Forest Carbon in the United States - 2007 Updated Edition

8/14/2019 Forest Carbon in the United States - 2007 Updated Edition

1/56

LAURIE A. WAYBURN, JERRY F. FRANKLIN, JOHN C. GORDON,CLARK S. BINKLEY, DAVID J. MLADENOFF, NORMAN L. CHRISTENSEN, JR.

ForestCarbonin the United States:

Opportunities & Options for Private Lands


2/56

The Pacific Forest Trust (PFT) is the only non-profitorganization solely dedicated to protecting, enhancingand restoring Americas private working forests for all

their public benefits. Since 1993, PFT has been implementing acomprehensive conservation strategy to Retain, Sustain and Gainthrough education, innovation and collaboration. PFT workswith landowners, forest managers, public agencies, localcommunities and others to sustain working forests for thewealth of goods and services they provide wood, water, wildlifeand a well-balanced climate.

PFT is a leading advocate for the climate benefits of forestsand works with landowners to demonstrate that conservation

and stewardship can deliver both economic and ecologicalrewards. PFT leads regional and national efforts to enact climatechange policies that incorporate forest conservation with market-based incentives to reduce net carbon dioxide emissions.

The Pacific Forest Trust believes maintaining long-term,ecologically based productivity is the key to forest preservation.Working forests will be preserved only if they remain productive;and they will produce only if they are preserved.

2000, 2007 Pacific Forest TrustAll rights reserved. Reproduction in wholeor in part permitted with attribution.

Cover photo by Larry Ulrich


3/56

LAURIE A. WAYBURNJERRY F. FRANKLIN

JOHN C. GORDON

CLARK S. BINKLEY

DAVID J. MLADENOFF

NORMAN L. CHRISTENSEN, JR.

Forest Carbonin the United States:Opportunities & Options for Private Lands


4/56

Contents

List of Figures and Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .iiiAcknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ivIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ivEmerging Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .v

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3Accounting Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5Ecological Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14Recent Trends in US Private Forest Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16Regional Implications

Pacific Northwest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20The Great North Woods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24Southeastern Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29

Economic Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43About the Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Inside Back Cover


5/56

Figures and Tables iii

Figure A: Managing Working Forests to Optimize Climate Benefits . . . . . . . . . . . . . . . . . . . . . . . . .vii

Figure B: How Carbon Dioxide Flows in Forests: Stores, Emissions & Reductions . . . . . . . . . . . .viii

Figure 1: Carbon Pools by Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

Figure 2: Forest Carbon Flux After Harvest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

Figure 3: Labile Forest Carbon: Growth, Harvest, Decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7

Figure 4: Effects of Harvest and Fires on Forest Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8

Figure 5: Forest Carbon During Timber Harvests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

Figure 6: Carbon Gains from Reforestation, Forest Management, and Conservation . . . . . . . . .13

Figure 7: Managing Forests for Greater Carbon Stock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

Figure 8: Regional Comparison of Forest Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

Figure 9: Trends in Private Forestland Gain and Loss (1982-1997) . . . . . . . . . . . . . . . . . . . . . . . . . . .17

Figure 10: Trends in Forest Carbon Stores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

Figure 11: Incremental Land Value for Medium-site Lobolly Pine in the South . . . . . . . . . . . . . . .38

Figure 12: Incremental Land Value for Medium-site Douglas Fir in the Pacific Northwest . . . . .40

Table A: Global Norms for CO2 Emissions Reductions Projects (CERs):Californias Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

Table 1: Carbon Gains and Losses Due to Growth and Harvest in 1996 . . . . . . . . . . . . . . . . . . . . . .19

Table 2: Growth Versus Harvest (Softwoods and Hardwoods) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19Table 3: Statistics for Private Timberland in Western Oregon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

Table 4: Trends in New England Carbon Stocks (1952-1992) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25

Table 5: Optimal Rotation for Medium-site Loblolly Pine in the South . . . . . . . . . . . . . . . . . . . . .36

Table 6:. Optimal Rotation for Medium-site Douglas-fir in the Pacific Northwest . . . . . . . . . . . .37

Table 7: Incremental Land Value for High-Site Lobolly Pine in the South(% increase over no-carbon case) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37

Table 8: Incremental Land Value for Medium-Site Lobolly Pine in the South(% increase over no-carbon case) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38

Table 9: Incremental Land Value for High-Site Douglas-fir in the Pacific Northwest(% increase over no-carbon case) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39

Table 10: Incremental Land Value for Medium-Site Douglas-fir in the Pacific Northwest(% increase over no-carbon case ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40

Table 11: Opportunity Cost Analysis of a High-site Douglas-fir Stand . . . . . . . . . . . . . . . . . . . . . . .41

Figures and Tables


6/56

iv Forest Carbon in the United States

AcknowledgmentsThis project has been made possible bythe generous financial support of theCatherine T. and John D. MacArthurFoundation, as well as the Wallace GlobalFund. Particular thanks to Melissa Dannfor her continuing support, and to DanMartin for his early and thoughtful enthu-siasm for our work.

Dr. Mark Harmon of Oregon StateUniversity provided essential advice and

review of the report. Dr. Peter Frumhoff of the Union of Concerned Scientistsalso reviewed the report. Constance Bestof the Pacific Forest Trust providedinvaluable testing of various assump-tions. Jennifer ODonnell of the PacificForest Trust was instrumental in editingand production. Jay Chamberlin of thePacific Forest Trust provided key supportin the early stages of this project.

There is no doubt that forests affect thenationaland globalcarbon budgets.Forest loss, gain, or changes in manage-

ment lead to significant gains and lossesin carbon stocks and atmospheric levelsof carbon dioxide. While there is agree-ment that forests are important to car-bon balances, there has been much dis-cussion, indeed controversy, over thepotential role of forest management andconservation in reducing net US emis-sions of carbon dioxide. Concerns haveranged from fears that the US will offsetall of its emissions through forestry and

forest conservation, to fears that a creditsystem, called for by some industrygroups, might lead to very short-termand highly uncertain gains creditingbusiness-as-usual. Further, characteri-zations of the accuracy and certainty of accounting for changes in carbon stocksgained or lost from forestry and forestconservation have often been misleading.

This report is intended to describe therange of potential forest conservationand management activities on private

forests that can lead to net increases incarbon stocks. It identifies and illustratescauses of forest carbon loss, outlines theprinciples and mechanisms for forestaccounting on a practical level, and pro-poses potential options for reversingexisting trends toward forest lossthrough market mechanisms to createchanges to business-as-usual. It furtherexplores how market systems can pre-vent forestland loss and increase forest

age, benefiting many biodiversity inter-ests. The focus of this report is on privateforests, as they make up the majority of US forests and are the ones most at riskfrom loss and conversion; however, manyof the principles concerning forest man-agement apply equally to public forests.

Introduction


7/56

Since this publication was first published, ashort six years ago, several key new issueshave emerged in the discussions about the

role of forests in climate change:

The clarity and immediacy of climatechangeas a reality

The need to manage forests for resiliencyand adaptation

The recognition that forest-producedwoody biomass could become a very signifi-cant part of our nations energy portfolio;

With appropriate policies, forests can, andshould, play a significant role in reducingnet carbon dioxide (CO2) emissions

Initial market responses show major economicreturns associated with managing forests toincrease their climate benefits. Furthermore,these benefits are complementary to sustainableproduction of timber and other forest products.

Past uncertainty regarding whether climatechange is occurring and its connection to ongoingand increased emissions of carbon dioxide is virtu-

ally gone. Current discussion centers on how toavoid the most intense disruptions that climatechange will bring. Hence, the relevance of work-ing with forests to rapidly and significantly reducenet atmospheric levels of CO2 is only greater andmore immediate.

Climate change will likely stress forest ecosys-tems, reinforcing the importance of forest man-agement that ensures will remain stable carbonbanks. This underscores the importance of incorporating management for resilience andadaptation in forest-based carbon emissionsreduction efforts. It also reinforces the need toencourage more naturalness in forest manage-ment. The complexity and diversity of naturalforests contributes enormously to their robust,resilient and stable character. The synergy andfeasibility of managing for both restoring natu-ralness and increasing net carbon (as well as for

increased timber stocks) is well illustrated inFigures A and 7. Thus, managing for stableclimate benefits in our forests also provides manyco-benefits for habitats, biodiversity, and water-shed function, as well as for long-term sustain-able supplies of timber products.

The efficiency and economic potential of woody biomass as a new, sustainable energy feed-stock is now foreseeable. As oil prices reach andexceed $80 per barrel, most energy experts agreethat the production of wood-based biomass energybecomes competitive with other alternativeenergy production, promising a new source of return for forest landowners. If such woodybiomass capacity is developed in the context of adaptive management and ensured stability of long-term forest carbon banks, positive syner-gistic effects are likely. On the other hand, if asingle-focus management approach is pursued,one that solely manages for woody biomass,synergistically negative impacts on the stabilityof forests are more likely.

More traditional forests products also have

climate benefits. High value, structural woodproducts that can substitute for more CO 2-inten-sive building materials like steel or aluminum alsohave a significant role to play. Reversing currenttrendswhere steel and aluminum have beenreplacing woodwould deliver a net CO 2 benefitin the forest, and outside of it. Further, managingour forests to remove excess small diameter mate-rials for biomass can help restore more natural,resilient and complex forest conditions. Doing soremoves excess fuels loads and releases theremaining stands to grow more fully and natural-ly. Forest products are not net new pools of carbon. Rather, they represent a transfer from oneforest pool of carbon where they are producedto another product pool where they decay.Nonetheless, products can still provide for a netsubstitution of other materials whose productioncauses more CO2 emissions (see Figure B).

CARBON REPORTv

Emerging Issues


8/56

When this report was first written, the carbonmarket was nascent. We speculated then on thepricing and demand that might occur as themarket developed could create positive changein forests. Six years later, there is a strong andgrowing global climate market, one with 2006revenues exceeding $30 billion and over $40billion forecasted for 2007. This market scalebrings real economic promise to drive significantchange in trends causing forestand carbon loss and depletion by creating new revenues basedon the conservation management of forest forclimate benefits.

There has also been a corresponding emer-gence of global norms (Table A) for projects thatcreate CO2 emissions reductions (CERs). Thesenorms, which concern the permanence, addi-tionality, and co-benefits of CERs, are linkedback to the principles articulated in the Kyotoprotocol. They seek to ensure the durability of net climate benefits, as well as to sustain otherforests benefits including biodiversity, water-sheds and rural economic livelihoods.

Another compelling development is the num-ber of US states that have risen to the challenge of addressing climate change and that are now seek-ing to benefit from the rapidly growing carbonmarketplace. In the absence of national action,

states are developing policies and regulations thatenable new climate-friendly business to take aleadership position in the global marketplace.California, with its landmark laws developingfirst a state-backed registry (2000-2001) and thenCO2 emissions caps (2006), was the first to developa comprehensive system that meets the expecta-tion of the Kyoto Protocol and incorporatesforests into that system (Table A). States underthe Regional Greenhouse Gas Initiative in NewEngland and the mid-Atlantic are now also devel-oping such systems and seeking to link themacross the continent. The Multi-State Registry(MSR), based on Californias Climate ActionRegistry, has 38 members as of this printing.

Such state actions are both necessary andhopeful as we seek to curb forest loss and deple-tionthe source of more than 40% of excess and to restore the significant forest carbon banksthat are key to stabilizing and then reducingCO2 levels. These action create the foundationfor a national, and possibly global, approach toleveraging the natural potential forests have

to sustain our climateif, of course, we firstsustain our forests.

Laurie WayburnAugust 2007

vi Forest Carbon in the United States


9/56

CARBON REPORTvii

Generalized forest carbon stores over time for U.S. forests

V O L U M E

TIME

Carbon stores at timeof business-as-usualtimber harvest dictatedby current market forces

Carbon stores at timeof timber harvest tooptimize climate benets

Revenue fromcarbon market buystime to allow treesto grow older and store more carbon

MANAGING WORKING FORESTS TO OPTIMIZE CLIMATE BENEFITSFigure A

ADDITIONALITYActions above standard, commonly accepted baselines;projects exceeding forestry laws and land use regulations

PERMANENCEEnsure projects durability and minimize risks through conservation easements

MINIMIZE AND AVOID LEAKAGEState-standardized accounting system applied to entire ownership as well asqualifying projects

TRANSPARENCY AND VERIFICATIONState licensed verifiers (California Energy Commission) and stateRegistered Professional Foresters work with standardized verification protocols

PROVIDE CO-BENEFITS; PROMOTE AND ENHANCE SUSTAINABILITYProgram promotes native forests; focus on conserving working forests,restoration and reserves

GLOBAL NORMS FOR CO2 EMISSIONS REDUCTIONS PROJECTS:Californias Approach

Table A


10/56

viii Forest Carbon in the United States

HOW CARBON DIOXIDE FLOWS IN FORESTS:STORES, EMISSIONS & REDUCTIONS

Figure B


11/56

Executive Summary1

The United States has a highly significantopportunity to reduce its net emissions of carbon dioxide (CO2) through actions on

private forests in three areas:

Reducing forest lossIncreasing reforestation of former forestsIncreasing forest age

Actions in these three areas could permanentlyincrease US carbon stocks by millions of tonsannually at a cost per ton equivalent to thelower end of the range of mitigation costs.

Forest Extent and Carbon RetentionForests occupy one-third of the US land mass

(747 million acres), with private ownership onalmost two-thirds of that area (424 millionacres). Private forests are the most productive and threatenedforests in the country. Theirstate of carbon accumulation or release has amajor impact on the US carbon balance. Whenforests accumulate and hold carbon (sequestra-tion), they contribute to lowering emissionsoverall, acting as carbon sinks or reservoirs.When forests are disturbed through harvest orconversion to other land uses, they release car-bon, adding to emissions overall. From the netatmospheric carbon balance, a molecule of CO 2removed from the atmosphere is equivalent tonot releasing a molecule of CO2 to the atmos-phere. Whether these forests accumulate andhold carbon, release carbon, or are lost entirelyas carbon sinks will be a major determinant inhow quickly and cost effectively the US can

meet its goals to reduce carbon emissions.Currently, the US counts on forests to help

reduce its net total emissions; for example,forests sequestered 310 million metric tons of carbon (MMT CE) in 1999. However, thisamount was less than the prior year and contin-ued a decline in stores from the past five years.When forests store less carbon, they are releasing

more. The two major causes of decline in storesof forest carbon are forest conversion and lossand the increase of harvest versus the amount of growth on private forest lands.

Forest cover in the US has declined by one-thirdto one-half its extent since European settlement.While there has been considerable re-growth of US forests since the early to mid-1900s, especiallyin the Northeast and Southeast, the US is never-theless currently losing forests at an increasingrate, with lost acreage in the five years from 1992to 1997 estimated to be twice as great as in the 10

preceding years, 1982-1992. Further, land in for-est use is projected to continue to decline ascompetition for land for development continuesto increase. As forests are lost, particularly olderforests, so too are carbon reservoirs, as olderforests accumulate and store more carbon thanyounger ones.

In addition to forests lost to development,increases in timber harvests outstripped theamount of tree growth on private forests in the1980s and 1990s, extending a trend from the

1960s. As a result, the US is beginning to losemore carbon in private forests than it is accumu-lating, especially in faster growing softwoods andthe most productive forest regions of the coun-try, the Pacific Northwest and Southeast.Overall, the US lost 11.5 million acres of existingforest between 1982 and 1997, and the averageage of forests on private lands declined in thisperiod as well. The results include diminishedforest habitat and watershed values as well asdecreased carbon stocks. Since 1990, the US hasstored less forest carbon each year. That trend isprojected to continue to 2020 through the lossand unsustainable harvest of private forests.

Increasing Carbon StoresWhile the total amount of forest carbon stor-

age is declining, especially on private lands, it isnot irreversible. The US has the opportunity to

Executive Summary


12/56

2 Forest Carbon in the United States

increase net forest carbon stores on privateforests significantly by addressing the causes of these trends and encouraging landowners toalter prevailing business-as-usual practicesthrough changes in management actions consis-

tent with increasing carbon stores. These actionsinclude preventing forest loss by conserving cur-rent forestland, reforesting former forest areas,and increasing average forest ages. With suchactions, forest carbon stores could increase in theUS by hundreds of millions of tons during thenext several decades and play a significant role indiminishing net US emissions of CO2.

These changes in management depend on thedevelopment of a new market for forest carbonsequestration services. Value added to standingtimber in the form of carbon credits, if suffi-ciently valued, could encourage private forestowners to make the management decisions thatwould reduce CO 2 emissions and increase carbonstores. To be an effective economic incentive, theprice of carbon needs to be at least $20/ton C($5.45/ton CO2). This enables carbon to providesome incremental value to landowners, resultingin either retention of land for forest or retentionof trees during harvest. At prices of $100/ton C

($27.25/ton CO2) and greater, the value of carbonalone begins to pay landowners to hold land andincrease forest age, competing directly withreturns from development and short rotationtimber harvest. However, carbon values are like-ly to have the most significant impact when usedto pay for partial interests: keeping the land intimber production, but changing forest manage-ment and paying part of the cost to keep landfrom conversion. In most cases of high develop-ment or very high timber values, carbon valuesmust be at least $150/ton C ($40.87/ton CO2) tobe competitive.

To create this marketplace, the US needs toestablish some essential infrastructure. This includes:

Formally identifying carbon rightsDeveloping a standardized carbon account-ing system that includes both credits anddebits and adjusts appropriately for riskEstablishing a credible registry at the federaland state levels

Accounting for forest carbon should followthe same principles, such as additionality (credit-ing above business-as-usual), permanence, andaccuracy levels, established for other carbonemissions sectors, such as energy and transporta-tion. Standard accounting rules for a forest car-bon market must:

Include both debits and creditsDiscount appropriately for riskDiscount for less-than-permanent storesRequire accuracy to the same level as forother emissions sectors

Increasing net and permanent forest carbonsequestration while decreasing forest carbonemissions is clearly a meaningful piece of the setof actions that the US needs to pursue to reduceglobal warming. With a transparent and credible

accounting system, a robust market for long-term and verifiable US forest carbon creditswould yield a significant new revenue base forprivate forest landowners and lower the cost of permanent emissions reductions for carbon pro-ducers. It would also lead to substantial benefitsfor biodiversity, watershed values, open spaceprotection, and long-term, sustainable domestictimber supplies. A carbon market would promoterestoration of forest timber inventories, and there-fore carbon, on private forests in the US, leading tomore sustainable forest economies overall.


13/56

Overview 3

ising atmospheric levels of carbon dioxide(CO2) are agreed to be a source of signifi-cant concern as a cause of global warm-

ing (IPCC 1995). CO2 is the target of most effortsto reduce greenhouse gases; though it does nothave the greatest global warming potential of the greenhouse gases, it is emitted in the greatestquantity. Thus, reducing net emissions of CO 2 tothe atmosphere is the goal of both national andinternational efforts to mitigate or avoid globalwarming. Achieving net reductions of CO 2 canbe done both by reducing direct emissions andby increasing the amount of CO 2 removed fromthe atmosphere through photosynthesis andstored in plant biomass for the long term(sequestration). A ton of carbon dioxideremoved permanently from the atmosphere

through sequestration is equivalent to prevent-ing a ton of carbon dioxide emissions.

The forest sector is the second-largest sourceof CO2 emissions globally; it is also the mostexpandable long-term sinkfor CO2 (Dixon et al.1994). Forest-based carbon emissions are largelydue to forest loss, such as through conversion toagriculture or development, and to harvest.While forests can recover from harvest losses,conversion of forest land to other uses eliminatescurrent stocks and future stores permanently. Withunsustainable management, such as when harvestexceeds growth and forests are degraded, forest car-bon stocks never fully recover.

As part of the implementation of the 1992United Nations Framework Convention onClimate Change, more than 160 nations agreed

TheRoleof Forestsin the Carbon Budget of the United States

Overview

R

Older forests, especially those never harvested or cleared for agriculture, hold great stocks of carbon. Conserving them helpsbalance the US carbon budget. These older forests also have irreplaceable biodiversity, habitat and watershed values.

L A R R Y U L R I C H


14/56


that the risks of global warming warranted con-certed international action. In response, theydeveloped the Kyoto Protocol in December 1997.This document provides international guidanceand a framework for actions to reduce globalwarming. As of September 2000, 84 countries hadsigned the Protocol. The second article of theProtocol acknowledges the importance of forestecosystems both as sources of CO2 emissions andas a critical means to reduce CO2 emissionsthrough maintaining and increasing sequestration:

Article 21. Each Party...shall:(a) Implement and/or further elaboratepolicies and measures in accordance with its

national circumstances, such as:... (ii) Protection and enhancement of sinksand reservoirs of greenhouse gases not con-trolled by the Montreal Protocol...; promo-tion of sustainable forest management prac-tices, afforestation and reforestation.(UNFCCC 1997)

While discussions continue as to how individual

nations will choose to proceed with implement-ing the Protocol and what degree of effortshould be placed on any single means of emis-sions reduction, there is strong agreement, espe-cially in the United States, that forests have a role

to play (IPCC 2000; US Department of State 2000).The 747 million acres of forestlands in the

United States are some of the most productive inthe world for biomass (Barbour and Billings1998). The 424 million acres of forest in privateownership are the most productive forests in theUS, producing the greatest amount of timberand having the capacity to store the greatestamount of carbon in the country. Paradoxically,these private forests are increasingly at risk of con-version and degradation (USDA 2000; Best andWayburn 2000). Nationwide, harvest of privateforests has been increasing relative to growth forthe past four decades. Loss and degradation of theseforests means greater forest carbon emissions,along with the loss and diminution of any futurepotential to increase critical carbon stocks.

On average, the US lost 1.275 million acres of forest each year during the last 15 years (USDA1999). In 1998, the National Research Councilprojected that the US will lose another 20 mil-

lion forested acres by 2020. The Southeast andNorthwest, the nations most productive forestlandareas, lead the country in forest loss. Although mil-lions of acres of pasture and crop land, primarilyin the Southeast, were planted to trees since1985, these areas will take many years to begin toaccumulate significant amounts of carbon anddecades more to replace the forest carbon thathas been lost (USDA 2000).

With these conditions, the US faces both achallenge and an opportunity to manage its forestcarbon stocks for the future. It can either enable acontinuation of business-as-usual, with con-comitant declines in carbon stocks, or encouragea broad portfolio of forest conservation andmanagement activities to enhance forest carbonsequestration that increases long-term forestcarbon stocks.

Deforestation from increasing development in forests acrossthe United States causes the loss of existing carbon stocks, aswell as any potential future stocks. Preliminary data from themost recent NRCS National Resources Inventory shows thatthe US lost forests to development at twice the rate in the peri-od 1992-1997 as it had in the prior ten years.

G A R Y B R A A S C H


15/56

Accounting Principles5

Forest carbon accounting differs from other,non-land-based sectors as forests bothsequester and emit carbon. Whereas a ton

not emitted, for example, through conservation,is simply not emitted, a ton sequestered, for

example, in an afforestation project, can later bere-released on harvest. The goal in the forestrysector, therefore, goes beyond achieving a carbonneutral state of zero emissions, which otheremissions sectors may target. The forest sectorcan achieve a target of carbon negative.Achieving carbon negative results in forest carbonstores that last millenniaand even longer.Forest carbon accounting thus must includeboth accumulation in, and release from,foreststhat is, both credits and debits account-ingand it needs to include both annual andlong-term accounting over a period of at least100 yearsthe time frame recommended by theKyoto Protocol.

As with other emissions sectors accuracy inaccounting for forest carbon varies depending onscale: global, national, and project or site-based.The larger the area considered, the greater theuncertainties. Global-level accounting for forestcarbon change or flux is the most uncertain.

National-level accounting is significantly moreaccurate. Project-level accounting accuracy is inthe same range as for other emissions sectors in theUS. Project-level accounting for sequestration andrelease of forest carbon can be achieved with 90% to95% accuracy (Brown 1995). The following sectionsidentify how accurate project-level accounting upto landscape scales can be accomplished.

What to Count and How to Count It Forests store carbon in virtually all their

components: soils, litter (forest floor), andunderstory, as well as trees (Figure 1) (Turner etal. 1995 a and b). Forest carbon is both organic(from biomass) and inorganic (mineral carbon carbonates). Organic carbon material varieswidely in its stability, from being easily releasedto the atmosphere (labile carbon) to not easilyreleased (stable carbon). Forest-soil carbon is alarge, stable pool, accounting for some 50% of total forest carbon and changing very slowlyover hundreds of years (Kimmins 1997). For timeframes of 100 years and less, forest accounting

50%

6%10%

1%

33%

CARBON POOLS BY SECTORFigure 1

Accounting Principles

Myths&Realities

Trees

Understory

Woody Debris

Forest floor

Soil

Stable C

Labile C

Source: Turner et al. 1995 a.

Soil carbon is relatively stable withminor change happening over time.It accounts for half of total forest car-bon. Only 10% of labile carbon is insoil organic matter.


16/56

can ignore this pool and focus on changes tomore labile forest carbon components.

The vast majority of forest carbon accumula-tion is from photosynthesis by trees, with under-story vegetation accounting for less than 5%

(Kimmins 1997). Accumulation of carbonthrough tree growth and the release of carbonfrom timber harvest, including from decay of dead material or from burning in post harvestsite management, therefore, constitute the pri-mary accounting focal points. (Figure 2) Harvestcauses a decrease of forest carbon, as forest car-bon is transferred from the site to the forestproducts pool and is released from decayingstumps and slash (branches, tops and leaf litter),from soil, and from burning the site in prepara-tion for re-forestation. When trees are harvested,increased exposure of soils increases the release of carbon from litter and soil (Harmon et al. 1990). Forexample, in the period 1972-1991, post-harvest-related carbon release from decay of slash, litter,and soil in Washington and Oregon alone is esti-mated at 11.8 million tons/year (Cohen et al. 1996).

Standing live trees account for 64% of forestlabile carbon, with roughly half of that carbon inthe bole (main tree trunk) and the remainder in

roots, bark, branches, and leaves or needles(Turner et al. 1995 a and b). Tree-bole volume ismeasured in standard accepted methodologies,with standard extrapolations for volumes of root, branch and other carbon (Birdsey 1996).The efficiency of the harvesting and millingdetermines how much of the harvested bole 1

ends up in forest products. Estimates vary, butroughly between 20% and 33% of labile forest car-bon ends up in forest products (Birdsey 1996;Skog and Nicholson 1998; Harmon et al. 1996a).

Up to 40% of this carbon is stored over the longterm in such products as saw timber and furni-ture, lasting for 20 to more than 100 years. Theremainder is stored for the short term in such

products as paper, lasting five years or less.However, the decay rates for these products areunpredictable at best as paper, for example, maybe landfilled and remain undisturbed for decadesor it may be burned and released immediately.Given these variations, forest products may bebest treated as a whole, rather than divided intoshort- and long-term stores. When taken as a

whole, the average estimate of the decay rate forall forest products combined is 2%/year (Figure3) (Harmon et al. 1996a).

Taking the high end of this range, if a harvestaffects 100 units of carbon, on average 35 units of carbon remain on site and 65 units leave withthe harvest. After manufacturing, however, 32.5units of that carbon remain in a forest product,

This figure illustrates thefate of carbon in aDouglas fir forest managedon 40-year rotations. Afterharvest, dead materialdecays and releases carbonto the atmosphere forapproximately 40 years.Thus while new carbon isabsorbed in trees (bole, root, branches) carbon emis-sions outweigh accumulations until after 20 years.


Years Since Initial Harvest

C a r

b o n

( t o n s / a c r e

)

FOREST CARBON FLUX AFTER HARVEST

350

300

250

200

150

100

50

00 40 80 120 160 200

Figure 2

1During harvest, parts of the bole are left behind such asthe stump, top, and imperfect sections. The harvestedbole is the bole minus these sections.

Bole CRoot CFoliage/Branch CSoil CDead CTotal

Source: Harmon et al. 1996c.


17/56


where it decays slowly over time. The other half becomes volatile or is burned in processing. (Infact, some of the material is burned for fuel or aspart of chemical processing and may offset otherfuel sources, though this is arguable.) Of the 35

units of carbon that remained in the forest,more than 30 are lost in subsequent site-prepara-tion fires and increased decay over time.Roughly 5% of the original 100 carbon unitsremain in the forest over time (Harmon et al.1996 a). Thus, nearly two-thirds of forest carbonin a harvest is released as emissions over time.

Changes in forest carbon stocks from forestgrowth, harvest, and decay can therefore beaccounted for through the inventory of treeswith the application of standard algorithms toinclude root, branch litter, and decay pools(Birdsey 1996). On commercial forestlands in theUS, tree inventory is typically done with 90-95%accuracy. Documentation and data on the domi-

nant forest types and tree species in the US arewell developed and widely available. Multi-bil-lion-dollar industries rely on the accuracy of these inventories and the US government hasinvested substantial sums in updating and main-

taining those data. Unlike many other forestregions worldwide, data and infrastructure toaccurately track and verify forest carbon flux byproject site and, increasingly, at the nationalscale, exist in the US.

As noted, the mechanics of forest carbonaccounting are relatively straightforward, oncethe basic data of tree growth and decay are col-lected. However, natural disturbances, such aspests, wildfires, storms, or disease, can alterforests outside of human intentions, makingsome carbon outcomes unpredictable. Further,the short-term and sometimes unpredictablenature of human decision-making that can serveto increase carbon removal, such as altering the

LABILE FOREST CARBON: GROWTH, HARVEST, DECAY

When forests are disturbed by harvest, 32.5% of the carbon is released to the atmosphere within 5 years. Thisincreases to 62.5% over time as the majority of the 35% of carbon initially retained on site (in stumps, roots, andcoarse woody debris) is released through decay. 32.5% of carbon is transferred to the forest products pool, where2% of this carbon is released per year through decay.

Forest Carbon Harvest Disturbance Carbon Transfers

32.5 % C released to

atmosphere within 5 yrs

32.5 % C in forestproducts. On average,2% lost/yr to decay

35% C initially retainedon site

C to Emmisions

C to Products

C Left On Site

16% C in coarse woody debris10% C in organic soil matter

74% Live Cin trees andshrubs

}Sources: Harmon et al. 1996 c; Turner et al.1995 a and b.

Figure 3

Photosynthesisand carbon uptake


18/56


19/56


Natural risks of fires, pests, etc.Human risk of plans and actions changingover time

These risks can be accounted for or discountedto ensure comparability with other sectors.

In addition, there are other factors that affectall sectors: establishing baselines; leakage, orthe risk of simply shifting carbon emissions fromone area to another; the risks of re-release orincreased emissions at later dates; or factorsunpredicted at the time a project is started. Asthese uncertainties affect all emissions sectors,standardized means must be developed to dealwith them uniformly. As with accuracy at theinventory level, forest accounting can and

should be held to similar uncertainty levels, ordiscounted to those levels, as described below.

Natural RisksNatural risks include natural disasters: fires,

catastrophic storms, pests, and disease. The natu-ral risks involved all release carbon, either direct-ly or through reduced growth. As such, theyhave historically been tracked by the timberindustry to account for risks to their timberinventory, which is effectively a surrogate for acarbon inventory.

Historically, this risk of loss to inventory is lessthan 1% over time, but this does not translate to aloss of 1% per year of timber inventory. Someinvestors may adjust projected forest earningsdownwards by some fraction of this 1% factorover time. As forest earnings are based on growthand harvest, this risk factor can reasonably betranslated to carbon credits and debits. Within a

market system where estimated carbon creditsand debits become the basis for trading in emis-sions, one can revise projections of carbon accu-mulations over time downward by some fractionof 1%. Perhaps more reasonably and accurately,one would require annual accounting to verifyactual stores and require diversified pools of insurance carbon to ensure actual carbon gains

are achieved. Maintaining such a guaranteed poolof carbon might be done in a fashion similar towhat the Federal Deposit Insurance Corporation(FDIC) requires of banks, with forestland ownersdepositing some measure of carbon credits with

the government as insurance.

Human Risk

Another risk in forest sector accounting ishuman: people shortening the time to harvestafter a project has occurred, re-releasing carbonaccumulated and credited by a project for whichcredits were traded. This risk can be dealt withthrough discounting non-permanent projectscredit as well as by permanently entering forests

into a trading system. Such a trading system willrequire the development of a national or poten-tially state-based carbon registry. The registrywould ensure the quality and standardization of these credits, as well as ensure there is no doublecounting of credits.

Forest species can be very long-lived. Manytree species live hundreds and even thousands of years, making forests the longest-term and mostexpandable of terrestrial carbon reservoirs. Yet

these ecological time frames are largely incom-patible with human economic and political timeframes that typically range from weeks to sever-al years. The Kyoto Protocol identifies the aver-age 100-year cycling time of carbon as the desiredtime frame in which to operate. Thus, forest car-bon projects need to have a longer time framethan other sectors to account for the potentialof re-release of carbon from forest harvest orconversion over time. Discounting projects from

a hundred-year or permanent duration can becombined with annual credit and debitaccounting. This would effectively encouragelonger-term projects with more durable carbongains as full annual credit would be given toprojects that store carbon for 100 or more years.Proportional credits, or discounted credits,would be given to shorter-term stores.


20/56


Creating Carbon Stores: Myths and RealitiesThere are several pervasive myths about how

forests should be managed to create the bestcarbon stores. These illustrate the importance of a standardized forest carbon accounting systemas, depending on how the accounting is done,different management systems will be put intoplace. Major myths include:

Younger trees (preferably plantations) arebetter than older forests as carbon storesbecause young trees grow faster, whereasolder forests are decadent.The US has lost and gained roughly thesame number of acres of forestland overthe last few decades and so is not losing for-

est carbon.Harvest does not have a negative effect onforest carbon because most of the carbon istransferred to products, creating a netincrease in overall forest carbon.

Forest age: Are younger or older forestsbetter?

The argument that younger forests rapidgrowth increases carbon stores most quicklyposits that the best way to increase forest carbon

is to increase tree planting and the extent of young plantations. This argument neglects thefact that existing older forests typically have agreater impact on total forest carbon thanyoung forests of the same acreage.

While a high rate of sequestration is impor-tant, the total amount of carbon sequestered(absorbed and stored) is equally, if not more,important in terms of annual carbon budgets.Thus, an acre of Douglas fir forest at 60 yearscontains roughly 125-150 tons of carbon andaccumulates carbon at 2.2 tons/year, while thesame acre at only 10 years may contain roughly50 tons of carbon and accumulate carbon atroughly 1.4 tons per year. An old-growth forestof more than 500 years may contain more than1,000 tons of carbon/acre and accumulate at

about a quarter-ton/year. All three forests are

valuable to maintain. Replacing the 60-year-oldforest or the old-growth forest with a 10-year-old one, however, would not result in a net gainof carbon despite the more rapid growth of theyoung trees. In fact, it would produce a signifi-cant net loss. Whether the practice is harvestingolder hardwood forests in the Southeast for chipand replacing them with pine plantations, orharvesting older conifer forests in the Northwestand replacing them with short-rotation planta-tions, a net negative carbon balance is createdthat takes many decades, or more, to rectify.

Forest extent: Are we losing forest carbon?The amount of land in forests significantly

influences total forest carbon stocks, but, asnoted, the age of forests on that land is the more

Years Since Initial Harvest

F o r e s t

C a r

b o n

( t o n s / a c r e

)

FOREST CARBON DURING TIMBER HARVESTS

400

300

200

100

0

-25 0 25 50 75 100 125 150

Figure 5

Extended Rotation40 yr. Rotations

80 yr. Rotations

To replenish carbon stock fully after harvest, the forestmust grow to the same age as it was at harvest.Repeated harvest diminishes overall stocks, as demon-strated by these 40-year rotations on Douglas fir.Source: Harmon et al. 1996 c.


21/56


22/56


23/56

T o n s /

A c r e s


carbon, or simply trees.Who takes credit for the carbon in wood prod-

ucts? When the products leave the forest, theentity that created the credit, i.e. the landown-er, no longer controls the fate of that wood prod-

uct. Thus, can they ensure the carbon still existsand has not been released? It might be initiallyincorporated in a building that then is burned ina fire, or it could last for many, many years. If thecredit, and right, leave with the product, who isresponsible for accounting for its decay overtime? One also needs to ensure that there is notdouble or triple sale of carbon credits in prod-ucts: one as the tree is grown, another for theproduct itself, or another for the product as itchanges hands yet again. One might argue that itis simpler not to count forest products as car-bon, though in an ideal construct, one wouldaccount for their benefits over time.

If a carbon right is real property, this could cre-ate constitutional issues as it leads to the poten-tial for new takings claims. The desirability of creating this potential is dubious. An examina-

tion of these questions suggests that a carbonmarket will require the development of an insur-ance system to provide for stability and guarantees.As suggested earlier in this report, one might con-sider a role for the federal or state government in

this regard in creating something similar to theFederal Deposit Insurance Corporation for bank-ing carbon.

In selling carbon credits, one retains the carbonright. But the value of that right is then dimin-ished by the amount of the value of the creditand this must be reflected as an encumbrance onthe title to the property from which the carboncredits were sold. Further, in Tort law, when acredit is sold apart from the right itself, it wouldbe a personal property interest that the buyeracquires. The owner of the underlying right thenhas a new duty not to interfere with the credit. Forexample, if a landowner sold carbon credits associ-ated with creating an older forest, the other rightsof the new owner that are associated with the car-bon may be affected, such as the right to harvest orotherwise manage the forest.

350

300

250

200

150

10050

0

Figure 6

RELATIVE CARBON GAINS FROM REFORESTATION, FOREST MANAGEMENT, AND CONSERVATIO

Carbon Gains

Reforestration

Extended Rotation

Conservation

In this example of Douglas fir, over a 50-yearperiod, carbon gains are almost doubled by extendingthe rotation of a 40-year-old stand to 90 years, or

by saving a 200-year-old forest from conversion overthose gained by planting a new stand. However,for many forests, carbon management will entail allthree strategies: reforestation of harvested or formerforest areas, extending forest age, and savingremnant old growth.

143.5

276.07 291.24


24/56


Implicit in the discussion of the need for car-bon sequestration is that there are continuingnegative consequences for not maintaining andeven increasing carbon stores. Namely, if more

CO2 continues to be added to the atmosphere,there will be a continuing risk of further, directglobal effects on the biosphere, the ecosystemswithin it, and human welfare. One aspect of theconsequences of forest management practices toincrease carbon retention are the effects on for-est biodiversity. These can be direct, physical andphysiological effects on organisms, or morecomplex, indirect effects that operate throughecosystem and biosphere feedbacks. These indi-rect pathways can further influence carbon

dynamics negatively or positively (Myers 1992).Most studies of direct effects of forest prac-

tices on carbon emissions and biodiversity havebeen done in the tropics. Here both areal loss of tropical forests and degradation of remainingareas through burning and other impacts haveresulted in widespread species extinction (Myers1989, FAO 1993) and deterioration in ecosystemhealth (Groom and Shumaker 1993).

In temperate regions, forest managementthat affects carbon dynamics can also havestrong effects on ecosystem health and biodiver-sity. Management strategies that restore totalforest area and increase mean forest stand ageand tree sizes in order to increase carbon storeshave parallel beneficial effects on biodiversity andforest ecosystem functioning. The question of whether greater species diversity in a particular

community enhances ecosystem functioningdirectly has not been answered unequivocallyand is the subject of a great deal of currentresearch. However, at a larger landscape or

regional scale, maintaining the full range of bio-diversity helps to assure that the system has thevariety and abundance of components neededfor long-term persistence and productivity in theface of an uncertain future environment.Ultimately, if long-term stability is enhanced ona landscape, all of these direct and indirecteffects can also enhance reductions in carbonemissions, contributing to further climatic sta-bility and more sustainable, long-term forestproductivity (Figure 7).

Increasing forest area improves biodiversityand also benefits the pattern of forests on thelandscape. Restoring forest area increases thetotal amount of habitat for organisms. Increasingforest area can also reduce landscape fragmenta-tion, thus affecting biodiversity directly byenhancing the habitat quality of the forest(Franklin and Forman 1987, Saunders et al. 1991).Also, larger forests provide connectivity withreserve areas, which can enhance the overallfunctional value of the landscape for large,broad-ranging species that typically have homeranges larger than most feasible reserves(Grumbine 1990, Mladenoff et al. 1997). Thesestrategies can be particularly important on pri-vate forests that occur amidst public forest landsas they improve habitat connectivity.

At the stand scale, managing for greater

Ecological Implications

INFLUENCE OF CARBON SEQUESTRATIONON BIODIVERSITY AND ECOSYSTEM FUNCTION


25/56

Ecological Implications15

carbon sequestration would result in forests of greater mean age overall, thus a more diversedistribution of stand ages and greater structuraldiversity. Such stands would contain more com-plex, multiple canopy layers, trees of largerstature, and more coarse woody debris on theforest floor. There would also be opportunitiesfor longer successional sequences and therefore abetter balance of regional tree species diversity.

By helping maintain a more diverse distributionof habitat, such stand-level changes would alsobenefit biodiversity at a variety of habitat andmicrohabitat scales. Maintaining a range of com-munities and forest age classes across landscapes,and their spatial relationships, can be anotherway that long-term diversity and productivitypersist. Natural, diverse landscapes often haverepeating juxtapositions of different habitats.(Mladenoff et al. 1993)

The relationships of adjacent communities canbe important for forest regeneration as well as foranimal species with multiple habitat needs. Thehealth of ecosystem services is also enhanced bysuch changes, such as movement and retentionof water, reduction of erosion potential, andretention and availability of nutrients. The land-scape and stand-level changes would also reduce

forest edge contrast and micrometeorologicaleffects that influence the understory forest envi-ronment, affecting animals, micro-organisms,and tree regeneration (Chen et al. 1992, Wilcoveet al. 1986).

These direct benefits to biodiversity andecosystem function may feed into longer-termand indirect effects. For example, forests that aremore complex and diverse in age and habitat

structure have greater capacity to buffer againstchanging climate and may afford better protec-tion against changing disturbance regimes suchas wind, fire, insects, and disease. More rapid cli-mate change itself will be a growing backgrounddisturbance since it may occur faster thanspecies can adapt to changing conditions ormigrate. Enhanced forest area and reduced frag-mentation can best allow species movementunder such conditions (Peters and Lovejoy 1992).

Similarly, diverse, healthy, functioning forestsare likely to be best able to respond to continuedforest harvesting under such potential environ-mental changes as these forests are more resilient.The net result can be enhanced biodiversity andecosystem services, more stable climate, andmore sustainable forest productivity on ourlandscapes in the long term.

Figure 7

MANAGING FORESTS FOR GREATER CARBON

Increases Forest Area

Decreases Fragmentation Increases Tree Age, Size and Diameter Increases Coarse Woody Debris

Forest NOT Managed for Carbon Result In: Decreased Biodiversity Decreased Ecosystem Health, diversity and

Resilience Decreased Carbon Stability Increased Carbon Emissions

Forests Managed for Carbon Result In: Increased Biodiversity Increased Ecosystem Health, Diversity and

Resilience Increased Carbon Stability Decreased Carbon Emissions


26/56


f the nine forest service regions identi-fied by the US Forest Service (USFS),four are most important in terms of

potential gains and losses in US forest carbonstores: the Northeast, Southeast, Midwest /Lakestates and Pacific Northwest regions. The forestsin these regions contain the majority of privateforest lands, are the most productive, mostintensively managed, and most threatened byconversion (Best and Wayburn 2000). Thus, theircarbon fate is a major determinant in nationalforest carbon flux. Of these four regions, theSoutheast and Pacific Northwest have the great-est forest productivity and ability to increase car-bon stores. These two regions are particularlyimportant in tracking overall carbon flux.

Three factors are significant in tracking forestcarbon:

The amount of area in forest (forestlandextent)Average forest ageThe balance of harvest to growth

Forestland Extent and Forest Carbon Reservoirs

Forests are the most significant, expandable

long-term future carbon reservoirs or sinks inthe US. While the Northeast and Midwest/Lakestates regions have gained some 1.5 million acresof forestland, the Pacific Northwest andSoutheast combined have lost 3.2 million acres,mainly to development. The USFS projectsdevelopment pressure will increase, affecting theafforestation of croplands as well and leading to

an accelerated decline in forest extent (Alig2000).Decreasing forestland losses would substantiallydecrease US forest-based carbon emissions and increase stores.

Recent Trendsin U.S. Private Forest Carbon

120

100

80

60

40

20

0

REGIONAL COMPARISONOF FOREST PRODUCTIVITY

The two most productive forest regions of the UnitedStates are in the Northwest and Southeast. Thesehave the greatest capacity to increase forest carbonstores in the short term and maintain them. These tworegions are experiencing a net loss of forest land (seeFigure 9). Maintaining stocks in the Northeast and

Midwest where forests have regrown over the pastdecades is equally important.

NE SE MW NW

Source: Powell et al. 1992.

G r o w

t h / A

c r e /

Y r .

( c u . f

t . )

Figure 8

Northeast =Maine, New Hampshire, New York, and Vermont

Southeast =Florida, Georgia, N. Carolina, S. Carolina, and Virginia

Midwest =Michigan, Minnesota, and Wisconsin

Northwest =Oregon and Washington

O


27/56

At an average carbon stock of 35 tons peracre, the 1982-1997 loss of 1.7 million acres of existing forests in Oregon, Washington, Georgia,Florida, Virginia and the Carolinas alone meantthe release of at least 60 million metric tons of

carbon, as well as the loss of additional potentialstores from the forests diminished capacity tostore carbon.

The US Forest Service is projecting the loss of another 20 million acres of timberland by 2050.Conserving these lands would prevent release of 700 million tons of carbon and 19 billion tons of carbon dioxide, not to mention the loss of future

stores. Increasing forest age could result in adoubling of carbon stocks in the major forestareas over the next 25 to 50 years.

There has been a significant focus on reforest-ing, but in many states this has not been success-

ful. Oregon, a highly productive forest state,estimates that some 775,000 acres of former forestremain in unforested condition (Cathcart 2000).

Reforestation and afforestation have substan-tial long-term potential, especially in the Southeastand in Midwest agricultural areas. The ConservationReserve Program has demonstrated the appealand effectiveness of tree planting in Midwestareas for net carbon and other ecological gains.

Combining these efforts of conservation, stewardship

reforestation could increase net long-term US carbon stoby hundreds of millions of tons by 2050.

Forest Age and Carbon StoresThe longer a forest is allowed to grow prior

to harvest or the greater the average age of astanding forest, the greater the carbon storessince older forests accumulate and store morecarbon than younger forests. There is a decliningaverage age of forests on private lands, continu-ing a long-term trend since settlement, whenvirgin forests began to be harvested. This is espe-cially the case on private lands, exacerbated bythe need to generate economic returns on short-er and shorter cycles. For example, in the PacificNorthwest, the average age of harvest of com-mercial species has declined from 80 to 40 yearsduring just 20 years (Haynes 1995). As illustratedin Figure 10, this trend is projected to lead to adecrease of more than 100 million metric tons of carbon stores between 1990 and 2010 based on

the loss of older age classes and gain in youngerage classes of forest.

There is a significant opportunity to reverse this trendcarbon stocks by extending rotations, retaining trees throone or more harvests, and rebuilding older age classes of forest on the landscape.

The estimates of forest age recorded in tradi-tional growth-and-yield tables for commercial

Recent Trends 17

1500

1000

500

0

-500

-1000

-1500

-2000-2500

-3000

TRENDS IN PRIVATE FORESTLANDGAIN AND LOSS(1982-1997)

Northeast =Maine, New Hampshire, New York, and Vermont

Southeast =Florida, Georgia, N. Carolina, S. Carolina, and Virginia

Midwest =Michigan, Minnesota, and Wisconsin

Northwest =Oregon and Washington

Figure 9

Between 1982 and 1997, the Southeast andNorthwest experienced losses of forestland while

the Northeast and Midwest experienced slightgains. The net loss of forestland between 1982and 1997 is 1,695,000 acres.

Source: USDA 1997.

( x . 1 , 0

0 0 a c r e s )

581929

-498

-2707NE SE MW NW


28/56


forests can be a useful surrogate for estimatingcarbon volume. The Forestry Inventory Analysis,or FIA, also gathers data on the growing stockvolume. This, in turn, is translatable to standingcarbon. Table 1 illustrates changes in standing

volume of carbon in the main forest regions; ithighlights overall declines in carbon in theSoutheast, Northwest, and Lake States andincreases in the Northeast.

Sustainability of Management,Harvest and Growth

When more biomass is accumulated throughthe growth of forests than removed in harvest, anet gain of carbon occurs. When more harvest

occurs than growth, a net decrease occurs.Overall, private forests in the US are experienc-ing an increase of harvest compared to growth,despite continuing reforestation of the Northeastsince the early 1900s and an increase in growth in the Southeast, especially in hardwoods. This

trend is expected to continue as illustrated inTables 1 and 2. Hardwood harvest is projected toaccelerate while harvest of softwoods declines assoftwood inventories are depleted (Haynes 1995).In the four major forest regions considered here,there is a net excess of harvest over growth in

three regions, with greater growth than harvestonly in the Northeast. The US has an opportunity toalter these business-as-usual trends, decrease harvest ovenext decades, and rebuild carbon and timber inventories.

Table 1 illustrating growth versus harvest byregion in 1996 shows that removal of forest car-bon exceeds growth of forest carbon by a net 7.5million tons, not including loss of carbon fromdecay or transfer to the products pool. As indi-cated above, roughly 60% of labile forest carbonis released over time due to increased decay.With these factors included, more than 50 mil-lion tons of carbon were released to the atmos-phere in that year alone. By allowing growth to exceed harvest, this trend could be substantially altered.

Figure 10

TRENDS IN FOREST CARBON STORES:THE NORTHWEST WEST REGION

From 1990 to 2010, the Northwest West Region(Western Oregon and Washington) is expected to lose

more than 97.4 million tons of carbon under business-as-usual management.

Source: Haynes et al., 1995.

1,500

1,4001,300

1,200

1,100

1,000 F o r e s t

C a r

b o n

( m i l

l i o n s t o n s ) 1,423.3

1,325.8

1990 2010

Clearcutting is common silviculture in many forest regions.Typically, harvested sites are then burned to remove slash andcompeting brush. Often the soils are then cultivated prior toplanting, releasing more carbon. With increasingly short rota-tions, this silviculture results in substantially reduced carbonstocks which do not recover by the next harvest.

P F T F I L E P H O T O


29/56


30/56


Ecological SettingThe temperate rain forests of the Pacific

Northwestwestern Washington and Oregonand northwestern Californiahave the abilityto accumulate immense stocks of carbon. In

fact, old-growth forests in this region1have thegreatest carbon accumulations of any ecosystem

on Earth. These record capacities for carbonaccumulation relate to several important vari-ables, including the following:

Long-lived conifers capable of continuedgrowthLarge amounts of decay-resistant litterEnvironmental conditions favoring hightree productivity

Infrequent natural disturbancesForests in western Washington, Oregon, and

California are composed primarily of very long-lived conifers, such as Douglas fir(Pseudotsuga men-ziesii), Hemlock (Tsuga heterophylla), Western Cedar(Thuja plicata), and Coast Redwood (Sequoia semper-virens). Many of these species continue to grow indiameter and volume throughout their livesand, most unusually, even in height for two cen-turies or more. These tree species typically rep-resent the largest and longest-lived representa-tives of their genera (Franklin and Dyrness 1973).Many of these tree species also produce largeamounts of decay-resistant litter, including largeboles, when they die. As a consequence, the

primeval forests contain large amounts of dead aswell as live organic matter.

Many of these Pacific Northwest forest sitesare highly productive because of environmentalconditions favorable to evergreen trees. Much of

the annual photosynthetic production actuallyoccurs outside of the summer growing season,during the spring, fall and even the relativelywarm, wet winter months. As a result, positivenet carbon balancesmeaning significantuptake of carbon from the atmospherearecharacteristic of these forests. Continued netcarbon sequestration appears to persist to mucholder ages (e.g., 500-year-old stands) than origi-nally believed; maximum organic matter accu-mulations may not occur in these forests untilstand ages of 800 years or more.

Major wildfires and windstorms are the mostusual natural disturbances, but they appear tooccur at long intervals allowing forests longperiods for recovery. For example, fire-returnintervals for major natural fires average 400 to500 years in western Washington and 100 to 150years in central western Oregon. Furthermore,while such disturbances kill many trees, they donot consume or remove much of the organic

matter, so most of the carbon stocks are subjectto very slow release by decay processes.Consequently, natural disturbances, unlikeclear-cut harvests that claim to mimic such dis-turbance, almost never draw carbon stocks to thelow levels encountered following forest harvest.

CARBON SEQUESTRATION OPPORTUNITIES ONPRIVATE LANDS IN THE PACIFIC NORTHWEST

Regional Implications

1 The record is held by the Sequoia sempervirens, atemperate rainforest species in northern California.


31/56


32/56

carbon losses from logged and intensively managedsites are rarely considered in carbon balance work.

Recent TrendsTrends in private forestland area and timbervolumes have not been positive during recentdecades. Table 3 indicates results from 1952 to 1992.

The private timberland 2 base declined at therate of about 0.5%/year between 1952 and 1992.This trend is particularly critical in westernWashington and Oregon and results from rapidincreases in population and urbanization, espe-cially along the I-5 corridor. Based on recentrates of conversion, another 1.8 million acres(nearly 16%) of private forest lands in westernOregon and Washington are projected to be lostto conversion in the next 50 years. The conver-sion of forest lands to other uses is a response toa variety of pressures: economicthe opportu-nity for large monetary returns; socialtheantagonism of neighbors toward forest harvest-ing, especially by clearcutting; and, potentially,regulatorythe increasingly restrictive regula-tory environment associated with wildlife andfisheries issues.

Harvest pressures on private forest lands havealso increased and are evident in declines of standing stocks and average stand age. Whilesome older data show an upturn in timber vol-

ume, they do not reflect major changes in theregion from greatly reduced harvests on federallands and increased regulatory pressures.Reductions of more than 80% in federal timberharvest levels in the 1990s (after an unprecedent-

ed increase in harvests in the 1980s) produceddramatic increases in stumpage values, providingmajor incentives for harvest on private lands.The effect of this situation, as reflected in morerecent statistics (see USDA 2000 and Tables 1 and2 of this report), has been accelerated harvest of timber on all private forestland ownerships,reducing the average timber volume and carbonstock levels on these lands. Data from westernOregon show significant reductions in averageper-acre growing stock volume and live carbontonnage in the decade ending in 1995.

Reductions in per-acre carbon stocks on pri-vate lands is probably much greater than sug-gested by inventory statistics, which consideronly live carbon, as discussed earlier in thisreport. Levels of non-living organic matter havealso been reduced by intense harvesting and sitepreparation practices that have increasedremovals and accelerated decomposition of organic residues.

Potential for Increasing Carbon StorageThe potential for increased carbon storage in

Pacific Northwest forests is immense. Theseforests have a huge capacity for carbon storage,which they have not begun to reach.

The regional capacity of forests to sequesteradditional carbon can be illustrated by a simpleexample. These forests could easily recover half or more of the carbon released during the 20th

century (Harmon et al. 1990) by modifying forestpractices throughout the region. If the regional for-est were managed so as to recover half of thereduced carbon (76 tons/acre) on about half of theforest land base (12,350,000 acres) the total addition-al carbon sequestered would be 9.4 million tons.Biologically, this could easily be accomplished inthree to five decades. Federal forest management

Table 3

AREA AND TIMBER VOLUMES FOR PRIVATETIMBERLAND IN WESTERN OREGON ANDWASHINGTON (1952 AND 1992)


Survey Year Area Volume(year) (thousands of acres) (mill ions of cubic feet)

1952 21,979 55,8021962 21,132 51,8571977 19,742 46,0511987 18,267 47,5721992 17,561 49,055

2 Timberland is a classification of highly productive forestland.


33/56

Regional Implications23

policies are already contributing significantly to thisgoal with the extensive series of forest reservesestablished in the Northwest Forest Plan. Tens of thousands of acres of cutover federal forest land arebeing managed for restoration of late-succes-

sional forest conditions and, coincidentally,much higher levels of carbon stocks (Tuchmannet al. 1996).

Maintaining and increasing carbon stocks inPacific Northwest forests, as nationwide, can uti-lize a variety of measures, including:

Maintaining the existing forest land baseAdding land to the forest land baseRetaining more carbon at harvestLengthening rotation periods

Keeping and Adding to the Private ForestLand Base

Private forest landowners will need financialincentives to retain existing forest land base aswell as add to that base through reforestationand afforestation. Creation of an active carbonmarket could be a major financial opportunityfor landowners interested in retaining and man-aging their forest lands, as discussed earlier in

this report.As noted earlier, we are currently losingabout 0.5 % of the private forest land base eachyear. This level of forest land loss is potentiallyadding about 2.2 million tons of carbon per yearto the atmosphere, assuming that each acre lostcurrently stores about 100 tons of labile carbonper acre. This significant leakage might bereduced dramatically through a carbon market.

Retention HarvestingRetention of additional carbon at time of

regeneration harvest is a practice that significant-ly increases average stand carbon levels. Variableretention harvesting techniques are being substi-tuted for traditional clearcutting practices on pub-lic and private forestlands throughout the region.Under variable retention harvesting, significantstructural elements, such as large live trees, snags,

and logs, are left behind to become a part of thenext stand (Franklin et al. 1996).

Calculations of carbon contributions fromstructural retention are straightforward. Theycan be approximated simply by multiplying the

level of structural retention that is being specified,such as retention of 15% of the basal area of livetrees, by the carbon content of those particularstructures. Under current private land manage-ment practices of clearcutting in a 35-40 year rota-tion, with 10% retention (current minimal prac-tices on Weyerhaeuser Companys BC CoastalDivision, for example) the additional carbonmaintained on site would be about 9 tons/acre onan average site (Wayburn and Richards 1999).

Retention harvest practices have numerousadditional ecological benefits beyond additionalcarbon sequestration. These include lifeboatinga broad array of organisms on cutover areas andstructurally enriching the subsequent forest stands.

Lengthening RotationsSignificant lengthening of rotations could be

a very effective change in practice on privatelands to increase sequestration of carbon inPacific Northwest forests. The potential effects of longer rotations on carbon stocks can be easilyillustrated using data provided by Birdsey (1996).These data show per-acre tree carbon stocks inan age sequence of fully stocked stands of Douglas fir forests grown following clearcuttingof high-site lands.

We assume a traditional current rotation ageof 35 years; stands at that age would have 140,000lbs/acre (70 tons/acre) of tree carbon. Increasingrotation age to 60 and 70 years, respectively, pro-duces the following increases in carbon:

Rotation Age Tree C Stocks Increase in C

Stocks60 years 333,000 lbs/acre 97t/acre (217 Mt/ha)70 years 389,000 lbs/acre 125 t/acre (280 Mt/ha)

Assuming that these increased rotation agesare achieved on only 50% of the private land baseof 5.8 million acres of western Washington and


34/56


35/56


in the 150- to 250-year range. The predominantlyglacial soils of the region are considered poor inagricultural terms, but they are clearly capableof supporting dense stands of healthy trees. Leftundisturbed, these forests will continue to grow

and develop for many years and continue tosequester more carbon in the structure of thetrees and in the soil.

Economic and social forces in the Northeasthave created these new forests. If trends thatdominated the region for the last 50 years con-tinue, forestland area will grow or remainroughly constant and tree volume and carboncontent will continue to increase with averageage for at least another 50 to 100 years.

There are three forces that could intervene inthe projected increase in volume of carbon: landuse conversion, harvest, and natural risks of fire,disease and insects. The determination of whether these forces will exert large or smallinfluences is critical to projecting the future of this carbon reservoir.

Land UseThe decrease in relative profitability of agri-

culture was the major reason for the reforesta-

tion of this region. It is highly unlikely that anew wave of land clearing for agriculture willoccur in the Northeast. The greatest land usethreat to this carbon reservoir today is conver-sion to small residential and commercial proper-ties through forest parcelization and frag-mentation. As long as conversion to residentialand commercial real estate remains a more prof-itable use of forestland, it will occur at a faster orslower pace, depending on the regions andnations economy. Conversion to building lots

does not in itself decrease carbon storage or for-est growth per unit of forest area immediately.However, some complete clearing occurs forroads and buildings even with the creation of sequestered forest dwellings on relatively largelots. A typical scenario for a forested property,perhaps all or a portion of a farm a century ortwo (or even three) old might be:

1. A timber sale that removes all or nearlyall merchantable pulpwood and sawtimber2. Subdivision into lots consistent with azoning minimum (sometimes 5 to 6 acres,but often smaller)3. The sale of lots and construction of housesand roads over a 5- to 10-year period4. Further development (lawns, pastures,ponds, etc.) of the subdivided property overanother decade

Each transition results in a loss of stored car-bon and in a loss of carbon storage capacity. Eventhe subdivision itself (step 2) results in a loss of

Table 4

TRENDS IN NEW ENGLAND CARBONSTOCKS (1952-1992)

a. Growing Stock Volume of PrivateTimberland (x1,000,000 cu ft.)

ME NH NY VT1992 23,47 6,833 19,243 6,0951987 21,617 6,339 18,189 5,2301977 22,183 5,900 12,167 4,5471962 18,376 4,043 10,689 3,3471952 15,275 3,129 9,662 3,145Source; Forest Resources of the United States, 1992 (GTR-RM-234)

b. Live Forest Carbon per Acre (tons)ME NH NY VT

1992 20.88 26.45 21.96 24.85

1987 18.96 24.73 20.88 22.441977 19.43 21.77 14.05 18.191962 16.05 14.48 14.24 13.761952 13.57 11.23 14.57 14.20Source: Birdsey (1992) and Forest resources of the United States, 1992(GTR-RM-234)

c. Live Forest Carbon (million tons)ME NH NY VT

1992 343.66 107.05 324.01 98.381987 316.29 99.30 304.55 84.471977 317.11 89.51 202.61 72.921962 266.07 61.41 178.27 53.411952 222.97 46.44 161.12 50.41

Source: Turner et al (1993) and Powell et al (1992) (GTR-RM-234)


36/56


New Hampshire may best illustrate the needfor concern about the future of theNortheastern forest carbon reservoir and seques-tration capacity. Of the four states included in theNortheast here, New Hampshire experiencedthe most dramatic reforestation as agriculturedeclined. Its young forests spread over its magnifi-cent hills and mountains support a stunning newarray of wildlife, tourists, retirees, and economicactivities. It is a splendid example of a place whereletting trees grow, growing trees on purpose, andbenefiting from them in many ways are easy toexplain and apparent to most. Most of its forestsare privately owned, though the White Mountain

National Forest occupies more than 10 percent of the state. Its forests, with an average age of 58years and about 33 tons of carbon contained inlive trees per acre, form an excellent carbon reser-voir and sequestration system. Its predominantlyyoung forests are just coming into their maxi-mum sequestration years and can be managed fora wide variety of purposes because of their variedspecies composition, age, size of trees, and extent.Most of the state has an excellent road systemthat facilitates tourism, forest protection, and for-est product extraction.

Given this picture, one would think that inNew Hampshire, if anywhere, natural forestscould be left to themselves to sequester carbonand provide other benefits. A series of reports,most prominently from the Society for theProtection of New Hampshire Forests (SPNHF),make it clear that complacency on this point isout of place. Consider the following excerpt:

Although forest conversion is probably the

greatest threat to the New Hampshire forestcarbon system, others loom. The effects of acidrain on forests was first seriously researched andreported at the Hubbard Brook ExperimentalForest in New Hampshire. It is likely that airpollution now substantially reduces NewHampshires forest carbon sequestration capaci-ty, although specific estimates have not beenmade. As parcelization and fragmentationaccelerate, so too will unwise pre-conver-sion timber harvests that draw down thecarbon reservoir and reduce sequestrationcapacity. As air pollution and high gradingharvests reduce forest health and photosyn-thetic capacity, the likelihood of catastrophicforest losses to insects and disease increases.

Conversion alone is predicted to reduceforestland in New Hampshire by 372,462 acresbetween 1997 and 2020. Using todays num-bers, that results in the loss of more than 12million tons of carbon storage in this relative-

ly small state. Because converted forests cantgrow, the long-term loss will be much larger.

Society may or may not take action tomaintain the health and extent of NewHampshires forests. But it seems clear thatunless action is taken, a substantial forest car-bon reservoir will be lost or degraded.

Loss, parcelization and fragmentation of the for-est land base have been identified as problems forforest industry and many other users of the New

Hampshire forest. For the first time in many decades,the forest cover in New Hampshire has declined. Thishas implications for wildlife habitat, surface waterquality, drinking water quality, recreation, tourism,and forest management. Forestland was being con-verted to development at a rate of about 13,000acres per year between 1982 and 1992 [13,000 acresequals about 430,000 tons of carbon in trees alone].Between 1983 and 1994, 44% of the white pine

New Hampshire: Threats to Forest Carbon

removals and 53% of red oak removals in NewHampshire occurred on land being converted fromforestry to another use.

In a study of forest cover fragmentation insoutheastern New Hampshire and northeasternMassachusetts, 49% of the New Hampshire townswere found to have experienced a moderate to majorincrease in forest cover fragmentation between 1973and 1988. Also of concern is forest parcelizationthedivision of forested ownerships into smaller andsmaller parcels with lower volumes. Lower volumesaffect the profitability of the harvest and lower thecarbon reservoir. (SPNHF, 2000)


37/56

capacity in that it makes manage-ment and protection of the forestharder and more expensive perunit area. Clearly, land use con-version of this kind is a threat to

carbon storage in the forests of theNortheast and will continue to beuntil there is an incentive for for-est owners to keep their forestsgrowing in relatively large blocks.As illustrated by New Hampshire,the threat of forest conversion isreal and increasing.

Harvest Types and PressuresWorld wood use is rising and

the Northeast has millions of acresof valuable timber. As prices fortimber rise, harvest rates are likelyto increase. If the harvesting is donein ecologically sound ways, theforests of the Northeast can becomeboth secure carbon reservoirs andenhanced creators of aesthetic andeconomic value. If much of the har-vesting is done poorly, both the car-

bon reservoir and carbon storagemechanism are threatened.Fewer than one-third of the

private forest owners in theNortheast have a management plan for theirland. Fewer still use the services of a professionalforester when they harvest trees. These habitshave profound implications for the future of car-bon storage in the region. The most commonkind of harvest remains the removal of thelargest and most valuable trees in one operation.This minimizes cost per unit of wood volumeremoved and may maximize return at a singleharvest time. It is also often destructive to thefuture value of the forest for wood productionand may reduce the overall value of the property,even when the receipts from the harvest areincluded in the calculation of value.

For the mixed forests of the region, this sort

of harvest has a potentially disastrous effect oncarbon storage. Usually the remaining trees arenot capable of rapid response to the new grow-ing space afforded them by the harvest, but arenumerous enough to slow or prevent naturalregeneration of new trees of the species that willmost quickly restore carbon storage capacity.Thus, this sort of harvest can diminish the carbonstorage capacity of the forest for many years.

This need not be the result. Mixed forests of this kind can be managed to enhance carbonstorage and timber value simultaneously,through the use of individual tree, group selec-tion, and patch cutting (small clearcuts). By con-centrating growth on the larger, more vigorous


Reforestation after harvest is essential to restore carbon stocks. Reforestation offormer forests, such as in the old fields of the Southeast and Northwest, willbuild significant new carbon reservoirs.

P F T F I L E P H O T O


38/56

trees, and by removing less vigorous trees thatstill occupy space in the forest, both carbon storageand value are increased. Forest owners will practicethis sort of management to the degree that it isavailable and attractive financially and otherwise.

Societal transfer payments to owners to compen-sate them for the carbon their forests store, forexample, would be a powerful, positive incentive toconservative management.

Fire, Air Pollution, Disease and InsectsThe forest

Documents

Forest Carbon in the United States - 2007 Updated Edition