solving the climate problem - princeton...

8 ENVIRONMENT DECEMBER 2004

The atmosphere’s concentration of car-bon dioxide (CO2) has increased bymore than 30 percent over the last 250

years, largely due to human activity. Two-thirdsof that rise has occurred in the past 50 years.1

Unless there is a change, the world will seemuch higher CO2 levels in the future—levelsthat are predicted to lead to damaging climatechange. Fortunately, many carbon mitigationstrategies are available to set the world on a newpath, one that leads to a lower rate of CO2 emis-sions than is currently expected.

The environmental community is currentlyplaying a prominent role in the developmentof the CO2 policies that will elicit these strate-gies. Until a few years ago, the environmentalcommunity was almost exclusively interestedin policies that promote renewable energy,conservation, and natural sinks. More recently,it has begun to explore alliances with tradi-tional energy supply industries on the groundsthat to establish the pace required to achieveenvironmental goals, parallel action on manyfronts is required.

Charting a New Path

Comparing the carbon mitigation potential ofdifferent strategies requires a fixed time frame.

A 50-year perspective may be best: It is longenough to allow changes in infrastructure andconsumption patterns but short enough to beheavily influenced by decisions made today.

If the world continues on its currently pre-dicted path for 50 years—relegating significantaction on global carbon to a later time—manymodels of this future show CO2 emissions fromhuman activities roughly doubling by 2054relative to today (see Figure 1a on page 11,dashed line).2 This path is likely to lead to atleast a tripling of the preindustrial (circa 1750)level of CO2 in the atmosphere—280 parts permillion (ppm)—by the middle of the next cen-tury.3 Such a high concentration is likely to beaccompanied by significant global warming,rising sea level, increased threats to humanhealth, more frequent extreme weather events,and serious ecological disruption.4

An alternate 50-year future is a world in whichemissions stay roughly flat (Figure 1a, solidline). Assuming natural sinks of carbon in theocean and on land continue to function as theyhave in the past (see the box on pages 14 and 15),such a world can avoid a doubling (550 ppm) ofthe preindustrial CO2 concentration if furtheremissions cuts are made after 2054. Avoiding adoubling of CO2 levels is predicted to reducesubstantially the likelihood of the most dramaticconsequences of climate change, such as shut-

SOLVING THE CLIMATE PROBLEMTechnologies Available to Curb CO2 Emissions

by Robert Socolow, Roberta Hotinski, Jeffery B. Greenblatt, and Stephen Pacala

BEYOND

KYOTO

This article was published in the December 2004 issue of Environment, volume 46,no. 10, pages 8–19. © 2004, Heldref Publications. www.heldref.org/env.php. Thisarticle was part of a special “Beyond Kyoto” issue. For copies of the entire specialissue, e-mail Customer Service at [email protected].

©R

OG

ER

RE

SS

ME

YE

R—

CO

RB

IS.I

NS

ET

S,

LEF

T T

O R

IGH

T:©

LES

TE

R L

EF

KO

WIT

Z—

CO

RB

IS;©

DA

VID

PA

RS

ON

S;©

CO

RB

IS.


down of the ocean’s thermohaline circula-tion (which transports heat from the equa-tor to northern climes) and disintegrationof the West Antarctic Ice Sheet.5

However, keeping global emissions attheir current level will require a monu-mental effort. As seen in Figure 1a, com-mitting to an emissions trajectory approx-imating the flat path entails an amount ofCO2 emissions reduction in 2054 roughlyequal to all CO2 emissions today.

In 2000, about 6.2 billion tons of car-bon were emitted into the atmosphere asCO2. Approximately 40 percent wasemitted during the production of elec-tricity and 60 percent when fuels wereused directly in vehicles, homes, com-mercial buildings, and in industries.Approximately 55 percent was emittedin the 30 member countries of theOrganisation for Economic Co-opera-tion and Development (one quarter ofthe world’s total in the United States),10 percent in transitional economies(Russia and other formerly Communistcountries), and 35 percent in developingcountries.6 To achieve the flat trajectory,therefore, no-carbon and low-carbonenergy strategies must be implementedon a massive scale across all sectors ofthe economy and in countries at allstages of economic development.

The Scale of the Problem:Stabilization Wedges

To assess the potential of various car-bon mitigation strategies, the concept of“stabilization wedges” is useful. The dif-ference between the currently predictedpath and the flat path from the present to2054 gives a triangle of emissions to beavoided (see Figure 1a), a total of nearly200 billion tons of carbon. This “stabi-lization triangle” can be divided intoseven triangles—or “wedges”—of equalarea (see Figure 1b on page 11). Eachwedge results in a reduction in the rate ofcarbon emission of 1 billion tons of car-bon per year by 2054, or 25 billion tonsover 50 years.

Fifteen carbon-reduction strategieshave been examined, each of which is

based on known technology, is beingimplemented somewhere at an industrialscale, and has the potential to contribute afull wedge to carbon mitigation.7 Scalingup from present capacity to an entirewedge would be challenging but plausiblein each of the 15 cases. Each wedgewould entail environmental and socialcosts that need to be carefully considered.

Expressing various strategies in termsof the amount of activity required to fillone wedge allows for comparisons acrossstrategies: area of forest plantation versusnumber of fuel-efficient cars versus num-ber of coal plants where carbon emissionsare captured and stored (sequestered fromthe atmosphere), for example. The strate-gies may be grouped into five categories:energy conservation, renewable energy,enhanced natural sinks, nuclear energy,and fossil carbon management.

Energy Conservation

History leads one to expect substantialincreases in energy efficiency in thefuture, even without an emphasis on cut-ting carbon emissions. Over the past 30years, advances in efficiency, in conjunc-tion with changes in the sources of energysupply, have led carbon emissions to growonly half as fast as the gross world prod-uct (1.5 versus 3 percent per year).8 As aresult, global carbon intensity—the ratioof global carbon emissions to gross worldproduct—has been steadily falling. If thepatterns of the past 30 years continue foranother half century, the world carbonintensity will fall by half, relative to today.This reduction is already taken intoaccount in the currently predicted path inFigures 1a and 1b. To count toward anywedge, changes in efficiency and all otherstrategies will need to result in an evenlower carbon intensity than that expectedfrom historical trends.

Global CO2 emissions come fromthree broad end-use sectors: power gener-ation (which in 2000 made up 42 percentof emissions), transportation (22 per-cent), and direct uses of fuel in industryand buildings (36 percent).9 There areopportunities everywhere: in power

plants and household appliances, in air-plane engines and city planning, in steelmills and materials recycling. In a worldfocused on carbon, all three sectors willbecome targets of more intense efforts toimprove energy efficiency.10

For example, a conspicuous source of awedge is increased efficiency for theworld’s light-duty vehicles—cars, vans,sports utility vehicles (SUVs), and lighttrucks. A recent study by the global autoindustry, the Sustainable Mobility Project(SMP),11 reports that the world’s light-duty vehicles emitted 0.8 billion tons ofcarbon as CO2 in 2000, one-eighth of allglobal emissions. SMP predicts that theseemissions will double in 2050, to 1.6 bil-lion tons of carbon per year. In this sce-nario, while miles driven by light-dutyvehicles increases 123 percent (1.52 per-cent per year), average fuel economy(miles per gallon) increases by only 22percent (0.4 percent per year). Represen-tative values yielding such carbon emis-sions in 2054, for example, are 1.6 billionlight-duty vehicles (versus about 600 mil-lion today), 10,000 miles per year of dri-ving per vehicle (about the same as today)and 30 miles per gallon average fuel econ-omy (versus somewhat more than 20miles per gallon today).

Almost a whole wedge is achieved(actually, 0.8 wedges) if 2054 fuel econ-omy doubles, relative to what SMP pro-jects, or if total distance driven in 2054 ishalf as much as SMP projects.12 Such areduction in driving could be accom-plished by increased reliance on masstransit and telecommuting, if urban plan-ning makes these options convenient andeconomical relative to travel in a privatevehicle. Achieving a wedge here isapproximately equivalent to keeping thetotal CO2 emissions rate from theworld’s light-duty vehicles in 2054 nohigher than today’s instead of letting itroughly double.

Another route to wedges of energy effi-ciency is adding a carbon focus to capitalformation (such as new power plants,steel mills, and apartment buildings).Decisions that determine the energy effi-ciency of the world’s capital stock are par-ticularly important, because they lock in a

VOLUME 46 NUMBER 10 ENVIRONMENT 11

particular level of energy efficiency formany decades.

For example, the details of the con-struction of an apartment building can dic-tate its future CO2 emissions for a centu-ry; of a power plant, for a half century; ofa truck, for a decade. Retrofitting an itemof capital stock after construction is gen-erally far more costly than making theitem energy efficient in the first place.

At present, much of the world’s addi-tion to its capital stock is taking place indeveloping countries, a trend that isexpected to remain true throughout thenext 50 years. More aggressive pro-grams to demonstrate low-carbon capitalfacilities—such as advanced powerplants—in industrialized countries mayreduce the reluctance of developingcountries to build their own low-carboncapital facilities in parallel. New finan-cial mechanisms that encourage globallycoordinated demonstration of new low-carbon technologies are justified,because wherever capital facilities arebuilt that are inappropriate for a futurewhere global carbon emissions are con-strained, the world bears the costs oftheir future emissions.

It is impossible to decide whichimprovements in energy efficiency willarrive only in a world focused on globalcarbon management and which will arriveregardless. For example, installing themost efficient lighting and appliancesavailable, along with improved insulation,could supply two wedges of emissionsreductions if applied in all new and exist-ing residential and commercial buildingsby 2054, provided one assumes that nosuch improvements would be installedwithout a carbon constraint. An Intergov-ernmental Panel on Climate Change(IPCC) report expects half of these reduc-tions to arrive without climate policy, so itis conservative to expect energy efficiencyimprovements in buildings to yield atmost one wedge, rather than two.13 Simi-larly, compact fluorescent bulbs alonecould provide one-fourth of a wedge overthe next 50 years if used instead of incan-descent bulbs—which are four times lessefficient—for all the lighting projected for2054.14 Energy efficiency also carries

1954

7

14

2004

Year

2054

Historicalemissions

Bill

ion

s o

f to

ns

of

carb

on

em

iited

per

yea

r

Currently

predicted path

Stabilizationtriangle

Flat path

Towardtripling

Avoiddoubling

Figure 1a. Historical carbon emissions with two potential pathways for the future

NOTE: Our currently predicted path (dotted black line) will probably lead to at least atripling of atmospheric carbon dioxide (CO2) relative to its preindustrial concentration,while keeping emissions flat (solid line) would put us on track to avoid a doubling ofCO2.

SOURCE: R. Socolow, R. Hotinski, J. B. Greenblatt, and S. Pacala.

Currently

predicted path

2004 2054

Year

7

14

1 “wedge”

7 wedges

are needed to build the stabilization

triangle

1 wedge

avoids 1 billion tons of carbon

emissions per year by 2054B

illio

ns

of

ton

s o

f ca

rbo

n e

mit

ted

per

yea

r

Flat path

Stabilizationtriangle

Figure 1b. Stabilization wedges

NOTE: The stabilization triangle in Figure 1a can be divided into seven equal “wedges”that represent activities capable of reducing one billion tons of carbon per year by 2054.

SOURCE: R. Socolow, R. Hotinski, J. B. Greenblatt, and S. Pacala.


intangible benefits in addition to financialsavings: It fosters elegant design, qualityconstruction, and careful maintenancethat result in more satisfying products.

Energy efficiency does not have to dothe whole job of keeping global carbonemissions constant for the next 50 years.There is room for energy use to grow,even when carbon emissions are constant,as long as the growth is in the form of no-carbon and low-carbon energy. In particu-lar, the prospect of a future much morefully powered and fueled by renewableenergy is tantalizing.

Renewable Energy

A useful division of CO2 emissionsdistinguishes those arising at powerplants and those arising where fuels areused directly. The introduction ofrenewable energy can reduce emissionsin both settings.

Renewable Electricity

Electricity can be produced by manyrenewable energy sources. Hydropow-er was the first renewable energy

source to gain a large market share.Now, wind power—growing globallyfor the past decade at about 30 percentper year—is playing a substantial rolein several countries, notably Germanyand Denmark. Solar photovoltaic elec-tricity (PV), from a much smaller

base, is also growing 30 percent peryear globally.

A wedge from renewable electricityreplacing coal-based power is avail-able from a 50-fold expansion of windby 2054 or a 700-fold expansion of PVrelative to today. The expansion factorfor geothermal energy is about 100.Tidal energy and wave energy areprobably too untested at a large scalefor a claim to be made for a wedgebased on existing technology.

A 50-fold expansion of wind amountsto deploying two million wind turbines(of the one-megawatt size that is cur-rently typical).15 The land demands areconsiderable: A wedge of wind requiresdeployment on at least 30 millionhectares (the area of the state ofWyoming or nearly the area of Ger-many). Because wind turbines slowdown the wind, the wind turbines on awind farm are typically separated by 5 to10 rotor diameters to make room for thewind speed to recover from one windturbine to the next. Current rotor diame-ters are approaching 100 meters, so oneshould expect perhaps two wind tur-bines per square kilometer, or five persquare mile. The spaces between tur-

bines on land can be used in many ways,including for agriculture and grazing.

Nonetheless, because modern windturbines are so tall (as tall as all but thevery tallest skyscrapers) and thereforevisible from far away, the expansion ofwind farms is already provoking strong

NIMBY (not in my backyard) reactions.In response, wind farms in Europe,where wind power is expanding mostrapidly, are being taken to offshore loca-tions—recently, far enough to be invisi-ble from the shore.16

For a renewable energy technology,land demands for PV are relatively lowbecause the efficiency of conversion ofsunlight to PV is relatively high: An entirewedge of PV electricity will require anestimated two million hectares (the areaof New Jersey).17 Some of this area can besupplied by the roofs and walls of build-ings. The only technology comparable inefficiency of conversion of sunlight is thesolar engine running on high-temperatureheat, produced by a solar concentrator (afocusing trough or dish). This technologywas commercialized for a brief time notvery long ago with the help of subsidiesand will probably return.

Wind, PV, and solar thermal energy allhave the obvious problem of intermitten-cy. The output of wind turbines is highlydependent on the strength of the wind,and solar cells and solar troughs provideelectricity only when the sun shines. Thisproblem grows more acute as these inter-mittent technologies gain market share.

To be practical for large-scale use in elec-trical grids, intermittent renewable energysources are best combined with energystorage technologies as well as energysupply technologies that can fluctuate inoutput yet can also operate a large fractionof the time. Natural gas turbines are well

Although the land demands associated with a wedge from wind power are considerable (30 million hectares—the area of Wyoming),land between turbines can be used for agriculture, grazing, and other purposes.

©D

AV

ID W

OO

DFA

LL—

PE

TE

R A

RN

OLD

, IN

C.

matched to intermittent renewables, as ishydropower.

Renewable Fuels

There are fewer routes to low-carbonfuels than to low-carbon electricity.Renewable biofuels can be producedfrom vegetation, and hydrogen—a no-carbon fuel—can be produced fromrenewable electricity.

Because plant matter is created by pho-tosynthesis using CO2 from the atmos-phere, combustion of “biofuels” simplyreturns borrowed carbon to the atmos-phere. Renewability requires that theplants are grown sustainably—that is,planting must keep pace with harvest-ing—so that there is no net effect on theCO2 concentration of the atmosphere.Renewability also requires that very littlefossil fuel is used in harvesting and pro-cessing. Renewable biofuels can displacethe gasoline for a car or the cooking fuelfor a stove.

The world’s two largest biofuels pro-grams today are the Brazilian program toproduce ethanol from sugar cane and theU.S. program to produce ethanol fromcorn. Between them, the two programsmake a total of 22 billion liters of ethanolper year.18 If the world’s ethanol produc-tion could be increased by a factor ofabout 50, with much less associated fossilfuel use, this strategy could provideenough renewable fuel to account for onewedge of the stabilization triangle,assuming that ethanol displaces gasoline.The land demands, however, would bemuch greater for a wedge of biofuel thanfor a wedge of PV or even a wedge ofwind—about 250 million hectares,19 anarea almost the size of India and one-sixthof the area now used globally for allcrops. Bioengineering might increase theefficiency of plant photosynthesis, butlarge-scale production of biofuels willalways be a land-intensive proposition.

Wind and PV can also be used to pro-duce hydrogen fuel, enabling an economywhere hydrogen is the dominant fuel (the“hydrogen economy”). The attractivenessof a hydrogen economy, from a globalcarbon perspective, stems from the simple

fact that hydrogen (H2), unlike almostevery other combustible gas or liquidtoday considered a usable fuel, does notcontain carbon. When conventional fuelsare used in individual cars, trucks, orbuildings, CO2 is produced at such smallscale that recovery of CO2 is, if not hope-less, then extremely costly. Substantiallyreducing the CO2 emissions associatedwith energy use in vehicles and buildings,therefore, requires replacing carbon-containing fuels either with hydrogen or,in some cases, electricity.

For hydrogen fuels to contribute awedge, they cannot be produced byprocesses that emit as much CO2 as wouldfossil fuels with the same energy content.Producing hydrogen by electrolysis afterproducing electricity from wind emits noCO2, while producing hydrogen from nat-ural gas or coal—the principal primaryfuels used to make hydrogen today—ordinarily emits such large amounts ofCO2 as to defeat the emissions-reductionobjective. Producing hydrogen from fos-sil fuels can lead to substantially reducedCO2 emissions (relative to the emissionsfrom the hydrocarbon fuel displaced)only if the CO2 emissions can be capturedand stored (discussed below).

One can compare the relative impact oncarbon emissions of using a given amountof renewable electricity to back out con-ventional coal-based electricity or to backout gasoline via the intermediate step ofhydrogen production by electrolysis.Because the production of hydrogenrequires extra energy, and because thecarbon content of coal is high, displacingcoal-based electricity with wind electrici-ty provides emissions reductions roughlytwice as great as displacing gasoline withwind-produced hydrogen fuel.20 Thus,from a climate perspective, the optimaluse of wind may not be as a primary ener-gy source for hydrogen.

Enhanced Natural Sinks

The strategy of compensating for CO2

emissions by deliberately enhancing nat-ural sinks challenges our capacity to uselarge areas of land in ways that are not

damaging to biodiversity. (See the box onpages 14 and 15 for futher discussion ofnatural sinks.) Enhancing natural sinksentails fostering the biological absorptionof carbon and increasing its storage aboveand below the ground by, for example,reducing deforestation, creating new for-est plantations on non-forested land, orexpanding conservation tillage.

Current deforestation is transferringcarbon from forests to the atmosphere at arate of approximately one billion tons ofcarbon per year (a rate that is still veryuncertain). If we arbitrarily assume thattoday’s deforestation rate declines by 50percent over the next 50 years in a worldthat does not pay attention to carbonissues, a half-wedge could be provided byhalting deforestation completely in thattime. Another half-wedge could be pro-vided by establishing plantations on 300million hectares of non-forested landworldwide—about the same area neededto supply a whole wedge of biofuels (seeabove) and five times the land area now intropical plantations.21

Conversion of natural vegetation toannually tilled cropland has resulted inthe loss of more than 50 billion tons ofcarbon from the world’s soils over his-

torical time.22 Conservation tillage, bywhich farmers avoid aeration of the soilto promote retention of carbon, couldprovide a full wedge, if used on allcropland around the globe; today, lessthan one-tenth of agricultural produc-


Natural carbon sinks can be enhanced bycreating forest plantations, such as thispine stand in Angus, Scotland.

©N

IALL

BE

NV

IE—

CO

RB

IS


tion uses conservation tillage.23 Theincreased need for weed control thataccompanies conservation tillage cangenerate its own environmental prob-lems. However, conservation tillagecan be implemented in many ways,ranging from heavy use of chemicalherbicides to planting cover crops

immediately after harvest, with little orno herbicide use.

Nuclear Energy

Those hoping for a revival of nuclearpower see support coming their way if

carbon management becomes a highersocietal priority. Yet in environmentaldiscourse, nuclear power is quite oftennot mentioned.

The connection between nuclearpower and reduced carbon emissions islikely to become an important part ofthe debate about the extension of the

The oceans and terrestrial ecosystemshave historically slowed the rise of atmos-pheric carbon dioxide (CO2), absorbingmore than half of the carbon that has beenemitted into the atmosphere since the startof the Industrial Revolution. The figurebelow depicts fossil fuel emissions,atmospheric growth, and the ocean andland sinks for the past 50 years and a pro-

jection for the next two centuries that sta-bilizes atmospheric CO2 at 500 parts permillion (ppm). The flat path shown inFigures 1a and 1b is assumed for theyears 2004–2054.

The future contributions of the land andocean sinks, however, will be affected bychanges in climate and other factors.Changes in these natural sinks could

change the size of emissions reductionsneeded for stabilization by up to threewedges in either direction.

Oceans

Over the past two centuries, the oceanshave absorbed roughly half of the 250 bil-lion tons of carbon that have been emittedfrom fossil fuel combustion.1 For the past20 years, the oceans have been absorbingalmost two billion tons of carbon per yearas CO2, and the rate is steadily increasing.Complete ocean mixing takes more than athousand years, so the continualupwelling of deep water with preindustri-al CO2 concentrations will allow theocean to be a significant sink of atmos-pheric CO2 for several centuries. In thefigure at left, the ocean sink is shown toincrease until 2054 and then to fall gradu-ally under the depicted fossil fuel emis-sions scenario.

The 50-year future of the ocean sink isuncertain by about two wedges—50 bil-lion tons of carbon over 50 years—ineither direction. The uncertainty is theresult of limitations in the accuracy ofhistorical data, modeling of interannualocean variability, interactions of marineorganisms with the carbon cycle, andregional mixing, particularly in the South-ern Ocean.2

Climate-change feedback compoundsthis uncertainty. If emissions are flatover the next 50 years, the ocean sinkwill almost certainly continue to increaseits carbon uptake from the atmosphere.However, there is a chance that climatechange will weaken the ocean sink, sub-stantially increasing the CO2 emissionsreductions needed for stabilization at 500ppm. While the ocean’s response towarming is not well understood, the factthat the solubility of CO2 in the oceandecreases as the ocean temperature

OCEAN AND LAND SINKS

0

1954 2004 2054 2104 2154 2204

1

2

3

4

5

6

7

Year

Fossil fuel emissions

Land uptake

Bill

ion

s o

f to

ns

of

carb

on

em

itte

d p

er y

ear

Ocean uptake

Increase inatmosphere

NOTE: Shown are total fossil fuel emissions (thick black line), atmospheric increase(light blue), land uptake (green), and ocean uptake (dark blue). The atmosphericincrease is the difference between fossil-fuel emissions and ocean and land sinks. Theocean sink is calculated with a simple model (see source below), and the land sink isheld constant after 2004 at 0.5 billion tons of carbon per year. Fossil-fuel emissionsare prescribed by a 500 parts per million stabilization path with the following four components: 1954–2004: historical emissions; 2004–2054: flat path emissions, as inFigures 1a and 1b; 2054–2104: linear descent to stabilization; after 2104: stabilization.“Stabilization” means there is no further build-up of CO2 in the atmosphere, becauseemissions are balanced by land and ocean sinks.

SOURCE: R. Socolow, R. Hotinski, J. B. Greenblatt, and S. Pacala. The ocean sink is calculated using a model described in U. Siegenthaler and F. Joos, “Use of a SimpleModel for Studying Oceanic Tracer Distributions and the Global Carbon Cycle,” Tellus44B (1992): 186–207.

Where does the carbon go?

plant life of existing reactors. If, overthe next 50 years, all of today’s nuclearpower plants were to be phased out infavor of modern coal plants, about halfa wedge of additional CO2 emissionsreductions would be required to com-pensate. This half-wedge would not berequired if current nuclear reactors were

replaced with new ones, one-for-one. Similarly, building a wedge with new

nuclear power requires tripling the cur-rent nuclear electricity production,assuming the new plants displace coal.This would mean building about 700new 1,000-megawatt nuclear plantsaround the world.24

The expansion of nuclear power is, ofcourse, a politically charged issue. Muchof the debate has focused on plant safetyand waste management. The nuclearpower community would welcome theopportunity to build a new generation ofadvanced nuclear reactors whose designsincorporate intrinsic safety features that


increases points to a weakening of thesink. This effect, not included in themodel behind the figure, could be asmuch as one wedge.3 Further, if globalwarming triggers a slowdown of theocean’s “conveyor belt”—the thermoha-line circulation that transfers cold CO2-rich surface waters to the deep ocean—this could further reduce carbon uptake,perhaps significantly. However, such aslowdown would require warming of sev-eral degrees Celsius that is unlikely tooccur by 2054.4

The continued influx of CO2 to theocean will change ocean chemistry. Addi-tion of CO2 to the ocean makes it moreacidic. Anthropogenic CO2 emissions havealready decreased surface-ocean pH byapproximately 0.1 unit globally, compara-ble (though opposite in sign) to the pHchange during the last ice age.5 Increasingacidity may adversely affect marineecosystems and alter the balance of dis-solved minerals.6 Addition of CO2 to theocean also reduces the concentration ofcarbonate, the ocean’s main bufferingagent, and will affect the ocean’s ability toabsorb CO2 on long time scales.7

Land

The terrestrial biosphere, in contrast tothe ocean, has been almost neutral withrespect to carbon exchange, neither asignificant net carbon source nor sink,when averaged over the past two cen-turies.8 Prior to about 1950 it was a netsource due to widespread land clearingfor agriculture, but it has since become anet sink due to a variety of factors.These include regrowth on the previouslycleared land, fire suppression (whichincreases the amount of carbon in stand-ing wood), entombment of carbon in sed-iments of reservoirs and other bodies ofwater, the buildup of wood products in

buildings and landfills, agricultural soilconservation, and, possibly, fertilizationof vegetation by the increased CO2 in theatmosphere or nitrogen in air pollution.Data show that for the past half centurythese factors have been taking more car-bon out of the atmosphere than has beenemitted to the atmosphere by the massiveforest clearing in the tropics.9 The netterrestrial sink over the past two decadeshas been somewhat less than half asstrong as the net ocean sink.10

The future land sink depends on therelative importance of such mechanisms,which are still uncertain. For example, ifthe sink is caused primarily by regrowthon previously cleared land and fire sup-pression, then it will probably decreaseover time as regrowth nears completionand as lands under fire suppression com-plete their adjustment to the new condi-tions.11 This decrease, along with otherpossible negative impacts on the landsink due to changes in temperature andprecipitation, could mean another twowedges of emissions reductions would berequired for stabilization at 500 ppmCO2. Some models predict that regionalshifts to hotter and drier climates willdominate the future of the terrestrial sinkand cause a catastrophic loss of globalbiodiversity and carbon.12

In contrast, if the sink is caused pri-marily by CO2 fertilization, then it willincrease with the buildup of atmosphericCO2 to absorb three billion tons or moreof carbon per year by mid-century.13 Thislast mechanism represents the single-largest uncertainty for carbon manage-ment in the 50-year time frame—3wedges’ worth of uptake that could sig-nificantly reduce emissions cuts needed.The figure on page 14 is based on theassumption that the net land sink willremain approximately the same size as ithas been over the last two decades, at

half a billion tons of carbon per year, forthe indefinite future.

All these considerations mean the futureof the land sink is quite uncertain, rangingfrom uptake surpassing that of the oceanto a significant loss of carbon.

1. C. L. Sabine et al., “The Oceanic Sink forAnthropogenic CO2,” Science, 16 July 2004, 367–71.

2. C. Le Quéré and N. Metzl, “Natural ProcessesRegulating the Ocean Uptake of CO2,” in C. B. Fieldand M. R. Raupach, eds., The Global Carbon Cycle:Integrating Humans, Climate, and the Natural World(Washington, DC: Island Press, 2004).

3. J. B. Greenblatt and J. L. Sarmiento, “Variabilityand Climate Feedback Mechanisms in Ocean Uptake ofCO2,” in Field and Raupach, ibid.

4. J. L. Sarmiento and C. Le Quéré, “Oceanic Car-bon Dioxide Uptake in a Model of Century-ScaleGlobal Warming,” Science, 22 November 1996,1346–50.

5. K. Caldeira and M. E. Wickett, “AnthropogenicCarbon and Ocean pH,” Nature, 25 September 2003,365.

6. U. Riebesell et al., “Reduced Calcification ofMarine Plankton in Response to Increased AtmosphericCO2,” Nature, 21 September 2000, 364–67; and B. A.Seibel and P. J. Walsh, “Potential Impacts of CO2 Injec-tion on Deep-Sea Biota,” Science, 12 October 2001,319–20.

7. I. C. Prentice et al., “The Carbon Cycle andAtmospheric Carbon Dioxide,” in J. T. Houghton et al.,eds., Climate Change 2001: The Scientific Basis (Cam-bridge, UK: Cambridge University Press, 2001).

8. Sabine et al., note 1 above.

9. Prentice et al., note 7 above; and R. A. Houghton,“Why are Estimates of the Terrestrial Carbon Balanceso Different?” Global Change Biology 9 (2003):500–09.

10. Sabine et al., note 1 above.

11. G. C. Hurtt et al., “Projecting the Future of theU.S. Carbon Sink,” Proceedings of the National Acade-my of Sciences 99 (1999): 1389–94.

12. P. M. Cox et al., “Acceleration of Global Warm-ing Due to Carbon-Cycle Feedbacks in a Coupled Cli-mate Model,” Nature, 9 November 2000, 184–87.

13. J-L. Dufresne et al., “On the Magnitude of Posi-tive Feedback between Future Climate Change and theCarbon Cycle,” Geophysical Research Letters 29(2002): doi: 10.1029/2001GL013777.


make minor events far less likely to esca-late into major releases of radioactivity.Workable long-term nuclear waste man-agement probably requires the develop-ment of a political consensus in favor ofchanging the objective from managementby passive barriers to management active-ly and indefinitely.

Even more challenging than wastemanagement, however, are the problemsfor international relations generated bythe linkages between the military atomand the civilian atom. A climate strategybased on the expansion of nuclear powermust contend with the need to reduce CO2

emissions all over the world. Is the worldready to permit nuclear power develop-ment in all countries? Currently, thenuclear power program in Iran is a sourceof international tension: Although Iranclaims its nuclear program is for peacefulpurposes only and is willing to acceptinternational safeguards, it appears deter-mined to build a uranium-enrichment

capability that would put it very close tonuclear weapons if it chose to go thatroute. Entirely new international arrange-ments that make unprecedented demandson the sovereignty of all countries areprobably required for wedges of nuclearpower to materialize.

The “mutual hostage” relationshipamong nuclear power plants creates yetanother obstacle to the creation of awedge of nuclear power. A reactor acci-dent at any power plant generates politicalpressure for the shutdown of all powerplants, to an extent not present elsewherein the energy system.

Nuclear power does not have to play amajor role in reducing global CO2 emis-sions. Nuclear power addresses, for now,

only the 40 percent of CO2 emissionsassociated with electricity production.(Someday, perhaps, nuclear reactors willalso produce hydrogen fuel.) To decar-bonize electricity, there are many alterna-tives. As providers of low-carbon electric-ity, wedges of nuclear power compete withwedges of wind and PV, and, as well, withwedges from the fossil energy strategies.

Fossil-Carbon Management

In the flat trajectory of Figures 1a and1b, fossil fuels hardly disappear in thenext 50 years; rather, their CO2 emissionsstay constant. In that same period, howev-er, global energy demand is predicted togrow even faster than global CO2 emis-sions, more than doubling. If the flat tra-jectory is followed, the fossil fuel indus-try’s share of the global energy system in2054 will be much less than its 85 percentshare today. How much less will be deter-

mined by the extent to which the CO2

emissions to the atmosphere associatedwith fossil fuels can be reduced.

Fossil fuel CO2 emissions can bereduced in two ways. The first strategy isto change the relative market shares ofcoal, oil, and gas. The second strategy isto interfere with the CO2 emissionprocess, preventing the CO2 from reach-ing the atmosphere.

Changing the Mix of Fossil Fuels

Shifts in the market shares of coal, oil,and natural gas have been a feature of fos-sil fuels throughout their history. Carbonconcerns have had only a minute role thusfar in these shifts, but such concerns have

formidable implications for the future.Each fossil fuel has its own ratio of hydro-gen to carbon. Carbon emissions per unit of energy content are highest for hydrogen-poor coal and lowest for hydrogen-rich natural gas. (Carbon emis-sions per unit of energy content areapproximately in the ratio of 5 to 4 to 3 forcoal, oil, and natural gas, respectively.)Carbon concerns, as they come into play,will therefore favor natural gas and thwartcoal. Greater-than-expected expansion inthe global use of natural gas, which emitshalf as much CO2 as coal for the sameamount of electricity production, is a pos-sible consequence of a carbon constraint.

Any major expansion of natural gas usewill require transporting large amountsover long distances, either in liquefiedform or in pipelines. A wedge’s worth ofpower from natural gas instead of coal ismatched to 50 large liquefied natural gas(LNG) tankers docking and unloadingevery day or building the equivalent of the

Alaskan natural gas pipeline, currentlyunder negotiation, every year.25

Carbon Capture and Storage

The second principal strategy forreducing the CO2 emissions associatedwith fossil fuel use—interfering with CO2

emissions—requires a two-step processknown as “carbon capture and storage.”The first step, carbon capture, typicallycreates a pure, concentrated stream ofCO2, separated from the other products ofcombustion. The second step, carbon stor-age, sends the concentrated CO2 to a des-tination other than the atmosphere.

Opportunities for CO2 capture are abun-dant. The natural gas industry routinely

Acid-gas injection wellsin Alberta, Canada

(near right), as a sulfurdisposal strategy, co-

store hydrogen sulfideand carbon dioxide.

Studying cement sealsin old wells (far right)

clarifies risks of CO2leakage underground.

PH

OTO

CO

UR

TE

SY

G.S

CH

ER

ER

—P

RIN

CE

TON

©R

.SO

CO

LOW

—P

RIN

CE

TON

generates capturable streams of CO2 whennatural gas, after coming out of theground, is stripped of CO2 before ship-ment by pipeline or tanker. Refineriesmaking hydrogen for internal use are gen-erating, as a byproduct, capturable streamsof CO2; refinery hydrogen requirements,in turn, are increasing to process heaviercrude oils and to remove more sulfur. Cap-turable streams of CO2 will also be gener-ated where technologies are deployed toconvert coal or natural gas into liquidfuels, and the geopolitics and economicsof oil are nearly certain to increase the useof these technologies. In a world focusedon the reduction of CO2 emissions, allthese streams are candidates for capture,instead of venting to the atmosphere.

The most promising storage idea is “geo-logical storage,” in which the CO2 is placedin deep sedimentary formations. (Alternatecarbon storage ideas include storage of CO2

deep in the ocean and storage of carbon insolid form as carbonates.) Carbon captureand storage has the potential to be imple-mented wherever there are large pointsources of CO2, such as at power plants andrefineries. The storage space availablebelow ground is probably large enough tomake carbon capture and storage a com-pelling carbon mitigation option.

For decades, oil companies have beeninjecting CO2 underground to scrubhydrocarbons from oil reservoirs in latestages of production, a tactic calledenhanced oil recovery (EOR). Storing awedge’s worth of CO2 would require scal-ing up the current global rate of injectionof CO2 in EOR by a factor of about 100.26

Because EOR projects to date have nothad the explicit objective of long-termCO2 storage, little is known about thelong-term fate of the CO2 injected.

If waste carbon can be captured andstored, hydrogen production from coaland natural gas could provide an alter-nate route to the “hydrogen economy.”And in all situations where CO2 captureand storage is under consideration, theremay be opportunities to “co-capture andco-store” other pollutants, like sulfur,with the CO2. With co-capture, the costsof above-ground pollution control willbe reduced, and perhaps pollution con-

trol costs and total environmental emis-sions will as well.

At this time, via demonstration proj-ects, the carbon capture and storagestrategy is undergoing its first testing(see the box on page 18). The environ-mental community is being asked to helpestablish the criteria—such as the condi-tions under which permits would begiven for CO2 storage below ground—toensure that the strategy is effective andsafe. Some in the environmental commu-nity want to keep their distance, callingcarbon capture and storage an “addict’sresponse” to climate change.27 Othersrecognize the importance of taking partin setting the ground rules for carboncapture and storage to increase the at-tention paid to its environmental per-formance. Still others see benefit inengaging in coalition politics, leading topolicies that advance several CO2 miti-gation strategies at once—to the benefitof technologies they prefer.

Getting Started with Wedges:The Next 10 Years

The challenges of carbon mitigation aredaunting. Unless campaigns to reducecarbon emissions are launched in theimmediate future across all sectors of theeconomy and in countries at every stageof economic development, there will belittle hope of avoiding a doubling ofatmospheric CO2.

To clarify the scale of the effort, it ishelpful to consider the emissions reduc-tions needed over the next 10 years to stayon the flat path of Figures 1a and 1b,assuming the alternative is the currentlypredicted path. By 2014, as a point of ref-erence, one might implement 20 percent

of each of 7 wedges. There are manyways to do this.

• Addressing demand, to reduce elec-tricity emissions, the world could accom-plish the first 20 percent of a buildingsefficiency wedge by replacing everyburnt-out incandescent bulb with a com-pact fluorescent bulb. Addressing supply,the world could develop the first 20 per-cent of a wind wedge by completing400,000 new wind turbines, or of anuclear power wedge by building 140new nuclear plants. It could implementCO2 storage projects with 700 times thecapacity of the Sleipner project (see thebox on carbon capture and storage onpage 18). As part of an augmented strate-gy to displace coal with natural gas inpower plants, it could build 10 natural gaspipelines having the capacity of the Alas-ka pipeline now under discussion.

• To reduce transport emissions, againaddressing demand, the world couldimprove average vehicle fuel economy

by 25 percent with the assumption thatthe amount of driving also increases by25 percent. Addressing supply, the worldcould convert 50 million hectares(200,000 square miles) to crops likesugar cane that can be converted toethanol with modest fossil fuel inputs. Itcould accelerate the arrival of hydrogen-powered vehicles and the production oflow-carbon hydrogen.

• To reduce emissions from the space-heating and -cooling of buildings, the worldcould embark on a campaign of imple-menting best-available design and construc-tion practices, especially for new buildingsbut also for the retrofit of buildings.

• The world could take some pressureoff the energy system by modifying the


Replacing burnt-outincandescent bulbswith fluorescentones could be one of a handful of significantbeginning stepstoward reduced CO2 emissions.

©P

ET

ER

FR

ISC

HM

UT

H—

PE

TE

R A

RN

OLD

, IN

C.


agricultural practices on nearly one-fifthof all cropland to bring about conservationtillage. It could create 60 million hectaresof sustainable plantations on nonforestedland and set a new course to eliminatetropical deforestation within 50 years.

It is hard not to feel overwhelmed bythis menu. Are there ways out? Perhaps anoptimum pace for a 50-year campaignwould result in bringing some of thesewedges forward more slowly, accomplish-ing less than 20 percent of the job in thefirst 10 years. But how much more slow-ly? Perhaps the number of parallel effortscan be reduced by getting two or threewedges from a single strategy; energy effi-ciency is the most likely area. But formany of the other strategies, bringing ontwo wedges is more than twice as hard asbringing on one because cheap and easyopportunities will be used up early on.

Perhaps the stabilization triangle willturn out to be smaller: The world econo-my might grow more slowly, and greenplants, in response to elevated CO2 levels,might store more carbon than predicted.But the stabilization triangle could just as

easily be larger (see the box on pages 14and 15). In this case, getting onto a paththat avoids doubling the preindustrialCO2 concentration might require morethan seven wedges.

The Road Ahead

Advocates of any one wedge shouldtake a clear-eyed look at the difficultiesinherent in cutting one billion tons of car-bon emissions per year using that strategy.Deep patterns in the energy system limitthe rate of introduction of energy efficien-cy. Land-use constraints limit the roles ofrenewable energy and natural sinks. Deepfear and distrust, as well as nuclearweapons proliferation, hobble nuclearpower. The long record of resistance topro-environment initiatives by the fossilfuel industries compromises their credi-bility. Aesthetic considerations limit thepenetration of several renewable options,including wind and hydropower.

Advocates of one kind of wedgeshould also not minimize its potential to

attract the support of other interestgroups who can help advance its adop-tion. The widespread use and economiccompetitiveness of natural gas andnuclear electricity mean they might beimplemented within the existing energysystem relatively quickly, bringing repre-sentatives from those industries to thetable. Coal used with carbon capture andstorage reduces not only CO2 emissionsbut conventional pollution as well, a ben-efit attractive to public health advocates.Photovoltaics are bringing electricity toremote areas, including poor villages farfrom any electricity grid, and appeal toadvocates for the developing world.Wind energy should engage rural com-munities, as it brings income to ruralareas and may slow migration to cities.Biofuels and plantations can helpreclaim degraded land, enhancingecosystem services. The pursuit of ener-gy efficiency will cut costs for business-es, providing joint environmental andeconomic gains. Attention to energy effi-ciency brings broad support from all whotake pleasure in well-made goods.

Environmental groups and others havecalled attention to the need to establishthe long-term effectiveness of under-ground storage and its safety. AlthoughCO2 is not flammable or explosive, largeCO2 releases into low-lying depressionscould lead to hazardous concentrations.CO2 percolating upward from deep stor-age could also increase the acidity ofground water and soil.

Two major projects are under way tostudy geological CO2 storage, each cur-rently injecting one million tons of CO2per year. In the Weyburn project inSaskatchewan, Canada, CO2 captured ata synthetic fuels plant in North Dakotaand transported north to Saskatchewanby pipeline is used for enhanced oilrecovery (EOR) in the aging Weyburnoil fields. In the Sleipner project in theNorth Sea off the coast of Norway,excess CO2 in the natural gas being pro-duced from one formation below the seafloor is stripped away and then reinject-ed into another reservoir, as opposed tothe customary practice of venting it to

the atmosphere. A third major project,the In Salah project in Algeria, is juststarting. Like the Sleipner project, it willstrip excess CO2 from natural gas, and itwill also store a CO2 stream of aboutone million tons of CO2 per year. In thisproject, the CO2 will be injected into theformation from which the gas was ini-tially recovered, aiding in gas recovery.This is in contrast to the Sleipner proj-ect, in which CO2 is injected into a non-hydrocarbon-bearing formation.

This approach to carbon mitigation isvery young and is still being reviewed bythe international community. Next year,

the Intergovernmental Panel on ClimateChange (IPCC) will release a specialreport on carbon dioxide capture andstorage (CCS) that will summarize cur-rent knowledge. If these initial findingsand CCS pilot projects suggest that long-term storage will be safe and effective,and if adequate permitting procedurescan be established, over the next 50 yearsCCS could provide one or more wedgesof emissions reductions. For each CCSwedge, storage programs equivalent to70 of the Sleipner, Weyburn, or In Salahprojects would have to be created everyyear and maintained through 2054.

CO2 STORAGE PROJECTS

At this facility in Algeria, excess CO2 is removed from natural gas for injectionunderground.

PH

OTO

CO

UR

TE

SY

BP

A multiple-wedge approach to CO2

policy will provide common ground andfoster consensus on mitigation policy.Most advocates of particular wedgesagree that it is too early now to settle onjust a few “winner” strategies, that therelative attractiveness of strategies willdiffer from one region to another, thatenvironmental problems associated withscale-up ought to be investigated, thatsubsidy of early stages is often merited,and that choices among mature alterna-tives should be determined mostly bymarket mechanisms. Framing the climateproblem as one requiring the parallelexploration of many stabilization wedgesmay help broaden the political consensusfor early action.

Robert Socolow is a professor in the Department ofMechanical and Aerospace Engineering at PrincetonUniversity, New Jersey, and co-director of the CarbonInitiative, a 10-year, university-wide research programwithin the Princeton Environmental Institute. The ini-tiative, sponsered by BP and Ford, explores the sci-ence, technology, and policy dimensions of global car-bon mitigation. Socolow’s current research focuses onglobal carbon management and mitigation and energyand environmental technology and policy. He is acontributing editor of Environment and may bereached at [email protected]. Roberta Hotinskiis an information officer for the Carbon MitigationInitiative. She oversees communication of the initia-tive’s research results to corporate sponsers, theresearch community, the media, and the general pub-lic. Her previous research focused on paleoceanogra-phy and ocean biogeochemical cycles. She may bereached at (609) 258-7253 or via e-mail at [email protected]. Jeffery B. Greenblatt is a research staffmember at Princeton University. His research interestsinclude modeling of global energy systems, windenergy, the ocean carbon cycle, and population. Hemay be reached at (609) 258-7442 or via e-mail [email protected]. Stephen Pacala is the PetrieProfessor of Biology at Princeton’s Department ofEcology and Evolutionary Biology and co-director ofthe Carbon Mitigation Initiative. His research focuseson the processes that govern ecological communities,the interplay between community and ecosystem-level processes, and the interactions between theglobal biosphere and climate. Pacala is currentlyworking on a new model of the terrestrial biosphere.He may be reached at [email protected]. Detailsof calculations given here and more complete refer-ences can be found in the paper S. Pacala and R.Socolow, “Stabilization Wedges: Solving the ClimateProblem for the Next 50 Years with Current Technolo-gies,” Science, 13 August 2004, 968–72, available atthe Carbon Mitigation Initiative Web site,http://www.princeton.edu/~cmi.

NOTES

1. C. D. Keeling and T. P. Whorf, “AtmosphericCO2 Records from Sites in the SIO Air Sampling Net-work,” in Carbon Dioxide Information Analysis Cen-ter, Trends: A Compendium of Data on Global Change(Oak Ridge, TN: Ridge National Laboratory, 2000).

2. S. Pacala and R. Socolow, “Stabilization Wedges:Solving the Climate Problem for the Next 50 Yearswith Current Technologies,” Science, 13 August 2004,968–72.

3. R. Socolow, S. Pacala, and J. Greenblatt,“Wedges: Early Mitigation with Familiar Technology,”to be published in the Proceedings of GHGT-7, the 7thInternational Conference on Greenhouse Gas ControlTechnology, Vancouver, Canada, September 5–9, 2004(forthcoming, 2004).

4. U. Cubasch et al., “Projections of Future ClimateChange,” in J. T. Houghton et al., eds., Climate Change2001: The Scientific Basis (Cambridge, UK: Cam-bridge University Press, 2001), 525–82; and J. Antle etal., “Ecosystems and their Goods and Services,” in J. J.McCarthy et al., eds., Climate Change 2001: Impacts,Adaptation, and Vulnerability (Cambridge, UK: Cam-bridge University Press, 2001), 235–342.

5. B. C. O’Neill and M. Oppenheimer, “DangerousClimate Impacts and the Kyoto Protocol,” Science, 14June 2002, 1971–72.

6. International Energy Agency (IEA), World Ener-gy Outlook 2002 (Paris: Organisation for EconomicCo-operation and Development/IEA, 2002), http://library.iea.org/dbtw-wpd/Textbase/nppdf/stud/02/weo2002_1.pdf.

7. Pacala and Socolow, note 2 above.

8. IEA, note 6 above.

9. IEA, note 6 above.

10. H. Geller, Energy Revolution: Policies for a Sus-tainable Future (Washington, DC: Island Press, 2003);and M. A. Brown, M. D. Levine, J. P. Romm, A. H.Rosenfeld, and J. G. Koomey, “Engineering-EconomicStudies of Energy Technologies to Reduce GreenhouseGas Emissions: Opportunities and Challenges,” Annu-al Review of Energy and the Environment 23 (1998):287–385.

11. World Business Council for Sustainable Devel-opment, Mobility 2030: Meeting the Challenges toSustainability, Full Report 2004 (Geneva, Switzerland:World Business Council for Sustainable Development,2004), http://www.wbcsd.org.


13. K. Blok et al., “Technological and EconomicPotential of Greenhouse Gas Emissions Reduction,” inB. Metz, O. Davidson, R. Swart, and J. Pan, eds., Cli-mate Change 2001: Mitigation (Cambridge, UK: Cam-bridge University Press, 2001), 167–299.



16. See M. J. Pasqualetti, “Wind Power: Obstaclesand Opportunities,” Environment, September 2004,22–38.


18. IEA, Biofuels for Transport: An InternationalPerspective (Paris: IEA, 2004).



21. Kauppi et al., “Technological and EconomicPotential of Options to Enhance, Maintain, and Man-age Biological Carbon Reservoirs and Geo-engineer-ing,” in Metz, Davidson, Swart, and Pan, note 13above, pages 301–43.


23. World Resources Institute, Building a Safe Cli-mate (Washington, DC: World Resources Institute,1988).




27. G. Muttitt and B. Diss, “Carbon Injection: AnAddict’s Response to Climate Change,” The Ecologist,22 October 2001.


BP Statistical Review of WorldEnergy, by British Petroleum;http://www.bp.com/subsection.do?categoryId=95&contentId=2006480

International Energy Annual 2002,by the Energy Information Administration; http://www.eia.doe.gov/emeu/iea/contents.html

International Energy Outlook, 2003,by the Energy Information Agency;http://www.eia.doe.gov/oiaf/ieo/index.html

Proceedings of the 6th International Conference on Greenhouse Gas Con-trol Technologies, 1-4 October, 2002,Kyoto, Japan, edited by John Gale andYoichi Kaya; http://www.ieagreen.org.uk/ghgt6.htm

World Energy Outlook 2002,by the International Energy Agency;http://library.iea.org/dbtw-wpd/Textbase/nppdf/stud/02/weo2002_1.pdf (by subscription)

Key World Energy Statistics, by theInternational Energy Agency;http://www.iea.org/dbtw-wpd/bookshop/add.aspx?id=144

Land Use, Land Use Change andForestry, edited by Robert T. Watsonet al.; http://www.grida.no/climate/ipcc/land_use/index.htm

IPCC Third Assessment Report: Cli-mate Change 2001, by the Intergov-ernmental Panel on Climate Change;http://www.ipcc.ch/index.html

Special Report on Emissions Scenar-ios, by the Intergovernmental Panel onClimate Change; http://www.grida.no/climate/ipcc/emission/index.htm

The Hydrogen Economy: Opportuni-ties, Costs, Barriers, and R&D Needs,by the Committee on Alternatives andStrategies for Future Hydrogen Pro-duction and Use, National ResearchCouncil; http://www.nap.edu/books/0309091632/html/

USEFUL ONLINE REFERENCES

solving the climate problem - princeton...

Documents