ownerkosskeb4

Tahiretal.
Agric&FoodSecur(2018)7:19https://doi.
org/10.
1186/s40066-018-0170-8RESEARCHEvaluationoflivestockfeedbalanceundermixedcrop–livestockproductionsysteminthecentralhighlandsofEthiopiaMeketeBekeleTahir1,2*,AshenafiMengistuWossen2andBerhanTamirMersso2AbstractBackground:Livestockproductionundersmallholdermixedcrop–livestock(MCL)productionsystemsisconstrainedbyfeedshortageandscarcityoflandonwhichtogrowfeed.
Livestockfeedsareobtainedfromdifferentsourcesincludingcropresidues(CR),grazinglands(GL),cropaftermath,fallowlandandpurchased.
Butthecontributionofthesefeedresourcesandtheextentoffeedbalanceatfarmlevelarenotquantitativelyexamined.
ThestudywasconductedtoassessthemajorfeedresourcesavailableandevaluatefeedbalancefortheprevailinglivestockinMCLsystem.
Methods:Feedtypesandamountavailable,livestockholdingsizeandfeeddemandwereestimatedfor159small-holderfarmersstratifiedintowealthstatus,whichwereselectedfollowingamultistagesamplingprocedure.
Thefeedbalancewasevaluatedasthedifferencebetweenrequirementsoflivestock(feeddemand)andamountofutilizablefeed(supply)peryearintermsofdrymatter(DM),metabolizableenergy(ME)anddigestiblecrudeprotein(DCP).
Results:Regardlessoffarmers'wealthstatus,CRmainlysourcedfromcereals,particularlybarley,contributedmorethanhalfoftheannualfeedsupply,followedbyGL.
ThecontributionofCRtototalfeedsupplysourcedon-farmandpurchasedcombinedwas55%.
Significantdifferencesinthesupply,demandandbalanceoflivestockfeedwereobservedacrosswealthgroupoffarmers.
Thewealthierhavehigherquantitiesoffeedsupplyanddemand,butsuf-feredmoreinfeedinsufficiency.
Overall,about51,19and38%annualfeeddeficitinDM,MEandDCPwereobserved,respectively.
Butwhentherateisconsideredbasedonfeedproducedon-farmonly,thedeficitworsenedandgoesupto60,34and52%inDM,MEandDCP,respectively.
Conclusions:Expansionofgrazinglandisnotapracticaloptiontoincreasefeedsupply.
Therefore,increasingfood-feedcropsproductionperunitarea,conservationofsurplusforages,strategicfeedingbasedonproductivityandtractionservicesoflivestock,purchaseoffeeds,andincreasinglivestockoff-takeduringtimeofscarcitywouldhelptocorrecttheobservedfeedshortage.
Moreover,refinementofthefeedbalanceanalysisatspecificnutrientslevelwouldbecompulsoryforeffectivestrategicinterventions.
Keywords:DebreBerhan,Drymatter,Metabolizableenergy,Digestiblecrudeprotein,Feedbalance,Feedsupplyanddemand,WealthstatusTheAuthor(s)2018.
ThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.
0InternationalLicense(http://creativecommons.
org/licenses/by/4.
0/),whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.
TheCreativeCommonsPublicDomainDedicationwaiver(http://creativecommons.
org/publicdomain/zero/1.
0/)appliestothedatamadeavailableinthisarticle,unlessotherwisestated.
BackgroundEthiopiaisanagrariancountryknownforpossessionofhugelivestocknumbers.
Thetotallivestockpopulationestimatedinmillionsduringthe2014/2015annuallive-stocksamplesurveyinthesedentaryareasofthecountrywasabout56.
7cattle,29.
3sheep,29.
1goats,2.
0horses,7.
4donkeys,0.
4mules,1.
2camelsand56.
9poultrywithtotaltropicallivestockunit(TLU)of52.
9,exclud-ingthenon-sedentarythreezonesofAfarandsixzonesofSomaliRegionofthecountry[1].
LivestockperformeconomicandsocialfunctionsbothatthenationalandOpenAccessAgriculture&FoodSecurity*Correspondence:meketebekele@gmail.
com2DepartmentofAnimalProductionStudies,CollegeofVeterinaryMedicineandAgriculture,AddisAbabaUniversity,POBox34,DebreZeit,EthiopiaFulllistofauthorinformationisavailableattheendofthearticlePage2of17Tahiretal.
Agric&FoodSecur(2018)7:19householdlevels.
Albeitvariationsamongdatasources,livestockcontribute15–17%ofnationalgrossdomesticproduct(GDP),35–40%ofagriculturalGDPand37–87%ofthehouseholdincomes[2].
Moreover,livestockcon-tributetoimprovethenutritionalstatusandincomegainofthepeoplebyprovidingmeat,milk,eggs,cheese,but-ter,etc.
andcommodities,suchasliveanimals,hidesandskinsforhomeuseandexport,andavertrisksintimesofcropfailures[1,2].
Despitethehugenumbersofpossession,thecurrentcontributionoflivestocktotheproducersandtothenationaleconomyisdismaltoitssize.
Ithasincreas-inglybeenunabletomeetthedemandsfortherapidlygrowingpopulation[3–5].
Amongthemanyfactorsthatcouldexplainthisdisproportionateroleofthesectoroftenmentionedaretheinadequatequantityandqualityoffeedsupplythroughouttheyeartosatisfytheannualdemandoflivestock[6,7].
Smallholderfarmersinthemixedcrop–livestock(MCL)systemskeepsomeformoflivestockinconjunc-tionwithcropproduction.
However,bettersoilsareallo-catedforfood-feedcrops,whereasmountainous,sloppyandlessfertilemarginallands,whicharenaturallyvul-nerableforsoilerosionandlanddegradation,areleftforlivestockgrazing.
Eventually,livestockbecamemoredependentoncommonfeedresourcesderivedfromlowbiomassproducingfood-feedcropsandpoorlyman-agedgrazinglands(GL).
Inorderlivestocktoexpresstheirproductivepotentialsandincreaseproductivity,theavailablefeedresourcesshouldmatchtheirdemandsfordrymatter(DM)andnutrients.
However,frequentlylivestockareexposedtoseasonalfeedshortagesbothinquantityandquality,especiallyduringthedrysea-son[8].
Moreover,thenutritivequalityofnativepastureislowespeciallyindryseasonanditismuchworseforcropresidues(CR)owingtothelowercontentofdigest-iblenutrients[9].
Furthermore,thecontinuingtrendsofexpansionincroplandcultivationattheexpenseofGLintheMCLsystemshaveresultedinshrinkageoftheareaandproductivityofGLandreversedtheproportionalcontributionofCRforfeedupward.
Theneedtoincreasebothcropandlivestockproduc-tionfromtheexistingresourcesrequiresresponsiveactionthroughouttheproductionsystem.
Uponexam-iningthelivestockfeedresourcesandfeedbalanceatsmallholderfarmlevel,Kassaetal.
[10]suggestedthepossibleuseoflivestockfeedbalanceaspotentialindi-catortoassesssustainabilityofthefarms.
However,theextentandpersistencyoffeeddeficitinrainfedMCLsystemsinthehighlandsofEthiopiadescribedinseveralpreviousreports[6,9–12]areaformidablechallengeforsustainedlivestockproduction.
AnegativefeedbalanceintheMCLsystemdisruptstheinteractionsbetweenthesystemcomponents,impairslivestockperformanceandcompromisesthepotentialrolesoflivestockindriv-ingtheeconomicdevelopment.
Forinstance,Kassaetal.
[10]reportedthatbetter-offandmediumwealthgroupsofsmallholdermixedfarmersdidnotproduceenoughfeedmorethanthepoorgrouptosupporttheirlive-stockintheHararhighlandsofeasternEthiopia.
How-ever,itisanticipatedthattheanalysisatfarmscalethatconstitutesthemajoravailablefeedsandthelivestockresourceswouldunveiltheprevailingstatusandhintstrategiestoaddressfeedshortageproblemsencounteredinthemixedfarmingsystems.
InthestudyarealikemostotherMCLsystemsareasofEthiopia,theongoinglandusechangefromgrazinglandtootherlandusesmainlytocroplandresultedinshortageofgrazinglands.
Ontheotherhand,feedobtainedasaby-productfromcroppro-ductiononconvertedlandfromgrazingtocroplandisnotlikelytocompensatethefeedsupplyduetotheinher-entlylowfeedvalueofcropresidues.
Moreover,feedfromotheralternativefeedsourcesislimited.
Thepre-sentstudywasaimedatassessmentofthepotentialfeedresourcessupplyfortheprevailinglivestockfeeddemandandevaluatetheannualfeedbalanceintheMCLsystem.
MethodsDescriptionofthestudyareaThestudywasconductedinDebreBerhanmilkshedintheAmharaNationalRegionalState,centralhighlandsofEthiopia,130kmawaytotheNortheastofAddisAbaba.
Theareaislocatedbetween9°30′and9°50′latitudesand39°20′and39°44′longitudes(Fig.
1).
Theelevationrangesfrom2840to2943masl[13].
Themeanannualminimumandmaximumtempera-turesaveragedbetween2000and2014yearsare6.
7and19.
9°C,respectively.
Themeanannualrainfallwithinthespanofthesameyearsis1026mmwithapoten-tialevapotranspirationof1396mm.
Rainfalldistribu-tionisbimodal,usuallythelongrainslastfromJunetothebeginningofSeptember,andtheperiodoftheshortrainsfallsbetweenFebruaryandMay.
About85.
5%oftherainfallsbetweenJuneandSeptemberwhichisthemaincroppingseason(Fig.
2).
Mostoftheareaiscoveredbymoderatelyandpoorlydrainedsoils,predominantlyblackVertisol[13,14].
RainfedMCLfarmingisthedominantsystemcarriedoutprimarilytomeetthesubsistencerequirementsformostofthefarmerswiththeirfamiliesinthestudyarea.
Small-scaleirrigatedfarmingislimitedtofewfarmerswithsmallpatchyareasalongBeressaRivermainlyforvegetables.
Theprincipalrainfedfood-feedcropsincludebarley,wheat,fababeanandfieldpea.
ThecerealswhichcoveredthelargestportionofcroppedareasarethemajorsourcesofCRforlivestockfeeding.
CultivationsPage3of17Tahiretal.
Agric&FoodSecur(2018)7:19ofteff(Eragrostisteff),lentils,chickpea,oats,linseedandvegetablesareintermittentonsmallplotsofcrop-land.
NaturalGLandCRarethemajorlivestockfeedresources,whilefallowland,cropaftermathgrazingandconcentratesareoccasionallyusedfeeds[8,15].
Thelive-stockspeciesrearedincludecattle,sheep,goats,donkeys,horses,mulesandpoultry.
Cattleproductionwithindig-enousandcross-bredanimalspredominatesthelivestockproductionfollowedbysheepproduction[8].
WealthstatusclassificationcriteriaThesettingofcriteriaforwealthstatuswasmadeincon-sultationwithdistrictexperts,developmentagentsandconfirmedbasedontheperspectivesoflocalfarmersduringgroupdiscussions.
Multiplecriteriafocusedonphysicalownershipofkeyassetsandtheiranticipatedvaluesatthetimeofthestudywereusedratherthanpre-cariousannualcashincome.
Ownershipofhouseswithcorrugatedironorthatchedgrassroofs,numberandtypesoflivestock,areaoflandandthecapacityofthefarmerstosatisfyannualhouseholdbasicneedswerethemajorfocuses.
Nonetheless,settinganabsolutecut-offpointtoeachcriterionwasnotpossible,andanoverlapintherangeofvaluesforthesetcriterionwasevident.
Insteadoffixingthejudgmentbasedonthevalueofasin-glecriterion,thecontributionofthewholewasassessedtogethertogroupafarmerunderoneofthethreewealthcategories(better-off,mediumandpoor).
Thedescrip-tionsofeachcriterionaresummarizedinTable1.
SurveydesignanddatacollectionSamplingproceduresandsamplesizeAmultistagesamplingprocedurewasemployedtoselectsamplefarmersengagedinMCLfarminganddeliverfluidmilktothenearbymilkcollectioncentres(MCC).
Fig.
1MapofEthiopiashowingthelocationofthestudyarea-50510152025050100150200250300350JanFebMarAprMayJunJulAugSepOctNovDecTemperature(0C)RainfallandET0(mm)RainET0TminTmaxFig.
2Averagemonthlyrainfall,potentialevapotranspiration(ET0),maximum(Tmax)andminimum(Tmin)temperaturesatDebreBerhanweatherstation(monthlydataaveragedfrom2000to2014)Page4of17Tahiretal.
Agric&FoodSecur(2018)7:19DebreBerhanmilkshedwaspurposelyselectedbasedonitsaccessibilityandpotentialrepresentativenessoftheMCLsystemincentralhighlandsofEthiopia.
From48villagesinthemilkshed,sixrepresentativeMCC(Wush-awushign,Angolela,Kormargefia,Kebele01,Kebele07andKebele09)wereselectedusingsimplerandomsam-plingprocedure.
FarmerssamplesizewasdeterminedusingG*power3.
1.
7softwareconsideringthefarmerwealthgroupasfixedeffectandassuming0.
25effectsize,α=0.
05and80%powerofthestatisticaltest[16].
ListoffarmersobtainedfromtheselectedMCCwascategorizedintothreewealthgroupstoformthesamplingframe.
Thenumberofsamplefarmersineachwealthgroupwasbasedonthe'probabilityproportionaltosize'samplingtechnique[17].
Finally,atotalof159farmers(50inthepoor,58inthemediumand51inthebetter-offwealthgroups)wereselectedusingsystematicrandomsamplingwithwhomthequestionnaireswereadministered.
DatacollectionAsocio-economicsurveyusingpretestedsemi-struc-turedquestionnaireswascarriedoutusingafacetofaceinterviewin2014and2015.
Thequestionnairescovereddataon:householddemographiccharacteristics,landandlivestockownership,livestockspeciescompositionandherdstructure,feedtypes,areaofprivateandcom-munalGL,typesoffood-feedcropsproduced,areacul-tivated,inputusedandcropyieldatfarmlevel.
FocusgroupdiscussionswereconductedwithfarmersateachoftheselectedMCC.
Atotalof40farmers(6–8perses-sion)representingwealthgroupshaveparticipatedinsixsessions.
Knowledgeablefarmersfromallwealthgroupswereselectedanticipatinganeffectivecommunicationbetweenthemoderatorandwithinthemselves.
Thesec-ondarydatawereextractedfrompreviousstudiesandinformationdocumentedatagriculturaldevelopmentandresearchoffices.
LocalclimatedatawereobtainedfromDebreBerhanagriculturalresearchcentre.
Dataenumeratorsrecruitedfromagriculturaldevelopmentagentsweretrainedtoassistonprimarydatacollectionduringfacetofaceinterviewandgroupdiscussions.
Estimationofannualfeedavailability(supply)QuantityoffeedDMavailableperyearwasestimatedfromthemajorfeedresourcesincludingCR,cropafter-math(stubblethatremainafterharvest)andGL.
CropresiduesDMobtainedwasderivedfromgrainyield,har-vestindicesandareaofcroplandcultivated.
Conversionfactorsderivedfromharvestindices,1.
5t/haforwheatandbarleyand1.
2t/haforfababeanandfieldpea[18],wereusedtoestimatecropresiduesfromgrainyields.
TheamountofCRDMcollectedperyearbyanindividualfarmerwasquantifiedbasedonthesizeofcroplandplotallocatedforgrowingaparticularcroptypeduringthecroppingseason.
Giventhefeedshortageandfarmers'prioritytotheuseofCRinthehighlandMCLsystems,itisassumedthatabout90%oftheCRusedasfeedand10%forotherpurposesandwastage[19].
AvailableDMfromcropaftermathgrazingoncroplandwasestimatedusingconversionfactorof0.
5tonDM/haperyear[20]andtheareacultivated.
AvailablefeedDMfromGLisestimatedbytakingtheprivateandcommunalownershippatternintoaccount.
FarmersareeligibletousewholeofavailablefeedontheirentitledprivateGL,butcanonlysharecertainamountoffeedfromopenlyaccessiblecommunalGL.
Itwasassumedthattheamountofsharefromcommunalgrazingisafunctionoflivestockdensity,whichisascribedtothesizeoflivestockownershipthathadaccesstousethiscommunalresource.
Alivestockdensityof14.
8TLU/haderivedbasedondataonthesizeTable1Descriptionofwealthstatusgroupingcriteriageneralizedbasedonfarmers'perspectivesinthemixedcrop–livestocksystem,centralhighlandsofEthiopiaNo.
CriteriaWealthcategoriesBetter-offMediumPoor1Croplandholding(ha)>1.
501–2.
50.
25–1.
52Numberofploughingoxen>2≥2≤13Numberofmilkingcows(localandcrossbreds)>2mostofcrossbredsAtleast2,fewcrossbreds<2andusuallylocalbreeds4Numberofsheep(adults)10–50ormore5–20≤105Numberofequines(adults)Amule,ahorseanddonkeysAtleasthorseanddonkeysNomule&horsemaybedonkey6RoofofhousesCorrugatedironsheetYesYes(often)NoGrassthatchedYesYesYes7AnnualfoodproductionandsupplytosustainhouseholdEnoughwithmoresurplusEnoughbutmeagresurplusNotenough,inneedofsupportPage5of17Tahiretal.
Agric&FoodSecur(2018)7:19oflivestockandGLareawasusedtoallocatethecommu-nalgrazingtoeachlivestockownerrelativetolivestockpossession.
ThetotalareaofprivatelyownedandpartofthecommunalGLallocatedwasconsideredtoestimatetheavailablefeedDMfromGLperhouseholdperyear.
ThefeedDMproductivityonGLwasestimatedbasedonmultiplierof2t/haestablishedfrompreviousGLcondi-tionandproductivitystudies[20].
Utilizationfactorof75%assuggestedby[21],forextensivegrassland,isusedtoquantifytheDMthatwouldbeutilizedbylivestock.
Thequantitiesofmetabolizableenergy(ME)anddigest-iblecrudeprotein(DCP)offeedresourceswerecalcu-latedbasedontheinvitrodigestibilityoforganicmatterindrymatter(IVDOMD)andcrudeprotein(CP)con-tentsofeachfeedtypereportedby[9]inthestudyarea.
Thefollowingequationswereusedtoestimatetheannualenergyandproteinsupplyatfarmlevelinrelationtothetypeoffeedresourceandamountobtainedperyear[22,23].
Theseasonalavailabilityoffeedresourceswasassessedbasedonfarmers'judgmentandscoresgivenforapar-ticularfeedtypeineachmonththroughouttheyear.
Availabilityoffeedovertheyearwasscoredonascaleof0–10,where10=excessfeedavailable,5=adequatefeedavailableand0=nofeedavailable[24].
Estimationoflivestockfeedrequirements(demand)LivestockholdingsperhouseholdwereaggregatedintoTLUconsideringtheannualaveragelivestockownershipderivedbasedonthenumberofanimalsatthebegin-ningandendofthestudyyear.
Thiswasdonetotakeintoaccounttheannualinflowandoutflowdynamicsoflivestockatfamerlevel.
Species-specificTLUconversionfactorsof0.
7forcattle,0.
1forsheepandgoats,0.
5fordonkeysand0.
8forhorseswereused[25].
Thedrymat-terdemand(DMD)wasestimatedbasedontheexpecteddailydrymatterintake(DMI)suggestedforthestand-ardTLUof250kgat2.
5%ofthebodyweight,whichisequivalentto6.
25kg/dayor2280kg/year[25].
Compa-rablerateofDMIwasalsosuggestedby[26].
TheMEandDCPrequirementsformaintenancewerecalculatedaccordingtothedailyaveragerecommendationsgivenby[26].
Basedonmetabolicbodyweight,118.
0,93.
0and103.
45kcalofME/Wkg0.
75and2.
86,1.
72and2.
51gofDCP/Wkg0.
75perdayforcattle,sheepandgoats,respec-tively,wereusedformaintenance[26].
Accordingly,thedailynutrientrequirementtablesformaintenancerecommendedforruminantsindevelopingcountriesby[26]wereusedtoestimatetheannualMEandDCPMEMJ/kgDM=0.
015IVDOMDg/kg;DCPg=0.
929CPg3.
48.
requirementsforcattle,sheepandgoats.
Inthecaseofdonkeysandhorses,thedailyMEandDCPmaintenancerequirementsof14.
9and27.
6MJand0.
18and0.
37kgrecommendedbyMcCarthyascitedin[9]wereused,respectively.
Then,farmlevelDM,MEandDCPrequire-mentsformaintenanceperyearwereextrapolatedrela-tivetothelivestockownershipperhousehold.
LivestockfeedbalanceLivestockfeedbalanceatindividualfarmerlevelovertheentireproductionyearwasdeterminedasthedifferencebetweentheannualfeedDM,MEandDCPsupplyesti-matedfrommajorfeedresourcesandtheannualfeedDM,MEandDCPdemandsfortheannualaveragelive-stockholdingoffarmers.
StatisticalanalysisMeans,standarddeviationsandpercentageswereusedtodescribevariablesobservedamongfarmersstratifiedintowealthstatus.
Tocomparethedifferencesacrossfarmerwealthgroupsintermsofharvestedgrainyields,netfeedsupply,netdemandandfeedbalance,aone-wayanalysisofvariance(ANOVA)wasused.
Theone-wayANOVAmodelisgivenby:whereYijisthejthobservationintheithwealthgroup,iisthecommoneffectforthewholewealthgroup,Fiistheeffectoftheithwealthgroupandεijistherandomerrorassociatedwiththejthobservationintheithwealthgroupassumedtobenormallyandindependentlydis-tributed,withmeanzeroandvarianceσ2ε,idesignatesthewealthgroup,andjdenotesaspecificobservation.
Thetestsweredoneat95%levelofconfidence(α=0.
05).
Tukeys'HSDmeancomparisonprocedurewasusedtotestmeandifferences.
TheanalysiswascarriedoutusingSPSSversion23.
0statisticalsoftware[27].
ResultsFarmhouseholdsandkeyfarmresourcescharacteristicsDescriptionsoffarmhouseholdsandkeyfarmresourcescharacteristicswithrespecttofarmerwealthgroupsaresummarizedinTable2.
Averagefamilysizeofthebetter-offandmediumwealthgrouphouseholdswassomehowcomparable,butforbothwealthgroupsitwashigherthanthepoorfarmers.
Thefamilylabourforcefollowedthesametrendoftrajectoryasthatofaveragefamilysize.
Incontrast,agedependencyratiowashigherinthepoorthanboththemediumandbetter-offwealthgroups.
Themagnitudeoflivestockandlandholdingsmatchedthewealthstatusgradientoffarmers,exhibitingadecreas-ingtrendfrombetter-offtopoorwealthgroups.
Live-stockholdingaggregatedintermsofTLUforthewholeYij=i+Fi+εij,Page6of17Tahiretal.
Agric&FoodSecur(2018)7:19andgrouped/disaggregatedintospecieswashigherinthebetter-offfollowedbymediumandpoorfarmers,respectively.
Thesizeofcultivatedcroplandthatincludescroplandrighteouslyentitledandcroplandtemporarilyacquiredbasedonlocalleaseagreementsbetweenfarm-ers,GLandthetotallandholdingssimilarlyreflectedtheimageoffarmers'wealthstatusconsistentlybeinghigherforthewealthier.
FoodfeedcropsproductionDescriptionsofcultivatedcropswhicharethestaplehumanfoodandmajorsourcesofCRforfeedinglive-stockarepresentedinTable3.
Thesametablealsopre-sentsthecomparisonsofmeangrainyields(ton)ofthemajorcropscultivatedperhouseholdbetweenfarmerwealthgroups.
Thetrendsinthesizeofplotofcroplandallocationtoaparticularcroptypeandproportionoffarmerswhohavegrownthecropduringthecroppingseasonweresimilarirrespectiveoftheirwealthstatus.
Barelywasthemostextensivelygrownfood-feedcropfollowedbyfababean,wheatandfieldpea,respectively,whichismanifestedbythelargersizeofcroplandplotallocatedforbarleygrowingthanfortheothercroptypesanditwascroppedbyeverywealthgroupoffarmers.
Thequantityofbothbarleyandwheatgrainyieldharvestedperhouseholdwassignificantlyhigherforbetter-offthanbothmediumandpoorfarmerwealthgroups(P<0.
001).
Thegrainyieldoffababeanharvestedperhouseholdbythebetter-offandmediumwealthgroupswassignifi-cantlyhigherthanthepoorgroup(P<0.
001).
However,nosignificantdifferenceinfieldpeagrainharvestedwasobservedamongthewealthgroups.
TypesandavailabilityoffeedresourcesEstimatedfarmlevelannualfeedsupplyfromdifferentsourcesintermsofDM,MEandDCPispresentedinTable4.
On-farmproducedfeedresourceswerederivedfromfood-feedcropsgrown,GL,cropaftermathandfal-lowland.
Inaddition,livestockfeedsweresourcedoff-farm,thoughthequantitiesprocuredweredependentonthecapabilityoffarmerstoafford.
Thewealthierwererelativelymorecapabletoaffordconsiderableamountofadditionalpurchasedfeedsources(hayandsomecon-centrates)off-farm.
Correspondingtotheamountofgrainyieldharvested,thequantityofCRobtainedfromindividualcroptypesandaggregatedwerehigherforthewealthiergroupoffarmers.
Moreover,thequanti-tiesofDM,MEandDCPobtainedfromon-farmfeedsourcesincludingGL,fallowlandandaftermathgrazingwerelowercomparedtotheamountattainedfromtotalCR;however,thetrendandextentofdisparitiesbetweenfarmergroupshavemirroredtheirwealthstatus.
Livestockfeedsupply,demandandfeedbalanceComparisonsoftheannualfeeddemand,supplyandbalancequantifiedonDM,MEandDCPbasisperfarmhouseholdacrosswealthgroupsoffarmersarepre-sentedinTable5.
ThedemandoflivestockestimatedforexpectedannualintakeofDM,MEandDCPwassignificantlydifferent(P<0.
001)acrosswealthgroupoffarmers,consistentlyincreasingalongwiththewealthgradientoffarmers.
Theaggregatedannualsupplyobtainedon-farmfromdifferentsources,aswellasthetotalannualsupply(on-farmobtainedcombinedwithpurchasedfeeds),differssignificantly(P<0.
001)acrossTable2Householdsdemographicandkeyfarmresourcescharacteristicsinthemixedcrop–livestocksystem,centralhighlandsofEthiopiaNnumberofrespondents;numbersinparenthesisarestandarddeviations(SD);ADRagedependencyratio;TLUtropicallivestockunit;1TLU250kgliveweight;ADRthedependencyratiorelatesthenumberofchildren(0–14yearsold)andolderpersons(65yearsorover)totheworking-agepopulation(15–64yearsold)HouseholdsandfarmresourcescharacteristicsWealthgroupsTotal(n=159)Poor(n=50)Medium(n=58)Better-off(n=51)Familysize4.
9(2.
0)6.
2(1.
5)6.
4(1.
6)5.
9(1.
8)Labourforce(inadultequivalent)2.
5(1.
2)3.
5(1.
1)3.
8(1.
1)3.
3(1.
3)Agedependencyratio(ADR)1.
0(0.
6)0.
7(0.
5)0.
6(0.
5)0.
8(0.
6)Livestockholding(TLU)5.
1(1.
2)8.
2(2.
0)12.
0(2.
4)8.
5(3.
4)Cattle3.
0(0.
9)4.
6(1.
2)6.
4(1.
4)4.
7(1.
2)Sheepandgoats0.
9(0.
5)1.
7(0.
8)2.
7(1.
2)1.
8(1.
1)Equines1.
2(0.
7)1.
9(1.
0)2.
9(1.
0)2.
0(1.
1)Croplandentitled(ha)1.
1(0.
5)1.
3(0.
6)1.
8(0.
6)1.
4(0.
6)Fallowland(ha)0.
1(0.
1)0.
1(0.
2)0.
3(0.
3)0.
2(0.
3)Cultivatedandgrazingland(ha)2.
3(0.
4)3.
2(0.
7)4.
1(0.
7)3.
2(1.
0)Croplandcultivated(ha)1.
6(0.
4)2.
2(0.
6)2.
5(0.
6)2.
1(0.
6)Grazingland(ha)0.
7(0.
2)1.
0(0.
4)1.
6(0.
5)1.
1(0.
5)Page7of17Tahiretal.
Agric&FoodSecur(2018)7:19wealthgroupoffarmers.
Similartothedemands,thewealthierobtainedmorequantitiesofDM,MEandDCPfromfeedssourcedon-farmandfromthetotalfeedsup-plyincludingtheadditionalpurchasedfeeds.
TheannualbalanceofDM,MEandDCPatfarmhouseholdlevelassessedbasedontheon-farmfeedsupplyandtotalfeedsupplythatincludedpurchasedfeedswassignificantlydifferent(P<0.
001)acrosswealthgroupsoffarmers,exceptthatthebalanceinMEwasnotsignificantlydiffer-entbetweenthepoorandmediumwealthgroups.
How-ever,inallcasesthewealthierencounteredmoredeficitthanthepoorgroupoffarmersinthequantitiesofDM,MEandDCPbalanceattained.
ThedegreeatwhichthedemandsforDM,MEandDCPwerefulfilledornotwithrespecttotheestimatedquan-titiesformaintenancerequirementoflivestockatfarmhouseholdlevelisillustratedinFig.
3.
ThemagnitudeofinsufficiencyinfeedDM,MEandDCPsupplyforlive-stockwasunlikebetweenthestratifiedwealthgroupsoffarmers.
Thebetter-offgrouphadmoreinadequatesupplyfollowedbythemediumandpoorinorderofimportance.
WhenthesupplyofDM,MEandDCPwasobservedatfarmerlevelinaparticularwealthstatusgroup,theDMsupplyshortagewasmorepronouncedfollowedbyDCPandMEshortagesinorderofthemagnitudeofinsuffi-ciency.
Itwasalsoobservedthatacquiringadditionalpur-chasedfeedshelpedtoreliefpartofthedeficienciesintheDM,MEandDCPsupplydifferentlybetweenthepoorandwealthiergroupoffarmers.
ThepoorfarmergroupwasabletoreducetheshortageintheDM,MEandDCPsupplyby6,9and9%,respectively,whereasthemediumandbetter-offwealthgroupshavereducedtheDMandMEshortageequallyby10and15%,respectively.
Thedefi-ciencyinDCPwasreducedby15%amongthemediumandby16%amongthebetter-offwealthgroups.
Theover-allsupplyshortageinDM,MEandDCPinthefarmingsystemwasmitigatedby9,14and14%,respectively,duetoadditionalpurchasedfeedsacquired.
TheproportionoffarmersunderwealthstatusgroupswhowereinapositivebalancearepresentedinTable6.
Overall,veryfewfarmerswereabletosourcesufficientfeedtosatisfythemaintenancerequirementsoflivestockthroughouttheyear.
RelativelymorenumberoffarmersfromthepoorwealthgroupwereinpositivestateoffeedbalanceintheDM,MEandDCPsupply.
ThesituationwasbetterforMEirrespectiveoffarmers'wealthstatus,espe-ciallywhenfarmersacquireadditionalpurchasedfeeds.
Basedonthefeedobtainedfromon-farmonly,noneinanywealthgroupwereinapositiveDMbalance,butlessthan2%,allofthemfromthepoorfarmersgroupwhenpurchasedfeedswereconsidered.
Generally,thepropor-tionoffarmersinapositivefeedbalanceforDM,MEandDCPwerecontrarytotheirwealthstatus,thewealthiersufferedmorethanthepoorinfeedinsufficiency.
Table3Areaofplotsofcroplandallocated,proportionoffarmerswhogrowthespecificfood-feedcroptypesandgrainyieldsperhouseholdinthemixedcrop–livestocksystem,centralhighlandsofEthiopiaValuesinparenthesesindicatestandarddeviations(SD)nnumberofrespondents,hahectare,tton,hhhouseholda,b,cMeansinarowwithdifferentsuperscriptsdiffersignificantly(P<0.
05),NSnotsignificantatthe0.
05probabilitylevel***Significantat0.
001probabilitylevelAreacultivated,grainyieldandproportionoffarmerswhogrowWealthgroupsTotal(n=159)Poor(n=50)Medium(n=58)Better-off(n=51)Landallocated(ha)Barley0.
9(0.
3)1.
1(0.
6)1.
2(0.
4)1.
1(0.
5)Wheat0.
3(0.
2)0.
3(0.
2)0.
5(0.
3)0.
4(0.
2)Fababean0.
4(0.
2)0.
6(0.
3)0.
7(0.
4)0.
6(0.
3)Fieldpea0.
1(0.
1)0.
1(0.
1)0.
1(0.
1)0.
1(0.
1)Farmerswhogrow(%)Barley100.
0100.
0100.
0100.
0Wheat78.
093.
1100.
090.
6Fababean100.
0100.
092.
297.
5Fieldpea20.
043.
149.
037.
7Grainyield(t/hh)Barley1.
7(0.
7)a2.
1(1.
0)ab2.
5(0.
9)c2.
1(0.
9)***Wheat0.
5(0.
4)a0.
7(0.
4)ab1.
2(0.
7)c0.
8(0.
6)***Fababean0.
6(0.
3)a0.
9(0.
5)bc1.
0(0.
5)c0.
8(0.
5)***Fieldpea0.
1(0.
2)0.
2(0.
2)0.
2(0.
2)0.
1(0.
2)NSPage8of17Tahiretal.
Agric&FoodSecur(2018)7:19RelativecontributionoffeedresourcesProportionalcontributionsoflivestockfeedresourcesproducedon-farmonly(Fig.
4a)andcombinedwithpur-chasedoff-farmsources(Fig.
4b)tothetotalfeedDMobtainedperyearatfarmerlevelareillustratedamongdifferentwealthgroupsoffarmers.
Regardlessoffarm-ers'wealthstatus,theforemoston-farmfeedsourcethatcoveredmorethanhalfoftheannualfeedDMsupplywasCRderivedfrommajorfood-feedcrops(Fig.
4a).
Evenwhenthecontributionofpurchasedfeedresourceswasconsidered,shareofCRprevailed(Fig.
4b).
FollowingCR,naturalGL,cropaftermathandfallowlandcontributetothebulkofon-farmannualfeedsupplyinorderofimpor-tance(Fig.
4a),whereaspurchasedfeedresourcestakethesecondlevelofcontributionexceptforthepoorfarmersgroupwhereittakesthefourthlevel,whenthepropor-tionalcontributionoffeedresourceswasconsideredbasedontheon-farmproducedandoff-farmprocuredsourcescombinedonaggregatedatfarmlevelofthedifferentwealthgroups(Fig.
4b).
Thecontributionoffallowlandwasverylimitedastherewasscarcityofcropland.
AmongCRthatmakeupthelargestpartofthelivestockfeedresourcebase,cereals(barleyandwheat)contributemorethanlegumes(fababeanandfieldpea).
Moreover,theproportionalcontributionofbarleystrawwasthehighestfollowedbyfababeanandwheatwhichhavecomparablecontributions,whereasthecontributionoffieldpeawastheleast,whentheshareofindividualcropswasconsid-eredirrespectiveoffarmers'wealthstatus(Fig.
4a,b).
SeasonalfeedresourcesavailabilityTheseasonalavailabilityoffeedresourcesthroughouttheyearisillustratedinFig.
5.
Cropresiduesobtainedfromon-farmcropproductionwerethemajorfeedsources,whichwereabundantlyavailablefrommonthsofDecem-bertoMarchandsteadilydeclinefromApriltoNovem-ber.
Grazinglandseventhoughutilizedthroughouttheyear,feedavailabilityislimitedtothemonthsofAugusttoDecember.
Duringtherestofthemonths,feedsavail-ablefromgrazinglandweremeagre.
Availabilityoffallowlandgrazingcoincideswiththetrendofgrazinglands,butwithverylowcontributionfromFebruarytoAugust.
However,thecontributionoffallowlandandnaturalgrazinglandisincreasingfromSeptembertoNovember.
AvailabilityofcropaftermathgrazingfollowsharvestingperiodsofcropsgrownanditisshortlivedfromDecem-bertoFebruary.
MonthsfromApriltoAugustarecriticalperiodsoffeedshortagewhenfarmersusedmostofpur-chasedfeedresources.
Duringthesecriticalfeedshort-ageperiods,conservedhayandcropresidueswereused,preferentiallyforfeedinglactatingcowsandoxenusedforcroplandpreparation.
DiscussionHouseholdcharacteristicsandfarmresources:implicationsonlivestockfeedavailabilityTheoverallaveragefamilysizeof5.
86inthepresentstudyishigherthanthenationalaverageof5.
14reportedTable4FeedresourcetypesandestimatedquantitiesofDM,MEandDCPobtainedperyearperfarmhouseholdinthemixedcrop–livestocksystem,centralhighlandsofEthiopiaValuesinparenthesesindicatestandarddeviations(SD)nnumberofrespondents,DMdrymatter,MEmetabolizableenergy,MJmegajoules,DCPdigestiblecrudeprotein,kgkilogram,tton,hhhouseholdFeedresourcetypesWealthgroupsTotal(n=159)Poor(n=50)Medium(n=58)Better-off(n=51)DM(t/hh)Cropresidues3.
8(1.
2)5.
1(1.
4)6.
6(1.
9)5.
2(1.
9)Barleystraw2.
3(0.
9)2.
8(1.
3)3.
4(1.
2)2.
8(1.
2)Wheatstraw0.
7(0.
6)0.
9(0.
5)1.
6(1.
0)1.
1(0.
8)Fababeanstraw0.
7(0.
4)1.
2(0.
7)1.
5(1.
0)1.
1(0.
8)Fieldpeastraw0.
1(0.
2)0.
2(0.
2)0.
2(0.
2)0.
2(0.
2)Grazingland1.
1(0.
3)1.
6(0.
5)2.
1(0.
5)1.
6(0.
6)Fallowland0.
1(0.
2)0.
1(0.
3)0.
4(0.
4)0.
2(0.
4)Cropaftermath0.
6(0.
2)0.
8(0.
2)0.
9(0.
2)0.
8(0.
2)Purchased0.
7(0.
5)1.
9(1.
7)2.
5(2.
0)1.
7(1.
7)ME('000MJ/hh)Cropresidues28.
4(9.
3)38.
1(10.
8)48.
8(14)38.
5(14)Barleystraw18.
3(7.
4)22.
2(10.
4)26.
7(9.
5)22.
4(9.
8)Wheatstraw4.
6(3.
8)5.
9(3.
2)10.
3(6.
5)6.
9(5.
2)Fababeanstraw4.
7(2.
5)8.
7(4.
9)10.
4(7.
1)8.
0(5.
7)Fieldpeastraw0.
8(1.
7)1.
2(1.
6)1.
3(1.
5)1.
1(1.
6)Grazingland9.
0(2.
6)12.
6(3.
7)17.
2(4.
4)12.
9(4.
9)Fallowland0.
7(1.
5)1.
1(2.
6)3.
4(3.
4)1.
7(2.
9)Cropaftermath4.
4(1.
2)5.
7(1.
8)6.
5(1.
6)5.
6(1.
7)Purchased5.
0(3.
9)13.
6(12.
3)18.
6(14.
8)12.
5(12.
6)DCP(kg/hh)Cropresidues128.
8(42.
0)182.
7(51.
2)230(70.
8)180.
9(68.
7)Barleystraw68.
3(27.
7)83.
0(38.
8)100(35.
4)83.
8(36.
6)Wheatstraw18.
4(15.
1)23.
6(12.
5)41.
0(25.
7)27.
5(20.
7)Fababeanstraw35.
8(18.
6)66.
1(37.
1)78.
5(53.
7)60.
5(42.
7)Fieldpeastraw6.
4(13.
7)10.
1(12.
8)10.
5(11.
9)9.
1(12.
9)Grazingland70.
1(20.
4)98.
0(28.
7)134(34.
2)100.
8(38.
1)Fallowland5.
4(11.
5)8.
2(20.
4)26.
5(26.
3)13.
2(22.
2)Cropaftermath11.
4(3.
1)15.
0(4.
2)16.
9(4.
0)14.
5(4.
4)Purchased36.
4(28.
6)99.
3(90.
2)136.
0(108.
5)91.
3(92.
4)Page9of17Tahiretal.
Agric&FoodSecur(2018)7:19by[28],whichisclosertothepoorbutlowerthanthewealthiergroupoffarmersinthestudyarea.
ItisstillmuchhighercomparedtotheaveragefamilysizeforboththeAmharaRegionof4.
62andtheNorthShewaZoneof4.
45,wherethestudysiteislocated[28].
Thelowerfam-ilylabourforceavailabilityamongthepoorfarmingfami-liesthantheotherwealthgroupscouldbeexplainedbytheobservedsmallnumberoffamilysizecoupledwithhigheragedependencyratioattributedtothepresenceofmoreproductivelyinactivedependentfamilymembers,especiallychildrenunder15yearsofage.
Theagedistri-butionoffamilymembersandtheconsequentstateofagedependencyratio,particularlyamongthepoorfarm-ingfamilies,areinagreementwiththenationaldemo-graphiccharacteristicsreportedby[1],whichexplainsthatabout45%ofthepopulationareunder15yearsofage,reflectingthedominanceofyoungsectionofthepopulationwithlesscontributiontothefarminglabourforcerequirement.
Availabilityoffarmresourcessuchasfamilylabour,landandlivestockhasaninfluenceonthelevelofproductionandavailabilityofCRforlivestockfeed-ing.
Thelimitedavailabilityoffamilylabourforceinthepoorfamilygroupmayhaverestrainedtheircapacitytointensifyhumanlabouruseinthefarmingoperation,whichmaynegativelycontributetothecauseforlowerfarmproductivity.
Itinfluencesfarmerstoapplyadequatelabourinputforoptimumagronomicactivitiesliketill-agefrequency,timelylandcultivationandweedingthatarerequiredtoincreasecropbiomassproductivity.
Inaddition,farmingfamilieswithsmallsizeofcroplandownershiparehardlyabletodiversifyandincreasethecropgrainyieldandassociatedCRbiomassproductionduetolimitationofavailablecroplandandsubsequentallocationofsmallerplotsfortheintendedcroptypes.
Moreover,thesmallnumberoflivestockownership,especiallyoxenamongpoorfarmingfamilies,hasalim-itinginfluenceontheuseofneededanimallabourfortractionservices,suchastimelycultivationofcroplandwithoptimumtillagefrequencyrequiredforthecroptypegrown.
Onthecontrary,wealthierfarmingfami-lieswithrelativelybetteravailabilityofploughingoxenTable5Meanannuallivestockfeedsupply,demandandbalanceperhouseholdinthemixedcrop–livestocksystem,cen-tralhighlandsofEthiopiaValuesinparenthesesindicatestandarderrors(SE)nnumberofrespondents,DMdrymatter,MEmetabolizableenergy,MJmegajoules,DCPdigestiblecrudeprotein,kgkilogram,tton,hhhouseholda,b,cMeansinarowwithdifferentsuperscriptsdiffersignificantly(P<0.
05),NSnotsignificantatthe0.
05probabilitylevel***Significantat0.
001probabilitylevelFeedsupply,livestockdemandandfeedbalanceWealthgroupsTotal(n=159)Poor(n=50)Medium(n=58)Better-off(n=51)DMdemand(t/hh)11.
6(0.
4)a18.
7(0.
6)b27.
4(0.
8)c19.
3(0.
6)***DMsupply(t/hh)On-farm5.
6(0.
2)a7.
6(0.
2)b10.
0(0.
3)c7.
7(0.
2)***On-farmandpurchased6.
3(0.
2)a9.
4(0.
4)b12.
6(0.
4)c9.
5(0.
3)***Balance(t/hh)On-farm6.
0(0.
5)c11.
2(0.
7)b17.
4(0.
7)a11.
5(0.
5)***On-farm&purchased5.
3(0.
5)c9.
3(0.
7)b14.
8(0.
7)a9.
8(0.
5)***MEdemand('000MJ/hh)54.
9(1.
8)a86.
4(2.
7)b122.
8(2.
9)c88.
1(2.
6)***MEsupply('000MJ/hh)On-farm42.
5(1.
5)a57.
5(1.
9)b75.
9(2.
2)c58.
7(1.
5)***On-farmandpurchased47.
5(1.
7)a71.
0(2.
7)b94.
4(3.
3)c71.
1(2.
1)***Balance('000MJ/hh)On-farm12.
3(2.
5)c28.
9(3.
4)b46.
9(3.
1)a29.
5(2.
1)***On-farmandpurchased7.
3(2.
6)cb15.
4(3.
8)b28.
3(3.
5)a17.
0(2.
1)***DCPdemand(kg/hh)392.
2(14.
6)a630.
4(23.
5)b907.
1(21.
8)c644.
2(20.
2)***DCPsupply(kg/hh)On-farm215.
7(6.
8)a303.
9(9.
3)b407.
4(11.
8)c309.
4(8.
2)***On-farmandpurchased252.
1(8.
3)a403.
1(16.
3)b543.
4(21.
9)c400.
6(13.
3)***Balance(kg/hh)On-farm176.
4(16)c326.
5(24.
7)b499.
7(17.
3)a334.
9(15.
5)***On-farmandpurchased140.
0(16.
9)c227.
2(26.
4)b363.
7(20.
2)a243.
6(14.
6)***Page10of17Tahiretal.
Agric&FoodSecur(2018)7:19Fig.
3Percentagesofdrymatter(a),metabolizableenergy(b)anddigestiblecrudeprotein(c)supplyanddemandbalancesoffarmhouseholdscategorizedintowealthgroupsinthemixedcrop–livestocksystem,centralhighlandsofEthiopiaPage11of17Tahiretal.
Agric&FoodSecur(2018)7:19areabletocultivatetheircroplandadequatelyintimeandacquireadditionalcroplandforsharecroppingfromotherfarmersthroughlocalleaseagreementandattainmorefeedfromtheshareofCRattainedinpartorasawholedependingontheagreementwiththecroplandowners.
Insupportofthepresentstudy,severalstudiesreportedthatinadequatecroplandpreparationduetolimitedaccesstoproductionfactorssuchasland,humanandanimallabouraffectthefood-feedcropbiomasspro-duction[29–32].
ThewealthierarealsorelativelymoreefficientincollectinghayfromprivateGLandstoreforlateruseintimesofscarcityorpracticestrategicfeed-ingbyprovidingforthemoreproductiveanimalssuchasmilkingcowsandworkingoxen[8,9].
TypesandavailabilityoffeedresourcesThetypesandamountoffeedobtainedperhouseholddependonlanduse,sizeoflandownershipandcrop-pingpatternduringtheproductionseason.
Irrespectiveofwealthstatus,largeproportionoflandisallocatedforcropcultivationthanotherlandusetypesincludinggraz-ing,whichisinagreementwithreportsof[14,6].
Theincreasingexpansionoflandcultivationforcropproduc-tionattheexpenseofGLhasresultedinshrinkageoftheareaandproductivityofGL[14].
Consequently,CRgen-eratedfromproductionoffood-feedcropscomprisethelargestproportionoflivestockfeedresources.
Inagree-ment,[33]reportedtheprogressivedeclineofGLandtheuseofCRasthemajorsourceoflivestockfeedparticu-larlyduringthedryseasoninthemixedfarmingsystemoftheBalehighlandsofEthiopia.
Inaddition,[9]alsoobservedtheheavyrelianceoffarmersatDebreDerhaninthecentralhighlandsofEthiopiaonCRforfeedingtheirlivestock.
Besidesthesuitabilityofthelocalagro-climateforbar-leygrowing[8,14],thehigherquantityofbarleystrawcollectedisassociatedwiththepreferenceoffarmersandtheconsequentallocationoflargerplotsofcroplandforthebarleycropcultivation.
Apparently,theremain-ingsmallerplotsoflandaresharedforothercroptypes,whichcorrespondinglyleadstolesserquantityofstrawobtainedfromcroptypessuchaswheat,fababeanandfieldpea.
Inlinewiththepresentstudy,[33]explainedthevariationinthequantityandqualityofCRproducedcouldbeduetothecroppingintensitybesidesthevaria-tionsinaltitude,soiltypeandrainfallpatternswherethecrophasbeengrown.
Inrelationtothesizeofcroplandownershipandpar-celallocatedtoaspecificcroptype,highervolumeofbiomassisharvestedbybetter-offandmediumwealthgroupsthanpoorfarmers.
Thiscropbiomassproduc-tivityisaccordinglytranslatedintothehigheramountofCRharvestwhichreflectedthegradientoffarmers'wealthstatus.
Inadditiontotheplotofcroplandalloca-tion,theavailabilityoffamilylabourforce,sizeoflive-stockownedparticularlyoxen,naturalfertilityofthecroplandandothervariablefactorsassociatedwiththecapacityoffarmerstoaffordvariousinputsandmanagetimelyagronomicpracticesinfluencethevolumeofCRproduceperhousehold.
Moreover,theamountoffeedderivedfromthecropfieldintheformofcropaftermathgrazingisintrinsicallyinfluencedbythecroppingpatternandplotofcroplandallocation,aspartofthefeedrecov-ered,istheremainderoftheparentcropobtainedafterharvestinadditiontothedifferentweedsandotherher-baceousplantsgrownonthecropfield.
Inlinewiththis,[34]reportedthequantityandqualityofstubbleleftvar-iesconsiderablydependingontheharvestingtechniquesappliedforthevariouscropsandthegrassweedsgrownunderneaththemaincrop.
Therefore,itispredictablethatthewealthiercouldbeabletobenefitmorethanthepoorfarmergroupsfromcropaftermathgrazingwhichisattributedtothelargersizeofcroplandownership.
Table6Numbersandproportionsoffarmerswithpositivelivestockfeedbalanceinthemixedcrop–livestocksystem,centralhighlandsofEthiopiaValuesoutsideandinsideparenthesesindicatefrequencies(n)andpercentages(%),respectivelynnumberofrespondentsFeednutrientsFeedsourcedfromWealthgroupsTotal(n=159)Poor(n=50)Medium(n=58)Better-off(n=51)DMOn-farm0(0.
0)0(0.
0)0(0.
0)0(0.
0)On-farmandpurchased2(4.
0)0(0.
0)0(0.
0)2(1.
3)MEOn-farm15(30.
0)3(5.
2)0(0.
0)18(11.
3)On-farmandpurchased23(46.
0)17(29.
3)7(13.
7)47(29.
6)DCPOn-farm2(4.
0)0(0.
0)0(0.
0)2(1.
3)On-farmandpurchased7(14.
0)7(12.
1)0(0.
0)14(8.
8)Page12of17Tahiretal.
Agric&FoodSecur(2018)7:19Fig.
4Relativecontributionsoffeedresourcesaobtainedon-farmandbobtainedon-farmcombinedwithpurchasedfeedstothetotalannuallivestockfeeddrymattersupplyinthemixedcrop–livestocksystem,centralhighlandsofEthiopiaPage13of17Tahiretal.
Agric&FoodSecur(2018)7:19ThesizeofavailableGLinthestudyareaissmallirre-spectiveofwealthstatus,thoughthewealthieraremorefavouredtoexploitavailablefeedresourcefromfreelygrazingthecommunallandduetoownershipofrelativelylargerlivestockherdsize.
Moreover,fallowlandsaslive-stockfeedsourcewereveryrare,asthereiscriticallandshortageforcropcultivationaswell.
Inagreement,[34]reportedthatitwasinthepastthatGL,hayandfallowlandgrazingwerethemainsourcesoffeed;however,atpresentthesituationhaschangedduetotherapidlyincreasinghumanpopulationandexpansionofcroppro-ductioninthegrazingareas.
Sameauthorassertedthatthereisverylittlethatlivestockcanforagefromthelim-itedcommunalgrazingduetocontinuousgrazingandthemismatchbetweenthefeedsupplypotentialofthelandandthenumberofanimalskept.
Consequently,cropresiduesareincreasinglybecomingthemainsourcesoflivestockfeedresourcesintheMCLsystems[34–36].
Otherseveralstudieshavealsodisclosedthat,inrecentyears,GLintheMCLsystemsaredevastatinglyover-stockedandthattheherbagebiomassproductivityisdecliningbothinquantityandquality[8,11,37–40].
Therefore,expansionofGLtoincreasefeedavailabilityisnotapracticaloptionastheincreasingtrendsofboththehumanandlivestockpopulationaregoingtoclaimmoreadditionallandfurtherforcropproduction,settlementandfeedsourcing.
Hence,strategicoptionstoimprovetheavailabilityofCRandotheralternativeand/oradditionalfeedresourcesshouldbeintervened.
Fromevidencescapturedinthepresentstudywhicharesupportedbyseveralotherfind-ings,mostofsmallholderfarmsintheMCLsystemsaredependentonCRasthemainsourceoffeedforlivestock.
Moreover,nowadaysstrawshavealsobecomeanimpor-tantsourceofrevenue[34]andusedforotherpurposesthatcompetelivestockfeedsupply.
References[41,34]suggestedclosecollaborationofcropandlivestocksci-entistsaimedatmultidimensionalcropimprovementtoincreasethequantityandfeedingvalueofCRwithoutnegativelyaffectingthegrainyieldandquality.
Inaddi-tion,Tolera[34]furthersuggestedtheintroductionofcovercropsthatcouldbegrazedaftershortrainysea-son,highyieldingforagecropswithpossibilityofmulti-pleharvestsperyearandmultipurposeleguminoustreesusedassupplementalgreenfoddercouldbeexploredtoalleviatethefeedshortagepertainingtotheexistingsitu-ation.
Inaddition,interventionstoincreasefeedbiomassperunitareaofGL,propergrazingmanagementandpurchaseofsupplementalfeedswheneverpossiblewouldbeamongtheoptionsthathelptoreducetheinfluenceoffeedshortageonlivestockhealthandproductivity.
Thefluctuationsintheseasonalavailabilityofmajorfeedresourcesaremainlyassociatedwiththerainfalldis-tributionandcroppingpatternoffood-feedcrops.
Cropresiduesareplentyduringthedryseason,whichisthe012345678910JanFebMarAprMayJunJulAugSepOctNovDecFeedAvailability(0-10scale)BarleystrawWheatstrawFababeanstrawFieldpeastrawCropresiduesNaturalgrazingCropstubblegrazingFallowlandPurchasedFig.
5Seasonalavailabilityoffeedresourcesasperceivedbyfarmersinthemixedcrop–livestocksystem,centralhighlandsofEthiopia.
Availabilityoffeedovertheyearonascaleof0–10,where10=excessfeedavailable,5=adequatefeedavailableand0=nofeedavailablePage14of17Tahiretal.
Agric&FoodSecur(2018)7:19periodofthreshingforharvestedcrops.
Availabilityandutilizationofcropaftermathgrazingisalignedwithhar-vestingperiodsofcropsinthedryseason.
Cropresiduesandhayarealsoconservedformitigatingfeedshortage,especiallyduringpeakperiodsofcroplandploughinginthewetseason.
Foragesfromgrazinglandsandfallowlandsareavailablefollowingthemain/longandshortrainyseasons,whenthemoistureissufficientenoughtosupportplantgrowth.
LivestockfeeddemandandfeedbalanceInthepresentstudy,livestockfeedbalanceevaluatedisgenerallyfoundtobenegativeirrespectiveoffarmers'wealthstatus,indicatingthecriticalfeedshortageinthesystemtofulfiltheDM,MEandDCPrequirementfortheprevailinglivestock.
Overall,onlyabout49,81and62%ofDM,MEandDCPrequirementsarefulfilled,respec-tively,withthecurrentamountoffeedsupplyfromon-farmproductioncombinedwithpurchasedfeeds.
SeveralpreviousstudyfindingssupportthepresentreportinthatintheMCLsystemwherelandislimited,thecon-tributionofCRisprominentandGLaredeterioratinginsizeandproductivity,andthelivestockproductionisperformedunderannualfeeddeficit.
Forinstance,Won-datiretal.
[9]reportedthataroundDebreBirhantheannualfeedsupplyonlysatisfies64%ofthemaintenanceDM81%MEand66%DCPrequirementsoftheanimalsperfarm.
StudybyAmsaluetal.
[11]inGummara-RibwatershedoftheLakeTanasub-basininAmharaRegionshowsthattheavailablefeedsourcesaddressonly72%oftheannualDMrequirementperhousehold.
Thesestudiesconfirmedthelong-standingimpedimentsforincreasinglivestockproductivityconsistentlyclaimedbysmallholderfarmersanddevelopmentpractitionersintheMCLsystems,whereseasonalfeedshortagebothinqualityandqualitycoupledwithinadequatesuccessfulinterventions,whichareremainedunabated.
ThisstudyandothersmentionedhereusedsameamountofaveragedailyDMIforastandardTLUof250kgat2.
5%oflivebodyweightsuggestedby[25],whichis6.
25kg/day,toarriveatanaggregateddemandforfeedDMathouseholdlevelperyear.
However,thedailyDMfeedintakeisaffectedbyvariousanimalandfeed-relatedfactors.
FAO[42]describedwherethereisgoodaccesstofeed,andhealthandwelfareareoptimum,theDMintakeforyoungruminantanimalsislikelytobe4%ofbodyweightanddeclineto2–2.
5%asanimalsaregettingmature.
Moreover,wherefeedsoflowdigestibil-ityarethemainoronlysources,voluntaryintakewillbesignificantlyinfluencedduetothelongertimeittakestoprogressthroughthedigestivetract,andasaresultlowdigestibilityisreflectedinlowerintakes.
LivestockfeedbalancereportedundersmallholdersintheMCLsystemisfrequentlynegative[10,43,44],andevensomanytimesdeclaredasbelowthemaintenancerequirements[6,9,11];however,infactlivestockcontin-uedtosurvive,reproduce,provideandperformtheirvitalfunctions,whereasthedifferentproductsandservicesobtainedcannotbeimaginedatmaintenancerequire-mentlevelletalonewhenanoverwhelmingannualfeeddeficitbelowmaintenanceisrevealed.
Hence,seriesofcriticalquestionscouldariseontheassumptions,meth-odsandvalidationsoflivestockfeedbalanceestimations,whichwerealsoiteratedbyKassaetal.
[10].
Forinstance,isthesuggesteddailyDMIamountat2.
5%oflivebodyweightislimitedtoonlysatisfythemaintenancerequire-ment,asdescribedinvariouslivestockfeedbalancestud-iesTowhatextentthesuggesteddailyallowancefulfilstheanimalrequirementDofeedtypesofanyqualityandformcanbeingestedatthesuggestedratebyanytypeand/orclassoflivestockthroughoutallseasonsoftheproductionyearIsthatamountoffeedcouldbeafforda-bledailybysmallholdersintheMCLsystemconsistentlythroughouttheproductionyearetc.
Thefeedbalanceresultsshouldbecautiouslyinter-preted,inthatthelivestockproductionsystemisrun-ninginanannualfeeddeficitdoesnotmeanthatthelivestockarecurrentlywithoutanybenefittooffer.
Itisnotunimportanteither,sinceitquantitativelyunveilsthepictureofwhatfarmersandexpertsclaimtobethemostimportantdeterminingfactorsoflivestockproduction,whicharefeedshortageandtheincreasingtrendoflandscarcityonwhichtoproducelivestockfeed.
Firstofall,thesuggestedratefortheDMIoffeeddoesnotindicatethequalityofthefeedsuppliedbuttheaveragequan-titythattheanimalscouldpotentiallyconsumedaily.
Inaddition,itdoesnotmeanthatwhateverisingestedisequallyavailabletoalltypesofanimals.
Thisisbecausethenutritionalrequirementofananimaldependsonsev-eralfactors:species,size,age,weight,physiologicalsta-tus(pregnancy,lactation,etc.
),levelofproduction(rateofgain,amountofmilkproduced,etc.
),generalhealth,amountofwork,weatherconditionandseason[19,42,45].
Secondly,thereisaneedtocloselyobservethedynam-icsoffarmelementsandsituationanalysisthroughouttheproductionyear.
Actually,thefeedbalanceestimateshowedthesituationsaggregatedonannualbasis,asifthefeedavailabilityandlivestockDM,MEandDCPrequirementsperdayaredistributedevenlythrough-outtheproductionyear.
However,alotofdynamicscanhappenintheseasonalfeedsourcesavailability,nutri-tionalcompositionandquality,andintheindividualani-malbodyandinthelivestockherdinthecourseoftheannualproductionprocess.
ItisanestablishedfactthatPage15of17Tahiretal.
Agric&FoodSecur(2018)7:19quantityandqualityoffeedavailablefromnaturalpas-tureareaffectedbyseason[10,35,43–46].
Feedsourcedfromcropproductionisalsoaffectedbytypeofcrop,croppingpattern,amountofinputusedlikefertilizersandagronomicpracticesapplied[36,47,48].
Differentfeedsourcesofvariedquantityandqualitymaybeavail-abletolivestockacrossdifferentseasons,whichcouldbeproducedon-farmorpurchaseddependingonthecapac-ityofproducerstoafford.
Theanimalsmayundergothephenomenonofcompensatorygrowth,i.
e.
loseweightduringperiodsoffeedscarcityandputonweight(gain)duringtimesofgoodfeedavailability.
Inthemeantime,livestockcoulddeliveranyformsofproductsandser-vicesduringperiodsoffeedavailabilityand/ormobilizewhatisdepositedduringgoodtimesoftheyear.
Atherdlevel,aninflowfromnaturalperpetuationandpurchaseofanimals,andanoutflowduetodeath,sellandslaugh-ter,andsharedrearingofanimalineithercasecouldallhappenwithintheyearfollowingthescarcityandavaila-bilityoffeedathouseholdlevel.
Hence,theseasonalfeedavailabilitycoupledwithfarmersfeedingpracticeandcopingmechanismoffeedshortage,theproductivitygainorlossandtheinflowandoutflowdynamicsoflivestockathouseholdlevelshouldbeconsidered,whileinterpret-ingthelivestockfeedbalance.
ConclusionsTheavailableannualfeeddrymatter,metabolizableenergyanddigestiblecrudeproteinsuppliesandrequire-mentsoflivestockdidnotmatchatpresent,inthestudyareacharacterizedbysmallholdermixedcrop–livestockproductionsystem.
Livestockfeedinsufficiencyunani-mouslyaffectsthewholegroupofsmallholderfarm-ersirrespectiveofwealthstatus.
However,thewealthiergroupoffarmerswhichownedlargersizeoflivestockcollectedmorequantityoffeedon-farmandacquiredadditionalfeedthroughpurchasefoundtosuffermoreintermsoffeeddeficitthanthepoorfarmers,whenlive-stockfeedbalancewasassessed.
ScarcityofGLduetoexpansionofcultivationonGLcoupledwithlowproduc-tivityofbiomassfromgrazingandfood-feedcropspro-ductionfurtherexacerbatethemismatchbetweenfeedsupplyanddemand.
ThetrendoflandscarcitytestifiesthatsmallholderlivestockproductionintheMCLsystemislikelytocontinuetorelyuponCR.
Cropresidueswhichareclaimedtohavesuboptimalfeedingvalueduetothehighfibrecontentandlowdigestibilitycomprisedthemajorportionofannualfeedsupply,andthustheinher-entlimitationsofCRshouldnotbeignored.
Farmerswhopromotedcross-breedingoflocalanimalswithbreedsoftemperateorigintoincreasemilkproductionarefoundunabletosupplysufficientquantityandqualityoffeedtoachievetheanticipatedlevelofproductivity.
Thepresentevidencesimpliedthatunlessthefeeddeficitisrelievedthroughincreasingtheavailabilityandqualityoffeedsandpracticingstrategicfeedingtheproductivityoflive-stockwillbemoreprofoundlyaffected.
AbbreviationsADR:agedependencyratio;CR:cropresidues;DM:drymatter;DMD:drymatterdemand;DMI:drymatterintake;ME:metabolizableenergy;MJ:megajoules;DCP:digestiblecrudeprotein;kg:kilogram;t:ton;hh:household;GDP:grossdomesticproduct;GL:grazinglands;masl:metresabovesealevel;MCC:milkcollectioncentres;MCL:mixedcrop–livestock;n:numberofrespondents;SD:standarddeviations;SE:standarderrors;TLU:tropicallivestockunit.
Authors'contributionsMBTdesignedthestudy,collectedandanalysedthedataanddraftedthemanuscript.
AMWandBTMsupervisedthedesignofthestudyanddataanalysisandrevisedthemanuscript.
Allauthorsreadandapprovedthefinalmanuscript.
Authors'informationMBTisalecturer(M.
Sc.
inanimalproduction)inDebreBerhanUniversityandaPh.
D.
candidateinanimalproductioninAddisAbabaUniversity.
AMWisanassociateprofessor(Ph.
D.
inanimalnutrition)intheDepartmentofAnimalProduction,CollegeofVeterinaryMedicineandAgriculture,AddisAbabaUniversity.
BTMisaprofessor(Ph.
D.
inanimalnutrition)andheadofAnimalProductionDepartmentintheCollegeofVeterinaryMedicineandAgriculture,AddisAbabaUniversity.
Authordetails1DepartmentofAnimalScience,CollegeofAgricultureandNaturalResourceSciences,DebreBerhanUniversity,POBox445,DebreBerhan,Ethiopia.
2DepartmentofAnimalProductionStudies,CollegeofVeterinaryMedicineandAgriculture,AddisAbabaUniversity,POBox34,DebreZeit,Ethiopia.
AcknowledgementsTheauthorsliketoacknowledgeAddisAbabaUniversityandDebreBerhanUniversityforfinancialsupport.
TheauthorsliketothankstaffoftheBosonaWoranaandAngolelaenaTeradistrictagricultureandruraldevelopmentoffices,developmentagentswhoassistedindatacollectionandfarmersofthestudyareafortheircooperation.
CompetinginterestsTheauthorsdeclarethattheyhavenocompetinginterests.
AvailabilityofdataandmaterialsThedatasetsusedand/oranalysedduringthecurrentstudyareavailablefromthecorrespondingauthoronreasonablerequest.
ConsentforpublicationNotapplicable.
EthicsapprovalandconsenttoparticipateNotapplicable.
FundingFundingforthisstudywasprovidedbyAddisAbabaUniversityandDebreBerhanUniversity.
BothUniversitieshadnoroleinthestudydesign,collection,analysisorinterpretationofthedata,writingofthemanuscript,orthedeci-siontosubmitthepaperforpublication.
Publisher'sNoteSpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpub-lishedmapsandinstitutionalaffiliations.
Received:24April2017Accepted:10February2018Page16of17Tahiretal.
Agric&FoodSecur(2018)7:19References1.
CentralStatisticalAgencyoftheFederalDemocraticRepublicofEthiopia(CSA).
Agriculturalsamplesurveyin2014/15.
Reportonlivestockandlivestockcharacteristics(PrivatePeasantHoldings).
AddisAbaba,Ethiopia;2015.
2.
GebreMariamS,AmareS,BakerD,SolomonA,DaviesR.
StudyoftheEthiopianlivecattleandbeefvaluechain.
ILRIDiscussionpaper23.
Nairobi,Kenya;AddisAbaba,Ethiopia:InternationalLivestockResearchInstitute(ILRI);2013.
3.
SwanepoelF,StroebelA,MoyoS.
Theroleoflivestockindevelopingcom-munities:Enhancingmultifunctionality.
Bloemfontein,SouthAfrica:TheTechnicalCentreforAgriculturalandRuralCooperation(CTA);2010.
4.
ShapiroBI,GebruG,DestaS,NegassaA,NigussieK,AbosetG,etal.
Ethiopialivestockmasterplan,roadmapsforgrowthandtransformation,acontributiontothegrowthandtransformationplanII(2015–2020).
Ethiopia:AddisAbaba;2015.
5.
AsresieA.
ContributionoflivestocksectorinEthiopianeconomy:areview.
AdvLifeSciTechnol.
2015;29:79–91.
6.
AssefaD,NurfetaA,BanerjeeS.
AssessmentoffeedresourceavailabilityandlivestockproductionconstraintsinselectedKebelesofAdami.
AfrJAgricRes.
2013;8:4067–73.
https://doi.
org/10.
5897/AJAR2013.
7096.
7.
YadessaE,EbroA,FitaL,AsefaG.
LivestockfeedproductionandfeedbalanceinMeta-RobiDistrict,WestShewaZone,OromiyaRegionalState,Ethiopia.
AcaResJAgricSciRes.
2016;4:45–54.
https://doi.
org/10.
14662/ARJASR2016.
072.
8.
HassenA,EbroA,KurtuM,TreydteAC.
LivestockfeedresourcesutilizationandmanagementasinfluencedbyaltitudeintheCentralHighlandsofEthiopia.
LivestResRuralDev.
2010;22:1–19.
9.
WondatirZ,MekashaY,WoutersB.
Assessmentofproductiveandreproductiveperformanceofdairycattlenexuswithfeedavailabilityinselectedperi-urbanareasofEthiopia.
JCellAnimBiol.
2011;5:308–15.
https://doi.
org/10.
5897/JCAB11.
037.
10.
KassaH,GibbonD,TamirB.
Useoflivestockfeedbalanceasapotentialindicatorofsustainabilityoftropicalsmallholdermixedfarms—prevail-ingknowledgegaps:acasestudyfromtheHararhighlandsofeasternEthiopia.
JSustainAgricLond.
2003;22:37–41.
11.
AmsaluT,AddisuS.
Assessmentofgrazinglandandlivestockfeedbal-anceinGummara-Ribwatershed,Ethiopia.
CurrAgricResJ.
2014;2:114–22.
https://doi.
org/10.
12944/CARJ.
2.
2.
08.
12.
TegeneN,DinkuG,MohammedB.
AssessmentofpotentialofnaturalpastureandotherfeedresourcesinsweetpotatoproductionsystemofShebedinoDistrict,SidamaZone,SNNPRS,Ethiopia.
IntJLivestProd.
2015;6:91–8.
https://doi.
org/10.
5897/ijlp2015.
0264.
13.
MollaA.
Farmers'knowledgehelpsdevelopsitespecificfertilizerraterecommendations,centralhighlandsofEthiopia.
WorldApplSciJ.
2013;22:555–63.
https://doi.
org/10.
5829/idosi.
wasj.
2013.
22.
04.
1196.
14.
AmsaluA,StroosnijderL,GraaffJD.
Long-termdynamicsinlandresourceuseandthedrivingforcesintheBeressawatershed,highlandsofEthiopia.
JEnvironManag.
2007;83:448–59.
https://doi.
org/10.
1016/j.
jenvman.
2006.
04.
010.
15.
BekeleM,MengistuA,TamirB.
Livestockandfeedwaterproductivityinthemixedcrop–livestocksystem.
Animal.
2017;11:1852–60.
https://doi.
org/10.
1017/S1751731117000416.
16.
FaulF,ErdfelderE,BuchnerA,LangA-G.
G*Power3:aflexiblestatisticalpoweranalysisprogramforthesocial,behavioral,andbiomedicalsci-ences.
BehavResMethods.
2007;39:175–91.
17.
AlamM,SumySA,ParhYA.
Selectionofthesampleswithprobabilityproportionaltosize.
SciJApplMathStat.
2015;3:230–3.
https://doi.
org/10.
11648/j.
sjams.
20150305.
13.
18.
KossilaV.
Theavailabilityofcropresiduesindevelopingcountriesinrela-tiontolivestockpopulations.
In:ReedJD,CapperBS,NeatePJH,editors.
Proceedingsofplantbreedingandthenutritivevalueofcropresidues,7–10December1987.
AddisAbaba,Ethiopia:InternationalLivestockCentreforAfrica(ILCA);1988.
p.
29–39.
19.
ToleraA,SaidA.
ProspectsforintegratingfoodandfeedproductioninWelayitaSodo,Ethiopia.
In:StaresJES,SaidAN,KategileJA,editors.
ThecomplementarityoffeedresourcesforanimalproductioninAfrica.
ProceedingsofthejointfeedresourcesnetworksworkshopheldinGaborone,Botswana,4–8March1991.
AddisAbaba,Ethiopia:ILCA(Inter-nationalLivestockCentreforAfrica);1992.
p.
309–18.
20.
FoodandAgricultureOrganizationoftheUnitedNations(FAO).
MasterLandUsePlan,EthiopianRangeLivestockConsultancyTechnicalReport,AG/ETH/82/020/FAO.
Rome,Italy;1987.
21.
WoodyBiomassInventoryandStrategicPlanningProject(WBISPP).
Astrategicplanforthesustainabledevelopment,conservation,andman-agementofthewoodybiomassresources.
AmharaRegionalState,Finalreport.
AddisAbaba,Ethiopia;2002.
22.
MinistryofAgricultureFisheriesandFood(MAFF).
Energyallowancesandfeedingsystemsforruminants.
ReferenceBook413.
HMSO,London,UK;1984.
p.
85.
23.
ChurchD,PondW.
Basicanimalnutritionandfeeding.
NewYork:Wiley;1982.
24.
InternationalLivestockResearchInstitute(ILRI).
FeedAssessmentTool(FEAST)individualfarmerinterviewquestionnaire[Internet].
Nairobi,Kenya:ILRI;2014.
p.
21.
https://www.
ilri.
org/feastmethodology.
Accessed27Dec2014.
25.
JahnkeHE.
LivestockproductionsystemsandlivestockdevelopmentinTropicalAfrica.
Kiel:KielerWissenschaftsverlagVauk;1982.
26.
KearlLC.
Nutrientrequirementsofruminantsindevelopingcountries.
Logan:UtahStateUniversity;1982.
27.
InternationalBusinessMachinesCorporation(IBMCorp.
Released).
IBMSPSSStatisticsforWindows,Version23.
0.
Armonk,NY,USA:IBMCorp.
;2015.
28.
CentralStatisticalAgencyoftheFederalDemocraticRepublicofEthiopia(CSA).
Agriculturalsamplesurveyin2015/2016.
Reportonlandutilization(privatepeasantholdings,Meherseason).
AddisAbaba,Ethiopia;2016.
29.
BlümmelM,SamadM,SinghOP,AmedeT.
Opportunitiesandlimitationsoffood–feedcropsforlivestockfeedingandimplicationsforlivestock–waterproductivity.
RangelJ.
2009;31:207–12.
https://doi.
org/10.
1071/rj09005.
30.
HaileslassieA,PedenD,GebreselassieS,AmedeT,WagnewA,TaddesseG.
LivestockwaterproductivityintheBlueNileBasin:assessmentoffarmscaleheterogeneity.
RangelJ.
2009;31:213–22.
https://doi.
org/10.
1071/RJ09006.
31.
KebebeEG,OostingSJ,HaileslassieA,DuncanAJ,deBoerIJM.
Strate-giesforimprovingwateruseefficiencyoflivestockproductioninrain-fedsystems.
Animal.
2015;9:908–16.
https://doi.
org/10.
1017/S1751731114003115.
32.
McdermottJJ,StaalSJ,FreemanHA,HerreroM,VandeSteegJA,VanDeSteegJA.
Sustainingintensificationofsmallholderlivestocksystemsinthetropics.
LivestSci.
2010;130:95–109.
https://doi.
org/10.
1016/j.
livsci.
2010.
02.
014.
33.
BogaleS,MelakuS,YamiA.
Matchinglivestocksystemswithavail-ablefeedresourcesintheBaleHighlandsofEthiopia.
OutlookAgric.
2008;37:105–10.
https://doi.
org/10.
5367/000000008784648960.
34.
ToleraA.
FeedresourcesforproducingexportqualitymeatandlivestockinEthiopia:examplesfromselectedWoredasinOromiaandSNNPregionalstates.
Ethiopia:AddisAbaba;2007.
35.
BogaleS,MelakuS,YamiA.
PotentialuseofcropresiduesaslivestockfeedresourcesundersmallholderfarmersconditionsinBalehighlandsofEthiopia.
TropSubtropAgroecosyst.
2008;8:107–14.
36.
DuncanAJ,BacheweF,MekonnenK,ValbuenaD,RachierG,LuleD,etal.
Cropresidueallocationtolivestockfeed,soilimprovementandotherusesalongaproductivitygradientinEasternAfrica.
AgricEcosystEnviron.
2016;228:101–10.
https://doi.
org/10.
1016/j.
agee.
2016.
05.
011.
37.
AsfawMA.
Livestock–waterinteractions:thecaseofGumarawatershedintheUpperBlueNileBasin,Ethiopia.
Ph.
D.
dissertation,DerHumboldt-UniversittzuBerlin,Germany;2013;p.
1–141.
38.
HaileslassieA,PedenD,GebreselassieS,AmedeT,DescheemaekerK.
Livestockwaterproductivityinmixedcrop–livestockfarmingsystemsoftheBlueNilebasin:assessingvariabilityandprospectsforimprovement.
AgricSyst.
2009;102:33–40.
https://doi.
org/10.
1016/j.
agsy.
2009.
06.
006.
39.
HussenM,KecheroY,MollaM.
Availability,yieldandutilizationpracticesoflivestockfeedresourcesinGilgelGibecatchmentsofJimmaZone,SouthwesternEthiopia.
GlobVet.
2016;17:78–94.
https://doi.
org/10.
5829/idosi.
gv.
2016.
17.
01.
1046.
40.
TschoppR,AseffaA,SchellingE,ZinsstagJ.
Farmers'perceptionsoflivestock,agriculture,andnaturalresourcesintheruralEthiopianhighlands.
MtResDev.
2010;4:381–90.
https://doi.
org/10.
1659/MRD-JOURNAL-D-09-00072.
1.
Page17of17Tahiretal.
Agric&FoodSecur(2018)7:1941.
SharmaK,PattanaikAK,AnandanS,BlümmelM.
Food-feedcropsresearch:asynthesis.
AnimNutrFeedTechnol.
2010;10S:1–10.
42.
FoodandAgricultureOrganizationoftheUnitedNations(FAO).
Conduct-ingnationalfeedassessments—FAOanimalproductionandhealthmanualno.
15.
CoughenourMB,MakkarHPS,editors.
Rome:FAO;2012.
43.
WondatirZ,MekashaY.
Feedresourcesavailabilityandlivestockproduc-tioninthecentralriftvalleyofEthiopia.
IntJLivestProd.
2014;5:30–5.
https://doi.
org/10.
5897/IJLP2013.
0158.
44.
YisehakK,JanssensGPJ.
Theimpactsofimbalancesoffeedsupplyandrequirementonproductivityoffree-rangingtropicallivestockunits:linksofmultiplefactors.
AfrJBasicApplSci.
2014;6:187–97.
https://doi.
org/10.
5829/idosi.
ajbas.
2014.
6.
6.
9168.
45.
MooreAD,BellLW,RevellDK.
Feedgapsinmixed-farmingsystems:insightsfromtheGrainandGrazeprogram.
AnimProdSci.
2009;49:736–48.
https://doi.
org/10.
1071/AN09010.
46.
AbegazA,VanKeulenH,OostingSJ.
Feedresources,livestockproductionandsoilcarbondynamicsinTeghane,NorthernHighlandsofEthiopiaTeghane,NorthernHighlandsofEthiopia.
AgricSyst.
2007;94(94):391–404.
https://doi.
org/10.
1016/j.
agsy.
2006.
11.
001.
47.
ValbuenaD,TuiSHK,ErensteinO,TeufelN,DuncanA,AbdoulayeT,etal.
Identifyingdeterminants,pressuresandtrade-offsofcropresidueuseinmixedsmallholderfarmsinSub-SaharanAfricaandSouthAsia.
AgricSyst.
2015;134:107–18.
https://doi.
org/10.
1016/j.
agsy.
2014.
05.
013.
48.
LenneJM,Fernandez-RiveraS,BlümmelM.
Approachestoimprovetheutilizationoffood-feedcrops-synthesis.
FoodCropRes.
2003;84:213–22.
https://doi.
org/10.
1016/s0378-4290(03)00152-7.