Download - Soil science and agricultural development in Rwanda: state ... · Soil science and agricultural development in Rwanda: state ... En pratique cependant ... integrated in 2012 into

BASE Biotechnol. Agron. Soc. Environ.201418(1),142-154 Focus on:

SoilscienceandagriculturaldevelopmentinRwanda:stateoftheart.AreviewPascalN.Rushemuka(1,2),LaurentBock(2),JeremiasGasperMowo(3)(1)RwandaAgriculturalBoard(RAB).Box617.Butare(Rwanda).E-mail:[email protected](2)Univ.Liege-GemblouxAgro-BioTech.DepartmentSciencesandTechnologiesoftheEnvironment.Soil-WaterSystems.AvenueMaréchalJuin,27.B-5030Gembloux(Belgium).(3)WorldAgroforestryCentre.Box30677.Nairobi(Kenya).

ReceivedonApril19,2012;acceptedonSeptember18,2013.

PooragriculturalproductivityremainsacrucialprobleminRwandainspiteofnumeroustechnologicalinterventions,includingaspectsof soilmanagement.Theobjectiveof this studywas todraw lessons from thepastwith theview tobetterorientfuture interventions insoil fertilitymanagement.Theliteraturereviewanditerativefieldobservationswere thesourcesofinformation.Findingsfromthisstudyshowthatsubstantialprogresshasbeenmadeintheidentificationofdifferentsoiltypesandtheirspatialdistribution.Factorsrelatedto lowlevelofproductivityhavebeenidentifiedandsustainablesoil fertilitymanagementoptionshavebeendevelopedatplotlevel.However, thewidespreadadoptionofthesetechnologieshasbeenproblematic.Themainreasonisthefailuretotailorsoilfertilitymanagementtechnologiestospecificsoiltypes.Thestudyhasdemonstrated that thesoilmapofRwanda(CPRforCarte Pédologique du Rwanda)–1:50,000–offersaremarkablepotentialtoconstituteatooltosolvethisproblem.Inpracticehowever,theCPRremainsunderutilized,mainlybecauseofitsinaccessibilitytoitspotentialusers(e.g.policymakers,soilfertilityexperts,agronomistsandextensionists).Foritseffectiveuse,thefollowingisrecommended:Rwandansoilscientistsneedtoincreasepolicymakers’awarenessabouttheusefulnessofthissoilmap;agriculturalresearchneedstoadaptfromtheconventionalmodeltoatrulyparticipatoryandintegratedapproach;theCPRlegendshouldbeelucidatedbyprovidinginformationonthelandunitsinwhichsoilsoccurandbybridgingSoilTaxonomywiththefarmers’soilnomenclature;regionalsoilreferencesystemsshouldbeestablishedthatallowlinkingsoiltypeswiththefertilitystatusofarablelandandcropyields.ThisimpliestheneedfortrainingofRwandansoilscientistsinbothSoilTaxonomy(thelanguageoftheCPR)andthefarmers’soilnomenclaturesothattheycanserveasinterpreterforscientistsfromotherdisciplinesandfarmers.RwandansoilscientistsshouldbetrainedintheuseofGeographicInformationSystem(GIS)softwaretoenablethemtoexploitthedigitizedversion/softcopyoftheCPRandtobecomefamiliarwiththeRwandanbiophysicalenvironment.Keywords.Soilsciences,soilmap,agriculturalresearch,informationexchange,ruraldevelopment,Rwanda.

Science du sol et développement agricole au Rwanda : état de la question (synthèse bibliographique). AuRwanda,malgréplusieurs interventions techniques, encecompris lesaspectsde lagestiondes sols, la faibleproductivitéagricoleresteunproblèmecrucial.L’objectifdecetteétudeétaitdetirerlesleçonsdupasséenvuedemieuxorienterlesinterventionsfuturesengestiondelafertilitédessols.Larevuedelalittératureetlesobservationsitérativessurterrainontservidesourced’information.Lesrésultatsdecetteétudemontrentqu’unprogrèssubstantielaétéréalisédansl’identificationdesdifférentstypesdesolsetdeleurrépartitionspatiale.Lesfacteursresponsablesdufaibleniveaudeproductivitédesterresontétéidentifiésetlesoptionsdegestiondurableontétédéveloppéesàl’échelledelaparcelle.Cependant,leuradoptionàgrandeéchelleestrestéeproblématique.Laraisonprincipaleapparaitêtrel’incapacitéd’adapterlestechnologiesdegestiondelafertilitéauxdifférents typesdes sols.Cette étudemontredoncque laCartePédologiqueduRwanda (CPR) -1:50000– constitueunoutilpossiblepourrésoudreceproblème.Enpratiquecependant,laCPRrestesous-utilisée,principalementàcausedesoninaccessibilitéàsesutilisateurspotentiels(planificateurs,expertsengestiondelafertilité,agronomesetvulgarisateurs).Poursonutilisationeffective,lesrecommandationssuivantesontétéformulées:lespédologuesduRwandadevraientsensibiliserlesplanificateursàproposdel’utilitédecettecartedessols;larechercheetvulgarisationagricolesdevraientpasserdel’approcheconventionnelleàuneapprocheréellementparticipativeetintégrée;lalégendedelaCPRdevraitêtreexplicitéeenyincluantlesunitéspaysagiques/morphologiquesetenétablissantdespontsdecommunicationentrelalégendetaxonomiquedelaCPRetlesnomsvernaculairesdessols;dessystèmesrégionauxderéférencesurlessolsdevraientêtreétablis,quipermettentdemettreenrelationlestypesdesols,l’étatdefertilitédesterrescultivéesetlesrendementsobtenus.Ceciimpliqueunbesoin

SoilscienceandagriculturedevelopmentinRwanda 143

1. INTRODUCTION

Soil is studied from both fundamental and appliedpointsofview.Theknowledgeacquiredbybasicsoilscience is published in scientific journals and books.However, theway the information generated by thisscientific sub-discipline is used to formulate soundpolicies and translated into soil-specific and user-tailoredtechnologiesinappliedsoilscienceiscomplexandcontroversial.

While Hartemink (2006) maintained that soilscience has contributed to the increase in worldagricultural food production over the last 50years,manyotherauthors(Papadakis,1975;Leeuwisetal.,2004; Ruellan, pers. com.) asserted that the increasein agricultural food production in the industrializedworld was made possible less by progress in soilscienceandacademicresearchthanbytheagronomicsciences, which developed responsive fertilizervarieties, pesticides and intensive use of fertilizers,agricultural engineering, value chain developmentandmarkets.Theproblem is that thiscapital-ledandnonsoil-specificintensificationoffoodproductionhasoccurredat theexpenseof thecapacityof thesoil tosustainablyproducefoodandsupportlife(Rainaetal.,2006;Ruellan,pers.com.;Herren,2011).

Despite the above concerns, in those developingcountrieswherefoodproductionhasstagnatedoverthelast50years,thereisagreattemptationto“imitate”thedevelopedworld.Forinstance,intheAfricanFertilizerSummit, the conclusions of which were endorsedby theAfrican Heads of State atAbuja, Nigeria, in2006,itwasarguedthatforagreenrevolutiontotakeplace inAfrica, fertilizerusemustbe increased fromthe then mean of 8kg.ha-1 to ~ 50kg.ha-1 by 2015.Accordingly, African governments were encouragedto takeconducivemeasures to increase fertilizeruse.Following these recommendations, several Africancountries have used subsidies in efforts to increasefarm-levelfertilizerapplications(Marenyaetal.,2012).It is also in this context that Rwanda has promoteda policy of agricultural “modernization” and “cropintensification”withlandconsolidation,mechanization,mono-cropping,highyieldingcropvarieties,intensiveuse of fertilizers and irrigation (MINECOFIN, 2000;MINAGRI,2002;MINECOFIN,2007).

Agro-ecologists,whilesharingthesameconcernsaboutregardinglowagro-systemproductivity,wouldprefer not to see developing countries repeatingthe past errors of the developed world.Within thiscontext, they consider “agro-ecological solutions”or Ecological Agriculture (EA) (minimum useof fertilizers and investment in agroforestry) tobe superior to conventional agriculture based onchemicals or Industrial Agriculture (IA). Theythus propose measures to governments to lead thedevelopmentandadoptionofsuchapproaches(Altieri,2002;deSchutter,2010;Herren,2011;Marenyaetal.,2012). Soil scientists, for their part, maintain that“agro-ecological solutions” are unable to contributesignificantly to food securityandpovertyalleviationwithin the context of acid and inherently poor soils,such as those found in many parts of sub-SaharanAfrica (Drechsel et al., 1996; Rutunga et al., 2006;Breman,2011;Keatingetal.,2011).

Several questions arise: is the debate about IAversus EA new? Has any progress been achieved?Whatpositionshouldgovernmentstake?Shouldtheywait forscientists toreachacompromise,or isevenanycompromisepossible?

The objective of this study is to analyze howsoil science has evolved in Rwanda, what has beenachieved, how these achievements have contributedtoagriculturaldevelopment,whattheconstraintshavebeen,andwhatmightconstitutethewayforward.

2. METHODOLOGICAL APPROACH

A literature review, including unpublished reports,maps and journal articles, was the main source ofinformation for this study. A historical perspectiveapproachwasusedtoanalyzethecontributionofsoilscience to agricultural development inRwanda.Thehistorical time-frame covers a period of ~80years(1930-2010).Threeyears(2010-2013)ofiterativefieldactivityobservationswereundertaken to support theliterature reviewwith concrete and recent examples.Figure 1 presents the location of Rwanda withinAfrica,theAgro-EcologicalZones(AEZs)ofRwanda(Verdoodtetal.,2003a)andthemainsitescitedinthetext.

pouruneformationdespédologuesduRwandaàlamaitrisedela«Soil Taxonomy »(langagedelaCPR)etdelanomenclaturevernaculairedessolsafinqu’ilsserventd’interprètespourlesnon-pédologuesetlespaysans.Aumêmemoment,ilsdevraientaussi recevoirplusdeformationsur l’utilisationdes logicielsdeSystèmesd’InformationGéographiques(SIG)afind’êtrecapablesd’exploiterlaversiondigitalisée/électroniquedelaCPRetdevenirfamiliersaveclemilieubiophysiquerwandais.Mots-clés.Sciencesdusol,cartedesols,rechercheagricole,échanged’information,développementrural,Rwanda.

144 Biotechnol. Agron. Soc. Environ. 201418(1),142-154 RushemukaP.N.,BockL.&MowoJ.G.

3. SOIL SCIENCE COMPONENTS AND ACHIEVEMENTS IN RWANDA

Soil science, as applied in agricultural research anddevelopmentinRwanda,istraditionallysubdividedintothreemaincomponents:soilsurvey,soilconservation,andsoilfertilitymanagement.

3.1. Soil survey

ThefirstsoilsurveyinRwandawasundertakenbytheteamofthe“Institut National d’Études Agronomiques au Congo” (INEAC), beginning in 1955 at RubonaStation(Figure 1).AfterIndependence(1962),INEACactivitieswerecontinuedbythe“Institut des Sciences Agronomiques du Rwanda” (ISAR), which wasintegratedin2012intotheRwandaAgricultureBoard(RAB).By1963,themajorsoiltypesofthecountryhad

been described (VanWambeke, 1963). In the 1980s,almost all soil knowledge acquired by the INEAC-ISARteamwassynthesizedintoasoilassociationmapat a scale of 1:250,000 (Prioul et al., 1981).Duringthe same period, Pietrowicz (1985) undertook a soilsurveyinthemandateareaoftheproject“Projet Agro-Pastoral (PAP) Nyabisindu”(Figure 1).

During the period 1981-1990, the project “Carte Pédologique du Rwanda”(CPR)(Birasaetal.,1990)conducted a comprehensive soil survey of Rwanda.The CPR project produced a soil association and amedium scale (1:50,000) soil map (43sheets) underthe“SoilTaxonomy”classificationsystem.TheCPRdatabase was created in 1990-1994. In 2002, thesoil map was digitized and the associated databasepublished (Verdoodt et al., 2003b). In 2003, a set ofsoil suitabilitymapswas published (Verdoodt et al.,2003a).

Figure 1.LocationofRwandawithinAfrica,RwandaAgro-EcologicalZonesandsitescitedinthetext(adaptedfromVerdoodt,2003aandSchörry,1991)—Localisation du Rwanda par rapport à l’Afrique, Zones Agro-Écologiques du Rwanda et sites cités dans le texte (d’après Verdoodt, 2003a et Schörry, 1991).

20 10 0 20 40 60 80km

LegendPAPMandateArea

Sites with terracesKisaroKigemeMuniniRubonaRuhandeKanyirandoriKigali

Agro-ecological zonesImboImparaKivulakeborderCongo-NileWatershedDivideEasternplateauBirungaBuberekahighlandsIslandKivulakeMayagaandperipheralBugeseraCentralPlateEasternSavannaandCentralBugesera

N


3.2. Soil conservation and erosion control

ErosioncontrolhasbeenformallypracticedinRwandasince 1937 at INEAC research stations (Kabiligi,1985).In1947,theprogramwaswidenedtothewholecountry and many extensionists (mostly known as“MONAGRIS”1)wererecruited.In1947,thecolonialadministrative“resident”decreemadethecreationofditchesandtheplantingofgrassandtreesobligatoryforalllandholders.

After Independence (1962), erosion control wasabandonedandmanyerosioninfrastructureswereevendeliberately destroyed (Kabiligi, 1985). In 1966, thegovernmentofRwandarevivedanationalprogramofsoilerosioncontrol.Followingthis,severalcompulsoryfive-year programs (1966-1970; 1977-1981; 1982-1986)wereimplemented.

In 2005, as the countrywas recovering from the1990-1994civilwarandgenocide,theerosioncontrolprogramwasre-startedwiththesameapproach.Sincethen,manybench terraceshavebeenconstructed.Asaresult,inmanyregionsofthecountry,thelandscapehas changed remarkably. At the same time, the oldinfiltrationditcheswererenewed.

Infiltration ditches. One of the most ancient andcommonexamplesoferosioncontrolinfrastructureinRwandaisintheuseofinfiltrationditches(imingoti)stabilized by grass verges parallel to contour lines(Figure 2). In addition to being labor-intensive andlackingpositiveimpactsoncropyields,themethodisinefficient incontrollingsoilerosiononsteepslopes.Infact, theditchesdonothaveanyimpactontillagepractices(therootcauseofhuman-inducederosion)or

onphysicalsoilproperties(e.g.improvementofwaterinfiltrationrate).

Thus,whensoilswith loworganicmattercontentonsteepslopesarecultivated(to~60cmdepth),atthebeginningoftherainyseasonwhenlandisbare,theyareexposedtohigherosionrisk.Itiscommonfortheditchestofillupwithinafewdays(Rooseetal.,1993).

Bench terraces. Bench terraces (Figure 3) wereintroduced in Rwanda in 1973 in the mountainousregion of Buberuka AEZ at Kisaro hill (Figure 1).In this region, terraceshavebeengreatlyappreciatedas an effective way of controlling soil erosion andmaintaining or progressively improving soil fertility.Since1992,theuseofbenchterraceshasbeenexpandedtotheunproductivesoilsofthetopographicallysimilarmountainous region of the Congo-Nile watersheddivideAEZ,atKigemehill(Figure 1).Inthisregion,incontrast,benchterraceshavenotbeenadopted,andtheterracedterrainshaveremainedunused.

More recently (2006), under the “Food forWorkSystem”, many Districts and Non GovernmentalOrganizations (NGOs) have created bench terracesonlargeareasinallAEZs.However,bothancientandrecently constructed terraces have led to a situationwheresometerracesareusedeffectively,whileotherswere totally abandoned just after their construction.This has been reported as embarrassing for policy-makers andothernon-soil scientists interested in theadoption of terraces (Bizoza, 2011). However, foran informed soil scientist, inorder for terraces tobeusedeffectively,theywouldneedtobeconstructedonproductivesoils,whichcanstillberesponsivetofarmerinput (organic input or organic input + fertilizers).Alternatively,effectiveterraceswouldbeconstructedon very strongly acid and inherently poor soils, butwith appropriate input supply (limestone, organicinput, fertilizers and improved seeds). Terraces that

Figure 2. Fields on a steep slopewith infiltration ditchesand grass verges parallel to contour lines for erosioncontrol—Champs sur forte pente avec fossés d’infiltration et bandes enherbées parallèles aux courbes de niveau pour lutter contre l’érosion.

1MONAGRISrefersto“moniteursagricoles”.

Figure 3. Bench terraces with farmers harvesting Irishpotatoes at Munini — Terrasses radicales avec les cultivateurs récoltant la pomme de terre à Munini.


becomeabandonedwouldbethoseconstructedonverystronglyacidic,depletedandunproductivesoilswithoutadequateinputsupply.Indeed,ithasbeendemonstratedthatwhereappropriateinputsarewellcombined,highyields are obtained (Figures 4a and 4c -1st and lastplots) onbench terraces constructedon the otherwiseunproductive soils of southern Rwanda (Rushemukaet al., 2011). In this region, zero yields have beenobtained when there was low input (Figures 4b and4c-2ndand3rdplots).

3.3. Soil fertility management

Soil fertilitymaybeexaminedeither fromaclassicalor conventional perspective or from an integratedagroforestryapproach.

Conventional soil fertility management. In the1970s,aseriesofpotexperimentswasundertakenon

over500soilsamplesinordertodiagnosesoilfertilitylimitingfactors(Iyamuremye,1983).Itwasnotedthatthemajor soil fertility limiting factorswere (in orderofimportance)P,K,Nand,insomesoils,CaandMg.Followingtheseinitialexperiments,anineyear(1971-1980)soilfertilityexperimentwasundertaken(Rutungaetal.,2006)atMata in theMuniniregion(Figure 1).ManyotherstudieshavebeenundertakenbyISAR,indifferentAEZs,onthemaincropsgrowninthecountry.Sinceitscreationin1979,theFacultyofAgricultureattheNationalUniversityofRwanda(NUR)hasconductedother soil fertility-related researches (Ndoreyaho,1985).TheprojectSFI2-FAO(1980-1990)carriedoutanimportantstudyonfertilizeruseincollaborationwithISARandNUR(Coursier,1985).Theprojectundertooksimplefertilizertrialscountrywideforthemostwidelygrowncrops.Pietrowiczetal.3(1987)citedbyDrechselet al. (1996) adopted a more “ecological” approach(greenmanureorfarmyardmanureplusfertilizer)inthe

PAPNyabisindumandatearea.While,inmanycases,experimentswere

undertaken assuming the homogeneity ofAEZs, results from those experiments,whatever theAEZ, enabled Rwandan soilstobecategorized into three fertilityclasses(Rutunga,1991):–fertilesoils;–mediumfertilitysoils;–infertilesoils.

The characteristics and proportion of eachof these fertility classes are summarized intable 1. Fertile soils are unresponsive tofertilizers. They need manure for fertilitymaintenance. Medium fertility soils arehighlyresponsivetofertilizers.Theyrequiremanure+ fertilizers. Infertile soils areunresponsivetofertilizersaloneandneedacombinationoflime,manureandfertilizersinordertobeproductive.

The experiments that led to theseconclusions were generally undertakenon a trial and error basis, assuming thehomogeneityofAEZs.Consequently,itisachallenging task, in the complex soilscapeof Rwanda, to replicate experimentalresults. This is one of the main obstaclesto the adoption of soil-related technologyrecommendations.

Figure 4.BenchterracesonextremelyacidanddepletedsoilsofsouthernRwanda—Terrasses radicales sur sols extrêmement acides et fortement lessivés du sud du Rwanda.

a.Irishpotatoeswithtravertine+manure+fertilizers—pomme de terre avec travertin + fumier + fertilisants;b.Irishpotatoeswithoutanyinput/control—pomme de terre sans intrant/témoin;c.layoutoffourplots:thefirstandlastplotscorrespondtothefarmer’streatmentmadeupoftravertine+DAP+manure;thesecondplotcorrespondstotreatmentwithtravertine+DAP.Thethirdplotcorrespondstotreatmentwithtravertine+manure—dispositif de quatre parcelles : la première et la dernière correspondent à du travertin + fumier + DAP ; la seconde à du travertin + DAP ; la troisième à du travertin + du fumier.

2SFI:SoilFertilityInitiative.3PietrowiczP.&NeumannI.,1987.Fertilisation et amélioration des sols.Étude sur l’application d’engrais vert, de la fumure organique et des engrais minéraux.ÉtudesetExpériencesn°11.Nyabisindu,Rwanda:GTZ.

a b

c


Agroforestry approach. The modern practice andscientificdisciplineofagroforestrywasintroducedinRwanda by the PAP Nyabisindu in 1975 (Neumannet al., 1985). The PAP Nyabisindu was a long-term(1969-1989)integratedprojectrunundertheauspicesof the German-Rwanda Cooperation Program. Inthisproject,agroforestrywaspartofawiderconceptof EA and was conceived as appropriate means toachieve site-adapted technologies and sustainableland use (Schörry, 1991;Drechsel et al., 1996).ThePAP Nyabisindu promoted a system whereby treesand shrubs, livestock and cropswere intended to beassociated within one farm of ~1ha, known as a“fermette”.Within this system, trees were primarilyconsidered for their fundamental role in soil fertilityimprovementunderthenutrientrecyclinghypothesis,but also for collateral multipurpose functions, suchas erosion control (hedgerows), fodder, fuel andconstruction wood and stakes for climbing beans.Livestockwasprimarilyseenasasourceofmanureandwasassumedtobekeptinzerograzingandnourishedfrom on-farm produced leguminous and herbaceousfodder.Theaimofagroforestery,aspromotedbythePAPNyabisindu,wastoanticipate,asfaraspossible,the inappropriate use of fertilizers (on tropical soilswithveryloworganicmattercontentandunprotectedagainst erosion) in a land-locked country, wheretheir productivity and sustainability were uncertain(Neumannetal.,1985).Moreover, in theacidichighlands(>1,800),withhighCcontent(3-16%),fertilizeruseefficiencywasnotguaranteedbecausecorrectingpHofsuchsoilsrequiresmuchmorelimeduetobuffereffect.

By the same token,during theperiod1980-1995,manyresearchanddevelopmentprojectsexperiencedtheconceptofimprovingsoilfertilityandcropyieldswith the help of planted fallows with wood andherbaceous legumes or green manuring in differentAEZsofRwanda(Drechseletal.,1996).

AlthoughtheconceptofEApromotedbythePAPNyabisinduandmanyotherprojectswasexciting,thetechnological packages proposed by these projects(“fermette”, legume fallows, green manuring) were,surprisingly,notadopteddespitetheirapparentappeal(Schörry,1991;Drechseletal.,1996).Instead,farmers

selectively chose just some components. The treecomponentwasthemostwidelyadopted.Forinstance,attheendofthePAPNyabisinduProject(1990),80%offarmershadtreesontheirfarms;butonly10%hadrespectedprojectmanagementprescriptions(biomasspruning and composting or green manuring). Thus,a contrast could be seen between the apparent highadoptionoftreesandtheextremelylowrateoffarmersadheringaccuratelytothespecifiedrequirements.

The above situation is explained by the fact thattrees are mainly accepted for their stated secondaryroleassourcesoffuelwood,timberforconstruction,andstakesforclimbingbeans,butlessfortheireffecton soil fertility (Drechsel et al., 1996). The pooracceptability of trees as a soil fertility managementfactor inRwanda is attributed to the highly variableand generally minimal effect of green manuring onsoilpropertiesandcropyields(Drechseletal.,1996).ThiswasquitedifferentfromthesituationreportedinotherAfricancountries,wheretheconceptofEAwasoriginated.Forinstance,inNigeriaandinZimbabwe,therotationofmaizeandMucunagreenmanurewasabletomaintainmaizeyieldsformorethan20years(Vine, 19534; Rattray et al., 19535 both cited byDrechseletal.,1996).Incontrast,inRwanda,theeffectofcontinuouscroppingofmaizefollowedbybeansfor8yearswasinsignificant(Rutungaetal.,1998).Itwasnotedthatthesustainableeffectofgreenmanurewasrestricted to rather fertile soils, as the experience ofZimbabwe allowed, on the control plots, continuousmaizecroppingover22years(Drechseletal.,1996).TheRwandanexperience,ontheotherhand,gavenoyieldincontrolplots(Rutungaetal.,1998).Contraryto the experience of Zimbabwe where maize yieldsobtained after the fallowing season were more thanable to compensate the “lost” season, in Rwanda,evenintherelativelyfertilesoils,theresidualeffectofgreenmanurewasofonlyhalfayearandwasunabletocoverthe“lost”season(Drechseletal.,1996).Thelow

4VineH.,1953.ExperimentsonthemaintenanceofsoilfertilityatIbadan,Nigeria.Empire J. Exp. Agric.,21,65-85.5RattrayA.&EllisB.S.,1953.MaizeandgreenmanuringinsouthernRhodesia.Rhodesia Agric. J.,49,188-199.

Table 1.Soilfertilityclasses,theircharacteristicsandtheirproportiontowardsarablelandinRwanda—Classes de fertilité de sols, leurs caractéristiques et leur proportion par rapport aux terres cultivables au Rwanda. Fertility classes Limitation level pH (water) Al (meq.100g-1soil) SEB (%) Proportion Fertilesoils Low >5.5 <1.5 >3 27.5Mediumfertilitysoils Mediumtohigh >5.2<5.5 >1.5<3 >1<3 29.6Infertilesoils Verystrongtoextremelystrong <5.2 >3 >1 43.2Al:ExchangeableAluminum—Aluminium échangeable;SEB:SumofExchangeableBases—Somme des Bases Échangeables;Sources:Birasaetal.,1990;Rutunga,1991.


effectofgreenmanureintherelativelyfertile soils of Rwanda is attributedto the rapid leaching ofN andK orinappropriate foliage incorporation.In the infertile soils, it is explainedby the very low nutrient reservesandvirtuallynonutrientavailabletorecycleinthesub-soil(Drechseletal.,1996).Ontheinfertilesoils,eventhecombination of green manure withfertilizerswasnotabletosignificantlyincrease crop yields (König, 1992;Rooseet al.,1997).Theexplanationis that while the principal limitingfactors of those soils are the basiccation concentrations (Ca2+, Mg2+,K+),itwasnotedthattheseelementswere unaffected by green manure(Drechsel et al., 1996). Overall, inthe complex soilscape of Rwanda,without systematic consideration ofdifferentsoiltypesandtheirleveloffertility, green manuring proved tobe a risky enterprisewith uncertainresidual effect, which was entirelyinconsistent with farmers’ strategyofriskminimization(Drechseletal.,1996). Therefore, the concept ofimproving soil fertilitywith the aidof improved fallow failed (Schörry,1991; Raquet et al., 19956 cited byDrechseletal.,1996).

Withtheaboveexperience,ithasbecomeclearthatintheacidsoilsofRwanda,soilfertilityconstraintscannot be solved with farm-produced inputs only(Drechseletal.,1996).Therecommendationwasthatthe production and availability of farmyardmanureand local mineral fertilizers, such as travertine andvolcanic ash, should be supported (Drechsel et al.,1996).

In this context,more recently, with agroforestryhedgerowsoncontourlines,attheKanyirandoliRABResearchStation,thecombinationoflime/travertine+manure+fertilizersonextremelyacidandinherentlypoor oxisols has generated spectacular crop yields(Figure 5).However,theadoptionoftheKanyiradolimodelhasnotyetbecomewidespread.Thefrequentlygivenexplanationisthatsuchanagroforestrysystemis a knowledge-demanding, labor-intensive andhigh investment technology (Drechsel et al., 1996).This means that designing an on-site experimental

sustainable agroforestry system might be relativelysimple compared with the necessary understandingof thecontributing factors toensure its replicability(Rhoades, 1999). Effective adoption depends onfactorssuchas:– farmers’investmentcapacity;– appropriatetechnologytransfer;– inputaccessibility;– cropproductmarket;– adequatebiomasspruningequipmentand,perhaps moreimportant;– farmers’perceivedbenefits(Coursier,1985).

Thesamefactorsalsoplayaroleintheadoptionofbenchterraces(Bizoza,2011).

In view of the above, contrary to what someagro-ecology proponents maintain, manuring maynotbeacheaperoptionforsoilfertilitymanagement(Breman,2011).Sometimesitisdescribedasalow-inputtechnologysincemanureislocallyavailableincontrasttoexternalinputssuchasinorganicfertilizers(Altieri, 2002).However, the real cost ofmanuringincludes substantial amountof inputs suchas labor,skillsandmanagement(Drechseletal.,1996;Altieri,2002).

Figure 5.Kanyirandolimodel:agroforestryhedgerows(progressiveterraces)onacidsoilswithhighinputsystem(lime+manure+fertilizers)—Modèle Kanyirandoli : haies vives agroforestières (terrasses progressives) sur sols acides avec système d’intrants intensifs (chaux + fumier + engrais minéral).

a.generaloverviewofprogressiveterraces—vue générale de terrasses progressives;b.Irishpotatoes—pomme de terre;c.wheat—blé;d.plotafterharvesting—parcelle après récolte.

6RaquetK.&NeumannI.F.,1995.IntensivbracheinderkleinbäueurlichenLandwirtschaft.In:EggerK.&KorusU.,eds.Öko-Landbau in den Tropen.Heidelberg,Deutschland:MüllerVerlag,199-214.

a

b c d


4. DISCUSSION

4.1. Soil survey, soil conservation and soil fertility management

Due to the fact that Rwanda is a hilly countrywithsteepslopesandbecausethepopulationhaslongbeenexposed toerosioncontrolcampaigns,policy-makershavecometoperceivesoilerosionasthemajorcauseof poor soil fertility in most Rwandan soils. As aconsequence,financialresourcesandextensioneffortshavebeenconcentratedonerosion.However,despitethe different erosion control methods and extensionapproaches used over the last 80years, the adoptionof erosion control measures has remained very low(Schörry, 1991; Bizoza, 2011). The reason for thismight be that, for most Rwandan soils (Table 1),underlying the problem of soil erosion is a graveproblem of low fertility due to the parentmaterials.Thesoilsthusrequirehigh-levelsofinvestmentinsoilamendments(Rutungaetal.,2006;Rushemukaetal.,2011).Withsuchsoiltypes,itisnotenoughtocontrolerosion,asproductivityisalreadyverylow;theyoftenproducezeroyieldsofcerealsandlegumes(Rutungaetal.,1998),andonlymarginalonesofsweetpotatoesor cassava (1-3t.ha-1) (Roose et al., 1997). In suchconditions,itiswisertofocuson“win-win”solutions:erosion control as part of Integrated Soil FertilityManagement (ISFM7) with crop yields and farmproductivityas indicators (Figures 4 and5). Inmostcases, the initial investment to convert unproductivesoil into productive soil is beyond the financialcapacityoffarmers(Bizozaetal.,2010;Gilleretal.,2011).Thus,thelowproductivityofaRwandanfarmisprimarilyduetopoorsoils,whilethemaincauseofthelowlevelofadoptionofproposedISFMtechnologiesis farmers’ lack of resources.However, agronomists,extensionistsandfarmerslackgoodunderstandingofhowbesttomanageavailableresources(lime,organicinputandfertilizers)(Gilleretal.,2011).Itisherethatdecision-makersneedtolistentosoilscientists,sothatappropriate investment in soil fertility managementmaybemadeandfarmershelpedtoefficientlyusethelimitedresources.

4.2. Industrial versus ecological agriculture

The debate regarding IA (IA: predominantly theuse of inorganic fertilizers) versus EA (EA: largelyagroforestry, organic manure and minimal use ofinorganicfertilizers)isnotnew.TheIAasproposedbyIFAD(2006)isareturntothe1960-1970swhentheuseofinorganicfertilizeralonewasthoughttobesufficientto improve and sustain yields (Fairhurst, 2012).ThevisionofEAproposedbyauthorssuchasdeSchutter(2010)isalsoareturnto1980swhenorganicinputusewasseenasappropriateoptiontosustainyieldswithaminimal role of inorganic fertilizers. InRwanda, theexperienceofIAwassynthesizedfor30years(1960-1990)(Rutunga,1991).Formore than20years,PAPNyabisindu’smainfocuswasEA,whileotherprojectspromotedgreenmanureforalmostaslong.

TheexperienceofRwandashowsthatrather thanseeingIAandEAasopposingconcepts,abetterwaytoviewthemisasconvergentconcepts.Afterall,bothIAandEAproponentsaimtobringaboutasustainable“greenrevolution”,whichcanonlybeattainedthroughoptimal investment in soil amendments (lime andorganic input), inorganic fertilizers and germplasmassumingoptimumclimaticconditions. In thissense,EA and IA have become synonymous with ISFM,as noted by Breman (2011). Today however, therealizationisthat,inthe21stcentury,northeIAnortheEAneithertheISFMhasmetexpectations:since1960per capita food production has declined by 20-40%inEast,Central andSouthernAfrica and small-scalefarmers remainpoorly servedby the linearResearchand Development (R&D) innovation model and itstop-downtechnologytransfer(Keatingetal.,2011).

The challengewith ISFM, aswith IA andEA, ishowtosystematicallytakeintoaccountsoilvariationswhen designing experiments, evaluating data andextrapolatingISFMrecommendationsinthecomplexsoilscape of Rwanda. This is normally the role ofthesoilsurveyand,inRwanda,itwastheaimoftheCPR (Carte Pédologique du Rwanda), Rwanda’scomprehensivesoilsurvey.However,sincetheCPR’scompletion, its results have not been used for thispurpose.

4.3. Constraints in using the soil map of Rwanda (CPR)

This studyhighlightsfive reasonshampering theuseof theRwanda’ssoilmapinagricultural researchfordevelopment.

Historical reasons. As with all sectors of life inRwanda, the CPR and its use did not escape thenegativeconsequencesof the1994genocide.Almostall personnel involved in soil map’s production

7 ISFMisdefinedas theapplicationofsoil fertilitymanagementpractices, and the knowledge to adapt these to local conditions.ISFM aims tomaximize both the efficient use of fertilizers andorganicresourcesandcropproductivity.Thepracticesnecessarilyinclude appropriate fertilizer and organic input managementcombinedwith the utilization of improved germplasm (Adesina,1996[FactorsaffectingtheadoptionoffertilizersbyricefarmersinCôted’Ivoire.Nutr. Cycling Agroecosys.t,46,29-39]citedbySanginga et al., 2009). In hilly country with acid soils, erosioncontrolandliming(travertineandvolcanicash)shouldbeimportantcomponentsofISFM.


and use were either killed or driven out, and muchinformationwaslost.It tookeightyears(1994-2002)torebuildthedatabaseandthenpublishthedigitizedversion.The scant input fromRwandanpersonnel inthe implementation stagemeant the loss of valuableexperience and understanding of the technology andphilosophyunderpinningthemap.Today,18yearsafterthegenocide,fewRwandansoilscientistsareinterestedinitsuse.Moreover,policy-makersarelargelyunawareoftheusefulnessofsoilmapsinagriculturalresearchand development planning. Attempts to build theirpositiveattitudetowardstheCPR’susefulnesshadtorestart from zero. However, these circumstances arenotenoughtoexplainallthedifficultiesthatunderminetheuseofsoilmapsindevelopingcountries,suchasRwanda.Indeed,intheregionandthroughoutAfrica,examples of the successful use of soil maps for thebenefitofsmallfarmeragriculturearefew,iftheyexistat all. InWestAfrica, Nachtergael (2000) noted theconfusionbetweensoilfertilitypotentialidentificationand landevaluationand the relationshipbetween thetwo.Muchenaetal.(1995)andLal(1995)discussedtheminimalcontributionofsoilscienceinEastAfrica(Kenya, Uganda and Tanzania). Habarurema et al.(1997) and Steiner (1998) have discussed this forRwanda.Similarly,Burundi’ssoilsuitabilitymapsarelittle used (Tessens, 1991).The recent IA-EAdebateandcurrentstateoftheartofISFM(Fairhurst,2012)demonstratetheslightconsiderationpaidtosoilmapsin discussions of soil fertility management. Due tothis difficulty, the African Soil Information Service(AFSIS) is constrained to experiment with a globalintegrated soil information service (Shepherd et al.,2010). Thus, beyond the historical reasons, otherfundamental causes for the low use of soil maps inRwandadeservedeeperexamination.

A complex biophysical environment. Due to thecomplexity of relief and parent materials, Rwanda’sbiophysicalenvironmenthasexceptionalsoilvariationacrossveryshortdistances(Dressler,1983;Pietrowicz,1985; Birasa et al., 1990; Steiner, 1998). Farmersrespondtovariationsinsoilswithapparentlycomplexfarming systems. Steiner (1998) noted the limitationofidentifyingasmall,welldefinedandrepresentativerecommendation zone in such circumstances. HeobservedthatsoilsvarybetweenAEZs,astheydowithinoneAEZ.WithinoneAEZ,soilsofdifferentsuitabilityvaryfromhilltohill(Dressler,1983).Evenononehill,soilsandsoilpropertiesvaryfromthehilltop/upperhillto the lowerslopeandvalleybottom(Steiner,1998).Fromasuitabilitypointofview,under thesamesoilfertility management system, lower slopes can yield20-50% less compared with upper slopes (Steiner,1998).Thissoil fertilitygradientalongtheslopeisacommonfeatureinRwanda(Nizeyimanaetal.,1988;

Steineretal.,1994)andinsub-SaharanAfrica(Tittonelletal.,2007).Thesoilfertilitygradientisnormallywellexplainedbytheconceptsofcatenaandtoposequence(Okusami, 2006). However, in Rwanda, the formsof slope are complex, so that the slope criterion israrelypracticalfordefininga“recommendationzone”(Steiner,1998).Thus,scientistsfaceadilemmainthiscomplex biophysical environment.On the one hand,soil fertility management recommendations needto be as soil-specific as possible to be replicable toanalogoussoiltypes.Ontheotherhand,becauseofthebiophysicalcomplexity,ithasbeenimpossibletodefinea“recommendationzone”.Underthesecircumstances,the conventional research and extension approachbecomeslessappropriatetodevelopandtransferISFMtechnologies(Steiner,1998).

Research mode and research institutions.Soilmapsarehamperedfrombeingthefoundationofsoil-specificand replicable ISFM technologiesby researchpolicymisconceptions and biased funding systems. Thesein turn lead to research programs oriented towardsscientificdisciplinesorcroporlivestockcommodities(Rainaetal.,2006).Forinstance,undertheRABanditspredecessors,ISARandINEAC,eachcrop(e.g.maize,rice, sorghum, beans, cassava) constitutes a researchprogram.Similarly, biophysical sciences suchas soilscience and forestry are subdivided into programs,suchaserosioncontrol,soilfertilitymanagementandagroforestry. Rigid frameworks, including researchagenda, experimental sites, reporting systems,and incentives, govern these autonomous researchprograms. Such a context prevents soil scientists(pedologists)frompartneringwithsoilfertilityexpertsor crop and animal production specialists. As partof this philosophy, even soil fertility managementand erosion control practices are implemented asseparateprograms,disconnectedfromthesoilresourceinformation contained in the CPR.Today it is clear,however, that no research program is likely tomakemuchprogressonitsowninthelightofthe21stcenturydriversandneeds,andthattechnicalconstraintscannotbesolvedwithoutbroad-basedinstitutionalinnovation(Keatingetal.,2011).

Itwasinthiscontextthatin1982theInternationalService for NationalAgricultural Research (ISNAR)observed that Rwanda’s agricultural research anddevelopmentwasenteringacrisisphaseandstronglyrecommendedreform(ISNAR,1982).PAPNyabisindualso recognized the limitation of the conventionalagricultural research and development frameworkwhenitintroducedtheconceptof“agricultureadaptedtothebiophysicalandsocio-economicenvironments”.

More recently, Rwanda’s ISAR appropriatedthe innovative concept of Participatory IntegratedWatershedManagement from theAfrican Highlands


8AHIwasaneco-regionalprogramoftheConsultativeGroupofInternationalAgriculturalResearch(CGIAR)andtheAssociationforStrengtheningAgriculturalResearchinEasternandCentralAfrica(ASARECA).

Initiative(AHI8)(Germanetal.,2006).TheAHImodelappearedtopromisetoovercometheproblemoftheconventional research and its top-down extensionapproach.However,ithasalsofacedtheinstitutionalrigidity of the linear R&D model and experiencedthe drawbacks ofmany other development conceptspromoted by international research (Rhoades, 1999;Keating et al., 2011). In this planning environment,thesoilmapfindsnoplace.

Internal limitations of soil maps: complex knowledge. In the current arena of participatoryintegratedresearchframeworks(Germanetal.,2006),and because traditional soil maps (soil-centeredapproach) are only understood by users who knowhow surveys are made, soil scientists have realizedthechallengesofusingthesemapstoworkinatrans-disciplinary fashion (Wielemaker et al., 2001; Bui,2004). Indeed, a soil-centered andmultipurpose soilmapliketheCPRismoredirectedtoapeeraudienceofotherclassifiers thanoutwardtoa largergroupofnon-soil surveyorpotentialusers (Wielemakeretal.,2001).

Tosolvethiscommunicationproblem,Wielemakeretal.(2001)proposedamulti-hierarchicallandsystemapproach. In this model, higher categories of maplegend are expressed in geomorphological terms.Lowercategoriesareoften landscapecomponents inwhichsoilsaredescribedaspatternsorassociations.Thisapproachisclosertothegeomorphopedologicalapproach of University of Liege- GemblouxAgro-Bio Tech. With this concept, Bock (1994) stressedthe need to identify soil fertility potential and itslink with topsoil fertility evaluation by means ofrepresentativecompositesoilsamples.Inagriculture,the multi-hierarchical land system approach isvery important because each landscape unit has itsown specific soil and related level of productivity(Steiner, 1998). There is clearly additional work todo if theCPR is to be used in amore participatoryandmultidisciplinarymanner.Forinstance,itshouldbe circumscribed in the large national biophysicalspatial organization/natural regions and its legendpresented in relation to the landscape context, inline with the multi-hierarchical Land InformationSystem(LandIS).Followingthislogic,thereismuchto be gained in intelligibility by using spatial landinformation, such asAEZ-watershed-hill-land units,before moving onto soil fertility management. Forsoilfertilitymanagement, it is important toestablishregionalsoilreferencesonsoilsthatallowtolinkdata

on soil types (morphological, physical and chemicalcharacteristics)tothestateofthecultivatedlands(bymeansofsub-surfacecompositesoilsamples)andtotheobtainedcropyieldsconsideringthesoilmap,thebaseofthefieldunderstandingandtheframeworkfordataarchivingfinally.

Reliance on soil taxonomy and the neglect of farmers’ soil knowledge.SoiltaxonomyisthelanguageoftheCPR(Birasaetal.,1990).This,forfamersandmanyotherpotentialusersoftheCPR,presentsanadditionalcommunication constraint. In themeantime, farmershavemaintainedtheirsoilknowledgesystemwithanimmensely practical soil nomenclature (Habaruremaet al., 1997). Since it would be unrealistic toexpect non-soil scientists to master soil taxonomy(Thomasson, 1981), Steiner (1998) recommendedthatRwandan scientistsbuild ISFM technologiesonthesynergybetweenfarmers’soilknowledgesystemsand soil taxonomy.He recommendedusing farmers’soil nomenclature to transfer ISFM technologies toanalogous soil types. In this context, soil scientistsmust be prepared to understand both technical (soiltaxonomy)andfarmers’soilnomenclaturesandtousetheseasan interfacebetweennon-soil scientists andfarmers. This task can be described as a translationprocess,initsbroadbiophysicalenvironmentmeaningandculturalcontext,ratherthaninanarrowlinguisticsense (Thomasson, 1981). Farmers should be fullpartnersinresearchandextensionprocesses.

5. CONCLUSIONS

The objective of this study was to outline the pastand current contribution of soil science toRwandanagricultural development and propose the wayforward.

Ourfindingsshowthatmuchtimeandfundshavebeen lost as a consequenceof poor considerationofsoilresourceinformationinagriculturalresearchandextensionplanningandimplementationduringthelast50years. This study shows clearly that the currentcontroversyaroundIA,EAandISFMisamisleadingdebate, largely because it is conducted with littleconsideration of the biophysical environmentalcontext.Morespecifically,itoverlookstheexistenceofsoiltypesofvaryingsuitabilityandtheneedtodevelopsite-adaptedandsoil-specifictechnologies.Giventhis,inRwanda,theCPRemergesasanimportantplanningdocument that can augment understanding of thecomplexRwandanbiophysicalenvironmentandassistinformeddecision-makingforitssoundmanagement.However,duetoitsinaccessibilitytomanypotentialusers, the CPR, completed in 1990, has remainedunusedfromthe1994genocidetodate.


In thefieldof ISFM, thechallengeahead ishowtousethissoilmaptoimplementsoil-specificISFMtechnologiesinRwandawhere:– soilscientistsareexpectedtousetheCPRwithits soil taxonomy language and collaboratewith crop scientistsandfertilityexpertswithlittlebackground insoilscience(pedology);– agricultureispracticedbysmallholderfarmerswith theirownsoilknowledgesystem;– ISFMtechnologiesareexpectedtobetransferredby extensionistswithinsufficientunderstandingofboth soilknowledgesystems.

This study proposes three complementaryalternatives:– to improve access to the CPR by completing its legendwithlandunitterms;– to build ISFM technologies based on the synergy betweentechnicalandfarmers’soilknowledge;– tousefarmers’landunittermsandsoilnomenclature fortechnologytransfer.

Inturn,thiswillrequiretwoimportantchangesintheconductofagriculturalresearchandextension:– the research modes and institutions will need to change from the current top-down extension approach to more participatory and integrated approaches;– the gap between technical and farmers’ soil knowledge will need to be bridged to improve communicationbetweenscientistsandfarmers.

For this, Rwandan soil scientists will need tomaster both soil taxonomy and the farmers’ soilnomenclature so that they can serve as interpretersbetweenscientistsfromotherdisciplinesandfarmers.Finally,soilscientistswillrequiremoretrainingintheuseofGeographicInformationSystems(GIS),sothattheyareabletousethedigitizedversion/“softcopy”of the CPR and increase their familiarity with thebiophysical environment.All of these, including theconstructionofamulti-hierarchicalLandInformationSystem(LandIS)willhelptobringaboutmuchneededsoil-specific ISFM recommendations in Rwanda andexpeditetheiradoptionbyfarmers.

List of abbreviations

AEZ:Agro-EcologicalZoneCPR: Carte Pédagogique du RwandaDAP:Diammoniumphosphate(Phosphate de diammonium)EA:EcologicalAgricultureIA:IndustrialAgricultureISFM:IntegratedSoilFertilityManagementPAP:Projet Agro-Pastoral

Acknowledgements

TheauthorsthanktheBelgianTechnicalCooperation(BTC)toprovidethenecessaryfinancialsupportforthisresearch.CatharineWatsonisalsoacknowledgedforEnglisheditingofthefinalversionofthisarticle.

Bibliography

AltieriM.A., 2002. Agro-ecology: the science of naturalresource management for poor farmers in marginalenvironments (a review).Agric. Ecosyst. Environ.,93,1-24.

BirasaE.C. et al., 1990. Carte pédologique du Rwanda.Kigali:MINAGRI.

BizozaA.R., 2011. Farmers, institutions and land conservation: institutional economic analysis of bench terraces in the highlands of Rwanda.PhD thesis: Wageningen School of Social Sciences(TheNetherlands).

BizozaA.R. & de GraaffJ., 2010. Financial cost-benefitanalysis of bench terraces in Rwanda. Land Degrad. Dev.,23(2),103-115.

BockL., 1994. Analyse des sols et gestion de l’espace.Plaidoyer pour leur cadrage géomophopédologiquedanslesprojets,expertisesetservicesdeconseil.Étude Gestion Sols,1,23-33.

BremanH., 2011. The agroecological solution: foodsecurity and poverty reduction in sub SaharanAfrica,with an emphasis on the EastAfrican Highlands. In:CIALCA et al., eds.Challenges and opportunities for agricultural intensification of the humid-highland systems of sub-Saharan Africa. Conference abstracts, October 24-27, 2011, Kigali, Rwanda, http://cialcaconference.org/2011/10/25/an-agroecological-solution,(24/06/2013).

BuiN.E., 2004. Soil survey as knowledge system.Geoderma,120,17-26.

CoursierC., 1985. Le programme Engrais FAO. In:MINAGRI, eds. Premier séminaire national sur la fertilisation des sols au Rwanda, 17-20.06.1985, Kigali, Rwanda. Kigali: MINAGRI

DeSchutterO., 2010.AgroEcology and the right to food.Reportpresentedatthe16thSessionoftheUnitedNationsHumanRightsCouncil[A/HRC/16/49],8March2011.

DrechselP.,SteinerK.G.&HagedornF.,1996.Arevueonthepotentialof improvedfallowsandgreenmanureinRwanda.Agroforestry Syst.,33,109-136.

DresslerJ., 1983. La structure agraire de la région du projet. Résultats d’enquête socio-économique. Étudeset Expériences n°2. Nyabisindu, Rwanda: PAPNyabisindu.

FairhurstT.,ed.,2012.Handbook for integrated soil fertility management.Nairobi:AfricaSoilHealthConsortium.

GermanL.et al.,2006.Participatory integratedwatershedmanagement:evolutionofconceptsandmethods inan


ecoregional program of the easternAfrican highlands.Agric. Syst.,94,189-204.

GillerK.E. et al., 2011. Communicating complexity:integrated assessment of trade-offs concerning soilfertilitymanagementwithinAfricanfarmingsystemstosupport innovation and development.Agric. Syst.,104,191-203.

HabaruremaE. & SteinerK.G., 1997. Soil suitabilityclassificationbyfarmersinsouthernRwanda.Geoderma,75,75-87.

HarteminkA., ed., 2006. The future of soil science.Wageningen, TheNetherlands: International Union ofSoilSciences.

HerrenH.R.,2011.ParadigmchangeforAfricanagriculture:whyandhowtomakethetransition.In:CIALCAetal.,eds. Challenges and opportunities for agricultural intensification of the humid-highland systems of sub-Saharan Africa. Conference abstracts, October 24-27, 2011, Kigali, Rwanda.

IFAD (International Fund for Agricultural Development),2006. Africa fertilizer summit, http://www.ifad.org/events/fertilizer/index.htm.,(27/12/2012).

ISNAR (International Service for National AgriculturalResearch), 1982. Le système national de recherche agricole au Rwanda. Rapport au gouvernement de la République rwandaise. The Hague, The Netherlands:ISNAR.

IyamuremyeF., 1983. Synthèse des acquis scientifiques et techniques de la recherche agricole au Rwanda.Butare,Rwanda:InstitutdesSciencesAgronomiquesduRwanda(ISAR).

KabiligiJ., 1985.La protection et la conservation des solsagricoles au Rwanda. In: Direction générale duGénieruraletdelaConservationdesSols.,Rwanda,ed.Premier séminaire national sur la fertilisation des sols au Rwanda, 17-20 juin 1985, Kigali, Rwanda.Kigali:MINAGRI.

KeatingB., CarberryP. & DixonJ., 2011. Sustainableintensification and the food security challenge. In:CIALCA et al., eds.Challenges and opportunities for agricultural intensification of the humid-highland systems of sub-Saharan Africa. Conference abstracts, October 24-27, 2011, Kigali, Rwanda.

KönigD., 1992.The potential of agroforestrymethods forerosioncontrolinRwanda.Soil Technol.,5(2),167-176.

LalR.,1995.DevelopmentinEastAfrica.Discussionpaper-Comment.Geoderma,65,159-164.

LeeuwisC. & van den BanA., 2004. Communication for rural innovation: rethinking agricultural extension.3rded.Oxford,UK:Blackwell.

MarenyaP. et al., 2012.Which policy would work betterfor improved soil fertility management in sub-SaharanAfrica,fertilizersubsidiesorcarboncredits?Agric. Syst.,110,162-172.

MINAGRI (Ministère rwandais de l’Agriculture et desRessources Animales), 2002. Politique Nationale Agricole.Kigali:MINAGRI.

MINECOFIN(MinistryofFinanceandEconomicPlanning),2000.Vision 2020.Kigali:MINECOFIN.

MINECOFIN(MinistryofFinanceandEconomicPlanning),2007. Economic development and poverty reduction strategy 2008-2012.Kigali:MINECOFIN.

MuchenaF.N.&KiomeR.M.,1995.Theroleofsoilsciencein agriculturaldevelopment inEastAfrica.Geoderma,67,141-157.

NachtergaeleF., 2000. New development in soilclassification: the World Reference Base for SoilResources.Conferencecommunication.In:The West and Central Africa Soil Correlation Meeting, Abomey, Benin,http://www.fao.org/docrep/005/Y3948F/y3948f04.htm,(29/07/2012).

NdoreyahoV.,1985.ContributiondelaFacultéd’Agronomiedel’UNRàlarecherchesurlafertilisationauRwanda.In :MINAGRI,eds.Premier séminaire national sur la fertilisation des sols au Rwanda, 17-20.06.1985, Kigali, Rwanda.Kigali:MINAGRI,194-202.

NeumannI. & PietrowiczP., 1985. Agroforesterie à Nyabisindu. Projet agro-pastoral de Nyabisindu.Étude et Expériences n°9. Nyabisindu, Rwanda: PAPNyabisindu.

NizeyimanaE.&BickiT.J.,1988.Soilandsoil-landscaperelationships in north central region of Rwanda, EastAfrica.Soil Sci.,153,225-236.

OkusamiT.A., 2006. Catena/Toposequence/SoilAssociation: Unifying Concept. In: Soil classification, soil genesis and land use in the West Africa sub-region-Nigeria, 18th World Congress of soil science, July 9-15, 2006, Philadelphia, Pennsylvania, USA,http://www.ldd.go.th/18wcss/techprogram/P15374.HTM,(11/06/2013).

PapadakisJ., 1975.The world food problem: another low cost technology is needed: the failure of conventional agronomy.BuenosAires:J.Papadakis.

PietrowiczP., 1985. Les sols de la région d’action du projet agro-pastoral de Nyabisindu, Rwanda.ÉtudesetExpériencesn°7.Nyabisindu,Rwanda:PAPNyabisindu.

PrioulC. & SirvenP., 1981. Atlas du Rwanda. Kigali:UniversitéNationaleduRwanda.

RainaR.S., SangarS., SulaimanR.V. & HallJ.A., 2006.ThesoilscienceinIndia:policylessonsforagriculturalinnovation.Res. Policy,35,691-714.

RhoadesR.E., 1999. Evolution of agricultural research and development since 1950: toward an integrated framework.GatekeeperSeriesNo12.London:IIED.

RooseE., NdayizigiyeF. & SekangeL., 1993.L’agroforesterie et la GCES au Rwanda: commentrestaurerlaproductivitédesterresacidesdansunerégiontropicale de montagne à forte densité de population?Cah. Orstom. Série Pédologie,28(2),327-349.

RooseE.&NdayizigiyeF.,1997.Agroforestry,water andsoil fertility management to fight erosion in tropicalmountainsofRwanda.Soil Technol.,11,109-l19.

RushemukaP.N. et al., 2011. Imperative combinationof lime, organic matter and fertilizers to foster the


productivityof theacidanddepletedsoilsofSouthernRwanda. In:Woomer P.L., ed. Integrated soil fertility management in Africa: from microbes to markets. Conference Information, Program and Abstracts. An international conference held in Nairobi, Kenya, 22-26 October 2012. CIAT.

RutungaV., 1991. Essai de synthèse des connaissances acquises sur la fertilisation des cultures au Rwanda (± 1960-1990).Kigali:MINAGRI.

RutungaV. et al., 1998. Continuous fertilization on non-humiferous acid oxisols inRwanda “PlateauCentral”:soilchemicalchangesandplantproduction.Biotechnol. Agron. Soc. Environ.,2(2),135-142.

RutungaV.&NeelH.,2006.Yieldtrendsinthelong-termcrop rotation with organic and inorganic fertilizerson alisols inMata (Rwanda).Biotechnol. Agron. Soc. Environ.,10(3),217-228.

SangingaN. & WoomerP.L., eds, 2009. Integrated soil fertility management in Africa: principles, practices and development process.Nairobi:TropicalSoilBiologyandFertilityInstituteoftheInternationalCentreforTropicalAgriculture.

SchörryH.,1991.Le projet agro-pastoral de Nyabisindu : sa conception, son historique et ses perspectives.ÉtudesetExpériencesn°11.Nyabisindu,Rwanda:PAPNyabisindu.

ShepherdK. & VagenT., 2010. A Globally Integrated African Soil Information Service (AFSIS),http://ongoing-research.cgiar.org/factsheets/a-globally-integrated-african-soil-information-service-afsis/,(27/01/2010).

SteinerK.G., 1998. Using farmers’ knowledge of soils inmakingresearchresultsmorerelevanttofieldpractice:experiencefromRwanda.Agric. Ecosyst. Environ.,69,191-200.

SteinerK.G., BischoffW.A., DrechselP. &NzabonihankuyeG., 1994. Relation entre position sur le versant, sol et rendement dans deux zones agro-écologiques du sud du Rwanda. Note technique.Rubona,Rwanda:InstitutdesSciencesAgronomiquesduRwanda(ISAR).

TessensE.,1991.Une méthodologie pour évaluer l’aptitude des terres du Burundi. Note technique n°156. Gitega,Burundi: Institut des Sciences Agronomiques duBurundi(ISABU).

ThomassonC.G.,1981.Maximizingparticipativeplanning:culturalandpsychologicalaspectsofusercenteredsoilresource inventory, data preparation and presentation.In: Soil resource inventory group, eds. Soil resource inventories and development planning. Technical monograph No 1. Selected papers from the proceedings of workshops, April 4-7, 1977 and December 11-15, 1978, Cornell University, Washington, USA.Washington,USA:USDA, 391-400.

TittonellP.,VanlauweB.,deRidderN.&GillerK.E.,2007.Heterogeneity of crop productivity and resource useefficiencywithinsmallholderKenyafarms:soilfertilitygradientormanagementintensitygradients.Agric. Syst.,94,376-390.

VanWambekeL.,1963.Cartes des sols et de la végétation du Congo, du Rwanda et du Burundi : note explicative de la carte des sols du Rwanda et du Burundi.Bruxelles:INEAC.

VerdoodtA. &Van RanstE., 2003a. Land evaluation for agricultural production in the tropics: a large scale land suitability classification for Rwanda. Gent, Belgium:GhentUniversity,LaboratoryofSoilScience.

VerdoodtA.&VanRanstE.,2003b.Soil map of Rwanda.Gent, Belgium: Ghent University, Laboratory ofSoil Science, http://www.labsoilscience.ugent.be/soilmaprwandaintro.html.,(13/12/2012).

WielemakerW.G.,deBruinS.,EpemaG.F.&VeldkampA.,2001. Significance and application of the multi-hierarchical landsystem in soil mapping. Catena, 43,15-34.

(58ref.)

Download - Soil science and agricultural development in Rwanda: state ... · Soil science and agricultural development in Rwanda: state ... En pratique cependant ... integrated in 2012 into

Top Related