Accorhdingto SNV (2008), the Eastern Province has more than 49% of all the cattlepopulation in Rwanda but is challenged by prolonged drought as it is the firstregion with high risk of drought (MIDMAR, 2014, Bazimenyeraet al., 2014).
). Dairy productionplays an important role in the area both as a food source and for cashgeneration. Cattle for instance,contribute 30% of the total household income of which 89% is from the sale ofmilk. The main cattle breeds kept arecrossbreed between exotic (mainly jersey and Holstein) and local East AfricanShorthorn.
Milk productivity perlactating animal in the area is normally between 3 and 10 kg per day in the dryseason, and 5 and 15 kg per day in the wet season. The dominant system of dairy production iszero grazing system (Cut and Carry system).Farmer interviews andobservationThe study will target smallholder farmers in zero-grazingdairy production system, who keep cattle in a common cow shed. A comprehensive list of common cow shedswill be obtained from the Girinka coordinator and District Animal Resources Officer(DARO). The list will be used to draw upa sample framework for the formal survey.
Five (5) farms (common sheds) from eachdistrict will be randomly selected for interview. This will give a total number of 35 farms inthe seven districts. Interviews will beconducted using a structured questionnaire and it will start in March 2018. The data to collect are classified in socialdemographic information about the shed manager, members, herd status (agecategory, sex, breeds) animal welfare, health management and feeds and feeding(types of grasses used, quantity and varieties of Napier used, the way Napiergrass is grown, available land, used conservation techniques and storage,Challenges/ season, best forage, needs in forages, their awareness on Napiergrass new varieties and manure management.Feeding management monitoringFromthe 5 selected farms per district for interview, one will be randomly selectedfor feeding monitoring. The feed will bemonitored in a total of seven sheds over one-week period in March, May, Augustand October 2018. March and August aredry season months associated with feed scarcity while May and October are thepeaks of the rainy season when feeds are more widely available.
From each shed, 10 for a total of 70 cows willbe selected for feed monitoring. A 50 kgspring scale with units of 0.5 kg will be used for weighing feeds at each farm. Samples of various varieties of feedwill be pooled separately and subsampled each week for chemical analysis.
Refusals will be weighed every afternoon,accumulated weekly, subsampled and prepared for chemical analysis. The types of feeds offered to cattle duringthe exercise will be classified into: Napier grass; weeds (roughage cut fromroad sides and cropland); fibrous crop residues; fodder legumes; other foragesgrass and commercially processed feeds (concentrates) which included dairymeal, wheat or rice bran and pollard (a by-product from the oil extractionprocess from maize grain). The dailymilk yield from each lactating cow in monitored sheds will be measured byfarmers using a scale and recorded in an exercise book for the duration of thestudy.
The collection of Milk samples willbe done in every season and kept in cold storage.Feed samples processingand nutrient analysisFeed sampleswill be dried at 65 °C to constant weight and ground to pass through a 1 mmscreen before analysis. Chemicalproximate analysis of feed will be done by wet chemistry (AOAC, 1995) for drymatter (DM), total N (To be calculated as N × 6.
25.), neutral detergent fibre(NDF), acid detergent fibre (ADF) and total condensed tannins. Pasture yield andresidual hay estimationThebiomass yield of the pasture will be estimated by placing an exclusion cage pereach monitored farm (seven in total) and extrapolating the yield to the area ofthe pasture. Banana pseudo-stems andlegume trees in the diet will be estimated from farmer recall.
Maize Stover biomass will be estimated byapplying farmer recall of grain yield to a harvest index (0.41) (Remison & Fajemisin, 1982).For forage pasture, the cutting intervals will be taken. The soil samples from the pasture will betaken and rainfall conditions data considered.Milk samples analysisMilk sampleswill be analyzed for butterfat (BF) using Gerber method (Kleyn et al.
,2001) and milk density (FAO, 1998) by New Kenya Co-operative Creameries (KCC),Kapsabet. The Richmond’s formula (Bector& Sharma, 1980) will be used to calculate milk solid non-fat (SNF) from BFand density.SNF=(milkdensity/4)+(0.22*BF)+0.72Milkenergy content (ECM) will be calculated from equation by Tyrell (1965);ECM=0.0386F(g ? (kg milk) )+0.
0205SNF(g?(kg milk) )-0.236Data analysisDatafrom the field exercise and chemical analyses of the feed samples will bestatistically analyzed using R (R, 1985). Standard deviations and means will becalculated for numbers of cattle per cow shed, areas of fodder crops grown, areof Napier grown, quantities of roughage, hay and concentrates offered to cattleduring rainy and dry seasons, folder conservation techniques and generalmanagement. Analyses of Variance (ANOVA)will be carried out to test for differences between farms, districts andbetween seasons, in quantities of roughage and concentrates offered, and inchemical composition of feeds and milk. During the rainy season, the daily refusalquantity and quality, pasture production and herd size will be used to estimatethe hay quantity and quality that can be made from the overproduction.Objective 2Estimateenteric methane emissions from common cow sheds in Eastern province of Rwandaacross seasons using Tier II method. Estimation of Enteric CH4emissionThegeneral approach of IPCC Tier 2 will be used to estimate enteric methaneemission. This approach integrates theanimal activity performance, production data and metabolizable energyrequirements (MER) to compute the daily methane production (DMP) and emissionfactor (EF).
Live weights of each cow inthe seven selected farms will be estimated through heart girth measurements usinga ‘Weighband’ (Dalton Supplies Limited, Nettlebed, UK) at the start of themonitoring. WhereDMP: daily methane production and DMI: Dry matterintake Mean DMPfor each class of animal in each season will be calculated. This will be then used to calculate an annualenteric methane EF (CH4kg/head/year): EF:emission factor Forthe Methane gas analyses, a multiplelinear regression model and analysis of variance (ANOVA) will be utilized. Descriptive statistics (mean and standarderror of means (SEM)) will be calculated for live-weight, live-weight change,daily milk yield, total metabolizable energy requirements (MER), DMI and DMPfor each district and season.
The linearregression model fitted tested the association of methane emissions with drymatter digestibility and the impact of milk production on energy required forlactation and total metabolizable energy. The dependent variables will be daily methaneproduction.Yi = ?0 + ?1X1,1 + ?2 X1,2+ ? Where;Yi =DMP/ MERL/ Total MER?0, =constant, ?1and ?2=regression parameters, X1,1 = dry matter digestibility, X1,2=daily milk production, ? = randomerrorThe effect of district and season on DMD and DMP, willbe analysed using ANOVA fitting season (SR, HD, LR and CD) and districts asfactors in a general linear model in the form:Yijk= ? + ?i + ?j+ ?ij with i= 1, 2, 3,4,5,6,7 and j= 1, 2, 3, 4 Where:Yijk= DMD/DMP, ?= overall mean, ?i= leveleffect of district, ?j=level effect of season (SR, HD, LR or CD). ?ij= randomerror Objective 3Determine theeffect of Napier grass (Pennisetum purpureum) supplemented with the hay ofClitoria ternatea and Calliandra spp as protein and tannin source on methaneemission, nitrogen and milk quality in lactating cattle.The study will be carriedout during the dry season in the Eastern province of Rwanda describedabove.
Twelve (12) lactating cows willbe purposively selected from the seven farms where the feeding ismonitored. The selection criteria willbe the use of Napier grass as basal diet by the owner, no use of concentrates,being from 1st to 3rd lactation, crossbreed, dewormedagainst endoparasites and sprayed against ectoparasites, and havingapproximately the same live weight. Theexperimental animals will be weighed for initial milk yield and body weight. Cows will be divided into six groups of two animalseach. The animals from each group willbe randomly assigned to six dietary treatment.Feeds and experimental dietsClitoria will established atthe onset of long rains (April) without fertiliser application in the pastureof the university of Rwanda (Eastern Province) and harvested after attaining50–60% flowering. Calliandra and minerallicks will be respectively purchased from farmers and local supplier.
The hays of Clitoria and Calliandra will bemade and stored. The control diet consistedof Napier grass fed ad libitum as usual and 60g of a mineral lick daily. The other five treatments consisted of thecontrol supplemented with 2kg (DM) of Clitoria and/or Calliandra hay.
Considering the daily DMI of 10 Kgs, cows will be getting 80% (8kg (DM)) of thebasal diet (Napier grass) and 20 % (2kg (DM)) of legumes (Clitoria and/ orCalliandra) as protein supplement.The six groups of lactating cows will be allocated tothe six dietary treatments (table 1) in a 6×6 Latin-square design. Thebasal diet will be offered after the half of the supplement had beenconsumed. In late afternoon the other remaininghalf of supplement will be offered. Eachexperimental period will last for 10 days, consisting of 5 days of adaptationto the respective diet, 5 days of measurement and collection. Each animalreceived the six diets in a different sequence.
The grass quantity will be measured as well asthe refusal. The milk samples will bealso taken for analysis. Data will be collected for a period of 60days including the adaptation period of cows to diets.Table 2:Different diet treatments Ingredient Basal diet (Napier grass) % Clitoria hay % Calliandra hay % Treatment 1 100 0 0 2 80 20 0 3 80 15 5 4 80 10 10 5 80 5 15 6 80 0 20 In vivo apparent diet digestibilityDuring data collection ofeach diet, total daily faecal output for two consecutive days per experimentalanimal will be collected in individual buckets as it was dropped. The faeces will be weighed, and weights willbe recorded.
Samples will be used for drymatter determination and ashing. Thesevalues will be used to determine the amount of dry matter and organic matter inthe faeces, that will help to calculate the percent DM and OM digestibility ofthe diets as described by Abdulrazak (1995).Forage andmilk sample analysisThe forages samples fromeach treatment will be taken for proximate analysis.
Tannins will be determinedaccording to the methods described by Abdulrazak and Fujihara (1999).Milk sampleswill be analyzed for butterfat (BF), milk density and non-fat (SNF) using themethod described in objective 1.Statisticalmodel and analysisThefollowing statistical linear model will be used:yij=µ+ti+eijwhereyij is the response due to ith treatment and jthanimal; µ is overall mean; ti is the effect of ith treatmentand eij is a random error component.Datawill be subjected to analysis of variance (ANOVA), using the General linear model(GLM) procedures of Statistical Analysis Systems (SAS, 1987). Means will be separated using the leastsignificant difference (LSD).
Objective4 Assessthe in vivo digestibility and methane gas production of the hay made of Clitoria ternatea and Calliandra spp with and without thePolyethylene glycol (PEG) added. Thehay with and without the polyéthylèneglycol ((PEG-6000) added will be analyzed for digestibility,gas production (GP) characteristics and methane production. Samples will be of five types of mixture of Clitoriaand Calliandra. The proportions of thetwo legumes will be: 4:0,3:1,2:2, 1:3 and 0.
six replicates will be done from each mixture type. Samples will be dried in forced oven, andground and analyzed for nutrient composition. The PEG will be added in three replicates of each type. In vitro GP and in vitro organicmatter digestibility (IVOMD) will be determined using rumen fluid collected,strained and anaerobically prepared. Asemi-automated system will be used to measure GP by incubating the sample in ashaking incubator at 39°C.The net gas volume data werefitted to the model p =a +b(1-e-ct) (Menke and Steingass,1988).
Where,a, the gas production from the immediately soluble fraction (mL); b, the gasproduction from the insoluble fraction (mL); c, the gas production rateconstant for the insoluble fraction (mL/h); a+b, potential gas production (mL);t, incubation time (h); y, gas produced at time “t”.Theeffect on microorganism evolution will also be determined.Objective 5 Evaluate the effect ofhigh yielding Napier ecotypes, Calliandra and Clitoria on methane emission,nitrogen and energy balance in growing lambs Thestudy will be conducted at BecA-ILRI Hub (Nairobi). It will consist of feeding trials ofdifferent diets on growing lambs. Diets willconsist of different proportions of dried, leafy Napier grass, the herbaceouslegume Clitoria and the foliages of the tannin-rich shrub legume Calliandra. Napier grass cultivar to be used in the studywill be two high nutritious ecotypes from Rwanda.Theforages will be used to form six diets: one pure Napier grass diet and fivediets consisting of 80% Napier grass and 20% legumes (on dry matter (DM) basis).
The legume part of the diets will consistof Clitoria or Calliandra separately or combined in ratios of: 3:1,2:2, 1:3.Sixcastrated, male lambs of 3 months of age of the same breeds will be weighted, allocatedto the six dietary treatments in a 6X6 Latin-square design. Each experimental period will last for 16days, consisting of 7 days of adaptation to the respective diet, 7 days of collectionand measurement of growth and nitrogen related parameters, and 2 days ofquantitative measurement of gaseous exchange in dual open-circuit respiratorychambers.Eachanimal will receive the six diets in a different sequence with a daily forageDM allowance of 60 g per kg of metabolic BW (BW 0.75). The diet sample will be taken for proximateanalysis. Animalswere weighed directly before feeding in the morning, at the beginning and theend of each experimental and adaptation periods. The food refusal will be recorded during theexperimental period.
Faeces and urine will be collected for more analysis. 4 hafter feeding, rumen fluid and blood samples will be taken.LaboratoryanalysesFeeds,refusals and faeces will be analysed for DM, total ash (to calculate organicmatter (OM)), NDF and ADF, feeds also for ADL, all by standard methods (AOAC,1990). NDF will be analysed without theaddition of sodium sulphite, and values of NDF and ADF will be corrected forash content. Extractable and fibre-boundCT will be determined according to the procedure suggested by Terrill et al.(1992) as modified by Barahona et al.
(2003). Nitrogen and C content will be determined infeeds, refusals, fresh faeces and urine. Gross energy (GE) contents of feeds, refusals and lyophilised faeceswill be assessed. Ciliate protozoa andbacteria will be microscopically enumerated. Determination of VFA willbe performed by HPLC. Blood plasmaconcentration of urea will be determined. Oxygen, carbon dioxide and methane fromeach lamb will be quantified.Calculationsand statistical analysisCand N concentrations will be used to calculate urinary energy (Hoffmannand Klein, 1980).
Methane energy lossand energy expenditure will be calculated from the gaseous exchange data (heatproduction) by the equations of Brouwer (1965) but corrected for thecarbon dioxide emitted from fermentation (Chwalibog et al., 1996). H= (3.866 x O2) +(1.200 x CO2) -(0.
518 x CH4)- (1.431 x N) where H = heat production (kcal), O2, CO2and CH4 represent volumes of oxygen consumed and of carbon dioxideand methane produced in litres and N: urine nitrogen (g).Totalbody energy retention (RE) will be calculated as metabolizable energy (ME) minusenergy expenditure. Body fat energy will be calculated as the differencebetween RE and the energy retained in protein.TheGLM procedure of SAS (version 9.1.3; 2006; SAS, Cary, NC, USA) will be used foranalysis of variance with diet, animal and experimental periods as sources ofvariation.