Productivity of America's forests and climate change

78 Pages·1995·5.3 MB·English

by  Joyce, L. A

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

©
D"r
Productivity of America's Forests
of
A
gr,cu .ure Climate Change
ForestService
Rocky Mountain
Forest and Range
ExperimentStation
Abstract
Joyce, Linda A., editor. 1995. ProductivityofAmerica's forests
and climate change. Gen. Tech. Rep. RM-271. Fort Collins, CO:
U.S. DepartmentofAgriculture, ForestService, RockyMountain
Forest and Range Experiment Station. 70 p.
This study assessed the impact of climatic change on the
forestry sector and carbon storage on timberlands. Productivity
changesreflectedecosystemsensitivitiestotemperature, precipi-
tation, and nutrient availability. The largest increases in NPP
occurred in the northerly ecosystems, and the least response to
climate change occurred in timber management types in the
southern regions. Increased timber inventories decreased prices,
imports, and shifted production to low cost regions. In this
analysis, only the most optimistic scenario shifts future forests
from sources of carbon to a carbon sink. The strong demand for
woodproductsinthefuturedampensanypositivegrowtheffects
on forests, for all but the maximum scenario.
Keywords: timber supply and demand, productivity, climate
models, carbon storage
"The United States Department of Agriculture (USDA) prohibits dis-
crimination in its programs on the basis of race, color, national origin, sex,
religion,age,disability,politicalbeliefs,and maritalor familialstatus. (Not
all prohibited bases apply to all programs.) Persons with disabilities who
requirealternativemeansofcommunicationofprograminformation(braille,
large print, audiotape, etc.) should contact the USDA Office ofCommuni-
cations at [phone].
"To filea complaint, write theSecretaryofAgriculture, U.S. Department
ofAgriculture,Washington, DC20250,orcall [phone](voice)or(202)
720-1127 (TDD). USDA is an equal employment opportunity employer."
This publication was printed on recycled paper.
USDA Forest Service September 1995
General Technical Report RM-271
Productivity of America's Forests and Climate Change
Editor:
Linda A. Joyce, Range Scientist,
Rocky Mountain Forest and Range Experiment Station1
Contributing Authors:
R.A. Birdsey, R.W. Haynes, LS. Heath, LA. Joyce, A.D. McGuire, W. Massman, and J.R. Mills
1Headquarters isin FortCollins, in cooperation with ColoradoState University.
Contents
Page
Executive Summary 1
1. Climate and Climate Modeling 3
W. Massman
C0
2. Responses of Net Primary Production to Changes in and Climate 9
2
A.D. McGuire and L.A. Joyce
3. Influence of Climate Change on Supply and Demand for Timber 46
J. R. Mills and R. W. Haynes
4. Carbon Changes in U.S. Forests 56
R. A. Birdsey and L. S. Heath
Productivity of America's Forests and Climate Change
Linda A. Joyce, RangeScientist, Editor
Contributing Authors:
R.A. Birdsey, R.W. Haynes, L.A. Heath, L.A. Joyce, A.D. McGuire, W. Massman, and J.R. Mills
EXECUTIVESUMMARY
MarketModel (TAMM), whichprovidesprojections
Increasing levels of atmospheric carbon dioxide of timber and wood product prices, consumption,
are expected to alter climate, with potential changes and production trends. For the economic baseline,
inforestproductivity and timbersupply. Thisstudy weused the 1993 Forest Service AssessmentUpdate
analyzestheimpactsofelevated carbondioxideand assumptions about population, economic activity,
climate changeon forested ecosystems, and theeco- andincome,technologicalandinstitutionalchanges,
nomic feedbacksonharvestpatterns and vegetation energy costs, capital availability, and public and
change onprivate timberlands, in the United States. private investments in forest management, utiliza-
The analysis used a framework that linked general tion, and research. Carbon storage and flux from
circulationmodeloutput,anecosystemmodel(TEM), forestlandswasestimatedwithFORCARB,whichis
models of the forest sector (TAMM/ATLAS/ tightly linked to the timber inventory model (AT-
NAPAP), and a carbon accounting model LAS). Although FORCARB includes all forest eco-
(FORCARB). Future climates were described with system components, only carbon in trees is pre-
output from four different general circulation mod- sented in the climate change scenarios. The projec-
els (GCMs): one from the Oregon State University tion period for this analysis is 50 years.
(OSU) model, one from the Goddard Institute of The effects ofclimate change in productivity var-
SpaceSciences (GISS) model, and two from theGeo- ied by timber management type within each region.
physical Fluid Dynamics Lab model (GFDL-1, and The largest increases in NPP occurred in the north-
GFDL-Q).Thefourfutureclimatesencompassglobal erly ecosystems, and the least response to climate
temperature increases of2.8°C to 4.2°C and precipi- changeoccurred in timber management types in the
tation increases from 7.8% to 11%. southern regions. Because most of the productivity
Thecontemporaryclimatedataandthefourfuture changeswerepositive,inventoriesincreasedformost
climates were used as input for the Terrestrial Eco- of the scenarios. All softwood forest types show
system Model (TEM) to estimate changes in net gains, while the hardwoods gain in all but the mini-
primary productivity (NPP) for temperate and bo- mum scenario; growth declines for southern hard-
real forests in the United States. The response to woods under the minimum scenario. These inven-
climate change was computed as the change in the tory changes reflect not only the changes in produc-
NPP response between the future climate runs and tivity, but also changes in timberland area and
thecontemporary climate run. Three scenarios were changes in harvest.
developed:anaverage,a minimum,anda maximum The increased productivity impacts harvest to-
productivity response for each timber management wardstheendoftheprojectionperiod.Growthcom-
type,withineachtimbersupplyanddemand region. pounds quickly and, by the end of the projection, is
Under the IPCC Business-as-usual scenario, car- 5% to 24% above the base (minimum and maximum
bon dioxide doubles(625 ppm) in about 2065. The projections). While the corresponding increase in
productivity responses were implemented gradu- inventory pushes stumpage prices lower (roughly
ally in the timber inventory model, the Aggregate 6% to 35%), harvest increases by only 1% to 3%.
Timberland Assessment System model (ATLAS); Consumptionlagsthegrowthand inventory fortwo
changes in productivity always were less than 1% reasons. First, early in the projection, mostgrowthis
peryear. ATLASislinked to theTimber Assessment on trees too young to harvest, and second, it takes
1
«
time to build processing capacity to take advantage Between 1952 and 1992, carbon storage on forest-
of more abundant resources. But given this, the land in the coterminous United States increased by
smaller gain in harvest is an indication that the 11.3 billion metric tons, an average of 281 millon
primary determinants of timber demand are items metric tons/year, an amount that offset about one
other than wood costs. quarter of U.S. emissions of carbon for ihe period.
Witha changeinavailablesuppliesofharvestable Most of the historical increase in carbon storage has
inventories,thereisaredistributionofharvestacross been on the private timberland analyzed in this
the United States, as demand in various regions study. Baseprojectionsforcarbonstorageontimber-
adjuststotakeadvantageoflowercostrawmaterials. landthrough2040showadeclineof20millionmetric
Thechangealsoisbetweensoftwood andhardwood tons peryearon timberland at the end ofthe projec-
fiber types, and between forest industry and the tion period. Responses with a minimum increase in
nonindustrial private ownerships. For example, as productivity aresimilar to the baseline, but increase
supply of softwood growing stock increases and more than 40 million metric tons per year with a
pricesdecline, theamountofhardwoodsused inthe maximum increase in productivity.
harvest mix declines. This increases the demand for This study presents a methodology to examine the
softwood pulpwood while reducing demand for effectsofmarketforcesonpotentialeffectsofclimateon
hardwood pulpwood. Increased softwood timber temperate and boreal forests in the United States, as
supplies also act to increase softwood sawtimber well as the potential shifts in carbon storage in those
capacity,andlowersoftwood lumberpricesresultin same forests. In this analysis, only the most optimistic
some increases in lumber production in the South. scenarioshifts futureforests from sourcesofcarbon to
Increased production also increases the amount of acarbonsink.Thestrongdemandforwoodproductsin
residues available for pulp, further lowering hard- the future dampens any positive growth effects on
wood pulpwood harvest. Differences in resource forests, for all but the maximum scenario. Market fac-
managementbetweentheNorthandSouthalsoplay tors need tobeconsidered inany analysisoftheuseof
a major role in the hardwood projections. forests for mitigating carbon emissions.
2
t
Climate and Climate Modeling
1.
W. Massman2
INTRODUCTION 1990). Consequently,otherrefinementstothisfigure
canbeexpectedasunderstandingoftheearthsystem
The possibility of climate change has been the improves. In addition to representing climate and
subject ofmany discussions and speculations in the the earth system, this diagram also can be used to
past few years. An awareness also has emerged that describe many smaller scale climate-driven ecosys-
human activities may contribute to these changes, tems. On virtually all scales, the processes and cou-
and thata continuationofthese changes could dras- plings between all the major components of this
tically alter humansocial and natural environments system are the same, but the strength of these pro-
(IPCC 1990). At present, it is not possible to accu- cesses and couplings may vary significantly from
ratelypredicthowclimatewillchangeorwhatfuture one scale to another and from one ecosystem to
global environments will be. Nevertheless, some another. The complexity or richness of climate de-
modelingtools(althoughimperfect)areavailablefor rives from the enormous variety of linkages and
developing climate scenarios that can be used to feedbacks between these components and from the
address issues concerning the impacts that climate nonlinear nature of their interactions in time and
change may have on ecosystems. This chapter pro- space.
vides a brief discussion of climate and some of the Thescienceofclimatemodelingbeganwithatmo-
complexities of climate modeling, and provides a spheric general circulation models (AGCMs or
brief discussion of the four GCMs used in other GCMs). These models, originally developed for
chapters of this report. weather forecasting, are numerical representations
Climate, as presently understood, is just one as- oftheatmospheric componentoftheclimate system
pect of the total earth system. The earth's climate is coupled with relatively simpler models for the hy-
driven primarily by solar radiation and the earth's drosphere, the cryosphere, the biosphere, and the
rate of rotation, and results from the interactions land surface. Like the climate system, these models
betweenthefivecomponentsoftheearthsystem: the are also complex, nonlinear, and dynamic. Using
atmosphere, the hydrosphere, the cryosphere, the equations representing the conservation of mass,
biosphere, and the land surface. Within this system, energy, and momentum, GCMs attempt to predict
physical, chemical, and biological processes are the state of the global climate. They are used to
coupled on virtually all temporal and spatial scales. simulate (at a discrete number of points distributed
Someofthecomplexity ofthissystem isrepresented in the vertical and the horizontal) a variety ofatmo-
by figure1.1,whichshowsa fewoftheprocesses tha sphere-related climate variables. Someof thesevari-
externally force the climate and that internally link ables include wind speed and direction, the atmo-
thefiveearthsystemcomponents. Butfigure 1.1 also spheric temperature and pressure, cloudiness, pre-
indicates thatourunderstanding ofdynamics ofthe cipitation, and the energy and water balance of the
climate and of the earth system is still evolving. For land surface. However, a GCM is only one type of
example, the original diagram (GARP 1975) from climate model. More sophisticated climate models
which this figure is adapted does not specifically coupleaGCMwithadynamic(andequallycomplex)
include methane (CH4) or chlorofluorocarbons ocean model that then can be used to simulate the
(CFCs) as important atmospheric trace gases. But, ocean-related climate variables (i.e., the ocean sur-
now we know that the combined effect of CH4 and face temperatures, the salinity of the ocean, and
CFCs is nearly as important forglobal warming asis position of the ocean currents). Ultimately a global
carbon dioxide (Rasmussen and Khalil 1981, IPCC climate model will require coupling detailed, com-
2Rocky Mountain Forest and Range Experiment Station, Fort plex and dynamic modelsofeachoftheearthsystem
Collins. CO. components (OIES 1992). However, regardless of
3
CLIMATE COMPONENTS: atmosphere, hydrosphere, CLIMATE PROCESSES: transpiration, evaporation,
cryosphere, land surface, biomass burning, precipitation,
biosphere
photosynthesis, heat exchange,
weathering of rocks, wind,
surface water runoff, etc.
Changes in tidal or Changes in solar
gravitational forces ii radUition SPACE
ATMOSPHERE
Terrestrial
1
radiation i
N2, O2, H2O, CO2, 03, CH4, CFCs,etc. Aerosols Clouds
Atmosphere Atmosphere
land surface ocean biomass
coupling coupling
A
Atmosphere
biomass Atmosphere Atmosphere
coupling ocean ice
coupling coupling
SNOW &
ICE SHEETS
BIOMASS
+ SOILS
LAND
EARTH
Changes
in
atmospheric
composition
Changes In land features,
obroigorgaepohcyh,emviecgae*ltactyicolni,nga,lbeetdco., Chsahnagpees, isnaloincietay,nebtacs.in
Figure 1.1—Schematic ofearth system. Solid arrows indicate forcing by external processes. Dashed arrows indicate forcing by internal
processes. Double arrows indicate internal coupling or two-way forcing. Processes most directly influenced by human activity are
denoted by an asterisk (*).
theircomplexityand sophistication, climatemodels, system hasnotbeenachieved. Several majorhurdles
like all mathematical models, will always be simpli- must be overcome before any model can be used
fied versions of the full climate system. confidently for predicting ecosystem response to
However, the science of global climate modeling climate change. These hurdles include the limita-
will continuetochangeand improveas understand- tions imposed by coarse spatial resolution, the in-
ing of the physical and numerical aspects of earth ability to accurately model the effects of clouds, the
system modeling improve. To date, GCMs have inability to accurately model the global hydrologic
shown to be extremely powerful tools for studying cycle, the lackofan interactive biosphere model, the
the earth's climate system. Furthermore, GCMs also lack of an interactive human-climate model, the in-
may prove useful for distinguishing between alter- herentcomplexityoftheclimatesystem,and thelack
natestrategiesforhumanresponsetoclimatechange. of knowledgeofhow to interactively coupleseveral
But a fully integrated numerical model of the earth dynamic nonlinear models.
4