how can land-use modelling tools inform bioenergy policies?

Interface Focus (2011) 1 212ndash223

doi101098rsfs20100023

Published online 2 February 2011

REVIEW

Author andBiology 120(davisscillin

One contribueconomy and

Received 15 NAccepted 10 J

How can land-use modelling toolsinform bioenergy policies

Sarah C Davis12 Joanna I House3 Rocio A Diaz-Chavez4Andras Molnar5 Hugo Valin67 and Evan H DeLucia12

1Energy Biosciences Institute and 2Department of Plant Biology University of Illinoisat Urbana-Champaign Urbana IL USA

3QUEST Department of Earth Sciences University of Bristol Bristol UK4Centre for Environmental Policy Imperial College London UK

5Research Institute for Agricultural Economics Budapest Hungary6Forestry Program International Institute on Applied Systems Analysis Laxenburg Austria7UMR Economie PubliquePublic Economics Research Unit Institute National de Recherche

Agronomique (INRA) Paris France

Targets for bioenergy have been set worldwide to mitigate climate change Although feed-stock sources are often ambiguous pledges in European nations the United States andBrazil amount to more than 100 Mtoe of biorenewable fuel production by 2020 As a conse-quence the biofuel sector is developing rapidly and it is increasingly important to distinguishbioenergy options that can address energy security and greenhouse gas mitigation from thosethat cannot This paper evaluates how bioenergy production affects land-use change (LUC)and to what extent land-use modelling can inform sound decision-making We identified localand global internalities and externalities of biofuel development scenarios reviewed relevantdata sources and modelling approaches identified sources of controversy about indirect LUC(iLUC) and then suggested a framework for comprehensive assessments of bioenergy Ulti-mately plant biomass must be managed to produce energy in a way that is consistentwith the management of food feed fibre timber and environmental services Bioenergy pro-duction provides opportunities for improved energy security climate mitigation and ruraldevelopment but the environmental and social consequences depend on feedstock choicesand geographical location The most desirable solutions for bioenergy production will includepolicies that incentivize regionally integrated management of diverse resources with lowinputs high yields co-products multiple benefits and minimal risks of iLUC Many inte-grated assessment models include energy resources trade technological development andregional environmental conditions but do not account for biodiversity and lack detaileddata on the location of degraded and underproductive lands that would be ideal for bioenergyproduction Specific practices that would maximize the benefits of bioenergy productionregionally need to be identified before a global analysis of bioenergy-related LUC can beaccomplished

Keywords indirect land-use change biofuels greenhouse gas ecosystem servicesenvironmental economics feedstocks

1 INTRODUCTION

Bioenergy provides an opportunity for mitigating cli-mate change and improving energy security byreplacing both liquid and solid fossil fuels Howeverconcerns about land availability competition with

address for correspondence Institute for Genomic6 West Gregory Drive Urbana IL 61801 USAoisedu)

tion of 9 to a Theme Issue lsquoBiorenewables the bio-basedsustainabilityrsquo

ovember 2010anuary 2011 212

food production and environmental impacts must allbe addressed when defining suitable land-use practicesfor bioenergy This paper explores the characteristicsof land-use change (LUC) scenarios that should be con-sidered when making land management decisions thataffect bioenergy production It also evaluates whethercurrent land-use data and models contain the necessaryinformation and processes to assess the consequences ofexpanding bioenergy production throughout the world

There are many possible scenarios of bioenergyproduction and the options vary with geographical

This journal is q 2011 The Royal Society

Review Assessing bioenergy-related LUC S C Davis et al 213

location Bioenergy feedstocks range from oil and sugarcrops to grasses and trees and conversion pathwaysrange from direct combustion to chemical processingfor liquid transportation fuels (eg biodiesel bio-ethanol) New conversion technologies for second-generation biofuels allow more diverse feedstocksincluding lignocellulosic biomass and residues andadvanced conversion systems to high-density energycarriers [1] These new feedstock opportunities arelikely to encourage expansion of biomass productionglobally in the future but the magnitude and locali-zation of LUC are highly dependent on the regionof production and the type of feedstock used

Because the bioenergy sector interacts with otherecosystem services a holistic approach to assessmentof land management is required With increasing com-petition for land resources there is a need forcomprehensive tools that will identify the best ways tooptimize many agricultural resources in an integratedway A few studies have recently addressed this chal-lenge in the context of specific regions (eg [2ndash4])There are also global models that aim to characterizeLUC (eg [56]) but the macro-level view of thesemodels can overlook some dimensions of potentiallybeneficial land-use options Delineation of positive andnegative scenarios for bioenergy is therefore necessaryto complement global approaches and highlight localopportunities

There has been much recent concern about indirectland-use change (iLUC) as a result of bioenergy cropsbeing grown (eg [7]) Many modelling studies that esti-mate LUC produce widely varying results as to thedegree of iLUC (see [8] for a review and [9] for discussionon uncertainty) Considering the large confidence inter-vals of model estimates it remains unclear how theseresults can be used for legislation At the same timeit is increasingly being recognized that for certaincrops integrated multiple-use land management canreduce the risk of unintended iLUC For examplebioenergy can be produced as a waste product or as aco-product with food feed or fibre Thus far tools forcarrying out proper assessments of multiple-useopportunities have not been developed

We will first present a typology that characterizesdesirable and undesirable effects of LUC associatedwith feedstock crops then provide examples of positiveand negative LUC scenarios for bioenergy productionNext we will explore how existing datasets andmodels characterize new LUC scenarios and the risksof iLUC We conclude with a discussion of how scien-tific tools that evaluate bioenergy-related LUC mightbe improved for informing policy decisions

2 CRITERIA FOR EVALUATINGBIOENERGY SCENARIOS

Different feedstock crop species bring different costsand benefits with respect to economic developmentfood security climate mitigation biodiversity andother land services A clear typology of positive andnegative effects is needed for a comprehensive evalu-ation of these different options For example planting

Interface Focus (2011)

bioenergy crops on lands with high carbon stockssuch as peatlands and forests has negative net green-house gas consequences and other costs [10] On theother hand deploying high-yielding bioenergy cropsthat require low inputs (fertilizers irrigation) canimprove ecologically degraded and economically under-productive landscapes [411ndash13] In this section we willfirst introduce the dimensions of bioenergy productionthat must be addressed in an evaluation of bioenergy-related LUC We then discuss LUC scenarios that opti-mize biofuel benefits regionally highlight scenarios ofintegrated land management for multiple uses andfinally provide contrasting examples of scenarios thatrisk multiple negative impacts

21 Dimensions of bioenergy

To assess different LUC scenarios including those forbioenergy some common rules must be applied toensure systematically consistent evaluations In thecase of bioenergy some dimensions that appear particu-larly critical for evaluation of costs and benefits aresystem boundaries counterfactual possibilities andexternalities When evaluating iLUC these definitionsare broad as system boundaries extend to land use inthe whole world and there is little agreement abouthow externalities at this scale are best regulatedinternationally

System boundaries define the start and end points ofthe production chain (eg cradle-to-grave) the geo-graphical space and the temporal limits that apply toa given analysis System boundaries are one of the keysources of difference in estimates of overall costs andbenefits of a biofuel production chain among variouslife-cycle analyses [14] Results of any estimate ofbioenergy-related impacts should be considered in thecontext of the system boundary that was selectedfor analysis

The counterfactual possibilities should be definedrelative to the condition of land if a biofuel project inquestion is not implemented This lsquobusiness-as-usualrsquotrend can vary over time according to profitability ofbiomass for other uses future demand effects of climatechange on yield and other temporally dynamic factorsThe costs and benefits of a land-use scenario withbusiness-as-usual (that continues for the same lengthof time as a bioenergy production scenario) are theminimum counterfactuals that should be consideredwhen evaluating impacts of LUC caused by bioenergyThe bioenergy scenario would displace the business-as-usual case and therefore offset expected costs andbenefits with new temporally dynamic costs andbenefits The difference between a bioenergy land-usescenario and its counterfactual is sometimes consideredin lsquopayback timesrsquo as in estimates of carbon debtsresulting from LUC [10]

Externalities are positive or negative effects of aneconomic activity that are not embodied in the priceof products and consequently cannot be optimally man-aged through pure market arbitration Effects can bediverse across bioenergy LUC scenarios because theydepend on feedstock species and regional managementrequirements Externalities of bioenergy production

Table 1 Effects of biofuel production scenarios across three dimensions (i) benefits versus costs (ii) local versus global and(iii) internalities versus externalities

local global

benefits direct economic benefits (internalities) direct economic benefits (internalities)mdash agricultural support to biofuel producers mdash higher prices for world farmersmdash higher prices for farmersmdash subsidization of meat sector through cheaper

co-products

positive social and environmental externalities positive social and environmental externalitiesmdash rural development eg biofuel projects in areas with

degraded land and declining resourcesmdash fossil fuel substitution and associated

emissions savings (partly internalized)mdash landscape management eg plantation in degraded

landmdash carbon sequestration in land and biomass

(partly internalized)mdash biodiversity eg rotation culture in monoculture

areasmdash soil quality eg cultivation of degraded landmdash water quality and quantity eg lower inputs in

areas previously under intensive management

costs direct economic costs (internalities) direct economic costs (internalities)mdash increase price for fuel consumer andor burden for

taxpayermdash increase price for fuel consumer and

or burden for taxpayermdash higher food price for consumer mdash higher food price for consumermdash higher feed price for meat producers

negative social and environmental externalities negative social and environmental externalitiesmdash landscape management eg maize cultivation in

grassland areasmdash direct carbon and GHG emissionsmdash carbon and GHG emissions from iLUC

mdash biodiversity eg palm plantation instead of primaryforests

mdash food insecurity and world hunger

mdash soil quality eg expansion of monoculture cropsunder intensive management used for biofuelcultivation (corn soya beans)

mdash water quality and quantity eg annually fertilizedcrop on grassland

214 Review Assessing bioenergy-related LUC S C Davis et al

are characterized in table 1 Change in biodiversity is anexample of an externality that is infrequently quantifiedwhen bioenergy scenarios are analysed but is neverthe-less an important driver of economic social andenvironmental impacts [15] Soil quality is anotherexternality that has effects on the value of bioenergy(eg production) as well as the potential benefits ofcarbon sequestration from new feedstock crops (eg per-ennial grasses on degraded lands) Some of these effectscan be partially internalized through intervention inmarkets that tend to correct a price signal to accountfor the additional cost or benefit (eg carbon pricingin the EU for electricity required for the transformationof biofuels) Such intervention can influence theperceived value of a bioenergy cropping systemwithin regions

To frame a discussion about LUC associated withbioenergy we can classify impacts of scenarios alongthree dimensions (i) localglobal (ii) costbenefitsand (iii) internalitiesexternalities (table 1) Theseeffects are measured as differences between the scenarioand its counterfactual within the system boundarieschosen Specific scenarios with benefits that outweighcosts will be defined first and then followed by adescription of scenarios where cumulated costs acrossthe three dimensions outweigh the expected benefitsIt should be noted that there are costs and benefits inall scenarios of bioenergy but lsquopositive LUC scenariosrsquo


have greater benefits than costs while lsquonegative LUCscenariosrsquo have greater costs than benefits

22 Positive land-use change scenarios

Generally land use for bioenergy is considered positiveif an additional ecosystem or social service is providedas a result of the establishment of a bioenergy crop(table 1) The new benefits should outweigh the costsand could be realized in the form of increased resources(biomass feed co-production co-generation etc)improved water quality reduced greenhouse gas emis-sions increased soil carbon sequestration or increasedbiodiversity Social welfare benefits include increasedproduction capacity that diversifies and stimulateslocal economies Not all of the benefits listed abovecan be accomplished simultaneously but those thatcan be achieved should outweigh the costs that mightoccur Realistic evaluations of bioenergy scenariosshould provide transparent accounting of risks thatwould occur in the context of the benefits expected

Highly productive crops with low input require-ments are ideal for bioenergy crops because theyhave the potential to restore environmentally degradedland Intensive land management for agricultural pro-ducts leads to a reduction in soil quality over time(eg [1617]) If more productive crops that requireless intensive management (reduced tillage herbicide


pesticide and fertilizer) can supplement agriculturalproduction without damaging food supplies it ispossible that land quality can be restored withbioenergy crops

Perennial grasses used in advanced second-gener-ation energy conversion pathways can lead to positivebenefits compared with land use for intensively mana-ged first-generation crops For example Miscanthus xgiganteus (miscanthus) has the potential to triple thebiomass that is currently available for ethanol feedstockfrom corn grain in the midwestern USA [11ndash1318] In ascenario that assumes the same land area that is cur-rently used for grain ethanol the agriculturallandscape can be changed from a net source of green-house gases to the atmosphere to a net sink ofgreenhouse gases to the land [11213] There are sometrade-offs in this scenario because corn seed marketswould be affected benefiting seed consumers (farmers)but causing a net loss for companies that sell corn seedAnother side effect of this scenario is increased wateruse because of the greater evapotranspiration thatoccurs with the high photosynthetic rates and thelonger growing season of miscanthus [18ndash20] In regionswhere rainfall is abundant as in much of the midwesternUSA this cost of water is sustainable but such a costwould be intolerable in regions that are water-limited

In a different scenario the conversion of pasture landto high-producing grass crops can result in minimallosses of ecosystem services while adding resource pro-ductivity in the form of biofuel feedstocks In Brazilfor example it is estimated that a small increase incattle stocking density from the current one head ofcattle per hectare of pasture (versus two in Franceand Ireland according to FAOSTAT) can yieldadditional land for feedstock production [121] In thisway there is no cost in displaced land use eventhough LUC occurs A caveat of this scenario is thatthe welfare of nomadic cultures could be compromisedif this land-use practice was implemented in someregions of the world (eg Africa) this risk is lowhowever in other regions (eg Brazil)

High-producing feedstocks that grow in arid or semi-arid regions also exemplify a bioenergy scenario of LUCfor biofuel crops that maximizes production while mini-mizing losses of other resources and ecosystem servicesPlants that have high water-use efficiency and droughttolerance like those that use a photosynthetic pathwayknown as Crassulacean acid metabolism (CAM) haverecently been introduced for such scenarios [2223]Specifically for areas of Mexico Africa and Australiawhere water is an extremely limiting resource for foodand crop production CAM plants like Agave maybe a viable option to maximize efficient productionwithout losing other land-based resources [2425]

The scenarios described above indicate that there arealready opportunities for positive bioenergy outcomeswith existing crops technologies and land resourcesWhile the scenarios we provided are regionally specificthey each exemplify the use of second-generationcellulosic feedstocks Second-generation bioenergyproduction chains (advanced conversion systems tohigh-density energy carriers) have a much lower landfootprint per unit of energy than current biofuel


feedstocks that were selected for high sugar or oil con-tents Some of the benefits described above will onlybe fully realized when the infrastructure is in place tosupport cellulosic ethanol production Thus the time-line established by policies can affect the temporalsystem boundary that is most relevant for these scen-arios A more detailed discussion of the effects oftimescales follows in sect54

Cellulosic ethanol facilities are already being built inthe USA and are an indication of the future second-generation bioenergy market that is emerging VercipiaBiofuels for example is initiating a commercial-scalefacility that will produce 36 million gallons annuallyfrom cellulosic feedstocks that will be grown on an esti-mated 20 000 acres in Highlands County Florida Othercompanies such as Dupont Danisco Cellulosic EthanolLLC and Verenium BP are optimizing demonstration-scale facilities and anticipate a near-term transition tocommercial-scale production These examples areindicative of a near-term temporal system boundarythat can be applied for cellulosic feedstock crops insome regions of the USA

Species composition interacts strongly with biogeo-chemical cycling and other ecosystem services and thepotential impacts of LUCs associated with an expand-ing biofuel industry on biodiversity remain largelyunexplored [26] Biodiversity is greatly diminished asnative ecosystems give way to intensively managed agri-culture However reinserting perennial feedstocksparticularly if it becomes practical to use species mix-tures or mixed landscapes has enormous potential torestore some of this diversity [127]

23 Integrated land management

The objective of integrated land use is to find a balancebetween economic social and environmental objectives[28] Land is limited and the fertility productivity andfunctionality of lands are widely variable across spaceIn the case of degraded or abandoned lands LUCmay contribute to improved utilization of availableresources that support local and global environmentalsustainability Accomplishing such sustainable utiliz-ation however can be problematic and often requirescomplex and well-balanced incentives The needs of sta-keholders must be balanced with the costs and benefitsto the environment and societal welfare

Competing interests at the local and global scale cancomplicate decision-making about integrated land use[29] but iLUC risk can be reduced with an integratedapproach to land resource management The scenarioof cattle intensification in Brazil that was describedabove is an example of an integrated approach becausetwo needs are being met on the same amount of landthat previously had a single purpose Another kind ofintegrated land use is exemplified by the use of Jatrophacurcas L on small-scale plantations in Zambia wherecrops in some cases serve both as exports and resourcesfor the local community (other side effects like invasive-ness notwithstanding)

There also are many scenarios of integrated land usewhere crop residues can be used as biofuel feedstocks InMexico 38 per cent of the biomass in Agave tequilana

Table 2 Illustration of US corn ethanol with respect to a baseline without biofuels (Plus symbols for positive effects andminus for negative effects italic for externalities)

corn ethanol versus nobiofuels baseline (USA) local cross-sectoral boundaries global cross-sectoral boundaries

benefits farm support to corn producers (thornthorn) higher price for world cereals farmers (thornthornthorn)energy security (thornthorn) direct GHG savings (2 or thorn)

costs cereal prices in the USA (222) food security in the world (222)burden for tax payer (22) indirect land-use change (carbon biodiversity)

(22)intensification of cultivation (water stressnitrogen) (22)


plants that are harvested for the purpose of fermenta-tion and distillation to tequila is in the leaves [3031]which are not used to make tequila This leaf biomasscould serve as biofuel feedstock Additionally bagasseand vinasse from the tequila-making process can beused as feedstocks for heat and power [32] If all ofthe resources are used three products can be producedon the same land that previously yielded only one Thisin theory maximizes the efficiency of the land-basedresources within a finite regional boundary withoutcompromising existing benefits

24 Negative land-use change scenarios

Generally LUC can be considered negative if there is aloss of ecosystem or social services that outweighs thebioenergy benefits that could be gained Consideredalone a small-scale bioenergy project may appear notto have an impact beyond the local vicinity Howeversince the US and EU policies resulted in a significantreallocation of croplands to the production of biofuelsa body of literature has emerged that attempts toevaluate the market-mediated effects of land diversion

From the perspective of climate mitigation conversionof land with large carbon stores to high-input annuallytilled crops can degrade or eliminate the potential advan-tages of bioenergy The climate services of an ecosystemcan be summarized as its greenhouse gas value quantifiedby summing its net uptake of greenhouse gases (CO2N2O CH4) with the losses of these constituents whenthe ecosystem is replaced by another Replacing a tropicalpeat forest for example with pasture greatly reduces thecapacity of this landscape to store GHG and mitigate cli-mate change [33] In some cases it can take centuries torecover The net losses of GHG to the atmospherecaused by converting large-stature ecosystems with largecarbon stocks such as forests to low-stature herbaceousfeedstocks with high rates of carbon turnover maynegate the potential for climate change mitigation Itwould take in excess of three centuries for example torepay the carbon debt associated with the conversion oftropical rainforest to palm or soya bean biodiesel [10]

While converting forests to herbaceous vegetationclearly demonstrates a large loss of above-ground bio-mass the net losses of soil organic carbon (SOC)associated with LUC also can be significant Conversionof prairie with its dominant perennial grasses toannually tilled row crops mostly soya bean and maizein the USA has contributed to substantial losses of


SOC and reduced soil fertility [3435] While losses ofabove-ground biomass can be repaid by convertingannually tilled land to perennial feedstocks such asswitchgrass miscanthus or native prairie the paybacktime for SOC losses can be very long in some soilsRestoring SOC following conversion of native forest tosugarcane is measured in decades to centuries [36]

The net effect of LUC on greenhouse gas emissions isof primary concern because greenhouse gas reductionis one of the main justifications for transitioning frompetroleum to biofuels but a myriad of ecosystem ser-vices probably will be affected by LUC associatedwith the widespread deployment of biofuel cropsCorn agro-ecosystems for example are notorious forinefficient nitrogen cycling expanding the use of cornfor ethanol production will increase nitrate contami-nation of ground water and N2O emissions to theatmosphere [37ndash39] Because of its high input costslow energy yield and negative effects on the environ-ment ethanol from corn grain is not a sustainableenergy source in the long term The economic andenergy benefits of corn-ethanol are outweighed by thenegative societal and environmental costs (table 2)

3 LAND-USE MODELS AND DATA

31 How do land-use models represent land-usechange and bioenergy

The nature of land use at a macro-scale makes it particu-larly difficult to represent in a model of LUC dynamicsIn a recent review of competing land-use interests glob-ally Smith et al [40] concluded that policies governingproduction and markets affect land-use practices morethan land-use competition affects the markets for terres-trial products Competition for land in itself is not adriver affecting food and farming in the future but isan emergent property of other drivers and pressuresThere is considerable uncertainty over projectionsof competition for land in the future and the regionaldistribution of this competition This means thatmodels used for land-use assessment need to incorporatea wide range of drivers from macro-economic indicatorsto local policy specifications

Different families of models are presented in Smithet al [40] Some models are top down such as the gen-eral equilibrium macro-economic model EPPA whichprojects the effects of economic growth on five landtypes including bioenergy cropland at 5 year intervals

Table 3 Classification schemes for land use The GlobalLand Cover (GLC 2000) scheme is recognized by mostEuropean nations and the Good Practice Guidance for LandUse Land Use Change and Forestry (GPG-LULUCF) isrecognized by the Intergovernmental Panel on ClimateChange (IPPC) and the United Nations

model GLC-2000GPG-LULUCF

classifications pristine forest forest landmanaged forestshort rotation tree plantationcropland (IIASAmdash

subsistence low input highinput irrigated)

cropland

grassland grasslandother natural vegetation wetlands

settlementsother land


(eg [6]) This model assumes that conversion of landback to a lsquonatural statersquo has no cost an assumptionthat overlooks the need to assess counterfactual con-ditions Two other models that categorize bioenergyas a unique land use IMAGE and MiniCAM areboth integrated assessment models [41ndash44] IMAGEincludes clear geographically delineated land-use cat-egories but must be coupled with a separateeconomic model to simulate dynamics of LUC overtime [4041] MiniCAM however simulates the feed-backs between profitability margins and landallocation to different categories [4042ndash44]

Feedbacks to ecosystem services are the least rep-resented (relative to effects on production andeconomics) in integrated assessment models like Mini-CAM and are more often modelled regionally withoutconsidering interactions with the global market Con-nections between regional responses of ecosystemservices including greenhouse gas mitigation andcarbon sequestration and LUC must be made inorder to assess global scenarios of biofuel-related LUC

To properly evaluate scenarios of LUC as positive ornegative we must integrate results from different typesof regional models that collectively include a wide var-iety of essential indicators and approaches This doesnot mean that more models are needed but ratherthat a wide variety of existing tools must be used inaggregate to assess LUC A combination of pro-ductivity biogeochemistry economics environmentalimpact and social impact models must be employed toclarify the potential consequences of bioenergy in differ-ent regions of the world These modelling tools havealready been developed and are subject to ongoing cri-tique within specialized disciplines that define each ofthese model classes The combined results of multidisci-plinary state-of-the-art models should be informativefor assessing LUC outcomes These models should ide-ally subscribe to similar criteria in all regions relyingon data sources analogous to a global unified databaseand ground-truth verifications of projections (throughdata collection and monitoring)

32 Do we have the data we need to address theimpact of land-use change

Model inputs depend on land cover information (agri-cultureforestrygrassland) This is available fromvarious products at different resolutions Products areimproving with satellite technology but there are stilldifferences among datasets that are partly due to classi-fication (eg per cent coverage of trees that classifiesland as a forest can vary from 20 to 60) Forexample when we compare two widely used schemesfor land-use classification (table 3) the categories areinconsistent How can a global comprehensive modelof LUC reconcile different classification schemes thatare applied to different regions or scenarios Standard-ization of land-use categories would increase therelevance of LUC models for global analysis andshould be inclusive of subdivisions with varied manage-ment practices that are employed throughout the world

Bioenergy cropland is not explicitly identified inmost internationally recognized land-use classification


schemes (eg table 3) because it has historically rep-resented a very small amount of land that isdistributed between several land categories (croplandshort rotation plantations) The amount of bioenergycropland has grown substantially in recent years andin 2009 11 per cent of US cropland was devoted tocorn for ethanol and 5 per cent of EU cropland to rape-seed for biodiesel Obviously this would be a keycategory for explicit analyses of LUC for bioenergy

Another limitation of the present categorical defi-nitions is the lack of information about productivitypotentials of biofuel crops relative to the existingproduction in each of these land-use types Globalestimates of ecosystem productivity under different landuses are available from the Agro-Ecosystem Zoning pro-ject (International Institute for Applied SystemsAnalysis Laxenburg Austria) and other sources butwithout a parallel characterization of the productionpotential both in a natural state and for biofuel cropsthe criteria that define positive versus negative scenarioscannot be assessed Databases such as GLASOD andGLADA characterize the degraded nature of croplands[4546] and may be a source of data for assessing theproduction potential of different regions

Other examples of categories that are not distinguishedby either the Global Land Cover (GLC) or the Good Prac-tice Guidance for Land Use Land Use Change andForestry (GPG-LULUCF) classification system (table 3)but are relevant to societies that may be regionally con-fined include livestock (intensive pasturerangeland)subsistence abandoned croplands urban protected highbiodiversity and peatland A database for abandoned crop-lands has been compiled by the Center for Sustainabilityand the Global Environment in Madison WisconsinUSA [47] and the protected and high-biodiversityclassifications are reported by the United Nations WorldConservation Monitoring Center [4849]

Lastly more information on carbon stocks of eco-systems is necessary for accurate model predictionsChanges in land use (primarily deforestation) have beenresponsible for about 20 per cent of all anthropogenicemissions of CO2 to the atmosphere over the past two


and a half decades but this term is the most uncertainvariable in the global carbon budget (IPCC FourthAssessment Report [50]) The estimated emissions fromLUC for the 1990s in the IPCC Fourth AssessmentReport (a negligible part of it being related to the biofuelsector that was only present in Brazil at the time of assess-ment) were 16 Gt C y21 (59 Gt CO2 y21) with a rangeof 05ndash27 Gt C y21 across different estimates The widerange is primarily due to differences in quantification ofthe area the fate of the carbon and the nature of theLUC Some elements that are critical for calculating acomplete LUC-related flux of carbon include

mdash the type of vegetation change (eg deforestationafforestation shifting cultivation logging croplandand pasture establishment and abandonment)

mdash the amount of carbon present in vegetation and soilsbefore and after the change

mdash the method used for the clearing event (eg loggingversus slash and burn)

mdash the fate of biomass following clearing (eg paperversus long-lived wood products)

mdash regrowth of vegetation following clearing andmdash the dynamics of soil carbon following clearing

While there is a growing wealth of data that des-cribe land use globally there is still a need for aglobally comprehensive categorization internationalstandardization and inter-disciplinary aggregation ofthose data with the variables that define sustainability

4 ASSESSING INDIRECT LAND-USECHANGE

The US environmental agencies (Environmental Protec-tion Agency (EPA) and California Air Resource Board)were the first to attempt incorporating the iLUC dimen-sion in legislation that defines which biofuel crops shouldbe supported for future domestic use These agenciesrelied on various agricultural energy and economicmodels to evaluate the effects of iLUC [5152] Althoughall models predict some iLUC associated with bioenergyexpansion policy makers had to face a large variation inthe estimates because of the inherent difficulty of trackingthe interactive variables that contribute to iLUC Never-theless it was clear that all crops were not equal incausing iLUC Different initial yields differences in thesensitivity of areas where land competition takes placeand trade configuration can induce significant variationin computed results [5354] The incorporation of theseeffects in legislation sparked strong controversies betweenvarious interest groups during public consultations andthe EPA position was attacked in Congress The finalEPA assessment presented a more moderate impact ofiLUC than in the 2009 draft and in the end qualified allmajor biofuel pathways even though several corn ethanolprocesses had initially been disqualified

In the EU the European Commission received withthe 2009 Renewable Fuel Directive the responsibilityof writing a report to the European Parliament on thequestion of iLUC by the end of 2010 This was intendedto open the way to legislative options for mitigating theimpact of biofuel policies The Commission ordered sev-eral studies and literature reviews on the topic in 2009


and 2010 [955ndash57] Although there was considerablecontroversy about the interpretation of a few selectedfigures these reports when analysed in detail did notdepart from the US findings High variation in theestimates of iLUC were presented with figures rangingfrom acceptable values to disqualifying ones raisingquestions but not offering conclusions on the environ-mental efficiency of biofuel policies Interestingly likein the US studies impacts depended on the cropconsidered suggesting that a beneficial regulationshould generally promote high-yielding feedstocks andsecond-generation biofuels [758]

The use of macro-sectoral models for computation ofiLUC in life-cycle assessments is now deemed prematurebecause of the uncertainties of parameters in a field thatis still in its early development Modelling efforts havehowever succeeded in illustrating the rationale ofiLUC and the need for market intervention to mitigateside effects of biofuel policies These side effects includeleakage owing to the feedback from biofuel consump-tion mandates on fuel prices and subsequent increasesin fossil energy consumption in the rest of the world[5960] Another side effect can be shuffling on the pro-duction side between inefficient and efficient productionunits without real eviction from the market of undesir-able production processes [61] Sustainability criteriawould remain partially ineffective as long as biofuel pol-icies and land-use decisions are not integrated in a moregeneral economic instrument of carbon regulation[6263] Modelling approaches reveal different crops anddifferent types of biofuels that vary in their potential forclimate change mitigation Measures limited to the pro-ject scale would be ineffective in addressing macro-scaleimpacts if applied only to a few regions or products

In summary of the issues associated with modellingiLUC recent studies find that iLUC is a problem thatmust be considered (eg [64]) but there is a significantamount of uncertainty in projecting iLUC Part of thisuncertainty results from complex economic factors thatare difficult to measure Generally we can say thatiLUC is important and should be considered but largeconfidence intervals prevent precise figures that canserve as regulatory guidelines Models are however appro-priate to identify trade-mediated externalities andhierarchical relationships of low- and high-risk optionsThere are regionally specific datasets that characterizeLUC interactions and policies that address the pro-duction of multiple resources within a regional boundarywill reduce externalities and lower the risk of iLUC

5 HOW CAN MODELS BE IMPROVED FORPOLICY DECISIONS

51 Regional details of land management

Regionally detailed costs and benefits of LUC for bioe-nergy must be further explored to identify moreintegrated bioenergy management choices For examplein Europe very little land is actually unused In contrastthere are regions in the USA that comprise relativelyunused land some of which has previously been in agricul-ture and abandoned In Africa there are areas of openrangeland for subsistence livestock communal grazing

Table 4 Illustration of a US cellulosic ethanol with respect to a baseline without biofuels (Plus for positive effects and minusfor negative effects italic for externalities)

cellulosic ethanol versus nobiofuels baseline (USA) local cross-sectoral boundaries global cross-sectoral boundaries

benefits farm support to feedstock producers (thornthorn) higher price for world cereals farmers (thorn)energy security (thornthorn) direct GHG savings (thornthorn)

costs food prices in the USA (2) food security in the world (2)burden for tax payer (222) indirect land-use change (carbon

biodiversity) (2)intensification of cultivation (water stress

nitrogen) (2)

Table 5 Illustration of a US cellulosic ethanol with respect to a baseline with biofuels (frac14 table 4 2 table 2) (Plus for positiveeffects and minus for negative effects italic for externalities)

cellulosic ethanol versuscorn ethanol (USA) local cross-sectoral boundaries global cross-sectoral boundaries

benefits farm support to feedstock producers (thornthorn) direct GHG savings (thorn to thornthornthorn)food prices in the USA (thornthorn) food security in the world (thornthorn)lesser intensification of cultivation (water stress

nitrogen) (thorn)indirect land-use change (carbon

biodiversity) (thornthorn)

costs farm support to corn producers (22) lower price for world cereals farmers (22)burden for US tax payer (2)


land collection of wood fuel traditional medicines un-protected nature reserves etc The opportunities associ-ated with these regions are unique and should be treatedas such in analyses of bioenergy options while stillacknowledging potential externalities

Common to all regions and economies are crop areasthat are underproductive relative to the potential ter-restrial yield Biofuel markets may become a strongeconomic driver to do something on land that is notbeing used efficiently or where people currently havelow earning power For bioenergy development it willbe essential to identify places where improvementscan be made in husbandry practices biologicalproductivity and the local economy

52 Toward integrated assessment

The concept of integrated assessment is often misrepre-sented in models because co-benefits of productionpractices are not fully evaluated This limitation does notdisqualify modelling but instead calls for a better charac-terization of possible co-benefits in any assessmentexercise Biofuel co-products for example were inade-quately represented by the first models addressing LUCfor biofuels [65] After the importance of co-products wasidentified it became a prerequisite for any reliable model-ling exercise On the other side many other mechanismsand co-benefits are still badly represented in models asemphasized by Babcock amp Carriquiry [66] in an evaluationof the potential for double cropping or the use of idle ordegraded land There remains room to improve models inthis respect and measures of iLUC might be substantiallymodified as a result

In order to determine good and bad scenarios of LUCrelated to bioenergy policies it is therefore necessary tomake a holistic assessment where all externalities are


taken into account The challenge is to set boundariesthat make the assessment tractable and at the sametime exhaustive enough and relevant to the most realis-tic business-as-usual baseline If we take the case of USethanol as an example we propose in table 1 a possibleassessment grid that comprehensive studies should con-sistently and completely address Table 2 illustrates aqualitative evaluation of corn ethanol that is currentlyproduced in the USA

The case of cellulosic ethanol from miscanthus isframed in table 4 similar to the corn ethanol case intable 2 Alternatively to accurately reflect current bio-fuel agriculture in the USA one could establish thebaseline business-as-usual counterfactual to include bio-fuels that are presently based on a first-generationtechnology (eg corn grain) Assessing biofuel scenariosbased on these current conditions leads to differentresults (table 5) than if one ignores the current baselineand the multiple products from a landscape with mixeduses (table 4)

The EPA attempted a quantitative and comprehen-sive costndashbenefit analysis of the US biofuel programme[52] finding some benefits to the policy if current mar-kets are ignored but negative effects are evident if theyare included in the assessment The methodology forsuch assessments remains highly controversial but themost practical framework for evaluating impacts ofnew biofuel feedstocks includes a baseline that reflectscurrent real-world conditions (as in table 5) andmultiple resources that would be produced within awell-defined geographical boundary

53 Uncertainty and gaps in knowledge

Models are only as good as the data used to parametrizethem It is fair to assume that model uncertainty for


land-use analysis is high because there are inconsistenttypes of data collection and incomplete process descrip-tions Nevertheless models can serve as tools toestimate broad outcomes or unexpected trends Forexample the magnitude of projections calculated bySearchinger et al [8] is now considered closer to theupper range of estimates [10] but the calculationsdescribed in the study led to more awareness aboutpotential iLUC Different models are fit for differentpurposes may be only regionally appropriate and maybe specific to particular processes Combinations ofcomplementary models and integrated approachesmay yield valuable information that can direct policiesand land management as long as prudence remainsthe rule with respect to any exact estimates ofeffects produced

We need to develop more comprehensive infor-mation on actual land use and management carbonstocks and ecosystem services that are consistent andcomparable globally Even for categories of land usethat are broadly agreed upon (eg agriculture or for-ests) there is a substantial variation in ecosystemservices For example there can be very different bio-mass different soil carbon different rotations andmanagement all of which can affect the overall green-house gas balance of changing land use Each of thesecharacteristics further introduces uncertainty aboutthe relative loss of services that the land was alreadyproviding (food timber biodiversity) We need thisinformation to judge how much land might be lsquoavail-ablersquo which land could be converted and what theconsequences would be for ecosystem services (eggreenhouse gases)

Consistent data monitoring is essential for compari-son of LUC criteria at varied scales Global datasetsand national datasets differ because of measurementresolution techniques and analyses More completenational datasets will allow validation of global datasetsand improve detail Datasets based on high-resolutionsatellites and country-level reporting achieve consist-ency for some land variables but there are others thatmust be measured on site even if only for validationFAO and EUROSTAT are examples of internationaldata reports that may inform international LUCquestionsmdashbut even in Europe these are not always con-sistent Some land areas will soon be more accuratelycharacterized as many countries are carrying out eco-system assessments in response to the MillenniumEcosystem Assessment [67] Other country-level datamay exist from national statistics scientific surveys etcIn sum the varying quality and quantity of data by acountryregion must be addressed

For near-term developments in bioenergy we suggestan emphasis on a smaller number of low-risk opportu-nities in regions where varied land use and multiplebenefits can be managed and evaluated in an integratedway and within a clear geographical boundary Devel-oping countries often with the least data may havethe greatest opportunities for positive LUC for bio-energy and will rely most heavily on models withminimal data For these opportunities to be realizedtools must be developed to measure the actual use ofland and ecosystem services Policies should ensure


that identified positive scenarios are realized and notreshaped for purely opportunistic purposes that couldcompromise the expected benefits

54 Scenarios and time-scale issues

Time scale is of critical importance when assessing theoverall impact of bioenergy projects particularly withrespect to net greenhouse gas balances and pay backtimes for initial impacts that can be weighed againstthe lifetime of bioenergy production Politicians aremost interested in 2020 targets and 2050 targets butthe best options are likely to depend on whether a 10year or 40 year perspective is adopted In evaluatingthe costs of LUC for bioenergy variables to considerinclude how long a particular land management laststime of establishment the permanence of ecosystem ser-vices lost or gained and the variability of impacts andservices over time

In the near future second-generation technologiesare likely to advance substantially It is importantthat policies do not get locked into first-generationbiofuel chains that might fail to promote or eveninhibit the development of second-generation fuelsIndeed first-generation biofuels from cereals areexpensive and their benefits are limited relative tosecond-generation biofuels [52]

Policy makers have mandated very ambitious targetsfor biofuel incorporation up to 2020 in the EU and 2022in the USA This might be accomplished by usingresources more efficiently this could include waste con-versions and improvement of management practices tooptimize production Since cellulosic feedstock pro-duction is expanding (as demonstration by VercipiaBiofuels Dupont Danisco Cellulosic Ethanol LLCetc) there is likely to be an opportunity to use agricul-tural residues that are low in sugar or oil increasing theamount of biomass that can be used from existing cropsThese are possibilities that require very little LUC andcould be achieved by 2020

There are many differences in the temporal assump-tions of LUC models and scenarios For example theFORESIGHT analysis of different LUC scenariosdemonstrates the difference in outcomes that areachieved when an LUC scenario is implemented up to2020 versus 2050 [40] Most of the outcomes that dif-fered in terms of the areas and types of LUCs are dueto the assumptions in the scenarios (IPCC MEAGEO etc) about future conditions These scenariosare implemented with different temporally dynamicdrivers (eg IMAGE MminiCAM etc) that are subjectto a variety of assumptions

A more long-term bioenergy outlook 2050 andbeyond would be primarily based on second-generationfeedstocks but the drivers of LUC will depend onadvances in technology and non-technological changesin society that become harder to project farther intothe future Changes in diet for example can lead tochanges in the amount of land that is used for livestockand pasture that competes with land available for fuelcrops While bioenergy-related LUC is a timely problemthat needs to be addressed bioenergy has not been themain driver of LUC in the past Thus policies that


incentivize integrated land management woulddiscourage unintended LUC in the future

The role of bioenergy in global LUC depends on theextent to which we favour options that do not need a lotof land that extend on low-quality soils or wherefeedstocks are co-produced with food feed and fibreLong-term changes in societal trends and productionpractices may thus have different effects on projectionsfrom LUC models that run at different timescales

6 CONCLUSIONS

Regionally specific strategies are needed to optimizeresource-use efficiency and reduce iLUC owing to bio-energy development At the same time there is aneed for cooperation among regions to maintain fairaccountability of land conversion at the global scaleThis underscores the importance of common standardsbetween regions with different needs and resources Thefundamental goal of policies should be to maximize thevalue of land services at the lowest social cost If thiscan be accomplished on a smaller land footprint thenlands will be available for energy feedstocks To mini-mize greenhouse gas emissions all resources that arecultivated should be subject to the same emissions stan-dards management of any agricultural resource can beimproved to use less land and lower carbon emissionsEfficient and sustainable land management should berewarded equally (eg by way of incentives) in thecase of food production feed and livestock cultivationand bioenergy feedstock plantations

Recently there has been an effort to develop modelsthat can assess iLUC but results are still so diversethat these models cannot usefully inform detailedmanagement decisions GLC datasets lack detailedinformation on actual land use (eg grassland beingused as pasture subsistence rangeland) and associatedecosystem services (eg carbon sequestration biodiver-sity) which is necessary to assess net environmentalimpacts and services displaced by using that land forbioenergy We conclude that global data are not yetsufficient for full assessment of the impacts of imple-menting bioenergy scenarios It is however possible todetermine impacts by aggregating different modellingtools and data that are available regionally Optionsthat guarantee low risk of iLUC high potential for posi-tive externalities and increased efficiency of resourceproduction should be promoted in the short termwhile maintaining a long-term view of societal demandsand technology advancement

REFERENCES

1 Somerville C Youngs H Taylor C Davis S C ampLong S P 2010 Feedstocks for lignocellulosic fuelsScience 13 790ndash792 (doi101126science1189268)

2 Campbell J E Lobell D B Genova R C amp FieldC B 2008 The global potential of bioenergy onabandoned agricultural lands Environ Sci Technol 425791ndash5794 (doi101021es800052w)

3 Varvel G E Vogel K P Mitchell R B Follett R F ampKimble J M 2008 Comparison of corn and switchgrass on


marginal soils for bioenergy Biomass Bioenergy 32 18ndash21(doi101016jbiombioe200707003)

4 Dale B E Bals B D Kim S amp Eranki P 2010 Biofuelsdone right land efficient animal feeds enable largeenvironmental and energy benefits Environ Sci Technol44 8385ndash8389 (doi101021es101864b)

5 Gurgel A Reilly J M amp Paltsev S 2007 Potential landuse implications of a global biofuels industry J AgricFood Indus Organ 5 Article 9 See httpwwwbepresscomjafiovol5iss2art9 (doi1022021542-04851202)

6 Havlık P et al 2010 Global land-use implications of firstand second generation biofuel targets Energy Policy(doi101016jenpol201003030)

7 Searchinger T Heimlich R Houghton R A DongF X Elobeid A Fabiosa J Tokgoz S Hayes D ampYu T H 2008 Use of US croplands for biofuels increasesgreenhouse gases through emissions from land-use changeScience 319 1238ndash1240 (doi101126science1151861)

8 Edwards R Mulligan D amp Marelli L 2010 Indirect landuse change from increased biofuels demand comparison ofmodels and results for marginal biofuels production fromdifferent feedstocks Report no JRC 59771 Joint ResearchCenter European Commission

9 Plevin R OrsquoHare M Jones A Torn M amp Gibbs H2010 Greenhouse gas emissions from biofuelsrsquo indirectland use change are uncertain but may be much greaterthan previously estimated Environ Sci Technol 448015ndash8021 (doi101021es101946t)

10 Fargione J Hill J Tilman D Polasky S ampHawthorne P 2008 Land clearing and the biofuelcarbon debt Science 319 1235ndash1238 (doi101126science1152747)

11 Heaton E A Dohleman F G amp Long S P 2008 Meet-ing US biofuel goals with less land the potential ofMiscanthus Global Change Biol 14 2000ndash2014 (doi101111j1365-2486200801662x)

12 Davis S C Parton W J Del Grosso S J Keough CMarx E Adler P R amp DeLucia E H 2010 Impact ofsecond-generation biofuel agriculture on greenhouse gasemissions in the corn-growing regions of the US In 2ndPan American Congress on Plants and BioenergyAugust 8ndash11 Sao Pedro Brazil

13 Davis S C Parton W J Dohleman F G Smith C MDel Grosso S Kent A D amp DeLucia E H 2010 Com-parative biogeochemical cycles of bioenergy crops revealnitrogen-fixation and low greenhouse gas emissions in aMiscanthus x giganteus agro-ecosystem Ecosystems 13144ndash156 (doi101007s10021-009-9306-9)

14 Davis S C Anderson-Teixeira K J amp DeLucia E H2009 Life-cycle analysis and the ecology of biofuelsTrends Plant Sci 14 140ndash146 (doi101016jtplants200812006)

15 Dale V H Kline K L Wiens J amp Fargione J 2010Biofuels implications for land use and biodiversity Bio-fuels and Sustainability Reports Washington DC TheEcological Society of America 13 pp

16 David M B McIsaac G F Darmody R G ampOmonode R 2009 Long-term changes in Mollisol organiccarbon and nitrogen J Environ Qual 38 200ndash211(doi102134jeq20080132)

17 Schulze E D et al 2009 Importance of methaneand nitrous oxide for Europersquos terrestrial greenhouse-gasbalance Nat Geosci 2 842ndash850 (doi101038ngeo686)

18 Dohleman F G Heaton E A Leakey A D B amp LongS P 2009 Does greater leaf-level photosynthesis explainthe larger solar energy conversion efficiency of miscanthusrelative to switchgrass Plant Cell Environ 32 1525ndash1537 (doi101111j1365-3040200902017x)


19 Hickman G C Vanloocke A Dohleman F G ampBernacchi C J 2010 A comparison of canopy evapotran-spiration for maize and two perennial grasses identified aspotential bioenergy crops Global Change Biol Bioenergy4 57ndash68 (doi101111j1757-1707201001050x)

20 Vanloocke A Bernacchi C J amp Twine T E 2010The impacts of Miscanthus x giganteus production onthe Midwest US hydrologic cycle Global Change BiolBioenergy 4 180ndash191 (doi101111j1757-1707201001053x)

21 Goldemberg J 2008 The Brazilian biofuels industryBiotechnology and Biofuels 1 6 (doi1011861754-6834-1-6)

22 Borland A M Griffiths H Hartwell J amp SmithJ A C 2009 Exploiting the potential of plants with cras-sulacean acid metabolism for bioenergy production onmarginal lands J Exp Bot 60 2879ndash2896 (doi101093jxberp118)

23 Smith A M 2008 Prospects for increasing starch andsucrose yields for bioethanol production Plant J 54546ndash558 (doi101111j1365-313X200803468x)

24 Davis S C Dohleman F G amp Long S P 2011 Theglobal potential for Agave as a biofuel feedstock GlobalChange Biol Bioenergy 3 68ndash78 (doi101111j1757-1707201001077x)

25 Holtum J A M Chambers D amp Tan D K Y 2011Agave as a biofuel feedstock in Australia Global ChangeBiol Bioenergy 3 58ndash67 (doi101111j1757-1707201001083x)

26 Fetcher Jr R J Robertson B A Evans J Doran P JAlavalaparti J R R amp Schemske D W 2010 Biodiver-sity conservation in the era of biofuels risks andopportunities Front Ecol Environ (doi101890090091)

27 Tilman D Hill J amp Lehman C 2006 Carbon-negativebiofuels from low-input high-diversity grassland biomassScience 314 1598ndash1600 (doi101126science1133306)

28 DeFries R S Foley J A amp Asner G A 2004 Land-usechoices balancing human needs and ecosystem functionFront Ecol Environ 2 249ndash257 (doi1018901540-9295(2004)002[0249LCBHNA]20CO2)

29 van Dam J Junginger M amp Faaij A P C 2010 Fromthe global efforts on certification of bioenergy towardsan integrated approach based on sustainable land useplanning Renew Sustain Energy Rev 14 371ndash394(doi101016jrser201007010)

30 Cedeno M C 1995 Tequila production Crit Rev Bio-technol 15 1ndash11 (doi10310907388559509150529)

31 Iniquez-Covarrubias G Lange S E amp Rowell R M2001 Utilization of byproducts from the tequila industryI Agave bagasse as a raw material for animal feedingand fiberboard production Bioresour Technol 7725ndash32 (doi101016S0960-8524(00)00137-1)

32 Mendez-Acosta H O Snell-Castro R Alcaraz-GonzalezV Gonzalez-Alvarez V amp Pelayo-Ortiz C 2010Anaerobic treatment of tequila vinasses in a CSTR-typedigester Biodegradation 21 357ndash363 (doi101007s10532-009-9306-7)

33 Anderson-Teixeira K J amp DeLucia E H 2010 Thegreenhouse gas value of ecosystems Global Change Biol17 425ndash438 (doi101111j1365-2486201002220x)

34 Tilman D Wedin D amp Knops J 1996 Productivity andsustainability influenced by biodiversity in grasslandecosystems Nature 379 718ndash720 (doi101038379718a0)

35 Lal R 2004 Soil carbon sequestration impacts onglobal climate change and food security Science 3041623ndash1627 (doi101126science1097396)

36 Anderson-Teixeira K J Davis S C Masters M D ampDeLucia E H 2009 Changes in soil organic carbon under


biofuel crops Global Change Biol BioEnergy 1 75ndash96(doi101111j1757-1707200801001x)

37 David M B Drinkwater L E amp McIsaac G F 2010Sources of nitrate yields in the Mississippi River basinJ Environ Chem 5 1657ndash1667 (doi102134jeq20100115)

38 Erisman J W van Grinsven H Leip A Mosier A ampBleecker A 2010 Nitrogen and biofuels an overview ofthe current state of knowledge Nutr Cycl Agroecosyst86 211ndash223 (doi101007s10705-009-9285-4)

39 Woli K P David M B Darmody R G Mitchell C A ampSmith C M 2010 Assessing the nitrous oxide mole fraction ofsoils from perennial biofuel and corn-soybean fields AgricEcosyst Environ 138 299ndash305 (doi101016jagee201006002)

40 Smith P Gregory P J van Vuuren D Obersteiner MHavlik P Rousevell MWoods J Stehfest E amp BellarbyJ 2010 Competition for land Phil Trans R Soc B 3652941ndash2957 (doi101098rstb20100127)

41 MNP 2006 Integrated modelling of global environmentalchange an overview of IMAGE 24 In Expert meetingon how to feed the world in 2050 Food and AgricultureOrganization of the United Nations Economic andSocial Development Department

42 Gillinghem K T Smith S J amp Sands R D 2008 Impact ofbioenergy crops in a carbon dioxide constrained world anapplication of the MiniCAM energy-agriculture and landuse model Mitig Adapt Strat Global Change 13 675ndash701(doi101007s11027-007-9122-5)

43 Wise M Calvin K Thomson A Clarke L Bond-Lamb-ertyB SandsR Smith S JanetosAampEdmonds J 2009The implications of limiting CO2 concentrations for land useland-use change emissions and bioenergy PNNL-18341Pacific Northwest National Laboratory

44 Wise M Calvin K Thomson A Clarke L Bond-Lamberty B Sands R Smith S J Janetos A ampEdmonds J 2009 Implications of limiting CO2 concen-trations for land use and energy Science 324 1183ndash1186 (doi101126science1168475)

45 Nachtergaele F O F amp Auricht C M 2004 GLADAGlobal Land Degradation Assessment Progress Report Seeftpftpfaoorgaglagllladaglada_project_statuspdf

46 Nachtergaele F O F amp Licona-Manzur C 2009 TheLand Degradation Assessment in Drylands (LADA) Pro-ject reflections on indicators for land degradationassessment In The future of drylands (eds C Lee ampT Schaaf) Rome Italy Food and Agriculture Organiz-ation of the United Nations

47 Goldewijk K K amp Ramankutty N 2004 Land coverchange over the last three centuries due to human activi-ties the availability of new global data sets GeoJournal61 335ndash344 (doi101007s10708-004-5050-z)

48 Chape S Spalding M amp Jenkins M (eds) 2008 Theworldrsquos protected areas status values and prospects inthe 21st century Berkeley CA University of CaliforniaPress

49 Groombridge B amp Jenkins M 2002 World atlas of bio-diversity earthrsquos living resources in the 21st centuryBerkeley CA University of California Press

50 Denman K L et al 2007 Couplings between changes in theclimate system and biogeochemistry In Climate Change2007 The Physical Science Basis Contribution of WorkingGroup I to the Fourth Assessment Report of the Intergovern-mental Panel on Climate Change Cambridge UKCambridge University Press

51 CARB 2009 Proposed Regulation to Implement theLow Carbon Fuel Standard Vol I Staff Report InitialStatement of Reasons Sacramento CA


52 EPA 2010 Renewable Fuel Standard Program(RFS2) Regulatory Impact Analysis US EnvironmentalProtection Agency Report no EPA-420-R-10-006Washington DC

53 Bouet A Dimaranan B V amp Valin H 2010 Modelingthe global trade and environmental impacts of biofuel pol-icies Discussion Paper no 01018 International FoodPolicy Research Institute

54 Hertel T W Golub A A Jones A D OrsquoHare MPlevin R amp Kammen D M 2010 Effects of US maizeethanol on global land use and greenhouse gas emissionsestimating market-mediated responses Bioscience 60223ndash231 (doi101525bio20106038)

55 Al-Riffai P Dimaranan B V amp Laborde D 2010 Globaltrade and environmental impact study of the EU biofuelsmandate International Food Policy Research InstituteReport to the Directorate General of Trade EuropeanCommission ATLASS Consortium

56 EC 2010 The impact of land use change on greenhouse gasemissions from biofuels and bioliquids literature reviewEuropean Commission See httpeceuropaeuenergyrenewablesconsultationsdocpublic_consultation_ilucstudy_3_land_use_change_literature_review_final_30_7_10 pdf

57 Fonseca M B Burrell A Gay H Henseler MKavallari A MrsquoBarek R Dominguez I P ampTonini A 2010 Impacts of the EU Biofuel Target onAgricultural Markets and Land Use a Comparative Mod-elling Assessment Report no JRC 58484 Joint ResearchCenter European Commission

58 Melillo J et al 2009 Indirect emissions from biofuels howimportant are they Science 326 5958 (doi101126science1180251)


59 Babiker M H 2005 Climate change policy market struc-ture and carbon leakage J Int Econ 65 421ndash445(doi101016jjinteco200401003)

60 Drabik D amp de Gorter H 2010 Biofuels and leakages in thefuel market In International Agricultural Trade ResearchConsortium Symposium Climate Change in World Agricul-ture Mitigation Adaptation Trade and Food Security June27ndash29 2010 Stuttgart Germany Universitat Hohenheim

61 Bushnell J Peterman C amp Wolfram C 2007 Local sol-utions to global problems policy choice and regulatoryjurisdiction Report no 13472 National Bureau ofEconomic Research

62 de Gorter H amp Just D R 2010 The social costs and benefitsof biofuels the intersection of environmental energy andagricultural policy Appl Econ Pers Policy 32 4ndash32

63 Searchinger T et al 2009 Fixing a critical climateaccounting error Science 326 527ndash528 (doi101126science1178797)

64 Dehue B Meyer S amp van de Staaij J 2010 Responsiblecultivation areas identification and certification of feed-stock production with a low risk of indirect effectsEcofys Report no September 2010 version 10

65 Prins A Stehfest E Overmars K amp Ros J 2010 Aremodels suitable for determining ILUC factors Nether-lands Environmental Assessment Agency May 2010

66 Babcock B A amp Carriquiry M 2010 lsquoAn exploration ofcertain aspects of CARBrsquos approach to modeling indirectland use from expanded biodiesel productionrsquo StaffReport 10-SR 105 Center for Agricultural and RuralDevelopment Iowa State University Ames Iowa

67 Millennium Ecosystem Assessment 2003 Ecosystemsand Human Well-being a framework for assessmentWashington DC Island Press


location Bioenergy feedstocks range from oil and sugarcrops to grasses and trees and conversion pathwaysrange from direct combustion to chemical processingfor liquid transportation fuels (eg biodiesel bio-ethanol) New conversion technologies for second-generation biofuels allow more diverse feedstocksincluding lignocellulosic biomass and residues andadvanced conversion systems to high-density energycarriers [1] These new feedstock opportunities arelikely to encourage expansion of biomass productionglobally in the future but the magnitude and locali-zation of LUC are highly dependent on the regionof production and the type of feedstock used

Because the bioenergy sector interacts with otherecosystem services a holistic approach to assessmentof land management is required With increasing com-petition for land resources there is a need forcomprehensive tools that will identify the best ways tooptimize many agricultural resources in an integratedway A few studies have recently addressed this chal-lenge in the context of specific regions (eg [2ndash4])There are also global models that aim to characterizeLUC (eg [56]) but the macro-level view of thesemodels can overlook some dimensions of potentiallybeneficial land-use options Delineation of positive andnegative scenarios for bioenergy is therefore necessaryto complement global approaches and highlight localopportunities

There has been much recent concern about indirectland-use change (iLUC) as a result of bioenergy cropsbeing grown (eg [7]) Many modelling studies that esti-mate LUC produce widely varying results as to thedegree of iLUC (see [8] for a review and [9] for discussionon uncertainty) Considering the large confidence inter-vals of model estimates it remains unclear how theseresults can be used for legislation At the same timeit is increasingly being recognized that for certaincrops integrated multiple-use land management canreduce the risk of unintended iLUC For examplebioenergy can be produced as a waste product or as aco-product with food feed or fibre Thus far tools forcarrying out proper assessments of multiple-useopportunities have not been developed

We will first present a typology that characterizesdesirable and undesirable effects of LUC associatedwith feedstock crops then provide examples of positiveand negative LUC scenarios for bioenergy productionNext we will explore how existing datasets andmodels characterize new LUC scenarios and the risksof iLUC We conclude with a discussion of how scien-tific tools that evaluate bioenergy-related LUC mightbe improved for informing policy decisions

2 CRITERIA FOR EVALUATINGBIOENERGY SCENARIOS

Different feedstock crop species bring different costsand benefits with respect to economic developmentfood security climate mitigation biodiversity andother land services A clear typology of positive andnegative effects is needed for a comprehensive evalu-ation of these different options For example planting


bioenergy crops on lands with high carbon stockssuch as peatlands and forests has negative net green-house gas consequences and other costs [10] On theother hand deploying high-yielding bioenergy cropsthat require low inputs (fertilizers irrigation) canimprove ecologically degraded and economically under-productive landscapes [411ndash13] In this section we willfirst introduce the dimensions of bioenergy productionthat must be addressed in an evaluation of bioenergy-related LUC We then discuss LUC scenarios that opti-mize biofuel benefits regionally highlight scenarios ofintegrated land management for multiple uses andfinally provide contrasting examples of scenarios thatrisk multiple negative impacts

21 Dimensions of bioenergy

To assess different LUC scenarios including those forbioenergy some common rules must be applied toensure systematically consistent evaluations In thecase of bioenergy some dimensions that appear particu-larly critical for evaluation of costs and benefits aresystem boundaries counterfactual possibilities andexternalities When evaluating iLUC these definitionsare broad as system boundaries extend to land use inthe whole world and there is little agreement abouthow externalities at this scale are best regulatedinternationally

System boundaries define the start and end points ofthe production chain (eg cradle-to-grave) the geo-graphical space and the temporal limits that apply toa given analysis System boundaries are one of the keysources of difference in estimates of overall costs andbenefits of a biofuel production chain among variouslife-cycle analyses [14] Results of any estimate ofbioenergy-related impacts should be considered in thecontext of the system boundary that was selectedfor analysis

The counterfactual possibilities should be definedrelative to the condition of land if a biofuel project inquestion is not implemented This lsquobusiness-as-usualrsquotrend can vary over time according to profitability ofbiomass for other uses future demand effects of climatechange on yield and other temporally dynamic factorsThe costs and benefits of a land-use scenario withbusiness-as-usual (that continues for the same lengthof time as a bioenergy production scenario) are theminimum counterfactuals that should be consideredwhen evaluating impacts of LUC caused by bioenergyThe bioenergy scenario would displace the business-as-usual case and therefore offset expected costs andbenefits with new temporally dynamic costs andbenefits The difference between a bioenergy land-usescenario and its counterfactual is sometimes consideredin lsquopayback timesrsquo as in estimates of carbon debtsresulting from LUC [10]

Externalities are positive or negative effects of aneconomic activity that are not embodied in the priceof products and consequently cannot be optimally man-aged through pure market arbitration Effects can bediverse across bioenergy LUC scenarios because theydepend on feedstock species and regional managementrequirements Externalities of bioenergy production


local global


co-products





























































cropland






































nitrogen) (2)



































6 CONCLUSIONS



REFERENCES












































































local global


co-products





























































cropland






































nitrogen) (2)



































6 CONCLUSIONS



REFERENCES















































































































cropland






































nitrogen) (2)



































6 CONCLUSIONS



REFERENCES


































































































nitrogen) (2)



































6 CONCLUSIONS



REFERENCES



























































































6 CONCLUSIONS



REFERENCES











































































how can land-use modelling tools inform bioenergy policies?

Documents