .70 30 .80 (MHM) 47

1970 1975 1900 1981

.71 .94 1.90 1.33 .87 1.22 2.22 .57 .90 1.05 1.57 1.571.36 1J1
ANNUAL CHANGE (ppWyr)

Figure 1.— Mean monthly concentrations of atmospheric C02 at Mau-
na Loa. The yearly oscillation is explained mainly by the annual cycle
of photosynthesis and respiration of plants in the northern
hemisphere (after National Research Council 1983).

Projections of future concentrations of these green-
house gases are based on forecasts of energy consump-
tion, energy efficiency and population growth. Current
projections indicate that with present technology and
population growth, the concentrations of the greenhouse
gases would double by the year 2030, and that even with
high levels of energy conservation and efficiency, con-
centrations would double by 2075 (Mintzer 1987).

3o ' — 3o" ' — W

Ago; thousand* of yoart

Prosont

Figure 2.— The Vostock record of temperature from Antarctica, and
concentrations of carbon dioxide in the atmosphere (from Fifield
1988).

Quantifying the Atmospheric and
Ecological Responses

Modeling the Atmospheric Response

General circulation models (GCM's) are the equations
representing the physical concepts of conservation of
mass, energy, and momentum. Such models describe the
atmosphere and oceans with a large number of discrete
points for which forecasts of temperature, pressure, water
(for the atmosphere), and salinity (for the oceans) are
made. These forecasts permit calculation of clouds, wind,
precipitation, and exchange of energy between the bi-
osphere and the geosphere (Dickenson 1982, Schneider
1988).

Major weaknesses and sources of uncertainty in ap-
plying these models to predict future ecosystem
responses are: (1) coarse spatial resolution, (2) inadequate
representation of the role of clouds in the energy balance,
and (3) inconsistent prediction of the hydrologic cycle.
Spatial resolution of these models is very coarse, 5-7
degrees of latitude and longitude, when compared to
ecosystem dimensions. Physical processes associated
with small physical dimension phenomenon such as in-
dividual clouds cannot be described in these models. In-
stead, such processes are represented by an expected
mean effect on the energy, mass, and momentum budg-
ets. This shortcoming is particularly acute in moun-
tainous regions where ecosystem dimensions are very
small and are strongly related to elevation, and where
local climate variations are large (Schlesinger 1988).

Representation of clouds in these GCM's is particularly
important because of their influence on both incoming
solar radiation and outgoing infrared radiation. Intercom-
parison of the different GCM predictions have attempt-

2

ed to reduce the uncertainty resulting from how clouds
influence the model results (Schlesinger 1988), but there
is still need for improvement before this problem can be
considered resolved (Cess et al. 1989). Effects of clouds
on the local energy exchange is three times that predict-
ed from the greenhouse gases (Ramanathan et al. 1989).

How the oceans are depicted affects the prediction of
hydrological cycles. In some GCM models, oceans are
represented by a shallow ocean in which predetermined
sea surface temperatures are used to regulate the at-
mospheric circulation. In other models, a deep ocean
with ocean current to redistribute the heat is used, but
these models do not show consistent results (Bryan 1988,
Schlesinger 1988).

The four models, Oregon State University (OSU), Na-
tional Center for Atmospheric Research (NCAR), NASA
Goddard Institute for Space Studies (GISS), and the Geo-
physical Fluid Dynamics Laboratory (GFDL), show a
degree of consistency in predicting the future tempera-
ture rise and the regional distribution of these tempera-
tures. Also, all four models predict that the global
precipitation will increase, primarily because a warmer
atmosphere has a higher saturation vapor pressure
(Mitchell 1988). Intercomparison of model results for
regional precipitation distribution shows far less con-
sistency (Kellogg and Zhao 1988). This lack of consisten-
cy is particularly troublesome because the hydrological
budget is more important than temperature in determin-
ing leaf area, vegetation structure, and the mass of vege-
tation (Woodward 1987).

The rate at which climate will change is important. If
the climate evolves slowly, the biosphere may be able to
adapt. Three of these models attempt only to predict the
equilibrium climate of the future. Only the GISS model
allows the greenhouse gases to increase with time and
gives an estimate of the rate of climate change.

Coupling the Biosphere to the Geosphere

Direct, interactive coupling of the biosphere to the at-
mosphere in global models so that there is a direct ex-
change of mass and energy (Abramopoulos et al. 1988)
will need to be improved dramatically before results use-
ful to resource managers will be available. Current ap-
proaches only attempt to describe the heat and water
vapor exchange and make little reference to the structure
and composition of the ecosystem, to the abundance of
individual species, or to any mediating effects of the bi-
osphere on the atmosphere. Heterogeneity of land and
water distribution on the surface of the earth contribute
to the difficulty of interpreting mean values for climate
over a varied region.

Modeling the Ecological Response

Without models that directly link atmospheric proc-
esses to ecological processes, climate projections from
GCM's are used to construct scenarios as a context in
which to examine the behavior of extant ecological
models. Attempts to quantify the role of terrestrial biota
in the global carbon cycle have their origins in global

models of carbon flux (Houghton et al. 1983). Such
models have been used to estimate where carbon is stored
globally: 617 109 tons in vegetation and 1,652 109 tons
in soils of terrestrial ecosystems (Woodwell 1983), 39,660
109 tons in oceans, and 740 109 tons in the atmosphere.
In addition, these models have been used to estimate the
net biotic flux of carbon (carbon release to the atmosphere
from deforestation and other changes in land use) which,
in 1980, was estimated to be 1.9 ± 0.99 109 tons carbon
with only 0.11 109 tons carbon released from outside the
tropics (Houghton et al. 1987). Models of this type quan-
tify the biotic contribution to atmospheric carbon diox-
ide and must be compared to models computing the
nonbiotic contribution of human activities to atmospher-
ic carbon dioxide (Nordhaus and Yohe 1983). World con-
sumption of fossil fuels contributes nearly 5 109 tons
( ± 10%) per year to the atmosphere with the United States
contributing about one-fourth of this emission (Carbon
Dioxide Assessment Committee 1983).

Current approaches to modeling the ecological
response to climatic change differ by the questions they
attempt to answer: physiologically-based plant models,
population models, ecosystem models, and regional or
global models (Agren et al. in prep.). The response of
individual plant processes to climate change is the fo-
cus of physiologically-based models. While such models
contribute to our understanding of biochemical reactions
within plants, these models lack mechanisms to examine
interactions of nutrient cycling or species competition
at a scale larger than a single plant.

Plant establishment, growth, seed production, and
death are simulated in population-community models.
Gap-phase models, one example of population-commun-
ity models, offer a way of predicting future forest spe-
cies composition under disturbance at a scale that is
meaningful to resource managers. These models predict
the establishment, growth, and death of individual trees
and implicitly account for competition for light, water,
and nutrients among trees (Botkin et al. 1972). Such
models allow individual species to die and to be replaced
by new species that are better adapted to the environment,
or that are more competitive for light, water, and
nutrients. A major uncertainty in these models is the rate
of species dispersal into a region and the lack of explicit
dispersal mechanisms (U.S. EPA 1988). In the current
models, a species is present or absent, and when present,
migrates at the same rate as climate change — an unlike-
ly assumption.

Ecosystem models focus on the biogeochemical proc-
esses of fixation, allocation, and decomposition of car-
bon, and the cycles of nitrogen, phosphorus, sulfur, and
other elements. Rates of nutrient cycling may change
more rapidly than species composition (days to years ver-
sus years to centuries) and, thus, change the environment
for species interactions. It is unclear how much climate-
induced process-level change (e.g., decomposition) oc-
curs within ecosystems before plant community turnover
occurs and, conversely, to what extent species changes
drive process-level changes in these systems (Davis 1988).
Incorporation of nutrient cycling in gap-phase models
in forests (Pastor and Post 1988) suggests that a synthesis

3

of these two approaches will be needed to unravel the
likely changes in species abundance and ecosystem
processes (Davis 1988).

Recent regional and global models use the coincidence
of climate and vegetation zones (correlation) to describe
the future distribution of plant communities. The Hold-
ridge Life Zone Classification system (Holdridge 1964)
is based on correlating vegetation type (e.g., spruce-fir)
to gradients of temperature, precipitation, and the ratio
of potential evapotranspiration to precipitation. Histori-
cal pollen records can also be correlated with records of
past climate, and these correlations can be used to predict
future vegetation distributions under a projected climate.
All of these approaches are based on the equilibrium rela-
tionship between climate and vegetation distribution and
can be helpful in determining the global distributions
of vegetation under steady state. The assumption of
steady state implies that under climate change, the vege-
tation at a site shifts to the plant community for which
the new environmental condition is the optimal climate.
If climate zones were displaced geographically, the forest,
as it looks now, would migrate with its preferred climate.
Such migration may be possible if climate were to change
slowly, but under rapid climate change, maintaining an
intact forest would be difficult or even impossible be-
cause each species in the forest will migrate at a differ-
ent rate (Davis 1981). This disassociation of species in
migrating forests was clearly observed during the early
Holocene (Bernabo and Webb 1977).

The major shortcoming of all current approaches is the
omission of disturbances such as fire, insects, disease,
and pollutants. These analyses do not address potential
changes in the frequency and severity of traumatic events
and how such changes, in turn, will impact primary
productivity, seed production, seedling establishment,
and species competition. Insects and animals, particu-
larly herbivores, represent a disturbance that could
significantly change species composition as well as
nutrient cycling processes of ecosystems under climate
change. For example, if hardwoods replace conifers, gyp-
sy moth defoliation will certainly influence primary
productivity unless mitigating action is taken (Wingert
1988). Also, if cottonwood (Populus deltoides) becomes
a more important species for pulp and paper, impact of
climate change on melampsora leaf rust (McCracken et
al. 1984) must be taken into account (Fosberg 1988).

Increased insect- and disease-caused losses in our na-
tion's forests will become one of the first observed effects
of climate change. Evidence can be found in the pest-
caused epidemics which now occur as a result of peri-
odic droughts or rainy periods. Changes in climate either
through effects on the pest or on the host may increase
or decrease pest-caused losses. High temperatures and
reduced precipitation cause insect epidemics when these
climatic factors stress the tree (host) to the point that the
hosts lose their inherent resistance to native pests. In-
creased moisture can increase losses where disease was
limited in distribution and infection was limited by low
moisture conditions. Under climate change, currently im-
portant pest problems may all but disappear and new epi-
demics will arise. These pest attacks will often determine

the new geographic distribution of tree species under the
new climatic conditions.

Fire frequency and severity is also missing in assess-
ing the impact of climate change on forests. Charcoal ana-
lyses of sea sediments (Herring 1985) have shown a weak
but definite trend of charcoal deposition over the last 50
million years. Charcoal analyses of lake sediments have
shown fire cycles and climatically induced changes in
fire regimes (Clark 1988). Combining these data with tem-
perature relations (Fifield 1988), we see a weak but posi-
tive correlation between temperature and charcoal,
suggesting increased fire frequency under warmer
climates.

There are few definitive studies of the direct effects of
climate change on fire frequency and severity. Direct ef-
fects would be the changes in drought frequency, humid-
ity, precipitation, and other weather elements that
determine day-to-day variation and interannual variabil-
ity in fire behavior. Fried and Torn (in prep.) compared
the changes in area burned under the current and a
double carbon dioxide climate. They found that there
would be a two-fold increase in modest-sized fires (a few
hundred hectares) and a three-fold increase in fires great-
er than 1,000 hectares. Fried and Torn (1988) based their
studies on an area of the California Sierra Nevada in
which the ecosystems are expected to remain unchanged
in a future climate.

For many regions, species composition of forests is
projected to change. Much of the structure and compo-
sition of a forest will remain long after climate change-
induced stress has prevented regeneration of those spe-
cies. New species will take hold during the transition
from one vegetation community to another, and a tran-
sitional forest will contain elements of both vegetation
types. As the structure, composition, and total biomass
of the forest change, so will the behavior of fire. A shift
on the uplands to hardwood savannahs will likely cause
an increase in fire severity in the Lake States (Fosberg
1989).

The size, shape, and distribution of forest land, forest
types, and successional stages create a mosaic across the
landscape that contributes to the production of wildlife
and the use of land for other resources such as grazing.
There is concern that the current increasing forest frag-
mentation will eliminate some species as functioning
members of certain regional faunal communities (Flather
and Hoekstra 1989). The spatial pattern of land use and
vegetation cover influences the migration and dispersal
of insects, birds, and animals, and these influences un-
der a changing climate remain to be considered in eco-
logical models. The impacts of changing forest type and
the associated changing interspersion on wildlife and
other resources have not been addressed.

Forest Changes Under Climate Change

Assessment of the forest resources 50 to 100 years from
now as a result of the greenhouse effect has not been done
in any systematic fashion. Coverage of the country is not
uniform, several estimates were made for some regions

4

and only one estimate for others, and, finally, different
methods have been used for different regions. While
research is in process on modeling the ecological re-
sponse to climate change, several existing models have
been used to examine GCM scenarios. Gap-phase models
have been used to examine changing species distribu-
tions in eastern and western forests. Other approaches
have included interpreting historical change as an indi-
cation of future change. Correlation of modern pollen dis-
tribution with climatic data gives a model that can be
compared to fossil pollen records under different cli-
mates and used to project species distribution into fu-
ture climates (Overpeck and Bartlein 1988, as described
in U.S. EPA 1988).

These ecological models use climate scenarios from
the GCM's as inputs in order to quantify the ecological
response. Interpretation of these results must consider
the uncertainties of GCM climate projections, the poten-
tial interactions for disturbances and interactions not cur-
rently in the ecological models, and, finally, the
uncertainties in the ecological models themselves. In
those regions where more than one model or method has
been applied, coincidence of prediction may give more
support to estimates.

Eastern Forests

Most projections indicate a movement of conifers north
with an inward migration of hardwoods from the South,
a movement of grassland and savanna types eastward on
the western boundary; however, the degree and magni-
tude of these changes vary. In the New England states un-
der a more severe GFDL-correlation projection, conifers
(spruce, northern pines) would retreat into Maine;
however, there would not be appreciable change under
a milder GISS-correlation scenario (U.S. EPA 1988). Us-
ing these same two models, sugar maple shows a simi-
lar pattern (U.S. EPA 1988). Using gap models, the New
England forests would be replaced by more hardwoods,
particularly by oak species from the eastern mid-United
States (Botkin et al. 1988; Overpeck and Bartlein 1988;
Zabinski and Davis 1988, as described in U.S. EPA 1988).

For the Great Lakes area (U.S. EPA 1988), conifers
(spruce) will likely remain in the northern portion of the
Great Lakes region and potentially migrate as far north
as James Bay with a GISS-correlation scenario. Under a
GFDL-correlation scenario, conifers (balsam fir) would
be totally lost from the Great Lakes region. Sugar maple
would show similar migration patterns under these two
scenarios. The gap-phase model simulations show less
dramatic changes. Under a GISS-gap-phase model
scenario, Botkin et al. (1988) predicted that the boreal
forests would disappear by 2040 from the northern Lake
States region and that sugar maple, oaks, and other hard-
woods would dominate on the drier and better sites in
this area, with the potential for increased biomass. On
more southerly maple-oak sites, tree biomass would
decline 37% to 99%. Solomon and West (1987), using the
NCAR GCM, also show hardwoods preferred over
conifers, but conifers remained in the region. This simu-
lation predicted a decrease in biomass. Using correlation

Figure 3.— Projected changes in loblolly range assuming doubled at'
mospheric C02 (after Solomon et al. 1984)

analysis, Zabinski and Davis (as cited in U.S. EPA 1988)
predicted local extinction of tree species such as eastern
hemlock and sugar maple in the Great Lakes region.
Southern pine species could migrate into the present
hardwood forest lands of eastern Pennsylvania and New
Jersey (U.S. EPA 1988).

South and Southeast

Most projections indicated a decline of southern forests
as we know them, a reduction in the biomass, conver-
sion of some forests to grassland, species regeneration
becoming difficult, and a movement of southern pine
species north. Using a gap-phase model and GFDL, GISS,
and OSU climate scenarios, southern pines would be
greatly reduced or eliminated in Mississippi, South
Carolina, and Georgia (Urban and Shugart 1988, as cited
in U.S. EPA 1988). Biomass would be reduced 30% in
Tennessee. An earlier analysis by Solomon et al. (1984)
also predicted a marked reduction in Mississippi, Geor-
gia, and South Carolina, and showed loblolly pine in-
vading into Tennessee (fig. 3). Under GFDL-correlation
and OSU-correlation scenarios, southern pines extended
northward but did not move out of the South and
Southeast (Winjum and Neilson 1988).

Miller et al. (1987) described the subtle implications
of these projections. The current distribution of loblolly
pine is overlain on relatively deep soils, whereas areas
into which loblolly is projected to move have a relative-
ly high proportion of shallow, steeply sloping, coarse-
textured soils on rocky uplands. Thus, even though the
area of the range has increased, productivity is likely to
decline because of the marginal productivity of these
sites. The seasonal distribution of precipitation and tem-
perature under climate change could also affect the qual-
ity of wood for loblolly. An extended period of drought
in the northern limits of its range would result in lower-
specific-gravity wood and, thus, wood of poorer quality.

West

Projections for the West focused on correlation analy-
sis and, consequently, focused on individual species
responses rather than community-level projections. Spe-

5

cies responses indicated either movement up an eleva-
tional gradient or northward migration. Leverenz and Lev
(1987), using correlation and an unspecified GCM,
predicted that Douglas-fir (fig. 4) would expand to higher
elevations as the lower limit of the continuous winter
snowpack climbs upslope. Increased summer drought
and rising winter temperatures result in declining im-
portance of Douglas-fir on the east slope of the Rocky
Mountains and in the southern limit of its current range.

The range of ponderosa pine (fig. 5) in interior
Washington and the eastern slope of the Rocky Moun-
tains would decrease with deficit spring water balances
(Leverenz and Lev 1987). While summer drought allows
ponderosa pine to expand in California and Oregon, in-
creased temperatures are projected to push ponderosa
pine upslope in these states and in Washington, Mon-
tana, Idaho, and the middle and southern Rocky Moun-
tains. The southern Rocky Mountains would have major
losses in ponderosa pine (fig. 5).

Western hemlock would be restricted to the wetter sites
west of the Cascade Mountains. In the northern Rocky
Mountains, western hemlock and western larch would
be lost in the Idaho panhandle and eastside of Oregon
and Washington (Leverenz and Lev 1987) . Lodgepole pine
would not be effected greatly in its western extent. No
estimates were made for redwood or other species.

Western larch was projected to increase on better sites,
and on sites where fire frequency is increased. On sites
where summer drought increases, western larch will be
restricted to upslope movement.

Sensitivity of Forest Species Predictions to
Uncertainties in the General Circulation
Models (GCM's)

The four GCM's predict that the global mean surface
temperature rise will range from 2.8°C to 4.2°C
(Schlesinger 1988). North American regional and
seasonal distributions of these temperature increases
differ by as much as 8°C for summer and by 4°C for winter
(Schlesinger 1988). Not only does the projected global
mean vary, but the implication to regional climate pat-
terns varies by model. When the spatial and seasonal dis-
tribution of precipitation is expressed as soil moisture,
the southwestern, southern, and southeastern states are
expected to be drier during the winter (fig. 6). During
the summer, the entire country is expected to be drier
(Kellogg and Zhao 1988). However, there are marked
differences between each of the model predictions in
both winter (fig. 6) and in summer (fig. 7). Areas of
greatest discrepancy and, therefore, uncertainty are in
winter precipitation for the West Coast, the Great Basin,

Figure 4.— The current distribution of Douglas-fir and projected dis- Figure 5.— Current distribution of ponderosa pine and projected dis-
tribution under a doubled C02 scenario. Hatching is directed tribution under a doubled C02 scenario. Hatching is directed
toward the zones of decreasing acreages (after Leverenz and Lev toward the zones of decreasing acreages (after Leverenz and Lev
1987). 1987).

6

Figure 6— Increases (clear areas) and decreases (scratches) of winter
soil moisture relative to the control when carbon dioxide is dou-
bled (after Kellogg and Zhao 1988).

Rocky Mountains, the Mid-West, and the Northeast.
There is greater consistency between the predictions and
confidence during the summer.

Natural variations in annual precipitation and mean
temperatures have always existed. During the past 100
years, the long-term temperature record for the United
States has not shown any systematic change, but has
ranged from 10.6°C to 12.8°C (Karl and Jones 1989). Over
the last 2,700 years, which includes the Little Ice Age of
the 17th century, North America has experienced a natur-
al variability of 1.5 °C (Bernabo 1981). Similarly, precipi-
tation has shown large year-to-year variability during the
last 2,000 years (Stahle et al. 1988). The Palmer Drought
Index shows abnormally dry or wet periods are more
common than normal precipitation during periods of
change (Stahle et al. 1988).

Given the uncertainty in the predictions, and the natur-
al variability that climate change must exceed before we
can detect effects, what can we say about impacts on the
ecosystem? Two independent analyses of climate change
impact have been completed for the Lake States. Both of
these studies used gap-phase models, and both were
based on GCM predictions of climate for a doubled car-
bon dioxide concentration in the atmosphere. The differ-
ence between these two predictions is that two different
GCM's were used. Solomon and West (1987) used the

Figure 7.— Increases (clear areas) and decreases (scratches) of sum-
mer soil moisture relative to the control when carbon dioxide is dou-
bled (after Kellogg and Zhao 1988).

NCAR model while Botkin et al. (1988) used the GISS
model. These two analyses provide us with some meas-
ure of the simulated ecological sensitivities to climate
change prediction. Differences between the two simula-
tions are that, in one, conifers will be totally replaced
by hardwoods and, in the other, conifers will be retained.
Also, the two differ in the number of trees per hectare,
one showing an increase and the other a decrease.
Similarities are that both simulations show a decrease
in total biomass. The comparison suggests that some con-
fidence may be placed on prediction of total biomass, but
less on the structure and abundance of individual
species.

Ecological Uncertainties in the
Projections of Climate Change

Current projections of the potential impacts of climate
change have limitations because of the omission of some
important processes controlling forest production, par-
ticularly in response to disturbance or stress. In addition,
the physiological responses of individual plants to elevat-
ed carbon dioxide under moisture, temperature, or other
nutrient limitations remain to be definitively described
for natural settings. The fertilization response reported
for some species to elevated carbon dioxide levels may

7

be a short-term or transient response, reflecting an ad-
justment to new and different levels of nutrient availa-
bility. As current forest growth is limited by water and/or
nitrogen, the impact of climate change will involve not
only elevated levels of carbon dioxide, but also changes
in the seasonal distribution of precipitation and temper-
ature, both poorly described variables at the regional
scale in the current GCM's.

Changes in atmospheric temperature and precipitation
will affect soil moisture and soil temperature, two en-
vironmental factors controlling the cycling of nutrients
in the soil. The rate of nitrogen mineralization has been
shown to be positively related to net primary productivity
of trees and wood production (Nadelhoffer et al. 1985)
(fig. 8). Potential increases in the carbon to nitrogen ra-
tios in aboveground plant parts, such as leaves, would
shift the carbon content in litter, resulting in lower qual-
ity litter. Declines in litter quality (less relative nitrogen)
could decrease mineralization rates and, in turn, produc-
tivity of the stand could decline. In other areas, elevated
soil temperature and moisture could enhance soil miner-
alization rates and improve stand productivity. Thus,
shifts in the mineralization rates of nutrients could im-
pact forest productivity.

Individual species migrate at different rates; thus, forest
vegetation will begin to uncouple as we currently know
it and species interactions, which do not currently oc-
cur, will begin to play a role in the development of fu-
ture plant communities. The migration of understory
species, important in early-successional stages of forest
development and for grazing and browsing animals,
could affect the future availability of forage. Potential in-
creased levels of drought will not only reduce vegetation
but provide open niches for invading species from near-
by geographic regions. Changes in climate are projected

1600

S 1400
E

I200H

k_

"° V 1000-

c >,

800H
"g o> 600

400-
200-

— aboveground, total r2 = .72**

— aboveground, perennial tissue r -.63*

— I —

4.0

6.0 8.0 10.0 12.0 14.0 16.0

2 I

Apparent N Uptake (Nu,g m yr )

Figure 8.— Aboveground total net primary production (bole and branch
plus leaf litter) and aboveground perennial tissue (bole and branch)
in relation to annual nitrogen (N) uptake. Symbols designate
dominant genera on sites: P = pine, S = spruce, B = birch, M =
maple, O = oak. Upper and lower cases designate aboveground
production and perennial tissue, respectively. Regression lines
through data were significant at the P < 0.01 level (after Nadelhoffer
et al. 1985).

to be greatest for the mid-latitudes with only a small in-
crease (1°C) in temperature at the equator. Thus, species
diversity in the most diverse plant communities, the
tropics, will change less as a result of projected climate
change than current land use. Species diversity in the
polar and mid-latitudes will likely have the greatest
changes. Potentially significant impacts could occur on
rare species which are currently found only in refuges
located on the basis of their current distribution.

Assessing the impact of resources dependent upon
forest production is even more limited. Numerous spe-
cies of plants and animals are confined to a coincidence
of environmental parameters. If jack pine are replaced by
hardwoods in northern Michigan (Botkin et al. 1988), the
Kirtland's warbler could become extinct. Even though
jack pine would fluorish north of the Great Lakes, there
is a lack of sandy soil north of the Great Lakes and the
Kirtland's warbler nests on sandy soil under jack pine.
Similar situations could arise in mountainous areas for
other species (Peters 1987). For many reptiles, sex of an
individual is determined from temperatures during the
egg incubation period. Thus, climate change will affect
not only the habitat of the animals but also the energet-
ics and reproduction of their populations. Examples ex-
ist where changes in the landscape facilitated an
expansion of species previously occupying only a small
part of the system. The greatest extent of the Kirtland's
warbler occurred after the fire frequency changed dra-
matically in northern Michigan following settlement in
the early 1900 's (Whitney 1989). These examples reflect
our current understanding of the complicated interac-
tions between plants, animals, and environment in the
present ecosystems and suggest subtle interactions not
easily determined for the future climates.

The spatial distribution of vegetation across the land-
scape influences the abundance of wildlife and fish as
well as the use of that land for agricultural and forestry
purposes. It is difficult to determine changes at the land-
scape level resulting from climate change and what these
changes will imply for land use. A very high percentage
of domestic and agricultural water comes from public
lands, particularly in the mountainous West. Increased
temperatures could be expected to cause early and more
rapid snow melt (King et al. 1988). Increased disturb-
ances such as fire will impact the management of that
forest and potentially impact water quality (U.S. EPA
1988). With changed seasonal precipitation, the ratios of
springwood and late wood will vary from those of today,
and frequency of insect and disease outbreaks and the
severity of outbreaks will change. Insect- and disease-
damaged wood is of lower quality (Becker 1987). The
visual impact of forest declines, as well as changing forest
species, will impact the recreational use of forest land.
And finally, the socioeconomic impacts of changing
forest species could include unemployment, community
instability, industrial dislocation, and increased net im-
ports of wood products. Changes in what people will ex-
pect from the forest will need to be addressed for future
climates.

Research to address these questions is ongoing or has
been proposed (Bartuska 1989; Committee on Earth

8

Sciences 1989; Fox and Krebill 1989; Sandberg and Bell
1989; Special Committee for the IGBP 1988; USDA Forest
Service 1988a, 1988b). Generally, the main elements of
these research programs are: biogeochemical dynamics,
ecological systems and dynamics, climate and hydrolog-
ical systems, the interactions between natural processes
and human activities, the study of past natural changes
in earth system history, interactions between the earth's
surface and the atmosphere, hydrosphere, cryosphere,
and biosphere, and solar influences. Forest Service
research is aimed specifically at forest and range systems
and includes the effect of the atmosphere on ecosystems
(changing physical and chemical environments), the ef-
fects of ecosystem change on the atmosphere (biogenic
emission of gasses, land management influences), long-
term changes in ecosystems, and the prediction of
ecosystem responses (USDA Forest Service 1988).

The Future

While we have yet to detect the first signals of green-
house warming, either through direct measurements of
temperature or through impacts on forest ecosystems, we
need to begin preparing for the inevitable changes. Our
policy options are to conserve what we currently have
in forest resources, to develop strategies to mitigate the
effects of climate change, to adapt to change, or some
combination of these three options. Each of these options
raises many questions concerning management actions
and our understanding of forest ecosystems.

The conservation option is undoubtedly the most
difficult to achieve. In those areas where forest produc-
tivity will be significantly reduced, many resources will
be diminished. While we could conserve some elements,
albeit at a high cost, the external force, climate, will ul-
timately prevail. Different ecosystems will evolve in those
areas where the future climate significantly differs from
the current situation. Conservation actions might include
installation of irrigation systems in plantations, or use
of fertilizers to compensate for reductions in growth rates.
Solomon and West (1987) suggest that it is uncertain
whether it would be possible to maintain the net produc-
tivity of commercial forest lands in the United States un-
der climate change. The implementation of such
conservation actions raises a policy question of future
land use. Which forest types should be conserved, if any,
and where should they be conserved? Competition for
land use will be strong because other uses, such as
agriculture or urban area, will also be adjusting to cli-
matic change.

The second option, that of mitigating the effects of cli-
mate change, involves the global community. Energy con-
servation or use of nonfossil fuel energy will slow global
warming. Such actions require a global policy rather than
a local land management policy. Energy conservation or
use of alternate energy sources can control the rate of
greenhouse gases build-up, but cannot reverse the build-
up of greenhouse gases that has already occurred in the
atmosphere. Vegetation production removes carbon di-
oxide from the atmosphere and stores some of it as car-

bon either in wood aboveground or as roots below
ground. Through aggressive reforestation and afforesta-
tion, we can offset some of the anthropogenic trace gases.
To effectively accomplish accelerated tree planting on
nonfederal lands would require close coordination and
cooperation among federal and state forestry profession-
als, consulting foresters, and the tree nursery industry
to ensure adequate supplies of quality trees of appropri-
ate species were available to private landowners and lo-
cal communities. Management questions that need to be
answered include what tree species and where. Sustained
technical assistance would be required to ensure that
proper planting, silvicultural treatments, and tree main-
tenance take place.

The third option, that of adaptation, offers the greatest
flexibility in managing forests in a changing climate.
Adaptive strategies involve developing new technologies
to utilize the resources of the future forest, importing new
industries or businesses which are compatible with the
resources of the future forests, or relocating existing ac-
tivities in anticipation of a changing climate. Adaptive
strategies also include developing or introducing species
which are compatible with the changing climate. Deter-
mining these strategies will involve an examination of
questions such as how much and which forest land
should be managed for timber production, and what kind
of forests. What is the role of federal lands — the location
of which was based on a previous climate — in the
production of resources, such as timber, forage, water
quality and quantity, wildlife, fisheries, and recreation?
What is the role of private lands in timber and forage
production, water quality and quantity, wildlife, fisher-
ies, and recreation? Land management agencies, such as
the Forest Service, are faced with great complexity and
the challenge of developing appropriate data bases and
models that will provide a reliable basis for decisions
about what to do in many different ecosystems and loca-
tions and under various conditions which involve a wide
range of external variables, in addition to the greenhouse
gases.

Because forests are complex ecosystems, and because
uses of the forests are so varied, there is no set formula
which can be prescribed for all forests. Future forest
management will undoubtedly contain elements of all
three options to address the problems arising from global
change. Because of the uncertainties in the current
prediction of impact of climate change on America's
forests, we will need to continue careful monitoring and
surveillance of our forest ecosystems, particularly those
components which are highly sensitive to the greenhouse
effect in order to refine management strategies. Also, be-
cause our current capability to predict impacts is impre-
cise, we must continue to carry out research on the effects
of multiple stresses on our forests in order to assure their
health and productivity in a changing atmospheric en-
vironment.

References

Abramopoulos, F.; Rosenzweig, C; Choudhury, B. 1988.
Improved ground hydrology calculations for global cli-

9

mate models (GCMs). Soil Water Movement and
Evapotranspiration of Climate. 1: 921-941.

Agren, G. I.; McMurtrie, R.; Parton, W. J.; Pastor, J. [In
prep.] State-of-the-art of models of production-
decomposition linkages in conifer and grassland eco-
systems. Paper presented at the SCOPE workshop, The
effects of climate change on production and decom-
position in coniferous forests and grasslands; 1989
April 9-14; Woods Hole, MA.

Bartuska, Ann. 1989. Southern ecosystem health and
productivity in a changing environment. Draft strategic
plan. Mimeo. Raleigh, NC: U.S. Department of Agricul-
ture, Forest Service, Southeastern Forest Experiment
Station. 54 p.

Becker, M. 1987. Quality and utilization of wood from
damaged trees in impact of air pollution damage to
forests for round wood supply and forest products mar-
ket. New York: United Nations: 1-8.

Bernabo, J. C. 1981. Quantitative estimates of temperature
changes over the last 2 ,700 years in Michigan based on
pollen data. Quartenary Research. 15: 143-159.

Bernabo, J. C; Webb, T., III. 1977. Changing patterns in
the Holocene pollen record of northeastern North
America. A mapped summary. Quartenary Research.
8: 64-96.

Botkin, D. B.; Janak, J. F; Wallis, J. R. 1972. Some eco-
logical consequences of a computer model of forest
growth. Journal of Ecology. 60: 849-873.

Botkin, D. B.; Nisbet, R. A.; Keynales, T. E. 1988. Effects
of climate change on forest of the Great Lakes. Final
report to EPA. Santa Barbara: University of California.
40 p.

Bryan, K. 1988. Climate response to greenhouse warm-
ing: The role of the oceans. In: Berger, A., ed. NATO
symposium on climate and geosciences. Dordrecht,
The Netherlands: Reidel Publishing Company.

Carbon Dioxide Assessment Committee. 1983. Synthesis.
In: National Research Council, Changing climate.
Washington, DC: National Academy Press: 5-86.

Cess, R. D; Potter, G. L.; Blanchet, J. P.; Boer, G. J.; Ghan,
S. J.; Kiehl, J. T; LeTreut, H.; Li, Z.-X.; Liang, X.-Z.;
Mitchell, J. F. B.; Morcrette, J.-J.; Randall, D. A.; Riches,
M. R.; Roeckner, E.; Schlese, U; Slingo, A.; Taylor, K.
E.; Washington, W. M.; Wetherald, R. T; Yagai, I. 1989.
Interpretation of cloud-climate feedback as produced
by 14 atmospheric general circulation models. Science.
245: 513-516.

Clark, James S. 1988. Effect of climate change on fire re-
gimes in northwestern Minnesota. Nature. 334: 233-235.

Committee on Earth Sciences. 1989. Our changing planet:
a U.S. strategy for global change research. Washington,
DC: Office of Science and Technology Policy. 38 p.

Darr, David. [In press.] The 1989 RPA assessment of the
forest and range land situation in the United States.
Washington, DC: U.S. Department of Agriculture,
Forest Service.

Davis, M. B. 1981. Quaternary history and the stability
of forest communities. In: West, D.C.; Shugart, H. H.;
Botkin, D. B., eds. Forest succession: concepts and ap-
plication. New York: Springer-Verlag: 132-153.

Davis, M. B., coord. 1988. Ecological systems and dynam-
ics. In: Committee on global change. Toward an un-
derstanding of global change. Washington, DC:
National Academy Press: 69-106.

Dickenson, R. G. 1982. Modeling climate changes due
to carbon dioxide increases. In: Clark, W. C, ed. Car-
bon dioxide review. New York: Oxford University Press:
100-135.

Fifield, R. 1988. Frozen assets of the ice cores. New Sci-
entist. 118: 28-29.

Flather, C. H.; Hoekstra, T. W. 1989. An assessment of
the wildlife and fish situation in the United States:
1989-2040. Gen. Tech. Rep. RM-178. Fort Collins, CO:
U.S. Department of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station.

Fosberg, M. A. 1988. Forest productivity and health in
a changing atmospheric environment. In: Berger, A.,
ed. NATO symposium on climate and geosciences.
Dordrecht, The Netherlands: Reidel Publishing
Company.

Fosberg, M. A. 1989. Climate change and forest fires. In:
Topping, J.C., Jr., ed. Proceedings, Second North
American conference on preparing for climate change;
1988 December 6-8. Washington, DC: Climate Institute:
292-302.

Fox, Douglas G.; Krebill, Richard G. 1989. Forest health
and productivity in a changing atmospheric environ-
ment. The western component: Rocky Mountain and
Intermountain Forest and Range Experiment Stations.
Mimeo. Fort Collins, CO: U.S. Department of Agri-
culture, Forest Service, Rocky Mountain Forest and
Range Experiment Station. 16 p.

Fried, J. S.; Torn, M. S. [In prep.] The altered climate fire
model simulating the effects of climate change on the
effectiveness of a wildland fire initial attack program.
Berkeley, CA: University of California.

Friedli, H.; Lotscher, H.; Oeschger, H.; Siegenthaler, U;
Stauffer, B. 1986. Ice core record of the 13C/12C ratio
of atmospheric C02 in the past two centuries. Nature.
324: 237-238.

Hansen, J.; Fung, I.; Lacis, A.; Lebedoff, S.; Rind, D.; Re-
udy, R.; Russell, G.; Stone, P. 1987. Prediction of near
term climate evolution: what can we tell decision mak-
ers now? In: Proceedings, First North American con-
ference on preparing for climate change; 1987 October
27-29. Washington, DC: Government Institutes, Inc.:
35-47.

Haynes, Richard. [In press.] An analysis of the timber sit-
uation in the United States: 1989-2040. Gen. Tech. Rep.
Fort Collins, CO: U.S. Department of Agriculture, Forest
Service, Rocky Mountain Forest and Range Experiment
Station.

Herring, J. R. 1985. Charcoal fluxes into sediments of the
North Pacific ocean: The Cenozoic record of burning.
In: The carbon cycle and atmospheric C02: Natural
variations archean to present. Geophysical Monograph
32. Washington, DC: American Geophysical Union:
419-442.

Hoffman, J. S.; Wells J. B. 1987. Forests: past and project-
ed changes in greenhouse gases. In: Shands, W. E.;

10

Wells, J. B., eds. The greenhouse effects, climate
change, and U.S. forests. Washington, DC: The Con-
servation Foundation: 19-41.

Holdridge, L. R. 1964. Life zone ecology. San Jose, Costa
Rica: Tropical Science Center.

Houghton, R. A.; Boone, R. D.; Fruci, J. R.; Hobbie, J.
E.; Melillo, J. M.; Palm, C. A.; Peterson, B. J.; Shaver,
G. R.; Woodwell, G. M.; Moore, B.; Skole, D. L.; My-
ers, N. 1987. The flux of carbon from terrestrial
ecosystems to the atmosphere in 1980 due to changes
in land use: geographic distribution of the global flux.
Tellus. 39B: 122-139.

Houghton, R. A.; Hobbie, J. E.; Melillo, J. M.; Moore, B.;
Peterson, B. J.; Shaver, G. R.; Woodwell, G. M. 1983.
Changes in the carbon content of terrestrial biota and
soils between 1960 and 1980: A net release of COz to
the atmosphere. Ecological Monographs. 53: 235-262.

Karl, T; Jones, P. D. 1989. Urban bias in area averaged
surface temperature trends. Bulletin of American
Meteorological Society. 70 : 265-270.

Keeling, C. 1984. Atmospheric COz concentration at
Mauna Loa Observatory Hawaii, 1958-1983. Rep.
NDP-011. Washington, DC: U.S. Department of Energy.

Kellogg, W. W.; Zhao, Zong-ci. 1988. Sensitivity of soil
moisture to doubling of carbon dioxide in climate
model experiment of climate. Part 1: North America.
Journal of Climate. 1: 348-366.

King, G. A.; DeVelice, R. L.; Neilson, R. P.; Worrest, R.
C. 1988. California. In: Smith, J. B.; Tirpak, D. A., eds.
The potential effects of global climate change on the
United States. Vol. 1, chapter 4. Washington, DC: U.S.
Environmental Protection Agency. 59 p.

Leverenz, J. W.; Lev, D. J. 1987. Effects of carbon dioxide-
induced climate changes on the natural ranges of six
major commercial tree species in the western United
States. In: Shands, W. E.; Hoffman, J. S., eds. The
greenhouse effect, climate change, and U.S. forests.
Washington, DC: Conservation Foundation: 123-156.

McCracken, F; Schipper, A. L.; Widin, K. D. 1984. Ob-
servations on occurrence of cottonwood leaf rust in the
central United States. European Forestry. 14: 226-233.

Miller, W. F; Dougherty, P. M.; Switzer, G. L. 1987. Ef-
fects of rising carbon dioxide and potential climate
change on loblolly pine distribution, growth, survival
and productivity. In: Shands W. E.; Hoffman, J. S., eds.
The greenhouse effect, climate change, and U.S. forests.
Washington, DC: Conservation Foundation: 157-189.

Mintzer, L. 1987. Energy policy and the greenhouse prob-
lem: A challenge to sustainable development. In:
Proceedings, First North American conference on pre-
paring for climate change. Washington, DC: Govern-
ment Institutes, Inc.: 18-34.

Mitchell, J. F. B. 1988. Climate sensitivity: Model depen-
dence of results. In: Berger, A., ed. NATO symposium
on climate and geosciences. Dordrecht, The Nether-
lands: Reidel Publishing Company.

Nadelhoffer, K. J.; Aber, J. D.; Melillo, J. M. 1985. Fine
roots, net primary production, and soil nitrogen avail-
ability: a new hypothesis. Ecology. 66: 1377-1390.

National Research Council. 1983. Changing climate.
Washington, DC: National Academy Press. 496 p.

Nordhaus, W. D; Yohe, G. W. 1983. Future carbon diox-
ide emissions from fossil fuels. In: National Research
Council, changing climate. Washington, DC: Nation-
al Academy Press: 87-185.

Pastor, J.; Post, W. M. 1988. Response of northern forests
to C02-induced climate change. Nature. 334:55-58.

Peters, R. L. 1987. Effects of global warming on biodiver-
sity: An overview. In: Proceedings, First North Ameri-
can conference on preparing for climate change.
Washington, DC: Government Institutes Inc.: 169-185.

Ramanathan, V.; Cess, R. D; Harrison, E. F; Minnis, P.;
Barkstrom, B. R.; Ahmad, E.; Hartmann, D. 1989.
Cloud-radiative forcing and climate: Results from the
earth radiation budget experiment. Science. 243:
57-63.

Sandberg, David; Bell, Enoch. 1989. Ecosystem health
and productivity in a changing environment. Draft
strategic plan. Mimeo. Berkeley, CA: U.S. Department
of Agriculture, Forest Service, Pacific Southwest Forest
Experiment Station.

Schlesinger, M. G. 1988. Model projections of the climat-
ic changes induced by increased atmospheric C02 cli-
mate and geosciences. In: Berger, A., ed. NATO
symposium on climate and geosciences. Dordrecht,
The Netherlands: Reidel Publishing Company.

Schneider, S. H. 1988. The greenhouse effect: What we
can or should do about it? In: Proceedings, First North
American conference on preparing for climate change.
Washington, DC: Government Institutes, Inc.: 18-34.

Solomon, A. M.; Tharp, M.L.; West, D. C; Taylor, G. E.;
Webb, J. W; Trimble, J. L. 1984. Response of un-
managed forests to C02-induced climate change: avail-
able information, initial tests, and data requirements.
DOE/NBB/0053 TR009. Washington, DC: U.S. Environ-
mental Protection Agency.

Solomon, A. M.; West, D. C. 1987. Simulating forest
ecosystem responses to expected climate change in
eastern North America: Applications to decision mak-
ing in the forest industry. In: Shands, W. E.; Hoffman,
J. S., eds. The greenhouse effect, climate change and
U.S. forests. Washington, DC: The Conservation Foun-
dation: 189-218.

Special Committee for the IGBP 1989. Global change.
Report No. 4 1988. Stockholm, Sweden: IGBP Secretari-
at, Royal Swedish Academy of Sciences, Box 50005,
S-10405. 200 p.

Stahle, D. W.; Cleveland, M. K.; Hehr, J. G. 1988. North
Carolina climate changes reconstructed from tree rings:
AD 372 to 1985. Science. 240: 1517-1519.

U.S. Department of Agriculture, Forest Service. 1988a.
Forest health and productivity in a changing at-
mospheric environment. Mimeo. Broomall, PA: U.S.
Department of Agriculture, Forest Service, North-
eastern Forest Experiment Station. 42 p.

U.S. Department of Agriculture, Forest Service. 1988b.
Forest productivity and health in a changing atmospher-
ic environment. Unnumbered report. Washington, DC:
U.S. Department of Agriculture, Forest Service, Forest
Fire and Atmospheric Sciences Research Staff. 56 p.

U.S. Environmental Protection Agency. 1988. The poten-
tial effects of global climate change on the United States.

11

Vols. 1-2. Draft report to Congress. Washington, DC. 267
+ 395 p.

Whitney, Gordon. 1989. Seminar presented at Ecosystem
Center, Marine Biological Lab, Woods Hole, MA. Cam-
bridge, MA: Fellow, Harvard University.

Wingert, C. H. 1988. Forest management strategies to ad-
dress climate change. In: Proceedings First North
American conference on preparing for climate change.
Washington, DC: Government Institutes, Inc.: 328-333.

Winjum, J. BC; Neilson, R. P. 1988. Forests. In: Smith, J.
B.; Tirpek, D. A., eds. The potential effects of global cli-
mate change in the United States. Vol. 2, chapter 8.
Washington, DC: U.S. Environmental Protection Agen-
cy. 38 p.

Woodward, F. I. 1987. Climate and plant distribution. Cam-
bridge: Cambridge University Press. 174 p.

Woodwell, G. M. 1983. Biotic effects on the concentration
of atmospheric carbon dioxide: a review and projection.
In: National Research Council, changing climate.
Washington, DC: National Academy Press: 216-241.

12

Joyce, Linda A.; Fosberg, Michael A.; Comanor, Joan M. 1990. Climate
change and America's forests. Gen. Tech. Rep. RM-187. Fort Collins,
CO: U.S. Department of Agriculture, Forest Service, Rocky Moun-
tain Forest and Range Experiment Station. 12 p.

Projections in the Forest Service Assessment assume a future in which
changes in timber production and land use are not abrupt discontinui-
ties from the past. These assumptions may not be met if the earth's cli-
mate changes rapidly. This document summarizes the current research
on the impacts of climate change on America's forests.

Keywords: Greenhouse effect, timber assessment, ecological response

Rocky
Mountains

Great
Plains

U.S. Department of Agriculture
Forest Service

Rocky Mountain Forest and
Range Experiment Station

The Rocky Mountain Station is one of eight
regional experiment stations, plus the Forest
Products Laboratory and the Washington Office
Staff, that make up the Forest Service research
organization.

RESEARCH FOCUS

Research programs at the Rocky Mountain
Station are coordinated with area universities and
with other institutions. Many studies are
conducted on a cooperative basis to accelerate
solutions to problems involving range, water,
wildlife and fish habitat, human and community
development, timber, recreation, protection, and
multiresource evaluation.

RESEARCH LOCATIONS

Research Work Units of the Rocky Mountain
Station are operated in cooperation with
universities in the following cities:

Albuquerque, New Mexico

Flagstaff, Arizona

Fort Collins, Colorado*

Laramie, Wyoming

Lincoln, Nebraska

Rapid City, South Dakota

Tempe, Arizona

'Station Headquarters: 240 W. Prospect Rd., Fort Collins, CO 80526