Life cycle assessment of carbon and energy balances in jatropha production systems of burkina faso

Life cycle assessment of carbon and energy balances in Jatropha production systems of Burkina Faso Inaugural-Dissertation zur Erlangung des Grades Doktor der Agrarwissenschaften Dr.. cu

Life cycle assessment of carbon and energy balances in

Jatropha production systems of Burkina Faso

Inaugural-Dissertation

zur Erlangung des Grades Doktor der Agrarwissenschaften

(Dr agr.)

der Hohen Landwirtschaftlichen Fakultät

der Rheinischen Friedrich-Wilhelms-Universität

zu Bonn

von SOPHIA EMILIA BAUMERT

aus BERLIN

1 Referent: Prof Dr Asia Khamzina

2 Referent: Prof Dr P L G Vlek

Tag der Promotion: 10.01.2014

Erscheinungsjahr: 2014

Diese Dissertation ist auf dem Hochschulschriftenserver der ULB Bonn http://hss.ulb.uni-bonn.de/diss_online elektronisch publiziert

ModernbioenergyoffersseveraladvantagestoBurkinaFaso,acountrythatisheavilydependent on imported fossil fuel and greatly relying on traditional biomass use. In

this context, Jatropha curcas has been recently introduced as a lowͲmaintenance

energycropwiththepotentialtoincreaseenergysecuritywhilecontributingtoland

rehabilitationandclimatechangemitigation.ThisstudyidentifiedJ.curcascultivation

systemspracticedinBurkinaFasoandanalyzedtheirbiomassdynamicsandcarbon(C)accrualovertimeaswellassoilͲCstocks.Thesedata,togetherwiththeinformationon

J. curcas seed transformation processes, were integrated in a life cycle assessment

(LCA) of the greenhouse gas (GHG) emission and energyͲsaving potential of thecompletebiofuelproductionpathways.

The studied J. curcas systems include interplanting with annual crops,

intenselymanagedplantations,afforestationofmarginalland,plantingsalongcontourstone walls, and traditional living fences. Destructive aboveͲ and belowͲgroundbiomassdeterminationenabledtheidentificationofgrowthstagesanddevelopmentofallometricequationsrelatingtotalshootandrootbiomasswiththestemdiameterthatshowedverygoodfits(R²>0.9).EmpiricalgrowthmodelsrelatedwoodybiomassandtreeagebyathreeͲparametricnonͲlinearlogisticfunction.Accordingtothemodel

results,thebiomassproductionofJ.curcasplantspeakedbetweenthe10thand15thyear after planting, with intercropping and intensely managed systems showing thehighest stock (21 t haͲ1). Afforestation systems on marginal land had the lowestbiomassstocks(<0.1thaͲ1),andcouldnotbemodeledduetodrasticmortalityatanearlyageintheabsenceofmaintenance.Soilanalysisdidnotrevealacleartrendofsoilorganiccarbon(SOC)dynamicsovertimewhencomparingthesoilcarbonstatusin

4ͲyearͲold J. curcas sites with that in the reference cropland. Only J. curcas living

fences exhibited significantly higher SOC stocks in the top 20 cm soil based on a

chronosequencestudycovering20yearsofJ.curcascultivation.

All J. curcas production pathways showed GHG emission reductions and

energy savings of up to 82% and 85%, respectively, as opposed to fossil fuel.Decentralized production of straight vegetable oil and its consumption in stationary

diesel engines showed the best performance. However, J. curcas plantation systems

hadverylowlandͲuseefficiency(6.5Ͳ9.5GJhaͲ1)andthusahighlandͲusereplacement

potential.CarbonͲstockgainswereattainedwhenintroducingJ.curcasoncroplands.

However,thedisplacementofagriculturalactivitiestootherareascanindirectlyresultin C losses. Human energy accounted for 24% of the total energy balance, indicating

highmanuallaborrequirementsinsmallͲscaleJ.curcassystems.Monetaryvaluationof

CoffsetsviacarbontradingschemesshowedreturnsbelowUS$350over20years.

Overall, J. curcas biofuel production can contribute to climate change

mitigation and national energy independency. However, due to low landͲuseefficiency,highlaborrequirementsandtheunsuccessfulcultivationonmarginalland,

J. curcas becomes a direct competitor with food crops and is a not viable option for smallholder farmers. Whereas J. curcas cultivation is yet to be intensified through

improvedplantmaterialandoptimizedagronomicmanagement,thetraditionalhedgesystemsareapreferableoptionforseedproductionastheyofferadditionalbenefitsof

et étudiés quant aux dynamiques de la biomasse et du carbone (C) dans les sols.Combinéesàdesinformationssurlatransformationdesgraines,cesdonnéesontétéintégréesdansuneanalyseducycledevie(ACV)pourcalculerlesémissionsdegazàeffetdeserre(GES)etlepotentield'économied'énergiedelachainedeproductiondebiocarburantsdanssonensemble.

CinqsystèmesdeculturedeJ.curcasontétéidentifiés:l’associationavecdes

culturesannuelles,lesplantationsavecunegestionintensive,lereboisementdesolsmarginaux,leshaiesvivestraditionnellesetleshaieslelongdescodonspierreux.Desmesures directes de la biomasse aérienne et souterraine ont permis d’identifier lesdifférentesphasesdecroissanceetdedévelopperdeséquationsallométriquesreliantla biomasse aérienne et souterraine au diamètre du tronc (R²>0.9). En outre, desmodèlesdecroissanceempiriquesontétédéveloppéspourchaquesystème,prédisantlaproductiondebiomasseaérienneenfonctiondel’âge.Lesrésultatsdecesmodèlesmontrentquelaproductiondebiomasseestmaximaleentrela10èmeetla15èmeannéeaprèslaplantation.Lesplusgrosstocksdebiomasse,jusqu’à21thaͲ1,sontobservésdans les systèmes en association avec des cultures annuelles et dans les plantationsintensives alors que le système de reboisement des sols marginaux présente laproductiondebiomasselaplusfaible(0.1thaͲ1).Acausedutauxdemortalitéélevédesjeunesplants,cesystèmen’apaspuêtremodélisé.

LesanalysesdesolcomparantlessolssousJ.curcasdepuisquatreansavec

lessolssousculturesannuellesn’ontpasmontrédedynamiqueévidenteduCdanslesol.Unechronoséquencede20anspourunehaieviveacependantpermisdemettreenévidenceuneaugmentationsignificativeduCdanslespremiers20cmdusol.

PourtouteslesfilièresdeproductiondeJ.curcas,l’analysedecycledeviea

montrédesréductionsdeGESjusqu’à82%etunetrèshauteefficacitéénergétiqueparrapportauxcarburantsfossiles.Laproductionlocaled’huilevégétaleetsonutilisationdans les moteurs stationnaires affiche la meilleure performance. Néanmoins, les

plantationsdeJ.curcasmontrentuneefficacitétrèsfaibleentermesd'utilisationdes

terres(6.5Ͳ9.5GJhaͲ1),augmentantainsilepotentielpourunchangementd’utilisation

dusol.BienquelesstocksdeCaugmententlorsdel’intégrationduJ.curcasdansles

terresencultures,ledéplacementd’activitésagricolespourraitindirectementrésulter

crédits carbone pour le marché international ne promettait pas de recettessignificatives.

Globalement,ilapuêtredémontréquelaproductiondebiocarburantdeJ. curcas pouvait contribuer à l’atténuation des changements climatiques et à

l’indépendance énergétique. Cependant, l’inefficacité de l'utilisation de terres, lebesoin de main d'œuvre très élevé et l’inaptitude des terres marginales pour la

BodenkohlenstoffͲDynamik untersucht. Zusammen mit Informationen zurWeiterverarbeitung der Samen wurden alle Daten in einem Life Cycle Assessment(LCA) zur Berechnung der Treibhausgasemissionen und des EnergieeinsparungsͲ

potenzialsderJ.curcasBioenergieͲProduktionssystemezusammengeführt.

Insgesamt konnten fünf J. curcas Systeme identifiziert werden: Mischanbau

mit einjährigen Kulturen, intensiv bewirtschaftete Plantagen, Aufforstung vonmarginalen Flächen, traditionelle Lebendhecken und Hecken entlang vonKontursteinmauern. Durch direkte Messungen von oberͲund unterirdischer Biomasse

der J. curcas Bäume konnten unterschiedliche Wachstumsphasen definiert und

allometrische Modelle zur indirekten Biomassebestimmung entwickelt werden. Eszeigtesicheinesehrstarke(R²>0.9)allometrischeBeziehungzwischensowohlHolzͲalsauch Wurzelmasse und Stammdurchmesser. Des Weiteren konnten empirischeWachstumsmodelle zur Vorhersage der Holzbiomasse in Abhängigkeit des Alterserstellt werden. Entsprechend der Modelle erreicht die Biomasseproduktion ihren

Höhepunkt zwischen dem zehnten und fünfzehnten Wachstumsjahr. Jatropha curcas

im Mischanbau und in intensiv bewirtschafteten Plantagen erreichte die höchstenBiomassewerte (21 t haͲ1), während das Aufforstungssystem mit einer Biomasse vonwenigerals0.1thaͲ1diegeringstenWerteaufwies.AufgrundderhohenMortalitätderjungen Bäume auf den marginalen Standorten konnte das Biomassewachstum dieses

Systemsnichtmodelliertwerden.VergleichendeBodenanalysenvonvierJahrealtenJ. curcas Standorten mit Flächen unter einjährigen Kulturen ergaben keine eindeutige

TendenzvonVeränderungendesBodenkohlenstoffs.NurineinerChronosequenzvonBöden unter Lebendhecken über 20 Jahre konnte ein signifikanter Anstieg desKohlenstoffsindenersten20cmdesBodensfestgestelltwerden.

Für alle Produktionswege der J. curcas Bioenergie konnten eine bis zu 82%

hohe Verringerung der Treibhausgasemissionen und bis zu 85% Energieeinsparungenim Vergleich zu fossilen Brennstoffen festgestellt werden. Die dezentrale ProduktionvonPflanzenölunddessenVerbrauchinstationärenDieselmotorenzeigtediebestenErgebnisse. Eine sehr geringe Landnutzungseffizienz (6.5Ͳ9.5 GJ haͲ1) der J. curcas

Plantagensysteme erhöhen jedoch den Druck auf andere Landnutzungsformen. Auch

wenn die Integration von J. curcas in landwirtschaftliche Systeme zu einer größeren

Kohlenstoffspeicherung führt, kann die Verdrängung der Nahrungsmittel von den

konstituiert. Eine monetäre Bewertung der Kohlenstoffeinsparungen durch dessenHandelaufinternationalenMärktenversprachnurgeringfügigeErträge.

Zusammenfassend kann gesagt werden, dass J. curcas Systeme in Burkina

FasosowohlzumKlimaschutzalsauchzurEnergiesicherungbeitragenkönnen.Durchdie sehr geringe Landnutzungseffizienz, den hohen Arbeitsaufwand und die fehlende

Ertragsleistung auf marginalen Standorten wird J. curcas jedoch zu einer direkten

Konkurrenz zu Nahrungsmitteln und stellt keine praktikable Option für Kleinbauern

dar. Solange der Anbau von J. curcas durch verbessertes Pflanzmaterial und optimiertes Management nicht intensiviert werden kann, sollte der Anbau von J. curcasinHeckensystemenvorgezogenwerden.DiesebietenvielfältigeVorteilefürdie

Bauern während die Samenproduktion zur Energieversorgung in ländlichen Gebieten

I came up with a total 14 t CO2 emitted to the atmosphere through mydissertation1.Asyouwillunderstandafterreadingthedissertation,approx.200mofJatrophalivingfenceorhalfahectareJatrophaplantationwouldbeneededtooffsetthisamountofcarbon.Currently,Iamnotinthepositiontoundertaketheplantingsand maintenance, therefore I decided to buy my way out. I donated € 322 from the

Dreyer research budget to atmosfair gGmbH who is investing money in energizing

projectsworldwide.NowIcansaythatthepreparationofmydissertationwasalmostcarbonneutral!

However, the achievements resulting from my dissertation shouldn’t beneutral but hopefully contribute to a sound policy of Jatropha biofuel productionfulfillingmostofthepromisesassociatedwithJatropha.

1 INTRODUCTION 1

1.1 Problemsetting 1

1.2 JatrophacurcasanditsrelevanceforBurkinaFaso 2

1.3 Researchneeds 4

1.4 Researchobjectives 6

1.5 Outlineofthethesis 6

2 STUDYREGION 8

2.1 Climateandvegetation 8

2.2 Soilsandlanduse 11

2.3 Agriculture 11

2.4 Energyusepattern 12

3 JATROPHAINBURKINAFASO 14

3.1 Introduction 14

3.2 Materialsandmethods 16

3.2.1 Samplingdesignanddatacollection 16

3.2.2 Geographicdistributionofthestudysites 19

3.2.3 ShadingeffectofJatrophacurcasplantings 21

3.2.4 Statisticalanalyses 22

3.3 Results 23

3.3.1 StakeholdersinJatrophacurcasactivities 23

3.3.2 Systemclassificationandcharacterization 27

3.3.3 LandallocationtoJatrophacurcascultivation 35

3.1 Discussion 37

3.1.1 ManagementpracticesinJatrophacurcassystems 37

3.1.2 Thelandusedilemma 40

3.2 Conclusionsandrecommendations 42

4 DYNAMICSINABOVEͲANDBELOWͲGROUNDBIOMASS 44

4.1 Introduction 44

4.2 Materialsandmethods 46

4.2.1 Sampledesignandcontrolforconfounders 46

4.2.2 Studysites 47

4.2.3 Measurementsoftreedimensionsanddrymatterproduction 49

4.2.7 Carbonstockestimation 56

4.3 Results 56

4.3.1 MorphologicalandphysiologicalattributesofJatrophacurcastrees 56

4.3.2 Fruitcharacteristicsandseedyield 58

4.3.3 Growthstages 59

4.3.4 Allometricrelationships 60

4.3.5 Empiricalgrowthmodels 66

4.3.6 CarbonstorageinJatrophacurcassystems 70

4.3.7 Modelvalidation 71

4.4 Discussion 72

4.4.1 SeedproductivityofJatrophacurcastrees 72

4.4.2 AllometryofJatrophacurcas 73

4.4.3 Biomassgrowthmodeling 76

4.4.4 CarbonsequestrationpotentialinJatrophacurcassystems 77

4.5 Conclusionsandrecommendations 78

5 DYNAMICSOFSOILORGANICCARBON 80

5.1 Introduction 80

5.2 Materialsandmethods 82

5.2.1 Soilsampling 82

5.2.2 Chronosequencestudy 83

5.2.3 13Cnaturalabundancetechnique 84

5.2.4 Leaffallandleafdecomposition 84

5.2.5 Soilanalyses 85

5.2.6 Soilcarbonbudget 87

5.2.7 Statisticalanalyses 87

5.3 Results 88

5.3.1 Soilproperties 88

5.3.2 Soilorganiccarbondynamics 91

5.3.3 Soilorganiccarbonchangeoversoilchronosequence 95

5.3.4 Changesinɷ13Cvalues 96

5.3.5 Leaflitterfallanddecompositionrates 97

5.3.6 Contributionoforganicmaterialtothesoilcarboncycle 100

5.4 Discussion 101

5.4.1 Soilcarbondynamicsincontourhedges 101

5.4.2 Soilcarbondynamicsinlivingfences 102

5.4.3 Soilcarbondynamicsinplantationsystems 103

5.4.4 Soilcarbondynamicsinafforestationsystems 104

5.4.5 Carboninputandturnover 104

5.5 Conclusionsandrecommendations 107

6 GREENHOUSEGASANDENERGYSAVINGSINJATROPHACURCASBIOFUEL PRODUCTIONSYSTEMS 108

6.1 Introduction 108

6.2 Methodology:Lifecycleassessment 110

6.2.1 Goalandscopedefinition 110

6.2.2 Inventoryanalysis 114

6.2.3 Jatrophacurcascultivation 116

6.2.4 BiomasscarbonstocksandlandͲusechange 118

6.2.5 TransformationphaseofJatrophacurcasseeds 120

6.2.6 Jatrophacurcasoilconsumptionandenergysubstitution 122

6.3 Results 123

6.3.1 Cultivationphase 123

6.3.2 LandͲusechangeandcarbonbalance 126

6.3.3 Fromwelltotank 127

6.3.4 Energyconsumption 130

6.3.5 Carbonoffsets 133

6.4 Discussion 133

6.4.1 Managementascarbonemittingfactor 133

6.4.2 LandͲuseeffects 135

6.4.3 PerformanceofJatrophacurcasbiofuelproductionpathways 137

6.4.4 EnduseofJatrophacurcasfuels 138

6.4.5 PotentialofglobalcarbontradingforproͲpoormitigation 139

6.5 Conclusionsandrecommendations 140

7 GENERALOVERVIEWANDOUTLOOK 142

7.1 JatrophacurcasinBurkinaFaso 142

7.1.1 Carbonandenergybalances 145

7.1.2 Potentialofcarbontrading 146

7.2 Methodologicalissues 147

7.2.1 Estimationofbiomasscarbon 147

7.2.2 Changesinsoilcarbon 149

7.3 Overallconclusions 149

8 REFERENCES 151

9 APPENDICES 165 ACKNOWLEDGEMENTS

OM  Organicmaterial

PD  Plantdensity

RED  RenewableenergydirectiveRD  Relativedifference





1 INTRODUCTION

1.1 Problemsetting

SubͲSaharanAfricaishometotheworld’spoorestpopulationwith90%livinginruralareas and depending on subsistence agriculture for their livelihoods (Bationo andBuerkert2001).Thehighlevelsofpovertyare,amongstothers,reflectedintheenergyconsumptionpattern,withaverylowshareofmodernenergyandahighrelianceontraditional biomass energy (Karekezi 2002) accounting for more than 80% of theprimary energy supply (IEA 2006). With an annual population growth rate of 2.5%(World Bank 2012) the need for energy is constantly increasing, leading to highlyunsustainable biomass consumption (Bugaje 2006; Tatsidjodoung et al. 2012). Trees,an essential element for the stability of ecosystems, are removed without providingtheopportunityforreͲgrowth(RutzandJanssen2012),andtheenergeticuseofcropresidueslimitsthereͲcyclingofsoilnutrients,whichleadstodecliningsoilfertility(Lal2006).ParticularlyinthelowͲinputagriculturalsystemswhereproductivityͲenhancingtechnologiesarelargelyoutofreach,soilqualityiskeytoagriculturalproduction(Vlek2005). Declining soil fertility and land degradation are among the major humanͲinduced problems currently facing agricultural production throughout SubͲSaharanAfrica(KatyalandVlek2000,Zida2011).

Growing public awareness of the energy dilemma prevailing in SubͲSaharanAfricahasdirectedinternationalattentionontheuseofmodernbioenergy2(Ndongetal.2009).ParticularlyinAfrica,whereoneͲthirdofthetotallandispotentiallyavailableforbiofuelproduction(Caietal.2010)andalargeshareofthepopulationisinvolvedinagriculture,biofuelproductioncanoffermanybenefitstotheruralpoor(Blinetal.2013).BiofuelscouldprovideresourceͲpoorcountrieswithameanstoinvestintheirownruralareasinsteadofexportingtheircapitaltopurchasefossilfuel.Moreover,thepositivecorrelationbetweeneconomicdevelopmentandaccesstoenergyresourcesislong recognized (Karekezi 2002; Bugaje 2006). Internationally, energy crops cancontributetoclimatechangemitigationthroughcarbonsequestrationinbiomassand



2  Modern bioenergy is defined as bioenergy relying on sustainably used biomass as opposed to

soil and through substitution for fossil fuels or unsustainably harvested fuel wood(Bass et al. 2000). The carbon offsets can then be monetarily valuated via carbontrading mechanisms (e.g., Clean Development Mechanism (CDM), Voluntary CarbonMarkets),whichisoftencitedasanadditionalincomeopportunityforAfricanfarmers(Bryanetal.2008).However,alsoinSubͲSaharanAfrica,therearerisksassociatedwithbioenergy production such as negative impacts on ecosystems (Ndong et al. 2009),competitionwithfoodproduction,andincreasedfoodprices(vonBraun2008).

In this context, the tree species Jatropha curcas has become popular as an

energycropbasedonearlyclaimsofhighproductivityunderlowwater,nutrientandmanagementrequirements.Accordingtotheclaims,thecropcanthriveonmarginallandinsemiͲaridregions,contributestolandreclamationanddoesnotcompetewithfoodcropsforscarceresources(e.g.,Heller1996;Francisetal.2005;Jongschaapetal.2007;Henning2009;Achtenetal.2010b;Contranetal.2013).

1.2 JatrophacurcasanditsrelevanceforBurkinaFaso

Jatropha curcas Linnaeus has its origin in Central America and Mexico and was

probablyimportedbythePortugueseseafarerstotheCapeVerdeIslandsandGuineaBissau in the 16th century and then distributed over wider parts of Africa and Asia

(Heller1996;DomergueandPirot2008;Henning2009).Jatrophacurcasbelongingto thegenusEuphorbiaceaeisasmalltreethatproducesfruitscontainingseedswithan

oilfractionof30to35%(Jongschaapetal.2007;Achtenetal.2008).Theoilistoxicand not edible for humans and animals, but it has a very good burning quality(Jongschaapetal.2007;Blinetal.2013).Thetreeishighlyadaptabletoavarietyof

growing conditions (the J. curcas belt is roughly situated between 30°N and 35°S

(Jongschaapetal.2007))andisexpectedtoyieldover50yearswithagestationperiod

of3to4years(Jongschaapetal.2007;vanEijcketal.2010).Traditionally,J.curcasis

planted as living fences protecting fields from animals and contributing to erosioncontrol. The oil is originally used for the production of soap and for medicinalpurposes. With the rising interest in biofuel, the use of the oily seeds as an energyfeedstock has internationally come into focus. The oil can be mechanically extracted

with a simple technology and used directly as straight vegetable oil (SVO) in dieselengines(Blinetal.2013)suchasinnationalpowerstationsandcanreplaceimportedfossilfuel(NonyarmaandLaude2010;Tatsidjodoungetal.2012).Moreover,theuseof SVO offers the possibility of decentralized production and consumption (e.g., foragricultural activities, power generation, rural industry, and cooking) avoiding longtransportationdistancesandcomplicatedtransformationprocessesasisthecasewithbiodiesel (FACT Foundation 2009; Blin et al. 2013). These decentralized schemes areparticularlypopularinWestAfricancountrieswithsevereenergypovertyinruralareas(Blinetal.2013).

Owing to its great potential, J. curcas became idealized as a solution for

energyͲpoor countries, and triggered largeͲscale investments (Achten et al. 2010b)with cultivation hotspots in India, Zambia, Madagascar, Tanzania, Brazil, Mexico and

Ghana (Gao et al. 2011). However, most J. curcas projects were not scientifically

grounded, but rather driven by overͲoptimistic claims leading to manifold project

failures(vanEijcketal.2010).Bynow,manylessonshavebeenlearntshowingthatthe

full potential of this tree species is not easily exploitable and particularly not simultaneously applicable (Coltran et al. 2013). Jatropha curcas is still an

undomesticatedplantwithagreatvariabilityinproductivity(e.g.,Achtenetal.2010c;Liyama et al. 2012; Contran et al. 2013). Under the current knowledge status, adefinition of siteͲspecific agronomic management regimes for optimal productionlevels is impossible (Singh et al. 2013) leading to subͲoptimal management practicesandlowyields(Liyamaetal.2012,Singhetal.2013).Moreover,ithasbeenrealizedthat trees grown on marginal soils with marginal inputs will produce marginal yields(Lal 2006; Elbehri et al. 2013), thus trading off marginal land restoration and biofuel

Since 2007, J. curcas has been one of the most strongly promoted biofuel

cropsinBurkinaFaso(Tatsidjodoungetal.2012).Studiesassessingthelandavailabilityfor biofuel production in semiͲarid regions excluding agricultural land and land withhigh biodiversity (Cai et al. 2011; Wicke et al. 2011; Dauber et al. 2012) showed

substantiallandavailabilityinBurkinaFaso(Wickeetal.2011).ThecontributionofJ. curcas cultivation to the national energy supply and to the amelioration of the soil resources could thus be significant. Understanding the potential and challenges of J. curcas, the Burkinabe government began to design a national biofuel policy in 2009,

prioritizing food security, environmental and biodiversity protection, and inclusion ofsmallͲscale farmers in biofuel activities (MMCE 2009; Nonyarma and Laude 2010;Tatsidjodoungetal.2012).InordertoavoidenvironmentallyfatallandͲusechangeand

competition between food and energy, J. curcas should be preferably grown in

combination with annual crops or on soils low in productivity (MMCE 2009). TheallocationoflandtolargeͲscaleplantationswasregardedwithcaution(MMCE2009).



1.3 Researchneeds

Overall, the productive capacity of J. curcas has been rarely studied in Burkina Faso

(Sop et al. 2012), and the effects of different production models on people andenvironment have not yet been evaluated (Tatsidjodoung et al. 2012). It is generallyagreed that sustainable bioenergy systems must provide net energy gains, haveenvironmental and local socioͲeconomic benefits, and produce bioenenergy in largequantities without impacting food supplies (Fritsche et al. 2005; Hill et al. 2006;

Mangoyana2008,Elbehrietal.2013).Further,theassociationofJ.curcaswithcarbonͲ

neutral biofuel and climate change mitigation remains to be justified for theproduction systems inBurkina Faso in view of agroͲinputs in energy crop productionand impacts bound to landͲcover change from ecosystems high in carbon stock toenergycrops(Fargioneetal.2008).CarbonͲoffsetcalculationsalsoprovideevidenceof

Life cycle assessment (LCA) is a common tool to evaluate environmentalsustainability of biofuel production systems in terms of energy efficiency and carbon

neutrality(Gnansounouetal.2009).Todate,noLCAhasbeenconductedforJ.curcas biofuel production in Burkina Faso, and J. curcas initiatives are proceeding without

knowledge of caseͲspecific environmental consequences. Ndong et al. (2009)presented a study for West Africa, but they did not include carbon stock changes inbiomassandsoilresultingfromlandconversion,andassumedmorethan50%higher

seed yields than actually observed in Burkina Faso. OverͲoptimistic J. curcas yield

estimationswerenamedbyGasparatosetal.(2012)asamajorerrorsourceinLCAs.Achtenetal.(2012)criticizedtheabsenceofcarbonstockchangesinbiomassandsoilinmostLCAcalculations,althoughbioenergyͲinducedlandͲuseandlandͲcoverchangesareknowntohavehighimpactsonenvironmentalsustainability(Fritscheetal.2005).

ForJ.curcassystems,thismeansthatabetterestimationofcarbonstocksisneededas

already called for by Reinhardt et al. (2007). Moreover, investigations of the soil

carbon dynamics under J. curcas systems are important for the assessment of their

claimedlandrehabilitationpotential.

LongͲterm observations of temporal biomass dynamics in J. curcas systems are out of reach, as most J. curcas systems are in their infancy. However, the

developmentofempiricalgrowthmodelsbyfittingchronosequencesoftreesdifferinginagecouldprovidebiomasspredictionsovertimewithinaveryshortperiodoftime(Walker et al. 2010). The establishment of allometric relationships between biomass

and stem diameter in J. curcas could further facilitate nonͲdestructive tree biomass

estimation.Thechronosequenceapproachisalsowidelyappliedforthedetectionofdynamics in soil organic carbon (Walkeret al. 2010). Some studies have investigated

allometric relationships and biomass dynamics in J. curcas (Ghezehei et al. 2009;

Achtenetal.2010a;Beheraetal.2010;Rajaonaetal.2011;Hellingsetal.2012),albeitbasedonamodestsamplesize.NosuchresearchhasbeenconductedinWestAfrica,

and only few studies investigated changes in soil after afforestation with J. curcas

(Ogunwoleetal.2008;Soulama2008).

1.4 Researchobjectives

ConsideringthelackofscientificknowledgeandtheexpandingcultivationofJ.curcas, the aim of this dissertation is to assess the environmental sustainability of J. curcas

biofuelproductionsystemsinBurkinaFaso.Tothisend,thecarbonͲandenergyͲsaving

potential of existing J. curcas production systems is analyzed under consideration of

carbon sequestration in biomass and soil. The findings are expected to support

classificationcriteriaforfivemanagementsystemsaredeveloped.Thefindingsoftheinventory serve as basis for all further investigations. Chapter 4 presents thequantification of the carbon sequestration potential in standing biomass of the

identified J. curcas systems. Allometric equations for nonͲdestructive biomass stock estimationsandempiricalgrowthmodelsdemonstratingbiomassgrowthofJ.curcas

stands over the years are developed and tested. The aspect of soil carbon

sequestration under J. curcas systems is elaborated in Chapter 5. Data from a soil

survey concentrating on soil organic carbon stocks and their changesunder J. curcas

systems relative to reference sites are presented. Chapter 6 integrates the results of

Chapter 3, 4 and 5 in a life cycle assessment and presents different J. curcas

productionͲtransformationͲconsumption pathways in regard to their potential forcarbonemissionreductionandenergysavings.Finally,inChapter7themainfindingsofthestudyaresummarizedanddiscussed,andrecommendationsforexploitationof

thepotentialofJ.curcasandsuggestionsforfurtherresearchareformulated.

2 STUDYREGION

Burkina Faso ("country of the honorable people") is a landlocked country situated inthe heart of West Africa. It covers an area of 274,000 km² located between 09°20’ Ͳ15°03’ N and 05°03 W Ͳ 02°20’ E and bordered by Niger, Mali, Ghana, CôteͲd’Ivoire,Benin and Togo (CIA 2012). The country is divided into 13 regions and 45 provinceswithOuagadougouasthecapitalcity.Thepopulationcounts17.813millionpeople(65people kmͲ²) with a population growth rate of3% (CIA 2012). More than 80% of thepopulation resides in rural areas and is engaged in smallͲscale lowͲinput agriculture(CIA 2012). Burkina Faso’s economy heavily relies on cotton and gold exports forrevenues, as it has only few natural resources and a weak industrial sector. Overall,high population density, lack of natural resources, poor industrial development andlow agricultural productivity are the main reasons behind the persisting poverty inBurkina Faso where 46% of the population live below the poverty line (World Bank2013b).



2.1 Climateandvegetation

BurkinaFasoisdividedintothreeagroͲecologicalzones(AEZ),i.e.,theSudanianinthesouth(9°3’Ͳ11°3’N),theSudanoͲSahelianinthecentralregion(11°3’Ͳ13°3’N)andtheSahelian in the north (13°5’Ͳ15°5’N). It has a tropical climate with two alternatingseasons: a long dry spell from November to May with the continental trade wind(Harmattan) coming from northeast and a short rainy season from June to Octoberwith moist air coming from oceanic high pressure (Figure 2.1) (Thiombiano andKampmann2010).

Located in the transition zone between the Sahara Desert to the north andcoastalrainforeststothesouth,BurkinaFasoispronetoextremeweathereventssuchasrecurrentdroughts,floodsandwindstorms(WorldBank2013a).InterͲannualandinterͲdecadalclimatevariabilitywilllikelyincrease;howeverahighlevelofuncertaintyisassociatedwithclimatechangeprojectionsforWestAfrica(IPCC2001;WorldBank2013a).

0 100

0 50 100 150 200 250

0 50 100 150 200 250

systems.IntheSudanoͲSahelianAEZ(NorthSudan)annualrainfallrangesfrom700to900mmfromnorthtosouth.Vegetationchangesfromgrassyandshrubbysteppesinthe north to shrubby and woody savannas in the southern parts with parkland tree

speciessuchasVittelariaparadoxa,Feidherbiaalbida,Adansoniadigitata,Tamarindus indica, Lannea microcarpa, Azadichta indica,and Bombax costatum (Thiombiano and

Kampmann 2010). The Sudanien zone (South Sudan) is characterized by mosaics ofcropland, fallow areas in various stages of regeneration, and typical agroforestry

parklandwiththemaintreespeciesFaidherbiaalbidia,Vittelariaparadoxa,andParkia biglobosa (Boffa 1999; Thiombiano and Kampmann 2010). Pressure on the natural

vegetation is particularly high due to expanding cultivation of cotton, and highmigration from the northern parts of Burkina Faso (Gray 1999) coupled withunsustainable firewood collection, annual bushfires, intensive pasturing andsettlements(CommuneRuraledeBoni2009).Charcoalexploitationnotonlyforlocalconsumption but also for supplies to Ouagadougou is additionally triggeringdeforestation(1.45%annually)(Ouedraogo2007;Ouedraogoetal.2010).

2.2 Soilsandlanduse

MostofBurkinaFasoiscoveredbyferriclixisols(WRB1998)(leachedferruginoussoils (CPCS 1967)) and leptosols or lithisols (poorly evolved soils of erosion). Cambisols, vertisols, glysols and ferralsols are of limited extent, and are found localized

throughout the country (Thiombiano and Kampmann 2010). Generally, the soils areinherentlylowinsoilfertility(organiccarbon<1%),havelowwaterholdingcapacity,andatendencytodevelopsoilsurfacecrusting(Zougmoré2003).Bushfiresforlandclearing make the soils susceptible to wind erosion during the dry season, and highrainfall intensities trigger water erosion at the onset of the rainy season (Zougmoré2003).AccordingtoFAO(2009),44%ofthetotallandareainBurkinaFasoisalreadyaffectedbyseverelanddegradation.

ThelandresourcesinBurkinaFasoaredividedasfollows(WorldBank2008):45% agricultural land (12 Mio ha) with 50% under cultivation (6.3 Mio ha) and 50%underpermanentcropsandpastures(includingabandonedcroplandandlandnotyetcultivated),24%forestarea,and10%undersettlement(other:21%).Highpopulationgrowthand acceleratinglanddegradationarekeydriversbehindcroplandexpansionwithanannualrateof0.2%(0.96%insouthernBurkinaFaso)attheexpenseofgrazingarea,forestsandwoodland(FAO2001inOuedraogoetal.2010).



2.3 Agriculture

The agricultural sector dominated by small family farms on rainfed land andcharacterized by low labor and input productivity (Breman et al. 2001 in Zougmoré2003) provides income to more than 80% of the population (MED 2003). Millet

(Pennisetumglaucum),redandwhitesorghum(Sorghumbicolour),maize(Zeamays), andcowpeas(Vignaunguiculata)arethemainsubsistencecropsandcover80%ofthe cultivated area. Cotton (Gossypium herbarceum), groundnuts (Arachis hypogaea L.), and sesame (Sesamum indicum) are the principal cash crops (Zougmoré 2003). ).

Extensive livestock production also plays an important role (cattle, small ruminantsandpoultry(INERA2006)).

Crop production particularly suffers from poor native soil quality, surfacecrusting, low waterͲholding capacities, highly irregular rainfall patterns and high soilandairtemperatures(BationoandBuerkert2001).Climatechangeprojectionsshowafurther increase in climate variability and adverse effects on crop yields (IPCC 2001;World Bank 2013a). SmallͲscale agricultural systems are most vulnerable to thesechangesinclimateduetotheirpooradaptivecapacity(IPCC2001;Mangoyana2009).All in all, low agricultural productivity continues to impede poverty reduction.Therefore, major governmental efforts target agricultural intensification throughmechanization, financial lending, water storage, crop diversification, and soilrestoration(Hanffetal.2011;WorldBank2013a).



2.4 Energyusepattern

Burkina Faso is facing a major energy crisis. More than 80% of the country’s energyconsumption is covered by traditionally used biomass such as fuel wood, dung andcropresidues(Hanffetal.2011).Withitsgrowingpopulationanditsincreasingneedfor energy, the consumption of biomass exceeds the capacity of biomass reͲgrowth(Bugaje 2006; Tatsidjodoung et al. 2012). Unsustainable use of biomass leads to soilerosion and land degradation, which are becoming the most serious environmentalissues linked to energy consumption (Bugaje 2006; Toonen 2009; Sawe 2012).Moreover, indoor air pollution from open cooking fires is estimated to cause 16,500deaths per year (WHO 2004). The remaining national energy need is covered byimported hydrocarbons used mainly for transportation and electricity production(Tatsidjodoung et al. 2012). As net importer of fossil oil, amounting to 50% of thenationaltradebalance,BurkinaFasoisheavilyaffectedbyrisingoilprices(Hanffetal.2011;Tatsidjodoungetal.2012).

Overall,BurkinaFasohasaverylowlevelofenergyconsumption(234kgoeper inhabitant compared with 1145 kgoe per inhabitant worldwide), and very pooraccesstoelectricity(<1%inruraland<15%inurbanareas)(Blinetal.2008;Hanffetal.2011).Ithaslongbeenrecognizedthatenergypovertyisdirectlylinkedtoeconomicpoverty (Karekezi 2002; Bugaje 2006). Therefore, the country’s renewable energy

sourcesurgentlyhavetobeharnessed(e.g.,solarenergy,biogas,biofuel)inordertosupplythegrowingdemandinenergytosupportthenation’sdevelopment,increasethe independency from imported fossil fuel, and reduce environmental degradationandhealthimpactsassociatedwiththetraditionalbiomassuse(Karekezi2002;Bugaje2006;Toonen2009,Hanffetal.2011).



lacking,seedharvestanddehuskingareverylaborintensive,mostJ.curcasplantation projects are economically not viable, and landavailabilityand suitability for J. curcas

production are not yet defined (e.g., van Eijck et al. 2010; Contran et al. 2013).

very prudent with the allocation of areas to the cultivation of J. curcas in largeͲscale

monoculture plantations as it fears to compromise food security. According to the

authorities, J. curcas should be predominantly cultivated in combination with annual

cropsandondegradedsoilsfortheirreclamation.Amaximumof500,000halandfor

bioenergyproductionistargetedsofar(MMCE2009).AccordingtoWickeetal.(2011)an estimated 11% (1.6 million ha) of the arid and semiͲarid areas in Burkina Faso,excluding agricultural land and land with high biodiversity, would be potentiallyavailableforenergycropproductionwithoutnegativelyaffectingfoodproduction.The

suitabilityofsuchlandforJ.curcasproductionforsocial,environmentalandeconomic

reasons, however, remains to be investigated (Dauber et al. 2012). Life cycleassessment(LCA)isacommontooltoevaluateenvironmentalsustainabilityofbiofuelproductionsystems(Gnansounouetal.2009),butneedslocationͲspecificdataonthe

management of J. curcas cultivation systems in order to correctly show the

environmentalconsequences.

Currently, several stakeholders are involved in J. curcas activities. NonͲ

governmentalorganizations(NGOs)mainlyworkontheestablishmentofvaluechainsfor the local energy supply, whereas private investors aim at largeͲscale biofuel

production.Morethan80,000haarecoveredwithJ.curcasinformofnewlyplanted

monoͲ and intercropped plantations and traditional hedges (Nonyarma and Laude2010; Ouedraogo 2012), and two biodiesel factories are already in place. Yet mostprojectsdonotfullyoperatealongtheentirevaluechain(Blinetal.2008)duetolowseedyields,highseedprices,hightransportationcosts,inferiorqualityofseeds(5kgseedsneededforprocessing1Ioil),immaturemarketsandlackingregulatorypolicies

allocation. Conclusions about J. curcas systems and their productivity can hardly be drawn.Also,researchonJ.curcascultivationandpropagationinBurkinaFasoremains scarce(Sopetal.2012).ThesoundinventoryanddocumentationofJ.curcasactivities

in Burkina Faso is, therefore, the main objective of this exploratory study. Specific

objectivesareto(i)identifystakeholdersinvolvedinJ.curcasproductionandtoassess their operational practice, (ii) classify and characterize existing J. curcas cultivation

systems in Burkina Faso, (iii) evaluate the effect on existing landͲuse patterns of

introducing J. curcas, and (iv) based on the inventory results, to identify a sampling

procedures. The J. curcas farmers in the project areas were visited, guided by an instructed person under the permission of the respective village leader. Jatropha curcassiteswerethenselectedfollowingtheideaoftheoreticsampling.Accordingto

this sampling approach, research sites are selected based on the new insights theymayprovideinregardtotheoverallobjective(Neuman2006).Inthisway,itcouldbe

guaranteed that the variability of J. curcas activities in terms of crop management,

organizational practices, and environmental settings is covered. In case theoretic

samplingwasnotsuitable,J.curcassiteswereselectedrandomly.Intheregionswhere traditionalJ.curcashedgesweregrown,11focusgroupdiscussions(FGD)wereheld,

with the goal to identify appropriate research sites. Altogether, inͲdepth

questionnaireswerecarriedoutwith111J.curcasfarmersattheselectedsites(Table

3.1).ExpertinterviewsweregenerallyconductedinFrench,whilequestionnaireswithfarmers were translated by an assistant into the respective local language. Thequestionnaire consisted of factual questions for quantitative information, and of

At 81 sites, measurements of selected trees (tree height, stem diameter, number offruits)wereundertaken(seesection4.2.3;Table3.1).Allsiteswerecharacterizedandclassified according to planting scheme, plantation management and cultivationpurposes.ThefieldworkwascarriedoutfromJuly2010toMarch2011.

In parallel, investigations were undertaken in regions where J. curcas

cultivationforbiofuelproductionhadnotyetexpandedbutwilllikelydosointhenearfuture(Table3.2).TheseinvestigationswereparticularlyimportanttoassesslandͲuse

managementsystemsbeforeintegratingJ.curcasandpossiblelandallocationeffects ofexpandingJ.curcascultivation.InthesouthͲwesternpartofBurkinaFaso,J.curcas

cultivationislikelytocontinuouslyexpandduetotheregion’sfavorableclimaticandpedologicconditions.Inthecentralregion,theclosevicinitytoOuagadougouislikely

to trigger the production of J. curcasͲbased biofuel as the fuel demand for

transportationisconstantlyrising.Inthenorthernpart,soildegradationisanongoing

process,makingsoilconservationmeasuressuchasstonewallsandJ.curcashedges

attractive.

Intheseregions,110householdsweresurveyedand7FGDswereconducted(Table3.2),coveringthecategorieslanduseandlandownership,cropproductionandmaintenance,croprotation,laborneedandavailability,andfarmers’perceptiononor

experience with J. curcas. A meeting with the chief of the respective village was

arrangedinadvanceinordertogetpermissionforthesurveyandtogetfamiliarwithbasic characteristics of the village such as household number. All households werethen randomly selected. A high variation in terms of cropping patterns within onevillagewasnotexpected,thussmallsamplesizesweredeemedsufficient.Inthelasttwovillagesnohouseholdswerevisited(Table3.2)asmostofthequestionscouldbeansweredduringtheFGD.Thequestionnairewaspretestedwitharandomsampleof20 farmers in order to ensure that the questionnaire was capable of collecting allinformation needed in an unambiguous and easily understandable way. The surveywas conducted with the help of enumerators, who were instructed beforehand andsupervised along the way. All questions were translated into the local language

(Dagara and Morré). This part of the field work was carried out from February toMarch2011(Table3.2).

communicationwithpersonsinvolvedinJ.curcasactivities,andfromsecondarydata

suchasfromstudiesinBurkinaFasoandnationalstatistics.



3.2.2 Geographicdistributionofthestudysites

Investigations on J. curcas production were done throughout Burkina Faso in seven

different regions spread over three AEZ (10°05’ Ͳ 13°18’ N and 04°58’ Ͳ 00°26’ W)(Figure3.1).

In the region CentreͲNord, research was conducted in the province Bam

aroundthecityKongoussi,whereJ.curcasisplantedalongcontourstonewalls(Figure

3.1). The construction of stone walls is a common soil conservation technique tocontrolwatererosion,andwasinitiatedinthe1990s,particularlyinthenorthernpartofBurkinaFaso(Zougmoréetal.2002). InBam,1500km²arealreadyprovidedwiththese stone walls contributing to reͲgreening and restoration of large areas of land

(Landolt 2010). Bam has the largest natural lake, an important source for irrigationagriculture,yet80Ͳ90%ofthecultivatedareaisstillcoveredbyrainͲfedcropssuchasmillet and sorghum. Overall, the production of cereals is not sufficient for the localdemand(MAHRH2010).



Figure3.1 SamplesitesanddistributionofJatrophacurcaslandͲusesystemsin

  BurkinaFaso.N:Numberofsitesvisited.

inthemunicipalityofMogtedo,asnumerousmatureJ.curcashedgesexistinthisarea.

IntheregionsCentreͲSued(provinceZoundwéogo,municipalityofManga)andCentreͲ

Est(provinceBoulgou,municipalityofBagré),theresearchalsofocusedonmatureJ. curcas hedges (Figure 3.1). The typical agrosilvopastoral landͲuse system integrates

trees, crops and livestock in the same land management unit (Ayuk 1997). As aconsequence, living fences are a common technique for controlling livestockmovementtoprotectfields(Ayuk1997).Theemergenceoflivehedgesonthecentralplateau of Burkina Faso dates back to the early 1980s, and they serve not only asprotectionagainstbrowsinganimalsbutalsoaswindbreaksanderosioncontrol(Ayuk1997). Altogether, the central region is one of the most highly populated in BurkinaFaso(INERA2006);theresourcesoilisthusunderenormouspressureandexhibitsthehighestdegradationratesinthecountry(Zougmoré2003).

IntheprovinceofIoba(regionSudͲOuest),researchwasconductedinDano,

whereJ.curcasisgrownintercroppedwithannualcropsonsmallholders’land.InTuy

(regionHautsͲBassins),theheartofthecottongrowingregion(Gray1999),largerscale

J.curcasmonocultureplantations,afforestedmarginalland,andintercroppedJ.curcas

can be found around the village Boni. In Orodara and Banfora (region of Cascades,

province of Comoé), investigations were conducted in smallͲscale J. curcas

intercroppingsystems(Figure3.1).AllthreestudysiteslieintheSudanianzonewithfavorablerainfallpatternsandrelativelyfertilesoils(Gray1999).



whereCdisthreesingletreecanopydiameter(m),measuredwithameasuringtapefromonesideofthecanopytotheother(seesection4.2.3formethodology).



Thespaceoccupationperhectare(CAtm²haͲ1)overtimet(years)wasfittedbyathreeͲparametricexponentialmodel(STATA12.0)   



CA tൌ„ͳ൅ሺ„ʹή„͵ሻ௧     (3.2)



relatingtheshadedareaperhectareCAtwiththerespectiveageoftheplantation.CAtwascalculatedfordifferenttreespacingtypesbymultiplyingtheaveragetreecanopyexpansion with the respective plant density. It was assumed that the tree canopygrowthisnotaffectedbytreedensityaslongastotalcanopyclosureperhectareisnotreached. In the case of living fences, a closed hedge surrounding a oneͲhectare field(400 m) was assumed and multiplied by the mean canopy diameter of 3 m. Modelswere developed for the spacing types 1 m x 4 m (2500 trees haͲ1), 2 m x 4 m (1250treeshaͲ1),4mx4m(625treeshaͲ1),4mx6m(417treeshaͲ1),andclosedhedges(undertheassumptionofcompletetreesurvivalandnotreepruning).b1,b2,b3weretheparameterstobeestimated.



3.2.4 Statisticalanalyses

The data from the semiͲstructured interviews were coded and analyzed by STATA(12.0). Descriptive statistics (means, standard error and frequencies) were used foranalyzing parameters. Nonlinear regressions were run for modeling the tree shadingeffects.



3.3 Results

3.3.1 StakeholdersinJatrophacurcasactivities

Old J. curcas trees can be found throughout Burkina Faso, isolated on farmers’

property for land demarcation or arranged in living fences protecting fields from

roaminganimals.MostofthefarmerswithJ.curcasontheirpropertywereawareof

thetraditionalutilizationoftheoilandtheplantsapforsoapproductionaswellasformedical purposes such as against skin lesions, toothache and bacterial infections. In

thelanguageoftheMossi,J.curcasiscalledWabenbanguemeenmeaning‘Ifyoueat me,yougettoknowme’,whichreflectsthepoisonousnatureofthistree.

The promotion of J. curcas as feedstock for biodiesel only started in 2007.

Campaignswereundertakenbyprivateenterprises,politiciansandNGOstosensitize

farmers regarding the potential of J. curcas and to convince them to integrate this

plant into their agricultural activities (Table 3.3). It was also attempted to organize

seed collection from mature J. curcas living fences, as this could substantially

contribute to seed supply for biofuel production and economic revenues to thefarmers. However, despite radio, newspapers and private extension services biofuelproduction still remains a niche in Burkina Faso; A Burkinabe scientist described thesituationasfollows:“Jatrophaisineverymouth–manyJatrophatreescanbefound–someseedsareharvested–littleoilispressed–nooilisonthemarket”(Anonymous2011;personalcommunication).

Based on stakeholder analysis in Burkina Faso, two organizational modelscouldbedistinguished:(1)PrivateinvestorsaimingatindustrializedoilproductionandtransformationtobiodieselforuseatthenationallevelmanagelargeͲscaleplantationsonlandtheygotfromthegovernment.Thegovernment,however,remainsrestrictiveinallocatinglargeareastobiofuelproduction,asitfearsbothathreattofoodsecurityandcompetitionforlandresources(Blinetal.2011).Therefore,enterprisesstartedto

contractsmallholderfarmersforthecultivationofJ.curcasinintercroppingandliving

fence systems on their own land. Approximately 80,000 farmers are participating inthiskindofoutgrowerscheme(Table3.3).Farmersaregivenseedlingsfreeofcharge,training on plantation establishment, and given an oral guarantee of seed purchase.Twofactoriesforoilextractionandetherificationtobiodieselareinplace.Atthetime

of the field research (2011), there were the first seed harvests; however, the seedsupplyisnotyetsufficienttotapthefullcapacityofthebiodieselplants.

plant oil cooking stoves and oil lamps. The planting of J. curcas as field borders is

increasinglyadvertizedinordertominimizethecompetitionwithfoodcrops.

In Table 3.3 stakeholders involved in J. curcas production are listed as of January2013.NewJ.curcasactivitiesareconstantlybeinginitiated,butthesecannot

beeasilyidentifiedinBurkinaFaso.

W.Kwendé2010 interview;A.K.Sanou 2011pc;

http://www.agritechgr oup.com/2013

3.3.2 Systemclassificationandcharacterization

Withregardtotheplantingscheme,twomain J.curcassystemswereidentified:the

plantation and the hedge system, which were then further divided according to

management practices and the cultivation purpose. This resulted in five J. curcas systems.ThemostwidespreadsystemsweresmallͲscaleJ.curcasintercropping(47.7%

of the surveyed sites) located in the southͲwestern part and living fences (38.7%) inthe central part of Burkina Faso. Intensely managed plantations (2.7%), afforestationplots on abandoned cropland (4.5%) and hedges along contour stone walls (5.4%)presentedonlynichesystems(Table3.1;Figure3.1).



(1)SmallͲscaleintercroppingsystem

This system encompassed J. curcas cultivation combined with annual crops on land

ownedormanagedbysmallholderfarmers.Theplantationswereusuallylessthan2ha(1.8±0.15ha)(n=53).Themostcommontreearrangementswere4mx4m(interͲrowdistance x interͲplant distance) with 625 trees haͲ1(53%), 6 m x 4 m (417 trees haͲ1;13%) and 3 m x 2 m (1667 trees haͲ1; 11%). Of the plantations visited 21% wereestablishedin2007,53%in2008,and26%in2009.Accordingly,theresearchin2010and2011covered1Ͳto4Ͳyearoldtrees.

The intercropping system (see Appendix 9.1 for picture) was predominantlylocated in the southͲwestern region of Burkina Faso with favorable climate and soil

conditions (section 2.1). Here, NGOs and private companies introduced J. curcas and

its potential as an energy crop to smallͲscale farmers and promoted its cultivation.Seedlingsfreeofchargeweredistributedamongthefarmers,andfuturepurchaseof

theseedswasorallyguaranteed.RegardingthetransplantingofJ.curcasseedlings,a

planting depth of 30Ͳ50 cm with plantingͲhole dimensions of 50 cm x 50 cm wasrecommended. Planting was ideally scheduled prior to the rainy season in June.Seedlings originated from nurseries built up by the respective organization and wereraised,forexample,byAGRITECHfromcertifiedlocalseeds(certifiedincollaborationwith the laboratory of 2iE in Ouagadougou). Seedlings were grown in polybags filledwithamixtureofsand,manureandsoil,wateredregularly,andtransplantedintothe

field 8 to 12 weeks after sowing. After the first year, as reported by the farmers, aseedlingsurvivalrateof74Ͳ94%wasobserved.

Incontrast,managementtechniquesforJ.curcastreeswerenotpartofthe sensitization campaign. Hence, J. curcas was generally grown as a rainͲfed plant and

wasneitherfertilizednorsystematicallypruned.Only19%oftheinterviewedfarmers

temporarilyirrigatedtheirJ.curcastreesduringthefirstdryseason.Mineralfertilizers

werenotapplied,andonly24%ofthefarmersusedmixedmanureexplicitlyfortheirtrees and 19% cut some of the branches. Regular pruning is named as an importantpracticetomaximizeproductivity(Jongschaapetal.2007).However,mostfarmersdidnot prune for fear of destroying the trees or reducing productivity. Overall, in the

southͲwestern region the practices of J. curcas plantation establishment were quite

homogenous,asthesamesensitizationhadtakenplace.Performanceofplantations,however, differed depending on management practices such as rotation, fertilization

profitability of cotton cultivation (Commune Rurale de Boni 2009) and the expected

productivityofJ.curcas.Managementinterventionsassociatedwiththeintercropsare

listedinTable3.4.