+ All Categories
Home > Documents > Producer Consumer Responsibility for GHGs in Agriculture in ...

Producer Consumer Responsibility for GHGs in Agriculture in ...

Date post: 09-Jan-2017
Category:
Upload: doanxuyen
View: 215 times
Download: 0 times
Share this document with a friend
13
Producer and consumer responsibility for greenhouse gas emissions from agricultural production—a perspective from the Brazilian Amazon This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2009 Environ. Res. Lett. 4 044010 (http://iopscience.iop.org/1748-9326/4/4/044010) Download details: IP Address: 82.152.168.222 The article was downloaded on 24/02/2011 at 14:25 Please note that terms and conditions apply. View the table of contents for this issue, or go to the journal homepage for more Home Search Collections Journals About Contact us My IOPscience
Transcript
Page 1: Producer Consumer Responsibility for GHGs in Agriculture in ...

Producer and consumer responsibility for greenhouse gas emissions from agricultural

productionmdasha perspective from the Brazilian Amazon

This article has been downloaded from IOPscience Please scroll down to see the full text article

2009 Environ Res Lett 4 044010

(httpiopscienceioporg1748-932644044010)

Download details

IP Address 82152168222

The article was downloaded on 24022011 at 1425

Please note that terms and conditions apply

View the table of contents for this issue or go to the journal homepage for more

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING ENVIRONMENTAL RESEARCH LETTERS

Environ Res Lett 4 (2009) 044010 (12pp) doi1010881748-932644044010

Producer and consumer responsibility forgreenhouse gas emissions fromagricultural productionmdasha perspectivefrom the Brazilian AmazonD P M Zaks14 C C Barford1 N Ramankutty2 and J A Foley3

1 Center for Sustainability and the Global Environment (SAGE) Nelson Institute forEnvironmental Studies University of Wisconsin 1710 University Avenue MadisonWI 53726 USA2 Department of Geography McGill University 805 Sherbrooke Street West MontrealQC H3A 2K6 Canada3 Institute on the Environment (IonE) University of Minnesota 1954 Buford AvenueSt Paul MN 55108 USA

E-mail zakswiscedu

Received 14 May 2009Accepted for publication 3 November 2009Published 17 November 2009Online at stacksioporgERL4044010

AbstractGreenhouse gases from the combination of land use change and agriculture are responsible forthe largest share of global emissions but are inadequately considered in the current set ofinternational climate policies Under the Kyoto protocol emissions generated in the productionof agricultural commodities are the responsibility of the producing country introducingpotential inequities if agricultural products are exported This study quantifies the greenhousegas emissions from the production of soybeans and beef in the Amazon basin of Brazil a regionwhere rates of both deforestation and agricultural exports are high Integrating methods fromland use science and life-cycle analysis and accounting for producerndashconsumer responsibilitywe allocate emissions between Brazil and importing countries with an emphasis on ultimatelyreducing the greenhouse gas impact of food production The mechanisms used to distribute thecarbon emissions over time allocate the bulk of emissions to the years directly after the land usechange occurred and gradually decrease the carbon allocation to the agricultural products Thecarbon liability embodied in soybeans exported from the Amazon between 1990 and 2006 was128 TgCO2e while 120 TgCO2e were embodied in exported beef An equivalent carbonliability was assigned to Brazil for that time period

Keywords Amazon deforestation carbon emissions producerndashconsumer responsibilitylife-cycle assessment land use Kyoto protocol

1 Introduction

Agriculture is now recognized as one of the dominanttransformative forces in the global environment (Foley et al2005) By the year 2000 croplands and pastures accounted forsim40 of the ice-free land surface on Earth and provided food

4 Author to whom any correspondence should be addressed

feed and fuel to meet the demands of the current population(Monfreda et al 2008 Ramankutty et al 2008) Globalagriculture is also a powerful economic force according tothe Food and Agriculture Organization (FAO) of the UnitedNations the value of exported agricultural products increasedfrom $32 to $720 billion between 1961 and 2006 with thefastest rate of increase in the last decade (FAOSTAT 2009)

1748-932609044010+12$3000 copy 2009 IOP Publishing Ltd Printed in the UK1

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

The current production methods of the global food systemhelp sustain our livelihoods but the extent and intensivepractices of modern agriculture have substantial negativeenvironmental consequences (Foley et al 2005 Roy et al2009 Schau and Fet 2008) For example agricultural landuse is responsible for the release of greenhouse gases (GHG)biodiversity loss eutrophication of waterways emergence ofdisease and changes in local and regional climates all ofwhich detract from human health and security (MEA 2005)In economic terms the extent and severity of these negativeconsequences are typically externalities of the economicsystem because they are rarely communicated to the consumeror accounted for in the price of agricultural products (Prettyet al 2000)

In addition agricultural products are part of anincreasingly globalized food system that separates producersand consumers by thousands of kilometers and lengthy supply-chains The impacts of production span from local (eg airand water pollution) to global (eg greenhouse gas emissions)scales (Tilman 1999 Smith et al 2008) and the mechanisms totrack and account for these impacts are poorly developed

As markets become more globalized the production ofcash crops and other export commodities is expected toincrease This will likely lead to expansion of agriculturalland in the tropics the region that has the most arable landnot currently in production (Alexandratos et al 2006 Barbier2000) Such expansion could have serious implications forGHG emissions as did land use emissions from tropicalregions in the 1990s (Houghton 2003) Moreover Gibbs et al(2010) found that more than half of new agricultural land in thetropics originated from intact forest with another third comingfrom previously cleared forests Although signatories to theKyoto protocol are working to reduce their GHG emissionsfrom within-country fossil fuel sources they have neglectedland use emissions including those stemming from theiragricultural imports

Rising concern about GHG emissions an increasinglyinformed public and the threat of regulatory action haveprompted producers in the global food system and otherenergy intensive sectors to measure the energy life cyclesof their products (Brentrup et al 2004 Jolliet et al 2003Goleman 2009) Some producers are voluntarily providingconsumers with estimates of the life-cycle energy costs of theproduction distribution and consumption of their products toenable consumers to choose goods with the smallest energyfootprints (Gallastegui 2002) Current proposals suggest thatcarbon will become a regulated commodity under future globalclimate agreements and the disclosure of the energy used inthe production of commercial goods including agriculturalproducts will be necessary (Bodansky et al 2004)

Previous analyses have estimated the carbon containedin internationally traded crop biomass (Ciais et al 2007) andthe embodied emissions from industrial production (Petersand Hertwich 2008b) and have highlighted the importanceof producer and consumer responsibility for carbon emissions(Bastianoni et al 2004) Recent studies have explicitly calledfor the inclusion of land use related greenhouse gas impactsof soybean and beef production (Garnett 2009 Lehuger et al

2009) This study extends previously developed methodsby aggregating new land use datasets and models to trackcarbon emissions from land use to the resulting agriculturalcommodities

This study aims to quantify the hidden GHG emissionsof food production from the Amazon basin of Brazil aregion where rates of both deforestation and agriculturalproduction for export are high and to develop mechanismsto quantify and ultimately reduce the GHG impact of foodproduction Specifically our study provides an analysis ofGHGs embodied in exported beef and soybeans from theBrazilian Amazon explicitly accounting for land use We alsopropose an approach to allocate GHG emissions associatedwith agricultural land use change between producers andconsumers by integrating methods from land use science andlife-cycle analysis

2 Producer versus consumer

In the current Kyoto protocol GHG emissions are allocatedto the country in which the emission occurred Futureinternationally-binding agreements are likely to incentivizecountries to reduce GHG emissions throughout the life cyclesof the goods they produce (Bodansky et al 2004) When goodsare destined for consumption in other countries the emissionsgenerated in their production are referred to as the lsquoemissionsembodied in tradersquo (Ahmad and Wyckoff 2003) This can be asignificant fraction of global carbon emissions using a globaltrade model Peters and Hertwich (2008a 2008b) estimatedthat in 2001 roughly 23 (or sim57 Gt CO2) of energy relatedemissions were embodied in trade

lsquoCarbon leakagersquo occurs when a country opts to limit itsown carbon emissions by importing goods from a country thatdoes not participate in carbon-reduction agreements Carbonleakage is a noted problem of the current Kyoto protocoland has been estimated to comprise 11 of productionemissions (Peters and Hertwich 2008b) Consumption-basedGHG inventories account for emissions from production andimports and subtracts embodied emissions exported in trade(Peters and Hertwich 2008a) Allocating embodied emissionsto the consumer avoids carbon leakage amongst otherdeficiencies of production-based greenhouse gas inventories

Assigning the responsibility for carbon emissions to eitherproducers or consumers should not be a binary decision afairer allocation scheme is needed (Munksgaard and Pedersen2001 Gupta and Bhandari 1999) If responsibility is givento the producer carbon leakage can occur and if it isassigned to the consumer not participating in a global GHG-reduction agreement the responsibility for the emissions arenot taken (Andrew and Forgie 2008) Hence several authorshave put forth allocation schemes in which carbon emissionsare shared between producers and consumers (Lenzen et al2007 Rodrigues and Domingos 2008 Bastianoni et al 2004)These shared allocation schemes provide economic incentiveto the consumer nation to favor products with the smallestenvironmental impacts and thereby push producers to reducethe carbon emissions embodied in their products

2

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 1 The lsquopulsersquo of GHG emissions from land use can be amortized over (a) a single year (b) 20 years constant or (c) 20 years linearlydeclining

To date calculations of emissions embodied in trade haveonly been completed for manufactured products where dataon emissions from fossil fuels are readily available Life-cycleemissions embodied in agricultural products require a differentset of methodologies that have recently been highlightedby the agro-fuels community (Gibbs et al 2008 Panichelliet al 2009 Reijnders and Huijbregts 2008) Quantifying theGHG emissions embodied in a product is a necessary stepto distribute the responsibility for the impacts between theproducers and consumers

3 Allocation of land use emissions

Life-cycle assessments (LCAs) have helped to illuminate thelsquocradle to graversquo ecological impacts for a select numberof manufactured and agricultural products The LCAs ofagricultural products are markedly different from those ofmanufactured products especially if the product originatedfrom an area that recently underwent land use change Withevery transformation of land for agricultural use biophysicalimpacts occur over various spatial and temporal scales (Foleyet al 2005) When the conversion process includes removingaboveground biomass from the site a large lsquopulsersquo of GHGsis released to the atmosphere by burning or decay of theremoved vegetation (Ramankutty et al 2007) Sometimes thepulse of GHG emissions is nearly instantaneous (from burningbiomass) or it may decay slowly as forest slash or secondaryproducts (eg paper wood products) When consideringagricultural life-cycle assessments the analysis domain mustinclude impacts from lsquofield to forkrsquo since activities such asland clearing can overshadow efficiency gains in other areas ofthe product life-cycle (Gibbs et al 2008 Fargione et al 2008)

The time frame of land use varies greatly Cleared landcan transition between forest agriculture fallow and bareground as the fertility of the land changes or changes to thecropping system are introduced Depending on the locationand intensity of the new agricultural operation the land mayremain in production for as little as a single season or as longas several millennia Each of these states have different netcarbon balances as vegetation biomass regrows or is cleared

(Ramankutty et al 2007) Therefore the GHG liability fromthe initial transformation needs to be tracked over time andallocated to the appropriate user

Several amortization schemes can be used to distributethe lsquopulsersquo of GHG emissions over the duration of subsequentland use although none have been widely adopted Herewe briefly describe current methodologies and present a newhybrid approach that combines the best features from otherdescribed methods

4 Methods

The approach taken in the Ecoinvent LCA database(wwwecoinventch) assigns all land use emissions to theproduct that is harvested in the year of land conversionwithout consideration of the ultimate duration of agriculturalproduction (Jungbluth et al 2007) (figure 1(a)) The rationalefor this approach is that since deforestation causes irreversibledamage its impacts should not be amortized over a long period(Jungbluth 2009) If these impacts are monetized on a carboncost-basis and conferred to the importing country the resultingelevated price of goods becomes a disincentive for the futureconversion of land with large stocks of carbon If the land isused for agricultural production in later years the successiveproducts would not incur any carbon debt and would be lsquofree-ridingrsquo on the price paid in the initial year

While assigning all emissions to the first year is not anoptimal solution dispersing the cost of the emissions overa very long time horizon (eg 500 years) is also untenableand a middle ground needs to be explored (Canals et al2007a) An analog can be found in international accountingstandards which assign lsquouseful livesrsquo for products (Canalset al 2007b) Several LCAs and carbon footprinting standards(Muys and Garcıa Quijano 2002 BSI 2008) have adoptedmethods that amortize emissions uniformly over a 20 yr timehorizon (figure 1(b)) This time frame is practical for continuedoccupation before the land is abandoned does not confer anundue economic burden on the producer and still values thecarbon emitted in the land conversion process

3

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

However neither of these methods satisfies the goal ofsending a price signal to reduce deforestation without applyingdisproportional financial pressure on either the producer orconsumer A hybrid approach would allocate the bulk ofemissions to the years directly after the land use changeoccurred and gradually decrease the carbon allocation to theagricultural products derived from the cleared land as timewent on (figure 1(c)) With higher additional costs for thefirst several years a disincentive signal is introduced into themarket but not with disproportionate force In later years theremainder of the carbon costs are captured but at a reducedamount per year Similar to the previous case the time horizonover which the cost of the carbon is collected is arbitrary butis a length that can be societally determined All three casesassume that the land would stay in agricultural production forthe entire duration of the allocation period which may not berealistic in a market and environmentally sensitive area such asthe Brazilian Amazon

These amortization schemes are designed with inherentflexibility that makes them applicable across a range of localto global carbon-trading mechanisms Proposed programshave taken a broad-brush approach to monetizing carbonemissions from deforestation and agricultural production andare important building blocks in accounting for the life-cycle environmental impacts of agricultural production Thereduced emissions from deforestation and degradation (REDD)mechanism would compensate tropical developing countriesfor reducing deforestation rates (Mollicone et al 2007 Gullisonet al 2007) On a smaller scale groups such as Alianca daTerra in Brazil (wwwaliancadaterraorgbr) provide paymentsto farmers for more sustainable production methods by sellingtheir products at a premium These programs rely on themonetization of carbon emissions that is currently determinedby regional carbon markets These examples assume that theprice of carbon emissions are enough to reduce profit marginsand create a disincentive to production on newly cleared land

5 Case study

51 Deforestation in the Amazon

While land useland cover change and deforestation aregrowing concerns in all tropical regions the Brazilian Amazonhas been under intense pressure from national colonizationand agriculture programs and more recently due to increasedproduction of soybeans and cattle for export (Barreto et al2006) The Amazon is the largest contiguous tropical foreston the planet with vast stores of biodiversity and carbonand provides essential ecosystem services to people withinthe basin and around the world (Foley et al 2007) but alsoaccounted for more than half of global deforestation from2000 to 2005 (Hansen et al 2008) and thus for a substantialportion of carbon emissions to the atmosphere lsquoBusiness asusualrsquo scenarios of future demand for goods and governmentalpolicies suggest that deforestation and its attendant problemswill continue (Soares Filho et al 2006)

By 2007 18 of the Legal Amazon had been deforestedby smallholder as well as large holder mechanized agricultural

operators loggers and cattle ranchers among other actors(Barreto et al 2006) In the last decade mechanized agriculture(primarily soybean cultivation) and intensive cattle grazinghave been the dominant drivers of land clearing (Simon andGaragorry 2006 McAlpine et al 2009) Between 1990 and2006 the cattle herd in the Amazon almost tripled in sizeand the area used for soybean cultivation quadrupled so thatby 2006 cattle occupied 95 of the pastoral landscape andsoybeans had more than doubled their share of land (IBGE2009)

52 Transition to an export market

When large-scale deforestation in the Amazon began inthe 1970s the resultant agricultural products were mostlyconsumed within the region The Amazon did not produceenough beef to feed its own population until 1991 (Kaimowitzet al 2004) Since that time national incentives and globaldemand have transformed Brazil into the worldrsquos largestexporter of soybeans and beef among other commodities(Nepstad et al 2006) Most exports are in the form of freshor frozen beef although there is an increasing trend of livecattle exports (ALICEweb 2009) Between 1990 and 2006market and trade reforms in addition to the eradication of foot-and-mouth disease helped exports of beef from the Amazon togrow over 500 (IBGE 2009 Walker et al 2008) This growthhas had environmental consequences such as carbon emissionsfrom deforestation nutrient pollution biodiversity loss anddisplacement of local people (Betts et al 2008 Fearnside 2008Gibbs et al 2010 Foley et al 2007)

While cattle production is the predominant land use inthe Amazon soybeans have recently begun to encroach fromthe southern and eastern boundaries and are responsible fornew land clearing and displacement of cattle pastures (Vera-Diaz et al 2008) Several factors ranging from developmentof moisture-tolerant soybean varieties to increasing globaldemand for animal rations have led to the dramatic increasein soybean production and its rising percentage of the globalgrains market (Nepstad et al 2006) The Brazilian agriculturalcomplex is highly integrated into the global market system asBrazil exports more than 10 of the internationally traded cropbiomass (Ciais et al 2007)

Here we present an illustrative example that distributes theresponsibility for GHG emissions from deforestation betweenBrazil and the eventual importing nation of commoditiesproduced in the Amazon (figure 2) While the data andmethods presented here are considered to be the lsquostateof the sciencersquo the exact parameters allocating emissionsbetween international actors were chosen to exemplify theimportance of GHG emissions embodied in internationallytraded agricultural commodities as a template for futurepolicies Future work can build upon this framework makingdifferent policy assumptions as appropriate

53 Model description

Here we use a land use transition and carbon emission modeldescribed in detail by Ramankutty et al (2007) to estimatethe fate of deforested land and the associated carbon dioxide

4

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 2 Three major phases in the process of converting forests for the production and eventual consumption of beef and soybeans (1) landuseland cover change (2) agriculture and (3) trade

Table 1 Average biomass mean annual deforestation and change in deforestation rate between 1990 and 2006 for the nine states within theBrazilian Amazon (Sources INPE 2009 Saatchi et al 2007)

StateMean biomassvalue (Mg haminus1)

Mean annualdeforestation1990ndash2006 (km2)

Increase (+) or decrease (minus)in deforestation rate between1990 and 2006

Acre 204 583 +Amazonas 270 861 +Amapa 263 75 minusMaranhao 125 910 +Mato Grosso 164 6781 +Para 230 5703 +Rondonia 211 2633 +Roraima 215 246 minusTocantins 95 331 minus

equivalent (CO2e) emissions (figures 3(a) and (b)) While thepreviously published model is robust several key changes weremade to incorporate the latest published data and changingagricultural practices in the Brazilian Amazon The modelwas run for each of the nine states in the legal Amazon from1990 to 2006 using (time-smoothed) annual deforestation ratesprovided by Brazilrsquos National Institute for Space Research(INPE) (2009) Initial aboveground forest biomass values forland identified as intact forest by Brazilrsquos program to calculatethe deforestation of the Amazon (PRODES) were summarizedby state from Saatchi et al (2007) (table 1) In each yearcarbon emissions from deforestation are calculated as thesum of combusted biomass (20) and the decay of biomassremaining in the slash (70) product (8) and elemental (2)pools (Ramankutty et al 2007) (figures 4(a) and (b)) For thestate of Mato Grosso the percentages of biomass allocated tothe slash and burn pools were set to 20 and 70 respectivelyto reflect the trend of highly mechanized agriculture wheremost limbs and stumps are removed and burned after forestclearing (Galford et al 2010) The model partitions deforestedland between cropland pasture and secondary forest using

transition rates initially described by Fearnside (1996) butland transition rates for the state of Mato Grosso a hotspot ofdeforestation were updated according to the patterns reportedin Morton et al (2006) and Defries et al (2008) In addition tocarbon emissions from deforestation added methane emissionsfrom enteric fermentation in cattle were included in the totalCO2 equivalent (CO2e) flux and were computed with IPCCvalues for tropical cattle assuming a 100 yr global warmingpotential (IPCC 2007 Robertson and Grace 2004) GHGemissions from the application of fertilizer were found to benegligible compared to fluxes from deforestation and methaneproduction

54 Carbon allocation methods

The model distributes deforested land as it shifts betweencropland pasture and secondary forests computing the GHGemissions (or sequestration for the secondary forests) ofeach land use type While the model estimates the netemissions for new pastures and croplands the major exportcommodities (cattle and soybeans) only occupy a percentage

5

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(b)

(a)

Figure 3 Model simulation of the legal Amazon (a) land use and (b) carbon emissions from 1990 to 2006 The model distributes deforestedland (and resulting carbon emissions) between farms pastures and secondary forests Methane emissions from enteric fermentation in cattleare included within the emissions from lsquopasturersquo

of the agricultural landscape For each product the relativedominance on the landscape each year is calculated usingthe ratio of area planted in soybeans to total agriculturalarea and a similar ratio of pasture area for cattle tobuffalo horses sheep and other pastured animals (IBGE2009) The landscape dominance modifiers were appliedto each statersquos carbon emissions in every year of themodel

Export modifiers are calculated by dividing the amountof soybeans exported by the amount produced For cattleapproximately 10 of the herd is slaughtered every year andthe average carcass yield is used to calculate the beef andassociated products for a given year (IBGE 2009 FAOSTAT2009) Live cattle exports are incorporated with the beefexport data using average carcass yields (FAOSTAT 2009)The ratio of beef exported to beef produced was then estimated

(ALICEweb 2009) The export modifiers were applied over theaggregate Amazon to avoid data inconsistencies resulting frominterstate trade for each year of the model

The emissions calculated for each year are the sum ofburnt decay slash and elemental carbon from land deforestedin previous years minus land that transitioned to secondaryforest The allocation of carbon emissions over time werecalculated using three scenarios (1) emissions were allocatedto the year they occurred (2) equally distributed over 20 yearsand (3) linearly decreasing over 20 years (figure 1) Carbonfluxes from deforestation prior to 1990 were not includedand emission rates should therefore be seen as underestimatesAlso as future land use agricultural production and exportpatterns are unknown carbon from recent deforestation eventscannot be definitively allocated but must be tracked in order toassign future responsibility

6

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 2: Producer Consumer Responsibility for GHGs in Agriculture in ...

IOP PUBLISHING ENVIRONMENTAL RESEARCH LETTERS

Environ Res Lett 4 (2009) 044010 (12pp) doi1010881748-932644044010

Producer and consumer responsibility forgreenhouse gas emissions fromagricultural productionmdasha perspectivefrom the Brazilian AmazonD P M Zaks14 C C Barford1 N Ramankutty2 and J A Foley3

1 Center for Sustainability and the Global Environment (SAGE) Nelson Institute forEnvironmental Studies University of Wisconsin 1710 University Avenue MadisonWI 53726 USA2 Department of Geography McGill University 805 Sherbrooke Street West MontrealQC H3A 2K6 Canada3 Institute on the Environment (IonE) University of Minnesota 1954 Buford AvenueSt Paul MN 55108 USA

E-mail zakswiscedu

Received 14 May 2009Accepted for publication 3 November 2009Published 17 November 2009Online at stacksioporgERL4044010

AbstractGreenhouse gases from the combination of land use change and agriculture are responsible forthe largest share of global emissions but are inadequately considered in the current set ofinternational climate policies Under the Kyoto protocol emissions generated in the productionof agricultural commodities are the responsibility of the producing country introducingpotential inequities if agricultural products are exported This study quantifies the greenhousegas emissions from the production of soybeans and beef in the Amazon basin of Brazil a regionwhere rates of both deforestation and agricultural exports are high Integrating methods fromland use science and life-cycle analysis and accounting for producerndashconsumer responsibilitywe allocate emissions between Brazil and importing countries with an emphasis on ultimatelyreducing the greenhouse gas impact of food production The mechanisms used to distribute thecarbon emissions over time allocate the bulk of emissions to the years directly after the land usechange occurred and gradually decrease the carbon allocation to the agricultural products Thecarbon liability embodied in soybeans exported from the Amazon between 1990 and 2006 was128 TgCO2e while 120 TgCO2e were embodied in exported beef An equivalent carbonliability was assigned to Brazil for that time period

Keywords Amazon deforestation carbon emissions producerndashconsumer responsibilitylife-cycle assessment land use Kyoto protocol

1 Introduction

Agriculture is now recognized as one of the dominanttransformative forces in the global environment (Foley et al2005) By the year 2000 croplands and pastures accounted forsim40 of the ice-free land surface on Earth and provided food

4 Author to whom any correspondence should be addressed

feed and fuel to meet the demands of the current population(Monfreda et al 2008 Ramankutty et al 2008) Globalagriculture is also a powerful economic force according tothe Food and Agriculture Organization (FAO) of the UnitedNations the value of exported agricultural products increasedfrom $32 to $720 billion between 1961 and 2006 with thefastest rate of increase in the last decade (FAOSTAT 2009)

1748-932609044010+12$3000 copy 2009 IOP Publishing Ltd Printed in the UK1

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

The current production methods of the global food systemhelp sustain our livelihoods but the extent and intensivepractices of modern agriculture have substantial negativeenvironmental consequences (Foley et al 2005 Roy et al2009 Schau and Fet 2008) For example agricultural landuse is responsible for the release of greenhouse gases (GHG)biodiversity loss eutrophication of waterways emergence ofdisease and changes in local and regional climates all ofwhich detract from human health and security (MEA 2005)In economic terms the extent and severity of these negativeconsequences are typically externalities of the economicsystem because they are rarely communicated to the consumeror accounted for in the price of agricultural products (Prettyet al 2000)

In addition agricultural products are part of anincreasingly globalized food system that separates producersand consumers by thousands of kilometers and lengthy supply-chains The impacts of production span from local (eg airand water pollution) to global (eg greenhouse gas emissions)scales (Tilman 1999 Smith et al 2008) and the mechanisms totrack and account for these impacts are poorly developed

As markets become more globalized the production ofcash crops and other export commodities is expected toincrease This will likely lead to expansion of agriculturalland in the tropics the region that has the most arable landnot currently in production (Alexandratos et al 2006 Barbier2000) Such expansion could have serious implications forGHG emissions as did land use emissions from tropicalregions in the 1990s (Houghton 2003) Moreover Gibbs et al(2010) found that more than half of new agricultural land in thetropics originated from intact forest with another third comingfrom previously cleared forests Although signatories to theKyoto protocol are working to reduce their GHG emissionsfrom within-country fossil fuel sources they have neglectedland use emissions including those stemming from theiragricultural imports

Rising concern about GHG emissions an increasinglyinformed public and the threat of regulatory action haveprompted producers in the global food system and otherenergy intensive sectors to measure the energy life cyclesof their products (Brentrup et al 2004 Jolliet et al 2003Goleman 2009) Some producers are voluntarily providingconsumers with estimates of the life-cycle energy costs of theproduction distribution and consumption of their products toenable consumers to choose goods with the smallest energyfootprints (Gallastegui 2002) Current proposals suggest thatcarbon will become a regulated commodity under future globalclimate agreements and the disclosure of the energy used inthe production of commercial goods including agriculturalproducts will be necessary (Bodansky et al 2004)

Previous analyses have estimated the carbon containedin internationally traded crop biomass (Ciais et al 2007) andthe embodied emissions from industrial production (Petersand Hertwich 2008b) and have highlighted the importanceof producer and consumer responsibility for carbon emissions(Bastianoni et al 2004) Recent studies have explicitly calledfor the inclusion of land use related greenhouse gas impactsof soybean and beef production (Garnett 2009 Lehuger et al

2009) This study extends previously developed methodsby aggregating new land use datasets and models to trackcarbon emissions from land use to the resulting agriculturalcommodities

This study aims to quantify the hidden GHG emissionsof food production from the Amazon basin of Brazil aregion where rates of both deforestation and agriculturalproduction for export are high and to develop mechanismsto quantify and ultimately reduce the GHG impact of foodproduction Specifically our study provides an analysis ofGHGs embodied in exported beef and soybeans from theBrazilian Amazon explicitly accounting for land use We alsopropose an approach to allocate GHG emissions associatedwith agricultural land use change between producers andconsumers by integrating methods from land use science andlife-cycle analysis

2 Producer versus consumer

In the current Kyoto protocol GHG emissions are allocatedto the country in which the emission occurred Futureinternationally-binding agreements are likely to incentivizecountries to reduce GHG emissions throughout the life cyclesof the goods they produce (Bodansky et al 2004) When goodsare destined for consumption in other countries the emissionsgenerated in their production are referred to as the lsquoemissionsembodied in tradersquo (Ahmad and Wyckoff 2003) This can be asignificant fraction of global carbon emissions using a globaltrade model Peters and Hertwich (2008a 2008b) estimatedthat in 2001 roughly 23 (or sim57 Gt CO2) of energy relatedemissions were embodied in trade

lsquoCarbon leakagersquo occurs when a country opts to limit itsown carbon emissions by importing goods from a country thatdoes not participate in carbon-reduction agreements Carbonleakage is a noted problem of the current Kyoto protocoland has been estimated to comprise 11 of productionemissions (Peters and Hertwich 2008b) Consumption-basedGHG inventories account for emissions from production andimports and subtracts embodied emissions exported in trade(Peters and Hertwich 2008a) Allocating embodied emissionsto the consumer avoids carbon leakage amongst otherdeficiencies of production-based greenhouse gas inventories

Assigning the responsibility for carbon emissions to eitherproducers or consumers should not be a binary decision afairer allocation scheme is needed (Munksgaard and Pedersen2001 Gupta and Bhandari 1999) If responsibility is givento the producer carbon leakage can occur and if it isassigned to the consumer not participating in a global GHG-reduction agreement the responsibility for the emissions arenot taken (Andrew and Forgie 2008) Hence several authorshave put forth allocation schemes in which carbon emissionsare shared between producers and consumers (Lenzen et al2007 Rodrigues and Domingos 2008 Bastianoni et al 2004)These shared allocation schemes provide economic incentiveto the consumer nation to favor products with the smallestenvironmental impacts and thereby push producers to reducethe carbon emissions embodied in their products

2

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 1 The lsquopulsersquo of GHG emissions from land use can be amortized over (a) a single year (b) 20 years constant or (c) 20 years linearlydeclining

To date calculations of emissions embodied in trade haveonly been completed for manufactured products where dataon emissions from fossil fuels are readily available Life-cycleemissions embodied in agricultural products require a differentset of methodologies that have recently been highlightedby the agro-fuels community (Gibbs et al 2008 Panichelliet al 2009 Reijnders and Huijbregts 2008) Quantifying theGHG emissions embodied in a product is a necessary stepto distribute the responsibility for the impacts between theproducers and consumers

3 Allocation of land use emissions

Life-cycle assessments (LCAs) have helped to illuminate thelsquocradle to graversquo ecological impacts for a select numberof manufactured and agricultural products The LCAs ofagricultural products are markedly different from those ofmanufactured products especially if the product originatedfrom an area that recently underwent land use change Withevery transformation of land for agricultural use biophysicalimpacts occur over various spatial and temporal scales (Foleyet al 2005) When the conversion process includes removingaboveground biomass from the site a large lsquopulsersquo of GHGsis released to the atmosphere by burning or decay of theremoved vegetation (Ramankutty et al 2007) Sometimes thepulse of GHG emissions is nearly instantaneous (from burningbiomass) or it may decay slowly as forest slash or secondaryproducts (eg paper wood products) When consideringagricultural life-cycle assessments the analysis domain mustinclude impacts from lsquofield to forkrsquo since activities such asland clearing can overshadow efficiency gains in other areas ofthe product life-cycle (Gibbs et al 2008 Fargione et al 2008)

The time frame of land use varies greatly Cleared landcan transition between forest agriculture fallow and bareground as the fertility of the land changes or changes to thecropping system are introduced Depending on the locationand intensity of the new agricultural operation the land mayremain in production for as little as a single season or as longas several millennia Each of these states have different netcarbon balances as vegetation biomass regrows or is cleared

(Ramankutty et al 2007) Therefore the GHG liability fromthe initial transformation needs to be tracked over time andallocated to the appropriate user

Several amortization schemes can be used to distributethe lsquopulsersquo of GHG emissions over the duration of subsequentland use although none have been widely adopted Herewe briefly describe current methodologies and present a newhybrid approach that combines the best features from otherdescribed methods

4 Methods

The approach taken in the Ecoinvent LCA database(wwwecoinventch) assigns all land use emissions to theproduct that is harvested in the year of land conversionwithout consideration of the ultimate duration of agriculturalproduction (Jungbluth et al 2007) (figure 1(a)) The rationalefor this approach is that since deforestation causes irreversibledamage its impacts should not be amortized over a long period(Jungbluth 2009) If these impacts are monetized on a carboncost-basis and conferred to the importing country the resultingelevated price of goods becomes a disincentive for the futureconversion of land with large stocks of carbon If the land isused for agricultural production in later years the successiveproducts would not incur any carbon debt and would be lsquofree-ridingrsquo on the price paid in the initial year

While assigning all emissions to the first year is not anoptimal solution dispersing the cost of the emissions overa very long time horizon (eg 500 years) is also untenableand a middle ground needs to be explored (Canals et al2007a) An analog can be found in international accountingstandards which assign lsquouseful livesrsquo for products (Canalset al 2007b) Several LCAs and carbon footprinting standards(Muys and Garcıa Quijano 2002 BSI 2008) have adoptedmethods that amortize emissions uniformly over a 20 yr timehorizon (figure 1(b)) This time frame is practical for continuedoccupation before the land is abandoned does not confer anundue economic burden on the producer and still values thecarbon emitted in the land conversion process

3

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

However neither of these methods satisfies the goal ofsending a price signal to reduce deforestation without applyingdisproportional financial pressure on either the producer orconsumer A hybrid approach would allocate the bulk ofemissions to the years directly after the land use changeoccurred and gradually decrease the carbon allocation to theagricultural products derived from the cleared land as timewent on (figure 1(c)) With higher additional costs for thefirst several years a disincentive signal is introduced into themarket but not with disproportionate force In later years theremainder of the carbon costs are captured but at a reducedamount per year Similar to the previous case the time horizonover which the cost of the carbon is collected is arbitrary butis a length that can be societally determined All three casesassume that the land would stay in agricultural production forthe entire duration of the allocation period which may not berealistic in a market and environmentally sensitive area such asthe Brazilian Amazon

These amortization schemes are designed with inherentflexibility that makes them applicable across a range of localto global carbon-trading mechanisms Proposed programshave taken a broad-brush approach to monetizing carbonemissions from deforestation and agricultural production andare important building blocks in accounting for the life-cycle environmental impacts of agricultural production Thereduced emissions from deforestation and degradation (REDD)mechanism would compensate tropical developing countriesfor reducing deforestation rates (Mollicone et al 2007 Gullisonet al 2007) On a smaller scale groups such as Alianca daTerra in Brazil (wwwaliancadaterraorgbr) provide paymentsto farmers for more sustainable production methods by sellingtheir products at a premium These programs rely on themonetization of carbon emissions that is currently determinedby regional carbon markets These examples assume that theprice of carbon emissions are enough to reduce profit marginsand create a disincentive to production on newly cleared land

5 Case study

51 Deforestation in the Amazon

While land useland cover change and deforestation aregrowing concerns in all tropical regions the Brazilian Amazonhas been under intense pressure from national colonizationand agriculture programs and more recently due to increasedproduction of soybeans and cattle for export (Barreto et al2006) The Amazon is the largest contiguous tropical foreston the planet with vast stores of biodiversity and carbonand provides essential ecosystem services to people withinthe basin and around the world (Foley et al 2007) but alsoaccounted for more than half of global deforestation from2000 to 2005 (Hansen et al 2008) and thus for a substantialportion of carbon emissions to the atmosphere lsquoBusiness asusualrsquo scenarios of future demand for goods and governmentalpolicies suggest that deforestation and its attendant problemswill continue (Soares Filho et al 2006)

By 2007 18 of the Legal Amazon had been deforestedby smallholder as well as large holder mechanized agricultural

operators loggers and cattle ranchers among other actors(Barreto et al 2006) In the last decade mechanized agriculture(primarily soybean cultivation) and intensive cattle grazinghave been the dominant drivers of land clearing (Simon andGaragorry 2006 McAlpine et al 2009) Between 1990 and2006 the cattle herd in the Amazon almost tripled in sizeand the area used for soybean cultivation quadrupled so thatby 2006 cattle occupied 95 of the pastoral landscape andsoybeans had more than doubled their share of land (IBGE2009)

52 Transition to an export market

When large-scale deforestation in the Amazon began inthe 1970s the resultant agricultural products were mostlyconsumed within the region The Amazon did not produceenough beef to feed its own population until 1991 (Kaimowitzet al 2004) Since that time national incentives and globaldemand have transformed Brazil into the worldrsquos largestexporter of soybeans and beef among other commodities(Nepstad et al 2006) Most exports are in the form of freshor frozen beef although there is an increasing trend of livecattle exports (ALICEweb 2009) Between 1990 and 2006market and trade reforms in addition to the eradication of foot-and-mouth disease helped exports of beef from the Amazon togrow over 500 (IBGE 2009 Walker et al 2008) This growthhas had environmental consequences such as carbon emissionsfrom deforestation nutrient pollution biodiversity loss anddisplacement of local people (Betts et al 2008 Fearnside 2008Gibbs et al 2010 Foley et al 2007)

While cattle production is the predominant land use inthe Amazon soybeans have recently begun to encroach fromthe southern and eastern boundaries and are responsible fornew land clearing and displacement of cattle pastures (Vera-Diaz et al 2008) Several factors ranging from developmentof moisture-tolerant soybean varieties to increasing globaldemand for animal rations have led to the dramatic increasein soybean production and its rising percentage of the globalgrains market (Nepstad et al 2006) The Brazilian agriculturalcomplex is highly integrated into the global market system asBrazil exports more than 10 of the internationally traded cropbiomass (Ciais et al 2007)

Here we present an illustrative example that distributes theresponsibility for GHG emissions from deforestation betweenBrazil and the eventual importing nation of commoditiesproduced in the Amazon (figure 2) While the data andmethods presented here are considered to be the lsquostateof the sciencersquo the exact parameters allocating emissionsbetween international actors were chosen to exemplify theimportance of GHG emissions embodied in internationallytraded agricultural commodities as a template for futurepolicies Future work can build upon this framework makingdifferent policy assumptions as appropriate

53 Model description

Here we use a land use transition and carbon emission modeldescribed in detail by Ramankutty et al (2007) to estimatethe fate of deforested land and the associated carbon dioxide

4

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 2 Three major phases in the process of converting forests for the production and eventual consumption of beef and soybeans (1) landuseland cover change (2) agriculture and (3) trade

Table 1 Average biomass mean annual deforestation and change in deforestation rate between 1990 and 2006 for the nine states within theBrazilian Amazon (Sources INPE 2009 Saatchi et al 2007)

StateMean biomassvalue (Mg haminus1)

Mean annualdeforestation1990ndash2006 (km2)

Increase (+) or decrease (minus)in deforestation rate between1990 and 2006

Acre 204 583 +Amazonas 270 861 +Amapa 263 75 minusMaranhao 125 910 +Mato Grosso 164 6781 +Para 230 5703 +Rondonia 211 2633 +Roraima 215 246 minusTocantins 95 331 minus

equivalent (CO2e) emissions (figures 3(a) and (b)) While thepreviously published model is robust several key changes weremade to incorporate the latest published data and changingagricultural practices in the Brazilian Amazon The modelwas run for each of the nine states in the legal Amazon from1990 to 2006 using (time-smoothed) annual deforestation ratesprovided by Brazilrsquos National Institute for Space Research(INPE) (2009) Initial aboveground forest biomass values forland identified as intact forest by Brazilrsquos program to calculatethe deforestation of the Amazon (PRODES) were summarizedby state from Saatchi et al (2007) (table 1) In each yearcarbon emissions from deforestation are calculated as thesum of combusted biomass (20) and the decay of biomassremaining in the slash (70) product (8) and elemental (2)pools (Ramankutty et al 2007) (figures 4(a) and (b)) For thestate of Mato Grosso the percentages of biomass allocated tothe slash and burn pools were set to 20 and 70 respectivelyto reflect the trend of highly mechanized agriculture wheremost limbs and stumps are removed and burned after forestclearing (Galford et al 2010) The model partitions deforestedland between cropland pasture and secondary forest using

transition rates initially described by Fearnside (1996) butland transition rates for the state of Mato Grosso a hotspot ofdeforestation were updated according to the patterns reportedin Morton et al (2006) and Defries et al (2008) In addition tocarbon emissions from deforestation added methane emissionsfrom enteric fermentation in cattle were included in the totalCO2 equivalent (CO2e) flux and were computed with IPCCvalues for tropical cattle assuming a 100 yr global warmingpotential (IPCC 2007 Robertson and Grace 2004) GHGemissions from the application of fertilizer were found to benegligible compared to fluxes from deforestation and methaneproduction

54 Carbon allocation methods

The model distributes deforested land as it shifts betweencropland pasture and secondary forests computing the GHGemissions (or sequestration for the secondary forests) ofeach land use type While the model estimates the netemissions for new pastures and croplands the major exportcommodities (cattle and soybeans) only occupy a percentage

5

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(b)

(a)

Figure 3 Model simulation of the legal Amazon (a) land use and (b) carbon emissions from 1990 to 2006 The model distributes deforestedland (and resulting carbon emissions) between farms pastures and secondary forests Methane emissions from enteric fermentation in cattleare included within the emissions from lsquopasturersquo

of the agricultural landscape For each product the relativedominance on the landscape each year is calculated usingthe ratio of area planted in soybeans to total agriculturalarea and a similar ratio of pasture area for cattle tobuffalo horses sheep and other pastured animals (IBGE2009) The landscape dominance modifiers were appliedto each statersquos carbon emissions in every year of themodel

Export modifiers are calculated by dividing the amountof soybeans exported by the amount produced For cattleapproximately 10 of the herd is slaughtered every year andthe average carcass yield is used to calculate the beef andassociated products for a given year (IBGE 2009 FAOSTAT2009) Live cattle exports are incorporated with the beefexport data using average carcass yields (FAOSTAT 2009)The ratio of beef exported to beef produced was then estimated

(ALICEweb 2009) The export modifiers were applied over theaggregate Amazon to avoid data inconsistencies resulting frominterstate trade for each year of the model

The emissions calculated for each year are the sum ofburnt decay slash and elemental carbon from land deforestedin previous years minus land that transitioned to secondaryforest The allocation of carbon emissions over time werecalculated using three scenarios (1) emissions were allocatedto the year they occurred (2) equally distributed over 20 yearsand (3) linearly decreasing over 20 years (figure 1) Carbonfluxes from deforestation prior to 1990 were not includedand emission rates should therefore be seen as underestimatesAlso as future land use agricultural production and exportpatterns are unknown carbon from recent deforestation eventscannot be definitively allocated but must be tracked in order toassign future responsibility

6

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 3: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

The current production methods of the global food systemhelp sustain our livelihoods but the extent and intensivepractices of modern agriculture have substantial negativeenvironmental consequences (Foley et al 2005 Roy et al2009 Schau and Fet 2008) For example agricultural landuse is responsible for the release of greenhouse gases (GHG)biodiversity loss eutrophication of waterways emergence ofdisease and changes in local and regional climates all ofwhich detract from human health and security (MEA 2005)In economic terms the extent and severity of these negativeconsequences are typically externalities of the economicsystem because they are rarely communicated to the consumeror accounted for in the price of agricultural products (Prettyet al 2000)

In addition agricultural products are part of anincreasingly globalized food system that separates producersand consumers by thousands of kilometers and lengthy supply-chains The impacts of production span from local (eg airand water pollution) to global (eg greenhouse gas emissions)scales (Tilman 1999 Smith et al 2008) and the mechanisms totrack and account for these impacts are poorly developed

As markets become more globalized the production ofcash crops and other export commodities is expected toincrease This will likely lead to expansion of agriculturalland in the tropics the region that has the most arable landnot currently in production (Alexandratos et al 2006 Barbier2000) Such expansion could have serious implications forGHG emissions as did land use emissions from tropicalregions in the 1990s (Houghton 2003) Moreover Gibbs et al(2010) found that more than half of new agricultural land in thetropics originated from intact forest with another third comingfrom previously cleared forests Although signatories to theKyoto protocol are working to reduce their GHG emissionsfrom within-country fossil fuel sources they have neglectedland use emissions including those stemming from theiragricultural imports

Rising concern about GHG emissions an increasinglyinformed public and the threat of regulatory action haveprompted producers in the global food system and otherenergy intensive sectors to measure the energy life cyclesof their products (Brentrup et al 2004 Jolliet et al 2003Goleman 2009) Some producers are voluntarily providingconsumers with estimates of the life-cycle energy costs of theproduction distribution and consumption of their products toenable consumers to choose goods with the smallest energyfootprints (Gallastegui 2002) Current proposals suggest thatcarbon will become a regulated commodity under future globalclimate agreements and the disclosure of the energy used inthe production of commercial goods including agriculturalproducts will be necessary (Bodansky et al 2004)

Previous analyses have estimated the carbon containedin internationally traded crop biomass (Ciais et al 2007) andthe embodied emissions from industrial production (Petersand Hertwich 2008b) and have highlighted the importanceof producer and consumer responsibility for carbon emissions(Bastianoni et al 2004) Recent studies have explicitly calledfor the inclusion of land use related greenhouse gas impactsof soybean and beef production (Garnett 2009 Lehuger et al

2009) This study extends previously developed methodsby aggregating new land use datasets and models to trackcarbon emissions from land use to the resulting agriculturalcommodities

This study aims to quantify the hidden GHG emissionsof food production from the Amazon basin of Brazil aregion where rates of both deforestation and agriculturalproduction for export are high and to develop mechanismsto quantify and ultimately reduce the GHG impact of foodproduction Specifically our study provides an analysis ofGHGs embodied in exported beef and soybeans from theBrazilian Amazon explicitly accounting for land use We alsopropose an approach to allocate GHG emissions associatedwith agricultural land use change between producers andconsumers by integrating methods from land use science andlife-cycle analysis

2 Producer versus consumer

In the current Kyoto protocol GHG emissions are allocatedto the country in which the emission occurred Futureinternationally-binding agreements are likely to incentivizecountries to reduce GHG emissions throughout the life cyclesof the goods they produce (Bodansky et al 2004) When goodsare destined for consumption in other countries the emissionsgenerated in their production are referred to as the lsquoemissionsembodied in tradersquo (Ahmad and Wyckoff 2003) This can be asignificant fraction of global carbon emissions using a globaltrade model Peters and Hertwich (2008a 2008b) estimatedthat in 2001 roughly 23 (or sim57 Gt CO2) of energy relatedemissions were embodied in trade

lsquoCarbon leakagersquo occurs when a country opts to limit itsown carbon emissions by importing goods from a country thatdoes not participate in carbon-reduction agreements Carbonleakage is a noted problem of the current Kyoto protocoland has been estimated to comprise 11 of productionemissions (Peters and Hertwich 2008b) Consumption-basedGHG inventories account for emissions from production andimports and subtracts embodied emissions exported in trade(Peters and Hertwich 2008a) Allocating embodied emissionsto the consumer avoids carbon leakage amongst otherdeficiencies of production-based greenhouse gas inventories

Assigning the responsibility for carbon emissions to eitherproducers or consumers should not be a binary decision afairer allocation scheme is needed (Munksgaard and Pedersen2001 Gupta and Bhandari 1999) If responsibility is givento the producer carbon leakage can occur and if it isassigned to the consumer not participating in a global GHG-reduction agreement the responsibility for the emissions arenot taken (Andrew and Forgie 2008) Hence several authorshave put forth allocation schemes in which carbon emissionsare shared between producers and consumers (Lenzen et al2007 Rodrigues and Domingos 2008 Bastianoni et al 2004)These shared allocation schemes provide economic incentiveto the consumer nation to favor products with the smallestenvironmental impacts and thereby push producers to reducethe carbon emissions embodied in their products

2

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 1 The lsquopulsersquo of GHG emissions from land use can be amortized over (a) a single year (b) 20 years constant or (c) 20 years linearlydeclining

To date calculations of emissions embodied in trade haveonly been completed for manufactured products where dataon emissions from fossil fuels are readily available Life-cycleemissions embodied in agricultural products require a differentset of methodologies that have recently been highlightedby the agro-fuels community (Gibbs et al 2008 Panichelliet al 2009 Reijnders and Huijbregts 2008) Quantifying theGHG emissions embodied in a product is a necessary stepto distribute the responsibility for the impacts between theproducers and consumers

3 Allocation of land use emissions

Life-cycle assessments (LCAs) have helped to illuminate thelsquocradle to graversquo ecological impacts for a select numberof manufactured and agricultural products The LCAs ofagricultural products are markedly different from those ofmanufactured products especially if the product originatedfrom an area that recently underwent land use change Withevery transformation of land for agricultural use biophysicalimpacts occur over various spatial and temporal scales (Foleyet al 2005) When the conversion process includes removingaboveground biomass from the site a large lsquopulsersquo of GHGsis released to the atmosphere by burning or decay of theremoved vegetation (Ramankutty et al 2007) Sometimes thepulse of GHG emissions is nearly instantaneous (from burningbiomass) or it may decay slowly as forest slash or secondaryproducts (eg paper wood products) When consideringagricultural life-cycle assessments the analysis domain mustinclude impacts from lsquofield to forkrsquo since activities such asland clearing can overshadow efficiency gains in other areas ofthe product life-cycle (Gibbs et al 2008 Fargione et al 2008)

The time frame of land use varies greatly Cleared landcan transition between forest agriculture fallow and bareground as the fertility of the land changes or changes to thecropping system are introduced Depending on the locationand intensity of the new agricultural operation the land mayremain in production for as little as a single season or as longas several millennia Each of these states have different netcarbon balances as vegetation biomass regrows or is cleared

(Ramankutty et al 2007) Therefore the GHG liability fromthe initial transformation needs to be tracked over time andallocated to the appropriate user

Several amortization schemes can be used to distributethe lsquopulsersquo of GHG emissions over the duration of subsequentland use although none have been widely adopted Herewe briefly describe current methodologies and present a newhybrid approach that combines the best features from otherdescribed methods

4 Methods

The approach taken in the Ecoinvent LCA database(wwwecoinventch) assigns all land use emissions to theproduct that is harvested in the year of land conversionwithout consideration of the ultimate duration of agriculturalproduction (Jungbluth et al 2007) (figure 1(a)) The rationalefor this approach is that since deforestation causes irreversibledamage its impacts should not be amortized over a long period(Jungbluth 2009) If these impacts are monetized on a carboncost-basis and conferred to the importing country the resultingelevated price of goods becomes a disincentive for the futureconversion of land with large stocks of carbon If the land isused for agricultural production in later years the successiveproducts would not incur any carbon debt and would be lsquofree-ridingrsquo on the price paid in the initial year

While assigning all emissions to the first year is not anoptimal solution dispersing the cost of the emissions overa very long time horizon (eg 500 years) is also untenableand a middle ground needs to be explored (Canals et al2007a) An analog can be found in international accountingstandards which assign lsquouseful livesrsquo for products (Canalset al 2007b) Several LCAs and carbon footprinting standards(Muys and Garcıa Quijano 2002 BSI 2008) have adoptedmethods that amortize emissions uniformly over a 20 yr timehorizon (figure 1(b)) This time frame is practical for continuedoccupation before the land is abandoned does not confer anundue economic burden on the producer and still values thecarbon emitted in the land conversion process

3

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

However neither of these methods satisfies the goal ofsending a price signal to reduce deforestation without applyingdisproportional financial pressure on either the producer orconsumer A hybrid approach would allocate the bulk ofemissions to the years directly after the land use changeoccurred and gradually decrease the carbon allocation to theagricultural products derived from the cleared land as timewent on (figure 1(c)) With higher additional costs for thefirst several years a disincentive signal is introduced into themarket but not with disproportionate force In later years theremainder of the carbon costs are captured but at a reducedamount per year Similar to the previous case the time horizonover which the cost of the carbon is collected is arbitrary butis a length that can be societally determined All three casesassume that the land would stay in agricultural production forthe entire duration of the allocation period which may not berealistic in a market and environmentally sensitive area such asthe Brazilian Amazon

These amortization schemes are designed with inherentflexibility that makes them applicable across a range of localto global carbon-trading mechanisms Proposed programshave taken a broad-brush approach to monetizing carbonemissions from deforestation and agricultural production andare important building blocks in accounting for the life-cycle environmental impacts of agricultural production Thereduced emissions from deforestation and degradation (REDD)mechanism would compensate tropical developing countriesfor reducing deforestation rates (Mollicone et al 2007 Gullisonet al 2007) On a smaller scale groups such as Alianca daTerra in Brazil (wwwaliancadaterraorgbr) provide paymentsto farmers for more sustainable production methods by sellingtheir products at a premium These programs rely on themonetization of carbon emissions that is currently determinedby regional carbon markets These examples assume that theprice of carbon emissions are enough to reduce profit marginsand create a disincentive to production on newly cleared land

5 Case study

51 Deforestation in the Amazon

While land useland cover change and deforestation aregrowing concerns in all tropical regions the Brazilian Amazonhas been under intense pressure from national colonizationand agriculture programs and more recently due to increasedproduction of soybeans and cattle for export (Barreto et al2006) The Amazon is the largest contiguous tropical foreston the planet with vast stores of biodiversity and carbonand provides essential ecosystem services to people withinthe basin and around the world (Foley et al 2007) but alsoaccounted for more than half of global deforestation from2000 to 2005 (Hansen et al 2008) and thus for a substantialportion of carbon emissions to the atmosphere lsquoBusiness asusualrsquo scenarios of future demand for goods and governmentalpolicies suggest that deforestation and its attendant problemswill continue (Soares Filho et al 2006)

By 2007 18 of the Legal Amazon had been deforestedby smallholder as well as large holder mechanized agricultural

operators loggers and cattle ranchers among other actors(Barreto et al 2006) In the last decade mechanized agriculture(primarily soybean cultivation) and intensive cattle grazinghave been the dominant drivers of land clearing (Simon andGaragorry 2006 McAlpine et al 2009) Between 1990 and2006 the cattle herd in the Amazon almost tripled in sizeand the area used for soybean cultivation quadrupled so thatby 2006 cattle occupied 95 of the pastoral landscape andsoybeans had more than doubled their share of land (IBGE2009)

52 Transition to an export market

When large-scale deforestation in the Amazon began inthe 1970s the resultant agricultural products were mostlyconsumed within the region The Amazon did not produceenough beef to feed its own population until 1991 (Kaimowitzet al 2004) Since that time national incentives and globaldemand have transformed Brazil into the worldrsquos largestexporter of soybeans and beef among other commodities(Nepstad et al 2006) Most exports are in the form of freshor frozen beef although there is an increasing trend of livecattle exports (ALICEweb 2009) Between 1990 and 2006market and trade reforms in addition to the eradication of foot-and-mouth disease helped exports of beef from the Amazon togrow over 500 (IBGE 2009 Walker et al 2008) This growthhas had environmental consequences such as carbon emissionsfrom deforestation nutrient pollution biodiversity loss anddisplacement of local people (Betts et al 2008 Fearnside 2008Gibbs et al 2010 Foley et al 2007)

While cattle production is the predominant land use inthe Amazon soybeans have recently begun to encroach fromthe southern and eastern boundaries and are responsible fornew land clearing and displacement of cattle pastures (Vera-Diaz et al 2008) Several factors ranging from developmentof moisture-tolerant soybean varieties to increasing globaldemand for animal rations have led to the dramatic increasein soybean production and its rising percentage of the globalgrains market (Nepstad et al 2006) The Brazilian agriculturalcomplex is highly integrated into the global market system asBrazil exports more than 10 of the internationally traded cropbiomass (Ciais et al 2007)

Here we present an illustrative example that distributes theresponsibility for GHG emissions from deforestation betweenBrazil and the eventual importing nation of commoditiesproduced in the Amazon (figure 2) While the data andmethods presented here are considered to be the lsquostateof the sciencersquo the exact parameters allocating emissionsbetween international actors were chosen to exemplify theimportance of GHG emissions embodied in internationallytraded agricultural commodities as a template for futurepolicies Future work can build upon this framework makingdifferent policy assumptions as appropriate

53 Model description

Here we use a land use transition and carbon emission modeldescribed in detail by Ramankutty et al (2007) to estimatethe fate of deforested land and the associated carbon dioxide

4

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 2 Three major phases in the process of converting forests for the production and eventual consumption of beef and soybeans (1) landuseland cover change (2) agriculture and (3) trade

Table 1 Average biomass mean annual deforestation and change in deforestation rate between 1990 and 2006 for the nine states within theBrazilian Amazon (Sources INPE 2009 Saatchi et al 2007)

StateMean biomassvalue (Mg haminus1)

Mean annualdeforestation1990ndash2006 (km2)

Increase (+) or decrease (minus)in deforestation rate between1990 and 2006

Acre 204 583 +Amazonas 270 861 +Amapa 263 75 minusMaranhao 125 910 +Mato Grosso 164 6781 +Para 230 5703 +Rondonia 211 2633 +Roraima 215 246 minusTocantins 95 331 minus

equivalent (CO2e) emissions (figures 3(a) and (b)) While thepreviously published model is robust several key changes weremade to incorporate the latest published data and changingagricultural practices in the Brazilian Amazon The modelwas run for each of the nine states in the legal Amazon from1990 to 2006 using (time-smoothed) annual deforestation ratesprovided by Brazilrsquos National Institute for Space Research(INPE) (2009) Initial aboveground forest biomass values forland identified as intact forest by Brazilrsquos program to calculatethe deforestation of the Amazon (PRODES) were summarizedby state from Saatchi et al (2007) (table 1) In each yearcarbon emissions from deforestation are calculated as thesum of combusted biomass (20) and the decay of biomassremaining in the slash (70) product (8) and elemental (2)pools (Ramankutty et al 2007) (figures 4(a) and (b)) For thestate of Mato Grosso the percentages of biomass allocated tothe slash and burn pools were set to 20 and 70 respectivelyto reflect the trend of highly mechanized agriculture wheremost limbs and stumps are removed and burned after forestclearing (Galford et al 2010) The model partitions deforestedland between cropland pasture and secondary forest using

transition rates initially described by Fearnside (1996) butland transition rates for the state of Mato Grosso a hotspot ofdeforestation were updated according to the patterns reportedin Morton et al (2006) and Defries et al (2008) In addition tocarbon emissions from deforestation added methane emissionsfrom enteric fermentation in cattle were included in the totalCO2 equivalent (CO2e) flux and were computed with IPCCvalues for tropical cattle assuming a 100 yr global warmingpotential (IPCC 2007 Robertson and Grace 2004) GHGemissions from the application of fertilizer were found to benegligible compared to fluxes from deforestation and methaneproduction

54 Carbon allocation methods

The model distributes deforested land as it shifts betweencropland pasture and secondary forests computing the GHGemissions (or sequestration for the secondary forests) ofeach land use type While the model estimates the netemissions for new pastures and croplands the major exportcommodities (cattle and soybeans) only occupy a percentage

5

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(b)

(a)

Figure 3 Model simulation of the legal Amazon (a) land use and (b) carbon emissions from 1990 to 2006 The model distributes deforestedland (and resulting carbon emissions) between farms pastures and secondary forests Methane emissions from enteric fermentation in cattleare included within the emissions from lsquopasturersquo

of the agricultural landscape For each product the relativedominance on the landscape each year is calculated usingthe ratio of area planted in soybeans to total agriculturalarea and a similar ratio of pasture area for cattle tobuffalo horses sheep and other pastured animals (IBGE2009) The landscape dominance modifiers were appliedto each statersquos carbon emissions in every year of themodel

Export modifiers are calculated by dividing the amountof soybeans exported by the amount produced For cattleapproximately 10 of the herd is slaughtered every year andthe average carcass yield is used to calculate the beef andassociated products for a given year (IBGE 2009 FAOSTAT2009) Live cattle exports are incorporated with the beefexport data using average carcass yields (FAOSTAT 2009)The ratio of beef exported to beef produced was then estimated

(ALICEweb 2009) The export modifiers were applied over theaggregate Amazon to avoid data inconsistencies resulting frominterstate trade for each year of the model

The emissions calculated for each year are the sum ofburnt decay slash and elemental carbon from land deforestedin previous years minus land that transitioned to secondaryforest The allocation of carbon emissions over time werecalculated using three scenarios (1) emissions were allocatedto the year they occurred (2) equally distributed over 20 yearsand (3) linearly decreasing over 20 years (figure 1) Carbonfluxes from deforestation prior to 1990 were not includedand emission rates should therefore be seen as underestimatesAlso as future land use agricultural production and exportpatterns are unknown carbon from recent deforestation eventscannot be definitively allocated but must be tracked in order toassign future responsibility

6

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 4: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 1 The lsquopulsersquo of GHG emissions from land use can be amortized over (a) a single year (b) 20 years constant or (c) 20 years linearlydeclining

To date calculations of emissions embodied in trade haveonly been completed for manufactured products where dataon emissions from fossil fuels are readily available Life-cycleemissions embodied in agricultural products require a differentset of methodologies that have recently been highlightedby the agro-fuels community (Gibbs et al 2008 Panichelliet al 2009 Reijnders and Huijbregts 2008) Quantifying theGHG emissions embodied in a product is a necessary stepto distribute the responsibility for the impacts between theproducers and consumers

3 Allocation of land use emissions

Life-cycle assessments (LCAs) have helped to illuminate thelsquocradle to graversquo ecological impacts for a select numberof manufactured and agricultural products The LCAs ofagricultural products are markedly different from those ofmanufactured products especially if the product originatedfrom an area that recently underwent land use change Withevery transformation of land for agricultural use biophysicalimpacts occur over various spatial and temporal scales (Foleyet al 2005) When the conversion process includes removingaboveground biomass from the site a large lsquopulsersquo of GHGsis released to the atmosphere by burning or decay of theremoved vegetation (Ramankutty et al 2007) Sometimes thepulse of GHG emissions is nearly instantaneous (from burningbiomass) or it may decay slowly as forest slash or secondaryproducts (eg paper wood products) When consideringagricultural life-cycle assessments the analysis domain mustinclude impacts from lsquofield to forkrsquo since activities such asland clearing can overshadow efficiency gains in other areas ofthe product life-cycle (Gibbs et al 2008 Fargione et al 2008)

The time frame of land use varies greatly Cleared landcan transition between forest agriculture fallow and bareground as the fertility of the land changes or changes to thecropping system are introduced Depending on the locationand intensity of the new agricultural operation the land mayremain in production for as little as a single season or as longas several millennia Each of these states have different netcarbon balances as vegetation biomass regrows or is cleared

(Ramankutty et al 2007) Therefore the GHG liability fromthe initial transformation needs to be tracked over time andallocated to the appropriate user

Several amortization schemes can be used to distributethe lsquopulsersquo of GHG emissions over the duration of subsequentland use although none have been widely adopted Herewe briefly describe current methodologies and present a newhybrid approach that combines the best features from otherdescribed methods

4 Methods

The approach taken in the Ecoinvent LCA database(wwwecoinventch) assigns all land use emissions to theproduct that is harvested in the year of land conversionwithout consideration of the ultimate duration of agriculturalproduction (Jungbluth et al 2007) (figure 1(a)) The rationalefor this approach is that since deforestation causes irreversibledamage its impacts should not be amortized over a long period(Jungbluth 2009) If these impacts are monetized on a carboncost-basis and conferred to the importing country the resultingelevated price of goods becomes a disincentive for the futureconversion of land with large stocks of carbon If the land isused for agricultural production in later years the successiveproducts would not incur any carbon debt and would be lsquofree-ridingrsquo on the price paid in the initial year

While assigning all emissions to the first year is not anoptimal solution dispersing the cost of the emissions overa very long time horizon (eg 500 years) is also untenableand a middle ground needs to be explored (Canals et al2007a) An analog can be found in international accountingstandards which assign lsquouseful livesrsquo for products (Canalset al 2007b) Several LCAs and carbon footprinting standards(Muys and Garcıa Quijano 2002 BSI 2008) have adoptedmethods that amortize emissions uniformly over a 20 yr timehorizon (figure 1(b)) This time frame is practical for continuedoccupation before the land is abandoned does not confer anundue economic burden on the producer and still values thecarbon emitted in the land conversion process

3

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

However neither of these methods satisfies the goal ofsending a price signal to reduce deforestation without applyingdisproportional financial pressure on either the producer orconsumer A hybrid approach would allocate the bulk ofemissions to the years directly after the land use changeoccurred and gradually decrease the carbon allocation to theagricultural products derived from the cleared land as timewent on (figure 1(c)) With higher additional costs for thefirst several years a disincentive signal is introduced into themarket but not with disproportionate force In later years theremainder of the carbon costs are captured but at a reducedamount per year Similar to the previous case the time horizonover which the cost of the carbon is collected is arbitrary butis a length that can be societally determined All three casesassume that the land would stay in agricultural production forthe entire duration of the allocation period which may not berealistic in a market and environmentally sensitive area such asthe Brazilian Amazon

These amortization schemes are designed with inherentflexibility that makes them applicable across a range of localto global carbon-trading mechanisms Proposed programshave taken a broad-brush approach to monetizing carbonemissions from deforestation and agricultural production andare important building blocks in accounting for the life-cycle environmental impacts of agricultural production Thereduced emissions from deforestation and degradation (REDD)mechanism would compensate tropical developing countriesfor reducing deforestation rates (Mollicone et al 2007 Gullisonet al 2007) On a smaller scale groups such as Alianca daTerra in Brazil (wwwaliancadaterraorgbr) provide paymentsto farmers for more sustainable production methods by sellingtheir products at a premium These programs rely on themonetization of carbon emissions that is currently determinedby regional carbon markets These examples assume that theprice of carbon emissions are enough to reduce profit marginsand create a disincentive to production on newly cleared land

5 Case study

51 Deforestation in the Amazon

While land useland cover change and deforestation aregrowing concerns in all tropical regions the Brazilian Amazonhas been under intense pressure from national colonizationand agriculture programs and more recently due to increasedproduction of soybeans and cattle for export (Barreto et al2006) The Amazon is the largest contiguous tropical foreston the planet with vast stores of biodiversity and carbonand provides essential ecosystem services to people withinthe basin and around the world (Foley et al 2007) but alsoaccounted for more than half of global deforestation from2000 to 2005 (Hansen et al 2008) and thus for a substantialportion of carbon emissions to the atmosphere lsquoBusiness asusualrsquo scenarios of future demand for goods and governmentalpolicies suggest that deforestation and its attendant problemswill continue (Soares Filho et al 2006)

By 2007 18 of the Legal Amazon had been deforestedby smallholder as well as large holder mechanized agricultural

operators loggers and cattle ranchers among other actors(Barreto et al 2006) In the last decade mechanized agriculture(primarily soybean cultivation) and intensive cattle grazinghave been the dominant drivers of land clearing (Simon andGaragorry 2006 McAlpine et al 2009) Between 1990 and2006 the cattle herd in the Amazon almost tripled in sizeand the area used for soybean cultivation quadrupled so thatby 2006 cattle occupied 95 of the pastoral landscape andsoybeans had more than doubled their share of land (IBGE2009)

52 Transition to an export market

When large-scale deforestation in the Amazon began inthe 1970s the resultant agricultural products were mostlyconsumed within the region The Amazon did not produceenough beef to feed its own population until 1991 (Kaimowitzet al 2004) Since that time national incentives and globaldemand have transformed Brazil into the worldrsquos largestexporter of soybeans and beef among other commodities(Nepstad et al 2006) Most exports are in the form of freshor frozen beef although there is an increasing trend of livecattle exports (ALICEweb 2009) Between 1990 and 2006market and trade reforms in addition to the eradication of foot-and-mouth disease helped exports of beef from the Amazon togrow over 500 (IBGE 2009 Walker et al 2008) This growthhas had environmental consequences such as carbon emissionsfrom deforestation nutrient pollution biodiversity loss anddisplacement of local people (Betts et al 2008 Fearnside 2008Gibbs et al 2010 Foley et al 2007)

While cattle production is the predominant land use inthe Amazon soybeans have recently begun to encroach fromthe southern and eastern boundaries and are responsible fornew land clearing and displacement of cattle pastures (Vera-Diaz et al 2008) Several factors ranging from developmentof moisture-tolerant soybean varieties to increasing globaldemand for animal rations have led to the dramatic increasein soybean production and its rising percentage of the globalgrains market (Nepstad et al 2006) The Brazilian agriculturalcomplex is highly integrated into the global market system asBrazil exports more than 10 of the internationally traded cropbiomass (Ciais et al 2007)

Here we present an illustrative example that distributes theresponsibility for GHG emissions from deforestation betweenBrazil and the eventual importing nation of commoditiesproduced in the Amazon (figure 2) While the data andmethods presented here are considered to be the lsquostateof the sciencersquo the exact parameters allocating emissionsbetween international actors were chosen to exemplify theimportance of GHG emissions embodied in internationallytraded agricultural commodities as a template for futurepolicies Future work can build upon this framework makingdifferent policy assumptions as appropriate

53 Model description

Here we use a land use transition and carbon emission modeldescribed in detail by Ramankutty et al (2007) to estimatethe fate of deforested land and the associated carbon dioxide

4

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 2 Three major phases in the process of converting forests for the production and eventual consumption of beef and soybeans (1) landuseland cover change (2) agriculture and (3) trade

Table 1 Average biomass mean annual deforestation and change in deforestation rate between 1990 and 2006 for the nine states within theBrazilian Amazon (Sources INPE 2009 Saatchi et al 2007)

StateMean biomassvalue (Mg haminus1)

Mean annualdeforestation1990ndash2006 (km2)

Increase (+) or decrease (minus)in deforestation rate between1990 and 2006

Acre 204 583 +Amazonas 270 861 +Amapa 263 75 minusMaranhao 125 910 +Mato Grosso 164 6781 +Para 230 5703 +Rondonia 211 2633 +Roraima 215 246 minusTocantins 95 331 minus

equivalent (CO2e) emissions (figures 3(a) and (b)) While thepreviously published model is robust several key changes weremade to incorporate the latest published data and changingagricultural practices in the Brazilian Amazon The modelwas run for each of the nine states in the legal Amazon from1990 to 2006 using (time-smoothed) annual deforestation ratesprovided by Brazilrsquos National Institute for Space Research(INPE) (2009) Initial aboveground forest biomass values forland identified as intact forest by Brazilrsquos program to calculatethe deforestation of the Amazon (PRODES) were summarizedby state from Saatchi et al (2007) (table 1) In each yearcarbon emissions from deforestation are calculated as thesum of combusted biomass (20) and the decay of biomassremaining in the slash (70) product (8) and elemental (2)pools (Ramankutty et al 2007) (figures 4(a) and (b)) For thestate of Mato Grosso the percentages of biomass allocated tothe slash and burn pools were set to 20 and 70 respectivelyto reflect the trend of highly mechanized agriculture wheremost limbs and stumps are removed and burned after forestclearing (Galford et al 2010) The model partitions deforestedland between cropland pasture and secondary forest using

transition rates initially described by Fearnside (1996) butland transition rates for the state of Mato Grosso a hotspot ofdeforestation were updated according to the patterns reportedin Morton et al (2006) and Defries et al (2008) In addition tocarbon emissions from deforestation added methane emissionsfrom enteric fermentation in cattle were included in the totalCO2 equivalent (CO2e) flux and were computed with IPCCvalues for tropical cattle assuming a 100 yr global warmingpotential (IPCC 2007 Robertson and Grace 2004) GHGemissions from the application of fertilizer were found to benegligible compared to fluxes from deforestation and methaneproduction

54 Carbon allocation methods

The model distributes deforested land as it shifts betweencropland pasture and secondary forests computing the GHGemissions (or sequestration for the secondary forests) ofeach land use type While the model estimates the netemissions for new pastures and croplands the major exportcommodities (cattle and soybeans) only occupy a percentage

5

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(b)

(a)

Figure 3 Model simulation of the legal Amazon (a) land use and (b) carbon emissions from 1990 to 2006 The model distributes deforestedland (and resulting carbon emissions) between farms pastures and secondary forests Methane emissions from enteric fermentation in cattleare included within the emissions from lsquopasturersquo

of the agricultural landscape For each product the relativedominance on the landscape each year is calculated usingthe ratio of area planted in soybeans to total agriculturalarea and a similar ratio of pasture area for cattle tobuffalo horses sheep and other pastured animals (IBGE2009) The landscape dominance modifiers were appliedto each statersquos carbon emissions in every year of themodel

Export modifiers are calculated by dividing the amountof soybeans exported by the amount produced For cattleapproximately 10 of the herd is slaughtered every year andthe average carcass yield is used to calculate the beef andassociated products for a given year (IBGE 2009 FAOSTAT2009) Live cattle exports are incorporated with the beefexport data using average carcass yields (FAOSTAT 2009)The ratio of beef exported to beef produced was then estimated

(ALICEweb 2009) The export modifiers were applied over theaggregate Amazon to avoid data inconsistencies resulting frominterstate trade for each year of the model

The emissions calculated for each year are the sum ofburnt decay slash and elemental carbon from land deforestedin previous years minus land that transitioned to secondaryforest The allocation of carbon emissions over time werecalculated using three scenarios (1) emissions were allocatedto the year they occurred (2) equally distributed over 20 yearsand (3) linearly decreasing over 20 years (figure 1) Carbonfluxes from deforestation prior to 1990 were not includedand emission rates should therefore be seen as underestimatesAlso as future land use agricultural production and exportpatterns are unknown carbon from recent deforestation eventscannot be definitively allocated but must be tracked in order toassign future responsibility

6

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 5: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

However neither of these methods satisfies the goal ofsending a price signal to reduce deforestation without applyingdisproportional financial pressure on either the producer orconsumer A hybrid approach would allocate the bulk ofemissions to the years directly after the land use changeoccurred and gradually decrease the carbon allocation to theagricultural products derived from the cleared land as timewent on (figure 1(c)) With higher additional costs for thefirst several years a disincentive signal is introduced into themarket but not with disproportionate force In later years theremainder of the carbon costs are captured but at a reducedamount per year Similar to the previous case the time horizonover which the cost of the carbon is collected is arbitrary butis a length that can be societally determined All three casesassume that the land would stay in agricultural production forthe entire duration of the allocation period which may not berealistic in a market and environmentally sensitive area such asthe Brazilian Amazon

These amortization schemes are designed with inherentflexibility that makes them applicable across a range of localto global carbon-trading mechanisms Proposed programshave taken a broad-brush approach to monetizing carbonemissions from deforestation and agricultural production andare important building blocks in accounting for the life-cycle environmental impacts of agricultural production Thereduced emissions from deforestation and degradation (REDD)mechanism would compensate tropical developing countriesfor reducing deforestation rates (Mollicone et al 2007 Gullisonet al 2007) On a smaller scale groups such as Alianca daTerra in Brazil (wwwaliancadaterraorgbr) provide paymentsto farmers for more sustainable production methods by sellingtheir products at a premium These programs rely on themonetization of carbon emissions that is currently determinedby regional carbon markets These examples assume that theprice of carbon emissions are enough to reduce profit marginsand create a disincentive to production on newly cleared land

5 Case study

51 Deforestation in the Amazon

While land useland cover change and deforestation aregrowing concerns in all tropical regions the Brazilian Amazonhas been under intense pressure from national colonizationand agriculture programs and more recently due to increasedproduction of soybeans and cattle for export (Barreto et al2006) The Amazon is the largest contiguous tropical foreston the planet with vast stores of biodiversity and carbonand provides essential ecosystem services to people withinthe basin and around the world (Foley et al 2007) but alsoaccounted for more than half of global deforestation from2000 to 2005 (Hansen et al 2008) and thus for a substantialportion of carbon emissions to the atmosphere lsquoBusiness asusualrsquo scenarios of future demand for goods and governmentalpolicies suggest that deforestation and its attendant problemswill continue (Soares Filho et al 2006)

By 2007 18 of the Legal Amazon had been deforestedby smallholder as well as large holder mechanized agricultural

operators loggers and cattle ranchers among other actors(Barreto et al 2006) In the last decade mechanized agriculture(primarily soybean cultivation) and intensive cattle grazinghave been the dominant drivers of land clearing (Simon andGaragorry 2006 McAlpine et al 2009) Between 1990 and2006 the cattle herd in the Amazon almost tripled in sizeand the area used for soybean cultivation quadrupled so thatby 2006 cattle occupied 95 of the pastoral landscape andsoybeans had more than doubled their share of land (IBGE2009)

52 Transition to an export market

When large-scale deforestation in the Amazon began inthe 1970s the resultant agricultural products were mostlyconsumed within the region The Amazon did not produceenough beef to feed its own population until 1991 (Kaimowitzet al 2004) Since that time national incentives and globaldemand have transformed Brazil into the worldrsquos largestexporter of soybeans and beef among other commodities(Nepstad et al 2006) Most exports are in the form of freshor frozen beef although there is an increasing trend of livecattle exports (ALICEweb 2009) Between 1990 and 2006market and trade reforms in addition to the eradication of foot-and-mouth disease helped exports of beef from the Amazon togrow over 500 (IBGE 2009 Walker et al 2008) This growthhas had environmental consequences such as carbon emissionsfrom deforestation nutrient pollution biodiversity loss anddisplacement of local people (Betts et al 2008 Fearnside 2008Gibbs et al 2010 Foley et al 2007)

While cattle production is the predominant land use inthe Amazon soybeans have recently begun to encroach fromthe southern and eastern boundaries and are responsible fornew land clearing and displacement of cattle pastures (Vera-Diaz et al 2008) Several factors ranging from developmentof moisture-tolerant soybean varieties to increasing globaldemand for animal rations have led to the dramatic increasein soybean production and its rising percentage of the globalgrains market (Nepstad et al 2006) The Brazilian agriculturalcomplex is highly integrated into the global market system asBrazil exports more than 10 of the internationally traded cropbiomass (Ciais et al 2007)

Here we present an illustrative example that distributes theresponsibility for GHG emissions from deforestation betweenBrazil and the eventual importing nation of commoditiesproduced in the Amazon (figure 2) While the data andmethods presented here are considered to be the lsquostateof the sciencersquo the exact parameters allocating emissionsbetween international actors were chosen to exemplify theimportance of GHG emissions embodied in internationallytraded agricultural commodities as a template for futurepolicies Future work can build upon this framework makingdifferent policy assumptions as appropriate

53 Model description

Here we use a land use transition and carbon emission modeldescribed in detail by Ramankutty et al (2007) to estimatethe fate of deforested land and the associated carbon dioxide

4

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 2 Three major phases in the process of converting forests for the production and eventual consumption of beef and soybeans (1) landuseland cover change (2) agriculture and (3) trade

Table 1 Average biomass mean annual deforestation and change in deforestation rate between 1990 and 2006 for the nine states within theBrazilian Amazon (Sources INPE 2009 Saatchi et al 2007)

StateMean biomassvalue (Mg haminus1)

Mean annualdeforestation1990ndash2006 (km2)

Increase (+) or decrease (minus)in deforestation rate between1990 and 2006

Acre 204 583 +Amazonas 270 861 +Amapa 263 75 minusMaranhao 125 910 +Mato Grosso 164 6781 +Para 230 5703 +Rondonia 211 2633 +Roraima 215 246 minusTocantins 95 331 minus

equivalent (CO2e) emissions (figures 3(a) and (b)) While thepreviously published model is robust several key changes weremade to incorporate the latest published data and changingagricultural practices in the Brazilian Amazon The modelwas run for each of the nine states in the legal Amazon from1990 to 2006 using (time-smoothed) annual deforestation ratesprovided by Brazilrsquos National Institute for Space Research(INPE) (2009) Initial aboveground forest biomass values forland identified as intact forest by Brazilrsquos program to calculatethe deforestation of the Amazon (PRODES) were summarizedby state from Saatchi et al (2007) (table 1) In each yearcarbon emissions from deforestation are calculated as thesum of combusted biomass (20) and the decay of biomassremaining in the slash (70) product (8) and elemental (2)pools (Ramankutty et al 2007) (figures 4(a) and (b)) For thestate of Mato Grosso the percentages of biomass allocated tothe slash and burn pools were set to 20 and 70 respectivelyto reflect the trend of highly mechanized agriculture wheremost limbs and stumps are removed and burned after forestclearing (Galford et al 2010) The model partitions deforestedland between cropland pasture and secondary forest using

transition rates initially described by Fearnside (1996) butland transition rates for the state of Mato Grosso a hotspot ofdeforestation were updated according to the patterns reportedin Morton et al (2006) and Defries et al (2008) In addition tocarbon emissions from deforestation added methane emissionsfrom enteric fermentation in cattle were included in the totalCO2 equivalent (CO2e) flux and were computed with IPCCvalues for tropical cattle assuming a 100 yr global warmingpotential (IPCC 2007 Robertson and Grace 2004) GHGemissions from the application of fertilizer were found to benegligible compared to fluxes from deforestation and methaneproduction

54 Carbon allocation methods

The model distributes deforested land as it shifts betweencropland pasture and secondary forests computing the GHGemissions (or sequestration for the secondary forests) ofeach land use type While the model estimates the netemissions for new pastures and croplands the major exportcommodities (cattle and soybeans) only occupy a percentage

5

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(b)

(a)

Figure 3 Model simulation of the legal Amazon (a) land use and (b) carbon emissions from 1990 to 2006 The model distributes deforestedland (and resulting carbon emissions) between farms pastures and secondary forests Methane emissions from enteric fermentation in cattleare included within the emissions from lsquopasturersquo

of the agricultural landscape For each product the relativedominance on the landscape each year is calculated usingthe ratio of area planted in soybeans to total agriculturalarea and a similar ratio of pasture area for cattle tobuffalo horses sheep and other pastured animals (IBGE2009) The landscape dominance modifiers were appliedto each statersquos carbon emissions in every year of themodel

Export modifiers are calculated by dividing the amountof soybeans exported by the amount produced For cattleapproximately 10 of the herd is slaughtered every year andthe average carcass yield is used to calculate the beef andassociated products for a given year (IBGE 2009 FAOSTAT2009) Live cattle exports are incorporated with the beefexport data using average carcass yields (FAOSTAT 2009)The ratio of beef exported to beef produced was then estimated

(ALICEweb 2009) The export modifiers were applied over theaggregate Amazon to avoid data inconsistencies resulting frominterstate trade for each year of the model

The emissions calculated for each year are the sum ofburnt decay slash and elemental carbon from land deforestedin previous years minus land that transitioned to secondaryforest The allocation of carbon emissions over time werecalculated using three scenarios (1) emissions were allocatedto the year they occurred (2) equally distributed over 20 yearsand (3) linearly decreasing over 20 years (figure 1) Carbonfluxes from deforestation prior to 1990 were not includedand emission rates should therefore be seen as underestimatesAlso as future land use agricultural production and exportpatterns are unknown carbon from recent deforestation eventscannot be definitively allocated but must be tracked in order toassign future responsibility

6

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 6: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 2 Three major phases in the process of converting forests for the production and eventual consumption of beef and soybeans (1) landuseland cover change (2) agriculture and (3) trade

Table 1 Average biomass mean annual deforestation and change in deforestation rate between 1990 and 2006 for the nine states within theBrazilian Amazon (Sources INPE 2009 Saatchi et al 2007)

StateMean biomassvalue (Mg haminus1)

Mean annualdeforestation1990ndash2006 (km2)

Increase (+) or decrease (minus)in deforestation rate between1990 and 2006

Acre 204 583 +Amazonas 270 861 +Amapa 263 75 minusMaranhao 125 910 +Mato Grosso 164 6781 +Para 230 5703 +Rondonia 211 2633 +Roraima 215 246 minusTocantins 95 331 minus

equivalent (CO2e) emissions (figures 3(a) and (b)) While thepreviously published model is robust several key changes weremade to incorporate the latest published data and changingagricultural practices in the Brazilian Amazon The modelwas run for each of the nine states in the legal Amazon from1990 to 2006 using (time-smoothed) annual deforestation ratesprovided by Brazilrsquos National Institute for Space Research(INPE) (2009) Initial aboveground forest biomass values forland identified as intact forest by Brazilrsquos program to calculatethe deforestation of the Amazon (PRODES) were summarizedby state from Saatchi et al (2007) (table 1) In each yearcarbon emissions from deforestation are calculated as thesum of combusted biomass (20) and the decay of biomassremaining in the slash (70) product (8) and elemental (2)pools (Ramankutty et al 2007) (figures 4(a) and (b)) For thestate of Mato Grosso the percentages of biomass allocated tothe slash and burn pools were set to 20 and 70 respectivelyto reflect the trend of highly mechanized agriculture wheremost limbs and stumps are removed and burned after forestclearing (Galford et al 2010) The model partitions deforestedland between cropland pasture and secondary forest using

transition rates initially described by Fearnside (1996) butland transition rates for the state of Mato Grosso a hotspot ofdeforestation were updated according to the patterns reportedin Morton et al (2006) and Defries et al (2008) In addition tocarbon emissions from deforestation added methane emissionsfrom enteric fermentation in cattle were included in the totalCO2 equivalent (CO2e) flux and were computed with IPCCvalues for tropical cattle assuming a 100 yr global warmingpotential (IPCC 2007 Robertson and Grace 2004) GHGemissions from the application of fertilizer were found to benegligible compared to fluxes from deforestation and methaneproduction

54 Carbon allocation methods

The model distributes deforested land as it shifts betweencropland pasture and secondary forests computing the GHGemissions (or sequestration for the secondary forests) ofeach land use type While the model estimates the netemissions for new pastures and croplands the major exportcommodities (cattle and soybeans) only occupy a percentage

5

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(b)

(a)

Figure 3 Model simulation of the legal Amazon (a) land use and (b) carbon emissions from 1990 to 2006 The model distributes deforestedland (and resulting carbon emissions) between farms pastures and secondary forests Methane emissions from enteric fermentation in cattleare included within the emissions from lsquopasturersquo

of the agricultural landscape For each product the relativedominance on the landscape each year is calculated usingthe ratio of area planted in soybeans to total agriculturalarea and a similar ratio of pasture area for cattle tobuffalo horses sheep and other pastured animals (IBGE2009) The landscape dominance modifiers were appliedto each statersquos carbon emissions in every year of themodel

Export modifiers are calculated by dividing the amountof soybeans exported by the amount produced For cattleapproximately 10 of the herd is slaughtered every year andthe average carcass yield is used to calculate the beef andassociated products for a given year (IBGE 2009 FAOSTAT2009) Live cattle exports are incorporated with the beefexport data using average carcass yields (FAOSTAT 2009)The ratio of beef exported to beef produced was then estimated

(ALICEweb 2009) The export modifiers were applied over theaggregate Amazon to avoid data inconsistencies resulting frominterstate trade for each year of the model

The emissions calculated for each year are the sum ofburnt decay slash and elemental carbon from land deforestedin previous years minus land that transitioned to secondaryforest The allocation of carbon emissions over time werecalculated using three scenarios (1) emissions were allocatedto the year they occurred (2) equally distributed over 20 yearsand (3) linearly decreasing over 20 years (figure 1) Carbonfluxes from deforestation prior to 1990 were not includedand emission rates should therefore be seen as underestimatesAlso as future land use agricultural production and exportpatterns are unknown carbon from recent deforestation eventscannot be definitively allocated but must be tracked in order toassign future responsibility

6

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 7: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(b)

(a)

Figure 3 Model simulation of the legal Amazon (a) land use and (b) carbon emissions from 1990 to 2006 The model distributes deforestedland (and resulting carbon emissions) between farms pastures and secondary forests Methane emissions from enteric fermentation in cattleare included within the emissions from lsquopasturersquo

of the agricultural landscape For each product the relativedominance on the landscape each year is calculated usingthe ratio of area planted in soybeans to total agriculturalarea and a similar ratio of pasture area for cattle tobuffalo horses sheep and other pastured animals (IBGE2009) The landscape dominance modifiers were appliedto each statersquos carbon emissions in every year of themodel

Export modifiers are calculated by dividing the amountof soybeans exported by the amount produced For cattleapproximately 10 of the herd is slaughtered every year andthe average carcass yield is used to calculate the beef andassociated products for a given year (IBGE 2009 FAOSTAT2009) Live cattle exports are incorporated with the beefexport data using average carcass yields (FAOSTAT 2009)The ratio of beef exported to beef produced was then estimated

(ALICEweb 2009) The export modifiers were applied over theaggregate Amazon to avoid data inconsistencies resulting frominterstate trade for each year of the model

The emissions calculated for each year are the sum ofburnt decay slash and elemental carbon from land deforestedin previous years minus land that transitioned to secondaryforest The allocation of carbon emissions over time werecalculated using three scenarios (1) emissions were allocatedto the year they occurred (2) equally distributed over 20 yearsand (3) linearly decreasing over 20 years (figure 1) Carbonfluxes from deforestation prior to 1990 were not includedand emission rates should therefore be seen as underestimatesAlso as future land use agricultural production and exportpatterns are unknown carbon from recent deforestation eventscannot be definitively allocated but must be tracked in order toassign future responsibility

6

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 8: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 4 Model representation of the fate of land deforested in the legal Amazon in 1990 (b) as the cohort of land progresses through timeand (a) fluxes of carbon from deforestation in 1990 assigned to each land use class The initial pulse of emissions in the first year is from theburning of biomass while the remaining fluxes are from the decay of slash product and elemental pools Carbon sequestration fromsecondary forests is also included

55 Producerndashconsumer

Following the example of Gallego and Lenzen (2005) thefinal responsibility for the carbon emissions was divided5050 between producer and consumer (where Brazil is thelsquoproducerrsquo and importing countries are lsquoconsumersrsquo) Becauseassigning the entire responsibility for carbon emissions toeither the producer or consumer has been shown to be asuboptimal solution (Peters 2008 Peters and Hertwich 2008a)sharing the responsibility between both parties is preferred(Lenzen et al 2007) While the 5050 division of emissionsliability between producer and consumer is arbitrary andwould need to be negotiated by the importing and exportingparties it serves to illustrate how emissions can be fairlydivided along the supply chain In the present case theallocation of GHG responsibility between Brazil and thecountry importing the agricultural goods would remain thesame in all three scenarios but the length of time needed to

complete the carbon debt obligation and annual distribution ofcarbon liability would vary

56 Results

Exports of beef and soybeans from the Brazilian Amazonincreased dramatically between 1990 and 2006 As increasedexports coincided with increases in deforestation in theAmazon carbon emission liability increased for both beefand soybeans The emissions embodied in exported beef andsoybeans can be compared using the three temporal allocationscenarios from figure 1 (figures 5(a) and (b)) Model resultsbetween 1990 and 2006 are derived from deforestation andcommodity export data results for later years depend on thisdata and assumptions about future land use and export patterns

Carbon emissions liability from recent deforestationevents continue into the future due to decaying biomass andto amortization by the 20 yr allocation scenarios Here we

7

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 9: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 5 Comparison of carbon liability for (a) soybeans and (b) beef exported from the legal Amazon between 1990 and 2006 using the1 yr 20 yr and 20 yr decline allocation methods Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levelsand deforestation either continues at 2006 levels or ceases

assumed that export rates for the future were equal to the exportrate of 2006 and either no further deforestation took place ordeforestation took place at the same annual rate as in 2006While the probability of realizing these scenarios is small theyillustrate the envelope of possibilities that land use could haveon the allocation of carbon emissions embodied in soybeansand beef into the future

The total amount of carbon to be allocated is equalfor the three scenarios although the annual distributionvaries according to each scenario The annual allocation ofcarbon is most greatly influenced by annual deforestation andexport rates Between 1990 and 2006 the annual carbonliability from the 1 yr allocation scenario is greater thaneither 20 yr allocation (figures 5(a) and (b)) After 2006whether deforestation is halted or continues carbon liabilityis generally less than either 20 yr allocation scheme asmost carbon is allocated soon after the deforestation eventComparing the 20 yr allocation schemes the annual allocationto the 20 yr decline scenario is greater than the 20 yr constantscenario before 2006 while the trend reverses after 2006

The temporal patterns for carbon emissions export liability aresimilar for both soybeans and beef

Using the 20 yr decline allocation scenario 128 TgCO2eembodied in soybeans were exported from the Amazonbetween 1990 and 2006 If deforestation and exports continueat 2006 levels until 2025 an additional 499 TgCO2e wouldbe embodied in soybean exports while 236 TgCO2e wouldbe exported if deforestation ceased For beef 120 TgCO2ewere exported from 1990 to 2006 and an additional 822ndash1369 TgCO2e could be exported by 2025 as calculated usingour chosen envelope of future land use and export patterns Therelatively low embodied emissions from the early 1990s aredue to minimal exports and ignoring decay emissions fromdeforestation previous to 1990 Increasing embodied carbonemissions between 1990 and 2006 are the result of risingexports and the distribution of decay emissions over time

Using the 20 yr decline allocation scenario emissionsfrom deforestation were assigned to importing regionsaccording their percentages of the total global importsAssuming imports for each region remained constant between

8

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 10: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

(a)

(b)

Figure 6 Annual carbon emissions liability for major importing regions of soybeans (a) and beef (b) from the Brazilian Amazon using the20 yr decline allocation method Carbon emissions between 2006 and 2025 assume that rates of export continue at 2006 levels anddeforestation in the Amazon either continues at 2006 levels or ceases

2006 and 2025 the envelope of carbon liability is shownbased on the continuation or cessation of deforestation inthe Amazon The major beef importing regions during thestudy period were Eastern Europe the EU Middle EastAfrica South America and Asia The EU and Asia were themajor soybean importers during this time period (figures 6(a)and (b)) Between 1990 and 2006 the EU was the largestimporter of soybeans from the Amazon importing 312 ofthe emissions embodied in soybeans during that time Majorimports from Asia began around 2000 with the largest singleyear increase between 2004 and 2005 Between 1990 and2000 the EU imported the majority of beef from the Amazoncomprising 618 of embodied emissions during this timeAfter 2000 imports by Eastern Europe the Middle East andother areas in South America rapidly increased as imports bythe EU decreased In the mid-1990s live cattle were exportedexclusively to other countries in South America while after2000 exports shifted to the Middle East The total carbonliability for soybeans and beef between 1990 and 2006 isshown in figure 7

The 20 yr decline scenarios suggest important outcomesof such an allocation scheme (figures 6(a) and (b)) Firstthe carbon emissions liability of importing countries decreasesdramatically if they import from regions where deforestationceased decades ago This is by design the relative impactof the allocation policy then depends on the price of carbonSecond the deforestation status of the exporting region canhave a greater effect on the carbon emissions liability ofthe importing country than does the amount of commodityimported This could confer a significant trade advantage onexporting countries that avoid new deforestation and intensifyproduction on already cleared land

6 Conclusions

In this paper we present the methodologies necessary tocalculate the embodied GHG emissions due to land conversionin agricultural products and compare three schemes forshifting some of the responsibility for these emissions from

9

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 11: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Figure 7 Cumulative carbon emission liability (1990ndash2006) for countries importing soybeans and beef from the Amazon The total carbonemissions liability of importing countries is equal to the liability assigned to Brazil

producer to consumer While many mechanisms have beenproposed to decrease rates of deforestation in the Amazonvery few of them include the ultimate drivers of deforestationconsumers of agricultural products Prior to 2000 exportsfrom the Amazon minimally contributed to the environmentalfootprints of importing countries but increasing exports havethrust the idea of embodied carbon onto the global stage Ifgreenhouse gas mitigation becomes globally and consistentlyvalued then distributing the responsibility for GHG emissionswill be a practical and feasible instrument to incentivizealternatives to deforestation

The Amazon case study presented here tracked theGHG emissions of land use and land cover change due tofarming soybeans and beef for export The study requireda fusion of techniques including calculating emissions fromdeforestation life-cycle analysis of agricultural systems andallocating emissions between producers and consumers Whilethe best available data were used in this case study the inputdata place several constraints on the analysis such as theinability to distinguish between exports from newly clearedland and exports from previously cleared land For this reasonwe used the commodity production data as relative modifiers ofthe carbon emission model It would have been inappropriateto assume the carbon emissions embodied in soybeans andcattle could be inferred from the production statistics alone dueto varied production practices Also decay related emissionsfrom later years of deforestation (eg post-2000) must beallocated using assumptions as future production and exportquantities are unknown

In addition to bypassing the current drawbacks of theKyoto protocol allocating emissions between producers andconsumers would incentivize both parties to adopt lesspolluting production practices Production practices in theAmazon which is increasingly becoming integrated intoglobalized markets can be leveraged to decrease rates ofdeforestation and increase yields on already-established farmsIf buyers must pay for the carbon embodied in the productsthey purchase economic logic dictates that they will buy

items whose prices are lower due to lower carbon intensityof production In turn producers will change their practicesto compete with producers with lower costs in this case bychanging management and land use practices

While land use can make up a considerable portion ofthe life-cycle impacts from tropical agriculture additionalcomponents need to be considered Emissions related toconsumption of beef including production of feed grainstransportation and processing must be quantified Soybeanson the other hand are used as an input for animal feed andindustrial processes (eg biodiesel production) while humansonly directly consume a small proportion of soybeans (mostlyas soybean oil) In the case of soybeans emissions fromprocessing and transportation need to be assessed in additionto emissions from the combustion of biodiesel or impacts oflivestock rearing Some of the methodologies required toassess these additional impacts have been considered in otherstudies (eg Roy et al 2009 Dalgaard et al 2008)

This study helps to lay the foundations for a much-neededglobal analysis of embodied emissions from agriculturalproduction and to develop methodologies needed to assignresponsibility for the impacts While a global analysis isfeasible for a selection of agricultural products as illustratedin this case study there are many uncertainties and unknownparameters needed to perform full GHG accounting forother managed agricultural systems Future research in thishighly interdisciplinary arena must take advantage of recentlyreleased research using remote sensing platforms (Palace et al2008 Wang and Qi 2008) field studies (Davidson et al 2008)and RFID technology (Kelepouris et al 2007) in order to makeprogress toward full lsquofield to forkrsquo life-cycle assessments foragricultural products

Acknowledgments

We thank Michael Coe and Holly Gibbs and three anonymousreviewers for insightful comments on the paper and Marcos

10

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 12: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

Costa for helpful insights throughout the research processWe also thank Mary Sternitzky for assistance with graphicsThis work was supported by the National Aeronautics andSpace Administrationrsquos (NASA) Large Biosphere Atmospherein Amazonia (LBA) project and a NASAWisconsin SpaceGrant Consortium fellowship

References

Ahmad N and Wyckoff A 2003 Carbon dioxide emissions embodiedin international trade of goods OECD Working Paper

Alexandratos N Bruinsma J and Boedeker G 2006 WorldAgriculture Towards 20302050 Interim Report Prospects ForFood Nutrition Agriculture and Major Commodity GroupsGlobal Perspective Studies Unit Food and AgricultureOrganization of the United Nations

ALICEweb 2009 Ministerio do Desenvolvimento Industria eComercio Exterior available online httpalicewebdesenvolvimentogovbr (last accessed 20 February 2009)

Andrew R and Forgie V 2008 A three-perspective view ofgreenhouse gas emission responsibilities in New Zealand EcolEconom 68 194ndash204

Barbier E B 2000 Links between economic liberalization and ruralresource degradation in the developing regions Agric Econ23 299ndash310

Barreto P Souza C Jr Nogueron R Anderson A andSalomao R 2006 Human Pressure on the Brazilian AmazonForests (Washington DC World Resources Institute)

Bastianoni S Pulselli F M and Tiezzi E 2004 The problem ofassigning responsibility for greenhouse gas emissions EcolEconom 49 253ndash7

Betts R Malhi Y and Roberts J 2008 The future of the Amazon newperspectives from climate ecosystem and social sciences PhilTrans R Soc B 363 1729ndash35

Bodansky D Chou S and Jorge-Tresolini C 2004 Internationalclimate efforts beyond 2012 a survey of approaches PewCenter on Global Climate Change

Brentrup F Kusters J Kuhlmann H and Lammel J 2004Environmental impact assessment of agricultural productionsystems using the life cycle assessment methodologymdashITheoretical concept of a LCA method tailored to cropproduction Eur J Agron 20 247ndash64

BSI 2008 Publicly Available Specification (PAS)2050mdashspecification for the assessment of the life cyclegreenhouse gas emissions of goods and services

Canals L M i et al 2007a Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess 12 2ndash4

Canals L M i et al 2007b Key elements in a framework for land useimpact assessment within LCA Int J Life Cycle Assess12 5ndash15

Ciais P Bousquet P Freibauer A and Naegler T 2007 Horizontaldisplacement of carbon associated with agriculture and itsimpacts on atmospheric CO2 Global Biogeochem Cycles21 GB2014

Dalgaard R Schmidt J Halberg N Christensen P Thrane M andPengue W A 2008 LCA of soybean meal Int J Life CycleAssess 13 240ndash54

Davidson E De Abreu Sa T D Carvalho C J R Figueiredo R D OKato M and Ishida F Y 2008 An integrated greenhouse gasassessment of an alternative to slash- and-burn agriculture ineastern Amazonia Global Change Biol 14 998ndash1007

Defries R Morton D Van Der Werf G Giglio L Collatz GRanderson J Houghton R Kasibhatla P andShimabukuro Y 2008 Fire-related carbon emissions from landuse transitions in southern Amazonia Geophys Res Lett35 L22705

FAOSTAT 2009 Food and Agriculture Organization of The UnitedNations

Fargione J Hill J Tilman D Polasky S and Hawthorne P 2008 Landclearing and the biofuel carbon debt Science 319 1235ndash8

Fearnside P 1996 Amazonian deforestation and global warmingcarbon stocks in vegetation replacing Brazilrsquos Amazon forestForest Ecol Manag 80 21ndash34

Fearnside P 2008 Amazon forest maintenance as a source ofenvironmental services Ann Acad Bras Cienc 80 101ndash14

Foley J et al 2007 Amazonia revealed forest degradation and loss ofecosystem goods and services in the Amazon basin Front EcolEnviron 5 25ndash32

Foley J A et al 2005 Global consequences of land use Science309 570ndash4

Galford G L Melillo J M Mustard J F Cerri C E P and Cerri C C2010 Global frontier of land-use change the expansion ofcroplands in the southwestern Amazon Global Change Biolin review

Gallastegui I 2002 The use of eco-labels a review of the literatureEur Environ 12 316ndash31

Gallego B and Lenzen M 2005 A consistent inputndashoutputformulation of shared producer and consumer responsibilityEcon Sys Res 17 365ndash91

Garnett T 2009 Livestock-related greenhouse gas emissions impactsand options for policy makers Environ Sci Policy 12 491ndash503

Gibbs H Johnston M Foley J Holloway T Monfreda CRamankutty N and Zaks D 2008 Carbon payback times forcrop-based biofuel expansion in the tropics the effects ofchanging yield and technology Environ Res Lett 3 034001

Gibbs H K Ruesch A S Achard F Holmgren P Ramankutty N andFoley J A 2010 Pathways of agricultural expansion across thetropics implications for global feed food and fuel demandsProc Natl Acad Sci USA in review

Goleman D 2009 Ecological Intelligence How Knowing the HiddenImpacts of What We Buy Can Change Everything (New YorkBroadway Books)

Gullison R E et al 2007 Tropical forests and climate policy Science316 985ndash6

Gupta S and Bhandari P M 1999 An effective allocation criterion forCO2 emissions Energy Policy 27 727ndash36

Hansen M C et al 2008 Humid tropical forest clearing from 2000 to2005 quantified by using multitemporal and multiresolutionremotely sensed data Proc Natl Acad Sci USA 105 9439ndash44

Houghton R A 2003 Revised estimates of the annual net flux ofcarbon to the atmosphere from changes in land use and landmanagement 1850ndash2000 Tellus B 55 378ndash90

IBGE 2009 Sistema ibge de recuperacao automatica Sidra databaseavailable online wwwsidraibgegovbr (last accessed 20February 2009)

INPE 2009 Monitoramento da floresta amazonica brasileira porsatelite projeto PRODES available online wwwobtinpebrprodesindexhtml (last accessed 20 February 2009)

IPCC (Intergovernmental Panel on Climate Change) Working GroupI 2007 Climate Change 2007 The Physical Science BasisSummary for Policymakers (Paris IPCC Secretariat)

Jolliet O Margni M Charles R Humbert S Payet J Rebitzer G andRosenbaum R 2003 Impact 2002+ a new life cycle impactassessment methodology Int J Life Cycle Assess 8 324ndash30

Jungbluth N et al 2007 Life Cycle Inventories of Bioenergy FinalReport ESU-Services Uster CH

Jungbluth N 2009 personal communicationKaimowitz D Mertens B Wunder S and Pacheco P 2004 Hamburger

Connection Fuels Amazon Destruction CIFORKelepouris T Pramatari K and Doukidis G 2007 RFID-enabled

traceability in the food supply chain Indust Manage Data Syst107 183ndash200

Lehuger S Gabrielle B and Gagnaire N 2009 Environmental impactof the substitution of imported soybean meal withlocally-produced rapeseed meal in dairy cow feed J CleanProd 17 616ndash24

Lenzen M Murray J Sack F and Wiedmann T 2007 Shared producerand consumer responsibilitymdashtheory and practice EcolEconom 61 27ndash42

11

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References
Page 13: Producer Consumer Responsibility for GHGs in Agriculture in ...

Environ Res Lett 4 (2009) 044010 D P M Zaks et al

McAlpine C Etter A Fearnside P Seabrook L and Laurance W 2009Increasing world consumption of beef as a driver of regionaland global change a call for policy action based on evidencefrom Queensland (Australia) Colombia and Brazil GlobalEnviron Change 19 21ndash33

MEA (Millennium Ecosystem Assessment) 2005 Ecosystems andHuman Well-Being Scenarios (Washington DC Island Press)

Mollicone D Freibauer A Schulze E D Braatz S Grassi G andFederici S 2007 Elements for the expected mechanisms onlsquoreduced emissions from deforestation and degradation REDDunder UNFCCC Environ Res Lett 2 045024

Monfreda C Ramankutty N and Foley J 2008 Farming the planet 2Geographic distribution of crop areas yields physiologicaltypes and net primary production in the year 2000 GlobalBiogeochem Cycles 22 GB1022

Morton D C Defries R Shimabukuro Y Anderson L O Arei EEspirito-Santo F D Freitas R and Morisette J 2006 Croplandexpansion changes deforestation dynamics in the southernBrazilian Amazon Proc Natl Acad Sci USA 103 14637ndash41

Munksgaard J and Pedersen K A 2001 CO2 accounts for openeconomies producer or consumer responsibility Energy Policy29 327ndash34

Muys B and Garcıa Quijano J 2002 A new method for land useimpact assessment in LCA based on the ecosystem exergyconcept Laboratory for Forest Nature and LandscapeResearch Katholieke Universiteit Leuven

Nepstad D Stickler C and Almeida O 2006 Globalization of theAmazon soy and beef industries opportunities for conservationConserv Biol 20 1595ndash603

Palace M Keller M Asner G Hagen S and Braswell B 2008Amazon forest structure from IKONOS satellite data and theautomated characterization of forest canopy propertiesBiotropica 40 141ndash50

Panichelli L Dauriat A and Gnansounou E 2009 Life cycleassessment of soybean-based biodiesel in Argentina for exportInt J Life Cycle Assess 14 144ndash59

Peters G 2008 From production-based to consumption-based nationalemission inventories Ecol Econom 65 13ndash23

Peters G and Hertwich E 2008a Post-Kyoto greenhouse gasinventories production versus consumption Clim Change86 51ndash66

Peters G and Hertwich E 2008b CO2 embodied in international tradewith implications for global climate policy Environ SciTechnol 42 1401ndash7

Pretty J N Brett C Gee D Hine R E and Mason C F 2000 Anassessment of the total external costs of UK agriculture AgricSyst 65 113ndash36

Ramankutty N Evan A T Monfreda C and Foley J 2008 Farming theplanet 1 Geographic distribution of global agricultural lands inthe year 2000 Global Biogeochem Cycles 22 GB1003

Ramankutty N Gibbs H Achard F Defriess R Foley J andHoughton R 2007 Challenges to estimating carbon emissionsfrom tropical deforestation Global Change Biol 13 51ndash66

Reijnders L and Huijbregts M A J 2008 Biogenic greenhouse gasemissions linked to the life cycles of biodiesel derived fromEuropean rapeseed and Brazilian soybeans J Clean Prod16 1943ndash8

Robertson G P and Grace P 2004 Greenhouse gas fluxes in tropicaland temperate agriculture the need for a full-cost accounting ofglobal warming potentials Environ Develop Sustain 6 51ndash63

Rodrigues J and Domingos T 2008 Consumer and producerenvironmental responsibility comparing two approaches EcolEconom 66 533ndash46

Roy P Nei D Orikasa T Xu Q Okadome H Nakamura N andShiina T 2009 A review of life cycle assessment (LCA) on somefood products J Food Eng 90 1ndash10

Saatchi S S Houghton R Alvala R C D S Soares J V and Yu Y 2007Distribution of aboveground live biomass in the Amazon basinGlobal Change Biol 13 816ndash37

Schau E M and Fet A M 2008 LCA studies of food products asbackground for environmental product declarations Int J LifeCycle Assess 13 255ndash64

Simon M and Garagorry F 2006 The expansion of agriculture in theBrazilian Amazon Environ Conserv 32 203ndash12

Smith P et al 2008 Greenhouse gas mitigation in agriculture PhilTrans R Soc B 363 789ndash813

Soares Filho B S et al 2006 Modelling conservation in the Amazonbasin Nature 440 520ndash3

Tilman D 1999 Global environmental impacts of agriculturalexpansion the need for sustainable and efficient practices ProcNatl Acad Sci USA 96 5995ndash6000

Vera-Diaz M D C Kaufmann R K Nepstad D andSchlesinger P 2008 An interdisciplinary model of soybean yieldin the Amazon basin the climatic edaphic and economicdeterminants Ecol Econom 65 420ndash31

Walker R Browder J Arima E Simmons C Pereira R Caldas MShirota R and Zen S 2008 Ranching and the new global rangeAmazonia in the 21st century Geoforum in press(doi101016jgeoforum200810009)

Wang C and Qi J 2008 Biophysical estimation in tropical forestsusing JERS-1 SAR and VNIRr imagery II Abovegroundwoody biomass Int J Remote Sens 29 6827ndash49

12

  • 1 Introduction
  • 2 Producer versus consumer
  • 3 Allocation of land use emissions
  • 4 Methods
  • 5 Case study
    • 51 Deforestation in the Amazon
    • 52 Transition to an export market
    • 53 Model description
    • 54 Carbon allocation methods
    • 55 Producer--consumer
    • 56 Results
      • 6 Conclusions
      • Acknowledgments
      • References

Recommended