Progress Report

Transcription

Progress Report

FEDERAL PUBLIC SERVICE
FOREIGN AFFAIRS,
FOREIGN TRADE AND
DEVELOPMENT CO-OPERATION
________________________
Directorate General
Development Co-operation
____________________
The Consortium for
Improving Agriculture-based
Livelihoods in Central Africa
(CIALCA)
Progress Report
November 2006 –
December 2007
The Consortium for Improving Agriculture-based
Livelihoods in Central Africa (CIALCA)
Progress Report
November 2006 – December 2007
EXECUTIVE SUMMARY
5
MAJOR FINDINGS
5
1. INTRODUCTION
9
2. BENCHMARK AREA CHARACTERIZATION AND ORGANIZATION
13
3. PROGRESS WITH CHARACTERIZATION ACTIVITIES
16
3.A. CHARACTERIZATION STRATEGY
3.B. BASELINE SURVEYS
3.C. FINAL CHARACTERIZATION ACTIVITIES
3.C.1. DETAILED CHARACTERISATION STUDY ON LEGUME PRODUCTION, MARKETING AND
CONSUMPTION, AND NUTRITIONAL STATUS OF RURAL HOUSEHOLDS
3.C.2. DETAILED CHARACTERISATION STUDY ON BANANA PRODUCTION, MARKETING AND
16
17
23
CONSUMPTION
4. PROGRESS WITH STRATEGIC SOIL FERTILITY-RELATED ACTIVITIES
23
29
32
4.A. RELATIONSHIP BETWEEN SOIL FERTILITY AND NUTRITIONAL QUALITY OF BIOFORTIFIED BEANS
32
4.B. ASSESSMENT OF NUTRIENT DEFICIENCIES IN SOILS ON THE WALUNGU AXIS IN
SUD-KIVU
4.C. ASSESSMENT OF BANANA – ARBUSCULAR MYCORRHIZAL FUNGI RELATIONSHIPS
36
39
5. PROGRESS WITH BANANA GERMPLASM-RELATED ACTIVITIES
40
5.A IN-SITU GERMPLASM EVALUATION
5.B STRATEGIC RESEARCH AT K.U.LEUVEN
40
42
6. PROGRESS WITH LEGUME GERMPLASM-RELATED ACTIVITIES
44
6.A. LEGUME GERMPLASM DEMONSTRATION AND EVALUATION
6.A.1. ON-STATION LEGUME GERMPLASM EVALUATION
6.A.2. LEGUME GERMPLASM EVALUATION WITH FARMER ASSOCIATIONS AT THE
44
44
ACTION SITES
47
6.B. LEGUME SEED MULTIPLICATION
6.B.1. ON-STATION LEGUME SEED MULTIPLICATION
6.B.2. FARMER ASSOCIATION-LED LEGUME SEED MULTIPLICATION
53
53
54
7. PROGRESS WITH NATURAL RESOURCE MANAGEMENT-RELATED
ACTIVITIES
57
7.A. NATURAL RESOURCE MANAGEMENT OPTIONS FOR LEGUME-BASED SYSTEMS
7.A.1. OVERVIEW OF OPTIONS CURRENTLY BEING TESTED
7.A.2. SOIL CONSERVATION TECHNOLOGIES TESTED IN SUD-KIVU (“ERO-1” AND “ERO-2)
7.A.3. IMPROVED AGRONOMY AND SOIL FERTILITY MANAGEMENT IN
CASSAVA-LEGUME SYSTEMS
7.A.4. OPTIONS FOR SOIL FERTILITY AMENDMENT ON THE WALUNGU AXIS IN SUD-KIVU
7.B. NATURAL RESOURCE MANAGEMENT OPTIONS FOR BANANA-BASED SYSTEMS
7.B.1. ON-FARM TRIALS
7.B.2. ON-STATION TRIALS
7.B.3. INITIALISATION OF BANANA DISEASE CONTROL STRATEGIES
62
64
65
65
66
67
8. PROGRESS WITH MARKET-RELATED ACTIVITIES
68
8.A. BANANA VALUE CHAIN ANALYSIS
8.B. LEGUME VALUE CHAIN ANALYSIS
68
69
9. PROGRESS WITH NUTRITION-RELATED ACTIVITIES
72
9.A. SOYBEAN PROCESSING AND UTILIZATION
9.B. INITIALISATION OF BANANA NUTRITION-HEALTH-RELATED ACTIVITIES
72
74
10. PROGRESS WITH MONITORING AND EVALUATION
75
11. PROGRESS WITH DEGREE-RELATED ACTIVITIES
76
12. ANNEXES
80
ANNEX 1. LOG-FRAME EVALUATIONS
ANNEX 1.A. LOG-FRAME OF THE TSBF-CIAT-LED PROJECT
ANNEX 1.B. LOG-FRAME OF THE BIOVERSITY-LED PROJECT
ANNEX 1.C. LOG-FRAME OF THE IITA-LED PROJECT
ANNEX 2: FINAL CHARACTERIZATION TOOL FOR THE LEGUME-BASED SYSTEMS
ANNEX 3: LEG-2: LEGUME GERMPLASM DEMONSTRATION TRIALS
ANNEX 4: QUESTIONNAIRE USED FOR LEGUME GERMPLASM EVALUATION BY
57
57
59
80
80
83
85
87
104
110
FARMER ASSOCIATIONS
ANNEX 5: ACTION PLAN FOR INFORMAL LEGUME SEED SYSTEMS, 2007B–2008B,
TSBF-CIALCA
ANNEX 6: TRAINING OF CIALCA STAFF ON LEGUME SEED SYSTEMS AND SEED
115
117
PRODUCTION
ANNEX 7: ATELIER DE FORMATION SUR LA MULTIPLICATION DES SEMENCES DES
LEGUMINEUSES
ANNEX 8: FICHE FOR DATA COLLECTION IN LEGUME MULTIPLICATION FIELDS
ANNEX 9: ERO-1: EROSION CONTROL IN SUD-KIVU
3
118
129
130
ANNEX 10: ERO-2: COMPARISON OF VARIOUS FORAGE SPECIES FOR
EROSION CONTROL
ANNEX 11: EVALUATION DES FOURRAGES EN MILIEU PAYSAN (ERO-2)
ANNEX 12: CAS-1: IMPROVED CASSAVA AGRONOMY
ANNEX 13: CAS-2: IMPROVED CASSAVA AGRONOMY
ANNEX 14: CAS-3: IMPROVED FERTILITY MANAGEMENT IN CASSAVA SYSTEMS
ANNEX 15: PREFERENCES DES ESSAIS D’AMELIORATION DES SYSTEMES
AGRICOLES BASES SUR LE MANIOC PAR LES AGRICULTEURS DE KABAMBA
ANNEX 16: FER-1: IDENTIFICATION OF INPUTS REQUIRED FOR SOIL FERTILITY
AMENDMENT
ANNEX 17: PROTOCOL FOR ON-STATION AND ON-FARM MULCH TRIALS
ANNEX 18: RELEVANT REPORTS, PRESENTATIONS, AND PUBLICATIONS
4
141
143
147
149
153
157
164
167
171
EXXEECCU
UT
TIIV
VE
E SSU
UM
MM
MA
AR
RY
Y
Following a call for proposals of the Directorate General for Development Cooperation
(DGDC - Belgium) in April 2004, three proposals were approved:
•
‘Sustainable and Profitable Banana-based Systems for the African Great Lakes Region’, led by
the International Institute of Tropical Agriculture (IITA), Kampala, Uganda.
•
‘Enhancing the resilience of agro-ecosystems in Central Africa: a strategy to revitalize
agriculture through the integration of natural resource management coupled to resilient
germplasm and marketing approaches’, led by the Tropical Soil Biology and Fertility Institute
of the International Center for Tropical Agriculture (TSBF-CIAT), Nairobi, Kenya.
•
‘Building Impact Pathways for Improving Livelihoods in Musa-based Systems in Central
Africa’, led by the International Network for the Improvement of Banana and Plantain of the
International Plant Genetic Resources Institute (INIBAP-IPGRI), Kampala, Uganda.
As the above projects proposed to operate largely in the same parts of Rwanda, Burundi,
and the Democratic Republic of Congo (DRC), with similar national partner institutes, and
due to the complimentary nature of the activities proposed, above institutes agreed to
operate as a Consortium to ensure cooperation and complimentarity and avoid technical
and financial duplication at the national level. The Consortium for Improving Agriculturebased Livelihoods in Central Africa (CIALCA) is a Consortium of the International
Agricultural Research Centers (IARCs) and their national research and development
partners that aims at close technical and administrative collaboration and planning in areas
of common interest, thereby enhancing returns to the investments made by DGDC and
accelerating impact at the farm level.
This report gives technical details and evaluates progress made in the 3 projects during the
period November 2006 – December 2007, and describes some of the strategies and
planned activities for 2008. Formal logframe evaluations are included in Annex 1.
M
MAAJJO
OR
R FFIIN
ND
DIIN
NG
GSS
•
•
After completion of the PRA activities and baseline survey, action sites were
selected in all mandate areas and currently, activities are on-going with about 40
farmer associations across the different action sites. Activities in satellite sites, in
which partner NGOs take the lead, have been initiated with 90 farmer associations.
Through the baseline survey, implemented with 2,800 households across the 10
mandate areas, detailed information on various livelihood dimensions of rural
livelihoods has been obtained. This information covered the areas of household
structure and economics, social capital, agriculture, market access and postharvest
processing and handling, food and nutrition, food security, and health. Some
interesting facts are that (i) most agricultural production is based on relatively small
amounts of organic inputs or no inputs at all, (ii) food insecurity is a major
problem for over 40% of all households although large differences between
mandate areas were observed, (iii) generally, most households have poor access to
large regional or urban markets, especially in the more isolated areas, and (iv) social
capital is quite extensive in most areas with relatively large proportions of
households belonging to farmer, credit and savings, women, self-help, or health
insurance groups.
5
•
•
•
•
•
•
•
•
Detailed characterization of banana systems was implemented with 30 farms per
action site and covered the thematic areas of production, marketing, and
consumption. Productivity varied widely, ranging from 21-63 ton per hectare per
cycle. In terms of pest and diseases, Fusarium was widespread while banana
Xanthomonas wilt and banana bunchy top virus affected specific mandate areas.
Deficiencies of P and Mg were prominent on strongly weathered soils. Lastly,
drought stress resulted in yield losses, estimated between 30 and 70%.
A detailed characterisation study was conducted to complement the earlier
baselines survey with quantitative information on aspects of legume cropping, soil
fertility status, marketing of legume produce and nutritional status. Following
preliminary conclusions were drawn:
•
Low soil fertility, drought, climatic variability and erosion are the dominant
constraints for crop production.
•
Legume grain price variability in time and space opens up market opportunities
through storage and transport.
•
Anthropometric measures identify notable levels of malnutrition in young
children, with 8 – 23% being at risk and 2 – 12% suffering moderate
malnutrition. Malnutrition was most pronounced in Sud-Kivu and least in
Rwanda.
An analysis was conducted on variability in Fe and Zn contents in grains of biofortified bean varieties. While in some varieties, micronutrient contents vary with
environmental conditions (and appears to be related to the organic matter content
to the soil), a number of varieties contain high amounts of Fe across environments
(e.g., ARA-4), and can therefore be recommended.
The Walungu area is very unproductive due to low soil fertility constraints. A pot
trial was conducted on a large number of soils to assess nutrient deficiencies.
Preliminary results identified low P as the major constraint, but symptoms of other
deficiencies were observed. Detailed measurements are at present pending.
A survey was carried out in 188 fields in Rwanda to identify arbuscular
mycchorizal fungi (AMF) infection and plant parasitic nematode infection on
banana roots. Highly variable infection rates were observed, providing an insight in
the role that AMF play in banana production systems and possible benefits for
future use of AMF to improve plant health and vigour.
Related to in-situ conservation of banana germplasm, various varieties obtained
through the SMIP project, existing tissue culture labs in the region, and ITC in
Leuven, were planted at about 20 sites in the various mandate areas. Measurements
on productivity, pest and disease tolerance, profitability, and genotype x
environment interactions will be taken. Local macro-propagation facilities have
been installed near each germplasm trial for rapid and clean multiplication of the
best varieties.
Strategic banana research at KULeuven, has focused on assessment of droughts
stress in Musa cell cultures, cryopreservation, promoter tagging as a basis for
developing cisgenic bananas, and understanding arbuscular-mycorrhizal fungal biocontrol and its impact on banana productivity and survival. Various activities also
were implemented around broadening the banana ITC, managed by KULeuven.
On-station legume evaluation activities focused on identifying varieties tolerant to
low-P conditions, and promiscuous varieties producing high amounts of biomass.
Promising varieties were identified but differed between mandate areas. Currently,
a selection of varieties is being fully characterised and multiplied to enable
homologation and official release in the countries.
6
•
•
•
•
•
•
•
Legume germplasm demonstrations and evaluations were conducted in 4 mandate
areas, involving a total of 44 farmer associations. Performance differed between
sites and mandate areas, and farmers used differing criteria to select preferred
varieties. Researcher-defined criteria were explained and taken into account for
final selection. In each site, at least 3 bush beans, 2 climbing beans and 2 soybean
varieties were retained.
Informal seed multiplication of farmer-preferred legume varieties was initiated and
involves at present both associations in action and satellite sites. About half of the
associations have produced sufficient seed quantities to satisfy their own needs and
activities have been initiated to promote, disseminate and commercialize the
production of improved germplasm seed. Associations are being supported
through training, facilitation by NGO partners and involvement of the national
seed service.
A number of technologies targeting the main constraints for legume cropping
(principally improved agronomy and nutrient input management) were
demonstrated, involving a total number of 56 farmer associations in the mandate
areas. Technologies for soil erosion control in Sud-Kivu and rain water harvesting
in Rwanda to counteract seasonal drought spells are currently being tested onstation:
•
Hedgerow planting and reduced tillage are a valid alternative to terrace
construction, and do not negatively affect yields while being relatively effective
in soil erosion control in the short term.
•
Improved agronomic practices, using modified spacing and high-biomass
yielding legumes can significantly improve legume yields in cassava
intercropping systems.
•
Fertilizer application, preferably applied in combination with organic resources,
considerably increases legume yields in different cropping systems and sites.
•
Demonstration activities have attracted large interest of farming communities
and will proceed into an adaptation phase in 2008.
Mulch application, manure application, soil and water conservation measures, and
general plantation sanitation and husbandry were the themes that were retained by
farmers in the context of improved management of banana plantations. About 15
on-farm trials per action site will be installed covering above themes. Parallel to the
on-farm trials, on-station trials have been established at 8 sties, focusing on mulch,
tillage, and bean intercropping. Specific trails looking at competition between mats
for light, water, and nutrients as affected by banana planting density have been
installed in 3 sites in Rwanda.
Initial steps have been taken to start Xanthomonas wilt activities in Rwanda,
focusing on screening of banana germplasm, Xanthomonas wilt control options
and replanting time. Additional work on systemicity of the bacteria will be
conducted in Uganda.
Market surveys with 400 traders and 150 transporters aiming at detailing banana
value chains revealed that most actors deal mainly with beer banana. Various costs
along the value chain were quantified and critical market constraints were also
identified. Cross-border studies between Rwanda, DR Congo, Burundi, and
Uganda were also completed.
Legume (soybean, bean, and groundnut) value chain analysis has been completed
in the Bas-Congo and Sud-Kivu mandate areas. Most farmers did not sell legumes
and for those that sell a proportion of their produce, revenues were mostly used to
meet urgent needs and not to re-invest in agriculture. Prices were strongly
7
•
•
•
influenced by periods of abundance and shortage. Most traders in the local markets
were women.
A strategy to train farmers and trainers in soybean processing and utilization has
been completed and training materials that accompany those training events have
been produced. At least 10 25-are demonstration gardens for soybean production
have been installed near health centers at the action sites. It is expected that in first
instance 200 trainers and 400 farmers will be trained. A strategy has also been
developed for the trainers to continue training individual farmers. Planning for
banana nutrition-related activities has also been completed.
A detailed M&E framework has been put in place to monitor and evaluate
progress with project implementation and project interventions. The baseline
survey will serve as a basis for evaluation of initial project-related impact to be
assessed towards the end of the first phase of CIALCA.
In terms of degree-related training, CIALCA is currently engaging 8 PhD students,
13 MSc students, and 13 undergraduates. Four post-doctoral fellows are leading
specific project activities. Various training sessions with farmer associations, for
instance, on seed multiplication and participatory evaluation of technologies, have
also been conducted. On-the-job training of national system scientists has resulted
in substantial improvements in the technical capacity of those partners.
8
1. IN
NT
TR
RO
OD
DU
UC
CT
TIIO
ON
N
Following a call for proposals of the Directorate General for Development Cooperation
(DGDC - Belgium) in April 2004, focusing on Central Africa, three proposals were
approved:
•
‘Sustainable and Profitable Banana-based Systems for the African Great Lakes Region’, led by
the International Institute of Tropical Agriculture (IITA), Kampala, Uganda.
•
‘Enhancing the resilience of agro-ecosystems in Central Africa: a strategy to revitalize
agriculture through the integration of natural resource management coupled to resilient
germplasm and marketing approaches’, led by the Tropical Soil Biology and Fertility Institute
of the International Center for Tropical Agriculture (TSBF-CIAT), Nairobi, Kenya.
•
‘Building Impact Pathways for Improving Livelihoods in Musa-based Systems in Central
Africa’, led by Bioversity International, Kampala, Uganda.
The purpose of the project led by IITA is to develop and disseminate in partnerships with
all stakeholders technologies that improve the sustainability and profitability of bananabased cropping systems. Emphasis is put on identifying and exploring markets as a driving
force for changing banana-based farming systems. Technologies promoted include
amongst others locally adapted natural resource management options (including
integration of legumes), integrated pest management options, the introduction of new
banana hybrids, and improved post-harvest technologies. The project emphasizes strong
partnerships and capacity building with NARS, Universities, non-governmental
organizations (NGO), community-based organizations (CBO), and the private sector. The
project will also put emphasis on strategic research on sustainable use of the natural
resource base through collaboration with UCL.
The purpose of the Bioversity-led project is to strengthen national and regional
mechanisms to plan and orient investments, projects and research for development
synergies by increasing the contribution of Musa to rural well-being. The project will also
strengthen national frameworks for conserving local Musa germplasm, introducing and
evaluating new cultivars and multiplying and disseminating clean planting material of
superior cultivars. An important part of the project will be to support the global collection
of Musa germplasm and research on stress responses on banana, both occurring in
Belgium. The project further aims to identify, with scientists, extension agencies, NGOs,
and farmers, market opportunities for bananas and banana products, to validate options
for integrated pest and soil fertility management, and to develop improved Musa
production systems. The project emphasizes strong partnerships and capacity building
with NARS, universities, NGOs, CBOs, and the private sector. In the INIBAP-IPGRI-led
project, strategic research backing is given by K.U.Leuven.
The purpose of the project led by TSBF-CIAT is to develop and disseminate in
partnerships with all stakeholders resilient agro-ecosystems through integration of stresstolerant and bio-fortified germplasm, inclusion of locally adapted natural resource
management (NRM) options, market-led diversification and intensification, and
revitalisation of research for development capacity of all stakeholders. The main entry
points are multi-purpose legumes that will address issues related to declining soil fertility,
low income, and food insecurity and malnutrition which are major constraints to improved
rural livelihoods in the target areas. The project aims at integrating strategic, applied, and
adaptive research for development with strong involvement of various partners with
9
expertise in all the above sectors. In the TSBF-CIAT-led project, strategic research backing
is given by K.U.Leuven.
As the above projects proposed to operate largely in the same parts of Rwanda, Burundi,
and the Democratic Republic of Congo (DRC), with similar national partner institutes, and
due to the complimentary nature of the activities proposed, above institutes agreed to
operate as a Consortium to ensure cooperation and complimentarity and avoid technical
and financial duplication at the national level. The Consortium for Improving Agriculturebased Livelihoods in Central Africa (CIALCA) is a Consortium of the International
Agricultural Research Centers (IARCs) and their national research and development
partners that aims at close technical and administrative collaboration and planning in areas
of common interest, thereby enhancing returns to the investments made by DGDC and
accelerating impact at the farm level. The overall goal of CIALCA is to facilitate all above
interactions in order to obtain integration between the activities of each project to the
extent that this is desirable. It is not to create a ‘super-project’ directing all activities within
each of the three projects.
In order to operationalize CIALCA,
various initiatives were taken both at the
administrative and at the technical level.
At the administrative level, a
Memorandum of Understanding between
IITA, Bioversity, and CIAT has been
approved; a CIALCA Consultative
Committee (CCC) has been constituted;
CIALCA offices or representations have
been installed and CIALCA facilitators or
representatives engaged; and
communication and reporting channels
have been formalized. At the technical
level, various annual planning meetings
have been organized (Photograph 1) and
a CIALCA website was set up
(WWW.CIALCA.ORG) (Photograph 2).
The CIALCA website is a knowledge
platform for CIALCA reports,
newsletters, publications, student theses,
but also contains info on partner
organisations and has links to other
interesting websites. It is envisaged to
upload most of the available regional
agricultural information on the CIALCA
website.
Photograph 1: Participants at an annual
CIALCA planning meeting.
Photograph 2: Homepage of the CIALCA
website.
The coordinates of the CIALCA offices are:
•
Burundi: Mr. Sylvestre Hakizimana, IRAZ, PO Box 91, Gitega, Burundi, Tel: (+257)
403020/21, Mobile: (+257) 903315, email: [email protected] or [email protected].
•
Rwanda: Mrs. Kantengwa Speciose, c/o CIAT Rwanda, Kacyiru, Boulevard the l'Umuganda,
Concorde building, 1st floor, Kigali, Tel:(+250) 55 104708 or 08518471, email:
[email protected].
10
•
DRC – Sud-Kivu: Mr Dieudonné Katunga Musale, Coordinator, 6 Av. Kasongo, Commune
d'Ibanda, Bukavu, Eastern D.R.Congo, Tel:(+243) 98 669793, email:
[email protected].
•
DRC – Bas-Congo: Mr Jean-Paul Lodi Lama, c/o INERA office, 13 Avenue des Cliniques,
Kinshasa-Gombe, B.P.2037 Kinshasa 1, Tél: (+243) 815136746. email:
[email protected] or [email protected].
Various benefits have been observed while operating as CIALCA in the target areas. At
the administrative level, (i) although no funds were originally budgeted for setting up
offices in various locations, due to the combined effort of the three projects, sufficient
funds were identified to have such offices installed, which has proven to be an essential
component in implementing project activities; (ii) managing CIALCA has been very
efficient because of a clear distribution of tasks between the three projects; and (iii)
participatory rural appraisals and baseline surveys were implemented jointly by the three
projects which has resulted in a very cost-effective way to obtain the data and in tools that
are richer in terms of covering a wider range of topics important for rural livelihoods in
the mandate areas.
In terms of partnerships, (i) CIALCA partners have been actively using facilities of the
CIALCA offices (e.g., meeting venue, internet access) and (ii) visibility of CIALCA in the
region has been relatively high since the various partners openly identify themselves with
the Consortium. Partners involved in CIALCA activities are summarized as follows:
•
Belgian institutes: Katholieke Universiteit Leuven (K U Leuven), Université Catholique de
Louvain-la-Neuve (UCL), Faculté Universitaire des Sciences Agronomiques de Gembloux
(FUSAGx)
•
Non-governmental organizations (NGOs) & extension: DRC: Diobass; Bureau
Diocésaine de Développement (BDD), Association pour la Promotion de la Démocratie et du
Développement de la République Démocratique du Congo (APRODEC); Burundi: Catholic
Relief Services (CRS); Rwanda: Rwanda Rural Rehabilitation Initiative (RWARRI), Rwanda
Development Organisation (RDO)
•
National agricultural research institutes (NARS): DRC: Institut National des Etudes et de
la Recherche Agricole (INERA), Centre de Recherche des Sciences Naturels (CRSN); Rwanda:
Institut des Sciences Agronomiques de Rwanda (ISAR): Burundi: Institut des Sciences
Agronomiques du Burundi (ISABU), Institut de Recherche Agronomiques et Zootechnique
(IRAZ).
•
Regional networks: BARNESA, AFNET, FOODNET, ECABREN
•
National universities: DRC: Université Catholique de Bukavu (UCB), Université Catholique
de Graben (UCG), Université de Kinshasa (UNIKIN); Rwanda: Université Nationale du
Rwanda (UNR), Burundi: Université du Burundi (UB), other universities in East and Southern
Africa.
•
Private sector: DRC: Gourmet Gardens; Burundi: Agro-biotech, Phytolab.
•
Health partners: Rwanda: Rwandese Health Environment Project Initiative (RHEPI); DRC:
Centre Olame; DRC, Burundi: Healthnet-TPO; DRC, Rwanda, Burundi: Various health and
nutrition centers, Ministries of Health.
•
Farmer groups (FGs): DRC, Rwanda, Burundi: Various community-based organizations
(CBOs).
At the technical level, (i) the different projects have been leading activities that are relevant
for all three projects, depending on their in-house capacity; (ii) due to the wide thematic
coverage of the three projects as a whole, CIALCA activities cover all most important
realms of rural livelihoods and all major components of the farming systems in the
11
mandate areas; (iii) various MSc or PhD-related activities are co-supervised by colleagues
who have research links to several of the CIALCA projects; and (iv) within the banana
research group, there are strong complementarities in skills, which is optimally exploited
by designing collaborative research activities. Last but not least, some extra proposals have
been accepted to strengthen CIALCA activities. Examples are (i) the project on ‘Mobilizing
Innovation Platforms for Bringing More Quality Benefits to More People in Post-Conflict Central African
Great Lakes Region’, supported by the CSO-CGIAR Competitive Grants Program, (ii) the
project on ‘Amélioration de la productivité agricole en incitant l’utilisation efficace et rendable des
intrants inorganiques dans le cadre de la gestion intégrée de la fertilité de sol dans la province de Sud-Kivu
au République Démocratique du Congo’, supported by VLIR, Belgium, and (iii) a project on
‘Banana macro-propagation technology for rapid multiplication of improved banana cultivars’, supported
by BTC through ISAR (i.e. technology introduced by IITA-CIALCA, project written by
IITA-CIALCA scientist, training provided by IITA-CIALCA staff).
This report describes progress with CIALCA activities between November 2006 and
December 2007, largely covering the second year of the 3-year project, and is a follow up
on the first progress report, covering the period September 2005 – October 2006. As for
the former project, based on advice from DGDC, the current project covers the various
activities implemented by the three projects, constituting CIALCA.
12
2. BEEN
NC
CH
HM
MA
AR
RK
KA
AR
RE
EA
AC
CH
HA
AR
RA
AC
CT
TE
ER
RIIZ
ZA
AT
TIIO
ON
N
A
AN
ND
DO
OR
RG
GA
AN
NIIZ
ZA
AT
TIIO
ON
N
Based on the project proposals, several mandate areas have been identified for focussing
activities of the projects (Figure 1). Mandate areas are the geographic boundaries within
which the project will operate and criteria for choosing these were: (i) high levels of
poverty and environmental degradation, associated with low food and nutrition security,
(ii) high potential for productivity increases, (iii) good potential access to local and regional
markets, and (iv) existence of active agricultural development networks. Furthermore, in
the mandate areas, (v) the major cropping systems are based on Musa or plantain (Musa),
cassava, and legumes, which are mandate crops for the 3 institutes constituting CIALCAII, (vi) mid-season drought is occurring more frequently, induced by climate change, (vii)
social and biophysical heterogeneity is substantial, and (viii) recent civil strife has reduced
research-for-development capacity and service and market infrastructure. Mandate areas
are defined as areas with similar agro-ecological conditions and poverty profiles that have
nonetheless relatively good access to large urban markets and where the target cropping
systems are important components of smallholder farmers. Mandate areas represent large
areas (i.e. hundreds to thousands of square kilometers), often corresponding to provinces,
e.g. Kibungo (Rwanda), Gitega (Burundi), South Kivu highlands (DRC) and the number
of people living in each mandate area can vary between 100,000 and 1,000,000. All the
mandate areas have been characterized in terms of population density, altitude, agroecological potential, and access to markets (www.cialca.org).
Figure 1: CIALCA mandate areas in DRC, Rwanda, and Burundi (www.cialca.org).
Within each mandate area, action sites and satellite sites are identified based on the relative
access to markets (Table 1). Action sites are geographical areas encompassing a
community or a limited cluster of communities in each mandate area in which the field
activities related to technology identification, evaluation, and adaptation will take place in
partnership with development agencies. Action sites are selected to reflect contrasts in
13
specific key variables, presumed to substantially influence the nature of best-bet
technologies and their mode of dissemination. The number of people living in each action
site can vary between 500 and 5000. Satellite sites are similar in terms of geographical area,
population, and other general characteristics as action sites but more numerous. These
sites will be used to evaluate best-bet options, developed in the action sites, under
leadership of associate partners. The selection of satellite sites is an on-going process and .
NGO partners will especially be implicated to select suitable satellite sites within each
mandate area.
Most NRM development, testing, and adaptation work will take place in the action sites
with the satellite sites serving as a means to scale out project products. Within each action
site, baseline information has been collected on farmer typologies, within-farm soil fertility
gradients, farming systems, post harvest value addition activities and potentials, markets,
social structures, nutrition and health status, indigenous coping mechanisms, the current
contributions of legumes and bananas to human and livestock nutrition and cropping
system productivity, etc. These data will provide a framework for impact assessment later
in the project. Another important aspect of these activities has been the identification of
active farmer groups that will lead all evaluation activities, as the current project works
mainly with communities rather than individual farmers.
14
Table 1: Summary of the active Action and Satellite sites, and associations engaged in the different Mandate Areas.
Mandate area
Action sites
Satellite sites
Area
Number of associations
Area
Number of associations
Bas-Congo
Lemfu
3 (ADERKI, APEKI, ADPN)
Lemfu
30 (managed by BDD-Kisantu, in ‘groupements’
Kisantu, Kiyanika and Ngufu)
Kanga-Kipeti
2 (ADEKO, ACKI)
Muala-Nzundu 2 (managed by APRODEC)
Mbanza-Nzundu 2 (ACDPP, APDKI)
Kinkewa
1 (managed by APRODEC)
Zenga
3 (CALDZ, ADN, AFEPA)
Kiwembo
Bovin
Zenga
4 (managed by CLD, in ‘groupements’ Nkolo,
Nkolo-Tava and Makuta)
Vunda
4 (managed by CLD)
Sadi
12 (managed by CLD)
Tumba
5 (managed by BDD-Matadi)
Sud-Kivu
Kabamba
3 (Tuungane, Maendeleo, ADEPB)
Kavumu
1 (managed by DIOBASS)
Luhihi
3 (ATM, Rhusimane, Rhubehaguma)
Lurhala
2 (APACOV, CINAMULA)
Mwegerera
4 (ALEMALU, Abagwasinye, Rhuchihane, Bololoke)
Umutara
Murambi
2 (Dufashanye, Iriba)
Nyakigando
30 individual households, involved through ISAR
Rugarama
3 (Twisungane, Imbaraga, Giribakwe)
Kabarore
2 (Isoko y’ubumwe, Abahujumugambi)
Nyakigando
2 (Ingandurarugo, Dufatanye)
KigaliKabare
2 (Duterimbere A, Duterimbere B)
Kibungo
Gatore
2 (Benyshiaka, Dutabarante)
Musenyi
2 (Turwanyinzara, Turwanyubwigunge)
Mayange
2 (Tubanenabose, Abiwguruye)
15
3. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HC
CH
HA
AR
RA
AC
CT
TE
ER
RIIZ
ZA
AT
TIIO
ON
N
A
AC
CT
TIIV
VIIT
TIIE
ESS
33..AA.. C
CH
HA
AR
RA
AC
CT
TE
ER
RIIZ
ZA
AT
TIIO
ON
N SST
TR
RA
AT
TE
EG
GY
Y
Three characterization-related activities were originally planned: (i) participatory rural
appraisals (PRA), (ii) a baseline survey, and (iii) specific follow-up or detailed
characterization studies (Figure 2). As explained above, the PRAs and baseline survey
were implemented as CIALCA, through common planning and cost-sharing between the
three projects. After having delineated the mandate areas (see above), PRAs were held in
about 8 ‘villages’ per Mandate area, thereby ensuring that villages with relatively good and
relatively poor access to markets were included. ‘Villages’ corresponded with political units
containing about 500 households and were named differently in different region (the
‘village’ equivalent is underlined):
• Sud-Kivu: province - territoire - chefferie/collectivité - groupement - localité (chef, 500-1000
HH) - village
• Bas-Congo: province - district - territoire - secteur - groupement - village (chef, 500-1000 HH)
• Rwanda: province - district - secteur (chef, 500 HH) - cellule - nyumba kumi
• Burundi: province - commune - zone - colline (chef, 500 HH) - secteur
Based on the information obtained through the PRAs, Action Sites were identified
following a set of specific
Baseline study
PRA/Specific studies
criteria, including the
presence of active farmer
X Draft tools
X PRA checklist
groups, accessibility, etc. A
(incl Sampling)
X Teams
baseline survey was then
X Training/testing
X Final tools
implemented with about 2800
X Team identification
X Training/testing
households across all Action
X PRA’s/confirm sites
Sites identified, focusing on
aspects of livelihoods,
markets, nutrition, bananas,
X Community
Selected
and legumes. The baseline
selection
X Baseline
Villages
survey also aimed at collecting
implementation
X Tools for
information to enable
spec studies
X Teams
constructing farmer
X Data entry
X Training/testing
typologies, based on the
Typologies
X
Data
analysis
presence of specific
X Specific studies
production units or access to
(markets, health,
soil, legumes,
resources (e.g., land, labour,
bananas)
capital, knowledge). As a last
step in the characterization
X Baseline report
X PM&E with
work, detailed
communities
characterization studies have
been implemented to get all
required information related
Field activities
to the specific themes of the
Figure 2: Strategy for the selection and characterization of
individual projects. The
Bioversity and IITA projects the Action Sites.
16
have focused on banana production systems and access to markets while the TSBF-CIAT
project have focused on legume production and current contribution to rural livelihoods,
market access, and nutritional status of the rural population. More details related to the
baseline survey, and the specific studies are given in the sections below.
33..BB.. B
BAASSEELLIIN
NE
E SSU
UR
RV
VE
EY
YSS
The baseline surveys are a follow-up and in-depth study of the above described PRA
exercise. The baseline surveys provide counterfactual data for ex-post impact assessment
and information for priority setting and technology assessment. The data cover the
following fields: household systems and socio-economic structures, farming system
agronomics and economics, access to markets and marketing patterns for the focus crops,
post harvest handling and processing of the focus crops, social structure of households
and households’ embedding in social structures within the sites, status and determinants of
food security, and health and nutritional status. The baseline survey was implemented in
30 villages across all mandate areas and involved about 2800 households (Table 2).
Table 2: Baseline sites and sample size.
Country
Action sites
DR Congo Sud-Kivu montagneux
Nord-Kivu montagneux
Bas-Congo
Rwanda
Umutara
Kigali-Kibungo
Gitarama
Kibuye-Gisenyi
Burundi
Gitega
Kirundo
Ruzizi plains
TOTAL
Villages
5
4
4
4
4
2
2
2
2
1
30
Households/village
100
100
100
50
100
100
100
100
100
100
Total
500
400
400
200
400
200
200
200
200
100
2,800
The baseline data has been entered into the SPSS software package. Copies of the database
containing all data have been given to all national partners. The major findings will be
published in a book in the course of 2008. A draft report summarizing the major findings
is already available, and some highlights and examples of these findings are provided
below for the major themes covered by the baseline survey:
•
•
•
•
•
•
•
Household systems and social and economic structures, including assets and financial
resources
Farming systems agronomics and economics
Access to markets and marketing patterns for the focus crops (banana, legumes, cassava)
Post harvest handling and processing of the focus crops
Social structure of households and households’ embeddedness in social structures
Status and determinants of food security, including food consumption patterns and nutrition
Health (including access to and utilization of health services)
Household structure and economics
Some key household characteristics are shown in Table 3. A relatively high percentage of
households have attained at least primary level of education in all the mandate areas.
However Sud-Kivu showed only 28.4% of the households having attained primary
education. Literacy level was also substantially lower in Sud Kivu (67.2) compared to
Rwanda East and Ouest (>94.5). The high percentage of respondents (i.e. mostly HH
17
head or spouse) having no formal education in areas such as Cibitoke (53.2%) and Sud
Kivu (47.4%) is likely to present some bottlenecks in technology uptake in these areas.
Bas-Congo
Nord-Kivu
Sud-Kivu
Est
Ouest
Sud
Total
Rwanda
Kirundo
HH head gender (%)
Male
78.2
Female
21.8
Level of education (%)
No formal education
53.2
Adult alphabetization
2.1
Primary education
36.2
Secondary education
7.4
Other
1.1
Literacy of respondent (%)
Literate
82.6
Non Literate
17.4
Age of respondent
45.0
Household size
5.5
DR Congo
Gitega
Cibitoke
Table 3: Household characteristics
Burundi
80.6
19.4
92.1
7.9
82.7
17.3
85.2
14.8
84.8
15.2
80.5
19.5
80.9
19.1
82.0
18.0
83
17
36.0
10.8
45.2
2.7
5.4
39.8
7.6
40.7
5.9
5.9
13.8
1.0
32.2
40.6
12.3
39.9
0.3
45.7
13.8
0.3
47.4
5.5
28.4
17.8
0.9
28.6
5.8
53.6
10.8
1.2
20.2
4.0
64.6
10.1
1.0
14.9 32
5.9 4.3
73.3 44
3.0 16
3.0
3
82.8
17.2
44.5
6.6
83.7
16.3
40.2
6.2
81.3
18.7
42.8
5.6
78.0
22.0
41.1
6.4
67.2
32.8
45.2
7.0
94.5
5.5
42.5
5.8
97.3
2.7
43.0
6.4
85.7
14.3
39.9
6.3
60
40
43
6
The average number of people with in each household is about 6 persons however, SudKivu shows a higher household size of about 7 persons while Cibitoke showed the lowest of
about 5 persons. Overall, the average age of the household head is about 43 years which
indicates that most of the household heads are adults that can take informed household
decisions. However, household members under the age of 18 take up the largest
composition (45.1-50.3%) of household members at all sites. The dependency ratio which
represents the economic burden for resources imposed on the working population was
computed as follows: Dependency ratio equals the number of individuals aged below 18 or
above 59 years, divided by number of individuals aged 18 – 59 years. The results indicate
that there is a high dependency ratio in all CIALCA sites; there are more people who are not
of working age and that need to be looked after. This was especially the case in the provinces
of Kirundo, and Nord & Sud-Kivu which have a dependency ratio of 1.17, 1.27 and 1.34,
respectively.
Social capital
Membership to a professional organization is of great importance in that households
benefit from these associations through collective marketing, credit access, sharing of new
ideas and experiences, and a high level of output and return on goods sold through better
bargaining power. Figure 3 highlights farmer membership to an association in Burundi,
Congo and Rwanda. The results reported indicate higher percentages (about 70%)
reported in Rwanda provinces for respondents who belonged to a health insurance group.
18
Other countries
however showed
minimal or no
involvement in a health
insurance group. The
East province in
Rwanda indicated that
about 41% of
respondents belonged
to a self help group.
Higher proportions of
42% and 45% of
respondents in Ouest
and Sud provinces
respectively belonged
to a credit and savings
group. Fewer
proportions of
respondents belonged
to a women’s group.
Figure 3: Household membership of social groups
Agriculture
Table 4 shows the average household land holding and land holding by land type and
province. Results show that households in Bas-Congo have a larger farm size of 2.12
hectares followed by Cibitoke at 1.97 hectares. Households in Gitega have the smallest
farm size. Households in Cibitoke had the largest farm size for field on hill (0.96ha) and
homestead area of (0.45ha). Households in Bas-Congo had the largest farm size for field
under marsh (0.40ha) and forested area (0.44ha) accounting for 18.9% and 20.8% of total
farm size respectively. Area under grazing fields was reported to be largest in Nord-kivu
(0.60ha). Cropped area that is, homestead area, field under marsh and field on hill account
for the biggest proportion. These figures show that land pressure is very high, especially in
Gitega, Kirundo, Sud Kivu and Ouest Rwanda.
Table 4: Household land
Land type
Homestead Field under Field on
area
marsh
hill
Burundi
Cibitoke
Gitega
Kirundo
Congo
Bas-Congo
Nord-Kivu
Sud-Kivu
Rwanda
Est
Ouest
Sud
Forested
area
Grazing
field
average
farm size
(ha)
0.45
0.10
0.20
0.06
0.04
0.04
0.96
0.09
0.24
0.36
0.05
0.06
0.14
0.22
0.37
1.97
0.50
0.91
0.36
0.19
0.3
0.40
0.34
0.20
0.53
0.50
0.34
0.44
0.19
0.09
0.39
0.60
0.20
2.12
1.82
1.13
0.25
0.27
0.33
0.17
0.12
0.16
0.62
0.44
0.95
0.12
0.15
0.10
0.26
0.09
0.40
1.47
1.07
1.94
19
Total
Table 5: Types of organic inputs applied in homestead plots
% of HH
Burundi
DR Congo
Rwanda
Cibi- Gite- Kirun- BasNord- SudEst Ouest Sud
toke
ga
do
Congo Kivu Kivu
Nothing
97.3
58.3
90.2
97.0
76.0
44.0
77.1
64.3
54.9
Manure
0
1.4
4.9
0
0
0
0
0
0
Compost
2.7
38.9
4.9
3.0
4.1
42.7
21.3
31.8
41.2
Biomass
0
1.4
0
0
15.7
9.2
1.4
3.9
3.9
Other
0
0
0
0
4.1
4.1
0.3
0
0
67.0
0.3
26.1
5.1
1.6
Homestead plots in areas with very high land pressure, such as Gitega, South Kivu, and
Rwanda Ouest, generally received more often (>35% of farmers) organic nutrient inputs
than plots in areas with less high land pressure such as Bas Congo and Cibitoke (<3%).
Manure is seldom applied in a pure
form, but is often mixed into
compost or left unused near the
kraal (Table 5). The availability of
manure is closely linked to the
number of cattle owned. Livstock
numbers are generally low at all
sites, with some 15% of farmers
owning cattle on average. However,
cattle ownership was particularly
low in Gitega, Cibitoke and DRC,
where 10% or less of the farmers
owned cattle (Figure 4). These
findings highlight the need for
CIALCA to look for biomassrelated methods to improve nutrient
recycling and availability to improve
Figure 4: Cattle ownership
crop production.
Table 6: Value of output by crop grown per season (USD)
Burundi
DR Congo
Cibi- Gite- Kirun- Bas- Nord- Sudtoke
ga
do
Congo Kivu Kivu
Banana beer
40.5
13.2
17.6
0
9.4
14.0
Dessert banana
0
1.9
11.0
9.0
2.7
4.3
Cooking banana
57.2
13.2
17.6
4.9
12.6
5.2
Plantain banana
0
0
0
17.9
10.7
4.5
Bitter cassava
18.6
5.2
11.6
43.7
14.3
22.6
Sweet cassava
0
29.0
0
35.8
1.4
35.8
Cassava leaves
0
1.9
8.6
5.4
1.1
0
Fresh beans
0
0
0
0.4
0
5.4
Grain beans
17.6
1.3
4.2
26.2
10.7
10.0
Bean leaves
0
0
0
0
0
17.9
Green beans
0
0
0
0
0
53.7
Ground nuts
0
8.4
3.2
15.2
10.7
15.2
Soybean
0
0
0
8.6
3.6
9.7
Cowpea
0
6.6
0
1.6
0
0
Cowpea leaves
0
0
0
0
0
0
20
Est
13.8
8.7
23.0
0
1.6
0.7
0.2
0.6
0.6
0.6
2.0
2.4
2.3
0
1.8
Rwanda
Ouest
Sud
41.4
18.4
70.0
0
1.1
1.5
0
19.3
0.8
0
2.8
0
0.5
0
0
12.1
24.8
23.0
0
2.8
1.0
0
0
1.0
0
1.3
22.1
9.4
0
0
In terms of cropping systems and the value of the crops produced, the baseline confirmed
that banana and legumes are amongst the dominant crops at all sites (Table 6). Bananas
are the primary crop in terms of output value in Cibitoke, Kirundo, Nord Kivu, and all
Rwandan sites. Legumes are very important in all sites, and cassava is particularly
important in Gitega, Bas Congo, and South Kivu.
Table 7: Most dominant market outlet used to sell farm produce.
% of HH
Burundi
DR Congo
Rwanda
Cibi- Gitega Kirun Bas- Nord- SudEst Ouest Sud
toke
-do Congo Kivu Kivu
On farm
7.6
17.0
18.9 31.1
34.8 22.4
21.5 22.2
13.5
Local market
37.5 47.9
63.3 31.4
21.4 30.3
44.4 25.1
53.4
Neighbor market 27.8 22.9
15.6 14.0
25.6 23.3
20.1 26.7
23.6
Urban market
16.7 12.2
2.2
19.4
12.5
5.7
6.9
17.3
4.1
Regional market
10.4
0
0
4.2
5.7
18.2
7.1
8.6
5.4
Total
Market access, agricultural marketing, post harvest processing and handling
Markets outlets that are dominantly used by farmers to sell their products are the farm gate
and the local markets (Table 7). Urban and big regional markets are still important for a
substantial portion of the farmers (>20%) in areas that are relatively close to big cities,
such as Cibitoke (Bujumbura), Bas-Congo (Kinshasa), South Kivu (Bukavu), and Rwanda
Ouest (Kigali). However, in general most households have poor access to the big regional
or urban markets, and this was particularly true for the more isolated areas such as
Kirundo, which have no big city or big regional market nearby.
24.5
35.0
21.5
11.3
7.7
Food and nutrition, food security
Food insecurity is a major problem for over 40% of all households in the CIALCA sites
(Table 8). However, large differences exist, and areas with particularly high levels of food
insecurity (>60% food insecure) are Gitega and Kirundo in Burundi and South Kivu in
DR Congo. Female headed households are in general more food insecure than male
headed households. These are also the areas with very high land pressure and relatively
little land available per household.
Total
Table 8: Percentage of male and female headed households that are food-insecure for at
least part of the year.
% of HH
Burundi
DR Congo
Rwanda
Cibi- Gitega Kirun- Bas- Nord- SudEst Ouest Sud
toke
do Congo Kivu Kivu
Male headed
34.0
61.2
62.7
47.5
8.8
56.4
33.3
28.2
47.4 40.0
Female headed 46.5
67.0
65.1
58.6
11.6
67.8
36.1
48.3
40.9 46.2
Health
We recorded the illnesses affecting children under 5 years (Table 9) and adults over 18
years of age. Illnesses can be a major set back to the household’s progress towards low
mortality rates, food security and poverty reduction. Among the illnesses that were mostly
afflicting the children and adults were malaria/headache. Higher proportions of above
80% were reported for adults over 18 years faced with malaria in the areas of Cibitoke,
Gitega, Kirundo, Est, Ouest and Sud compared to 45% in Bas-Congo which was the
lowest. The highest proportion among children less than 5 years faced with malaria was
reported in Rwanda Est and Ouest (78%) and the lowest (38%) was recorded in NordKivu. Diarrhea for children under 5 was particularly important (>12% of HH) in Burundi
and Nord Kivu. The data suggests that health, food security, and the available land for
21
Table 9: Percentage of households that recorded illnesses for children under 5 years
% of HH
Burundi
DR Congo
Rwanda
Cibi- Giteg Kirun Bas- Nord- Sud- Est Ouest Sud
toke
a
-do Congo Kivu Kivu
fever/flu
3.9
19.0 10.5 29.9
3.5
6.8
3.2
0.7
5.3
vomiting
0
0
0
0
0
3.1
1.4
0.7
0
worms
2.0
0
3.5
1.9
0
8.6
8.8
12.8 13.3
malaria/headaches
68.6 43.1 68.6 47.5 37.8 48.8 77.9 78.0 69.3
cough/bronchitis/tbc
0
0
0
10.2
1.8
18.5
3.7
3.5
1.3
stomach
0
0
0
1.3
0.4
2.2
0
0
0
diarrhea
13.7 26.7 12.8
1.9
12.0
4.0
1.6
2.1
2.7
others
11.8 10.3
3.5
2.9
44.5
7.1
3.0
2.1
8.0
measles
0
0
0
4.5
0
0
0
0
0
Total
each household are strongly related; e.g., areas such as Gitega, Kirundo, and South Kivu
are amongst the highest in terms of diseases recorded, while also having the highest levels
of food insecurity and the smallest average land holdings.
9.8
0.9
5.7
58.0
6.5
0.7
6.3
11.0
0.8
The way forward with the baseline study
Although a draft report with results of the baseline survey is available, CIALCA would like
to publish a book with the major findings and descriptives for this study. This book should
be printed and released in the course of 2008. In addition, we will continue to further
explore the data with in-depth studies on the following topics:
• Production systems of the focal crops
• Marketing systems of the focal crops, in particular bananas
• Potential for processing and value adding, in particular for bananas
• Sociology, nutrition and health studies
22
33..CC.. F
FIIN
NA
AL
LC
CH
HA
AR
RA
AC
CT
TE
ER
RIIZ
ZA
AT
TIIO
ON
NA
AC
CT
TIIV
VIIT
TIIE
ESS
3.C.1. DETAILED CHARACTERISATION STUDY ON LEGUME
PRODUCTION, MARKETING AND CONSUMPTION, AND
NUTRITIONAL STATUS OF RURAL HOUSEHOLDS
A detailed characterisation study was
conducted during June and July 2007 in the
mandate areas of the TSBF-CIAT project.
Prior to the implementation of the study, a
two-day training and discussion session was
organized at the CIALCA office in Kigali on
24-25 May 2007, involving agronomists,
socio-economists and nutritionists from the
various regions. Local training sessions were
then organized with poll-taker teams of the
various disciplines, followed by trial runs with
a number of households in the field
(Photograph 3).
Photograph 3: Field testing of the survey
with chief enumerators.
This study complemented the earlier conducted baseline survey with quantitative
information on aspects of legume cropping, soil fertility status, marketing of legume
products and nutritional status of rural livelihoods. Households were randomly selected
within the set of households interviewed during the baseline survey. Between 15 and 20
households in each of the 4 action sites in all mandate areas were fully characterized;
nutritional status was evaluated in twice as many households. The characterisation study
entailed detailed questionnaires with farmers, soil and plant sampling, agronomic
measurements in legume-grown fields, collection of essential socio-economic data for
market chain analysis, anthropometric measurements in children between 2-5 years old,
and an assessment of dietary intake and diversity. The questionnaire used is presented in
Annex 2. Data entry has currently been concluded and some preliminary analyses have
been conducted (presented below). In depth examination will involve factor and
multivariate analysis. Soil and plant sample analysis are at this time pending.
Legume production
In each household, a map of the farm was drawn by the household head, indicating the
location of the manure/compost storage system, livestock facilities and the different fields,
relative to the homestead. The farmer was asked to specify the crops grown in the past
two seasons. All fields cultivated by legumes were than highlighted and visited by an
agronomist and a household member (see example in Figure 5). Detailed measurements
were taken, including soil sampling, and measurement of crop and weed densities. Farmers
gave details on crop management and inputs applied, appraised the soil fertility according
to the local classification system and indicated the major constraints for crop production in
the fields. Finally, compost and manure facilities were sampled for determination of
organic matter quality and nutrient contents, and legume grain samples were taken for
analysis of nutritional quality. Analysis results are pending.
23
SK/02/54
livestock
2 cows, grazing, with manure collection
7 goats, tethered, daily manure collection
organic inputs
compost (FYM, crop residues, green
manure, HH waste)
2007 A
banana
2007 A
cassava
2007 B
banana
2007 B
cassava
2007 A
beans
P7
10km
80m
P5
400m
120m
P1 (civu)
owned
P2 (civu)
owned
160m
P6 (civu)
owned
2007 B
sorghum
cocoyam
beans
320m
P3 (civu)
owned
2007 A
banana
soybean
beans
2007 A
beans
2007 B
banana
beans
2007 A
banana
P4 (civu)
owned
2007 A
soybean
sugarcane
2007 B
beans
maize
cassava
2007 B
banana
beans
2007 B
beans
sorghum
sugarcane
Figure 5: An example of a farm map of relatively wealthy household in Luhihi (Sud-Kivu),
with livestock and manure facilities, and 7 fields (all owned by the household). Five fields
are cultivated by legumes (yellow fields are of medium fertility and green fields of high
fertility, according to the farmer’s appraisal).
Common legume systems differ between the mandate areas (Figure 6). Legumes are
commonly associated with cassava in Bas-Congo, with cassava and/or sweet potato in
Sud-Kivu, and with cereals in Umutara. In Kibungo, legumes are frequently grown in
association with both cereals and banana. Mixed cropping systems, with 3 or more crop
types grown in the same field, are also common (except in Bas-Congo); these include
mostly associations of root and tuber crops with cereals and legumes, and to a lesser extent
banana with cereals and legumes (except in Kibungo). Legume mono-cropping is
uncommon and almost never practiced during two consecutive seasons; farmers are aware
of the disease accumulations, particularly for beans. Pure rotation systems are likewise rare,
and some legumes are usually planted in association during the cereal season. Planting in
line is very rarely practiced for legumes or cereals; seeds are usually simply broadcast.
24
Kibungo and Bugesera
Umutara
Bas-Congo
Sud-Kivu
cereal-legume association
cereal-legume rotation
root/tuber-legume association
banana-legume association
legume mono-cropping
mixed cropping
Figure 6: Relative importance of common legume production systems in the 4 mandate
areas of the TSBF-CIAT project.
In Sud-Kivu and Rwanda, more than 70% of the legume-grown fields are positioned on
slopes. While in Rwanda, conservation structures are common and well-maintained, in
Sud-Kivu these are almost entirely absent. Almost 90% of the legume-grown fields on
slopes are unprotected (only some physical embankments without hedgerows were
observed), and two thirds of the fields show visible signs of erosion. Farmers however
consider low soil fertility, drought and climatic variability as the major constraints for
legume production.
Legume commercialisation
Legume varieties were characterized and farmers specified the minimal and maximal prices
at which they sold their legume grains during the year on the local market. A preliminary
analysis was conducted in Sud-Kivu (Figure 7).
Groundnuts are primarily produced in Kabamba, and are sold at a much higher price (min.
price = 1.6 $ kg-1) than beans and soybean (min price = 1.0 $ kg-1 and 0.6 $ kg-1,
respectively). Maximal groundnut purchase prices in Kabamba are 1.9 $ kg-1 during periods
of scarcity. Soybean prices differ between sites. Prices are lowest in Kabamba and Luhihi
(on average 0.65 $ kg-1), where soils are relatively more fertile and soybean is more
commonly produced than in the Walungu area (Lurhala and Mwegerera). Minimal soybean
purchase prices in the Walungu area are almost twice as high as on the northern axis
(Kabamba and Luhihi). For beans, purchasing prices fluctuate around 1 $ kg-1, and are
slightly lower in Lurhala. Maximal price increases during periods of scarcity occur in
Kabamba (up to 0.5 $ increase kg-1). Price differences in space and time open up a number
of marketing opportunities through transportation and storage, allowing producers to sell
where and when the price is highest. Moreover, prices depend on grain traits such as size
and colour. Prices for grains of the preferred colour (red and white) are higher than for
less-preferred colours (black), and usually increase for larger grain sizes.
25
-1
maximal price increase during year (USD kg )
-1
minimum purchase price (USD kg )
2.0
A
1.5
a
a
ab
Aab
Bbc
1.0
c
Bbc
Cc
0.5
0.0
groundnut
beans
2.0
Kabamba
Luhihi
Lurhala
1.5
Mwegerera
1.0
Aa
0.5
b
Ab
Ba
a
c
a
Bb
0.0
soybean
beans
soybean
Figure 7: Minimal legume grain purchasing prices (left) and price increases during
periods of scarcity (right) as reported by farmer-producers in the 4 action sites in the SudKivu mandate area. Letter labels indicate a significant (P<0.05) difference in price
between species (capital letters) or sites (lower-case letters).
50
Rwanda
girls
40
boys
30
20
10
proportion of age group showing
symptoms of malnutrition (%)
Nutritional status of rural households
In Sud-Kivu, malnutrition is very prevalent in younger children; more than 30% of 2- to 3year-old children show at least mild symptoms of marasmus or suffer from kwashiorkor
(Figure 8). Lack of muscular tissue, swollen abdomen, wrinkled or flaky skin, and scanty,
pale hair were the most commonly observed symptoms. Relatively less symptoms of
malnutrition were observed in Rwanda (across both mandate areas). In Bas-Congo,
symptoms of kwashiorkor or marasmus were very rarely observed. In Rwanda and in BasCongo, malnutrition was more pronounced in girls than in boys, while in Sud-Kivu,
symptoms of marasmus or kwashiorkor were more often observed in boys than in girls.
0
Sud-Kivu
girls
40
boys
30
20
10
0
24-35
months
50
50
36-47
months
48-59
months
Bas-Congo
60-71
months
24-35
months
36-47
months
48-59
months
60-71
months
girls
40
boys
30
20
10
0
24-35
months
36-47
months
48-59
months
60-71
months
Figure 8: Prevalence of malnutrition symptoms (marasmus/kwashiorkor) in 2- to 5-yearold children of rural households in Rwanda, Sud-Kivy and Bas-Congo.
26
Anthropometric measures were taken in 2- to 5-year-old children, and included the weight,
height and mid-upper-arm circumference (MUAC). In Sud-Kivu, depending on the
measures and cut-off points used, between 2 and 12% of the children suffer moderate
acute malnutrition, and between 14 and 25% are at risk of malnutrition (Figure 9). In
Rwanda, anthropometric measures identified very few children with moderate
malnutrition, and up to about 10% were at risk. In Bas-Congo, somewhat conflicting
results for malnutrition prevalence were observed for the both measures. While MUAC
measurements suggested mild and moderate malnutrition in over 20% and almost 10 % of
2- to 5-year-old children, respectively, malnutrition was almost entirely absent according to
weight-for-height (WFH) measurements. Estimates of malnutrition rates tend to be larger
using MUAC measurements than WFH measurements (Bairagi and Ahsan, 1998).
80
(a)
Rwanda
girls
boys
60
40
20
100
Malnutrition prevalence in children
according to WFH (%)
according to MUAC (%)
100
0
11-12.5cm
12.5-13.5cm
(c)
Sud-Kivu
girls
60
40
20
80
-3<z-score<-2 -2<z-score<-1
-1<z-score
(d)
Sud-Kivu
girls
boys
60
40
20
0
11-12.5cm
12.5-13.5cm
>13.5cm
z-score<-3
(e)
Bas-Congo
girls
boys
60
40
20
100
<11cm
20
100
0
80
40
z-score<-3
boys
100
boys
60
>13.5cm
80
(b)
girls
0
<11cm
100
80
Rwanda
0
80
-1<z-score
(f)
Bas-Congo
girls
boys
60
40
20
0
<11cm
11-12.5cm
12.5-13.5cm
>13.5cm
z-score<-3
-1<z-score
Figure 9: Malnutrition prevalence according to (a,c,e) mid-upper-arm circumference
(MUAC) and (b,d,f) weight-for-height (WFH) in 2- to 5-year-old children of rural
households in Rwanda, Sud-Kivu and Bas-Congo. Categories signify (from left to right):
severe, moderate, mild and absent acute malnutrition (Lancet and Morley, 1974); z-scores
are relative to the median of the respective populations.
27
Legume consumption in Sud-Kivu
Diets of 2- to 5-year-old children were determined by asking mothers to recall the foods
given to their child during the past week and month. Presented below are the data for the
action sites in the Sud-Kivu mandate area.
Legumes constitute the principal source of protein for children in Sud-Kivu. In Lurhala
and Mwegerera, more than 50% of the children consume meat- or fish-derived protein less
than once a week, and usually only once or twice a month (Figure 10). About 30% of the
households feed their children with meat or fish once or twice a week. In Kabamba and
Luhihi, which are close to the lakeside, consumption of small fish (‘frétins’) is very
common, but meat is less frequently eaten. The consumption of eggs is very rare in all
sites. Among the four legumes grown in the area, soybean and bean are the most
important legumes in the diets of young children. Cowpea is rather uncommon and
groundnut is principally grown as a cash crop. In Luhihi, a productive soybean area, about
three quarters of the children are fed with soybean more than five times a week (Figure
11). Soybean is either prepared as soymilk, tea, or porridge (commonly mixed with maize
and/or sorghum). In the other sites, soybean remains an important constituent of young
children and is consumed on average 3-4 times per week. Beans are also important, and
fed at least once or twice a week in 70% of the households.
Kabamba
Luhihi
100
proportion of children (%)
100
80
60
40
20
0
80
60
40
20
0
0
1-2
3-5
>5
-1
frequency of consumption (week )
0
Lurhala
Mwegerera
100
100
1-2
3-5
>5
-1
80
60
40
20
0
fish
meat
80
eggs
60
40
20
0
0
1-2
3-5
>5
-1
0
1-2
3-5
>5
-1
Figure 10: Frequency of consumption of meat, fish and eggs by 2- to 5-year-old children
in the 4 action sites in the Sud-Kivu mandate area.
28
Kabamba
100
100
80
60
40
20
0
60
40
20
0
1-2
3-5
>5
-1
Lurhala
100
80
0
0
100
Luhihi
80
60
40
20
0
1-2
3-5
>5
-1
Mwegerera
beans
80
soybean
60
40
20
0
0
1-2
3-5
>5
-1
0
1-2
3-5
>5
-1
Figure 11 Frequency of consumption of beans and soybean by 2- to 5-year-old children in
the 4 action sites in the Sud-Kivu mandate area.
3.C.2. DETAILED CHARACTERISATION STUDY ON BANANA
PRODUCTION, MARKETING AND CONSUMPTION
The objective of the farm-level diagnostic studies was to identify and quantify the
biophysical and economic parameters driving the banana cropping systems. During the
CIALCA planning meetings, the objectives, tools and sampling strategies for this activity
were further developed and the final tool was completed and field-tested in October 2006.
Subsequently, staff and students executing the diagnostic surveys in North Kivu, Rwanda,
South Kivu, and eventually Burundi all underwent a minimum 2 day on-site training by
IITA-Uganda research staff. At each action site, 30 farms representing a subsample of the
baseline survey were visited. The 30 farms per site represented 10 wealthy, 10 medium,
and 10 poor farmers, following local wealth classifications as set during the PRA’s and as
identified by key local resource persons. This sampling framework resulted in a total
sample size of over 540 farms in 18 sites.
Banana production and constraints and farmer perceptions.
Banana productivity averaged by site ranged from 21-43, 25-53 and 35-63 Mg ha-1 cycle-1
for Burundi, Rwanda, and South Kivu, respectively. With the cycle duration (i.e. period
between two harvests from a single mat) averaging about 1 year, these production levels
are much higher than those reported in general by FAO for the region (6-15 Mg ha-1 yr-1).
Banana pest (i.e. nematodes and weevils) and disease (i.e. Black Sigatoka) damage
parameters varied distinctly between regions, but were strongly negatively related to
altitude. Fusarium wilt was widespread and limited productivity of exotic cultivars (i.e.
Pisang Awak, Apple banana, Gros Michel). Banana Xanthomonas wilt (BXW) severely
affected production in certain sites in Central Uganda, Western Rwanda and Eastern DRC,
while banana bunchy top virus (BBTV) was observed in the Rusizi valley. However, BXW
29
and BBTV currently affect a limited geographic area and still have relatively small impact
on national banana production. Highest productivity was observed near the Albertine rift,
where soils are relatively young, rainfall is high (> 1400mm yr-1), and plant densities are
high (1800-3300 mats ha-1), compared to much of Eastern and Central Rwanda and
Burundi, which have strongly weathered soils, low rainfall (<1100mm yr-1), and low plant
densities (1000-1700 mats ha-1). These findings suggest that much of the current research
focus is based on flawed data and assumptions. Water stress, which has previously been
overlooked, is an important production constraint, while several ‘traditional’ banana pests
and diseases may not be as important as is often assumed.
Lab analysis of foliar samples revealed that nutrient deficiencies such as P and Mg are
indeed predominant on strongly weathered soils, such as the Walungu axis south of
Bukavu (Figure 12). Potassium deficiency generally predominates on soils that have a low
stock of weatherable nutrients (quartzite and granite) such as in Kibuye, Ruhango (both
Rwanda) and Gitega (Burundi). Areas that know high levels of crop management and
where external nutrient inputs (e.g. external mulch) is frequently applied (e.g. Kibungo)
tend to have less nutrient deficiency problems.
5.0
5.0
4.5
4.5
4.0
4.0
3.5
3.5
3.0
2.5
N (% dry matter)
K (% of dry matter)
3.0
2.5
2.0
1.5
2.0
1.5
1.0
go
an
uh
R ga
n
ze
N oli
un
M ala
rh
Lu hi
hi Lu do
n
ru
K i ye
bu
Ki ngo
bu
Ki a
liv a
K a mb
ba 2
Ka gaite 1
G gaite gu
G an
n
ya e
C tok
o
ib
C ale
rh
Bu o
ng
Bi
go
an
uh
R ga
n
ze
N oli
un
M ala
rh
Lu hi
hi Lu do
n
ru
Ki y e
bu
Ki ngo
bu
Ki a
liv a
Ka mb
ba
Ka a-2
g
ite 1
G gaite gu
G an
n
ya e
C tok
o
ib
C ale
rh
Bu o
ng
Bi
.5
1.2
1.0
.4
.8
.6
MG (% dry matter)
P (% dry matter)
.3
.2
.1
.4
.2
0.0
go
an
uh
R ga
n
ze
N oli
un
M ala
rh
Lu hi
hi Lu do
n
ru
K i ye
bu
Ki ngo
bu
Ki a
liv a
K a mb
ba 2
Ka gaite 1
G gaite gu
G an
n
ya e
C tok
o
ib
C ale
rh
Bu o
ng
Bi
go
an
uh
R ga
n
ze
N oli
un
M ala
rh
Lu hi
hi Lu do
n
ru
K i ye
bu
Ki ngo
bu
Ki a
liv a
K a mb
ba
Ka a-2
g
ite
G a-1
g
ite gu
G an
n
ya e
C tok
o
ib
C ale
rh
Bu o
ng
Bi
Figure 12: Foliar nutrient concentrations (in % dry matter weight) for the diagnostic
survey sites in Burundi (Gitega, Cibitoke, Kirundo), South Kivu (Burhale, Kabamba,
Luhihi, Lurhala), North Kivu (Bingo, Kaliva, Munoli, Nzenga), and Rwanda (Cyangugu,
Kibungo, Cyangugu, Ruhango)
We also compared farmers’ perceptions of constraints and coping strategies related to
banana production, and verify these with field assessments. Average farmer bunch weight
estimates were variable, ranging from large (76%) over estimations in Bugesera (Rwanda)
to large under estimations (63%) in Karongi. In most Rwandan and Burundian sites,
30
Yield (t/ha/cycle)
farmers reported drought stress (85%)
90
and poor soil fertility (74%) as major
80
constraints to banana productivity,
70
whereas a minority (26%) of farmers
mentioned pest and diseases as major
60
constraints. Farmer perceptions are in
50
line with our assessments. Drought stress
40
yield losses based on average rainfall
30
were estimated at between 30 and 70%
20
for sites in central and eastern Rwanda
and Burundi. Banana weevil
10
(Cosmopolites sordidus) damage was low
0
at all sites (<3.5%) and root necrosis
0
100
200
300
400
500
600
moderate (13 – 23%), with the exception
Penetrometer resistance
of Cibitoke where root necrosis was only Figure 13 Penetrometer measurements from
8%, due to the dominance of the
Rwanda, Burundi and South Kivu suggest that
nematode resistant Yangambi km5
maximum attainable yields (t/ha/cycle)
decrease with increasing penetrometer
cultivar. Fusarium wilt strongly affected
resistance.
exotic banana production (i.e., AB and
ABB beer and dessert bananas) in all sites and 10– 67% of plantations. Farmers have
therefore resorted to replacing Kayinja (ABB) with AAA-EA beer bananas. Farmers’ crop
management was varied. Banana yield was positively correlated (r2=0.13, P<0.001) with
amount of mulch applied. However, few farmers (22%) applied external mulch. While
most farmers interviewed (86%) owned some cattle or small ruminants, only a minority of
them (39%) reported application of manure/compost. None of the farmers applied
mineral fertilizers. This study shows that farmers correctly perceive abiotic stress factors as
the most yield limiting. However, only a minority of farmers seem to adopt technologies
(i.e., application of mulch, manure, and compost) to overcome these yield-limiting factors.
31
4. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
H SST
TR
RA
AT
TE
EG
GIIC
C SSO
OIIL
L
F
FE
ER
RT
TIIL
LIIT
TY
Y-R
RE
EL
LA
AT
TE
ED
DA
AC
CT
TIIV
VIIT
TIIE
ESS
44..AA.. R
REELLAATTIIO
ON
NSSH
HIIPP B
BE
ET
TW
WE
EE
EN
N SSO
OIIL
L FFE
ER
RT
TIIL
LIIT
TY
YA
AN
ND
D
N
NU
UT
TR
RIIT
TIIO
ON
NA
AL
LQ
QU
UA
AL
LIIT
TY
YO
OFF B
BIIO
O--FFO
OR
RT
TIIFFIIE
ED
DB
BE
EA
AN
NSS
A G by E analysis of Fe contents in grains of beans grown in Sud-Kivu and Umutara
In the legume evaluation trials, 27 bush bean and 9 climbing bean varieties, along with the
local variety, were tested by 2 farmer associations in 4 sites in Sud-Kivu, DRC and in 4
sites in Umutara, Rwanda (16 associations in total). In each association, separate blocks
were set up with a control and a treatment with goat manure application at 5 t dry matter
(DM) ha-1. After harvest, a representative sample of grains was taken, oven-dried and
manually ground using an agate mortar and pestle. A subset of the grain samples was
analysed for Fe contents using radial ARL ICP-AES by ARI laboratories, Adelaide.
A preliminary analysis was conducted by fitting following general linear model:
Fe_Ca,t,g = µ + αa + βt + γg + θa,t + θ’g,t + εa,t,g
with Fe_Ca,t,g = grain Fe content for genotype ‘g’ in association ‘a’ with treatment ‘t’ [mg Fe kg-1], µ = grand mean, αa =
environment (association) mean deviations, βt = treatment mean deviations, γg = genotype mean deviations, θa,t =
association × treatment interaction residuals, θ’g,t = genotype × treatment interaction residuals and εa,t,g = the error term;
the association CINAMULA (Lurhala) was excluded from the model as no observations in the control treatment were
available.
Analysis of variance demonstrates that grain Fe content is largely determined by
environment (association) and genotype and unaffected by FYM application (Table 10).
Genotypic and environmental effects alone can explain 17 and 44% of the total variation,
respectively. Interaction effects between genotype and environment are part of the error
term (35.1% of the total variation).
Table 10: ANOVA for significance of genotype, environment (association), treatment and
environment × treatment and genotype × treatment interaction effects on Fe contents in
bean grains
source of variation
df
SS
MS
F-value
P-value
% of total SS
total
214
27684.3
model
53
17971.4
339.1
5.62
<0.0001
64.9
environment (assoc.)
10
12304.1
1230.4
20.40
<0.0001
44.4
treatment
1
0.4
0.4
0.01
0.9335
0.0
genotype
16
4682.9
292.7
4.85
<0.0001
16.9
environment × treat.
10
534.3
53.4
0.89
0.5480
1.9
genotype × treat.
16
449.6
28.1
0.47
0.9599
1.6
error
161
9712.9
60.3
35.1
Bean varieties Marungi and ARA4 generally contained highest grain Fe contents while
BRB194, CIM9314-36 and Kiangara contained lowest grain Fe contents (Table 11).
Highest Fe contents were observed in the ALEMALU and APACOV associations (which
are rather infertile sites), while lowest Fe contents were observed in the MAENDELEO,
RUSINAME and IRIBA associations (which are rather fertile sites).
32
Table 11: Adjusted grain Fe content means and standard errors for the different bean
varieties (across associations) and for the different associations (across varieties)
association
action site
mean
SE
species variety
mean
SE
ALEMALU
Lurhala
82.55
3.11 BB
Marungi*
74.06
1.91
APACOV
Burhale
77.58
1.64 BB
ARA4*
72.91
1.85
DUFATANYE
Nyakigando 73.24
1.39 BB
ZAA5/2
72.82
2.11
TWISUNGANE
Rugarama
71.53
1.90 BB
ECAPAN021
72.44
2.04
RHUBEHAGUMA Luhihi
71.00
2.01 CB
VCB81013*
72.36
2.80
ABAGWASINYE Burhale
70.26
4.19 BB
HM21-7*
71.90
2.03
TUUNGANE
Kabamba
64.84
2.00 BB
ZKA93-10m*
69.77
1.80
ISOKO Y’UB’MW. Kabarore
62.59
1.44 CB
MLV06*
69.63
2.99
MAENDELEO
Kabamba
59.52
1.74 BB
AFR708
68.47
2.10
RUSINAME
Luhihi
59.36
1.68 BB
CODMLB003*
68.28
2.19
IRIBA
Murambi
58.93
1.77 CB
AND10*
68.08
2.39
CB
VCB81012*
67.78
3.01
CB
local variety CB
67.16
3.00
BB
local variety BB
65.50
2.19
BB
BRB194*
61.95
2.19
BB
CIM9314-36
59.64
2.04
CB
Kiangara*
58.51
2.68
The AMMI model analyses G×E interactions by combining ANOVA (with additive
parameters) and PCA (with multiplicative parameters). As it requires a fully balanced
dataset (i.e. each genotype in each environment), only a sub-selection of the full dataset
was submitted. This sub-selection included 10 bush bean varieties (AFR708, ARA4,
BRB194, CIM9314-36, CODMLB003, ECAPAN021, HM21-7, Marungi, ZAA5/2 and
ZKA93-10m/95) in 7 associations (APACOV, RHUBEHAGUMA, RUSINAME,
TUUNGANE, MAENDELEO, ISOKO Y’UBUMWE and DUFATANYE. This subselection includes both some of varieties with the highest and lowest grain Fe contents but
does not include the associations with highest (ALEMALU) or lowest (IRIBA) grain Fe
contents. Applying the simple regression model to the data sub-selection shows that purely
genotypic effects remain significant (explaining 33% of the total variation) but purely
environmental effects become insignificant (P<0.37). The AMMI analysis can be used to
study G×E interactions in the data sub-selection; however, it does not take full account of
the environmental variability in the entire dataset. As manure application did not
significantly affect grain Fe contents, treatments were considered as replications in the
AMMI analysis. Missing values (18) were predicted using the model described in the
preliminary analysis (i.e. assuming no G×E interaction) to balance the dataset.
The AMMI model is described as:
N
Fe_Ca,g = µ + αa + γg + ∑ λ n ζ g ,n ηa ,n +ρa,g + εa,g
n −1
with Fe_Ca,g = grain Fe content for genotype ‘g’ in association ‘a’ [mg Fe kg-1], µ = grand mean, αa = environment
(association) mean deviations, γg = genotype mean deviations, N = the number of singular value decomposition (SVD)
axes retained in the model, λn = singular value for SVD axis n, ζg,n = genotype singular vector value for SBD axis n, ηa,n =
association singular vector value for SVD axis n, ρa,g = AMMI residuals, and εa,g = the error term.
The AMMI analysis shows that grain Fe contents are significantly affected by environment
(association) and genotype, which explained 5 and 46% of the model variation,
respectively (Table 12). G×E interaction accounted for 49% of the total model variation.
Two IPCA factors could significantly explain 74% of the G×E interaction variation.
The IPCA factors were tested for correlation with soil characteristics (pH, org. C, total N,
extractable P and exchangeable bases). IPCA 1 was found to be negatively correlated with
33
soil organic C and total N contents (r = -0.71 and -0.77 with P-values 0.07 and 0.04,
respectively). Environments with higher IPCA 1 scores would thus be less fertile
environments with lower organic C and total N contents. Sites with higher soil organic C
and total N contents were RUSINAME, RHUBEHAGUMA, MAENDELEO and
APACOV. Sites with lowest soil organic C and total N contents were ISOKO
Y’UBUMWE and DUFATANYE.
Table 12: AMMI analysis of variance for significance of genotype, environment
(association) and genotype × environment (association) interaction effects on grain Fe
contents, and the partitioning of interaction effects into AMMI axes
source of variation
df
SS
MS
F-value
P-value
% of G×E SS
total
139 17636
126.9
model
69 12656
183.4
2.41
0.0003
genotype
9 5795
643.9
8.45
<0.0001
environment (assoc.)
6
615
102.5
3.96
0.0020
block
7
181
25.9
0.34
0.9326
genotype × environ.
54 6246
115.7
1.52
0.0553
IPCA1
14 2496
178.3
2.34
0.0113
40.0
IPCA2
12 2126
177.2
2.33
0.0156
34.0
residuals
28 1623
58.0
0.76
0.7846
26.0
error
63 4799
76.2
The AMMI biplot (Figure 14) shows 71% of the model variation with 46%, 5% and 20%
due to genotype, environment and G×E interaction (IPCA 1 only), respectively. For any
G-E combination in the biplot, the AMMI-calculated grain Fe content can be estimated by
adding the G and E means minus the grand mean (69.9 mg Fe kg-1) to the product of the
G and E IPCA 1 scores. For variety BRB194 grown in ISOKO Y’UBUMWE for example,
this becomes:
Fe_CAMMI
= 76.0 (G mean) + 68.7 (E mean) – 69.9 (grand mean) + 2.99 (G IPCA 1 score) × 4.37 (E IPCA 1 score)
= 87.9 mg Fe kg-1
This fits the observed value Fe_Cobserved = 90.3 mg Fe kg-1. Actual calculated values by the AMMI model (in this case:
88.2 mg Fe kg-1) may differ slightly from these simple calculated values as the AMMI model also considers the G×E
interaction effect accounted for by ICPA 2 (17% of the model variation).
This example illustrates the positive interaction between this genotype and environment.
Inversely, growing AFR708 in the same association would lead to a negative interaction.
Genotypes can be selected based on high Fe contents and/or on stability (i.e. less variable
grain Fe contents across environments). When selecting genotypes for high Fe contents,
genotypes need to be chosen with high means and positive interaction with a given
environment. For ISOKO Y’UBUMWE, TUUNGANE and DUFATANYE, these
genotypes would thus be BRB194 and ZKA93-10m, while for MAENDELEO,
RHUBEHAGUMA and APACOV, these genotypes would be ARA4 and AFR708.
Varieties with stable grain Fe contents across environments are ECAPAN021 and
ZAA5/2. However, these varieties both have low Fe contents (below the grand mean).
The most suitable variety across all environments would thus be ARA4, with the highest
mean Fe content (80.0 mg Fe kg-1) and a relatively small ICPA 1 score (-1.56 mg Fe kg-0.5),
indicating relatively high stability.
34
5
environment (association)
ISOKO Y'UBUMWE
genotype
4
IPCA 1 (mg Fe kg-1)0.5
3
BRB194
2
ZKA93-10m
Marungi
CIM9314-36
TUUNGANE
1
DUFATANYE
RUSINAME
0
ECAPAN021
ZAA5/2
-1
ARA4
HM21-7
-2
CODMLB003
RHUBEHAGUMA
APACOV
MAENDELEO
AFR708
-3
55
60
65
70
75
80
85
grain Fe content (mg Fe kg-1)
Figure 14: AMMI biplot showing the main and IPCA 1 effects of both genotypes and
environments (associations) on grain Fe content; an estimate of the G×E interaction effect
for a specific genotype – environment (association) combination equals the product of
their corresponding IPCA1 scores.
Identification of varieties with highest Fe contents for specific environments can also be
done through plotting the AMMI-calculated grain Fe contents for each genotype in
function of the environments’ IPCA 1 scores (Figure 15). Genotypes thus become ranked
for grain Fe content in each environment. Varieties ARA4, AFR708, ZKA93-10m and
BRB194 are amongst the 4 varieties with highest Fe contents in most associations (Table
13). Except for AFR708, these are known biofortified varieties with elevated Fe and/or Zn
contents in the grains. These varieties also show opposite trends in grain Fe content versus
IPCA 1 score relationship: while BRB194, and ZKA93-10m show increasing grain Fe
contents with increasing IPCA 1 score (decreasing soil C and N content), ARA4, AFR708
and CODMLB003 show decreasing grain Fe contents with increasing IPCA 1 scores
(increasing soil C and N content).
Table 13: Varieties ranked following grain Fe contents based on AMMI-calculated values
in each environment; varieties marked with an asterisk (*) are known biofortified varieties
with elevated grain Fe and/or Zn contents.
association
action site
1st variety
2nd variety
3rd variety
APACOV
Burhale
ARA4*
AFR708
CIM9314-36*
RHUBEHAGUMA Luhihi
ARA4*
AFR708
ZKA93-10m*
RUSINAME
Luhihi
ARA4*
ZKA93-10m*
BRB194*
TUUNGANE
Kabamba
ZKA93-10m*
BRB194*
ARA4*
MAENDELEO
Kabamba
ARA4*
AFR708
CODMLB003*
ISOKO Y’UB’MW. Kabarore
BRB194*
ZKA93-10m*
CIM9314-36*
DUFATANYE
Nyakigando CIM9314-36*
ARA4*
AFR708
35
90
BRB194
ZKA93-10m
80
75
70
CIM9314-36
ARA4
Marungi
65
ZAA5/2
AFR708
60
ECAPAN021
CODMLB003
55
ISOKO
Y'UBUMWE
TUUNGANE
DUFATANYE
45
RUSINAME
50
HM21-7
RHUBEHAGUMA
APACOV
MAENDELEO
AMMI-calculated grain Fe content (mg Fe kg-1)
85
40
-3
-2
←higher soil C+N content
-1
0
1
2
IPCA 1 (mg Fe kg-1)0.5
3
4
5
lower soil C+N content→
Figure 15: Calculated grain Fe contents of 10 bean varieties based on the AMMI model
equation across environment IPCA 1 scores. Full lines are regressions for varieties
BRB194, ZKA93-10m and CIM9314-36; dotted lines are regressions for varieties ARA4,
AFR708 and CODMLB003.
Conclusion
A simple linear regression model applied to the entire dataset showed that 44% of the total
variation in grain Fe contents could be attributed to purely environmental effects, while
17% is related to purely genotypic effects. An AMMI analysis on selected varieties and
environments (associations) was conducted to study G×E interaction effects. Although
the sub-selection did not represent the environmental
variability of the entire dataset, it revealed significant G×E
interaction effects accounting for 35% of the total
variation in the data sub-selection. This demonstrates the
necessity of taking G×E interactions into account when
selecting bean varieties for high grain Fe contents.
The variety ARA4 (Photograph 4) can be recommended
as a genotype with high Fe contents (73 mg Fe kg-1) across
environments. However, other varieties such as BRB194
and ZKA93-10m have higher grain Fe contents in specific
environments, likely characterized by less fertile soils with
lower organic C and total soil N contents.
Photograph 4: ARA4, a
biofortified bean variety with
stable and high Fe contents.
44..BB.. A
ASSSSEESSSSM
ME
EN
NT
TO
OFF N
NU
UT
TR
RIIE
EN
NT
TD
DE
EFFIIC
CIIE
EN
NC
CIIE
ESS IIN
N SSO
OIIL
LSS
O
ON
NT
TH
HE
EW
WAALLU
UN
NG
GU
UA
AX
XIISS IIN
N SSU
UD
D--K
KIIVVU
U
The Walungu area in Sud-Kivu is very unproductive due to low soil fertility constraints
(see also section 7.A). The exact nature of the soil constraints remains unknown. Results
from a set of field trials established in February 2007 (“FER-1”) instigated a series of
greenhouse pot trials, set up between August and December 2007 at the research center of
36
INERA (Institut National pour l’Etude et la Recherche Agronomique) in Mulungu, Sud-Kivu
(Photograph 5). Preliminary results from these trials are highlighted below.
Pot trial I
At each of the 8 locations of the field
trial, 60kg of soil was sampled from the
0-20 soil layer, sun-dried and sieved to
pass 4 mm. These soils were then used to
set up a pot trial with 8 different
treatments: a reference treatment with all
nutrients applied (540 mg N kg-1 soil, 54
mg P kg-1, 450 mg K kg-1, 225 mg Ca kg-1,
99 mg Mg kg-1, 36 mg S kg-1, 3.2 mg Fe
kg-1, 3.2 mg Mn kg-1, 2.0 mg Zn kg-1, 2.0
mg Cu kg-1, 0.6 mg B kg-1 and 0.6 mg Mo
kg-1), six treatments with K, P, Mg, S, Zn Photograph 5: Missing nutrient pot trial
and B omitted, respectively, and a
established at a greenhouse at INERAtreatment with application of N, P and K Mulungu, Sud-Kivu, DRC.
only (at the same rates as in the reference
treatment). The above rates were assumed to eliminate deficiencies for the respective plant
nutrients. Following salts were used for the nutrient additions: KH2PO4, NH4H2PO4,
Mg(NO3)2.6H2O, NH4NO3, KNO3, Ca(NO3)2.4H2O, MgSO4.7H2O, (NH4)2S04, ZnCl2,
CuCl2.2H2O, FeCl3, MnCl2.4H2O, (NH4)6Mo7O24.4H2O, and Na2B4O7.10H2O. The pots
were filled with 2.5 kg of soil, after which the soil of each pot was mixed with nutrient
solutions according to the different treatments. The pots were then placed on tables in the
greenhouse following a randomized complete block design with 3 replicates, and rotated
daily. In each pot, 3 maize (Zea mais L., cv. Katumani) seeds were sown, and thinned to
one plant per pot after one week (1 WAP). Moisture conditions were kept optimal in the
course of the trial. The average minimum and maximum temperatures during the growth
period equalled 15.6°C and 43.0 ºC, respectively. Regularly, the height of the plants (i.e. the
distance from the plant basis to the highest tip of the three youngest fully developed leafs)
was measured. At 5 WAP, plants were harvested and the biomass was sun-dried in closed
paper bags. Subsequently, plants were oven-dried (65ºC) and weighed.
Photograph 6: Visual symptoms of P
deficiency in young maize plants.
All plants showed severe visual signs of P
deficiency (Photograph 6), and no significant
differences in height or final dry weight were
observed between soils or treatments. At 5 WAP,
all plants had a similar height as low as 45 cm. As
such, the assumed rate of 54 mg P kg-1, which
corresponds to a (broadcast) field rate of about
120 kg P ha-1, is inadequate to lift available P levels
sufficiently to eliminate P deficiency for maize. In
conclusion, P deficiency was the principal
constraint for crop growth in the soils tested.
Pot trial II
A second pot trial was established to investigate the extent of P deficiency in the two
action sites. To this end, a wider range of soils was collected from other the sites, including
the fields chosen for legume evaluation, legume seed multiplication fields and randomly
selected soils sampled during the final characterisation study. While the former two sets of
37
soils were presented by farmer associations for project activities and of variable fertility
according to the association’s ability to meet the expenses of providing land, the latter soils
are typically cultivated by legumes following the local practices. The selected soil collection
is considered a representative set of soils used for legume cultivation in the Walungu area.
DM yield (g)
plant height (cm)
The pot trial was set up following procedures for planting, watering and measuring plants
during plant growth equal to the first trial. Six different treatments were imposed in a
single replicate: a control without nutrient additions, a reference treatment with additions
of all nutrients (396 mg N kg-1, 360 mg P kg-1, 360 mg K kg-1, 210 mg Ca kg-1, 92 mg Mg
kg-1, 42 mg S kg-1, 2.9 mg Mn kg-1, 1.9 mg Zn kg-1, 1.9 mg Cu kg-1, 0.6 mg B kg-1 and 0.6
mg Mo kg-1), 3 treatments with N, P and K omitted respectively, and a treatment with
application of N, P and K only (at the same rates as in the reference treatment). The P rate
was increased 10-fold as compared to the first trial to ensure P deficiency was entirely
eliminated. The average minimum and maximum temperatures during the growth period
equalled 15,6°C and 45,6°C, respectively. Plants were cut at 38 days after planting (DAP).
At harvest the youngest fully developed leaf was cut and dried separately for leaf nutrient
analysis.
Treatment differences became
100
apparent at about 3 WAP
control
(Figure 16, Photograph 7).
PK + µnutrients
80
NK + µnutrients
At 32 DAP, plant heights in
NP + µnutrients
the control and the treatments
NPK
60
without P addition were
NPK + µnutrients
similar and as low as 50 cm,
40
while an average height of 85
cm was observed in the
20
reference treatment. Visual
SED treatment
SED time
signs of P deficiency were
0
observed in the former two
0
5
10
15
20
25
30
35
treatments, and were very
time (DAP)
pronounced in all 30 soils
10
(without exception); final dry
legume evaluation soils (n=3)
SED (a)
multiplication soils (n=12)
weights were more than 5
final characterisation soils (n=15)
8
times lower, compared to the
SED (b)
reference treatment. In the
other treatments with N, K or
6
micronutrients omitted,
responses were differential,
4
but no consistent differences
with the reference treatment
2
were observed. In some soils,
however, maize plants clearly
0
showed to suffer from
control
PK +
NK +
NP +
NPK
NPK +
nutrient deficiencies other
µnutrients µnutrients µnutrients
µnutrients
than P. Leaf analyses
Figure 16: Maize plant height in function of time (top)
and dry biomass weight after 38 days of growth (bottom), (pending) are required to
diagnose these.
as affected by different nutrient application regimes,
observed in a pot trial conducted at INERA-Mulungu,
Sud-Kivu, DRC; error bars represent SED for comparison
of treatment and time effects (left), and soil group (a) and
treatment (b) effects (right).
38
No significant differences in maize
response were observed between
soils used for legume germplasm
evaluation, legume multiplication
and soils sampled during the final
characterisation study. The
similarity between the soil types was
confirmed by soil analysis (Table
14). All soils were similarly
characterized by a similar and low
organic C content (0.6%), soil pH
(5.2) and low cation exchange
capacity (6 cmolc kg-1).
Photograph 7: In most soils, treatments effects
followed trends as observed in the example, from
left to right: control, N omitted, P omitted, K
omitted, NPK only, and reference treatment).
Table 14: Soil properties for project soils (sampled in fields for legume germplasm
evaluation and multiplication) and final characterisation soils (typically cultivated with
legumes, sampled in farmers’ fields).
org. C
pH (H20)
CEC
(g kg-1)
(cmolc kg-1)
Project soils
6.63
5.2
5.88
Final characterisation soils
5.36
5.3
6.71
SED
0.82
0.2
0.46
44..CC.. A
ASSSSEESSSSM
ME
EN
NT
TO
OFF B
BA
AN
NA
AN
NA
A –– A
AR
RB
BU
USSC
CU
UL
LA
AR
R
M
MY
YC
CO
OR
RR
RH
HIIZ
ZA
AL
L FFU
UN
NG
GII R
RE
EL
LA
AT
TIIO
ON
NSSH
HIIPPSS
A survey was carried out in 188 fields in Rwanda to identify arbuscular mycorrhizal fungi
(AMF) and plant parasitic nematode infection on banana roots in five regions: Ruhengeri
(young volcanic soils), Gitarama–Butare (soils derived from granitic rocks), Kibungo
(weathered soils from schistose materials), Gashonga (clay ferralsol on basalt) and
Bugarama (volcanic alluvial soils). Data were recorded for the single cultivar Intuntu
(AAA-EA). We recorded management practices, root health parameters, root colonization
by arbuscular mycorrhizal fungi (AMF) and nematode infection in roots. Highly varying
AMF colonization was observed in different soil types. Highest colonization was recorded
in roots of banana plants grown on clay soils on basalt with frequency 62.6% and intensity
35.4%, while the lowest was on poor weathered soils on granite with 17.2% and 16.7% of
frequency and intensity, respectively. Other soil types had intermediate infection. Higher
AMF frequency was associated with slightly increased height (r=0.190, p<0.001) and girth
of banana plants (r=0.144, p<0.05), significantly higher number of functional roots
(r=0.185, p<0.05), higher root number per 20cm3 of soil volume (r=0.222, p<0.01) and
much lower root necrosis (r= - 0.235, p<0.01). This study provides an insight on the role
that AMF play in existing highland banana production systems and possible benefits of
future use of AMF to improve plant health and vigor. Further studies should respond on
the question how AMF may help to improve and sustain the yield of highland banana
production systems.
39
5. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HB
BA
AN
NA
AN
NA
AG
GE
ER
RM
MP
PL
LA
ASSM
MR
RE
EL
LA
AT
TE
ED
DA
AC
CT
TIIV
VIIT
TIIE
ESS
55..AA IIN
N--SSIIT
TU
UG
GE
ER
RM
MPPLLA
ASSM
ME
EV
VA
AL
LU
UA
AT
TIIO
ON
N
The objectives of these activities are to introduce, test and disseminate new germplasm to
farmers in order to improve productivity and profitability under the given constraints (i.e.
pests, soils, climate, and management practices) and given the actual and potential market
and consumption requirements. This activity is executed in close collaboration with the
INIBAP-IPGRI-led project. A protocol for the establishment, management, and data
collection has been developed in collaboration with our partners. Germplasm trials have
been installed in late 2006 / early 2007 across the region, representing the existing
variation in soils and climate (i.e altitude) within the region (Photograph 8).
Sites:
1. South Kivu
a. Luhihi
b. Burhale
c. Lurhale
d. Kabamba
2. North Kivu
a. Maboya
b. Munoli
c. Mutwanga
d. UCG
3. Rwanda
a. Kibungo
b. Ruhango
c. Kibuye
d. Cyangugu
e. Kayonza
f. Bugesera
4. Burundi
a. Mugina-Cibitoke
b. Kirundo-Kirundo
c. Busoni-Kirundo
d. Giheta-Gitega
e. Mutaho-Gitega
Photograph 8: CIALCA banana
germplasm trial at the Mulungu
station in Sud Kivu, DR Congo.
Besides providing farmers with improved banana
cultivars, the trials will provide much scientific
insight into genotype × environment interactions,
which will help breeding, IPM and agronomic
research in the longer term. The germplasm trials
will also serve as farmer demonstration sites for
best-bet banana management practices, including
management practices such as timely deleafing,
weeding, mulching and desuckering, in
combination with proper planting practices and
Photograph 9: Mg deficient plants in
cleaning techniques (boiling water treatment,
germplasm trial in Lurhala, Sud Kivu
paring) of sucker-derived planting material.
40
Germplasm sources are: (i) three tissue-culture derived IITA highland banana hybrids
developed under the DGDC-funded ‘Strategic Musa Improvement Project’ (NSH 20, 22,
and 42), (ii) tissue-culture derived exotic cultivars already available in the region (FHIA 01,
FHIA 03, FHIA 21, FHIA 25, Yangambi km5) in existing tissue culture labs (i.e.
Agrobiotec, Burundi), (iii) tissue-culture derived exotic cultivars obtained from the banana
International Transit Center in Leuven, (iv) 4 sucker derived local checks and best-bet
highland varieties from within the Great lakes region. The three IITA hybrid varieties were
added to the varieties that were mutplied in the course of 2006-2007 and were added to
the field trials by late 2007, early 2008. Unfortunately, the banana varieties coming from
ITC-Belgium and that were multiplied at Agrobiotech in Burundi got infected by a fungal
pathogen in the lab and these plants have therefore not been established in the field yet. A
second batch of ITC tissue culture plantlets has been sent to IRAZ for multiplication
there. These plants should be added to the germplasm trials by September 2008.
Management practices are as much as possible uniform across sites. These will include:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Watering of the initial planting material if deemed necessary for the survival of the plants.
Planting pits of 60x60x60 cm.
Manure/Compost application in planting pit
Mulching of plants 2 x per year at the end of the wet season, preferably with grasses.
Desuckering of plants in order to have a 1-2-3 system (mother, daughter, granddaughter),
preferably with a circular movement of plants, so that original planting density are respected
as much as possible.
Deleafing of leaves that have <50% of functional leaf surface area.
Bunding/ water traps in sloping areas.
All plant residues (leaves, stems, etc.) will remain within the plot around the mat of origin.
However, ALL bunches will be exported, even when not sold.
Methods for selling or distribution of harvested bunches will be negotiated with the partners
managing the trial.
No use of pesticide/insecticide – damage parameters will be taken at flowering (nematodes)
and harvest (weevil).
Herbicide treatment or superficial removal of weeds by hand or hoe, but without tilling the
soil = damaging of roots.
Data collection in these germplasm trials is
ongoing, and a common data collection tool
has been developed for all the sites.
In addition, presence of the BBTV vector,
pentalonia nigronervosa is monitored in the 3
countries using “yellow traps” within the
framework of the PhD project of Celestin
Niyongere. Field trials on BBTV reinfection
rates in newly planted fields (with both in
vitro and sucker-derived plants) comprising 9
genotypes were established in December
2007.
Photograph 10: BBTV screening plot at
the ISABU Mparambo station in Burundi.
Close to each germplasm trial, local macro-propagation facilities will be (and have been)
developed for rapid, low-cost, and relatively clean multiplication of banana planting
material at or near the sites of the germplasm trials. After the initial training at IITAUganda in 2006, Macro-propagation sites have been established at INERA-Mulungu in
DRC, at ISAR-Butare in Rwanda, and at the RWARRI NGO in Kibungo, Rwanda. As we
envisage expanding these activities, these first three sites will act as a site for the training of
trainers on selection, cleaning and rapid multiplication of banana planting material.
41
55..BB SSTTRRAATTEEGGIICC RREESSEEAARRCCH
HA
AT
TK
K..U
U..L
LEEU
UV
VE
EN
N
ITC
During early 2007, 20 plantain accessions collected in the Congo-Basin were shipped from
the Kisangani University in DRC to the ITC. In July, a selection of 36 accessions of EastAfrican highland bananas, diploids and triploids, collected in the Morogoro and Mbeya
region of Tanzania in 2005, were sent to the ITC for duplication in vitro. In the course of
2007, the active collection increased with 30 new accessions, and currently includes a total
of 1,212 accessions. Currently, the collection comprises 808 accessions (66.6%) that are
virus indexed negative and available for international distribution. Accessions in storage
were indexed routinely for bacterial endophytes at annual subculturing. A total of 253
tests were performed on 58 accessions in storage. The majority concerned samples of
accessions that were checked after virus therapy. Also 58 accessions that were newly
introduced and were initiated in vitro for the first time were tested pre-storage.
Rejuvenation of the collection was continued for 9 accessions and started for another 51.
The plants were decapitated in the greenhouse and will be re-initiated in vitro early 2008.
The lyophilized leaf collection presently counts 598 accessions. In 2007, 2,234 samples of
411 accessions were processed in the leaf bank; per accession a variable number of
samples (1-37 replicates) are stored at -20°C. In 2007, 509 accessions, represented by 1654
individual tissue samples, were distributed from the ITC gene bank. Most materials were
supplied as proliferating cultures (PT) (55%) followed by rooted plantlets (RP) (34%) and
lyophilized leaf tissues (LT) (11%). A significant increase in the demand of lyophilized leaf
tissue for molecular (DNA) study purposes was recorded. The International Transit
Centre processed a total of 27 orders from 22 scientists in 18 countries. Samples of 408
accessions were supplied. The exchange of plant material from the ITC gene bank
involved 265 different accessions keeping the utilization ratio of germplasm available for
international distribution constant at about 30%. A small number of 84 accessions were
distributed to underpin the gene bank activities including virus indexing and therapy,
characterization and regional multiplication and distribution activities.
A feasibility study for the virus pre-indexing operation in collaboration with FUSAGx, was
started. The applicability of the PhytoPASS technology for preliminary indexing of banana
germplasm on different types of tissue (leaf tissue of field plants, corm tissue, and leaf
tissue of in vitro plants), pre- and post- entry, and for the most common banana viruses, is
currently being assayed on the new introductions from the Democratic Republic of Congo
and Tanzania. Preliminary results of the material from DRC indicate that the pre-indexing
allows the identification of material with an ‘initially better’ health status, and thus more
suitable for introduction in the collection, reducing the need for post-entry therapy and
repeated virus indexing.
Preliminary research towards induction of drought stress in Musa cell cultures
A better understanding of plant response to water deficit has become an increasingly
important challenge. The effects of water deficit on plants can be investigated at different
levels: (i) in the field (ii) in the green-house (iii) and in vitro. However, whole-plant factors
such as differences in morphology, carbohydrate assimilation and partitioning render it
very difficult to study the effects of induced drought in se. We explored the possibility to
use cell cultures in an in vitro model mimicking drought stress in banana, induced by
polyethylene glycol (PEG). The water content of ‘Cachaco (ABB)’, ‘Williams (AAA)’ and
‘Musa balbisiana (BB)’ cell samples were on average 89.6, 88.7 and 91.2% respectively. The
biggest influence of the lowest PEG concentration (7.5%), was noted for ‘Musa balbisiana’
42
losing 5.4% of its water content in comparison to 3.5% measured for ‘Cachaco’ and 3%
for ‘Williams’ cells. For all cell lines, a decrease in water content with 5-6% was obtained
after increasing PEG concentration from 7.5% to 15% and also from 15% to 22.5%. As
the water content decreases with increasing PEG concentration, PEG is indeed suitable to
induce dehydration in banana cell cultures. Regeneration tests also showed that PEG
negatively affected regrowth, and that this regrowth is variety-dependent and more
affected by higher concentrations. In summary PEG can effectively be used to induce
dehydration in Musa embryogenic cell cultures. The value of the in vitro model should be
evaluated by research performed ex vitro, submitting plants from the same varieties to
drought stress in the greenhouse and/or in the field.
Cryopreservation
In 2007, 108 accessions belonging to 19 genomic groups were successfully cryopreserved.
For 12 accessions the genomic constitution is unknown. This brings the total to 527
accessions completely and definitively stored in liquid nitrogen. Also 192 accessions were
shipped with a dry shipper to IRD and safely stored as “black box”.
Molecular biology: Promoter tagging as a basis for developing cisgenic banana
Based on a previously developed gene- and promoter tagging platform, a genome-wide
screening strategy was performed for the identification and characterization of novel
promoters in banana. Embryogenic Musa cell suspensions of the plantain were
transformed with Agrobacterium tumefaciens containing a promoterless, codon-optimized
luciferase (luc+). Tens of thousands of transgenic cell colonies were selected and screened.
The frequency of low temperature-responsive cell colonies ranged between 0.17-1.69%. 94
colonies were regenerated to plantlets and screened throughout different regeneration
stages. In total, twenty-four banana flanking DNA sequences were cloned in seven
independently selected lines. So far we isolated and characterised a novel banana promoter.
Understanding AMF biocontrol and its impact in banana-based cropping systems
Fourteen intercrops with high mycorrhizal compatibility and low nematode susceptibility
were selected. Effects of intercrops with different nematode susceptibility levels on
nematode population build-up in an intercrop set-up were studied. Common bean looks
promising as a banana intercrop for nematode control in a mixed cropping system. AMF
and rhizobial colonisation both resulted in reduced nematode populations and improved
growth of common bean. However, dual colonisation did not provide extra plant growth
when compared to single AMF colonisation. Mechanisms responsible for the AMF bioprotection effect were unraveled: the split-root experiments indicate a combination of a
locally and systemically induced effect. R. similis is less attracted to and penetrates less
(+50%) in AMF colonized plant roots. Strong indications exist that root exudates play an
important role in this process. However, based on the attraction bio-assays and exudates
experiment it could not be determined whether the decreased penetration of the AMF
colonised plants was partly due to a repellent effect of the root exudates. Once penetrated
in the root, R. similis development and reproduction in the AMF-colonised roots goes
slower than in non-colonised roots.
43
6. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HL
LE
EG
GU
UM
ME
EG
GE
ER
RM
MP
PL
LA
ASSM
MR
RE
EL
LA
AT
TE
ED
DA
AC
CT
TIIV
VIIT
TIIE
ESS
66..AA.. L
LEEGGU
UM
ME
EG
GE
ER
RM
MPPL
LA
ASSM
MD
DE
EM
MO
ON
NSST
TR
RA
AT
TIIO
ON
NA
AN
ND
D
E
EV
VA
AL
LU
UA
AT
TIIO
ON
N
6.A.1. ON-STATION LEGUME GERMPLASM EVALUATION
On-station evaluation of bean, soybean,
groundnut and pigeon pea germplasm started
in March 2006 to identify potential varieties
for testing at action site level with the various
farmer associations. In these evaluation trials,
specific measures were included to identify
varieties tolerant to low-P conditions, and
varieties with effective nodulation and
fixation of atmospheric nitrogen. In
September 2007, additional characterisation
Photograph 11: Soybean germplasm
activities were started to fully characterize
evaluation trial at INERA-Mulungu
selected varieties, which is essential for
(season 2008 A), Sud-Kivu, DRC.
homologation of the germplasm. Table 15
presents an overview of the various on-station germplasm evaluation and characterisation
activities conducted at the ISAR and INERA stations.
Table 15: On-station legume germplasm evaluation since the project start; in March ’07,
characterisation of germplasm was initiated to enable homologation of selected varieties.
station
2006 B
2007 A
2007 B
2008 A
(Mar’06-Jun’06)
(Oct’06-Jan’07)
(Mar’07-Jun’07)
(Oct’07-Jan’08)
RubonaCB, BB, SB
ISAR
KaramaGN, CP, PP
ISAR
Nyagatare- BB, CB, SB
ISAR
(+characterisation)
Mulungu- CB, BB, SB, GN,
CB, BB, SB
BB, CB, SB
INERA
CP, PP
(+characterisation)
M’vuaziBB, CB, GN, CP
GN, CP
BB, CB, SB
GN, SB
INERA
(+characterisation)
(+characterisation)
CB: climbing bean; BB: bush bean; SB: soybean; GN: groundnut; CP: cowpea; PP: pigeon pea.
Biomass production, nodulation and atmospheric N fixation
Specific observations on biomass production, nodulation and atmospheric N fixation were
carried out in the trials at the various stations, for all of the species tested. Presented below
is a selection of results obtained. Emphasis was given to soybean and groundnut
germplasm because of their known capacity for soil fertility enhancement.
Large differences between varieties were observed in nodulation, biological N fixation and
biomass production. At all sites, a generally positive response to P fertilization was
observed. At the ISAR-Rubona station, several soybean varieties showed higher
44
5000
SB17
-1
biomass yield [kg DM ha ]
SB6
SB20
4000
SB2
SB19
SB6
SB20SB15
SB2
3000
SB14
SB12
SB14
SB4
SB15
SB17
SB19
SB12
2000
SB4
1000
control
with TSP application
0
0
100
200
300
400
500
-1
nodulation [kg FM ha ]
Figure 17: Left: Nodulation and biomass yield as
affected by TSP application at 25 kg P ha-1 for selected
soybean varieties grown at the ISAR-Rubona station,
Rwanda; FM = fresh matter, DM = dry matter.
nodulation and biomass
production when supplied with
TSP fertilizer (e.g., SB2, SB4,
SB19, SB20); some varieties,
however, did not respond (e.g.,
SB6, SB12) (Figure 17). An
analysis using the 15N-isotopic
dilution for soybean varieties
grown at the INERA-Mulungu
station showed that most of
the newly introduced varieties
are equally effective in fixing
atmospheric N as the local
variety (Figure 18). In most
varieties, P fertilization
increased the effectiveness of
N fixation (proportion of N
derived from the atmosphere).
% N derived from atmosphere
Biomass production, nodulation and
100
atmospheric N fixation are of
90
interest, as they determine the
80
potential of the varieties to improve
70
soil fertility by supplying organic
60
matter and atmospheric N that can
benefit a subsequent crop. These
50
characteristics therefore constitute a
40
first researcher-defined criterion for
30
variety selection. Several of the newly
20
varieties are superior to the local
10
control
variety in terms of biomass
0
production, in particular soybean
SB2 SB4 SB6 SB12 SB14 SB17 SB19 SB20 local
varieties SB19 and SB24 in Sud-Kivu Figure 18: Biological N fixation determined by
and Rwanda, and SB20 in Bas15N-isotopic dilution method as affected by TSP
Congo.
application at 25 kg P ha-1 for selected soybean
varieties grown at the INERA-Mulungu station,
Sud-Kivu, DRC; Imperial was used as the local
variety.
Also within the 63 groundnut varieties tested (Spanish Bunch, Valencia and Virginia
growth types), a wide variability in nodulation and biomass production was observed.
Several varieties performed superiorly to the local variety (JL24) at INERA-M’Vuazi, BasCongo (Figure 19). Particularly varieties ICGV-SM93555 and ICGV-SM99551 were
retained for further testing as they produced up to twice as much biomass as the local
variety (both with and without additional P fertilisation.
Grain yield and response to P application
Grain and biomass yield were evaluated with and without P application, with the aim to
identify varieties that are tolerant to low-P conditions and/or respond to P application.
These observations were taken both on-station, and in germplasm evaluation trials in
farmers’ fields. A multi-locational assessment is required to identify varieties that are
45
4000
biomass yield (kg DM ha-1)
consistently tolerant to low-P
conditions and/or respond to
P application. Data from the
on-station trials will be merged
with yields observed in
farmers’ fields (with and
without TSP fertilizer
application) to identify
promising varieties (currently
on-going). Detailed studies will
then be undertaken to unravel
mechanisms of low-P tolerance
in these selected varieties.
Spanish Bunch type
(control)
Spanish Bunch type
(with TSP application)
3000
Valencia type
(control)
Valencia type
(with TSP application)
2000
JL24
1000
0
0
5
10
15
20
6
25
30
-1
nr of nodules (x 10 ha )
Figure 19: Nodulation and biomass yield as affected by
TSP application at 25 kg P ha-1 for selected groundnut
varieties (Spanish Bunch and Valencia type) grown at
the INERA-M’vuazi station, Bas-Congo, DRC; JL24 is
the local variety; DM = dry matter.
biomass yield with P application [kg ha-1]
At the INERA-M’vuazi
station, several soybean
varieties showed to be tolerant
to low-P conditions and at the
same time responsive to P
application (e.g., SB19, SB20,
Bossier and SB44) (Figure 20).
In addition, the variety SB44
produced high amounts of
biomass without P application.
Some varieties responded to P
application but performed
poorly under low-P conditions
(e.g., SB24, 449/16/6), while
others performed well under
low-P conditions but did not
respond to P application (e.g.,
SB46, SB51).
6000
SB42
TGM1781
SB24
4000
SB20
SB9
3000
SB38
SB44
TGx814-26D
SB19
2000
Vuangi
Duiker
SB14
SB51
1000
non-responsive
tolerant to low-P
non-responsive
susceptible to low-P
0
0
1000
2000
3000
4000
5000
6000
biomass yield in control [kg ha-1]
grain yield with P application [kg ha-1]
2000
The tolerance to low-P
conditions and responsiveness
to P application constitute a
second researcher-defined
criterion for variety selection,
as P deficiency is one of the
major constraints for legume
production in the areas (see
also section 4.B).
responsive
tolerant to low-P
responsive
5000
responsive
1500
responsive
tolerant to low-P
SB9
449/16/6
SB24
TGM1781
Bossier
SB25
SB20
SB44
Vuangi
1000
Peka6
SB15
SB19
SB46
SB51
TGx814-26D
SB2
Duiker
500
SB14
non-responsive
0
0
500
non-responsive
tolerant to low-P
1000
1500
grain yield in control [kg ha-1]
Figure 20: Biomass (top) and grain (bottom) yield for
selected soybean varieties as affected by NPK
application at 25 kg P ha-1 for selected soybean varieties
grown at the INERA-M’vuazi station, Bas-Congo, DRC;
local varieties tested include TGx 814-26D and Vuangi;
quadrants classify varieties based to their tolerance to
low-P conditions (along the X-axis) or their response to
P application (along the Y-axis).
46
Legume variety characterisation
In March 2007 in Bas-Congo, and in October 2007 in Rwanda, on-station trials were
established to fully characterize a selection of soybean and bean varieties, retained after
selection based on farmer and researcher criteria. Detailed measurements are being taken
during 2 consecutive seasons on the physiology and the resistance to pests, diseases and
environmental stresses. This information will be used to produce technical leaflets (“fiches
techniques”) that are required for the homologation of new germplasm. These new
varieties can then be introduced in national formal seed multiplication and dissemination
systems.
6.A.2. LEGUME GERMPLASM EVALUATION WITH FARMER
ASSOCIATIONS AT THE ACTION SITES
Bean and soybean germplasm evaluation
trials were initiated in October 2006 with all
associations in the action sites in the 4
mandate areas (Table 16). During the period
2007 A – 2008 A, a total of 44 farmer
associations have been involved in
germplasm evaluation activities across the 4
mandate areas. When variety selection was
affected by abnormal drought or excessive
rain, or when farmers insisted on evaluating
the germplasm during multiple seasons, trials Photograph 12: Soybean germplasm
evaluation trial in Murambi, Umutara,
were repeated during one or two additional
Rwanda.
seasons. Groundnut and pigeon pea
germplasm was included in season 2007 B. Cowpea germplasm poorly adapted to high
altitude and was only tested with farmer associations in Bas-Congo. In season 2008A,
germplasm evaluation trials were set up with a limited number of bean and soybean
varieties through NGO partners with associations in satellite sites in Bas-Congo.
Table 16: Legume germplasm evaluation trials installed with farmer associations in
between October 2006 and January 2008.
mandate
2007 A
2007 B
2008 A
area
(Oct’06-Jan’07)
(Mar’07-Jun’07)
(Oct’07-Jan’08)
BC
BB, CB, SB: 8 associations SB: 2 associations
BB, CB: 6 associations
GN: 8 associations
CP: 8 associations
KK
BB, CB, SB: 7 associations BB, CB, SB: 5 associations BB, CB, SB: 6 associations
PP: 4 associations
PP: 5 associations
GN: 1 association
GN: 1 association
UM
BB, CB, SB: 8 associations GN, PP: 2 associations
BB, CB: 1 association
SB: 5 associations
GN: 2 associations
SK
BB, CB, SB: 8 associations BB, CB: 4 associations
GN: 11 associations
GN: 8 associations
Mandate areas: BC: Bas-Congo; KK: Kigali-Kibungo, UM: Umutara, SK: Sud-Kivu.
Species: CB: climbing bean; BB: bush bean; SB: soybean; GN: groundnut; CP: cowpea;
PP: pigeon pea.
47
A number of measurements were taken in the germplasm evaluation trials (Annex 3).
Firstly, grain yields were determined with and without P fertilizer application for soybean
(to test for low-P tolerance), and with and without manure application (to test for
resistance to low soil fertility) for the other species. Secondly, biomass samples were
collected to assess atmospheric N fixation and potential for soil fertility improvement, and
grain samples were collected to assess the genetic and environmental variability in
micronutrient contents (particularly Fe and Zn) in bean grains (see section 4.A.).
Furthermore, pest and disease occurrence was assessed and scored for bean varieties.
Finally, participatory farmer evaluation events were organized to assess farmers’ selection
criteria and preferences for the various legume species tested.
Grain yield
An analysis was conducted to assess the effects of site, input application and variety on
grain yields of the various legume species evaluated. Generally, yields of the varieties were
significantly site-dependent. In Sud-Kivu for example, some varieties performed well on
the northern axis (e.g., SB4, SB6), but not on the southern axis (Figure 21). Other
varieties, however, performed well across all sites (particularly Peka-6, SB19 and SB24). In
all sites, new varieties performing equally well or better than the local variety could be
identified. Soybean grain yields on the northern axis (Luhihi and Kabamba) were much
larger than on the southern axis (Lurhala and Mwegerera). On the southern axis, soil
fertility is very poor compared to the northern axis, and due to the high altitude (around
2000m) and short growing season, only early-maturing varieties were able to produce (e.g.,
SB19, Peka-6).
800
800
Lurhala
Mwegerera
SED
local
TGM1781 non-estimatable
Peka6 non-estimatable
Soprosoy
Duiker
Ogden
Bossier
SB25
449/16/6 non-estimatable
SB24
SB20
SB19
SB17 non-estimatable
SB6 non-estimatable
SB15
SED
grain yield [kg ha-1]
3000
2500
2000
2500
2000
1500
1000
500
local
TGM1781
Peka6
Soprosoy
Ogden
Duiker
Bossier
449/16/6
SB25
SB24
SB20
SB19
SB17
SB15
SB14
SB6
SB4
SB2
local
TGM1781
Peka6
Soprosoy
Ogden
Duiker
Bossier
449/16/6
SB25
SB24
SB20
0
SB19
SB15
SB6
SB14
1500
SB2 non-estimatable
SB4
SB14
Luhihi
SED
3000
0
SB4
3500
Kabamba
500
200
0
3500
1000
400
local
TGM1781 non-estimatable
Peka6
Soprosoy non-estimatable
Duiker
Ogden
Bossier
SB25
449/16/6
SB24
SB19
SB2
0
SB6 non-estimatable
200
SB14
400
600
SB2 non-estimatable
600
SB4 non-estimatable
SED
Figure 21: Soybean grain yields observed in the 4 action sites in Sud-Kivu during the 2007 A season;
the interaction between variety and site was significant at P<0.05; TSP application significantly
increased yields, independent of site or variety.
48
Input application (TSP for soybean, manure for beans) generally increased yields,
independent of site or variety. Figure 22 presents bush bean yields observed in Kibungo
during the 2007 A and 2007 B (averaged for both sites), as affected by manure application.
All varieties responded significantly to manure application; bean grain yields increased
from on average 1200 kg ha-1 to 1900 kg ha-1. Some varieties performed superior in the
control (e.g., AFR619, AFR708 and ZAA5/2); these are known “BILFA” varieties, which
are resistant to low soil fertility and yield 60 – 80% more grains than the local variety
without manure application. Other varieties perform relatively well in the control and
responded strongly to manure application (e.g., CODMLB003, CNF5520, MLB49-89A
and HM21/7). Several varieties performed equally well or better than the local variety.
3500
control
with manure application
(a)
3000
grain yield (kg ha-1)
(b)
2500
2000
1500
1000
500
AFR619
AFR708
ARA4
BRB194
CIM9314-36
CIM9331-1
CNF5520
CODMLB003
CODMLB007
CODMLB033
CODMLB078
ECAPAN021
GNP585
HM21/7
LSA144
Maharagi soja
Marungi
MLB49-89A
M'Sole
Rab618
Rab619
Rjb7
T8426F11-6F
UBR(92)25
VEF88(40)L1P4T6
ZAA5/2
ZKA93-10m/95
local variety
0
Figure 22: Bean grain yields observed in Kibungo, Rwanda, averaged for
the two action sites and seasons A and B 2007, as affected by manure
application at 5 t DM ha-1; error bars represent the SED for comparison
between varieties (a) and treatments (b); the interaction between variety
and manure application was not significant.
Biomass yield
The biomass of the various species was
sampled in farmers’ fields at the 50%
podding stage, when biomass
accumulation is maximal. High biomass
production is an important selection
criterion as it is associated with the
variety’s ability to improve soil fertility.
Biomass samples are currently being
analysed for δ15N, which allows estimating
the proportion of N fixed from the
atmosphere. This is particularly relevant
Photograph 13: High biomass-yielding
for soybean because of its potential
soybean varieties in an evaluation trial in
rotational benefits on a subsequent crop.
Nyakigando, Umutara, Rwanda.
49
Several of the newly introduced soybean varieties, particularly soybean lines SB 38 – SB54,
produced large quantities of biomass (up to 6 t ha-1, Figure 23). These varieties also
responded strongly to TSP application. However, although these varieties are favourable
for soil fertility enhancement, they are little suited for the Rwanda and Sud-Kivu mandate
areas; in several action sites, the too short growth season did not permit these longduration varieties to reach maturity and produce grains.
control
biomass yield [kg ha-1]
6000
(a)
(b)
4000
2000
SB2
SB4
SB6
SB9
SB14
SB15
SB17
SB19
SB20
SB24
SB25
SB38
SB39
SB42
SB44
SB45
SB46
SB47
SB49
SB51
SB54
449/16/6
Bossier
Duiker
Ogden
Peka6
Soprosoy
local
0
Figure 23: Soybean biomass yields observed in Kibungo, Rwanda,
averaged for the two action sites and seasons A and B 2007, as affected
by TSP application; error bars represent the SED for comparison
between varieties (a) and treatments (b); the interaction between variety
and manure application was significant at P<0.1.
Other varieties such as SB19, SB24, SB25 and Peka-6 are
more suitable as they have short- to medium-duration
cycles and produce at the same time a reasonable amount
of biomass and yield relatively high amounts of grains.
Apart from soybean, climbing beans also produce high
amounts of biomass (up to 8 t ha-1). Several of the newly
introduced varieties, particularly VCB81012, produce
much higher amounts of biomass than the local variety.
There are indications that climbing beans can have
substantial rotational benefits on a subsequent cereal crop;
this is currently being tested in some of the demonstration
trials with the associations in the Kabamba action site in
Sud-Kivu (see section 7.A.).
Photograph 14: High
biomass-producing climbing
Micronutrient contents in bean grains
bean varieties in Umutara,
Measurements are currently being conducted on grain
samples collected from the various bush and climbing bean Rwanda.
evaluation trials to assess variability in Fe and Zn contents as affected by environment (site
× input) and genetic (variety) effects. This is presented in detail under the process research
activities in section 4.A.
50
Farmer evaluations
Initially, two evaluation events were
organized in each legume evaluation trial – a
first evaluation between flowering and
podding, and a second after harvest – to
examine whether farmer-preferred varieties
can already be identified at an early stage.
Male and female association members were
separated and defined their criteria for
selection. They then visited the trial and
specified positive and negative traits of each
Photograph 15: Farmer evaluation of bean
variety, and finally selected five varieties
germplasm in Zenga, Bas-Congo, DRC
which they scored according to their criteria.
The questionnaire used for farmer evaluation is presented in Annex 4.
Farmers used a wide range of criteria for evaluating the germplasm. These criteria
depended on the growth stage of the plants, differed between mandate areas and crop
species, and were influenced by the sex of the evaluating farmers. Presented below are the
evaluation criteria for beans across the mandate areas (Figure 24).
100
100
after harvest
Bas-Congo
60
Figure 24: Relative importance of evaluation criteria defined by farmer groups between flowering
and podding (left) and after harvest (right) in the mandate areas of the TSBF-CIAT project. The
index was calculated as the frequency of the criterion (%) divided by its average rank.
Between flowering and podding, farmers commonly used the number of flowers (or pods)
as principle criterion for the expected yield. In addition, farmers in Bas-Congo evaluated
the ‘germination capacity’ of the varieties, which relates to the swiftness at which the
variety grows in the early stages and its ability to out-compete weeds. In Rwanda, farmers
appreciated early maturity and drought resistance because of the short growing season
with unreliable rains. In Sud-Kivu, both high rainfall tolerance as well as drought resistance
were favoured traits, as climate unpredictability is a major constraint for bean production.
At the evaluation events after harvest, evaluation criteria comprised yield, followed by
grain traits (size, colour, lustre, decay, density,…). Some of the criteria identified during the
season were retained, but had much less weight in the choice of varieties to be retained. In
most cases, between one and three out of the five varieties preferred at the floweringpodding stage were retained after the harvest evaluation, and rarely the same variety was
most preferred at both evaluation stages.
Men and women evaluators used somewhat different evaluation criteria, and usually had at
least 2 or 3 common varieties in their top 5 of preferred varieties. After the final
51
disease
resitance
resistance to
poor soil fertility
drought
resistance
resistance to
heavy rain
early maturity
ease of cooking
taste
resistance to
weevils
market
preference
grain density
grain lustre
biomass
production
disease
resitance
resistance to
poor soil fertility
drought
resistance
0
resistance to
heavy rain
0
germination
20
early maturity
20
grain decay
40
grain colour
40
Sud-Kivu
grain size
60
Rwanda
80
yield
relative importance index
80
number of
flowers/pods
at flowering-podding
evaluation, a discussion session was organized where the association decided in plenary on
a selection of maximally five bush bean, three climbing bean and five soybean varieties to
be multiplied in the following season. For soybean, research teams suggested including at
least one soybean variety with high biomass production, in cases where farmers opted to
multiply early-maturing varieties only.
Because of the performance of the varieties varies between sites and regions, and because
farmers base their selection on differing criteria, varieties retained differed between
mandate areas, action sites and even individual associations. Nevertheless, a number of
varieties could be identified that were widely acceptable to farmers across sites (Table 17).
Table 17: Three most frequently retained bush bean, climbing and soybean varieties by
farmer groups at the evaluation event after harvest for the different mandate areas (or
regions within mandate area); numbers between brackets represent proportion of farmer
groups that retained the variety.
region
bush bean
climbing bean
soybean
BC
Lola (100%)
Lola (100%)
Vuangi (63%)
PVO14/2 (67%)
Tuta (100%)
SB19 (63%)
ZAA5/2 (53%)
MLV06-90B (78%)
TGx814-26D (78%)
KK, Bugesera
local variety (75%)
AND10 (50%)
not available yet
Marungi (63%)
G59/1-2 (50%)
ARA4 (50%)
VCB81012 (50%)
KK, Kibungo
AFR708 (80%)
AND10 (83%)
not available yet
CNF5520 (80%)
local variety (83%)
ZAA5/2 (60%)
VCB81012 (50%)
UM
local variety (71%)
Kiangara (80%)
not available yet
Maharagi-soja (57%)
MLV06-90B (60%)
M’Sole (43%)
local variety (60%)
SK, Katana axis
ZKA93-10m/95 (55%) VCB81012 (75%)
Peka6 (100%)
Marungi (45%)
AND10 (75%)
SB24 (79%)
BRB194 (45%)
G59/1-2 (75%)
Soprosoy (79%)
SK, Walungu axis Marungi (69%)
Kiangara (75%)
Peka6 (67%)
ZKA93-10m/95 (56%) LIB1 (75%)
SB19 (67%)
BRB194 (50%)
AND10 (50%)
Soprosoy (67%)
Mandate areas: BC: Bas-Congo; KK: Kigali-Kibungo, UM: Umutara, SK: Sud-Kivu; Bugesera
comprises action sites Mayange and Musenyi, Kibungo comprises Kabare and Gatore, the Katana
axis in Sud-Kivu comprises action sites Luhihi and Kabamba, the Walungu axis comprises Lurhala
and Mwegerera.
52
66..BB.. L
LEEGGU
UM
ME
E SSE
EE
ED
DM
MU
UL
LT
TIIPPL
LIIC
CA
AT
TIIO
ON
N
6.B.1. ON-STATION LEGUME SEED MULTIPLICATION
On-station seed multiplication activities started in October 2006, with the purpose of
providing seeds for research and dissemination activities. Initially, seed was produced for a
large number of varieties through the evaluation activities. When promising varieties were
identified, these were multiplied on a larger scale. Currently, seed stocks are maintained at
the national research centres to support activities within and beyond the project. Table 18
presents an overview of the on-station germplasm multiplication activities conducted at
the ISAR and INERA stations.
Table 18: On-station seed multiplication of legume germplasm since the project start.
station
2007 A
2007 B
2008 A
(Oct’06-Jan’07)
(Mar’07-Jun’07)
(Oct’07-Jan’08)
Nyagatare-ISAR
BB, CB, SB
Mulungu-INERA BB, CB, SB, FB, GP
BB, CB, SB
M’vuazi-INERA
BB, CB, SB, GN, CP
BB, CB, SB, GN, CP
BB, CB, SB, GN, CP
Species: CB: climbing bean; BB: bush bean; SB: soybean; GN: groundnut; CP: cowpea;
FB: faba bean; GP: garden peas.
A strategy has been developed to accelerate diffusion of promising varieties by multiplying
and disseminating seed through multiple channels, both formal and informal, starting from
the seed base produced at the ISAR and INERA stations (Figure 25). While formal seed
multiplication is led by the national legume programs, informal seed multiplication and
dissemination is primarily facilitated by NGO partners in collaboration with the ISAR and
INERA legume programs. National seed services (RADA and SENASEM) are involved in
training and accreditation of farmer associations and individual seed multipliers. A detailed
action plan, indicating the different sub-activities, responsibilities, timeline and reporting
format is presented in Annex 5).
Currently, selected promising varieties are being fully characterized by the national legume
programs, as a first step in the formal seed multiplication channel. In Rwanda, the ISAR
bean and soybean program are collecting necessary information for the production of
technical fiches, which are required for officially inscribing new varieties in the national
catalogue. At the same time, the legume programs are producing the initial clean seed
source, which will be multiplied by the Unité semencière de l’ISAR. Once the new varieties are
homologated, foundation seed will be supplied to the Rwanda Agricultural Development
Authority (RADA), who will formally distribute the varieties through its national network
of accredited seed multipliers. In DRC, formal seed multiplication has not yet been
initialised; however, SENASEM (Service National de Semence) has been actively involved in
the seed multiplication through informal channels, by providing technical backstopping,
training and accreditation to multiplying farmer associations.
53
INERA/ISAR semence de base
INERA/ISAR
+ NGOs
INERA/ISAR
Instituts nationaux
(SENASEM / RADA)
Multiplicateurs individuels de
semence
Associations paysannes
dans les sites d’action
formation par
SENASEM/RADA
INERA/ISAR
+ NGOs
formation par
SENASEM/RADA
INERA/ISAR
+ NGOs
Membres qui se spécialisent
comme multiplicateurs
NGOs
Réseau national de
multiplicateurs certifiés
Commerçants agriculteurs
(associations ou individus)
Associations paysannes
dans les sites satellites
Diffusion
paysan ↔ paysan
voie formelle
voie informelle
Figure 25: Seed system strategy and involvement of various partners for rapid seed
multiplication and dissemination through formal and informal channels.
6.B.2. FARMER ASSOCIATION-LED LEGUME SEED
MULTIPLICATION
Legume seed multiplication through informal
channels was initialised through the farmer
associations in the action sites and facilitated
by NGO partners. After selection of
promising varieties during the farmer
evaluation after harvesting the legume
evaluation trials, a discussion session was
organised with each farmer association to
decide on practical aspects of seed
multiplication. Generally, associations
decided to organize seed production as a
group activity, mostly in communal fields.
The activity started in March 2007 with 38
farmer associations (Table 19) in action and
satellite sites. In October 2007, an additional
5 associations were involved.
54
Photograph 16: A member of the farmer
association Dutabarane multiplying an
improved climbing bean variety in Gatore,
Kibungo, Rwanda.
Table 19: Legume germplasm multiplication activities conducted by farmer associations in
between March 2007 and January 2008.
mandate
2007 B
2008 A
area
(Mar’07-Jun’07)
(Oct’07-Jan’08)
BC
BB: 10 varieties, 5 associations
CB: 4 varieties, 4 associations
SB: 8 varieties, 21 associations
GN: 1 variety, 2 associations
GN: 2 varieties, 4 associations
CP: 2 varieties, 21 associations
CP: 2 varieties, 21 associations
KK
CB: 3 varieties, 1 association
SB: 4 varieties, 1 association
UM
SK
Mandate areas: BC: Bas-Congo; KK: Kigali-Kibungo, UM: Umutara, SK: Sud-Kivu.
Species: CB: climbing bean; BB: bush bean; SB: soybean; GN: groundnut; CP: cowpea.
During the first season, members of the
farmer associations involved were trained
on technical aspects of seed production
(Photograph 17), following the guidelines
in the CIAT manual for small-scale seed
production (David, 1998). Prior to these
training sessions, CIALCA staff were
acquainted with the guidelines and technical
requirements of seed multiplication during a
three-day training in February 2007 in
Photograph 17: Training session on seed
Butare, Rwanda (Annex 6). In Rwanda, the
production techniques organized by
ISAR legume program initially provided
DIOBASS and SENASEM with farmer
association representatives on the Walungu training and facilitation to farmer
associations, but currently NGO partners
axis, Sud-Kivu, DRC.
(World Vision, RWARRI, RDO and
RHEPPI) are increasingly more involved in this activity. In DRC, seed multiplication
activities are led by NGO partners (DIOBASS in Sud-Kivu; BDD and APRODEC in BasCongo), with technical backstopping by INERA and SENASEM. The first training
sessions focused on agronomic practices and regulations for seed multiplication (planting
in line, spacing, purging and disease management) (Annex 7). After the first season, an
evaluation event was organized with each association in order to discuss successes and
identify bottlenecks. Constraints were in the first place related to land availability and
availing sufficient organic inputs (compost or manure) for communal activities.
Capacities for seed multiplication differed between farmer associations. Some associations
were committed to continue multiplying in group, while others opted for a decentralised
system, whereby association members receive a portion of seed, multiply in their own
field, and return a proportion of the seed produced to the association at the end of the
season. Terms and conditions are discussed within each association, and facilitated by
NGO members. Performance in each individual multiplication field is being documented
by a technical team appointed in the association, or by a local animateur engaged by the
NGO partner (Annex 8). Some of the bottlenecks identified in the first season were
55
addressed in the second season, through
closer follow-up, providing technical advice
and assisting in discussions to improve the
organisation of the activity. In Sud-Kivu, for
example, SENASEM officers visited all fields
of multiplying individuals or associations,
made observations and recommendations,
and finally conferred official certificates to
the seed producers (Photograph 18).
Depending on the progress and
Photograph 18: SENASEM officer making
organisational skill of the association,
additional training is provided (e.g., related to observation in a soybean multiplication
field in Kabamba, Sud-Kivu.
post-harvest management).
Currently, about half of the associations have reached a level where seed quantities suffice
to satisfy the seed needs of all members within the association (between 50 – 100 kg per
variety). In the on-going season (season B 2008), activities are planned to promote and
disseminate the seed produced through informal channels, primarily by farmer-to-farmer
diffusion. Exchange visits will be organized between multiplying farmer associations, in
preparation of farmer field days. During these events, key stakeholders (representatives
from farmer associations inaction and satellite sits, local and regional NGOs, the national
seed service (RADA or SENASEM), local authorities and policy makers, journalists and
the local radio) will be invited to promote the improved varieties. Leaflets and farmer
fiches will be distributed during these events (Figure 26). NGO partners will play a crucial
role in linking the multiplying farmer associations (or individuals) with potential buyers
and creating incentive for commercialisation of seed production.
AFR708
haricot nain
Rendement potentiel :
Maturité :
-1
[2500 kg ha ]
haut
(au milieu paysan)
précoce
Taille des graines :
[75-85 jours]
large
Production de biomasse :
[50g pour 100 graines]
moyenne
Croissance sur sol pauvre :
Résistance aux fortes pluies :
Résistance à la sécheresse :
Résistance aux maladies :
Les paysans ont apprécié :
le haut rendement
la précocité
l’adaptation au milieu
la grosse taille des graines
la couleur des graines (rouge-tachetée)
l’apparence des graines
la cuisson facile/rapide
la préférence au marché
Gestion :
bonne
moyenne
pauvre
bonne
Semez en ligne : 40cm entre les lignes et
10-15cm sur la ligne (une graine par poquet).
Pour un rendement optimal, appliquez 50kg de
fumier ou compost par are. Cette variété donne
aussi bien sans application d’intrants.
Cette variété est bio-fortifiée : elle est riche en matières minérales et
bonne pour la santé.
Figure 26: An example of a farmer fiche of an improved (bio-fortified) bush bean
variety, presenting technical and farmer-preferred characteristics identified during
the germplasm evaluation trials.
56
7. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HN
NA
AT
TU
UR
RA
AL
LR
RE
ESSO
OU
UR
RC
CE
E
M
MA
AN
NA
AG
GE
EM
ME
EN
NT
T-R
RE
EL
LA
AT
TE
ED
DA
AC
CT
TIIV
VIIT
TIIE
ESS
77..AA.. N
NAATTU
UR
RA
AL
LR
REESSO
OU
UR
RC
CE
EM
MAAN
NA
AG
GE
EM
ME
EN
NT
TO
OPPTTIIO
ON
NSS FFO
OR
R
L
LE
EG
GU
UM
ME
E--B
BA
ASSE
ED
D SSY
YSST
TE
EM
MSS
7.A.1. OVERVIEW OF OPTIONS CURRENTLY BEING TESTED
Testing of natural resource management (NRM) options started in March 2007 (season
2007 B) or in September 2007 (season 2008 A). These options were chosen to address
some of the major constraints identified, and were targeted towards the specific conditions
in the regions and action sites. Partners as well as farmer groups were involved in the
process of selecting and developing options. In most sites, proven NRM options that were
successful elsewhere (improved agronomic practices and nutrient management in legumecereal rotation or intercropping, and cassava-legume intercropping systems) were
immediately implemented as demonstration trials with farmer associations. Other
technologies, requiring additional research and adjustment to the local conditions, were
implemented on-station before engaging with farmer associations. These include
technologies for soil erosion control, tested in Sud-Kivu (DRC), and rain water harvesting
in combination with efficient nutrient
management to counteract seasonal drought
spells in Umutara (Rwanda). At present, a
total number of 56 farmer associations are
actively involved in the testing, management
and evaluation of demonstration trials in the
various action sites across the 4 mandate
areas (Table 20). In addition, 50 households
in the Nyakigando action site (Umutara,
Rwanda) started implementing adaptation
trials in September 2007 (season 2008 A)
Photograph 19: An improved bean-maize
after having appraised the technologies
intercropping system, demonstrated in
demonstrated earlier in legume-cereal
Kabarore, Umutara, Rwanda.
intercropping systems (Photograph 19).
Presented below are selected results for NRM options on erosion control, improved
cassava-legume intercropping, and soil fertility amendment on the soils around Walungu
(Sud-Kivu).
57
Table 20: An overview of NRM research, demonstration and adaptation trials conducted in 2007 in the various action sites of the TSBF-CIAT project.
NRM option
Erosion control using legume hedgerows and
reduced tillage as alternatives to terrace
construction
Rainwater harvesting and interaction with
nutrient management to counteract seasonal
drought spells
Improved agronomy options in cassava-legume
intercropping systems
acronym
“ERO-1”
trial type
on-station
(research)
“WANU-1”
on-station
(research)
Improved agronomy options in cassava-legume
“CAS-2”
Improved soil fertility management in cassava
systems
“CAS-3”
Erosion control using leguminous and nonleguminous forage hedgerows as an alternative
to terrace construction
“ERO-2”
Soil fertility amendment using various organic
and inorganic inputs on poor soils
“FER-1”
Improved cereal-legume intercropping options
“SYS-1”
“CAS-1”
Demonstration of rotational benefits of high“SYS-2”
biomass- yielding legume varieties and the
micro-dose fertilizer technique
Demonstration of rotational benefits of soybean “SYS-3”
and Mucuna on a subsequent maize crop
Adaptation of cereal-legume intercropping
options
“ADA-1”
location
on-station at Mudaka
(Sud-Kivu)
on-station at ISARKarama and ISARNyagatare (Rwanda)
demonstration 2 locations, in Zenga
trials
and Nkamu in BasCongo
demonstration 6 locations, in
trials
Kabamba in Sud-Kivu
demonstration 2 locations, in Kisantu
trials
and Mbanza-Nzundu
in Bas-Congo
demonstration 6 locations, in action
trials
sites in Sud-Kivu
demonstration 8 locations, in
trials
Mwegerera and
Lurhala in Sud-Kivu
demonstration 8 locations, in action
trials
sites in Umutara,
Rwanda
demonstration 6 locations, in Luhihi
trials
in Sud-Kivu
demonstration 30 locations, in Lemfu
trials
in Bas-Congo
adaptation
trials
50 locations in
Nyakigando, Umutara,
Rwanda
58
period
installed in March 2007
(currently running the
3rd season)
installed in March 2007
(currently running the
3rd season running)
installed in April 2007
(on-going)
involvement of farmer associations
on-station activity but the local farmer
community is involved in the management of
the trial.
on-station activity – the activity aims to
identify promising options before engaging
with farmer associations
2 farmer associations are involved in the
management and evaluation of the
demonstrated options.
installed in September 3 farmer associations are involved in the
2007 (on-going)
installed in April 2007 2 farmer associations are involved in the
(on-going)
installed in March 2007 6 farmer associations are involved in the
(on-going)
evaluation of the forage species. The trial has
been combined with multiplication of
improved cassava germplasm.
installed in March 2007 6 farmer associations have been involved in
for one season
the testing of soil fertility management
(concluded)
options.
installed in September 8 farmer associations have been involved in
2006 (currently running the management and evaluation of the
the 3rd season)
installed in September 3 farmer associations are involved in the
2007 (on-going)
installed in October
30 farmer associations (through the network
2007 (on-going)
of NGO partner BDD) are involved in the
installed in September 50 individual households are testing and
2007 (on-going)
adapting the previously demonstrated options
in their own farms.
7.A.2. SOIL CONSERVATION TECHNOLOGIES TESTED IN SUDKIVU (“ERO-1” AND “ERO-2)
In Sud-Kivu, high population has driven
agriculture towards fragile land and soil
erosion has become one of the major threats
to agricultural production (Photograph 20).
Legume-grown fields are predominantly
located on lands with strong slopes and
commonly unprotected against erosion (see
section 3.C.1.). Inadequate soil conservation
measures have given rise to rapid loss of
topsoil and land degradation. In Rwanda,
Photograph 20: Severe soil degradation in
effective policies and community work
land with a high slope cropped with beans
arrangements are in place for large-scale
in Sud-Kivu, DRC.
terrace construction and combating soil
erosion. These are however absent in Sud-Kivu, where the situation calls for alternative,
less labour-intensive technologies that are adoptable by individual households.
Two trials were set up in March 2007 to evaluate alternative options for combating soil
erosion. A first trial (“ERO-1”) aimed to compare the effectiveness in conserving soil of
reduced tillage and planting hedgerows of a leguminous perennial, Calliandra callothyrsus,
with the construction of physical terraces. This trial was set up on-station in 3 replicates,
on a site with a strong slope (41 %), following a complete factorial design with factors (i)
terrace construction, (ii) tillage, and (iii) Calliandra hedgerows. A second trial (“ERO-2”)
was installed in 6 sites in farmers’ environment and aimed to examine the adaptability of
various forage species when grown as hedgerows on representative slopes, and to obtain
farmers’ feedback on the adoption potential of these forages.
On-station testing of soil conservation measures (“ERO-1”)
In this trial, the soil is cropped with soybean in
rotation with maize, and the short- and long-term
effects of reduced tillage, planting Calliandra
hedgerows, and installing physical terraces on
crop production and soil conservation are
assessed. The measurements conducted in this
trial include: crop grain yield, changes in slope
and soil loss, soil water profiles, and changes in
soil fertility. A detailed description is given in the
trial protocol (Annex 9). Presented below are
selected results from the first and second season,
grown with soybean and maize, respectively.
Photograph 21: Soybean cropping
without (top) or with (bottom)
physical embankments.
Extension programs recommend a vertical
distance of 1.6 m between two contour lines for
terrace construction or hedgerow planting. This
translates into a plot width of 4 m for the given
slope at the trial site. The installation of terraces
by embanking the soil up-hill and hedgerow
planting entail a loss of surface area available for
cropping, equal to 27% and 20%, respectively
59
25
without physical embankments
terrace surface area (m2)
with physical embankments
20
SED (a)
15
SED (b)
10
5
0
without Calliandra
with Calliandra
Figure 27: Plot surface area (plot length = 5m and
vertical interval = 1.6m) as affected by terrace
construction and hedgerow planting; error bars
represent SED, comparing (a) with and without
terraces, and (b) with and without hedgerows.
2.0
grain yield (kg per terrace)
1.5
1.0
SED
0.5
0.0
trad. tillage
zero tillage
trad. tillage
Calliandra
hedgerows
zero tillage
Calliandra
hedgerows
Figure 28: Grain yields obtained per plot (for a plot
length of 5m and a vertical interval of 1.6m) as
affected by tillage, terrace construction and
hedgerow planting; the error bars represents SED.
1.0
(Photograph 21, Figure 27). This
implies that grain yields per unit area
need to increase by 37% and 25%,
respectively, to compensate for these
area losses, and to justify the use of
these techniques by farmers.
In the first season, soybean grain
yields obtained per plot (dimensions:
length = 5m, vertical interval = 1.6
m) were significantly lower in plots
with physical terraces (Figure 28).
This was primarily related to the loss
of surface area available for
cropping. Plots without physical
embankments could generally hold 5
soybean lines while after terrace
construction, plots could only hold 4
soybean lines. However, terrace
construction also reduced yields per
unit area (630 kg ha-1 vs. 770 kg ha-1
without terraces). Most likely, the soil
embankment brought unfertile, acid
subsoil to the surface, which
negatively affected crop
performance. Reduced tillage did not
affect soybean yields when terraces
were installed or Calliandra
hedgerows were planted, but higher
yields were observed in plots without
terraces and hedgerows. Planting of
Calliandra hedgerows only decreased
yields in plots without tillage and
terrace construction.
Soil erosion was significantly reduced
by terrace construction and Calliandra
0.6
hedgerow planting (Figure 29);
SED (b)
tillage management did not affect soil
0.4
loss. During the first month after
planting (2nd season), the soil loss
0.2
amounted to almost 1 kg m-2in plots
without conservation measures,
0.0
which approximates a loss of 1 mm
without Calliandra
with Calliandra
of the soil profile. This soil loss was
Figure 29: Soil loss during the first month after
reduced by 80% in plots with
planting (2nd season) as affected by hedgerow
physical embankments. Calliandra
planting and terrace construction; error bars
hedgerows
were less effective in
represent SED, comparing (a) with and without
reducing
soil
erosion. Calliandra
terraces, and (b) with and without hedgerows.
initially grows slowly and the
hedgerows are at present not yet fully developed. Further measurements are required to
assess the effectiveness in soil conservation in the longer term.
soil loss (kg m-2)
0.8
SED (a)
60
As a preliminary conclusion, terrace construction is most effective in the short-term to
reduce soil erosion, but the reduction in surface area and upturning of subsoil considerably
reduces crop yields. Planting Calliandra hedgerows has less influence on crop yield, but is
also less effective in reducing soil erosion. Long-term effects on crop yield and soil
stabilisation need to be assessed to appraise Calliandra hedgerow planting as an alternative
to terrace construction.
Evaluation of hedgerow forages for erosion control in farmers’ environment (“ERO-2”)
Ten forage species (Brachiaria brizantha, Brachiaria
decumbens, Brachiaria ibrido, Brachiaria ruziziensis,
Calliandra calothyrsus, Leucaena diversifolia, Penisetum
purpureum, Setaria sphacelata, Tithonia diversifolia and
Tripsacum laxum) were established as hedgerows in 6
sites (Lurhala, Mwegerera, Luhihi, Kabamba,
Cijingire and Mudaka). Measurements included
survival rate and biomass accumulation, and slope
and soil accumulation. A detailed trial protocol is
presented in Annex 10.
Farmers evaluated the forages about 10 months
after establishment (Photograph 22). The
procedure used for forage evaluation was similar as
for the legume germplasm evaluation (Annex 11).
In each site, the male and female members of the
association were separated and first defined their
criteria for evaluation. They then visited the trial
and specified positive and negative traits of each
variety, and finally selected five forages which they
scored according to their criteria.
Photograph 22: Farmers evaluating
different forages in Lurhala, SudKivu.
The criteria defined by farmers primarily comprised the use of the biomass as a green
manure, effective rooting (as an indicator for its capacity to contain the soil), use as a
forage and production of high amounts of biomass (Figure 30). These criteria were
mentioned by at least 80% of the evaluating farmer groups. Other minor criteria included
the potential for using in construction (mainly as roofing for houses), being noncompetitive with crops (shading) and producing poles for climbing bean cultivation.
10
Brachiaria ibrido
Leucaena
Brachiaria decum.
Brachiaria ruzizi.
Penisetum
Brachiaria briz.
0
Setaria
produces poles
non-shading
use for construction
drought-resistant
non-competitive
biomass production
good forage
0
green manure
10
Calliandra
20
20
Tripsacum
30
30
Tithonia
40
good rooting
criterion importance index
50
Figure 30: Left: Farmer-defined criteria for evaluation of forages. Right: Forage species selected by
farmer groups; the relative importance index was calculated as the frequency of the criterion or
forage species (%) divided by its average rank.
61
Farmers selected and scored the forages based on these criteria. Tithonia, Tripsacum and
Calliandra were the most preferred forages, selected in the top 5 by 67, 75 and 92 % of the
evaluating farmer groups, respectively. Tithonia and Tripsacum were generally ranked higher
(average rank = 2) than Calliandra (average rank = 3). Farmers appraised Tithonia and
Setaria as the most effective for soil erosion control, followed by Calliandra, Tripsacum and
Penisetum.
7.A.3. IMPROVED AGRONOMY AND SOIL FERTILITY
MANAGEMENT IN CASSAVA-LEGUME SYSTEMS
In Bas-Congo, and to a lesser
extent in Sud-Kivu, legumes are
predominantly cultivated in
association with cassava (see
section 3.C.1.). Options for
improved agronomic practices and
nutrient management in these
systems were discussed with
partners and farmer groups. Three
sets of demonstration trials were
then installed, including options for
Photograph 23: A demonstration trial on improved
improving productivity through (i)
cassava-legume intercropping in Kabamba, Sudimproved germplasm, (ii)
Kivu, DRC.
alternative spacing, (iii) application
of locally available green manures and/or fertilizer, (iv) reduced tillage, (v) planting
alternative legume species, and (vi) planting climbing beans during the second season. The
first set of trials was installed in April 2007 with two farmer associations in Bas-Congo
(two sites, 3 replicates per site) and focuses on improved agronomic practices (“CAS-1”,
Annex 12). The second set likewise focuses on improved agronomic practices and was
installed in September 2007 with 3 farmer associations in 6 sites in the Kabamba action
site in Sud-Kivu (“CAS-2”, Annex 13). The third set of trials focuses specifically on
nutrient input management using green manures and/or fertilizer to improve cassava
production, and was installed in April 2007 with two farmer associations in Bas-Congo
(two sites, 3 replicates per site (“CAS-3, Annex 14). Specific measurements included
legume biomass and grain production, cassava tuber production and tuber
quality/tradability, and detailed labour assessments. Farmers evaluated the trial at different
stages: firstly at peak biomass production of the legumes (2 months after planting),
secondly after harvest of the legumes (4 months after planting), and finally at the cassava
harvest (12 months after planting).
Biomass production in the traditional system was generally low (for example 2 t DM ha-1
in Nkamu, Bas-Congo; Figure 31). Biomass production can be considerably increased by
fertilizer application or replacing the traditional legume species (groundnut in Bas-Congo,
beans in Sud-Kivu) by soybean. In Nkamu (Bas-Congo), biomass yield was three times
higher for soybean than for groundnut. This has important implications for soil fertility
management, as a biomass production of 6 t DM ha-1 may supply a net input of 30 – 40 kg
N ha-1 (to be verified by BNF measurements), and entail significant rotational benefits for
subsequent crops (to be verified in successive seasons).
62
biomass yield (kg DM ha-1)
6000
SED (a)
4000
SED (b)
2000
0
groundnut
control
groundnut
NPK
soybean
control
Figure 31: Biomass yield obtained for groundnut
(with and without NPK applied at
100 kg ha-1) and soybean grown in association
with cassava in Nkamu, Bas-Congo, DRC; error
bars represent SED for comparison of effects of
NPK application (a) and legume species (b).
SED
2000
NPK application
improved variety
traditional
0
2x0.5m spacing
1000
1x1m spacing
pod yield (kg ha-1)
3000
Figure 32: Pod yields for common beans
obtained by successively changing the spacing
(1 × 1m), variety, spacing (2 × 0.5m) and
applying NPK at 100 kg ha-1 in Kabamba, SudKivu, DRC.
women
20
men
15
10
soybean intercrop
groundnut intercrop
NPK application
2x0.5m spacing
improved variety
0
1x1m spacing
5
traditional
farmer preference (%)
25
Figure 33: Farmer preference of improved
agronomic practices demonstrated in Kabamba,
Sud-Kivu, as compared to the traditional
cassava-legume intercropping system by
successively changing the spacing (1 × 1m),
variety, spacing (2 × 0.5m), applying NPK at 100
kg ha-1 and replacing the common legume
(beans) for groundnut or soybean.
63
Legume grain yields can also be
substantially increased using improved
agronomic practices and fertilizer
application. In Kabamba (Sud-Kivu),
for example, pod yields for the
traditional legume (beans) were
increased by 50% using an alternative
intercropping spacing that favours the
legume; fertilizer application doubled
yields compared to the traditional
system (Figure 32). Soybean generally
performed very well in association with
cassava, when planted at a spacing of 2
× 0.5m, allowing sufficient space for 4
lines of soybean (400,000 plants per
hectare) between cassava lines (10,000
plants per hectare).
Farmers have currently evaluated the
trial twice, at podding and harvest of
the legume. At the podding stage,
farmers primarily evaluated based on
the production of biomass, the number
of pods or flowers, the lustre
(greenness) of the leaves, and the
presence of diseases (Figure 33,
Annex 15). Farmers particularly
preferred the option with cassava
planted at 2 × 0.5m and intercropped
with soybean (and to a lesser extent
with an improved bean variety) as well
as the option with NPK application.
These trials have attracted large interest
by the farmer associations and
neighbouring farming communities. At
present (season 2008 B), this activity
has proceeded into an adaptation
phase. Individual members of farmer
associations have been given access to
improved legume varieties, cassava
cuttings and fertilizer, and have been
trained to test and adapt the
demonstrated options in their own
fields.
7.A.4. OPTIONS FOR SOIL FERTILITY AMENDMENT ON THE
WALUNGU AXIS IN SUD-KIVU
In the germplasm evaluation trials, a generally poor crop performance was observed in the
action sites on the southern axis (Walungu axis) in Sud-Kivu (see section 6.A.). The region
is ill-reputed for its acid and unfertile soils. During the baseline and characterisation
studies, farmers expressed low soil fertility as one of the major constraints for crop
production (see section 3.C.1.). Farmers are limited in their options for soil fertility
restoration. Due to the limited cattle numbers, use of farm yard manure (FYM) is scarce,
and chemical fertilizer is absent in the region. In addition, preliminary studies and
observations in farmers’ fields suggested potential micronutrient deficiencies. A set of
exploratory trials (“FER-1”) was installed with 6 farmer associations in the two action
sites to investigate the potential of increasing crop yields using FYM (5 t DM ha-1), NPK
(20 kg P ha-1), mavuno fertilizer (NPK enriched with micronutrients, 20 kg P ha-1), lime (4
t ha-1), and combinations of these resources. Application of Tithonia leaf residues (5 t DM
ha-1) was included as an option that is relatively readily available to farmers. Climbing
beans and maize were selected as test crops (Annex 16).
maize
climbing beans
grain yield (kg ha-1)
1500
(a)
1000
(b)
500
Tithonia
lime+mavuno
lime+NPK
lime
FYM+mavuno
FYM+NPK
FYM
mavuno
NPK
control
0
Figure 34: Grain yields for climbing beans and
maize as affected by different inputs (NPK
fertilizer, mavuno fertilizer, farm yard manure, lime
and Tithonia leaf residues) on the Walungu axis in
Sud-Kivu, DRC.
Grain yields were significantly
increased by all inputs, except by
lime application (Figure 34).
However, both maize and bean
yields generally remained much
below the potential of the crops.
Highest yields observed in one of
the more fertile fields were 2.7 t ha-1
for maize and 1.8 t ha-1 for climbing
beans (in treatments with combined
application of FYM and fertilizer).
There was a significant interaction
between species, treatment effects
and the soil fertility status; in the
poor fields, maize failed in all
treatments, while climbing beans
responded significantly to FYM and
fertilizer application.
Only in one out of the eight sites, a striking difference could be visually observed between
the treatments with NPK application and mavuno application, which suggests a nutrient
other than N, P or K was limiting crop growth (Photograph 24). In other sites, responses
to NPK and mavuno were comparable. However, micronutrient deficiency may have been
masked by P deficiency (see section 4.B.). Currently, specific analyses are conducted on
young bean leaves and maize ear leaves to identify potential micronutrient deficiencies.
64
Photograph 24: Maize growth as affected by input application in a demonstration trial in
Lurhala, Sud-Kivu, DRC; from left to right: control, NPK(17:17:17) at 20 kg P ha-1 and
mavuno (NPK enriched with micronutrients) at 20 kg P ha-1.
The results from these field trials instigated a series of pot trials on a wider range of soils
from the area to rapidly investigate major nutrient deficiencies occurring on the Walungu
axis. Preliminary results for these trials are presented in section 4.B.
77..BB.. N
NAATTU
UR
RA
AL
LR
REESSO
OU
UR
RC
CE
EM
MAAN
NA
AG
GE
EM
ME
EN
NT
TO
OPPTTIIO
ON
NSS FFO
OR
R
B
BA
AN
NA
AN
NA
A--B
BA
ASSE
ED
D SSY
YSST
TE
EM
MSS
7.B.1. ON-FARM TRIALS
The selection of technologies for on-farm testing with farmers was done through feedback
workshops, which have taken place in Rwanda and the DRC. Feedback workshops for
Burundi are planned for early 2008. At each site where the diagnostic survey was
conducted 50 persons were invited to participate in the feedback workshops. The
participants included: 30 farmers who participated in the survey at the sites as well as
NGO and farmer association representatives and local leaders. Invitations aimed to
achieve gender balance in the participation of the feedback workshops. The form of the
feedback workshops was as follows:
•
•
•
•
•
The results of the diagnostic survey were presented to the participants.
Building consensus on these perceived constraints. In some cases more constraints were
added to the list. The new list of constraints were ranked using the pair-wise ranking
technique.
Farmers suggested possible solutions to their perceived constraints. In some instances the
farmers solutions coincided with interventions that researchers wanted to test with
farmers. Research also made contribution to the list of solutions/interventions that could
be tested.
The solutions/interventions were then assessed for their sustainability in terms of
availability of resources and capacity to manage the option over time, technical and social
feasibility (technical possibility and social acceptability), Cost, effectiveness in solving the
constraint and time to success (time that the technology bear fruit).
For each action site participants of the feedback workshops selected their considered best
technologies for testing.
65
The following technologies were identified and adopted for on-farm trials:
1. Mulch applications – type and quantities can vary depending on local availability – the
trial will consist of treatments, namele (i) soil tillage and mulch removal, (ii) selfmulch and zero-tillage, and (iii) as the second treatment and application of external
mulch at a rate of 25 t dm/ha. Farmers are allowed to continue planting of
intercrop beans, but this should not interfere with the mulch and tillage practices as
described in these treatments.
2. Manure application – different manure management options will be tested, this will
include (i) application of fresh manure in small pits in between banana plants,
versus (ii) open storage of manure besides the kraal before application. Quantities
of manure applied in terms of dm/ha will be similar in both treatments.
3. Soil and water conservation measures – this includes the introduction of (i) contour
bunds in combination with contour mulching, (ii) with and without the planting of
leguminous intercrops on the contour bunds.
4. General plantation sanitation/husbandry – Scientist will show farmers how to improve
crop husbandry in a small section of his field.
Protocols for the above mentioned trials have been developed and will be similar across all
sites (Annex 17). Through the feedback workshops, farmers could indicate whether they
would volunteer to host any of the above-indicated trials. In general 2-5 farmers per site
per trial type will be involved, resulting in some 10-15 on-farm trials per action site.
Besides our focus on banana productivity, measurements will be done on the productivity
of legume intercrops if farmers have these. Hence, legume intercropping will not be
imposed, but if farmers do practice legume (i.e. mostly common bean) intercropping, then
measurements will be done on the legume intercrop to quantify the impact of zero-tillage
and mulching on the overall productivity of the plot.
7.B.2. ON-STATION TRIALS
Parellel to the on-farm trials, the PhD students in Burundi and DRC have started by late
2007 to prepare the installation of on-station trials. These researcher-managed trials are
located on the following sites and soil types:
Burundi:
DR Congo:
Rwanda:
1. Gitega – Ferralsol (FAO Acrisol)
2. Cibitoke – Vertisol (FAO Vertisol)
3. Kirundo – Ferrisol (FAO Nitisol)
4. Mulungu – Brown soil – tertiary basalt (FAO Nitisol-Ferralsol)
5. Walungu – Red soil – tertiary basalt (FAO Ferralsol)
6. Butare-Gitarama – Ferralsol (FAO Acrisol)
7. Kibungo – Ferrisol (FAO Nitisol on schiste)
8. Ruhengeri – Brown soil (FAO Andosol)
Table 21: Treatment structure of the on-station trial to study effectsof tillage and mulching
on nutrient fluxes, soil physical properties, and banana rooting and plant performance.
Treatment External mulch
Removal mulch Tillage
Beans
T0
No
Yes
Yes
Yes
T1
No
No
No
Yes
T2
Yes (Hyparrhenia diplandra) No
No
Yes
T3
Yes (Tripsacum laxum)
No
No
Yes
Each treatment will be repeated 4 times. Every plot consists of 6 x 6 plants, leaving a net
plot of 4 x 4 plants. The plants will be planted at a density of 2 x 2 m. The quantity of
mulch applied should be approximately 5 cm thickness, equivalent tot 25 t/ha dry matter.
66
The objective of these trials will be to study the impact of tillage and mulch application on
nutrient pool fluxes and soil physical properties, and their subsequent impact on the
banana rooting systems and the plant performance (Table 21). These trials will be entirely
handled by the PhD students in the IITA-led project. As in the on-farm trials, bean
intercropping will also be captured in these trials, in order to have a full economic an
agronomic evaluation of the bean intercrop on the total productivity of these mixed
cropping systems. The measurement protocols on the bean intercrops will be established
in collaboration with the TSBF-led legumes project staff.
Within the framework of the PhD research
of Telesphore Ndabamenye, trials have been
established to assess the effect of planting
density on banana (Musa spp) productivity,
soil fertility dynamics, and nutrient uptake
(Photograph 25). The aim of this research
project is to demonstrate and quantify the
effects of planting density on banana (AAAEAHB) productivity and nutrient dynamics
by establishing the relationship between soil
fertility parameters, environmental factors
(light, rainfall, soil moisture) and banana
Photograph 25: Banana planting density
cultivars. A special focus is also given to
trial at the ISAR, Rubona research station.
quantifying the nutrient balance during the
cropping cycle. Three planting density trials were established at three contrasting agroecological zones (Rubona ISAR station, Kibungo, ISAR station and Ruhengeri/Kinigi). An
initial soil characterization has been carried out at the 3 sites. Five different planting
densities are being assessed (1,428 plants ha-1 at a spacing of 3.5x2.0 m, 2,500 plants ha-1
spaced at 2.0x2.0 m, 3,333 plants ha-1 at 1.5x2.0 m, 4,444 plants ha-1 at 1.5x1.5 m and 5,000
plants ha-1 at 1.0x2.0 m). Three AAA banana cultivars were used (two cooking cultivars:
‘Injagi’ and ‘Ingaju’, and one beer cultivar: ‘Intuntu’). Data collection of growth parameters
and environmental characteristics is being conducted.
7.B.3. INITIALISATION OF BANANA DISEASE CONTROL
STRATEGIES
Initial steps have been taken to start Xanthomonas Wilt (XW) activities in Rwanda in
partnership with RADA. Mr. Frank Turyagyenda was recruited as a consultant to carry out
XW work in the Central African region.
The XW activities in Rwanda will focus on:
•
•
•
•
Screening banana germplasm for Xanthomonas wilt tolerance;
Systemicity studies after infections with contaminated garden tools (e.g. during de-leafing,
de-suckering) and after an inflorescence infection in both East African highland bananas
(Musa AAA-EA group) and “Pisang Awak” (Musa AABB group);
Seasonal influence of the systemicity and speed of bacterial movements;
Replanting trials targeting different agro-ecological zones.
67
8. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HM
MA
AR
RK
KE
ET
T-R
RE
EL
LA
AT
TE
ED
D
A
AC
CT
TIIV
VIIT
TIIE
ESS
88..AA.. B
BAAN
NA
AN
NA
AV
VA
AL
LU
UE
EC
CH
HA
AIIN
NA
AN
NA
ALLY
YSSIISS
As shown in Figure 35 the rural
assemblers mostly deal in beer
types and so do the transporters
and urban wholesalers. The
same patterns were observed in
Burundi. The rural retailers and
the urban retailers mostly sell
the cooking types. From the
above it is evident that the
banana value chains are as
dipictued in Figure 36.
Average daily sales of bananas (bunches)
From February to March 2007, market surveys were conducted together with ISAR,
ISABU, and INERA to identify opportunities and constraints in the banana value chains.
A total of 400 traders and 150 transporters were interviewed. Presented below are some of
the summary statistics of the respondents.
160
Beer
140
Cooking
Dessert
Plantain
Urban
wholesalers
Urban
retailers
120
100
80
60
40
20
0
Rural
assemblers
Rural
retailers
Transporters
Category of traders
Figure 35: Average daily sales of banana types by
category of traders in Rwanda.
Costs incurred by the traders varied with
the location of the trader. Whereas the
rural traders have high handling costs
representing 70-90% of their total costs,
urban traders have much lower handling
costs (<30%) and spend most of their
costs on communication (50%) and
storage (20%), despite the fact that
transport costs per bunch per kilometer
are highest in the urban centra.
Farmers
Rural assemblers
Rural retailer
Transporters
Taxes and dues are the major constraints
incurred by traders while selling the
scarcity of buyers and insufficient
finances are also among the constraints
mentioned.
Urban
Wholesalers
Processors
Urban
Retailers
Restaurants,
Hotels,
Institutions
Consumers
Similarly taxes and tolls are mentioned as
the most pressing constraints encountered
Key
by traders during purchase of bananas
Major Flow
Minor Flow
while the other constraints include pricing
mechanisms, transport, insufficient
Figure 36: The marketing channels of all
finance are also mentioned.
banana types in the Great Lakes region.
68
Taxes and road tolls are among the
most critical constraints
highlighted by the transporters
while other constraints mentioned
include road drudgery, delays and
police interceptions, difficulty in
assembling produce and
insufficient finance and difficulty in
selling produce as well (Figure 38).
Problems encountered in selling bananas
The transport in the banana trade sector heavily relies on trucks for the long distances (60100km), whereas pick-ups are still frequently used for medium distances (40-80km). In
Rwanda and DRC, part of the long distance transport is also happening by boat over the
Lake Kivu. Bicycle transport is used for the shorter distances (<30 km) and carrying
bananas on the head is often done for relatively short distances (<15km). The costs for
short distance transport by bicycle or head is also the highest (4-12ct US$ per km)
compared to transport by truck (2-4ct US$ per km). Whereas transport distances for most
rural and provincial markets are relatively short (15-65km), urban markets are often
supplied from much further distances; e.g., the average distance for bananas supplied to
Kigali was 170km.
Theft
2%
Poor Quality
2%
Scarcity of buyers
Storage
16%
5%
None
15%
Taxes and dues
Pricing
Transport and distance
29%
4%
6%
In the first half of 2007, banana
Inadequate supply
7%
cross-border trade studies were
Insufficient finances
14%
also conducted at the Rwandan,
0
5
10
15
20
25
30
35
Burundian, DR Congo, and
Intensity of problem (% of respondents)
Ugandan border posts. These data
are currently analysed. Preliminary Figure 38: Problems encountered in selling bananas
results highlight an important flow in Rwanda according to traders.
of bananas from Uganda and DRC to Kigali, but only minor flow of produce between
Burundi and DRC and between Burundi and Rwanda.
88..BB.. L
LEEGGU
UM
ME
EV
VA
AL
LU
UE
EC
CH
HA
AIIN
NA
AN
NA
AL
LY
YSSIISS
The overall objective of this study is to identify the most important market channels of
grain legumes (common beans, soybeans and groundnuts) in the Bas-Congo, Sud-Kivu
Montagneux, Kigali-Kibungo, and Umutara Mandate areas from the production to the
final consumption level, analyze market structure, performance, and conduct, and calculate
margins at each identified market level along the chain. The specific objectives are: (i) to
identify the most important market channels (one, two or three) for each commodity
(interviewing the operators at earlier levels along the marketing chain helps to reveal the
next level until we arrive at the level where most of the traded commodity are sold to final
consumers), (ii) to identify where producers take their commodities to, those who
purchase from them (the producers), (iii) to identify and analyse the different on-farm
profit, expenditure, investment made over grain legume production, (iv) to assess the role
of women on legume production and its impacts on household livelihoods improvement,
(v) to conduct a survey at each identified major point along the channel (to ascertain all the
costs associated with the marketing of the commodities and the prices at which
commodities are sold), (vi) to ascertain that all costs and benefits incurred on legumes
commercialization are relevant on people’s daily livelihoods, (vii) to determine market
structure, performance, conduct and profit margins at each level along the marketing
69
chain, and (viii) to make appropriate recommendations for improving marketing efficiency
and equity in the marketing of the selected grain legumes in the study area.
Preliminary market studies were carried out in the Action sites in order to trace out the
chain of the commodities to be studied (common beans, soybeans and groundnuts).
Detailed information collection tools were developed for the household and market level
surveys and enumerator were trained for administering these tools. Finally, a detailed
sampling frame was constructed, encompassing selection of households and market
operators for the different grain legumes. The obtained data will be analyzed using
descriptive statistics (percentages, frequencies, graphic displays) and quantitative
techniques (relationships will be determined using correlations and ANOVA).
Preliminary results show that at the household level, most farmers do not sell legumes but
consume these at the household level because the production is not sufficient for
marketing. For farmers who sell their produce, revenues derived from legume sales are
mostly used to meet basic needs such as school fees payment, clothing, food staff, medical
care. At the market level, it was found that most traders were women, indicating that the
commercialization of legumes is headed by women (Table 22). An exception was Kavumu
market where 59% of the traders were men. Most traders were rural wholesalers (Table
23). Another particularity of Kavumu market was that traders have the highest initial
capital (1,200,000 FC or 2,400 USD) while the lowest is Kabamba (6,500 FC or 13 USD)
(data not shown).
Table 22: Trader’s gender in the target markets in the Sud-Kivu mandate area.
Market name
Legume
Traders gender distribution
Male
Female
Cabwine-Mwami
Groundnut
1
10
Common beans
0
12
Soybeans
0
11
Total
1
33
Kadutu
Groundnut
3
14
Common beans
3
15
Soybeans
0
18
Total
6
47
Kavumu
Groundnut
10
7
Total
10
7
Mudaka
Soybeans
0
16
Total
0
16
Mugogo
Groundnut
7
5
Common beans
0
6
Soybeans
7
11
Total
14
22
Total
21
125
Proportion (%)
20
80
total
11
12
11
34
17
18
18
53
17
17
16
16
1
6
18
36
156
100
In terms of costs and margins, there was a significant difference in purchase price and sale
price between different common bean, groundnut, and soybean varieties. This could be to
a certain extend because most traders sell their commodities using different measuring
units and at times the scales they use are not well calibrated. The price of common beans,
groundnuts and soybeans was also affected by periods of food abundance and shortages
because during food abundance sales happen at the lowest prices. Most traders are
informed on market prices (Table 24) and obtain such information through other traders
(data not shown).
70
Table 23: Trader categories in the target markets in the Sud-Kivu mandate area.
Market name
Legume
Category of traders
Rural
Urban
Urban
wholesalers
wholesalers
retailers
Cabwine-Mwami Groundnut
11
0
0
Common beans
12
0
0
Soybeans
11
0
0
Total
34
0
0
Kadutu
Groundnut
0
12
5
Common beans
0
12
6
Soybeans
0
0
18
Total
0
24
29
Kavumu
Groundnut
17
0
0
Total
17
0
0
Mudaka
Soybeans
16
0
0
Total
16
0
0
Mugogo
Groundnut
12
0
0
Common beans
6
0
0
Soybeans
18
0
0
Total
36
0
0
Total
103
24
29
Proportion (%)
66
15
19
Total
11
12
11
34
17
18
18
53
17
17
16
16
1
6
18
36
156
100
Table 24: Knowledge on market prices in the target markets in the Sud-Kivu mandate area.
Name of the
Information on
Legumes
Total
market
market pricing Groundnut Common bean
Soybean
Cabwine Mwami Yes
9
7
10
26
No
2
5
1
8
Total
11
12
11
34
Kadutu
Yes
15
18
16
49
No
2
0
2
4
Total
17
18
18
53
Kavumu
Yes
13
0
0
13
No
4
0
0
4
Total
17
0
0
17
Mudaka
Yes
0
0
10
10
No
0
0
6
6
Total
0
0
16
16
Mugogo
-77
0
0
1
1
Yes
8
3
12
23
No
4
3
5
12
Total
12
6
18
36
Total
57
36
63
156
Proportion (%)
Yes
79
78
70
-
71
9. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HN
NU
UT
TR
RIIT
TIIO
ON
N-R
RE
EL
LA
AT
TE
ED
D
A
AC
CT
TIIV
VIIT
TIIE
ESS
99..AA.. SSO
OY
YB
BE
EA
AN
N PPR
RO
OC
CE
ESSSSIIN
NG
GA
AN
ND
DU
UT
TIIL
LIIZ
ZA
AT
TIIO
ON
N
In the context of the legume activities, it was felt that substantial efforts need to be made
to expose farmer associations to the processing and utilization of soybean since the other
legumes (beans, groundnut) are well known to the farming communities. In 2007, various
training tools were prepared and a detailed strategy was developed with partners, including
health centers at the Action site level, to be implemented in 2008.
In terms of training tools, one training manual
for trainers was developed and translated in
French (Photograph 26), covering the
following topics: bases de la nutrition, présentation
du soja, hygiène de base, méthodes domestiques de
préparation des produits de soja, contenu nutritif des
produits de soja, recettes du soja, cartes postales des
menus de soja, teste d’acceptabilité par le consommateur,
and évaluation de la connaissance sur la transformation
et l’utilisation du soja.
In terms of strategy, activities are: (i) assessment
of the knowledge of soybean processing, (ii)
establishment of demonstration gardens, (iii)
training of trainers, and (iv) training of farmer
associations. Assessment of the knowledge of
soybean processing will aid organizing the
curriculum of the processing activities. Such
issues will be included in each training session
as a pre-test evaluation. During the training, a
Photograph 26: Cover of the soybean
wide range of products for different targets
processing training manual.
(markets, small scale business, household
consumption, there is potential food AID
market for NGOs on USAID Title II programs advocating for locally procured food AID
etc) will be produced and after acceptability studies, the project will promote what is
preferred by farming communities. Near the health centers, demonstration gardens will be
established, focusing on soybean production. An area of about 2500 m2 (25 are), will give
about 200 kg of soybean grains. Of this 200 kg, 20 kg will be reserved for planting the next
season. About 80 kg will be used for the processing training sessions and about 100 kg will
be used for giving out to people who have followed the training to try on their own plot
(to about 200 g per mama at 500 g per visiting female farmer). The gardens will be
managed by the health center with technical supervision by ISAR or INERA. Training
events will be organized by health center staff (see next paragraph on training of trainers)
around the demonstration gardens with attendance of female farmers visiting the health
centers. These will be given 2 documents (besides the soybean seeds): (i) a one-page folder
in local language related to the production (see draft attached) (Photograph 27) and (ii)
recipes that include soybean. Monitoring tools and monitoring timeframes will be
developed in order to follow up if the female farmers are really planting the soybean, how
72
this is working, if they are able to increase soybean consumption in their families, if this
has an impact on children’s health, if they would be interested in disseminating to other
people, etc.
Photograph 27: Soybean production flyer in French (left) and Kinyarwanda (right) to be
distributed to female farmers interested in producing soybean through the health centers.
In terms of training of trainer sessions (ToT), there will be one session per action site with
about 20 people attending. The ToTs will take about 3 days per session and need to be
held before end of April 2008 in order to have the receipt document ready to be used in
training the farmers belonging to the associations. The technical document will be
multiplied in 300 copies (50 for BasCongo, 50 for Kivu, 100 for Rwanda) to satisfy the
demand of the ToT sessions and have extra copies for extra demand by partners. For the
trainers to advance the training through training sessions with farmers, the project will (i)
facilitate trainers and (ii) develop tools to monitor if the trainers are organizing farmer
training events, who they train, what is happening after training, etc.
In terms of training of farmer associations, the associations that we are directly working
with will be trained by the nutritionists working in the different Mandate areas. This
includes about 8-12 associations with about 20-40 members or about 400 people. The
training sessions will be held before the end of June 2008 (before the next growing season)
but after having finalized the ToT sessions during which the final recipe book will be
developed. The recipe document in local languages will be multiplied and distributed
during the training. During these events, an acceptability test will be implemented for the
processed products. The soybean production flyer and the commodity fact sheet will also
be distributed. Again, monitoring tools and monitoring agendas will de developed for
creating the necessary feedback.
73
99..BB.. IIN
NIIT
TIIA
AL
LIISSA
AT
TIIO
ON
NO
OFF B
BA
AN
NA
AN
NA
AN
NU
UT
TR
RIIT
TIIO
ON
N--H
HE
EA
AL
LT
TH
H-R
RE
EL
LA
AT
TE
ED
DA
AC
CT
TIIV
VIIT
TIIE
ESS
Mrs. Beatrice Nakhauka Onyango Ekesa, a nutrition-health expert, was recruited as a
Dutch associate expert at Bioversity International. Beatrice will carry out nutrition-health
work in the framework of CIALCA and will specifically be looking at the potential
contribution of banana-based systems to the nutrition of small holder communities. She
will mainly carry out her CIALCA nutrition activities around Butembo/Beni in north
Kivu, DR-Congo and in Burundi. She submitted a CIALCA nutrition-health proposal to
the NGO ‘HealthNet TPO’. This proposal got approved and totals 22,330 $ (with both
HealthNet TPO and CIALCA financial contributions).
The different CIALCA/HealthNet TPO nutrition-health activities to be carried out
comprise:
•
•
•
•
•
Analysis of CIALCA baseline survey datasets;
Carry out nutrition surveys on the contribution of banana and plantain to the diet of
community members and on post harvest technologies in north Kivu, DR-Congo and
Burundi;
Carry out research on the links between agriculture, nutrition and health in Burundi and
north Kivu, DR-Congo;
Train community-own resource persons on sustainable agricultural interventions for better
nutrition and health in Burundi and north Kivu, DR-Congo;
Establish comprehensive demonstration gardens (including bio-fortified beans, soybean
and banana) in Burundi and north Kivu, DR-Congo.
74
10. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HM
MO
ON
NIIT
TO
OR
RIIN
NG
GA
AN
ND
D
E
EV
VA
AL
LU
UA
AT
TIIO
ON
N
Monitoring and evaluation activities take place at different stages of project
implementation and can be sub-divided in the following broad set of activities: (i)
monitoring and evaluating progress with implementation of the various projects, (ii)
monitoring and evaluation the various project interventions in collaboration with farmer
associations and development partners, and (iii) monitoring and evaluating the impact of
project interventions on farmer livelihoods.
M&E of project implementation
Project planning happens at three levels (ii)
across the three projects constituting CIALCA,
(ii) within a single project but across the
various mandate areas (Photograph 28), and
(iii) within a single project, within a specific
mandate area. A common agenda point of each
of these meetings is review of progress, either
through presentations of and discussions on
progress reports or through formal logframe
evaluations (Annex 1). Based on this, project
activities are adjusted, if necessary, and
prioritization is done for initiating new
activities. During subsequent meetings, this
process is repeated. This approach entails that,
although the level and range of activities
initiated across the different Mandate areas
were similar at project inception, this is not
longer the case today since progress with
activities in some Mandate areas is relatively
greater than progress with activities in other
Mandate areas.
Photograph 28: The Director General of
INERA, the Director of Science of ISAR
and the TSBF-CIAT Director cochairing a session during a TSBF-CIAT
general planning meeting.
M&E of project interventions
Various project interventions are being
evaluated with farmer associations, including
improved legume and banana germplasm and
natural resource management options
Photograph 29: Training session with
farmers on participatory evaluation of
(Photograph 29). A critical stage of each of
improved legume germplasm.
those activities is to get gender-differentiated
feedback on the various positive and negative aspects of the interventions demonstrated in
order to guide the demonstration of future interventions and to identify researchable
issues that need to be addressed either through controlled on-station field experimentation
or through more strategic greenhouse and laboratory experiments.
M&E of project impacts on rural livelihoods: During the baseline, several livelihood-related
indicators were quantified and these will be evaluated again towards the end of 2008 (end
of phase I of CIALCA) with an initial focus on the Action sites. At the end of a potential
CIALCA-II project, the impact assessment activities will be broadened to include the
various Satellite sites and eventually the total Mandate areas.
75
11. PRRO
OG
GR
RE
ESSSS W
WIIT
TH
HD
DE
EG
GR
RE
EE
E-R
RE
EL
LA
AT
TE
ED
D
A
AC
CT
TIIV
VIIT
TIIE
ESS
Capacity building is a crucial component of CIALCA since all countries in which CIALCA
operates are recovering from civil strife which has had a detrimental impact on the
capacity of various partners in the mandate areas.
MSc projects
In the context of CIALCA, 13 MSc-related projects are currently supported (Table 25).
Most of these are implemented in DR Congo and Rwanda since the TSBF-CIAT-led
project is not directly operating in Burundi. MSc projects are also supported for students
form the Belgian university partners.
Table 25: MSc projects supported by CIALCA.
Name
Nationality University
Julie
DR Congo Facultés
Lunzihirwa
Catholiques de
Kinshasa, DR
Congo
Rachel Zozo
DR Congo Makerere
University, Uganda
Muke
Manzekele
DR Congo
Aime Herikazi
DR Congo
Idja Sikyolo
DR Congo
Agnes
Mukandinda
Placide
Rukundo
Rwanda
Edouard
Rurangwa
Rwanda
Anaclet
Nibasumba
Burundi
Oswald
Ntakirutimana
Burundi
Rwanda
Topic
The impact of beans and groundnut
channels on the productivity and
agricultural income of households in the
cataractes area’.
Assessing the socio-economic importance
legumes-based on the livelihoods of
farmers at Mugogo and Mudaka Markets
in Ngweshe and Katana axes, Democratic
Republic of Congo.
Université de
Techniques d’amélioration de la
Kinshasa, DR
production agricole et de la stabilisation
Congo
des sols en pente au Sud-Kivu
Montagneux.
Makerere University The impact of soil tillage and mulch
Kampala, Uganda
application on soil physical properties and
productivity of banana-bean intercrop
systems.
UCG, Butembo,
Altitude effects on plant performace in
DR Congo
banana and plantain demonstration plots
and Musa collections in North-Kivu, DR
Congo
National University Nutrient flows in banana based cropping
of Rwanda, Rwanda systems.
Katholieke
Banana biotechnology.
Universiteit Leuven,
Belgium
Jomo Kenyatta
Initial survival of tissue culture bananas as
University of
affected by inoculation with arbuscularAgriculture and
mycorrhizal fungi.
Technology, Kenya
Université
Relationship between nutrients (cations)
Catholique de
in the soil mineral and organic pools and
Louvain-la-neuve,
nutrients at the banana root surface.
Belgium
Université de
Contribution a l'etude de l'état
Burundi
phytosanitaire du bananier dans les
Provinces de Gitega, Kirundo et Cibitoke'
76
Geoffroy
Germeau
Belgium
Ellen
Vandamme
Belgium
Julie
Vandamme
Belgium
Université
Catholique de
Louvain-la-neuve,
Belgium
Katholieke
Universiteit Leuven,
Belgium
Université
Catholique de
Louvain-la-neuve,
Belgium
Explaining banana yield differences in
Rwanda through quantification of banana
crop performance, soil fertility, pest and
diseases and crop management practices.
Nutrient deficiency and unavailability in
the soils of Walungu, South-Kivu,
Democratic Republic of Congo.
Analysis of stakeholder perceptions of
constraints and solutions in the banana
sector in Rwanda.
PhD projects
CIALCA is currently supporting 8 PhD students (Table 26). In addition, CIALCA is
actively supporting research staff to pursue further scholarship opportunities that can build
on the ongoing research. In that respect, there’s an outlook for two Belgian PhD students
and one IITA-Uganda staff to do their PhD within the CIALCA project on farming
systems, soil nutrient pools and recycling, and banana value chain and market analysis.
Table 26: PhD projects supported by CIALCA.
Name
Nationality University
Dowiya
DR Congo
Sokoine
Nzawele
University,
Benjamin
Tanzania
Tony Muliele
DR Congo
Université
Catholique de
Louvain-la-neuve,
Belgium
Svetlana
Rwanda
Université
Gaidashova
Catholique de
Louvain-la-neuve,
Belgium
Telesphore
Rwanda
University of
Ndabamenye
Pretoria, South
Africa
Josaphat
Rwanda
Katholieke
Rusisiro
Universiteit
Mugabo
Leuven, Belgium
Leon
Nabahungu
Rwanda
Syldie
Bizimana
Burundi
Célestin
Niyongere
Burundi
Wageningen
University, the
Netherlands
Université
Catholique de
Louvain-la-neuve,
Belgium
JKUAT, Nairobi,
Kenya
77
Topic
Characterization of Musa germplasm in
Eastern
DR Congo.
Soil moisture and soil physical constraints
in highland banana systems.
Research on banana-soil fertility-soil
biology interactions, with special emphasis
on the role of plant-parasitic nematodes
and abuscular mycorrhizal fungi (AMF).
Planting density, soil fertility, leaf nutrient
status and nutrient absorption
Agricultural intensification under
population pressure in Rwanda: An analysis
of fertilizers policy and legume-based
systems economic incentives.
Competing Claims on Wetland in Eastern
Rwanda: Challenges and opportunities.
Effect of soil management on nutrient
availability and nutrient recycling in
highland banana cropping systems.
Banana Bunchy Top Virus (BBTV) in the
Great Lakes region.
Undergraduate projects
CIALCA is currently supporting 13 Undergraduate students (Table 27).
Table 27: Undergraduate projects supported by CIALCA.
Name
Nationality
University
Topic
Matara Murhonyi
DR Congo
Université
Identifying fungal diseases affecting
(Memoire-Ingénieur)
Catholique de banana production in South Kivu.
Bukavu
Bahati Lukangira
DR Congo
Université
Quantifying the spread and importance
Catholique de of banana bunchy top virus (BBTV) in
Bukavu
South Kivu.
Kambale Mboho
DR Congo
Université
Pest and disease problems in banana
Catholique de systems in Nord Kivu.
Graben
Sereka Saghasa
DR Congo
Université
Understanding soil management in
Catholique de banana-based farming systems in Nord
Graben
Kivu.
Kakule Lukalango
DR Congo
Université
Characterizing and understanding
Catholique de banana germplasm diversity in Nord
Graben
Kivu.
Sondirya Tsongo
DR Congo
Université
Identifying socio-economic constraints
Michel
Catholique de in banana-based farming systems in
Graben
Nord Kivu.
Janvier Bashagaluke
DR Congo
Université
Assessment of erosion features in
Bigabwa
Catholique de farmers’ fields.
(Stage-Ingénieur)
Bukavu
Rehani Jumaine
DR Congo
Université
Demonstration of the microdosing
Catholique de fertilizer technique and of benefits of
Bukavu
high biomass-yielding legumes in
cereal-based rotation systems
Wivine
DR Congo
Université
Demonstration of improved agronomic
Zirhahwakuhingwa
Catholique de practices in cassava-legume
Munyahali
Bukavu
Chantal Karondo
Burundi
Université de
Etude de la diversité génétique du
(Stage-Ingénieur)
Bujumbura
germoplasme de bananier au Burundi.
Fidès Barigenera
Burundi
Université de
Evaluation de l'état phytosanitaire des
(Stage-Ingénieur)
Bujumbura
bananiers dans les communes les plus
productrices de banane de Gitega:
Giheta, Itaba et Makebuko.
Léonidas Ndikuriyo
Burundi
Université de
Détermination des équivalents
(Stage-Ingénieur)
Bujumbura
taxonomiques en nomenclature
Américaine (Soil Tax.) et FAO-INEAC
comme une façon de définition des
zones potentielles de culture du
bananier.
Félix Gatoto
Burundi
Université de
Enquête de prospection de BBTV sur
(Stage-Ingénieur)
Bujumbura
base de symptômes caractéristiques et
les pertes causées par le BBTV dans la
province de Cibitoke.
78
Post-doctoral positions
CIALCA is supporting post-doctoral positions:
•
Dr Joyce Mnyazi Jefwa, a Kenyan national, was recruited as a post doc to work on
AMF in banana systems in the framework of the Bioversity project. She will assess
AMF species associated banana cultivars and evaluate banana cultivars inoculated with
indigenous AMF for their performance in the nursery and in farmer’s fields.
•
Dr Pieter Pypers, a Belgian national, was recruited to work as a soil scientist within
the TSBF-led legume project. He’s backstopping the project activities in DRC and
Rwanda, and is particularly focusing on soil fertility aspects and the beneficial role of
legumes within the overall productivity of cropping systems.
•
Dr. Charles Murekezi, a Ugandan national, has been recruited as a project scientist
with IITA. Dr. Murekezi is a specialist in the field of banana agronomy and virology
and is based in Rwanda. He will spend 60% of this time to train and backstop the
banana research team at ISAR. Another 40% of his time will be spent in Burundi and
DR Congo to backstop other project activities.
Dr. Emily Ouma, a Kenyan national, has been recruited as a project scientist on
socio-economics with IITA. Dr Ouma has a background in agricultural economics in
the CGIAR system, having conducted her PhD on cattle traits in East Africa within
the framework of ILRI –led projects. She finished her PhD at the University of Kiel in
Germany. Dr. Ouma will be based at IRAZ in Burundi. Her role is to lead the socioeconomics research in the IITA-led project and to backstop and train partners in the
national instates of Rwanda, DR Congo and Burundi. DR. Emily Ouma will join
CIALCA as of February 1, 2008.
•
Training of national institute scientists
• Formal training events have been organized with national institute scientists and NGO
partners in the context of the PRA’s, baseline surveys, legume germplasm evaluation,
seed systems, banana nematology training, banana macro-propagation training,
monitoring and evaluation, and final characterization studies. Such training events are
aiming at cross-regional exchange of expertise.
•
On-the-job training of national institute scientists and NGO partner has taken place in
the context of agronomy, data collection, and sample processing.
•
Various MSc and PhD projects, mentioned above, are implemented by colleagues
from the national research institutes.
Training of farmer associations:
• In the context of legume activities, various farmer associations in each of the mandate
areas have been trained in participatory germplasm evaluation (including training on
planting, fertilizer application, weeding and harvesting techniques) and seed
multiplication and storage.
•
Plans are underway to initiate training events on soybean processing, market linkages,
and participatory monitoring and evaluation.
79
12. AN
NN
NE
EX
XE
ESS
A
AN
NN
NE
EX
X 11.. L
LO
OG
G--FFR
RA
AM
ME
EE
EV
VA
AL
LU
UA
AT
TIIO
ON
NSS
ANNEX 1.A. LOG-FRAME OF THE TSBF-CIAT-LED PROJECT
Objectively verifiable indicators/milestones
Work Package 1: Baseline information and site selection
1.1. By June 2006, all partners are involved in the planning and implementation
of the project activities in all action sites.
Achieved (see planning and meetings and review mission); new NGO partners engaged in all 4
mandate areas.
1.2. By June 2006, all necessary information is available to select action and
satellite sites for all mandate areas.
Achieved for action sites (see baseline and PRA reports); activation of satellite sites is currently
on-going (9 active in Bas-Congo).
1.3. By Dec 2006, sufficient information is available to direct marketing-related
activities for all mandate areas.
Only preliminary market surveys were conducted through student MsC and PhD projects
1.4. By Sept 2006, at least 5 action and 30 satellite sites have been identified
across all mandate areas.
Achieved for action sites (16 active); potential satellite sites are identified (see baseline and
PRA reports) but only 10 are currently active.
1.5. By March 2007, baseline information on livelihood status has been collected
in the action sites.
PRAs, baseline study and detailed characterisation studies have been conducted. Production of
the reports is currently on-going
1.6. By March 2007, farmer groups have been identified in all action sites.
Achieved: 39 farmer associations involved in the action sites, 61 farmer associations involved in
satellite sites.
1.7. By March 2007, farm typologies have been constructed that will form the
basis for evaluation of the appropriateness of specific technologies to specific
groups.
Detailed characterisation studies have been conducted at action site level; data analysis and
reporting is pending.
Work package 2: Participatory evaluation of best-bet options
2.1. By June 2006, a list of promising NRM options is available for initial testing,
taking into consideration the overall action site characteristics.
Achieved: NRM options have been identified and are currently being tested at action and
satellite site level.
2.2. Between Sept 2006 and the Dec 2008, the number of on-going on-farm trials
increases from 50 to 1000, across all mandate areas.
At the end 2007, over 200 on-farm trials have been conducted (not taking into account seed
multiplication fields). These include demonstration trials (about 120) and adaptation trials
(about 80). In addition, an estimated number of 350 seed multiplication fields have been
installed with farmer associations.
2.3. By Dec 2008, the impact of NRM options on various aspects of rural
livelihoods is evaluated in all action sites.
Baseline information has been collected; the activity itself is to be implemented in year 3.
80
Level of
success
100%
AS: 100%
SS: 20%
33%
AS: 100%
SS: 33%
75%
100%
33%
100%
100%
N/A
yr 3 activity
2.4. By the Dec 2008, the role of access to markets and centres de santé in
improving livelihoods is evaluated for all action sites.
2.5. By March 2007, seed multiplication is on-going in all action sites to satisfy
the demand in the action and the satellite sites.
Achieved: seed multiplication is currently on-going in all action and satellite sites.
Work package 3: Understanding mechanisms and contributions
3.1. By Dec 2008, sufficient knowledge on mechanisms driving tolerance to
drought and low soil P is available to guide breeding efforts.
On-going: a selection of potential P-efficient germplasm is being identified through the legume
evaluation activities; specific trials to unravel mechanisms of P tolerance will then be conducted.
3.2. By Dec 2008, relationship between soil fertility status and the nutritional
quality of bio-fortified crops is used by development partners to target
production of these crops.
On-going: micronutrient analysis of bean grains are pending – a preliminary analysis has been
conducted.
3.3. By Dec 2008, the potential for occurrence of positive interactions between
organic and mineral inputs is evaluated for the most common cropping systems
in each mandate area.
On-going: one or more NRM options in each of the mandate areas include specific treatments
for testing of positive interactions between organic and mineral inputs.
3.4. By Dec 2008, the contribution of resilient germplasm in driving overall
system resilience is understood for the conditions occurring in all mandate areas.
A long term trial will be initiated in 2008, using specific information from the detailed
characterisation study to investigate the contribution of improved germplasm to system resilience
and soil fertility.
3.5. Throughout the project life, new questions generated in the evaluation
efforts of Work Package 2 are addressed and fed back to these evaluation
activities.
On-going: such questions include for example nutrient deficiency studies in Sud-Kivu and an
investigation into the effect of nutrient inputs on cassava tuber quality.
Work package 4: Trade-off analysis and impact assessment
4.1. Once each year an annual planning meetings and once each season, an action
site meeting is organised.
Achieved (see planning and meetings and review mission)
4.2. By Sept 2006, a monitoring and evaluation framework is established and
operationalised.
Partly achieved: training was done; monitoring and evaluation of project activities is done
informally; an external review mission by DGDC was done in July 2007. A formal farmer
monitoring and evaluation framework needs to be put into operation.
4.3. By Dec 2006, local and scientific indicators have been identified to measure
progress with project interventions against baseline information.
In progress: a baseline was conducted; indicators are currently being prioritized as part of the
baseline report.
4.4. By the Dec 2007, products of the trade-off analysis are guiding the
introduction and evaluation of alternative NRM options, better suited to the
farmer production objectives and the environment of the actions sites.
In progress: discussions on-going for increased involvement of AfricaNUANCES to do tradeoff analysis; necessary information is available through baseline and detailed characterisation
studies. Selection of NRM options is principally guided by identification of constraints through
the baseline study and discussions with farmer associations.
81
Level of
success
N/A
yr 3 activity
100%
25%
50%
50%
0%
100%
100%
60%
50%
20%
Level of
success
N/A
yr 3 activity
4.5. By the Dec 2008, the impact of the project activities in the action sites,
satellite sites, and mandate areas is quantified, against the baseline information
collected.
Work package 5: Scaling up and out
5.1. By Sept 2006, partners in the action and satellite sites are aware of the project
100%
and ready to collaborate.
mandate areas.
5.2. By Sept 2006, partners have identified the optimum way of communicating
100%
project-related information.
Achieved: an arrangement is made to have a communication specialist to serve needs of all
partners in the 3 countries.
5.3. From Sept 2006 onwards, various initiatives are taken to facilitate farmer-to10%
farmer dissemination, including seasonal field days and farmer exchange visits
between action and satellite sites.
Not achieved: exchange visits were only organized in Sud-Kivu around the seed multiplication
activities. Farmer-to-farmer dissemination will be strengthened in 2008 through exchange visits
and field visits, led by NGO partners in action and satellite sites.
5.4. By Dec 2008, sufficient information is available to advice on optimum ways
0%
to disseminate and scale up project products, taking into account the overall
conditions of the mandate areas.
Not achieved: this requires strengthening by involving NGO partners and an in-depth analysis
of the baseline and characterisation studies on principal information distribution systems in the
mandate areas.
5.5. By Dec 2008, at least 10% of the farmers and 50% of local policy makers are
N/A
aware of the project products in the mandate areas.
yr 3 activity
The current status of activities allows an increased number of farmer adaptation trials in 2008;
knowledge of the project will be increased through dissemination events organized around the
various activities (including nutrition and processing activities with involvement of ‘Centres de
Santé’).
Work package 6: Capacity building
6.1. By Sept 2006, specific training needs for all stakeholders are identified.
75%
Partly achieved: training needs of project research, NGO partners and farmer associations are
addressed ad-hoc rather than through formal need assessments.
6.2. By Sept 2006, research for development teams have been identified in each
100%
action sites, comprising partners from the NARS and NGOs.
mandate areas.
6.3. Between Sept 2006 and Dec 2008, at least 50 farmer groups and 1000
100%
farmers across the various mandate areas have acquired the necessary skills to
test, evaluate, and adapt alternative NRM options.
Achieved: currently almost 100 farmer associations (with on average 15 members per
association) are involved in project activities of testing and multiplying germplasm and/or testing
NRM options.
6.4. By Dec 2008, at least 2 input dealers in each action and satellite site have
N/A
acquired sufficient knowledge to guide farmers in the most appropriate
yr 3 activity
management of inputs for their respective environments.
Not achieved: input dealer networks are poorly functional – alternatives using credit systems for
access to inputs and potential collaboration with other projects are explored.
82
6.5. By Dec 2008, at least 6 MSc projects have been submitted for defences and
at least 3 PhD projects are nearly completed.
On-going: currently 5 MsC and 3 PhD projects are being conducted
6.6. By Dec 2008, proceedings of a final symposium are submitted for
publication.
Level of
success
100%
N/A
yr 3 activity
ANNEX 1.B. LOG-FRAME OF THE BIOVERSITY-LED PROJECT
Level of
success
Work Package 1: Establishing Musa sector linkages within each country (INERA, ISAR,
ISABU, IRAZ)
1.1. Musa sector development framework established.
100%
Musa sector strategic plans (for Burundi, DR-Congo and Rwanda) developed by a wide range of
Musa stakeholders during a Bioversity-led planning meeting in Butare, Rwanda.
1.2. GIS-based compilation of information on agro-climate, production zones,
85%
socio-economics of farm communities and organisations.
Participatory rural appraisal [PRA], baseline and diagnostic surveys carried out across the 3
counties. GIS activities carried out in the CIALCA framework by a TSBF GIS expert.
1.3. Musa production zones characterized and pilot sites selected.
85%
Participatory rural appraisal [PRA], baseline and diagnostic surveys carried out across the 3
counties. Benchmark sites selected for demo-plot establishment, PhD, MSc and BSc studies, and
farmer participatory research.
1.4. Strategy and resource mobilization approach formulated and implemented.
70%
Musa sector strategic plans (for Burundi, DR-Congo and Rwanda) developed by a wide range of
Musa stakeholders during a Bioversity-led planning meeting in Butare, Rwanda.
1.5. Final plan developed to guide future sector development and resource
N/A
mobilization.
yr 3 activity
1.6. Methods guide for national sector development compiled.
N/A
yr 3 activity
Work package 2: Building Musa partnerships regionally (Bioversity-BARNESA,
Bioversity-MUSACO)
2.1. Country perspective shared with region / GIS based compilation of information
80%
/ pilot sites prioritized according to the regional perspective.
Musa sector strategic plan for the CEPGL region developed by a wide range of Musa
stakeholders from Burundi, DR-Congo and Rwanda during a Bioversity-led planning meeting in
Butare, Rwanda. GIS activities carried out in the CIALCA framework by a TSBF GIS
expert. Benchmark sites selected for demo-plot establishment, PhD, MSc and BSc studies, and
farmer participatory research according to the regional needs.
2.2. Regional agenda initiated.
85%
Musa sector strategic plan for the CEPGL region developed.
Research topics chosen according to the regional priorities.
2.3. Final results shared with regional networks and plan for future developed.
N/A
yr 3 activity
Work package 3: Serving international germplasm needs (Bioversity-KUL ITC:
INIBAP Transit Centre)
3.1. Germplasm collected and maintained
40%
The INIBAP Transit Centre (ITC) continues to maintain 1183 accessions and efforts are
underway to introduce more germplasm into the collection from different countries, especially from
the Democratic Republic of Congo (38 plantain accessions were received at the ITC in
December06-January07), Tanzania, Kenya, Republic of Central Africa, Uganda, Rwanda
and Burundi. This will ensure the conservation of Musa germplasm in face of the BXW
epidemic.
83
Level of
success
40%
3.2. Collection rejuvenated and cryo-preserved
Most of the rejuvenated germplasm is grown in the field, and during the next 2 years information
on their trueness-to-type is expected. Every day still 5-7 accessions are distributed from the ITC
worldwide and for the moment 598 accessions are already cryo-preserved.
3.3. PATHOGENS treated and germplasm disseminated
40%
Activities start on June 1st 2007
Work package 4: Integrating local and improved germplasm (INERA, ISAR, ISABU,
IRAZ)
4.1. Plan for collection, characterisation and conservation developed with regional
85%
perspective.
Musa germplasm collections are being established in north and south Kivu, DR-Congo by an
INERA PhD student. Info on the accessions will be entered into the Bioversity MGIS software
which will make it possible to link the DR-Congo germplasm to already established Musa
collections in Rwanda, Burundi, Tanzania and Uganda.
4.2. Local germplasm inventoried/characterised.
70%
A PhD student from INERA DR-Congo is collecting and characterising Musa germplasm
from eastern DR-Congo, and comparing this germplasm with Musa collections in Rwanda
[ISAR], Burundi [IRAZ] and Uganda [NARO/IITA]
IRAZ is strongly involved in the Musa germplasm activities.
4.3. In situ conservation piloted
pending
Pending the identification of local germplasm
4.4. Cultivar performance data compiled.
N/A
yr 3 activity
4.5. Cultivars in trials.
60%
21 Musa germplasm demo-plots (each containing over 20 genotypes) have been established in
contrasting agro-ecological zones across the 3 countries
4.6. Feasibility of alternate seed multiplication and dissemination systems
40%
diagnosed.
A macro-propagation training course was given to regional stakeholders (including CIALCA
partners) in the framework of a USAID funded project (C3P).
4.7. Regional plan on cultivar introduction, evaluation and seed multiplication
N/A
established.
yr 3 activity
Work package 5: Understanding stress resistance (supportive and strategic research –
KULeuven)
5.1. Abiotic stress resistance measured
40%
Activities start on June 1, 2007
5.2. Gene tagging and transgenic lines
40%
5.3. Fungal resistance genes characterized
40%
Work package 6: Developing improved production systems (INERA, ISAR, ISABU,
IRAZ with TSBF)
6.1. Technical options for soil fertility, plant nutrition and pest and disease
50%
management compiled from the region and beyond.
Info available in Bioversity’s Musalit database and several Bioversity project final workshop
proceedings.
Bioversity, France is also developing a Musa resource knowledge centre.
6.2. Plans established for on farm work with farmer research groups.
60%
Identified constraints and farmer’s needs emerging from the PRA/baseline surveys and diagnostic
surveys will determine the best-bet technologies to be tested on farm.
On farm trials with best-bet technologies to be established by both IITA and Bioversity during
September 2007.
84
6.3. Market oriented improved production systems developed.
September 2007.
6.4. Soil fertility and plant nutrition constraints identified.
PRA, baseline surveys and diagnostic surveys carried out across the 3 counties.
6.5. Biological and agronomic feasibility of soil improvement and plant health
options determined in research trials.
September 2007. A PhD study (ISABU) on Banana Bunchy Top Virus is ongoing. A PhD
study (ISAR) on planting density, soil nutrient uptake and leaf nutrient status is ongoing.
6.6. Farmer participatory research groups involved in on farm studies on soil
enhancement technologies.
September 2007.
6.7. Methods and examples compiled in a manual for use by other extensionists
and farmers.
Level of
success
45%
85%
55%
55%
N/A
yr 3 activity
ANNEX 1.C. LOG-FRAME OF THE IITA-LED PROJECT
Work Package 1: Baseline assessment (BASELINE)
1.1. Gather and synthesize available information on banana pest and disease and
soil constraints, banana markets, existing production and post-harvest
technologies, and cultivar distriubtion at national and regional levels; identify
NGO, farmer group, and development partners; select benchmark sites.
Achieved (see various planning meeting outputs and consultancy report).
1.2. Assess demand patterns, price and income elasticities, consumer preferences,
regional and international trade for fresh and processed banana products
Partially achieved – (Market surveys 100%, Farm gate prices 100%, Post-harvest prod. 75%).
1.3. Identify nutritional constraints to banana production in major production
zones
On-farm diagnostic data collected at all sites but few lab analysis left
1.4. Assess economic characteristics of existing technologies at national and
regional levels.
Achieved– all farm economics data collected, but modelling (see 1.6) in progress
1.5. Characterize in detail for 6 benchmark sites banana-based farming systems
(soil, pest and disease constraints, cultivars, farm and household characteristics,
and production patterns).
Finished in all sites.
1.6. Economically assess banana production constraints at sites
All data collected. Economic modelling of constraints and technologies ongoing.
Work package 2: Integrated banana systems development, evaluation, and
demonstration (SYSTEMS)
2.1. Develop and select with farmers three best-bet technologies per site for
further evaluation
Feedback workshops and technology selection done in Rwanda, South Kivu, and ongoing in
Burundi and North Kivu.
85
Level of
success
100 %
85%
90%
100%
100%
60%
75%
2.2. Multiply banana germplasm in low-cost multiplication centers in 3 countries;
production of starter material in tissue culture lab in Uganda
Macro-propagation training was provided in 2006. Centres have been established in DR Congo
and Rwanda, but Burundi is still trailing. TC material is continuously multiplied at IITA and
IRAZ
2.3. Demonstrate to farmers and development partners improved germplasm,
IPM and soil-improving technologies (and their integration) on-station and onfarm; obtain farmer feedback
Germplasm trials established. On-station mulch trials and on-farm trials started in Dec 2007
2.4. Evaluate the economic and yield benefits of best-bet banana integrated
practices (IPM, nutrient, and agronomic) imposed upon existing fields; discuss
and obtain feedback from farmers, adapt trials based on farmer feedback
2.5. Evaluate the economic and yield benefits of best-bet banana integrated
practices (germplasm, IPM, nutrient, and agronomic) imposed upon newly
established fields on-farm and on-station; discuss and obtain feedback from
farmers, adapt trials based on farmer feedback
2.6. Quantify on-farm nutrient flow dynamics with the objective of optimizing
nutrient cycling
PhD studies ongoing. First abstracts/papers submitted.
2.7. Disseminate production packages (germplasm, IPM and soil management) to
farmers within benchmark sites through trainings.
Work package 3: Post-harvest (POST-HARVEST)
3.1. Evaluate potential post-harvest options (processing and value adding) from
inside and outside the study areas.
Finished in Rwanda, but validation of results in Burundi, DRC needed. Also product
quality tests in laboratories remain.
3.2. Demonstrate to and train farmers in novel post harvest technologies to
farmers; adapt technologies with farmer feedback
3.3. Farmers trained in business plans for the post-harvest options identified in
year 2
Work package 4: Capacity-building (CAPACITY)
4.1. 2 PhD dissertation studies on topics relating to banana, soil types,
rhizosphere processes, nutrient uptake and pest/disease tolerance.
PhD students identified. Managed to be more cost efficient so recruited 3 PhD students for
UCL.
4.2. Public awareness of project goals and outputs increased throughout project.
Good interaction with partners and farmers but this will remain an continuous effort in the
project
Work package 5: Monitoring and evaluation (M&E)
5.1. Project monitored annually; progress on milestones assessed; logframe and
budgets adjusted as necessary; next year's activities planned; project staff
evaluated
Progressive activity; i.e. logframes and budgets are continuously adjusted when drawing new
agreements with partners. Planning through meetings with partners. Project staff evaluated
annually
86
Level of
success
75%
75%
N/A
yr 3 activity
N/A
yr 3 activity
50%
N/A
yr 3 activity
50%
N/A
yr 3 activity
N/A
yr 3 activity
50%
60%
66%
A
AN
NN
NE
EX
X 22:: F
FIIN
NA
AL
LC
CH
HA
AR
RA
AC
CT
TE
ER
RIIZ
ZA
AT
TIIO
ON
NT
TO
OO
OL
L FFO
OR
RT
TH
HE
E
LLE
EG
GU
UM
ME
E--B
BA
ASSE
ED
D SSY
YSST
TE
EM
MSS
CIALCA – TSBF-CIAT PROJET LEGUMINEUSES
SYSTEMES/SOLS, ACCES AU MARCHE et NUTRITION/SANTE
CARACTERISATION FINALE
Outils nécessaires pour cette activité
Chaque équipe aura besoin d’au moins:
- 1 GPS (toutes GPS doivent être calibrés avant le départ de l’activité)
- 1 sonde (à 0-20cm)
- 1 camera digital
- papier manila (A3) + crayons couleur (pour dessiner les cartes de la ferme)
- 1 × 1 m cadre pour mesurer la densité des légumineuses
- fil marqué avec des nœuds chaque 20cm pour mesurer la couverture des mauvaises herbes
- 1 calculateur à calculer le couverture des mauvaises herbes
- des sachets à stocker des échantillons des graines et les sols
- crayons pour étiqueter les échantillons
- mètre à ruban (MUAC tape – mesures anthropométriques)
- un pèse-personne (mesures anthropométriques)
- tableau de taille (mesures anthropométriques)
Procédure
La caractérisation finale est réalisée pour un sous-échantillon des ménages interviewés dans le
baseline. Cette sélection est faite par hasard. Dans chacun des villages (ou 20 ménages ont été
sélectionnés pour le baseline), 3 (au Rwanda) ou 4 (au Sud-Kivu/Bas-Congo) ménages sont choisis
arbitrairement pour la caractérisation finale. L’objectif est de caractériser 3 ménages par jour en
terme du rôle des légumineuses dans les sols/systèmes, nutrition/santé, et accès au marché+autres
aspects socio-économiques. Pour les aspects de nutrition/santé, le double de ménages doit être
caractérisé parmi les ménages avec au moins un enfant d’âge entre 2 et 5 ans. La caractérisation
finale vise à ajouter des données quantitatives au baseline. Avant la visite, une partie du baseline est
copiée au questionnaire de la caractérisation finale dans l’intention de faciliter le déroulement de
l’interview. Ces cellules sont délimitées en gras. Une équipe d’enquêteurs est constituée d’au moins
1 socio-économiste, 1 nutritionniste et 2 agronomes. Chaque membre de l’équipe est responsable
pour des sections spécifiques dans le questionnaire. Le questionnaire comprend une introduction
(Sections A-B), qui inclue le profil du ménage, suivi par la section sur la nutrition et la santé (Section
C), les sections agronomiques (Sections D-F) et les questions socio-économiques (Section G-J). La
dernière section (Section K) est un ensemble des questions agronomiques spécifiques sur les
parcelles comprenant des légumineuses.
Les membres d’équipe doivent optimiser le temps disponible. Les sections différentes peuvent être
complétées avec des différents membres du ménage. Après avoir rempli le profile du ménage, ces
membres peuvent être identifiés. Il faut faire attention de sélectionner les membres qui sont le
mieux placés à répondre certaines questions. Il est par exemple possible que l’épouse connaisse
mieux les caractéristiques des légumineuses que le chef du ménage. Les questions sur la nutrition et
santé doivent être posées à la maman. Les questions spécifiques sur les parcelles cultivées avec des
légumineuses peuvent être posées au chef, ou à un autre membre qui est impliqué dans la gestion
de la ferme et les opérations de champs. Les agronomes assistent d’abord ensemble avec le dessin
de la carte de la ferme (Section D). La carte est dessinée par le paysan, ou sous ses instructions. Un
deuxième membre doit être présent, qui doit joindre un des agronomes dans le champ pour
collecter les informations spécifiques sur les parcelles (après avoir dessiné la carte de la ferme).
L’autre agronome remplit d’abord les Sections E-F avant de joindre le premier agronome au champ.
Dans le cas ou il y a des champs à grande distance par rapport à la maison, les deux agronomes
peuvent travailler séparément (dans ce cas, deux kits de matériels sont nécessaires).
Alternativement, les deux agronomes peuvent partager les tâches. Le premier agronome fait
87
l’échantillonnage de sol et les mesures de densité de légumineuses et couverture par les mauvaises
herbes. Le deuxième agronome fait les autres observations et complète l’interview avec le paysan.
____________________________________________________________
Note: ‘0’ signifie mesuré et la valeur est zéro; ‘-99’ signifie manque d’information; ‘-88’ signifie
non-applicable; ‘-77’ signifie que le répondant ne connaît pas la réponse.
Détails sur les observations
1. détermination de la pente
Une personne monte 10 m sur la
pente (à mesurer à pas de pied),
pendant que une personne reste
en bas. Il faut alors estimer la
distance verticale entre les deux
personnes. La pente peut être
calculé à base de cette distance
verticale (cf. tableau):
distance verticale (m)
0m
entre 0 et 0.5m
entre 0.5 et 1m
entre 1 et 2m
entre 2 et 4m
plus que 4m
pente (%)
0%
entre 0 et 5%
entre 5 et 10%
entre 10 et 20%
entre 20 et 40%
plus que 40%
code
0
1
2
3
4
5
2. dessiner la forme des parcelles, numérotage des coins, prise des coordonnées géographiques et altitude
Des piquets sont mis à chaque coin et dans le centre principal de la parcelle. Apres, les
coordonnées (longitude et latitude) et l’altitude du centre principal sont notées, en utilisant un
GPS. Les coins des parcelles sont numérotés (en utilisant des codes ‘C1’, ‘C2’, ‘C3’, etc.) et les
coordonnées sont prises. La forme de la parcelle est dessinée et les positions des coins avec
leur numéro sont indiquées.
3. mesurage de la densité des légumineuses
La densité des légumineuses est déterminée en plaçant un cadre de 1 × 1 m au milieu d’une
diagonale qui connecte le centre avec un coin de la parcelle. Le nombre de plantes est compté.
Ceci est répété pour 3 diagonales choisies par hasard. La somme pour les 3 cadres est notée. (cf.
exemples pour des parcelles de forme triangulaire, rectangulaire, pentagonale et irrégulière)
4. mesurage de couverture par des mauvaises herbes
Un bout de fil de 5m avec des marquages chaque 20cm par un nœud est place au milieu d’une
diagonale qui connecte le centre avec un coin de la parcelle. La proportion des nœuds ou il y a
des mauvaises herbes sont calculée. Ceci est répété pour 3 diagonales choisies par hasard. La
moyenne pour les trois bouts est calculée et notée. (cf. exemples pour des parcelles de forme
triangulaire, rectangulaire, pentagonale et irrégulière)
5. échantillonnage et étiquetage des sols
Des échantillons de sol (0-20cm) sont pris sur chaque diagonale qui connecte le centre avec un
coin de la parcelle. Pour des champs avec 4 ou moins de coins, 2 échantillons sont pris sur
chaque diagonale plus un échantillon dans le centre. Pour des champs avec 5 ou plus coins, un
échantillon est pris sur chaque diagonale, plus un échantillon dans le centre (cf. exemples).
Toutes les sondes collectées dans la parcelle sont mélangées et un sous-échantillon de 250g est
gardé (une grande tasse). L’échantillon est étiqueté avec l’ID de l’exploitation (zone mandataire
/ site d’action / numéro de l’exploitation), la date d’échantillonnage et le code de la parcelle
(‘P1’, ‘P2’, ‘P3’, etc.). Les sols sont séchés au soleil le plus vite possible et stockés.
6. échantillonnage et étiquetage des graines
L’agronome laisse des sachets étiquetés au chef de ménage (zone mandataire / site d’action /
numéro de l’exploitation, le code de la parcelle (‘P1’, ‘P2’, ‘P3’, etc.), l’espèce de légumineuse et
le nom de la variété). Le paysan est demandé de garder un échantillon de 50 graines de la
récolte de chaque parcelle avec des légumineuses (en utilisant le dessin de la ferme). Il faut bien
expliquer au paysan comment prendre cet échantillon d’une manière représentative, et de bien
conserver ces 50 graines, protégés contre des ravageurs et humidité. Après collection, les
échantillons sont stockés pour analyse.
88
7. échantillonnage et étiquetage de compost /fumier
L’agronome laisse des sachets papier étiquetés au chef de ménage (zone mandataire / site
d’action / numéro de l’exploitation et le numéro suivant l’ordre dans le tableau en Section D.1.).
Le paysan est demandé de garder un échantillon d’environ 500g en mettant un petit peu de
matière dans le sachet chaque fois qu’il remplit un panier à appliquer dans le champ. Il faut
mettre le sachet au soleil et sécher l’échantillon. Cet échantillon sera collecté après le début de
la saison A’08.
8. prise des photos des parcelles et les systèmes de stockage de compost / fumier
Pour chaque parcelle avec des légumineuses, une photo est prise avec un membre de l’équipe
ou le paysan au milieu de la parcelle, tenant une étiquette avec le numéro de l’exploitation et le
code de la parcelle (‘P1’, ‘P2’, ‘P3’, etc.). Pour les systèmes de stockage de compost ou fumier,
le numéro de l’exploitation et le numéro suivant l’ordre dans le tableau en Section D.1 sont
indiqués.
89
C1
C2
C2
C1
1/
2
1/2
1/2
1/
2
C0
C0
C3
C4
densité des légumineuses – parcelle de forme triangulaire
C3
densité des légumineuses – parcelle de forme rectangulaire
C1
C2
C2
C1
5m
5m
C0
C0
C3
C4
couverture par des mauvaises herbes – parcelle de forme triangulaire
C3
couverture par des mauvaises herbes – parcelle de forme rectangulaire
C2
C1
C2
1/3
C1
1/3
1/3
1/3
1/3
1/3
C0
C0
C4
C3
échantillonnage de sol – parcelle de forme rectangulaire
échantillonnage de sol – parcelle de forme triangulaire
Légende:
piquet à démarquer le centre (C0) et les coins (C1, C2, C2,…) de la parcelle
ligne imaginaire connectant le centre avec un des coins de la parcelle
cadre de 1m x 1m à mesurer la densité des légumineuses
ficelle de 5m avec noeuds après chaque 20cm à mesurer la
couverture par des mauvaises herbes dans la parcelle
point à prendre des échantillons de sol
90
C3
C1
C1
C2
C2
C0
1/2
1/2
1/2
1 /2
C0
C5
C5
C6
C0’
C4
C3
C3
densité des légumineuses – parcelle de forme irrégulière
C4
densité des légumineuses – parcelle de forme pentagonale
C1
C1
C2
C2
C0
5m
C0
C5
C5
C6
5m
C0’
C4
C3
C3
couverture par des mauvaises herbes – parcelle de forme irrégulière
C4
couverture par des mauvaises herbes – parcelle de forme pentagonale
C1
C1
C2
C2
C0
1/2
1/ 2
C0
1/2
1/ 2
C5
C5
C6
C0’
C4
C3
C3
C4
échantillonnage de sol – parcelle de forme pentagonale
Légende:
échantillonnage de sol – parcelle de forme irrégulière
piquet à démarquer le centre (C0) et les coins (C1, C2, C2,…) de la parcelle
ligne imaginaire connectant le centre avec un des coins de la parcelle
cadre de 1m x 1m à mesurer la densité des légumineuses
ficelle de 5m avec noeuds après chaque 20cm à mesurer la
couverture par des mauvaises herbes dans la parcelle
point à prendre des échantillons de sol
91
CIALCA – TSBF-CIAT PROJET LEGUMINEUSES
SYSTEMES/SOLS, ACCES AU MARCHE et NUTRITION/SANTE
CARACTERISATION FINALE
SECTION A: INFORMATION GENERALE (10’)
1. Date de l’interview _______________________
2. Nom du chef de l’équipe __________________________________________________________
3. Noms des autres membres de l’équipe: ________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
4. ID de l’exploitation: (zone mandataire / site d’action / numéro de l’exploitation)
(même numéro que celui qui a été utilisé pendant le baseline) __________ /___________ /___________
Bas-Congo = BC
Kanga-Kipeti = 1
Lemfu = 2
Mbanza Nzundu = 3
Zenga = 4
Kigali-Kibungo = KK
Gatore = 1
Kabare = 2
Mayange = 3
Musenyi = 4
Sud-Kivu = SK
Kabamba = 1
Luhihi-Centre = 2
Lurhala-Centre = 3
Mwegerera = 4
Umutara = UM
Kabarore = 1
Murambi = 2
Nyakigando = 3
Rugarama = 4
5. Coordonnées GPS de la maison du chef de ménage (en dégrées décimales – à copier du baseline):
latitude (N/S) ______________; longitude (W/E) ______________; altitude: _____________ masl.
6. Observations sur la qualité de la maison:
Les murs (1=terre glaise; 2=bois; 3=briques; 4=autres: spécifiez) _____________________________________
Le toit (1=herbe; 2=tôles; 3=tuiles; 4=autres: spécifiez) ___________________________________________
92
SECTION B: PROFILE DU MENAGE (30’) [nom de l’enquêteur: _________________________________; temps début Section B: _______________]
Quelle activité
prend le plus de
(ans) niveau plus temps de ce
élevé atteint membre ?
0=aucun
1=primaire
2=secondaire
3=tertiaire
1=au champ
(propre ferme)
2=labour au champ
des autres paysans
3=ménage à la
maison
4=école
5=autres: spécifiez
répond aux questions?
1 = section B (profile de ménage);
2 = section C (nutrition et santé)
3 = section D (carte de la ferme);
4 = section E-F (questions agronomiques);
5 = sections G-J (questions soc.-économ.);
6 = section K (données des champs).
----------------------------------------------$ = a participé dans le PRA;
$$ = a participé dans le baseline;
# = membre d’une association participante;
## = accueille un essai de démonstration.
enfant sélectionné(e) entre 2 et 5
ans ? (marquez avex ‘×’)
1=chef
2=époux/épouse
3=fils/fille
4=parent du chef
5=petit-enfant
6=belle-famille
7=relatif du chef
8=relatif de l’époux/se
9=serviteur/servante
10=autres: spécifiez
scolarisation
(0=peu; 1=moyen; 2 =beaucoup)
relation par
rapport au chef
Combien de temps au champ
par rapport aux autres membres?
résident (0=non; 1=oui)
sexe (1=male; 2=féminin)
(définition: un ménage = un groupe de gens qui
vivent et mangent ensemble)
âge (ans)
members de ménage
nom
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[temps fin Section B: _______________]
93
SECTION C: NUTRITION & SANTE (60’-90’)
[nom de l’enquêteur: ___________________________; temps début Section C: _______________]
Note: dans les ménages additionnels à caractériser pour la nutrition, il faut d’abord faire la liste des enfants en utilisant le tableau de
Section B, et remplir les 6 premières colonnes (nom, âge, sexe, résidence, relation par rapport au chef et scolarisation).
Etat socio-démographique du ménage
1. Qu’est-ce qui est la source principale d’eau à boire pour l’enfant sélectionné(e) du ménage à ce moment?
(1=eau de robinet dans la maison; 2=eau de robinet dans la parcelle; 3=eau de conduite public; 4=puits; 5=rivière; 6=réservoir à
collecter eaux pluviales; 7=autres: spécifiez) ____________________________________________
2. Comment est-ce que vous préparez l’eau à la maison avant de boire? (1=bouillir; 2=filtrer; 3=laisser se déposer;
4=javelliser; 5=pas de préparation; 6=autres: spécifiez – plusieurs numéros possible) ______________________
3. Comment est-ce que vous stockez l’eau à la maison avant de boire? (1=réservoir fermé; 2=réservoir ouvert; 3=non
applicable (p.ex. eau de conduite); 4=autres: spécifiez) __________________________________________
4. Quel type de toilette utilisez-vous? (1=toilette à chasse; 2=latrine puisard; 3=toilette à trou; 4=pas de toilette; 5=autres:
spécifiez) ___________________________________________________________________
5. Qu’est-ce qui est la source principale de combustible à cuire dans le ménage? (1=électricité; 2=gaz; 3=pétrole;
4=charbon de bois; 5=du bois; 6=autres: spécifiez) _________________________________________
6. Conservez-vous les restes de nourritures après le repas? (0=non; 1=oui) ________________________
Si oui, pendant combien de temps? (0=mangé le même jour; 1=pendant 1-2 jours; 2=pendant 3-5 jours; 3=>5 jours)?
_____________________________________________________________________________
Si oui, décrivez comment vous conservez les restes. (Décrivez seulement pour les nourritures principales)
____________________________________________________________________________________
______________________________________________________________________
Informations spécifiques sur un enfant sélectionné(e) arbitrairement d’âge entre 2 et 5 ans
7. Décrivez comment vous avez alimenté votre enfant pendant les premiers 6 mois.
(1=allaitement au sein exclusif; 2=allaitement au sein + autres aliments (spécifiez); 3=lait de bétail frais; 4=lait en poudre;
5=lait soja; 6=autres aliments: spécifiez – plusieurs numéros possible) ____________________________________
________________________________________________________________________________
8. Décrivez comment vous avez alimenté votre enfant entre 6 et 12 mois.
(1=allaitement au sein exclusif; 2=allaitement au sein + autres aliments (spécifiez); 3=lait de bétail frais; 4=lait en poudre;
5=lait soja; 6=autres aliments: spécifiez – plusieurs numéros possible) ____________________________________
________________________________________________________________________________
9. Est-ce que votre enfant a une carte de santé ? (0=non; 1=oui) _______________________________
Si oui, note le poids et la date à la dernière visite au centre de santé. _________________________
_____________________________________________________________________________
Si oui, note la tendance de croissance après les dernières trois visites (1=bon gagne de poids; 2=pas de gagne de
poids; 3=perte de poids) ____________________________________________________________
10. Mesures anthropométriques à prendre:
-Poids actuel de l’enfant (kg) ______________________________________________________
-MUAC (circonférence du mi-haute bras gauche) (cm) ___________________________________
-Taille de l’enfant (cm) ___________________________________________________________
11. Est-ce que votre enfant a déjà été admis dans un centre de nutrition ? (0=non; 1=oui) ____________
Si oui, combien de fois? __________________________________________________________
Si oui, pour combien de temps? (spécifiez pour chaque admission) _________________________
_____________________________________________________________________________
Si oui, quel aliment est-ce que l’enfant avait reçu? ______________________________________
_____________________________________________________________________________
12. Présence de kwashiorkor/marasme après examen physique ? (0=non; 1=oui) __________________
Si oui, quels symptômes ? (1=manque de masse musculeux; 2=abdomen étendu; 3=peau ridé; 4=poils rêches et bruns;
94
5=oedème; 6=autres: spécifiez – plusieurs numéros possible) ________________________________________
_____________________________________________________________________________
13a. Votre enfant, a-t-il eu la diarrhée pendant les 3 mois passés ? (0=non; 1=oui) __________________
13b. Votre enfant a-t-il eu du sang dans les selles pendant les 3 mois passés ? (0=non; 1=oui) _________
13c. Votre enfant a-t-il eu difficulté à respirer pendant les 3 mois passés ? (la toux, respiration accélérée, respiration
nez/bouche bloquée, inhalation thorax difficile - 0=non; 1=oui) _____________________________
14. Votre enfant a-t-il été admis à l’hôpital pendant les 3 mois passés? (0=non; 1=oui) ______________
Si oui, combien de fois ? _________________________________________________________
Si oui, pourquoi ? ______________________________________________________________
Nutrition de l’enfant sélectionné(e)
15. Combien de fois votre enfant mange-t-il les aliments suivants? (Spécifiez le nombre de fois)
nombre de fois pendant les 7
derniers jours
lait frais de bétail
lait en poudre
lait soja
viande
poisson
œufs
légumes spécifiez
haricots
soja
niébé
arachide
pois cajan
manioc
banane
patate douce
mais
riz
fruits
spécifiez:
aliment de bébé formulé / fortifié, spécifiez:
bouillie des céréales
autres
spécifiez:
autres
spécifiez:
autres
spécifiez:
95
nombre de fois pendant les 30
derniers jours
20. 24-heures rappel de nourriture de l’enfant sélectionné(e) – date (jj/mm/aa): __________ /__________ /__________
Notez les aliments; spécifiez les quantités ou possible (p.ex. 1 tasse de thé: (2): 1, 1 tasse).
boisson
(1)
produits
laitiers (2)
viande/poisson/ de l’huile/ céréales
œufs (3)
lipides (4) (5)
banane/plantain racines/
(6)
tubercules (7)
légumineuses
(8)
légumes
(9)
fruits
(10)
petit-déjeuner
pause du matin
repas de midi
pause de
l’après-midi
repas du soir
(1): 1=thé; 2=lait; 3=de l’eau; 4=jus de fruit; 5=sucrée; 6=autres: spécifiez.
(2): 1=yaourt; 2=fromage; 3=autre: spécifiez.
(3): 1=boeuf; 2=chèvre; 3=porc; 4=poulet; 5=intestins; 6=poisson; 7=oeufs; 8=autres: spécifiez.
(4): 1=graisse d’animal; 2=l’huile végétale; 3=beurre; 4=margarine; 5=autres: spécifiez.
(5): 1=riz; 2=mais; 3=millet; 4=sorgho; 5=sosome; 6=pain; 7=autres: spécifiez.
(6): 1=banane dessert; 2=banana à cuire; 3=plantain; 4=matoke; 5=farine de banane; 6=chips de banane; 7=autres: spécifiez.
(7): 1=manioc; 2=yams; 3=patates douces; 4=pommes de terre; 5=autres: spécifiez.
(8): 1=haricots; 2=soja; 3=niébé; 4=arachide; 5=petits pois; 6=pois cajan; 7=autres: spécifiez.
(9): 1=chou; 2=amarante; 3=tomate; 4=oignons; 5=feuilles de manioc; 6=feuilles vertes (comme les épinards); 7=autres: spécifiez.
(10): 1=orange; 2=mangue; 3=papaye; 4=ananas; 5=fruits de passion; 6=avocat; 7=autres: spécifiez.
[temps fin Section C: _______________]
96
SECTION D: CARTE DE LA FERME (45’)
[nom de l’enquêteur: _______________________; temps début Sections D-F: _______________]
- A dessiner par le paysan ou sous son instruction sur un papier A3. Dépendamment de la situation, le dessin peut
être photographié et rester chez le paysan, ou peut être amené et réduit à taille A4.
- Indiquez où se trouve la maison du chef, les étables du bétail, les pâturages et les systèmes de stockage de compost
et/ou fumier.
- Indiquez toutes les parcelles de la ferme (soit louée, propriété ou commune); les parcelles qui appartiennent au chef
mais qui se sont louées à un autre paysan peuvent être indiquées, mais il faut alors demander ce qui se passe avec la
récolte. Indiquez si les parcelles sont contiguës (attachées) ou éloignées (détachées) de la maison du chef du ménage.
Pour les parcelles détachées, indiquez la distance entre la parcelle et la maison du chef, si nécessaire calculée en
termes de temps à se promener (1km = 12 minutes).
- Numérotez chaque parcelle (‘P1’, ‘P2’, ‘P3’, etc.).
- Indiquez les cultures actuellement (saison B’07) dans les champs (ou jachère) en bleu, et les cultures (ou jachère)
qui étaient aux champs dans la saison passée (saison A’07) en rouge.
Les informations spécifiques pour les parcelles (Section K), ne doivent être complétés que pour les
parcelles actuellement cultivées par une légumineuse (soit seule ou en association), et pour les parcelles
cultivées avec une légumineuse pendant la saison passée. Des parcelles qui appartiennent au ménage mais
qui se sont louées à autres paysannes, dont la récolte ne revient pas au ménage interviewé, ne doivent pas
être caractérisées.
Les informations suivantes doivent être copiées du baseline et peuvent aider à réaliser avec le dessin de la carte
de la ferme (copiez du baseline sections 4.2 et 4.5):
superficie
(en ha)
ordre
d’importance
1
2
3
4
5
6
7
8
9
10
distance par rapport
à l’habitation
champs de case
1
2
3
4
5
6
champs de marais/bas-fond
1
2
3
4
champs de collines
1
2
3
4
5
boisement
1
2
pâturages
1
2
97
liste de cultures
SECTION E: PROFILE DE LA FERME (30’)
1. Utilisation des intrants organiques
type de matière organique
(1)
spécifiez la source
Est-ce que c’est composté?
(0=non; 1=oui) si oui, pendant combien
de temps ? (nombre de mois)
quantité de matière organique système de
appliqué sur la ferme pendant stockage (2)
cette saison B’07 (en unités locales;
spécifiez le poids d’une unité locale)
(1): 1=fumier; 2=résidus des cultures; 3=engrais verts; 4=autres: spécifiez – indique plusieurs numéros pour des mélanges;
(2): 1=tas en plein air; 2=tas sous des arbres; 3=tas sous un toit; 4=fosse en plein air; 5=fosse sous des arbres; 6=fosse sous un toit;
7 = autres: spécifiez.
2. Gestion de bétail et fumier
nombre qui appartiennent
au ménage
nombre totale sur la ferme
type
gestion
matières utilisées comme
d’aliment aliment
(1)
(à remplir seulement pour le bétail en
stabilisation)
combien de fois Le fumier est-il collecté?
par semaine?
(0=non; si oui, combien de
1=oui) fois par mois?
vache locale
vache améliorée
mouton
chèvre
porc
(1): 1=pâturage individuel; 2=pâturage collectif; 3=divagation; 4=en stabulation avec aliments venant de l’extérieur; 5=en stabulation
avec plantation des fourrages/herbes; 6=achat des suppléments; 7=autres: spécifiez.
98
SECTION F: CARACTERISTIQUES DES VARIETES LEGUMINEUSES CULTIVEES (60’)
pure ou
mélangée?
(6)
% qui sera
vendu cette
saison (%)
taille des
graines
(5)
résisite à la
basse fertilité
de sol (4)
résiste aux
pestes et
maladies (4)
résiste aux
fortes pluies
(4)
(4)
résiste à la
sécheresse
(3)
précocité
besoin de
tuteur? (2)
(1)
haricots
1.
2.
3.
4.
niébé
1.
2.
3.
arachide
1.
2.
soja
1.
2.
3.
pois cajan
1.
2.
autres: spécifiez
production
Note 1: toutes réponses sont des opinions du paysan, pas des mesures absolus; le paysan donne des scores relatifs par rapport aux autres variétés.
Note 2: le marché local est le marché le plus proche (cf. baseline 7.4a); le marché régionale est le marché avec plus probablement le prix le plus élevé dans la région (soit urbain ou marché régional). Variations en prix
sont les prix minimum et maximum observés pendant l’année passée.
VARIETE
couleur
à remplir si le paysan connaît les prix, même s’il
ne vend pas
min-max prix sur le min-max prix sur le
marché locale
marché régional
(en FR/kg)
(en FR/kg)
-
-
-
-
0
0
-
-
0
0
0
-
-
0
0
-
-
-
-
(1): 1=très pauvre; 2=pauvre; 3=moyen; 4=bon; 5=très bon; (2): 0=non; 1=oui; (3): 0=précoce, 1=moyenne; 2=longue durée; (4): 1=résistance très pauvre (avec beaucoup de perte de production);
2=résistance pauvre (avec perte de production significatif); 3=résistance moyenne; 4=résistance bonne (mais le paysan connaît ou cherche encore des variétés meilleures); 5=résistance très bonne (la variété est
assez bonne et il n’y a pas de problème en terme de production; (5): 1=très petit; 2=petit; 3=moyen; 4=large; 5=très large; (6): 1=pure; 2=mélangé.
[temps fin Sections D-F: _______________]
99
SECTION G: APPUI INSTITUTIONEL (30’)
[nom de l’enquêteur: ____________________________; temps début Sections G-J: _______________]
Formation
1. Avez-vous suivi une formation au champ ? (0=non; 1=oui) ______________________________________
Si oui, combien de jours pendant les 5 dernières années? _____________________________________
2. Combien de fois par mois écoutez-vous des programmes d’agriculture? __________________________
3. Combien de fois avez-vous assisté à une réunion des associations paysannes / jour de champ pendant l’année
passée ? ____________________________________________________________________
Quel type de réunion ? _______________________________________________________________
Information sur le marché
4. Y a-t-il une organisation qui vous a appuyé sur la connaissance comment améliorer la commercialisation de vos
produits (0=non; 1=oui) _________________________________________________________
Si oui, quelle est le nom de cette organisation ? ____________________________________________
Si oui, sur quelle culture/élevage est-ce qu’ils vous ont aidé ? __________________________________
_________________________________________________________________________________
Si oui, quel type d’appui ont-ils fourni ? __________________________________________________
_________________________________________________________________________________
Si oui, quand était le dernier contact avec l’organisation ? _____________________________________
5. Comment obtenez-vous l’accès aux informations suivantes ? (marquez dans chaque cellule si 0=pas de source
d’information, 1=source principale d’information, ou 2=source secondaire d’information)
information sur
radio
journal
information obtenu par
visites au marché commerçants
voisins
ONG
autres:
spécifiez
engrais
minéraux
fertilisants
organiques
accès aux
sémences
caractéristiques
des variétés
qualité des
graines
prix des
graines
Commercialisation collective
6. Connaissez-vous des initiatives de commercialisation collective des produits agricoles? (0=non; 1=oui) ____
7. Avez-vous déjà participé dans une de ces initiatives ? (0=non; 1=oui) _____________________________
Si oui, quand était la dernière fois que vous avez participé ? ___________________________________
Si oui, pour la commercialisation de quelle culture/élevage ? __________________________________
_________________________________________________________________________________
Si oui, qui avait organisé cette initiative de commercialisation collective ? _________________________
_________________________________________________________________________________
Si oui, quels avantages en avez-vous tirés ? Expliquez pourquoi c’était avantageux. _________________
_________________________________________________________________________________
Contrats de production
8. Connaissez-vous des initiatives de production sous contrat ? (0=non; 1=oui) __________________________
9. Etes-vous actuellement impliqué dans une de ces initiatives ? (0=non; 1=oui) _______________________
Si oui, pour quelle culture/élevage ? _____________________________________________________
_________________________________________________________________________________
Si oui, quelle organisation achète vos produits ? ____________________________________________
Si oui, quels avantages en avez-vous tirés ? Expliquez pourquoi c’était avantageux. _________________
_________________________________________________________________________________
100
SECTION H: FONCTIONS ET CONTRAINTES DE COMMERCIALISATION (60’)
1. Avez-vous reçu une formation sur une des fonctions de commercialisation suivantes?
activité
formation
reçu?
(0=non;
1=oui)
de quelle organisation?
combien de fois pendant
l’année passée?
Ca a change la mise en oeuvre
de cette activité?
(0=non; 1=peu; 2=oui;
3=beaucoup)
rassemblage
triage
normalisation
mise en emballage
transport
transformation
Note: rassembler = collecter et mélanger des graines des différents parcelles/producteurs; trier = enlever les graines de mauvaise qualité; normaliser =
sélectionner les graines d’une certaine taille; emballer = mettre en certaines emballages ce qui augmente les possibilités de commercialisation; transport =
amener à un marché plus éloigné que le marché local; transformation = toute forme de conversion qui augmente les possibilités de commercialisation.
transport
embalage
normalisation
triage
rassemblage
légumineuse
transformation
2. Quelles légumineuses vendez-vous ? _____________________________________________________
____________________________________________________________________________________
Avant de vendre vos produits légumineuses, est-ce que vous faites les activités suivantes?
prix du
produit
brut
prix du
produit
converti
Trouvez-vous l’activité problèmes avec les activités
rentable par rapport à (question ouverte)
l’effort qu’elle
(en FR/kg) (en FR/kg) demande ?
(0=non; 1=peu; 3=oui;
4=beaucoup)
haricot
niébé
arachide
soja
pois cajan
autre:
spécifiez
Note: les prix des produits brut et converti se comparent à un moment donné par rapport à un paysan qui n’a pas fait l’activité.
3. Stockage et pertes pendant le stockage
légumineuse
normalement
stocké?
(0=non;
1=oui)
quelle
proportion?
(spécifiez le
%)
pendant
combien de
semaines?
raison pour
le stockage
(1)
perte pendant
le stockage
(2)
pourquoi cette perte ?
haricots
niébé
arachide
soja
pois cajan
autres:
spécifiez
(1): 1=impossible de vendre au moment de la récolte; 2=prix bas au marché; 3=stocké à consommer plus tard; 4=stocké comme semence;
5=autres: spécifiez (plusieurs numéros possible).
(2): 0=pas de perte; 1=peu de perte (moins que 10%); 2=entre 10 et 25% perdu; 3=entre 25 et 50% perdu; 4=plus que 50% perdu.
101
4. Quelles sont les contraintes principales pour la commercialisation de vos légumineuses?
Note: ces contraintes peuvent se rencontrer dans toute la filière de la production.
légumineuse
1er constrainte principale
2ieme constrainte principale
haricots
niébé
arachide
soja
pois cajan
autres:
spécifiez
3ieme constrainte principale
1=manque de terrain; 2=loyer de terrain cher; 3=manque de capital/crédit; 4=manque de labour; 5=bas fertilité de sol; 6=sécheresse;
7=pestes/maladies; 9=manque des variétés améliorées/adaptées; 10=manque de semences; 11=manque des engrais minéraux;
12=manque des intrants organiques; 13=manque de marché; 14=distance au marché; 15=mauvaise route au marché; 16=tracasseries
pour arriver au marché; 17=hautes taxes; 18=prix bas au marché; 19=haute fluctuations/incertitude des prix; 20=autres: spécifiez.
SECTION I:
INVESTISSEMENT DES BENEFICES OBTENUES PAR LA VENTE DES PRODUITS
(30’)
Comment dépenserez-vous l’argent réalisé avec la vente des produits agricoles?
(Rangez d’abord les dépenses, puis estime le pourcentage en utilisant 30 cailloux à partager parmi la liste de dépenses)
dépense
rang
% des
raison pour la décision sur la dépense
dépenses
loyer de terrain
labour pour les opérations de
champ
semence/matérielle de
plantation
engrais minéraux
intrants organiques
expérimentation et formations
d’agriculture
achat des aliments
frais de scolarité des enfants
frais médicaux de la famille
dépenses sociaux (divertissement
des visiteurs, mariages, enterrements,...)
autres: spécifiez
autres: spécifiez
autres: spécifiez
total
100%
SECTION J: CONCLUSION (10’)
Quelles sont vos 3 suggestions par ordre d’importance pour la commercialisation des légumineuses dans votre milieu ?
(question ouverte)
Priorité 1 ____________________________________________________________________________
_________________________________________________________________________________
Priorité 2 ____________________________________________________________________________
____________________________________________________________________________________
Priorité 3 ____________________________________________________________________________
____________________________________________________________________________________
[temps fin Sections G-J: _______________]
102
SECTION K: INFO PARCELLES
Code de la parcelle (numéro suivant le code (‘P1’, ‘P2’, ‘P3’, etc.) dans la carte de la ferme)
Altitude du centre principal de la parcelle (‘C0’)
Coordonnées GPS du centre principal (‘C0’):
Coordonnées GPS de chaque coin de la parcelle: coin 1 (‘C1’):
Croquis de la parcelle et numéro de chaque coin: coin 2 (‘C2’):
coin 3 (‘C3’):
coin 4 (‘C4’):
coin 5 (‘C5’):
coin 6 (‘C6’):
Attaché / détaché de la domaine de la maison (1=attaché; 2=détaché)
Position dans le paysage (1=plateau; 2=haute-pente; 3=mi-pente; 4=basse-pente; 5=vallée)
Pente (0=0%, 1=0-5%, 2=5-10%, 3=10-20%, 4=20-40%, 5=>40%)
Cultures actuellement au champ (liste les cultures)
S’il y a actuellement une légumineuse, spécifiez l’espèce
mesure la densité des légumineuses (nombre de plantes par 3m2)
S’il y a une légumineuse en association, spécifiez l’espèce en association
mesure la densité de l’espèce en association (nombre de plantes par 10×10m)
Mesure et calcule la couverture par des mauvaises herbes (%)
Type dominant mauvaises herbes (1=herbes, 2=feuilles larges, 3=impérata, 4=autres: spécifiez)
Présence des of roches, pierres, cailloux ou gravier à la surface
m
E/O
E/O
E/O
E/O
E/O
E/O
E/O
S
S
S
S
S
S
S
(échelle 1=0-5%; 2=5-25%; 3=25-50%; 4=50-75%; 5=75-95%; 6=95-100%)
Erosion visible (0=non; 1=à nappe; 2=à sillon; 3=à rigole)
Techniques d’in situ collection d’eaux pluviales
(0=absente; 1=zai; 2=plates-bandes; 3=plates-bandes serrées; 4=demi-lune; 5=autres: spécifiez)
Présence des structures anti-érosifs (0=absente; 1=végétatif; 2=structurel; 3=les deux)
si oui, type de structures anti-érosifs (spécifiez l’espèce dominante pour les structures végétatifs; spécifiez le
type pour les autres structures, p.ex. murs de pierre, fanya juu; fanya chini; terrasses; autres: spécifiez)
si oui, nombre de structures anti-érosifs (note: les structures en bas ne font pas partie de la parcelle)
Echantillon de sol (0-20cm) pris? (0=non; 1=oui)
Photo prise ? (0=non; 1=oui)
Propriété ? (1=propre propriété; 2=loué; 3=communal; 4=autres: spécifiez)
Quand est-ce que la parcelle a été prise en cultivation pour la 1ère fois ? (donne l’année)
Appréciation du paysan sur la fertilité du sol (1=pauvre; 2=moyenne; 3=bonne)
Nom local du type de sol (si connu)
Contrainte principale pour la production de la légumineuse
(1=érosion; 2=bas fertilité du sol; 3=mauvaises herbes; 4=pestes/maladies; 5=pierres; autres: spécifiez)
Labour de terrain ? (0=pas préparé; 1=avec hoe; 2=labouré avec vaches; autres: spécifiez)
Cultures au champ pendant la saison passée (A’07)
Cultures envisagées pour la saison suivante (A’08)
Spécifiez le nom de la variété légumineuse
Spécifiez la source de semence de la légumineuse
Engrais minéraux appliqués? (0=non; 1=oui)
si oui, spécifiez le type et la dose (en unités locales; spécifiez le poids d’une unité locale) type:
si oui, spécifiez le temps d’application (1=à la plantation, 2=autres: spécifiez)
mode d’application (1=épandu; 2=épandu et incorporé; 2=dans la ligne; 3=dans le poquet; 4=autres: spécifiez)
Intrants organiques appliqués? (0=non; 1=oui)
si oui, spécifiez le type et la dose (en unités locales; spécifiez le poids d’une unité locale) type:
si oui, spécifiez le temps d’application (1=à la plantation, 2=avant la plantation, pendant la
préparation du terrain; 3=à différents moments irrégulières pendant la saison; 4=autres: spécifiez)
mode d’application (1=épandu; 2=épandu et incorporé; 2=dans la ligne; 3=dans le poquet; 4=autres: spécifiez)
Insecticide ou herbicide appliqué? (0=non; 1=oui)
si oui, spécifiez le type (1=local: spécifiez; 2=produit chimique: spécifiez)
estime la production attendue (en unités locales; spécifiez le poids d’une unité locale en kg)
estime la production attendue des espèces en association (en unités locales)
Quoi sera/était fait avec la récolte de la légumineuse? (1=vendue entièrement, 2=consommée,
3=partiellement vendue, partiellement consommée, 4=autres: spécifiez)
Quoi sera/était fait avec les résidus de la légumineuse? (>1 option possible) (1=incorporés;
2=compostés; 3=aliments bétail; 4=mangés par la famille; 5=utilisé comme combustible; 6=vendus; 7=autres: spécifiez)
Le paysan a-t-il été informé de garder 3 cuillères de graines ? (0=non; 1=oui)
103
dose:
dose:
A
AN
NN
NE
EX
X 33:: L
LE
EG
G--22:: L
LEEGGU
UM
ME
EG
GE
ER
RM
MPPL
LA
ASSM
MD
DE
EM
MO
ON
NSST
TR
RA
AT
TIIO
ON
N
T
TR
RIIA
AL
LSS
Objectives:
This set of trials aims
(i) to obtain detailed farmer feedback on germplasm demonstrated;
(ii) to obtain a multi-locational (1 site = 1 replicate) agronomic evaluation of the germplasm demonstrated (G × E);
(iii) to obtained information on BNF of all germplasm demonstrated, on low-P tolerance in selected soybean
cultivars, on tolerance to low soil fertility in selected bean cultivars and on micronutrient contents in produce of
selected biofortified bean cultivars;
(iv) to multiply seeds for the following 2007 season.
Sites:
Demonstration trials are installed in a field identified by the associations.
Number of sites = 4 mandate areas × 4 action sites × 2 associations = 32 sites
mandate area
action site
association
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Kigali-Ngali/Kibungo
Musenyi
Mayange
A KAMEGA
B TUZAMURANE
A BENISHYAKA
Gatore
Kabare
Umutara
Nyakigando
Kabarore
Rugarama
Murambi
Sud-Kivu
Lurhala
Luhihi
Kabamba
Burhale
Bas-Congo
Zenga
Mbanza Nzundu
Kanga Kimpeti
Lemfu
A DUTERIMBERE A
B DUTERIMBERE B
A DUFATANYE
B INGANDURARUGO
A ISOKO Y’UBUMWE
B ABAHUJUMUGAMBI
A TWISUNGANE
B IMBARAGA
A DUFASHANYE
B IRIBA
A CINAMULA
B ALEMALU
A RUSINAME
B RHUBEHAGUMA
A MAENDELEO
B TUUNGANE
A APACOV
B ABAGWASINYE
A CALDZ
B ADESCO
A ADEKO
B ACKI
A APDKI
B ACDPP
A ADERKI
B APEKI
Germplasm:
Soybean and bean have been identified as test crops.
-soybean: selected TGx cv. from Nairobi, selected TGm cv. from Nairobi adapted to low-P conditions, 2 cv.
from Uganda, 6 cv. from ISAR adapted to high altitude.
-beans:
BIWADA cv. (drought-tolerant), BILFA cv. (tolerant to low soil fertility) and biofortified cv.
104
microplots
soybean
local variety
cv1: SB2(TGx1831-32E)
cv2: SB4(TGx1871-12E)
cv3: SB6(TGx1895-4F)
cv4: SB9(TGx1895-49F)
cv5: SB14(TGx1878-7E)
cv6: SB15(TGx1889-12F)
cv7: SB17(TGx1893-10F)
cv8: SB19(TGx1740-2F)
cv9: SB20(TGx1448-2E)
cv10: SB24(MAKSOY1a)
cv11: SB25(NAMSOI4m)
cv12: Duiker(ISAR)
cv13: Bossier 06B(ISAR)
cv14: 449/16/6 06B(ISAR)
cv15: Soprosoy 06B(ISAR)
cv16: Peka6 06B(ISAR)
cv17: Ogden(ISAR)
cv18: TGM1781(ISAR)
bush bean
local variety
climbing bean
local variety
BIOFORTIFIED:
cv1: BRB194
cv2: CODMLB003
cv3: CODMLB007
cv4: CODMLB030
cv5: CODMLB078
cv6: MAHARAGI-SOJA
cv7: MARUNGI
cv8: MLB49-89A
cv9: M’SOLE
cv10: ZKA93-10m/95
BIOFORTIFIED:
cv1: AND10
cv2: G59/1-2
cv3: KIANGARA
cv4: LIB1
cv5: MLV06-90B
cv6: MLV59/97B
cv7: VCB81012
cv8: VCB81013
cv9: VNB81010
BILFA:
cv11: AFR619
cv12: AFR708 (+biofort)
cv13: ARA4 (+biofort)
cv14: CIM9314-36
cv15: CIM9331-1
cv16: CNF5520
cv17: VEF88(40)L1P4T6
cv18: ECA-PAN021
cv19: HM21-7 (+biofort)
cv20: LSA144 (+biofort)
cv21: T8426F11-6F
cv22: UBR(92)25
cv23: ZAA5/2
BIWADA:
(only to be included in Umutara and BasCongo mandate areas)
cv24: GNP585(BIWADA)
cv25: Rab618(BIWADA)
cv26: Rab619(BIWADA)
cv27: Rjb7(BIWADA)
single lines
cv19: SB38(TGx1903-2F)
cv20: SB39(TGx1903-3F)
cv21: SB42(TGx1904-4F)
cv22: SB44(TGx1908-8F)
cv23: SB45(TGm1909-3F)
cv24: SB46(TGM1420)
cv25: SB47(TGm1511)
cv26: SB49(TGm1360)
cv27: SB51(TGm1196)
cv28: SB54(TGm1576)
note: BIWADA cultivars only to be tested in Umutara and Bas-Congo mandate areas, where drought problems
occur; if seeds available, also to be tested in Kigali-Ngali/Kibungo;
note: in each trial (every association) a local variety of soybean, climbing bean and bush bean variety should be
included;
note: varieties of which enough seed is available are included in microplots (4 lines of 2 m), varieties of which limited
seeds are available are included as single lines (1 line of 2 m).
Plant spacing
-soybean:
4 rows, 2 m long (plot = 4 × 0.75 × 2 = 6 m2)
0.75 m between rows and 0.05 m between plants within a row
drill seeds (2 cm depth) and thin to 5 cm between-plant distance at 3 weeks after planting
-bush beans:
4 rows, 2 m long (plot = 4 × 0.40 × 2 = 3.2 m2)
plant 1 seed per hill
-climbing beans:
4 rows, 2 m long (plot = 4 × 0.50 × 2 = 4 m2)
plant 1 seed per hill
note: area needed = 5.6 are without border zones; total area should be about 7 are.
105
Treatments
-soybean:
-bush beans:
-climbing beans:
with and without TSP applied at a rate of 30 kg P ha-1
(band application: make trenches and apply the TSP in the trench, cover with a little soil and
plant seeds at 5 cm distance between seeds)
25 kg P ha-1 = 125 kg TSP ha-1 = 12.5 g TSP m-2 = 75 g TSP per plot = 18.75 g TSP per row
with and without FYM (from goats) applied at a rate of 5 t fresh matter ha-1 (broadcasted over
the entire plot)
10 t fresh matter ha-1 = 1 kg m-2 = 3.2 kg per plot
with and without FYM (from goats) applied at a rate of 5 t fresh matter ha-1 (broadcasted over
the entire plot)
10 t fresh matter ha-1 = 1 kg m-2 = 4.0 kg per plot
note: amounts of inputs per demonstration trial (association) required are:
-TSP: 21 plots × 75 g TSP per plot + 10 single lines × 18.75 g TSP per row = 1.763 kg TSP per association
-FYM: 30 plots × 3.2 kg per plot + 12 plots × 4.0 kg per plot = 144 kg FYM (fresh matter) per association
note: the FYM needs to be well homogenized and representatively sampled. The moisture content needs to be
determined and a dried sample needs to be sent to Nairobi for analysis.
Trial design
Split plot design: 1 association = 1 replicate; main plots = species × inputs; split plots = cultivars
The various species and treatments are blocked in main plots; the trial comprises 6 blocks:
1. soybean, no inputs
2. soybean + TSP
3. bush bean, no inputs
4. bush bean + FYM
5. climbing bean, no inputs
6. climbing bean + FYM
Each block consists of two sub-blocks, a first comprising the microplots with different cultivars and a second
comprising cultivars in single lines. Within the various sub-blocks, the location of the various cultivars is completely
randomized. A possible design with the 6 blocks is attached.
note: each block includes also two microplots where maize and sorghum are grown (alternatively, a long and a
short duration maize can be grown) as reference crops for BNF assessment. For maize, 2 seeds are planted per
hill and thinned to 1 plant at 3 weeks after planting.
note: keep border areas between the various sub-blocks and between the various microplots.
note: keep the area weed-free.
note: spray insecticide as needed.
Biophysical observations
1. Soil sampling
Soil samples are taken at the block-level, prior to application of inputs. Per block, 9 cores (0-15 cm) are taken and
mixed, air-dried and stored. At least 2 kg is required for soil analysis. Following sampling strategy is followed:
1/6
1/6
1/3
1/3
1/6
1/3
1/3
1/6
2. Weather data
If possible, precipitation is recorded on a daily basis, but this is not essential as satellite data is also available.
106
3. Physiology and disease scoring
Physiology and disease scorings are recorded according to ECABREN protocols.
4. Biomass and BNF assessment
Aboveground biomass is sampled at 50 % podding (“50% formation des gousses”). Within the middle two rows of
the microplots, a random section of 50 cm of plants (equivalent to 10 soybean plants, or 5 bean plants) is cut at the
ground level. The random section should be at least 20 cm away from the border. The total fresh weight of the
biomass is taken, using an accurate electronic balance. Subsequently, the biomass is separated in leaves, stems and
pods, and the fresh weights of the 3 fractions are also recorded. The plant fractions are then dried (65 oC) and dry
weights are recorded, using the same balance.
At each time biomass assessments are made for species/cultivars that have reached the 50% podding stage (this
will occur at different times during the season), 3 maize and 3 sorghum plants are taken from the reference
plots. This is extremely important as these crops serve as a reference for determination of BNF in the legume
cultivars.
note: since nodule sampling, counting and analysis take a large effort, and since nodule numbers relate weakly with
BNF, nodulation will not be assessed.
5. Harvest
At crop maturity, all pods (“gousses”) should be harvested from the net plots (the net plot is obtained by removing
both outside rows and removing 20 cm – equal to 2 bean plants or 4 soybean plants – on each end of the middle
rows). The total fresh weight of the pods is taken (using the 2 kg balance) and a sub-sample is taken and
immediately weighed (using the 200 g balance). The sub-sample is air-dried, oven-dried (65 oC overnight), and
separated into grains and husks (“fanes”). The dry grains and husks are then weighted with the same 200g balance.
note: use of grains harvested: The grains will be used for (i) taste tests at harvest, (ii) use in the next generation of
demonstration trials (only some varieties), (iii) further multiplication by the farmer groups (some varieties), and (iv)
further multiplication on-station to keep a minimal stock of seeds (all varieties).
Activities around the demonstrations
1 .Farmer group training
At each important trial management event (planting, fertilizer application, weeding, spraying, harvest), it will be
essential for the associations to be trained on the most appropriate way to implement these management practices.
Members of the associations should be facilitated through the members of the ‘comités techniques’ of each
association. It is thereby essential that all members of the associations are present during these training events.
Information should be obtained on the members present during each such event (name, association membership,
gender).
2. Participatory evaluation of the germplasm
At flowering and harvest of the legumes tested, these should be evaluated by all members of the farmer
associations. At flowering, traits related to general growth, biomass production, and pest/disease resistance will be
evaluated, while at harvest, traits related to grain quality (taste, size, colour, cookability, etc) will be evaluated. A
standard protocol, drafted by CIAT, will be used as a tool for carrying out these evaluations. Team members in
all sites will be trained on how best to implement the protocol.
note: since the bush beans, climbing beans, and soybeans are going to mature after different times, this evaluation
events should take place at different times during the season.
3. Linking farmers to markets
Since the project puts a lot of emphasis on linking farmers to markets in order to get better output prices and
produce those crops that attract considerable market interest, activities will be initiated to foster these linkages.
Team members in all sites will be trained on how to initiate these activities towards November 2006.
4. Training on soybean processing
Since soybean is like going to raise issues of processing and marketing in the various locations, it is envisaged that
two events will be organised during the second season of 2006: (i) training on processing of soybean into
products for local consumption and (ii) visits of association representatives to sites where soybean processing
and marketing is on-going. Note that for the latter, local inventories of such activities are needed for the
different mandate areas. These activities are planned to be implemented around November – December 2006.
107
5. Integration of partners to prepare selection of satellite sites
Since we need to identify development partners that are potentially interested to disseminate products of the
current project through satellite sites, managed by these partners, we need to start engaging these around our own
activities at the action site level. This could happen best through the following activities:
(i) Organisation of a meeting with all potential stakeholders to understand their objectives and activities and
evaluate potential synergies.
(ii) Institutional analysis of partner organisations that are interested in collaborating.
(iii) Three-monthly meetings, preferably around the demonstration sites, maybe in the form of a field day each
season.
(iv) Exchange visits between the action sites.
Trial layout
Split plot design
1 association = 1 replicate, main plots = species × inputs, split plots = cultivars
mandate area
action site 1
action site 2
association B
association A
association A
association B
soybean microplots
L
soybean single lines
soybean microplots
L
soybean microplots
L
soybean microplots
L
soybean microplots +TSP
soybean single lines
climbing bean microplots
climbing bean microplots +FYM
bush bean microplots
bush bean microplots +FYM
action site 4
action site 3
association A
association B
association A
association B
soybean microplots
L
soybean microplots
L
soybean microplots
L
soybean microplots
L
108
3m
2m
0.75m
3m
2m
0.75m
3m
soybean microplots (no inputs)
0.75m
2m
0.75m
3m
2m
0.75m
3m
cv27
cv19
cv24
cv21
cv23
cv10
cv26
cv22
3m
cv18
2m
0.75m
7.5m
3m
cv4
0.75m
0.75m
cv20
cv25
cv18
cv24
cv27
cv19
3m
cv6
0.75m
cv22
cv26
cv21
cv23
2m
0.75m
soybean single lines (+TSP)
soybean single lines (no inputs)
7.5m
BLOCK 2
BLOCK 1
BLOCK 4
cv16
cv25
cv4
cv20
3m
cv11
0.75m
cv7
2m
L
0.75m
cv12
2m
cv8
0.75m
3m
2m
3m
0.75m
3m
2m
3m
0.75m
3m
2m
2m
0.75m
cv3
3m
3m
2m
3m
0.75m
cv16
0.75m
M
cv17
3m
2m
2m
0.75m
3m
0.75m
2m
cv1
3m
2m
3m
cv9
0.75m
cv14
0.75m
3m
2m
2m
S
3m
cv2
0.75m
cv5
0.75m
3m
2m
2m
cv15
cv6
0.75m
3m
cv13
0.75m
0.75m
soybean microplots (+TSP)
3m
2m
3m
cv12
0.75m
cv17
3m
2m
0.75m
cv8
0.75m
0.75m
cv15
2m
2m
cv5
0.75m
3m
cv2
0.75m
S
3m
2m
2m
3m
0.75m
3m
cv11
0.75m
2m
3m
2m
2m
3m
0.75m
3m
cv3
0.75m
3m
cv14
2m
1.6m
Detailed layout (set up in every association)
2m
2m
2m
2m
2m
2m
2m
3m
0.5m
2m
M L
0.5m
2m
cv8
0.5m
2m
2m
2m
cv9
0.5m
cv1
0.5m
S
0.5m
2m
2m
2m
2m
2m
2m
cv5
0.5m
L
0.5m
cv4
0.5m
2m
2m
2m
2m
2m
2m
cv3
0.5m
cv7
0.5m
cv6
0.5m
2m
2m
2m
2m
2m
2m
M
0.5m
cv8
0.5m
cv2 L
0.5m
climbing bean microplots (+FYM)
2m
2m
2m
BLOCK 6
cv2
0.75m
cv7
2m
2m
2m
2m
2m
3m
0.5m
2m
cv4
0.5m
2m
cv9
0.5m
cv7
0.75m
cv9
2m
3m
cv13
0.75m
cv1
2m
3m
L
M
0.75m
2m
2m
2m
2m
1.6m
2m
0.5m
2m
cv1
0.5m
2m
S
L
0.5m
2m
cv10
2m
2m
2m
2m
1.6m
0.75m
0.5m
2m
cv6
0.5m
2m
cv3
0.5m
cv5
1.6m
2m
2m
2m
2m
bush bean microplots (+FYM)
BLOCK 3
2m
BLOCK 5
note: L = local variety; M = maize; S = sorghum
2m
2m
2m
1.6m
109
cv9 cv21 cv4 cv13
2m
M
2m
0.4m
cv17
2m
2m
2m
2m
1.6m
1.6m
1.6m
0.4m
0.4m
0.4m
0.4m
0.4m
cv5 cv10 cv18 cv20
2m
cv3
2m
cv7
2m
1.6m
1.6m
0.4m
0.4m
0.4m
0.4m
0.4m
0.4m
1.6m
1.6m
cv16 cv24
2m
2m
2m
1.6m
L
2m
cv12 cv23 cv27
2m
1.6m
1.6m
0.4m
0.4m
0.4m
0.4m
0.4m
0.4m
1.6m
1.6m
1.6m
1.6m
1.6m
cv6
cv25 cv14 cv22 cv11 cv2
2m
2m
2m
1.6m
1.6m
0.4m
0.4m
0.4m
0.4m
0.4m
0.4m
2m
2m
2m
1.6m
cv8 cv19 cv15
cv26 cv1
S
1.6m
1.6m
0.4m
0.4m
0.4m
2m
2m
2m
1.6m
1.6m
1.6m
1.6m
1.6m
2m
2m
2m
2m
2m
0.4m
0.4m
0.4m
2m
2m
2m
cv18 cv9 cv13 cv3 cv24 cv11
2m
0.4m
0.4m
0.4m
1.6m
1.6m
1.6m
1.6m
1.6m
1.6m
0.4m
0.4m
0.4m
0.4m
2m
2m
2m
L
2m
0.4m
0.4m
0.4m
cv21 cv17 cv6 cv26
0.4m
0.4m
cv25
2m
1.6m
1.6m
1.6m
S
2m
2m
2m
2m
1.6m
1.6m
1.6m
2m
cv4 cv19 cv15 cv23 cv10
2m
2m
2m
1.6m
1.6m
1.6m
0.4m
0.4m
0.4m
0.4m
0.4m
0.4m
cv5 cv12 cv20
2m
cv1
2m
2m
1.6m
1.6m
1.6m
1.6m
1.6m
1.6m
0.4m
0.4m
cv14 cv8
2m
cv16 cv27 cv2
M
0.4m
0.4m
0.4m
0.4m
1.6m
1.6m
1.6m
0.4m
cv7 cv22
1.6m
1.6m
1.6m
2m
2m
2m
0.4m
0.4m
0.4m
0.4m
0.4m
1.6m
1.6m
1.6m
2m
2m
2m
climbing bean microplots (no inputs)
bush bean microplots (no inputs)
A
AN
NN
NE
EX
X 44:: Q
QU
UE
ESST
TIIO
ON
NN
NA
AIIR
RE
EU
USSE
ED
D FFO
OR
RL
LE
EG
GU
UM
ME
EG
GE
ER
RM
MPPL
LA
ASSM
M
E
EV
VA
AL
LU
UA
AT
TIIO
ON
NB
BY
Y FFA
AR
RM
ME
ER
RA
ASSSSO
OC
CIIA
AT
TIIO
ON
NSS
EVALUATION DES VARIETES D’HARICOT ET ESSAIS EN MILIEU PAYSAN
Données générales
Nom de l’Association : _________________________ Village : ________________________________
Nombre de femmes présentes : __________ Nombre d’hommes présents : __________
Date de l’évaluation : __________ Noms et sexes des facilitateurs : __________________________________
Première étape : Discussion en groupe
FICHE 1: Quelles sont les caractéristiques que vous voudriez trouver dans une bonne variété d’haricot nain?
Qu’est ce que vous considérez quand vous voulez sélectionner une bonne variété d’haricot nain?
FICHE 2 : Quelles sont les CINQ CARACTERISTIQUES les plus importantes que vous utiliserez pour évaluer
les différentes variétés ?
Critères d’évaluation
Ordre d’importance
Proposez d’ajouter les critères suivants s’ils n’ont pas été mentionnés, en expliquant que ceci intéresse la recherche:
1. performance en conditions de basse fertilité des sols (sans fumier) ; 2. réponse au fumier
110
CIALCA-TSBF-CIAT
Deuxième étape : Evaluation au champ
FICHE 3. EVALUATION OUVERTE DES VARIETES D’HARICOTS NAINS
Gendre de groupe : Hommes ou Femmes …………………………………… Nombre de participants : …………………………………………
Codes pour « Réponse au fumier »:
No
Variété
0= Moins que le traitement sans fumier; 1= Pas de différence avec le traitement sans fumier; 2= Seulement une légère amélioration;
3= Beaucoup plus que le traitement sans fumier; 4= Plus du double que le traitement sans fumier
Aspects positifs (+)
Nombre
rubans
BRB194
CODMLB003
CODMLB007
CODMLB030
CODMLB078
MAHARAGI-SOJA
MARUNGI
MLB49-89A
M’SOLE
ZKA93-10m/95
AFR619
AFR708
ARA4
CIM9314-36
CIM9331-1
CNF5520
VEF88(40)L1P4T6
111
Aspects négatifs (-)
Nombre
Reponse au
Rubans
fumier
ECA-PAN021
HM21-7
LSA144
T8426F11-6F
UBR(92)25
ZAA5/2
GNP585
Rab618
Rab619
Rjb7
VARIETE LOCALE
112
Troisième étape : Analyse préférentielle
FICHE 4. ANALYSE PREFERENTIELLE DES VARIETES D’HARICOT NAIN (5-6
VARIETES SEULEMENT)
Combien de personnes classent une variété comme 1iere, 2ieme, 3ieme, 4ieme et 5ieme ?
Variétés
1er
2e
3e
4e
5e
En cas de divergence, indiquer les raisons:
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
FICHE 5. MATRICE D’EVALUATION PREFERENTIELLE D’HARICOT NAIN
Caractéristiques
Variétés
Production
sans fumier
Réponse au
fumier
Demande au
marché
113
Total
Rang
Ordre
Quatrième étape : Restitution et planification pour la saison suivante
1. Restitution des résultats en groupe
1. Groupe de femmes : ______________________________________________________
2. Groupe d’hommes : _______________________________________________________
3. Commerçants : ___________________________________________________________
2. Demandez aux commerçants d’expliquer les filières haricot et les possibilités pour les associations de
participer à cette filière ?
_________________________________________________________________________________
_________________________________________________________________________________
3. Inviter les paysans à poser des questions aux commerçants. Rassurez-vous que les questions ciaprès sont posées.
3.1. Seriez-vous intéresser à acheter les haricots directement chez les paysans? Si oui, quels pourraient
être les termes et conditions d’achat ?
Conditions
Prix
Volumes
Minimum
Maximum
Qualité (pureté, humidité,
triage, emballage,…)
Quelles sont les variétés les
plus préférées ?
3.2. Que devraient faire les paysans s’ils doivent vendre mieux ?
_________________________________________________________________________________
_________________________________________________________________________________
_________________________________________________________________________________
Suggestions pour la saison suivante
1. Quelles sont vos suggestions pour la saison prochaine ?
_________________________________________________________________________________
_________________________________________________________________________________
_________________________________________________________________________________
2. Quelles sont les variétés que vous voudriez multiplier et réévaluer ?
Variétés à multiplier
Variétés à réévaluer
3. Quel sera le sort des variétés que vous n’avez pas sélectionnées ?
_________________________________________________________________________________
_________________________________________________________________________________
_________________________________________________________________________________
114
A
AN
NN
NE
EX
X 55:: A
ACCTTIIO
ON
N PPL
LA
AN
N FFO
OR
R IIN
NFFO
OR
RM
MA
AL
LL
LE
EG
GU
UM
ME
E SSE
EE
ED
D SSY
YSST
TE
EM
MSS,, 22000077B
B––22000088B
B,, T
TSSB
BFF--C
CIIA
ALLC
CA
A
Activity
Variety evaluation by
farmers
Responsible
Agronomists in
charge of sites
Farmer based variety
description
Training of farmers on
post-harvest management
Agronomists in
charge of sites
Phytopathology
technician.
Others to be involved
-Socio-economists
-Local leadership/extension
-Local partner organizations
-Traders
Farmers and site managers
Time frame
To be reported by
2nd week of August
2007
End of the 2007 B
season
ISAR/INERA legume
depending on
programs, NGO partners, local progress and need
sector staff
Association
Supported by agronomists,
depending on
members
NGO partners
progress and need
NGO partners
Local government extension
depending on
staff, technicians in charge of
progress and need
sites
NGO partners or -Local leadership and extension before planting
-Local partners
technicians in
charge of sites
Carry out post harvest
management (PHM)
Training/information
sharing on group and social
dynamism
To facilitate farmers seed
multiplication:
-determine the amount of
seeds and preferred varieties
-land preparation
-multiplication site visits in
order to produce
quality/quantity seeds
To facilitate the contracting NGO partners
of association members in
case of extra land required.
Crop establishment
Association
members
Association leadership and
before planting
members, technicians in charge
of sites and local partner
organizations
NGO partners, technicians in
at planting
charge of sites
Carry out field data
recording
Technical teams
or animateurs
charge of sites
during crop growth
115
Indicators (tracking changes)
Number of preferred varieties identified in
each site
Tools /necessary
Forms available
Number of bean flyers / brochures
produced
Number of training sessions held and
farmers trained
Excel sheets available
-Minimized storage losses
-Amount of bean dusted
Number of people trained and training
sessions held.
-Demonstration materials
(pesticide)
-Training manuals
-Pesticides
-Training of farmers
-Training documents (PPMR:
Project de promotion pour microrealization)
Area availed
Areas under contract availed
-Varieties and areas planted in each
association
-Date of planting
-Agronomic practices
Measurement tools, ropes,…
farmer fiches (provided by
Pieter)
Activity
Training of farmers
(associations) in seed
production techniques /
pest and disease control
Training of farmers on
seed/grain business
organization/marketing
Crop harvest
Evaluation of season’s
results
Carry out exchange visits
and field days
Facilitate seed marketing
and dissemination
Characterization of
-existing seed systems /
local varieties/mixtures
-existing seed channels and
their importance
-farmer profiles
Responsible
NGO partners or
technicians in
charge of sites
Others to be involved
charge of sites, legume
programs, RADA/SENASEM
Socio-economists NGO partners, technicians in
Farmer
associations
charge of sites
NGO partners or NGO partners, technicians in
technicians in
charge of sites
NGO partners
farmer association
representatives, NGO partners,
technicians in charge of sites,
legume programs,
RADA/SENASEM, local
policy makers, local radio
NGO partners or farmer association
technicians in
representatives, NGO partners,
charge of sites
technicians in charge of sites,
legume programs,
RADA/SENASEM, local
policy makers
Agronomists in
legume programs
charge of sites
Time frame
mid-season
Indicators (tracking changes)
sessions held
Tools /necessary
Training manuals
depending on
progress and need
sessions held
Training manuals
end of season
Varieties and area under the crops, quantity
of seed produced
farmer fiches (provided by
Pieter)
evaluation sheet (provided by
Pieter)
Physiological
maturity
Number of visites / field days and presence
of stakeholders
Refreshment/transport
standardized reporting format
depending on
progress and need
-Amount / varieties marketed
-No of farmer seed producers selling seeds
-No of farmers accessing or buying seeds
(which varieties?)
along the project
time
-Existing varieties before the interventions
-% of new varieties in the farmer
associations along the duration of the
project
-Characterization of access / availability of
new varieties
after harvest
116
A
AN
NN
NE
EX
X 66:: T
TRRAAIIN
NIIN
NG
GO
OFF C
CIIA
AL
LC
CA
A SSTTAAFFFF O
ON
NL
LE
EG
GU
UM
ME
E SSE
EE
ED
D
SSYYSSTTEEM
MSS A
AN
ND
D SSE
EE
ED
D PPR
RO
OD
DU
UC
CT
TIIO
ON
N
Date: 4-8/02/07; Venue: Butare, Rwanda
Trainees: Adrien Bahizi Chifizi (PF DIOBASS/BUKAVU), Kasereka Bishikwabo
(CIAT/BUKAVU), Sanginga JM (CIAT/BUKAVU), Mugiraneza Thierry (ISAR-Kibungo),
Habitegeko Francis (ISAR-Umutara), Ngoga Tenge Gislain (ISAR-Karama)
Subject maters covered during the training
• Elements and types of Seed Systems (both formal and decentralized) and how to link to improved
bean varieties. Objectives were:
1.
Understand the seed sector in DRC/Rwanda
2.
Characterize seed system actors
3.
Appreciate the actors’ roles in seed systems
• Strength, weakness and opportunity and threat (SWOT) analysis of existing seed systems in the
different projects/countries (DRC and Rwanda) (formal and informal/decentralized)
1.
Assess the strength of seed systems existing DRC/Rwanda
2.
Assess their weakness regarding the accessibility of improved varieties to farmers
3.
Assess the threats/opportunities they offer to accelerate improved access of varieties
• Planning of decentralized seed systems schemes (very important because the commercial seed
sector is quasi existing in DRC/Rwanda)
1.
Understanding elements necessary to establish a decentralised seed system
2.
Assess stakeholders and their roles
3.
Development of incentive systems
• Choosing the right crop and variety
1.
Select the best and preferred varieties by farmers
2.
Facilitate farmers’ selection
3.
Link variety selection to seed systems
• Partnership for strengthening community-based seed systems scheme
1.
To identify partners relevant in decentralized seed systems
2.
To appreciate their roles
3.
To map relationship models
4.
To sustain relationships and partnerships
• Designing community seed systems schemes: To use the above-mentioned elements and design a
decentralized seed scheme adapted to the participants’ conditions
Training methodology
-Plenary Participatory presentations and discussions/comments
-Case studies analysis conducted in groups and plenary presentation of the results
Workshop Evaluation and gaps to be filled
The workshop was assessed very well by the participants; however, since the workshop gave the
overview on existing seed systems in DRC/Rwanda without crop-specific seed management practices,
the following points/wishes were expressed by participants:
1. Soybean and bean seed multiplication sites
2. Seed business skills
3. Disease control and post harvest management
4. Crop-specific seed production (related to soil fertility)
5. Need to help participants to set up decentralised seed systems for both beans and soybeans
117
A
AN
NN
NE
EX
X 77:: A
ATTEELLIIEERR D
DE
E FFO
OR
RM
MA
AT
TIIO
ON
N SSU
UR
RL
LA
A
M
MU
UL
LT
TIIPPL
LIIC
CA
AT
TIIO
ON
ND
DE
ESS SSE
EM
ME
EN
NC
CE
ESS D
DE
ESS L
LE
EG
GU
UM
MIIN
NE
EU
USSE
ESS
PLATE FORME DIOBASS AU KIVU
219, Avenue P.E. Lumumba/Nyawera
C/o Bâtiment E.C.C/Diaconie et Développement
B.P. 1914 Bukavu/ RD Congo
E-mail : [email protected]
RAPPORT
ATELIER DE FORMATION SUR LA MULTIPLICATION DES SEMENCES
Période
: Du 04 au 05 avril 2007, Axe Katana
Du 09 au 10 avril 2007, Axe walungu
Bénéficiaires : Associations partenaires à la PF DIOBBASS
Filière
: Multiplication des semences autour des essais
agronomiques
Sigles et abréviations
ALEMALU: Association pour la Lutte contre la Malnutrition à Lurhala
APACOV: Actions Paysannes pour l’Amélioration des Conditions de Vie
A.D.E.P.B: Association de Développement pour les éleveurs de Petits Bétails
A.D.E.A: Association pour le Développement de l’Elevage et de l’Agriculture
CIALCA: Consortium for Improvement Agriculture-Based Livelihoods in Central Africa
CIAT: Centre International pour l’Agriculture Tropical
DIOBASS: Démarche pour une Interaction entre Organisations à la Base et Autres Sources de Savoirs
SENASEM: Service National de Semences
ITM: Institut des Techniques Medicales
FOMULAC: Fondation Médicale de l’Université Libre de Belgique en Afrique centrale
I. Introduction et contexte
L’introduction des nouvelles variétés de haricot et de soja, en septembre 2006 par le CIAT, a nécessité
de procéder à des essais démonstrations pendant 2 saisons dans les 4 sites d’actions.
Culture
Haricots nains
Haricots volubiles
Soja
Nbre variétés
introduites en début
saison A 2007
28
10
32
Quantité de
semences
Variétés adoptées en fin
saison A 2007
60 graines
60 graines
200 graines
5
5
5
A l’issu de ces essais et de l’évaluation des variétés qui s’en est suivi, les paysans ont choisis 5 bonnes
variétés de haricot et de soja à la fin de la saison A 2007 comme repris dans le tableau ci- dessus.
L’adoption de ces variétés a permis donc d’entrevoir la possibilité de les disponibiliser auprès des
paysans pour leur multiplication dans les champs paysans avec des perspectives de marché.
118
Ia. Durée de la formation
L’atelier de formation en multiplication qui s’est déroulé du 04 au 05 avril à Katana et du 09 au 10 avril
2007 à Walungu, avait pour but de former les paysans sur les aspects techniques de la multiplication en
perspective de la saison B 2007.
Ib.Objectifs de la formation
La formation poursuivait des objectifs suivants: (i) doter les paysans membres des associations, des
connaissances techniques nécessaire en matière de production des semences selon les normes du
Senasem; (ii) permettre aux paysans de prendre une décision de devenir multiplicateur de semences et
d’en faire une activité génératrice de revenu; (iii) outiller les paysans sur les éléments pouvant leur
permettre de faire une étude de marché tenant compte des réalités de leur terroir.
II. METHODES UTILISEES
IIa. Choix du lieu de formation
La formation en la multiplication s’est déroulée à l’ITM/FOMULAC Katana pour les paysans des
localités Kabamba et Luhihi, du 04 au 05 avril. Pour les paysans des localités de Lurhala et Burhale,
WALUNGU-CENTRE a servi de cadre à la formation en multiplication, du 09 au 10 avril 2007. Ces
lieux de formations ont été choisis en fonction de réalités sociologiques de chaque axe, l’accessibilité au
site et visualisation des essais en milieu ouvert et de la facilité de communication entre paysans de
même terroir.
IIb. Choix des associations participantes
La formation en multiplication des semences, a concerné 12 associations dont 6 sur l’axe Katana et 6
sur l’axe Walungu. Ces associations ont été désignées pour la plupart, en fonction de leur participation
active à toutes les phases des activités d’introduction, d’évaluation et de diffusion des germoplasmes de
légumineuses depuis septembre 2006
IIc. Choix des participants
Le choix des participants à la formation en multiplication, s’est faite en tenant compte de la
représentativité dans l’association et par sexe, comme repris dans le tableau suivant :
Participants
Axe
Localités
Association
Katana
Luhihi
Kabamba
Walungu
Kavumu
Lurhala
Burhale
Rhusimane
Rhubehaguma
Tuungane
Maendeleo
ADEPB
ADEA
Alemalu
Cinamula
Apacov
Abagwasinye
Rhucihangane
Bololoke
Hommes
2
2
2
3
1
1
4
1
3
1
1
1
Femmes
2
2
2
1
0
3
1
3
-
Total
4
4
4
4
1
1
4
4
4
4
1
1
La lecture de ce tableau montre que chaque fois 4 paysans ont pu être formé par association, à
l’exception de ADEPB, ADEA, Rhucihangane et Bololoke qui ont disponibilisé chacun 1 membre.
Cette discrimination est due au fait que les associations évoquées ci- haut ont joint la dynamique après
les autres.
119
Le critère de choix de ces 4 participants a été:
- la capacité de suivre la formation pendant toute sa durée
- la capacité de restituer les apprentissages acquis aux autres membres de l’association
- la présence d’au moins une femme dans l’équipe
La plupart des associations ayant un effectif moyen de 40 membres, il est donc possible qu’une
personne puisse restituer à 10 autres ; ce qui a conduit au choix de 4 participants par association.
Ainsi, le total a été de 18 participants par axe.
IId. Choix de formateurs
La formation en multiplication des semences a été conduite par une équipe d’Ingénieur agronome
ayant une certaine expérience dans le monde associatif et de la vulgarisation du matériel végétal.
Il s’agit de:
1. Bahizire Chifizi Adrien, Agronome PF DIOBASS: Chargé de l’organisation, du suivi et de
l’accompagnement des associations partenaires des sites d’actions, dans les activités CIAT-TSBF à
l’Est de la RDC. Expérience également dans l’accompagnement des associations partenaires de
DIOBASS sur les questions touchant le renforcement de la sécurité alimentaire et l’agriculture
durable.
2. Sanginga Jean-Marie, Agronome CIAT: Chargé des implantations et du suivi des essais de
démonstrations dans les sites d’action à l’Est de la RDC. A participé à des ateliers de formations
sur la multiplication des semences au FHI en RDC
3. Byakombe Mazambi, Coordonnateur de SENASEM Sud-Kivu: Formateur dans le domaine de
multiplication des semences.
IIe. Approche de la formation
La formation des associations en multiplication entre dans un processus de capacitation des paysans
membres, en outils pouvant leur permettre d’influencer positivement leur avenir. Ceci passe par la
formation et l’information, 2 clés essentielles pour ouvrir les portes du progrès au milieu rural et faire
reculer les murs de la misère, le tout dans une perspective de marché.
La finalité de la rencontre était également de renforcer les capacités des organisations dans le but de
trouver des solutions endogènes et durables aux questions que se posent les communautés.
La démarche Diobass vise à :
Améliorer la communication et les échanges entre tous les acteurs de développement :
paysans, techniciens, cadres ou animateurs paysans.
Développer une dynamique de réflexion au sein des communautés rurales, autour de
thèmes qui intéressent directement ces communautés
Promouvoir la recherche paysanne et valoriser les savoirs paysans.
Cette finalité implique la formation conjointe de cadres et de paysans au cours de périodes de stages ou
d’ateliers résidentiels, dans des situations concrètes de terrain, au sein d’organisation paysannes
structurées
IIf. Répartition des thèmes par formateur
Les thèmes de formation ont été repartis comme suit entre différents intervenants :
Module 1 : Processus de la production des semences
(Ir Sanginga Jean-Marie / CIAT-TSBF)
Celui-ci a tout d’abord brossé succinctement les activités CIAT dans cette partie de RDC et les
partenaires locaux, régionaux et internationaux qui collaborent dans la réalisation des projets agricoles
en milieu paysan et dans le cadre de la recherche. Il a montré comment, à partir 60 graines de
différentes variétés de haricot (18 au total) et de soja (32 au total), introduites en septembre 2006, il a
été possible d’en sélectionner 5 bonnes variétés, qui font l’objet d’un atelier de formation en ce jour.
120
Il a, en plus, attiré l’attention des participants sur l’importance alimentaire et nutritionnelle des
légumineuses notamment le haricot et le soja. A partir de cette dernière culture, il est donc possible de
produire, le lait, la viande et même l’huile de cuisine.
Processus de production de semences
Le processus de production ou chaîne de production des semences a été donné comme suit :
- Semences de souches (Breeder seeds) : Au niveau des sélectionneurs
- Semences de pré-base (selected seeds) : Après 3 générations
- Semences de base (registred seeds) : après 1 génération
- Semences certifiées : avec les agriculteurs- multiplicateurs, les associations, les privés, sous contrôle
du SENASEM
Module 2 : Elements techniques sur la production des semences de soja et de haricots (Ir
Byakombe/ Senasem)
L’intervenant a donné la fiche technique du haricot et du soja :
1.. Fiche technique relative à la production des semences des variétés de
Soja (Nom Scientifique : Glycine max)
1.1. Les variétés :
Impérial, TGX888-49C, MUNANGA (TGX814-26D), AFIA (TGX849-249D), TGX573-209D,
JUPITER, UFV, SIATA 194.
Ces variétés sont celles qui sont inscrites, jusque là, au catalogue national des semences en RDC et
cerfifiées par le Senasem
1.2. La qualité des semences à multiplier (Normes)
Les exigences concernant la production de semence du soja sont les mêmes que pour que l’arachide. Il
faut souligner que la germination du soja démunie sérieusement pendant le stockage. C’est pourquoi,
avant le semis, une nouvelle analyse de la faculté germinative est absolument nécessaire pour pouvoir
assurer une densité normale de la culture. La parcelle utilisée n’aura pas portée du soja pendant 2
saisons, sauf s’il s’agit de la même variété et génération ou d’une génération antérieure et que cette
multiplication ait été agréée. Elle sera vierge de toute repousse accidentelle de la même espèce.
1.3. L’isolement
Pour la production de semences de base, l’espace d’isolement entre différentes variétés de soja est
de 30 mètres, et de 5 mètres entre mêmes variétés. Pour la production de semences certifiées, cet
espace est de 5 mètres entre différentes variétés de soja et de 1 mètre entre même variété.
1.4. Préparation du terrain et fumure de base
Sur les sols légers (sableux à sablo-limoneux), un labour léger est suffisant. Sur les sols plus lourds
(texture limoneuse), un labour profond (15à 20 cm) suivi d’un hersage énergétique pour écraser les
mottes est à recommander. La végétation grossière qui ne peut être enfouie avec le labour doit être
sortie du champ semencier. Une fumure de base de 100 à 150 kg/ha d’engrais NPK (17 :17 :17) est à
recommander de l’inoculum pendant les semis avec le Rhizobium
1.5.Le semis.
Le semis du soja se fait soit en ligne continues soit en poquets en utilisant 40à 80 kg/ha et démarier à
1-2 plantes par poquet, 7 jours après levée.
121
•
Les périodes de semis : les trois saisons (A, B et C avec irrigation d’appoint) conviennent à la
culture du soja. En saison A, on sème après les premières pluies de manière à ce que la maturation et la
récolte coïncide avec la petite saison sèche, afin d’éviter la pourriture des gousses.
•
Les écartements : Selon les variétés et la fertilité du sol, les écartements suivants sont
recommandés :
Semis est mécanique (ligne continue) : 60cmx5cm ce qui correspond à 50-80kg/ha
Semis manuel: 60cm x 5cm ; 40-60 cm x 20 cm avec 2 plants par poquet ce qui correspond à 40-60 kg
de semences par hectare.
•
La profondeur : 2 à 4 cm
1.6. Les travaux d’entretien
Les travaux d’entretien consistent notamment en sarclages suivis de binages, afin d’aérer le sol et
détruire les mauvaises herbes. A cet effet, au moins deux (2) sarclages sont indispensables : le premier
intervient 7 jours après la levée ; le deuxième à la fructification et le troisième selon le degré
d’envahissement par les mauvaises herbes.
1.7. Récolte, conditionnement et conservation
•
Récolte : Elle intervient avant 4 mois après le semis selon la variété, dès la maturité qui se
remarque par les signes ci-après :
•
•
•
Les feuilles jaunissent complètement et tombent ;
Les gousses se dessèchent ;
La graine résiste à la pression des doigts et prend sa couleur caractéristique.
Il faut récolter avant l’éclatement des gousses, surtout pour les variétés à gousses déhiscentes ; couper
les tiges à ras du sol, et non les arracher pour ne pas priver le sol des nodosités développées sur les
racines.
•
Battre les tiges pour détacher les gousses, mais décortiquer les gousses à la main pour éviter de
blesser l’embryon du soja qui est très sensible ; les graines sont ensuite triées en vue d’éliminer
celles qui sont malformées, trop petites, blessées ou attaquées, ainsi que les corps étrangers.
•
Enrober les semences, au Super Homai¨ à la dose de 10gr de produit pour 10 kg de semences ou
du ALMTHIO, un mélange de Thirame et de Lindane à la dose de 250 gr pour 100kg de
semences, afin de les protéger contre les insectes et les champignons ; puis les ensacher, les
exposer au soleil pendant une journée, pour éliminer l’humidité due à l’enrobage et conserver
dans un endroit sec, aéré et frais.
1.8. Le contrôle des cultures
Un minimum de deux (2) contrôles sont indispensables : le premier à la floraison, le second avant la
récolte, après la chute des feuilles. Lors des contrôles on observe les hors-types et vérifie la bonne
exécution des épurations. Il est parfois nécessaire de procéder à des inspections supplémentaires en cas
de problèmes particuliers.
En pratique, l’inspecteur s’assurera que les plants du soja présentent bien les caractéristiques de la
variété, puis examinera la bordure du champ afin de vérifier l’isolement. Il inspectera ensuite le champ
dans son ensemble fera une évaluation des plantes adventices présentes et de la situation
phytosanitaire.
Lots de cette inspection, l’inspecteur examinera avec soin 150 plantes choisies au hasard en cinq
endroits différents du champ (à raison de 30 plantes par endroit) et établira séparément le nombre des
plants non conformes aux caractéristiques de la variété et le nombre de plants d’autres espèces
cultivées dont les graines sont de dimension comparable. Le champ sera déclaré impropre si l’on
dénombre plus de trois (3) hors-types (2%), ou plus de trois (3) plants d’autres espèces cultivées (2%).
122
Lors des contrôles, l’attention doit être sur les plants malades (bactérioses).Un champ portant plus de
1 plant malade sur 200plants contrôlés sera refusé comme champ semencier. Les champs de
production de semences du soja ne porteront pas plus de 0,1 % d’adventices nuisibles.
Pour les travaux du laboratoire, on prélève un échantillon de 1.000 gr pour le dénombrement et les
autres analyse, dont 400gr pour l’analyse de pureté.
2. FICHE TECHNIQUE RELATIVE A LA PRODUCTION DE SEMENCES DES
VARIETES DE HARICOTS (Nom scientifique : PHASEOLUS VULGARIS)
2.1. Les variétés
Pour les régions des Kivu :
• Variétés VOLUBILES : ALIYA, VCB 81012 (jaune), AND 10 (blanche striée de rouge),
kiangara, pigeon vert,
• Variétés NAINS : KIRUNDO, G2858, M’MAFUTALA, M’solé, D6 Bean, Sugar beans,
blanket,…
Ces variétés sont celles qui sont inscrites, jusque là, au catalogue national des semences en
RDC et cerfifiées par le Senasem
2.2. Le choix de terrain
Pour produire des semences de haricot, le terrain doit avoir une texture moyenne, assurant un bon
drainage. Les sols neutres légèrement acides, bien labourés, propres (sans mauvais herbes) sont
préférables.
2.3. La qualité des semences à multiplier
Les normes de qualité pour les semences de haricot et de niébé sont les suivantes :
Semences de « base »
Semences « certifiées »
Pureté spécifique (minimum)
97%
95%
Pureté spécifique (variété) (minimum)
98%
97%
Matières inertes (maximum)
2%
3%
Graines de mauvaises herbes
0,005%
0,1%
Graines d’autres plantes cultivées
4%/kg
10%
(maximum)
70%
70%
Germination (maximum)
12%
12%
Humidité (maximum)
2.4. L’antécédent cultural
Les meilleures culturales précédant le haricot sont la pomme de terre, le mais et le sorgho ayant été
bien entretenus par des sarclages, binages et désherbage. Il est conseillé d’attendre une campagne avant
de faire revenir le haricot sur le même terrain.
2.5. L’isolement
Pour la production des semences certifiées cet espace est de 5 mètres entre différentes variétés de
haricots et de 1 mètre entre même variété. Pour la production des semences certifiées, cet espace est 5
mètres entre différentes variétés de haricot et de 1 mètre entre même variété.
2.6. Préparation du terrain et fumure de base
Le labour à une profondeur de 20 cm est indiqué. Avant le semis, on plane le terrain avec un ou deux
passages de pulvérisateurs. Avant le pulvérisation, on apporte la fumure de base: 30 à 50 unités d’azote
de 30 à 40 unités de phosphore (150-200kg/ha NPK 17:17:17).
123
2.7. L’époque de semis
Elle est en fonction des régions et du cycle végétatif de la variété. On cherchera toujours à arriver avec
la maturation en fin de saison de pluies et début de saison sèche.
2.8. Le semis
Sur les petites parcelles semencières, le semis se fait à la main, en lignes de 60 cm et en assurant 4 à 5
cm entre chaque poquets, de telle manière qu’on réalise une densité de semis de 33à 40 graines/m2 sur
les superficies plus importantes, le semis est réalisé avec le semoirs de précision, en assurant la même
densité. Avant le semis, un traitement des semences avec un fongicide (cryptodine, bénomy) est
nécessaire pour protéger les jeunes plantules, contre les bactérioses, l’anthracnose, etc. La profondeur
de semis est de 3à 4 cm, en fonction de la texture du sol.
2.9. Les travaux d’entretien
Même si on utilise des herbicides, un ou deux sarclages sont nécessaires pour éliminer les mauvaises
herbes et aérer le sol. Le premier sarclage sera exécuté 10 à 15 jours après la levée et le deuxième 20
à30 jours après le premier sarclage. A l’occasion de chaque sarclage, il est indiqué d’arracher toutes les
mauvaises herbes .Pour les variétés volubiles, il faut faire le tuteurage au moment opportun (1ère feuille
trifoliolée) La quantité de semences par hectare es t de 60 à 70 kg.
2.10. La récolte et le battage
•
La récolte se réalise manuellement, en arrachant les plantes ou en les coupant avec la faucille.
L’époque de la récolte d’une culture semencière doit coïncider avec la maturation physiologique. Les
plantes récoltées sont ramassées en meules pour que le séchage continue pendant 2 à 3 jours.
Dans les zones de haute pluviométrie, les meules ne donnent pas satisfaction.
En ce qui concerne les haricots volubiles dont la récolte est échelonnée, il faut récolter les gousses au
fur et à mesure de la maturité, en plusieurs passages.
•
Le battage : se fait au fléau pour les petites quantités et à la batteuse, quand il s’agit de grosses
quantités. Pour éviter l’écrasement des graines, la rotation du batteur ne doit pas dépasser 300 tours par
minute.
2.11. Le séchage, le conditionnement et le stockage des semences
L’humidité des semences ne doit pas dépasser 9 % pendant le stockage de longue durée. C’est
pourquoi, les dépôts doivent être bien sec, bien aéré et pourvus d’installations de ventilation. Pour les
grandes unités de production de semences, les installations de séchage sont indispensables. Pendant le
stockage, une attention particulière doit être accordée à la protection contre les charançons. Pour cela,
la fumigation des locaux de stockage avec tétrachlorure à la dose de 1kg/m3 de semences est indiquée,
en prenant les précautions nécessaires contre les risques d’inflammation et d’intoxication. On peut
utiliser également l’Actelic (2%) à la raison de 200 grammes pour une tonne de semences. Avant la
commercialisation, il faut enrober les semences avec du Super Homai¨¨ (10 gr de produit pour 10 kg de
graines) ou de Almithio (mélange de Thirame 250 gr et de Lindane 200gr) en raison de 250 gr pour
100kg de graines qui protégera les jeunes plantules.
2.12. Le contrôle du champ semencier
Les champs de multiplication doivent être inspectés au moins deux fois : le premier contrôle doit être
effectué pendant la floraison date où sont observés les hors types et où ont fait l’épuration aussi des
plants malades. Le deuxième contrôle se fait à la maturation pâteuse pour éliminer les plantes qui sont
soit très précoces, soit trop tardives par rapport à la majorité des plantes. Il est parfois nécessaire de
procéder à des inspections supplémentaires en cas de problèmes particuliers. En pratique, l’inspecteur
s’assurera que les plants de haricot présente bien les caractéristiques de la variété, plus examinera les
bordures au champ afin de vérifier l’isolement. Il inspectera ensuite le champ dans son ensemble dans
son ensemble et fera une évaluation des plantes adventices présentes et de la situation phytosanitaire.
124
Lors de cette inspection, l’inspecteur examinera avec soin 150 plantes choisies au hasard en cinq
endroits différents du champ ( à raison de 30plantes par endroit) et établira séparément le nombre de
plants d’autres espèces cultivées dont les graines sont de dimension comparable. Le champ sera déclaré
impropre si l’on dénombré plus de trois hors types (2%), ou plus de trois plants d’autres espèces
cultivées (2%).
Le troisième contrôle se fera à la récolte où on prélève les échantillons :
• 1.000 gr pour l’échantillon à soumettre ;
• 400 gr pour la pureté ;
• 1.000 gr pour le dénombrement et les autres analyses.
Module 3 : Compétence en gestion : Elements d’étude de marché (Ir
Bahizire Chifizi/PFDiobass)
L’intervenant a d’abord tenu à préciser que la question relative au marché sera approfondie dans les
prochaines activités prévues dans le projet. En revanche, il était important que les participants puissent
avoir une certaine généralité sur les outils de gestion, notamment des outils de contrôle du marché.
Il s’agit des éléments suivants:
• la connaissance du client
• Fournir un bon service à son client
A cet effet, il faudra faire connaître les avantages et les désavantages de chaque variété exploitée en
être prêts d’en parler.
Egalement, il serait utile de défendre et expliquer comment vous produisez et la différence entre
vos variétés et les variétés dites locales.
• Les relations avec les concurrents
Il faut donc connaître ce que font les autres concurrents car certaines de leurs semences peuvent
ressembler au votre. Ainsi donc, des essais démonstratifs sur les endroits accessibles peuvent se faire
pour permettre à vos concurrents de voir et de suivre votre manière de travailler.
• Pour accroître votre demande
- Pour accroître donc la demande de vos produits, il faut :
- accroître votre marché en étendant le rayon de vente au delà de votre village, votre localité.
- Chercher des nouveaux clients en ne vendant pas uniquement aux paysans. Etendre le rayon
de vente aux écoles, marchés locaux…
- Changer fréquemment de variétés : cela peut se faire sur petite échelle puis sur grand terrain.
• Le conditionnement: Pour les haricots par exemple, utiliser des sacs translucides permettant
aux clients de voir ce qu’ils achètent. Il est possible de vendre également en des petits
emballages. Sur l’étiquette, on pourra lire :
- Le nom et adresse du producteur
- Nom de la variété
- La qualité de la semence
- Avertir si la semence est traitée
- La classe de la semence
• Le transport: Il faut s’assurer de livrer, si les moyens le permettent, les semences à vos clients et
protéger suffisamment les semences pendant leur déplacement
125
III. RESULTATS OBTENUS
1.
Niveau d’assimilation des thèmes et points de vue des formés
A l’issue de la formation, les participants formés ont manifestés leur contentement par rapport aux
thèmes retenus et développés.
Ils ont donc apprécié les méthodes de formation retenues, notamment la visite au terrain qui a marqué
la dernière journée de formation sur les 2 axes.
Quant aux formateurs, ils ont été captivés par l’intérêt des participants pendant les débats et les visites
au terrain. Ceci a poussé le formateur sur les fiches techniques, à formuler des propositions ci-après
pour les prochains ateliers :
prévoir une formation sur les fiches techniques des autres cultures
accroître la capacité organisationnelle des associations intéressées par la multiplication des
semences
faire précéder un atelier de formation sur la multiplication par des estimations chiffrées
sur la quantité de semences à semer en fonction de la quantité de semences recherchées à
la récolte.
Organiser un atelier à part sur les aspects liés au marché et à la commercialisation des
produits agricoles.
Négocier avec le Senasem de la possibilité de devenir multiplicateur des semences sur de
terrains de dimensions inférieurs aux dimensions exigées par les normes.
2.
Visite des champs de multiplication : réaction des agriculteurs
Les visites au champ se sont effectuées dans le but de renforcer l’observation et confronter les
techniques apprises aux réalités du terrain. La 1ère visite a concerné les paysans de l’axe Katana. Elle
s’est effectuée au champ de multiplication d’une superficie de 1 ha, de l’association ADEA/ Kavumu
qui est un site satellite. La 2eme visite s’est effectuée au champ de multiplication de haricot nain de
APACOV/ Burhale, pour les paysans de l’axe Walungu. Dans ou l’autre site, les paysans étaient
curieux de savoir ce qu’il faut faire pour être un multiplicateur agrée par le Senasem. En plus des
préoccupations techniques liées à la multiplication des semences, les formés ont voulu également
savoir la manière dont ils vont s’y prendre pour vendre à des prix compétitifs lors de la
commercialisation.
3.
Comment les paysans formés estiment intégrer les nouvelles connaissances dans
leurs pratiques agricoles
Les participants qui ont été pour la plupart des paysans, ont estimé qu’à la prochaine campagne, ils
seront prêts à se lancer dans les activités de multiplication des semences.
Cependant, les difficultés pour les associations d’avoir des grands terrains demeurent. A cet, les
organisations voudraient solliciter au Senasem, la possibilité de présenter des terrains de dimensions
réduites. Ces terrains mis ensemble pourraient constituer la superficie exigée par le Senasem.
Egalement, les paysans voudraient bénéficier d’un appui en intrant par CIALCA, à la phase de
multiplication des semences.
IV. COMMENTAIRES DU REDACTEUR DU RAPPORT
Les réalités socio-économiques du Sud- Kivu et la situation de conflit qui a prévalue dans cette partie
de la RDC, n’ont pas épargné le mouvement associatif dans les sites d’action. Les structures
travaillant avec la PF DIOBASS sur l’introduction des nouvelles semences, sont confrontées à des
problèmes d’ordre structurel et technique autour des activités agricoles.
La formation en multiplication avait justement pour but de doter les structures intéressées en
technique de multiplication ainsi qu’en éléments méthodologiques d’organisation et de gestion dans la
filière.
126
Ce faisant, la durée de 2 jours impartie à l’atelier sur la multiplication des semences a paru nettement
insuffisant. 4 jours auraient suffis pour aborder également d’autres modules sur d’autres cultures que le
haricot et le soja.
Les travaux ont suscités beaucoup d’espoirs .D’abord au niveau des capacités d’organisation des
acteurs en tant que synergie de multiplicateurs de semences et enfin en tant que cadre ouvrant les
perspectives de marché pour lutter contre la pauvreté endémique dans le milieu.
V. CONCLUSION
La formation sur la multiplication des semences a permis :
De regrouper les paysans membres des associations en un seul endroit. Ce cadre a favorisé des
échanges d’expériences fructueux entre les paysans des milieux différents
De comprendre les difficultés d’intégration des nouvelles semences, les difficultés organisationnelles
des groupes ainsi que les pistes de solutions retenues par les paysans eux-mêmes et d’autres part de
comprendre les difficultés rencontrées dans l’adoption des nouvelles semences, les difficultés
organisationnelles des groupes et les pistes de solutions adoptées par les paysans eux-mêmes ont été
également mise en surface.
C’était aussi selon l’approche Diobass un rendez-vous entre paysans et scientifique. Une occasion
offerte aux paysans d’échanger avec les spécialistes de la filière notamment le SENASEM et les
agronomes CIALCA
Des leçons ont été tirées de part et d’autres, nous citerons :
L’opportunité est offerte aux privés et aux organisations d’encadrement des paysans intéressées de
devenir multiplicateur des semences en respectant certaines directives ;
La semence qui ainsi produite est un produit commerçable soumis aux lois de l’offre et de la demande,
à la concurrence et aux fluctuations du marché. Le multiplicateur devra se préparer en conséquence et
ne pas s’improviser dans la filière.
Ceci offre également une porte aux autres activités Ciat qui peuvent être menées dans le milieu d’être
favorablement accueillies.
La PF DIOBASS a des perspectives suivantes pour l’activité de multiplication des semences dans les
sites d’action et satellites: Faire des associations intéressées par la production, la distribution et le
stockage des semences, des structures dotées d’outils de gestion et de suivi pour faire de cette activité
un gagne –pain du paysan; Développer une dynamique de business autour de l’activité, en vue de
rehausser le revenu du paysan et faciliter la circulation de l’argent dans le milieu rural; Ouvrir les
paysans multiplicateurs sur le marché régional et sous régional en livrant leur production à des prix
compétitifs.
127
ANNEXE 1
Liste des participants :
1. Axe Katana
N°
01
02
03
04
05
06
07
09
10
11
12
13
14
16
17
18
19
20
21
22
23
Noms
Matunguru Mirindi
Dieudonné Byende Kabira
M’Kaseremba Marceline
Bahati cidabagizi Jean
Mweze kabale
Skolastika
Eugénie M’mukonda
Mwila M’nyamukamwa
Mirindi Kabujege
Eugenie M’moka
M’mlangota
M’rungora
Mulume Bijaci Berlin
Alphonse Bisusa
Olinabanji Dieudonné
Sangara Désiré
Bahati Rugina
Cizungu Luhirika
Fidèle Mupenda
Ir Gaetan Mazombo
Ir Adrien Bahizire chifizi
Organisation
Maendeleo
Maendeleo
Maendeleo
Maendeleo
Tuungane
Tuungane
Tuungane
Rhubehaguma
Tuungane
Rhubehaguma
Rhubehaguma
Rhusimane
ADEPB
ADEA
Marafiki wa mazingira
Rhusimane
Rhusimane
Rhubehaguma
Rhubehaguma
Senasem
PF Diobass
Site d’action
Kabamba
Kabamba
Kabamba
Kabamba
Kabamba
Kabamba
Kabamba
Luhihi
Kabamba
Luhihi
Luhihi
Luhihi
Kabamba
Kavumu
Katana
Luhihi
Luhihi
Luhihi
Luhihi
Bukavu
Bukavu
24
25
Ir Jean Marie SANGINGA
Augustin Chyoka M.
Ciat/Cialca
PF Diobass
Bukavu
Bukavu
26
Dr Katunga Musale
Cialca
Bukavu
Tél. et e-mail
0994262306
+2439942236060
+243997796109
+243997740827
+243997742028
+243 -997252039
[email protected]
+ 243 998666101
+243 813176372
[email protected]
2. Axe Walungu
N°
Noms
Organisation
Site d’action
Tél.
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
Jonas Balola Luhene
Jean Maheshe Chyoka
Ngwasi visita
Dieudonné Kusinza
Ursula Camunani
Bahati Venant
Geneviève M’mushaba
Odette Lushunju
Fabien kolondwa
Vincent Kiriza
Masimango Claude
Eugenie Basimara
Vumilia lorensi
N’simire M’birindwa
Kalyo kahirwe D.
Buhendwa Shaba Deux
Kaboza Mutabesha
Désiré Birindwa
Ir Jean-marie Sanginga
Ir Byakombe mazambi
Ir Adrien Bahizire chifizi
Alemalu
Cinamula
Alemalu
Alemalu
Cinamula
Apacov
Abagwasinye
Cinamula
Rhucihangane
Bololoke
Abagwasinye
Abagwasinye
Abagwasinye
Apacov
Apacov
Alemalu
Alemalu
Apacov
Ciat/Cialca
Senasem/SK
PF Diobass
Lurhala
Lurhala
Lurhala
Lurhala
Lurhala
Burhale
Burhale
lurhala
Burhale
Burhale
Burhale
Burhale
Burhale
Burhale
Burhale
Lurhala
Lurhala
Burhale
Bukavu
Bukavu
Bukavu
+243810258908
128
+243994066752
+243997788911
+243994304755
+243998666101
+243998669463
+243997252039
A
AN
NN
NE
EX
X 88:: FFIIC
CH
HE
E FFO
OR
RD
DA
AT
TA
AC
CO
OL
LL
LE
EC
CT
TIIO
ON
N IIN
NL
LE
EG
GU
UM
ME
E
M
MU
UL
LT
TIIPPL
LIIC
CA
AT
TIIO
ON
N FFIIE
EL
LD
DSS
Fiche de caractérisation et suivi des champs de multiplication
Association :
Numéro de champ :
(chaque association doit compter le nombre de champs de multiplication
et numérotez chaque champ)
Décrivez ou se trouve le champ :
Position du champ dans le paysage :
(plateau, en pente, ou dans la vallée)
Nom local de type de sol :
Cultures cultivés la saison passée :
(spécifiez les cultures, ou jachère)
Superficie du champ :
(en mètres carrés, ou largeur et longueur en mètres)
Nom du propriétaire du champ :
(notez le nom de l’individu ou de l’association pour les champs
communs)
Appréciation de la fertilité du champ (par le propriétaire) :
(pauvre, moyenne ou bonne)
Espèce :
(haricot nain, haricot volubile, soja, niébé ou arachide)
Nom de la variété :
Quantité de semence reçue :
(s’applique seulement si la multiplication est géré par un individu) (en kg)
Quantité de semence semée :
(en kg)
Entrants appliqués ?
Si oui, spécifiez le type d’entrant (fumier, composte, résidus de culture,
engrais verts, engrais chimiques) et la source.
Si oui, spécifiez la dose appliquée (en kg, ou en paniers).
Date de semis :
Spécifiez les écartements :
(distances entre les lignes, et entre plantes dans la ligne ; si ne pas
semée en lignes, comptez le nombre de plants dans 3 cadres de 1×1m)
(les écartements recommandés sont 75× 5 cm pour le soja et l’arachide,
50× 20 cm pour l’haricot volubile, et 40×10 cm pour l’haricot nain)
Dates de sarclage :
(spécifiez toutes les dates de sarclage)
Nombre de plantes arrachées pour le démariage :
Nombre de plantes arrachées de couleur jaune :
(à ajouter à différents moments pendant la saison)
Nombre de plantes arrachées avec signes de maladie :
(à ajouter à différents moments pendant la saison)
Date de récolte :
Rendement obtenu :
(en kg)
Quantité remise à l’association :
(s’applique seulement si la multiplication est géré par un individu) (en kg)
129
A
AN
NN
NE
EX
X 99:: E
ER
RO
O--11:: E
ERRO
OSSIIO
ON
NC
CO
ON
NT
TR
RO
OL
L IIN
N SSU
UD
D--K
KIIVVU
U
Objectives:
(i) to investigate the relative importance of mechanically formed embankments (with/without
hedgerows planted for stabilization) vs. surface management (traditional tillage / zero tillage) for
reducing soil erosion on sloping land in S-Kivu.
(ii) to evaluate Calliandra planted as hedgerows to reduce soil erosion (in presence or not of
mechanically formed embankments) and determine Calliandra biomass production;
(iii) to assess agronomic performance of a cropping system planted on the terraces, and investigate
effects of physical embankments, Calliandra hedgerows and surface management on crop production
and erosion.
Sites:
Replicated trial on-station (controlled conditions) on a field with a constant (representative) slope.
Treatment structure
- Main plots: physical embankments (combination of “fanya juu” and “fanya chini”) vs. no physical
embankments;
- Sub-plots: soil tillage (traditional tillage or zero tillage) × contour hedgerows (control or Calliandra
hedgerow);
- 3 replicates (blocks), positioned down-hill.
rep 3
TT-Ca
rep 2
TT-Ca
ZT-0
ZT-Ca
rep 1
ZT-Ca
ZT-0
grass strip
TT-0
ZT-0
ZT-Ca
TT-0
TT-0
TT-0
TT-Ca
TT-0
TT-Ca
ZT-Ca
ZT-0
ZT-0
ZT-0
ZT-Ca
ZT-Ca
TT-Ca
TT-0
grass strip
TT-Ca
grass strip
slope direction
fanya juu
no physical embankment
Figure
TT traditional tillage
0
ZT zero tillage
Ca Calliandra
control
ERO-1 trial layout
Three grass strips (length = 4 m) are included. The outer 1 m (on both sides) serves as a grassed
waterway to allow excess water to run down. The middle 2 m serves as a reference area where erosion
is zero (or very minimal) and is sampled at later stages to determine erosion using isotopic methods.
130
1m
(drainage canal)
2m
(reference strip with minimal erosion)
1m
(drainage canal)
Calliandra is included as hedgerow species to improve soil fertility. At regular times, plants are pruned
and the residues are applied to crop on the terraces. Other interesting species are: Tithonia, sugarcane,
Desmodium+Paspalum, Penisetum, Setaria, Trypsacum, Panicum, Brachiaria, Tephrosia, Vetiver…
Performance and biomass production of these species is tested in a 2nd trial (ERO-2).
Trial management
Every plot consists of 3 hedgerows (strips) and 3 alleys (terraces). The first strip is constructed as a
“fanya chini” (earth moved down the slope) to avoid run-off into the plots down-slope. Water coming
down the slope is first trapped in a trench and looses erosive power. The trench is sown with grass
and needs to be well maintained. Sediment accumulating in the trench needs to be removed regularly,
so that water is conducted laterally towards the grassed channels going down the slope. The
embankment in the first strip is physically enforced by wooden boards.
All 3 alleys are cropped (with a short duration bean cv.). The first alley serves as a buffer and no data
are recorded. Crop performance is assessed in the 2nd alley; biomass production in the hedges is
assessed in the 3rd hedgerow. A 3rd alley is included to avoid recording data from strips that borders a
physically enforced “fanya chini” strip. In the treatments where Calliandra is used as a hedgerow
species, the hedges are regularly pruned and the residues are exported from the trial.
Treatment 1: with fanyas, with hedgerows
strip 1 (fanya chini with physical stabilization)
terrace 1 (cropped)
=
A
(d at
ep le
en a
ds st
on 2
s lo m
pe
)
H
=
strip 2 (fanya juu)
1.
5m
terrace 2 (cropped)
(crop yield assessment)
(fanya juu) strip 3
(biomass assessment)
terrace 3 (cropped)
(fanya chini with physical stabilization) strip 4
W = 8m
131
Treatment 2: with fanyas, without hedgerows
terrace 1 (cropped)
=
A
(d at
ep le
en a
ds st
on 2
s lo m
pe
)
H
=
strip 2 (fanya juu)
1.
5m
terrace 2 (cropped)
(fanya juu) strip 3
terrace 3 (cropped)
W = 8m
Treatment 3: without fanyas, with hedgerows
terrace 1 (cropped)
=
a
A
+
(d t le
ep a
en s
ds t 2
on .7
s lo 5
pe
)
H/
2
strip 2
m
terrace 2 (cropped)
strip 3
(biomass assessment)
terrace 3 (cropped)
W = 8m
132
Treatment 4: without fanyas, without hedgerows
terrace 1 (cropped)
=
at
A
+
H/
2
(d le
ep a
en s
ds t 2
on .7
s lo 5
pe
)
strip 2
m
terrace 2 (cropped)
strip 3
terrace 3 (cropped)
W = 8m
Figure
Individual plot lay-out.
In treatments with physical embankments, the embankments between the 1st and 2nd, and between the
2nd and 3rd alley are constructed using “fanya juu” (earth moved slope upwards). They have the
advantage not to increase the slope of the land in the alleys (unlike “fanya chini”) and can gradually
develop into bench terraces. However, the risk of breakage during heavy storms is larger than for
“fanya chini” embankments. The 4th strip is again constructed as a “fanya chini” embankment and
coincides with the 1st hedgerow of the next replicate plot down the slope.
The width of each plot is 6 m (following the slope contours). The length of the plot down the slope
depends on the slope and is based on a 1.6 m change in elevation. The length (down the slope) of the
alley should be at least 2 m; the length of the hedgerows (down the slope) is 1.5 m. The total area
reserved for the trial should thus be 64 m wide (following the slope contours) by 40 – 60 m long
(down the slope). ERO-1 is intended as a long-term trial. It needs to be ensured that the site is
reserved for at least 5 years.
133
A=
D/s
in(α
)
-H
H (1.5
d
0.6m
m)
0.6m
D
1.6m
0.6m
L = D / tg(α)
Figure
α
0.6m
Calculation of distance between two hedgerows as affected by slope ( , [%]) and change in
elevation (D, [m]).
The Calliandra hedgerows are established as two lines at 50 cm between lines and 50 cm between
plants as presented in the Figure below.
Figure
0.5 m
0.5 m
Calliandra hedgerow spacing
134
ERO-1: Soybean planting protocol
Land preparation:
- For treatments with traditional tillage, the land on the terraces is cleared and tilled using a hoe.
- For treatments with zero tillage, the land is simply cleared using a cutlass (except for the 1st .
Establishment of soybean planting lines and soil sampling:
The number of soybean lines in each terrace plot will vary because of (i) the size of the plot, (ii)
presence of fanyas and/or Calliandra hedgerows. Planting lines are not necessarily straight, because of
the irregular shape of the terrace plot and should more or less follow the contour line. Distance
between lines can vary between 50 and 75 cm, but should as much as possible be kept at 75 cm.
Examples of how soybean lines are delineated in the different treatments are shown below.
1. treatment with fanya juu and Calliandra hedgerows:
ditch
contour line
soybean planting row 5
50-75cm
50-75cm
50-75cm
50-75cm
50-75cm
Calliandra hedgerow
50cm
fanya juu embankment
50cm
100cm
Calliandra hedgerow
contour line
60cm
ditch
6m
2. treatment with fanya juu and without Calliandra hedgerows:
ditch
contour line
50-75cm
50-75cm
50-75cm
50-75cm
25cm
fanya juu embankment
100cm
grass fallow
contour line
60cm
ditch
6m
3. treatment without fanya juu and with Calliandra hedgerows:
135
Calliandra hedgerow
contour line
50-75cm
Calliandra hedgerow
50-75cm
50-75cm
50-75cm
50-75cm
50-75cm
50-75cm
50cm
Calliandra hedgerow
25cm
25cm
contour line
Calliandra hedgerow
100cm
6m
4. treatment without fanya juu and without Calliandra hedgerows:
contour line
50-75cm
50-75cm
50-75cm
50-75cm
50-75cm
50-75cm
50-75cm
25cm
contour line
6m
Soybean lines in each plot are numbered starting from the lowest line in the plot (closest to the lower
contour). Each plot should minimally have 3 soybean lines; control plots (without fanyas and without
Calliandra hedgerows) can have up to 8 soybean lines. The soybean line is established by digging a
trench (about 10 cm deep). Seeds are planted at 5 cm within-line distance. Variety PEKA-6 is used.
Installation of erosion pins:
Erosion pins are installed only in the measurement plots (i.e. plot numbers 5, 6, 7, 8, 17, 18, 19, 20, 29,
30, 31, 32, 41, 42, 43, 44, 53, 54, 55, 56, 65, 66, 67, 68). Per plot, 6 erosion pins are installed: 3 at the
upper end of the plot and 3 at the lower end of the plot (see figure below).
136
1.5m
1.5m
1.5m
1.5m
6m
In treatments with Calliandra hedgerows, erosion pins are installed 25 cm away from the Calliandra
planting line. In treatments with fanya juus, the upper erosion pins are installed 25 cm down the slope
from the ditch; the lower erosion pin is installed 25 cm away from the Calliandra hedgerow in
treatments with planted embankments and 125 cm away from the lower ditch in treatments with bare
embankments. In control treatments (without fanya juu and without Calliandra hedgerows), erosion
pins are installed on the contour line. This is schematically presented in the figures below.
1. with fanya juu and with Calliandra hedgerows:
3. without fanya juu and with Calliandra hedgerows:
25cm
25cm
25cm
25cm
2. with fanya juu and without Calliandra hedgerows:
4. without fanya juu and without Calliandra
hedgerows:
125cm
25cm
contour line
contour line
137
Trial management and observations:
-Soil profile description.
-Soil sampling. Soil samples are taken separately in between each of the soybean line, but only in the
measurement plots (i.e. plot numbers 5, 6, 7, 8, 17, 18, 19, 20, 29, 30, 31, 32, 41, 42, 43, 44, 53, 54, 55,
56, 65, 66, 67, 68). In the middle between each soybean line, a composite soil sample (4 cores per line)
is taken, using a mass balance auger. Two soil depths are sampled: 0-7.5 cm and 7.5-15 cm. Samples
must be well labelled, indicating date, plot number, soybean line numbers in between which the sample
was taken, the soil depth, and the number of cores (this should normally be 4). Samples are air-dried,
weighed and stored.
-Precipitation is measured on a daily basis, using a rain gauge.
-Weeding (when necessary).
-Clearing and reparation of fanya chini ditches is performed when necessary (not of fanya juu
ditches!!).
-At 50% flowering and at 50% podding, soybean aboveground biomass is sampled, separately for every
row (only in the measurement plots!). In every row, the number of plants is counted in the “net line”,
i.e. the middle 5 m without the outer 50cm. Subsequently, 5 plants are cut (randomly within the row,
but not from the outer 0.5 m at both ends). The plants are dried, weighed, ground and stored pending
on analysis.
-During the season, the distance between each line and the upper and lower boundary of the soybeangrown area is assessed at 1.5, 3 and 4.5 m (see file with examples).
-Harvest of soybean is conducted separately for every row in the measurement plots (only the “net
row”, i.e. not the outer 0.5 m at both ends). In all other plots, the entire plot is harvested at once
(excluding the outer 0.5 m at both ends of the rows).
-Moisture profile measurements. Moisture tubes are installed up to a depth of ***cm, in treatments
without fanya juu embankments. The first tube is installed at ***cm from the contour line or the
upper Calliandra line downslope. The second tube is installed at ***cm from the contour line or the
upper Calliandra line downslope. Weekly, moisture profiles are assessed using a diviner probe.
-Slope assessment at planting (after land preparation) and at harvest of every season, using a triangle.
-Erosion pin soil levels at planting and harvest of every season.
-Initially, Calliandra hedges are allowed to obtain a height of 1.5 m; subsequently, hedges are pruned to
a height of 75 cm. From then onwards, Calliandra hedgerows are pruned regularly and not allowed to
be taller than 1 m. At each pruning event, residues are dried, separated in wood and leaves, cut and
subsampled for analysis. The wood is exported from the trial; the leaf residues are surface-applied
to the terrace located upslope from the hedge [or exported as livestock feed is the most likely
entry point for adoption?].
-At the end of the trial, at harvest of the final season, Calliandra hedges are removed and root depth
and root distribution of Calliandra and soybean are measured.
138
ERO-1: Sediment traps protocol
A number of modifications need to be made to allow installing sediment traps for assessing soil loss in
the various treatments.
Plot separations:
Plots need to be separated by making small ridges between the plots to prevent run-off from a given
plot to enter neighbouring plots. The outer plots similarly need to be separated from the grass buffer
strips. Ridges need to be about 50 cm wide and 30 cm high and need to be installed without affecting
the soil surface in the plots. Therefore, these need to be constructed using topsoil (0-20cm) dug out
from outside the trial. Ridges are planted with a grass, and maintained and trimmed regularly.
W = 7.5m
7.5m
ridge separating treatments
(H=30cm, W=50cm)
UPPER BORDER
PLOTS
grass strip
MEASUREMENT
PLOTS
grass strip
Sediment traps:
Sediment traps will be installed in the
ditch of the fanya chini in each replicate
treatment (24 in total). These traps do
not need to be installed in the top ditch
protecting the first replicate from
erosion occurring upslope, and do not
need to be installed in the ditches of the
ERO-2 trial. Walls are constructed
inside the ditch to separate treatments.
These walls are 0.5m wide (occupying
25cm at both outer ends of each
treatment) and reinforced using
wooden boards. The figures on the
right indicates the modifications to be
made (example showing 4 treatments);
the figure below shows the entire plot
lay-out with the position and
numbering of the sediment traps.
0.5m
LOWER BORDER
PLOTS
reinforced wall separating sediment traps
(D=90cm, W=50cm, L=60cm)
sediment trap
(D=60cm, W=7.5m, L=60cm)
139
ZT-Ca
16
TT-Ca
18
19
20
TT-0
ZT-0
ZT-Ca
TT-Ca
11
12
21
22
23
24
rep 3
ZT-0
17
rep 2
TT-0
8
grass strip
ZT-Ca
grass strip
7
rep 1
TT-0
15
ZT-Ca
TT-Ca
10
14
TT-0
TT-0
9
13
ZT-0
ZT-0
6
4
TT-Ca
ZT-Ca
grass strip
slope direction
5
TT-Ca
ZT-0
3
ZT-0
TT-0
2
ZT-Ca
TT-Ca
1
A plastic sheet is placed inside the ditch to catch the run-off. The sheet is perforated at the bottom
(holes of ~0.5cm diameter) to allow water to drain. The front and back wall need to be covered by the
sheet in such a way that run-off can freely flow into the trap.
Observations:
Sediment accumulating in the traps needs to be determined every two weeks. At all times, it needs to
be prevented that large amounts of sediment accumulate and restrain water infiltration; in such cases,
sediment needs to be collected more frequently.
At least every two weeks, the total amount of sediment in each of the traps is scooped out and weighed
(the date of sampling is noted in the field book). Subsequently the sediment is thoroughly mixed and a
subsample of exactly 1 kg (use an accurate balance) is collected immediately. The sediment is then airdried and weighed again. After weighing, a subsample of about 100 grams is stored for analysis. Each
sample needs to be carefully labelled with the treatment and replicate, number of the sediment trap
(see picture above), and the date of sediment collection.
140
A
AN
NN
NE
EX
X 1100:: E
ER
RO
O--22:: C
CO
OM
MPPA
AR
RIISSO
ON
NO
OFF V
VA
AR
RIIO
OU
USS FFO
OR
RA
AG
GE
E
SSPPE
EC
CIIE
ESS FFO
OR
RE
ER
RO
OSSIIO
ON
NC
CO
ON
NT
TR
RO
OL
L
Objectives:
(i) to compare biomass production and biomass quality of various forage species when grown as
hedges on sloping land;
(ii) to assess effectiveness of various forage species to stabilize the soil (assessment of soil erosion and
rooting depth and distribution).
Sites:
6 sites: (i) down slope from the ERO-1 trial, (ii) degraded slope at INIBAP site in Cijingiri and (iii) 4
sites with various associations in the action sites of the Sud-Kivu mandate area.
Treatment structure:
- 8 treatments: (i) control, (ii) Calliandra at 50 cm planting density, (iii) Calliandra at 25 cm density,
(iv) Leucena diversifolia, (v) Penisetum, (vi) Brachiaria, (vii) Setaria and (viii) Tithonia;
- 1 site = 1 replication (no replications on site).
viii
ii
ERO-2 possible trial layouts
Species are planted in hedges of 6 m long, as 2 parallel lines at densities recommended by local
extension services. The position of the hedges is randomized. A small initial dose of NPK fertilizer can
be applied to aid establishment (rate?), particularly on degraded slopes. Upslope from every hedge, a
fanya chini is installed to avoid treatments higher up the slope to affect the hedge down the slope.
Distances between each fanya chini and hedge are based on a vertical interval of 1.6 m. The terraces
between the hedges and fanya chinis should not be cleared and kept fallow (natural vegetation). Fanya
chinis canals are 60 cm deep and 60 cm wide and planted with grass. These canals need to be well
maintained; sediment accumulating in the trenches needs to be removed regularly. Fanya chini
embankments are reinforced by wooden boards. At the ERO-1 site the fanya chini coincides with the
final fanya chini of the ERO-1 trial.
141
grass strip
vi
iii
v
viii
grass strip
vii
i
iv
ii
slope
grass strip
vii
i
iii
v
Figure
iv
vi
grass strip
slope direction
fanya chini
grass strip
One of the following two layouts can be used, depending on the land area available (36 x 16 m on the
left, or 66 x 6 m on the right):
fanya chini
fanya chini
fanya chini
A=
D/s
0.6m
Figure
1.6m
d
0.5m
0.6m
in(α
)
D
α
L = D / tg(α)
Calculation of distance between the hedgerow and the fanya chini slope upward as affected by
slope ( , [%]) and change in elevation (D, [m]).
Trial management and observations:
-Precipitation assessed on a daily basis.
-Initial soil description and soil fertility evaluation.
-Soil accumulation and soil loss assessed by erosion pins installed before and behind the hedge.
-Isotopic measurements for assessment of soil erosion
-At regular times, plants are pruned, biomass production is determined and the residues are exported
(not applied to the above terrace). The quality of the biomass as a green manure or forage is assessed
(C:N ratio, % lignin and polyphenols, protein and carbohydrate contents, mineral nutrient contents).
As this trial is established at different sites with different slopes and/or soil fertility, the relation
between biomass production/quality and soil physicochemical measures can be studied.
-At the end of the trial (x years), hedges are removed and root depth and root distribution is measured.
142
A
AN
NN
NE
EX
X 1111:: E
EVVAALLU
UA
AT
TIIO
ON
ND
DE
ESS FFO
OU
UR
RR
RA
AG
GE
ESS E
EN
NM
MIIL
LIIE
EU
U
PPA
AY
YSSA
AN
N ((E
ER
RO
O--22))
Données générales
Nom de l’Association …………………………… Village …………………………………………….
Nombre de femmes présentes …………….… Nombre d’hommes présents………………
Sexe du groupe interviewé ……………………
Date de l’évaluation ………………………… Noms et sexes des facilitateurs ………………………
Première étape : Discussion en groupe
FICHE 1: Quelles sont les caractéristiques que vous voudriez trouver dans un bon fourrage? Ou
Qu’est ce que vous considérez quand vous voulez sélectionner un bon fourrage?
FICHE 2 : Quelles sont les CINQ CARACTERISTIQUES les plus importantes que vous allez
utiliser pour évaluer les différentes variétés ?
Criteres d’évaluation
Ordre d’importance
Proposez d’ajouter les critères suivants si ils n’ont pas été mentionné, en expliquant que ceci intéresse
la recherche: 1. production de biomasse ; 2. efficacité de lutter l’érosion.
143
Deuxième étape : Visite au champ
Gendre de groupe : Hommes ou Femmes …………………………………… Nombre de participants : …………………………………………
Liste des fourrages installés (à remplir basé sur les observations au champ par l’enquêteur):
nom du fourrage
fourrage compris dans l’essai ?
signes de non-adaptation (maladies,
(0 = non, 1 = oui)
pauvre biomasse,…) ? (0 = non, 1 =
oui)
1
Brachiaria brizantha
2
Brachiaria decumbens
3
Brachiaria ibrido
4
Brachiaria ruziziensis
5
6
Calliandra callothyrsus
(à 0.25m)
Calliandra callothyrsus
(à 0.5m)
7
Leucaena diversifolia
8
Penisetum purpureum
9
Setaria sphacelata
10
Tithonia diversifolia
11
Tripsacum laxum
144
vol ou mangé par des betes au
champ ?
(0 = non, 1 = oui)
FICHE 3. EVALUATION OUVERTE DES FOURRAGES
Production de biomasse / efficacité contre l’érosion : 0 = très pauvre; 1 = pauvre; 3 = moyen; 4 = bon; 5 = très bon
No
Fourrage
Aspects positifs (+)
Nombr Aspects négatifs (-)
e
rubans
Brachiaria brizantha
Brachiaria decumbens
Brachiaria ibrido
Brachiaria ruziziensis
Calliandra
callothyrsus (à 0.25m)
Calliandra
callothyrsus (à 0.5m)
Leucaena diversifolia
Penisetum purpureum
Setaria sphacelata
Tithonia diversifolia
Tripsacum laxum
145
Nombre production de
rubans
biomasse
efficacité
contre érosion
FICHE 4. ANALYSE PREFERENTIELLE DES FOURRAGES (1)
Combien de personnes classent un fourrage comme 1er, 2e, 3e, 4e, 5e ?
Fourrage
1er
2e
3e
4e
5e
Ordre
En cas de divergence, indiquez les raisons :
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
FICHE 5. ANALYSE PREFERENTIELLE DES FOURRAGES (2)
Caractéristiques
Fourrages
Producti
Efficacit
on
Critère 1 Critère 2 Critère 3 Critère 4
é contre Total
biomass
érosion
e
146
Rang
A
AN
NN
NE
EX
X 1122:: C
CA
ASS--11:: IIM
MPPR
RO
OV
VE
ED
DC
CA
ASSSSA
AV
VA
AA
AG
GR
RO
ON
NO
OM
MY
Y
Objectives:
(i) to evaluate the contributions of various legumes to cassava production;
(ii) to evaluate the use of alternative agronomic practices (tilled vs no-till; varying cassava planting
density; use of inputs) on cassava and legume production;
(iii) to evaluate the potential to include a second legume crop in a cassava field.
Treatment structure
DESCRIPTION OF THE FACTORS
The following factors will be considered (Table 1):
- Factor ‘Cassava variety’: One treatment will be cropped using local cassava while all the other
plots will have CMD-resistant varieties that are known to still have full resistance to CMD;
priority should be given to varieties that delay the formation of vigorous branching and leaf
production.
- Factor ‘Legume species/variety’: Dual purpose soybean will replace the commonly grown
groundnut in specific treatments.
- Factor ‘Planting density’: Cassava will be planted at a distance of 1 x 1m or 2 x 0.5 m (2 m
between the lines, 0.5 m between plants within a line).
- Factor ‘Planting time’: Since the best results are usually obtained when both the legume and
the cassava are planted at the same time (maximally one week difference), this factor has only one
level: simultaneous planting.
- Factor ‘Tillage’: Some plots will be conventionally tilled while on other plots, cassava will be
grown without tillage (‘au plat’).
- Factor ‘Second legume’: The treatments with a second legume will have a bush bean variety
planted at the start of the second season.
- Factor ‘Inputs used’: Some treatments will have no inputs while other treatments will be
treated with NPK fertilizer at 2 bags per hectare, with the fertilizer equally distributed over the
cassava and the legumes and applied in the planting hole.
Trt
1
2
4
6
7
8
5
9
10
Table 1: Detailed description of the various proposed treatments. Note that improved
components are written in bold.
Cassava
Tillage
Planting
Second
Cassava
Legume
legume
species/variety
planting
time
variety
density
local
Gnut/improved
1x1m
Tilled
Simultaneous None
CMD-resist. Gnut/improved
1x1m
Tilled
Simultaneous None
2x0.5m
Tilled
Simultaneous None
2x0.5m
No tillage Simultaneous None
2x0.5m
Tilled
Simultaneous Cl beans
2x0.5m
Tilled
Simultaneous None
CMD-resist. Soybean/improved
2x0.5m
Tilled
Simultaneous None
CMD-resist. Cowpea/improved
2x0.5m
Tilled
Simultaneous None
CMD-resist. Bush-beans/improved 2x0.5m
Tilled
Simultaneous None
EXPERIMENTAL DESIGN
- The design will be a completely randomized block design with 3 replicates per site.
- The trial will be laid out in 2 of the 4 Action Sites.
- One site should be place on a ‘savanna’ soil and one on a ‘forest soil’
- Fields with homogeneous history of management and a minimal slope should be chosen that
are representative for the farmer environment (not having been under a long fallow period; not
strongly eroded; not too many trees in the plots; etc).
147
Inputs
used
None
None
None
None
None
Yes
None
None
None
Trial implementation and management
PLOT SIZE AND ARRANGEMENT
- Plot size will be 10 by 10 m (10 lines of cassava in the 1x1 m arrangement and 5 lines of cassava
in the 2x0.5 m arrangement) (Figure 1).
- The legumes will be planted at distances of 40 cm between 2 lines. The distances between plants
within a legume line will be: 10 cm for soybean, 20 cm for groundnut; 20 cm for cowpea; 10 cm
for bush beans and 50 cm for the second season climbing beans.
Cassava-groundnut intercrop [1 x 1 m]
0.3m
0.4m
0.3m
Cassava hill
Legume line
1m
1m
Cassava-groundnut intercrop [2 x 0.5 m]
0.4m
0.4m
0.4m
0.4m
0.4m
0.5m
2m
Figure 1: Sketch of a cassava-legume intercrop with 1x1 m and 2x0.5 m cassava spacing.
CONTROL MAIZE PLOT
- Per replicate, one plot needs to be included with maize as a reference crop for the BNF
measurements.
MANAGEMENT
- The plots will be researcher-managed.
- All operations will be implemented following proper agronomic principles (e.g., timely
weeding).
Observations
- Initial soil description and soil fertility evaluation (at least 8 cores per plot at 0-20 cm using a
‘W’ desing; bulked; air-dried; stored pending shipment and analysis).
- Labour (chronometer time to perform certain operations with a special emphasis on planting,
weeding, and harvesting).
- Cassava biomass at harvest (roots, stems, etc) from the net plot (8 middle lines of cassava in the
1x1 m spacing or 3 middle lines in the 2x0.5 m spacing).
- Legume biomass at mid-podding (biomass sampling, also for BNF measurements).
- Legume grain yield at harvest.
148
A
AN
NN
NE
EX
X 1133:: C
CA
ASS--22:: IIM
MPPR
RO
OV
VE
ED
DC
CA
ASSSSA
AV
VA
AA
AG
GR
RO
ON
NO
OM
MY
Y
Objectives:
(i) to evaluate the contributions of various legumes to cassava production;
(ii) to evaluate the use of alternative agronomic practices (varying cassava planting density; use of
additional NPK input) on cassava and legume production;
(iii) to evaluate the potential to include a second legume crop in a cassava field.
Sites
The trial will be established with the associations in Kabamba (TUUNGANE and
MAENDELEO), at 2 replicates per association. If possible, neighbouring associations can be
involved and install additional replicates (maximally 8 fields in total). Sites should be chosen
carefully and be as homogeneous as possible, have the same position in the landscape (not on
sloping land), a similar cropping & management history and a comparable fertility status.
Treatment structure
- Factor ‘Cassava variety’: two treatments will be cropped using local cassava while all the other
plots will have CMD-resistant varieties that are known to still have full resistance to CMD;
priority should be given to varieties that delay the formation of vigorous branching and leaf
production.
- Factor ‘Legume species/variety’: In the 1st two treatments, the traditional cassava and bush
bean varieties are used. In the other treatments, improved varieties (e.g. Sawasawa for cassava
and ZKA93-10m/95 or Marungi for bush bean) are used. Dual purpose soybean (SB24 or SB25)
or groundnut will replace the commonly grown bush beans in specific treatments.
- Factor ‘Planting density’: In the 1st treatment, farmers use their traditional spacing
(broadcasted seeds, random placement of cuttings). In all the other treatments, exact spacing is
used, with the beans planted in lines. Cassava will be planted at a distance of 1x1m or 2x0.5 m (2
m between lines, 0.5 m between plants within a line).
- Factor ‘Second legume’: The treatments with climbing beans as a second legume, planted at
the start of the second season and climbing the maturing cassava crop.
- Factor ‘Inputs used’: All treatments receive a blanket of FYM at 5 tonnes fresh matter ha-1, as
farmers commonly apply organic inputs in their cassava systems. A treatment is included where
additional NPK is applied at 2 bags per hectare, with the fertilizer equally distributed over the
cassava and the legumes and applied in the planting hole.
Table 1: Detailed description of the various treatments.
Tr
Cassava
Legume
Cassava
variety
species/variety
planting
density
1 Local
Bean/local
Traditional
2 Local
Bean/local
1x1m
3 CMD-resist. Bean/improved
1x1m
2x0.5m
2x0.5m
2x0.5m
7 CMD-resist. groundnut/improved
2x0.5m
8 CMD-resist. soybean/improved
2x0.5m
149
Second legume
Inputs used
None
None
None
None
Climbing beans
None
None
None
FYM
FYM
FYM
FYM
FYM
FYM+NPK
FYM
FYM
- Plot size will be 10 by 6 m (10 lines of cassava in the 1x1m arrangement and 5 lines of cassava
in the 2x0.5m arrangement) (Figure 1).
- The legumes will be planted at distances of 40 cm between 2 lines. The distances between plants
within a legume line will be: 10 cm for soybean, 20 cm for groundnut, 10 cm for bush beans and
50 cm for the second season climbing beans.
Cassava-groundnut intercrop [1 x 1 m]
0.3m
0.4m
0.3m
Cassava hill
Legume line
1m
1m
Cassava-groundnut intercrop [2 x 0.5 m]
0.4m
0.4m
0.4m
0.4m
0.4m
0.5m
2m
Figure 1: Sketch of a cassava-legume intercrop with 1x1 m and 2x0.5 m cassava spacing
(see additional details in protocol CAS-2 annex).
FYM AND NPK APPLICATION
FYM will be broadcasted and incorporated at 5 tonnes ha-1 at trial establishment (per plot of
10x6 m, 30 kg of FYM is incorporated).
NPK will be applied at two bags per hectare, more or less equally distributed over the cassava
and legume species (i.e. one bag for each species).
In each planting hole of cassava, 2 bottle caps of NPK are applied (2 × ~3.65 g per planting
hole). Per 1.5 m of legume, 1 bottle cap (~3.65 g) is applied. As such, per planting line of 6 m, 4
bottle caps are applied, and as much as possible concentrated at the seed planting spots.
CONTROL MAIZE PLOT
Per site, two small plots (2x3m) need to be included with a maize reference crop for BNF
measurements, with and without NPK application. The maize is replanted in the 2nd season at the
same time as the climbing beans. In both plots, the maize receives a basal FYM application at the
same rate as the cassava (5 tonnes ha-1 = 3 kg per plot of 2x3m). In one of the two reference
plots, additional NPK is added at one bag per hectare (1 g per planting hole).
FIELD LAYOUT
A detailed field layout for both spacings, with the delineation of the net plot, location of the
cassava cuttings, legume planting lines and the planting beds, are presented in following two
figures:
150
10m
1m
1m
0.4m
0.6m
6m
parcelle utile
(6m x 4m)
bouture manioc
ligne legumineuse
echantillonnage de
biomasse (1m x 0.5m)
« Ecartement 1 × 1 m »
« Ecartement 2 × 0.5 m »
10m
0.4m
0.8m
2m
0.4m
0.5m
6m
parcelle utile
(6m x 4m)
bouture manioc
lignes legumineuses
echantillonnage de biomasse
(2 fois 0.5m x 0.5m)
OBSERVATIONS
- Initial soil description and soil fertility evaluation (at least 8 cores per plot at 0-20 cm using a
‘W’ design; bulked; air-dried; stored pending shipment and analysis).
- Economic analysis (chronometer time to perform certain operations). The number of man-days
required for overall land preparation (for the entire trial area) needs to be estimated. All other
operations need to be timed using a chronometer on a per plot basis, particularly:
(1) planting bed preparation;
(2) fertilizer application;
(3) cassava and legume planting;
(4) weeding (1st, 2nd, 3rd and 4th weeding);
(5) pesticide spraying (if conducted);
(6) legume harvesting (including pod collection and shucking), possibly conducted in
several subsequent pod collections;
(7) cassava tuber harvesting;
151
note: operations performed for scientific purposes (biomass sampling, BNF assessment,…)
should be differentiated from essential agronomic operations and not be included in the
economic analysis.
In addition, the market price of all inputs (seeds, cuttings, fertilizer, pesticide,…) should be
collected. Market price information on all produce (groundnuts, cowpea, soybean, beans, maize
and cassava) should be collected at a nearby market on a monthly basis, during a one year period.
Also price information on side-products (e.g. cassava leaves and stems,…) should be gathered.
- Legume biomass sampling at 50% podding (or 75% flowering for groundnut). Two random
0.50 m strips within two neighbouring legume lines in the net plot area are cut; the number of
plants cut is counted and recorded.
In treatments with 1 × 1 cassava spacing, one biomass sample is collected. In treatments with 2 ×
0.5 cassava spacing, two biomass samples are collected: a first in a legume line bordering a
cassava line (within the net plot area), and a second in a central legume line, bordered by two
other legume lines (within the net plot area). At each biomass sampling event, 3 random maize
plants are cut from the maize reference plots (3 replicates). As legume species are likely to attain
the 50 % podding at different times, repeated maize reference samples may need to be taken. All
biomass samples need to be well labelled (with the date of sampling clearly marked), and sundried. After oven-drying and recording the dry weight, the biomass samples are ground and
stored pending analysis.
- Legume grain yield at harvest.
Similarly as for the biomass yield assessment, the yield in rows bordering cassava and rows
bordered by two other legume rows are collected and sub-sampled separately in treatments with 2
× 0.5 cassava spacing.
- Cassava harvesting at 12 months after planting.
All cassava within the net plot are harvested and separated in tubers, stems and leaves. The
number of stems and the number of tubers are recorded. The total weight of stems, leaves and
tubers within the net plot is determined. Subsequently, the stems are cut in pieces of 20cm and 30
random pieces are subsampled. A subsample of 16 tubers and a subsample of about 1 kg of fresh
leaves are collected. The weights of the subsamples are recorded at the same time as total yield
measurements. Subsamples are taken to the station and dried. After recording the oven-dry
weight, the samples are ground and stored pending analysis.
152
A
AN
NN
NE
EX
X 1144:: C
CA
ASS--33:: IIM
MPPR
RO
OV
VE
ED
D FFE
ER
RT
TIIL
LIIT
TY
Y
M
MA
AN
NA
AG
GE
EM
ME
EN
NT
T IIN
NC
CA
ASSSSA
AV
VA
A SSY
YSST
TE
EM
MSS
Objectives:
(i) to evaluate potential of green manure application to improve cassava production, as compared
to the traditional slash and burn systems;
(ii) to evaluate additional and interactive effects of fertilizer application and green manure
application on cassava production;
(ii) to assess nutrient uptake during growth, as affected by nutrient availability;
Sites
The trial will be established as a researcher-led on-farm trial with an association at Kisantu
(CIALCA-managed) and one association at Mbanza Nzundu (VLIR-managed). Sites should be
chosen carefully and be as homogeneous as possible, neither on strong-sloping land (avoid
erosion/run-off problems) nor in low valleys (avoid flooding), and have a moderate soil fertility
status (where a good response to inputs can be expected). The trial will be established in a
completely randomized block design, with 3 replicates set up as separate blocks.
Note: An improved, CMD-resistant cassava variety must be used (same variety for both sites),
e.g. Butama or Disanka.
Treatment structure
- Factor ‘natural vegetation’: the natural vegetation is either cut and removed, or applied under
the planting hills. Only in the 3rd treatment, the natural vegetation is cut and burnt before being
buried under the ridges.
- Factor ‘OM application’: Tithonia and Chromolaena are applied at 2.5 t DM ha-1, supplying
approximately 75 – 100 kg K ha-1. DM contents need to be calculated and application rates need
to be corrected for moisture content. Moisture contents for Tithonia and Calliandra are 81% and
72%, respectively (determined on a subsample taken by W. Biponda).
- Factor ‘fertilizer application’: NPK is applied at different rates (40, 120 or 200 kg K ha-1).
Higher rates are split-applied (an initial 40 kg K ha-1 and additional application of 80 kg K ha-1
after 2 (and 4) months, depending on the treatment).
Table 1: Detailed description of the various treatments
OM application
Nr. natural vegetation
1
cut & removed
2
cut & placed under hill
3
cut, burnt & placed under
hill
4
cut & removed
Tithonia, 2.5 t DM ha-1
5
cut & removed
Chromolaena, 2.5 t DM ha-1
6
cut & removed
7
cut & placed under hill
8
cut & removed
9
cut & removed
10 cut & removed
11 cut & removed
12 cut & removed
13 cut & removed
-
153
fertilizer application
40 kg K ha-1
40 kg K ha-1
40 kg K ha-1
40 kg K ha-1
120 kg K ha-1
120 kg K ha-1
120 kg K ha-1
200 kg K ha-1
Plot size will be 12 by 8 m (6 lines of cassava on ridges of 12 m long and 1 m wide, 40cm
between ridges; Figure 1). On the ridge, the distance between cassava plants is 1 m (within the
row). Care should be taken that the planting beds are installed perpendicular to the slope
direction.
12m
(a)
II
1m
(ridge)
1.4m
1m
slope direction
40cm
(furrow)
8m
I
12m
(b)
II
8m
Note:
1m
(ridge)
Figure 1:
1.4m
1m
Layout of a cassava plot with 1x1.4 m cassava spacing and 2 net plots to
be harvested at 6 months (net plot I) and at 11-12 months (net plot II)
after planting. The selection of the net plots can follow either the outline
in (a) or in (b) (randomly selected).
Each trial consists of 3 blocks (replicates) of 13 plots each. Total land area required per site equals
42 are (including border zones between plots).
MANAGEMENT OF NATURAL VEGETATION
In all plots, the natural vegetation is slashed and exported (not burned!). In a separate area of
864m2 (24m × 36m), however, the vegetation is slashed separately and kept isolated. This will
provide the organic matter for treatments 2, 3 and 7. The total amount of vegetation slashed in
154
slope direction
40cm
(furrow)
I
this area is weighed, cut in smaller pieces (<20cm), thoroughly homogenized and divided in 9
equal quantities (3 treatments × 3 reps). A subsample is taken by mixing a handful out of each of
the 9 quantities of residues. The fresh weight of the subsample is recorded; the subsample is then
taken to the station for drying, determination of the dry weight, grinding and storage pending
analysis.
In each of the 9 plots where residues of the natural vegetation are retained, a quantity of residues
is spread equally in 6 strips of 8 m long and 1 m wide, where the ridges will be established. These
are then either burnt and buried (treatment 3), or immediately buried (treatments 2 and 7) by
creating the ridges. These preparations should be performed about two weeks before
planting, to allow the residues to decompose.
ORGANIC MATTER APPLICATION
Organic matter (either Tithonia or Chromolaena residues, chopped < 20cm) is collected from
nearby sites. Only younger plant parts should be gathered; woody stems must be avoided. The
material collected is thoroughly homogenized, and a representative subsample is taken. After
recording the fresh weight, the subsamples are taken to the station for drying, determination of
the dry weight, grinding and storage pending analyses.
The residues are applied similarly as the natural vegetation residues, by spreading the residues
equally in 6 strips of 8 m long and 1 m wide, where the ridges will be established. Fresh matter is
applied, but rates need to be corrected for moisture content. The application rate of 2.5 t ha-1
corresponds to 4 kg DM per strip of 8 m (24 kg DM per plot = 6 strips). This corresponds to 21
kg fresh matter per strip for Tithonia and 14 kg fresh matter per strip for Chromolaena. The
residues are immediately buried by creating the ridges. These preparations should be
performed about 2 weeks before planting, to allow the residues to decompose.
Note:
Note:
Per site, 216 kg Tithonia and 216 kg Chromolaena dry residues are required (3 treatments × 3 reps ×
24 kg per plot), corresponding to 1137 kg fresh Tithonia residues and 758 kg fresh Chromolaena
residues.
Per site, the organic residues (Tithonia and Chromolaena) and natural vegetation (both the slashed
material and the ashes after burning) should be representatively subsampled. The subsample is
weighed fresh, oven-dried, weighed after drying to determine the MC, and subsequently ground
and stored pending analysis for nutrient contents.
FERTILIZER APPLICATION AND PLANTING
NPK (0.17:0.17:0.17) is applied at different rates (40, 120 and 200 kg K ha-1). Fertilizer is applied
at planting, in the “planting hole” at an initial rate of 40 kg K ha-1 (23.53 g NPK per planting
hole). With the hoe, a small pit is dug in which the fertilizer is applied, after mixing thoroughly
with some soil to avoid chemical burning. The pit is subsequently covered with soil and a cassava
cutting is planted vertically on the pocket.
In treatments with NPK application at 120 kg K ha-1, additional fertilizer (80 kg K ha-1) is applied
at 2 months after planting, by digging a small trench next to each plant, and applying the NPK
(47.06 g NPK per planting hole) in the trench. In treatments with NPK application at 200 kg K
ha-1, two additional doses of fertilizer (each 80 kg K ha-1) are applied at 2 and 4 months after
planting respectively, by digging a small trench next to each plant, 20 cm away from the plant,
and applying the NPK (47.06 g NPK per planting hole) in the trench.
Note:
The quantities of 23.53 and 47.06 g of NPK do not need to be weighed with a balance.
Application can be facilitated by finding a cap that approximately contains 23.53 g or 47.06 g of
NPK.
Observations
- Initial soil description and soil fertility evaluation: per replicate block, a composite topsoil (020 cm) sample is collected by mixing two cores from each plot within the replicate block. The
soil is air-dried and stored pending shipment and analysis. Soils should be sampled before trial
establishment, i.e. after slashing but before the application of organic matter and ridging.
155
- Economic analysis (chronometer time to perform certain operations). The number of man-days
required for overall land preparation (for the entire trial area) needs to be estimated. All other
operations need to be timed using a chronometer on a per plot basis, particularly:
(1) organic matter collection, transport, cutting and application;
(2) burning (treatment 3);
(3) planting bed preparation;
(4) fertilizer application;
(5) cassava planting;
(6) weeding (1st, 2nd, 3rd and 4th weeding);
(7) cassava tuber harvesting;
note: operations performed for scientific purposes (biomass sampling,…) should be differentiated
from essential agronomic operations and not be included in the economic analysis.
In addition, the market price of all inputs (cuttings, fertilizer,…) should be collected. Market price
information on produce (cassava tubers) should be collected at a nearby market on a monthly
basis, during a one year period. Also price information on side-products (cassava leaves and
stems,…) should be gathered.
- Three cassava biomass samplings and a final yield assessment are included at 6 months after
planting and at harvest (11-12 months after planting). At each event, one of the 2 net plots (see
Figure 1) is randomly harvested. The biomass is separated in tubers, stems and leaves. The
height of the stems and number of stems and the number of tubers are recorded. The total
weight of stems, leaves and tubers within the net plot is determined. Subsequently, the stems are
cut in pieces of 20cm and 30 random pieces are subsampled. A subsample of 12 tubers and a
subsample of about 1 kg of fresh leaves are collected. The weights of the subsamples are
recorded at the same time as total yield measurements. Subsamples are taken to the station and
dried. After recording the oven-dry weight, the samples are ground and stored pending analysis.
Note:
In treatments with OM application or buried natural vegetation, OM decomposition is
assessed qualitatively (or quantitatively if possible, by digging up, drying and weighing the
remaining residues).
156
A
AN
NN
NE
EX
X 1155:: P
PRREEFFEERREEN
NC
CE
ESS D
DE
ESS E
ESSSSA
AIISS
D
D’’A
AM
ME
EL
LIIO
OR
RA
AT
TIIO
ON
ND
DE
ESS SSY
YSST
TE
EM
ME
ESS A
AG
GR
RIIC
CO
OL
LE
ESS B
BA
ASSE
ESS SSU
UR
R
L
LE
EM
MA
AN
NIIO
OC
C PPA
AR
RL
LE
ESS A
AG
GR
RIIC
CU
UL
LT
TE
EU
UR
RSS D
DE
EK
KAABBAAM
MB
BA
A
Consortium for Improving Agriculture-based Livelihoods in Central Africa
= CIALCA /Sud-Kivu =
DGDC Legume project of TSBF-CIAT
Préférences des essais d’amélioration des systèmes agricoles basés sur le manioc par les
agriculteurs de Kabamba
Résultats de l’évaluation à la floraison des légumineuses associées avec le manioc
Bukavu, janvier 2008
INTRODUCTION
L’évaluation à la floraison des essais CAS-2 s’est déroulée du 18 au 19 décembre 2008.
L’évaluation a été assurée par Kasereka Bishikwabo et Adrien Chifizi respectivement socioéconomiste au CIAT/Bukavu et responsable des activités avec les associations au sein de la
plateforme DIOBASS.
L’objectif a été d’évaluer les préférences par les agriculteurs des essais de gestion des systèmes
agricoles avec manioc.
Ce rapport préliminaire conçue pour permettre de décider des activités en saison B2008 présente
les méthodes et matériel utilisées, les résultats de l’évaluation et quelques commentaires ainsi que
des suggestions pour préparer l’acceptation des technologies qui sont préférées.
1. METHODES ET MATERIEL
1.1 Description de l’essai
L’essai d’association manioc-légumineuses comportait cinq traitements dans huit parcelles
(Tableau 1).
Tableau 1 : Traitements et contenu des parcelles dans les essais CAS-2
No
1
Traitement
Effet de la
technique de
semis
Numéro parcelle et contenu
Numéro parcelle et contenu
Parcelle no 1 : Traditionnel (en vrac)
Parcelle no 2 : Ligne
Variétés locales de manioc en semis
traditionnel (en vrac) et de haricot nain.
Variétés locales de manioc à
écartement 1mx1m et de haricot
nain.
Parcelle no 3 : Améliorées
Variétés améliorées de manioc à
écartement 1mx1m et de haricot
nain
Parcelle no 4 : 2mx0,5m
écartement 2mx0,5m et de haricot
nain.
Parcelle no 6 : Avec NPK
écartement 2mx0,5m et de haricot
nain avec NPK.
Parcelle no 8 :
écartement 2mx0,5m et soja.
2
Effet de la
variété
Parcelle no 2 : Locales
Variétés locales de manioc à écartement
1mx1m et de haricot nain
3
Effet de
l’écartement
Parcelle no3 : 1mx1m
écartement 1mx1m et haricot nain
4
Effet du NPK
5
Effet d’autres
espèces de
légumineuses
Parcelle no 5 : Sans NPK
écartement 2mx0,5m et de haricot nain
sans NPK.
Parcelle no 7 :
écartement 2mx0,5m et arachide.
157
Ce protocole a été respecté. Une erreur a été remarquée au niveau de l’association Maendeleo où
il y a eu deux variétés différentes de haricot nain, l’une dans la parcelle no 1 et l’autre dans la
parcelle no 2. Dans la parcelle no 2 la variété de manioc était améliorée au lieu d’être locale.
1.2 Les agriculteurs évaluateurs
Trente et un agriculteurs évaluateurs membres de trois associations ont évalué les essais de
gestion CAS-2 installés dans le site d’action de Kabamba (Tableau 2). Il y avait parmi eux 16
femmes et 15 hommes, tous membres des associations partenaires Maendeleo, Tuungane et
ADEPB.
Tableau 2 : Nombre d’agriculteurs évaluateurs des essais CAS-2 par sexe et par association
Association
Femme
Homme Total
Maendeleo
3
4
7
Tuungane
8
6
14
ADEPB
5
5
10
Total
16
15
31
1.3 Procédure d’évaluation
Une réunion a été tenue avec les membres évaluateurs au niveau de chaque association. Au cours
de la réunion, les participants ont décrit leur système traditionnel d’association manioc légumineuse. Ils l’ont par la suite comparé au système amélioré. Les objectifs des essais ont été
discutés dans l’intérêt que les agriculteurs comprennent mieux l’activité qu’ils sont en train de
réaliser. Par la suite, deux groupes par association ont été constitués : l’un composé uniquement
de femmes et l’autre d’hommes. Chaque groupe a généré et rangé selon l’ordre d’importance ses
critères d’appréciation du haricot nain, du manioc, de l’arachide et du soja. Dans l’essai, chaque
groupe a apprécié les différentes parcelles l’une après l’autre en donnant les observations
négatives et positives sur chaque parcelle d’essai. Ensuite il y a eu une comparaison des parcelles
par traitement. Chaque femme a reçu 10 billes de couleur verte et chaque homme en reçu pareil
mais de couleur bleue. Un sachet noir était suspendu sur l’étiquette dans chaque parcelle et
chaque agriculteur évaluateur était invité à partager ses billes entre les parcelles préférées en
prenant soin de mettre plus de billes dans le sachet à la parcelle la plus préférée. A la fin de
l’opération, les membres des deux groupes sont passés ensemble parcelle par parcelle pour
compter le nombre de billes bleues et celui des billes vertes. Une discussion s’en suivait chaque
fois pour justifier les écarts. (N.B. : dans les résultats, bille=cailloux). Toutes les données
récoltées ont été notées sur des fiches préalablement préparées par Mr Pieters. Pour élaborer ce
rapport préliminaire, quelques données synthèses sur chaque fiche ont été saisies dans Excel et
analysées dans SPSS12. Les graphiques ont été produites en utilisant le logiciel Excel.
2. RESULTATS DE L’APPRECIATION DES ESSAIS CAS-2
2.1 Les critères d’appréciation
Dans l’ensemble, les paysans ont généré 15 critères (Figure 1). Adaptation est la résistance de la
culture aux intemperies (eau abondante et sécheresse). Biomasse implique l’abondance des
feuilles et/ou tige ainsi que leur vigueur. Cynophoresol égal cynophore de la plante bien fixé
dans le sol. Feuilleaspect est l’apparence de la feuille ; quand les agriculteurs évoquent ce critère,
c’est souvent pour vanter la couleur vert-foncée mais aussi des feuilles larges. Fleursgousseab
est l’abondance des fleurs et/ou des gousses. Germinationbon veut dire que la plupart des
plants avaient levé. Maladiepas est l’absence de maladies ou d’autres signes inquiétant sur la
plante. Eauresiste est la résistance de la plante contre une abondance d’eau. Planthaut est la
hauteur de la plante ; une plante d’une hauteur élevée est mieux appréciée sauf pour l’arachide où
la faible hauteur ‘taillereduite’ est la qualité recherchée. Floraison2pas a été cité pour
l’appréciation des arachides. D’après les évaluateurs paysans, quand les arachides fleurissent deux
fois on ne s’attend plus à une bonne production. Ravageurspas est que la plante n’a pas été
affectée par les ravageurs. Meristemeeleve est le méristème élevé qui est préféré par les
agriculteurs. Tigegrosse est la grosse tige qui est préférée tandis que Tigetachepas est une tige
sans tâche.
158
Les critères Feuilleaspect et Maladiepas ont été utilisés pour apprécier toutes les quatre cultures
dans l’essai : manioc, arachide, haricot et soja. Tigegrosse a été appliqué sur toutes les cultures
sauf l’arachide. Le haricot nain et le soja sont plus préférés pour l’abondance de leurs gousses
et/ou fleurs. Le critère Maladiepas a plus servi pour l’appréciation du manioc et de l’arachide. Ces
deux cultures sont plus malades à Kabamba comparativement au haricot et au soja. Le manioc a
depuis plus de quatre ans été attaqué par la mosaïque. Les variétés locales d’arachide sont très
vulnérables aux maladies à tel enseigne que certains paysans se découragent à pratiquer cette
culture qui avec le manioc sont perçus comme les cultures qui rapportent le plus d’argent au
paysan de Kabamba.
Certains critères s’appliquent uniquement à l’appréciation de l’arachide : c’est le cas de
taillereduite, meristemeeleve, Cynophoresol et Floraison2pas.
9
8
7
6
arachide
5
haricot
4
manioc
3
soja
2
1
0
taillereduite
Tigetachepas
Tigegrosse
meristemeeleve
ravageurspas
Planthaut
eauresiste
maladiepas
Germinationbon
Floraison2pas
Fleursgousseab
Feuilleaspect
Cynophoresol
Biomasse
Adaptation
Figure 1 : Critères générés et utilisés par les agriculteurs pour l’appréciation des cultures
d’arachide, haricot, manioc et soja dans les essais CAS-2
Le critère Feuilleaspect est très important dans l’appréciation des cultures. Il a été relevé 11 fois
comme premier critère sur les 14 fois qu’il a été cité (Figure 2).
Adaptation
12
11
Biomasse
Cynophoresol
10
9
Feuilleaspect
Fleursgousseab
8
Floraison2pas
6
Germinationbon
maladiepas
5
4
4
3
3
2
2
1
1
1
0
Planthaut
2
1
0 0
eauresiste
3
3
0
0
1
1
0
0
0
0
ravageurspas
meristemeeleve
Tigegrosse
Critere 1
Critere 2
Critere 3
Tigetachepas
Figure 2 : Fréquence de critères d’appréciation par rang (critère 1 = premier critère, plus
important).
159
Les agriculteurs disent que la qualité de feuilles est le premier indicateur pour espérer ou non une
bonne production. Aussi est-il qu’à part le soja et l’arachide, les feuilles de manioc et des haricots
constituent un aliment consommé à Kabamba.
Le critère Fleurgousseab est cité trois fois comme premier, neuf fois comme deuxième et trois
fois comme troisième. C’est un critère important d’autant plus que les agriculteurs, acculés par le
besoin de manger d’abord, voudraient avoir des technologies qui entraînent une plus grande
production. Un plant ou une variété qui porte beaucoup de gousses et de fleurs est perçu comme
devant avoir un haut rendement. Ce critère s’applique aux légumineuses.
Le critère maladiepas est cité trois fois au rang premier, cinq fois au rang deuxième et quatre fois
au troisième rang. La maladie des cultures est perçue comme l’obstacle à la production.
Tigegrosse, cinq fois cité au deuxième rang et deux fois au troisième rang est un critère
important pour apprécier le manioc surtout.
2.2 Préférences des essais de gestion : association manioc-légumineuses
Les traitements évalués étaient les suivants : VARLOECART : variété locale écartement locale ;
VARLO1X1 : variété locale écartement 1 m x 1m ; VARAM 1x 1 : variété améliorée écartement
1mx1m ; VARAM 2x 05 : variété améliorée écartement 2mx0,5m ; VARAM2L2x05 : variété
améliorée parcelle prévue pour la deuxième légumineuse et NPK non appliqué ;
VARAM2L2x05N : variété améliorée parcelle prévue pour la deuxième légumineuse et NPK
appliqué ; VARAMAR2X05 : variété améliorée arachide comme légumineuse ;
VARAMSO2X05 : variété améliorée soja comme légumineuse.
En considérant la moyenne totale du nombre de cailloux pour les associations Maendeleo,
Tuungane et ADEPB et celle des rangs totaux pour les différents traitements, les traitements les
plus appréciés à Kabamba lors de l’évaluation à la floraison sont : variété améliorée soja comme
légumineuse (VARAMSO2X05) avec une moyenne de 11 cailloux et le rang moyen 2 , variété
améliorée parcelle prévue pour la deuxième légumineuse et NPK appliqué (ARAM2L2x05N)
avec une moyenne de 10 cailloux et le rang moyen 3 et variété améliorée écartement
2mx0,5m (VARAM 2x 05) avec une moyenne de 9 cailloux et le rang moyen 3 (Figure 3). Le
traitement le moins préféré est la variété améliorée parcelle prévue pour la deuxième
légumineuse et NPK non appliqué (VARAM2L2x05).
12
10
8
MCaillouxTO
6
MRangTO
4
2
0
VARLOE VARLO1 VARAM VARAM VARAM VARAM VARAM VARAM
CARL
X1
1X1
2X05
2L2X05 2L2X05N AR2X05 SO2X05
MCaillouxTO
5
6
5
9
3
10
6
11
MRangTO
6
6
6
3
7
3
5
2
Figure 3 : Fréquences totales des moyennes des cailloux (MCailluxTO) et des rangs (MRangTO)
par parcelle d’essai
La moyenne des cailloux (MCailluxTO) est e relation inverse de celle des rangs (MRangTO). Plus
une parcelle avait de cailloux, moins son rang était élevé ; ce qui veut dire qu’elle était plus
préférée.
Pour l’appréciation de trois traitements préférés (en gras) et du seul traitement moins préféré (en
italique), 48 appréciations positives et 19 appréciations négatives ont été données (Tableau 3). La
proportion du nombre d’avis négatifs sur l’ensemble des avis émis est très élevée pour la parcelle
moins préférée VARAM2L2X05 (6/16) ; elle est faible dans la parcelle la plus préférée
VARAMSO2X05 (4/13).
160
Tableau 3 : Avis positifs et négatifs émis par critère pour chacune des quatre parcelles.
Fleursgousseab
Biomasse
Feuilleaaspect
Hauteur manioc
Eaurésiste
Maladiepas
Tigegrosse
Germinationbon
Précocité
Ravageurspas
Total
VARAM2X05
Avis
Avis
positifs négatifs
4
1
1
2
2
1
2
3
1
VARAM2L2X05
Avis
Avis
positifs négatifs
5
1
1
1
2
3
1
1
1
VARAM2L2X05N
Avis
Avis
positifs négatifs
5
2
2
1
2
3
VARAMSO2X05
Avis
Avis
positifs négatifs
3
2
4
2
3
1
1
1
13
4
1
1
13
4
10
6
12
5
De quatre parcelles reprises dans le Tableau 3, les légumineuses (haricot et soja) sont préférées
pour l’abondance de leurs fleurs et gousses. Mais la variété de haricot utilisée CODMLB003 n’a
pas résisté à l’eau. La plupart des plants ont des feuilles ‘brûlées’ par l’eau, surtout dans la parcelle
non préférée VARAM2L2X05. La hauteur des plants de manioc est faible dans les parcelles où la
biomasse des légumineuses est abondante : c’est le cas de la parcelle avec soja où la variété SB24
ombrage le manioc ainsi que la parcelle avec NPK où la biomasse des haricots nains avec feuilles
toujours vertes est abondante.
En examinant l’effet des traitements, le haricot dans la parcelle avec NPK est plus préféré que le
haricot dans la parcelle sans NPK qui est cité sept fois comme étant beaucoup inférieur et
seulement une fois comme étant beaucoup supérieur (Tableau 4). La parcelle avec NPK a plus de
fleurs et gousses que la parcelle sans NPK. La préférence est inverse pour le manioc. Le manioc
dans la parcelle avec NPK est moins préféré que le manioc dans la parcelle sans NPK qui est
quatre fois supérieur et trois fois beaucoup supérieur. L’aspect du manioc n’est pas bon sous
l’ombrage d’une légumineuse ayant une biomasse élevée.
Tableau 4 : Résultats de la comparaison des parcelles sans et avec NPK ;
VARAM2X05≤ < = > ≥ VARAM2L2X05NPK.
Critère
Feuilleaspect
Fleursgousseab
Tigegrosse
Maladiepas
Eauresiste
Planthaut
Total
Culture
Haricot
Manioc
Haricot
Manioc
Haricot
Manioc
Haricot
Manioc
Haricot
Manioc
Haricot
Manioc
Haricot
Manioc
Nbre ≤
2
0
4
0
1
0
0
0
0
0
0
0
7
0
Nbre <
0
1
0
0
0
1
0
0
0
1
0
0
0
3
Nbre =
1
1
1
0
0
1
1
3
0
0
0
0
3
5
Nbre >
1
2
1
0
0
1
1
1
1
0
0
0
4
4
Nbre ≥
0
2
0
0
0
0
0
0
1
0
0
1
1
3
N.B. : ≤ : beaucoup inférieur que, < : inférieur que, = pas de différence, > : supérieur que
≥ : beaucoup supérieur que.
En comparant les parcelles avec écartements de 1mx1m et 2mx0,5m la différence n’est pas
évidente entre les plants de manioc sur les deux parcelles, sauf au niveau du critère aspect de
feuilles où la parcelle à 1mx1m est cité trois fois comme inférieure à la parcelle à écartement de
161
2mx0,5m. Le haricot sur la parcelle à écartement de 1mx1m est cité sept fois comme étant
inférieur à celui dans la parcelle à écartement de 2mx0,5m contre trois fois supérieur.
La différence liée au sexe n’est pas évidente entre les hommes (McaillouHO) et les femmes
(McaillouFE) pour la moyenne de cailloux utilisés lors de l’évaluation (Figure 4 ).
15
10
McaillouFE
McaillouHO
5
0
VARLOE VARLO1 VARAM1 VARAM2 VARAM2 VARAM2 VARAM VARAMS
CARL
X1
X1
X05
L2X05 L2X05NP AR2X05 O2X05
McaillouFE
5
6
5
9
3
10
6
9
McaillouHO
4
3
5
9
3
10
5
12
Figure 4 : Effet du sexe sur la préférence des parcelles d’essais
Il en va de même pour la moyenne de rangs entre les deux sexes.
3. QUELQUES COMMENTAIRES, OBSERVATIONS FAITES SUR LE
TERRAIN ET SUGGESTIONS
Les résultats de cette évaluation montrent que les parcelles avec technologies améliorées (engrais
NPK et écartement 2mx0,5m) ont été plus préférées que celle n’ayant pas ces technologies
(parcelles sans engrais et écartement de 1mx1m). Cette observation s’applique surtout à la culture
de haricot dont la tendance de rendement était déjà perceptible à partir des gousses. Pour le
manioc, il est trop tôt de juger de la pertinence de la préférence.
Bien que les parcelles préférées soient celles avec écartement de 2mx0,5m pour le manioc, les
agriculteurs ont dit qu’ils reconnaissent la facilité d’entretien et l’économie des semences dans un
champ semé en ligne. Mais le semis en ligne prend plus de temps de semis par rapport au semis
en vrac : ce qui fait qu’il peut faire rater la saison à un agriculteur n’ayant pas suffisamment de
main d’oeuvre. Lors du semis à ADEPB, à presque même durée, huit personnes auraient semé
une parcelle en ligne contre deux femmes qui ont semé en vrac la même dimension de parcelle.
Parfois les parcelles avec technologies améliorées ont un reçu un score d’appréciation inférieur à
celui des parcelles sans ces technologies. Les cultures sur une parcelle avec technologie non
améliorée ont dans certains cas eu la même préférence ou ont été mieux préférées (selon les
évaluateurs paysans) que celles sur une parcelle avec technologie améliorée. Ce résultat inattendu
ne semble pas lié au traitement appliqué mais plutôt aux conditions de terrain. Au niveau de
l’association ADEPB, la pente de la parcelle avec engrais était plus prononcée que celle de la
parcelle correspondante sans engrais. Cette dernière où l’érosion n’est pas manifeste était sur un
terrain plat, elle accumule aussi des sols des couches supérieures érodés des pentes situées en
amont dans le même essai. Au niveau de l’association Maendeleo, les paysans se sont demandés si
l’ombrage des bananiers n’a pas déterminé un mauvais comportement de haricot
comparativement à une parcelle qui était totalement ensoleillée. Claude Rubyogo s’était aussi posé
la même question mais en l’absence des agriculteurs évaluateurs.
S’agissant du NPK, au niveau de l’association Maendeleo, les femmes n’ont pas préféré la parcelle
avec NPK qui était la plus préférée par les hommes. Les femmes ont reconnu que cette
appréciation était liée à leurs observations faites dans un autre champ où les cultures sur la
parcelle avec engrais n’ont pas poussé.
162
Eu égard à ce qui précède, nous suggérons :
- La variété améliorée de haricot nain COODMLB003 appliquée dans les essais CAS-2
n’a pas résisté à l’eau. Elle n’était pas une des meilleures variétés. Pour les prochains
essais, il serait mieux d’utiliser une des variétés naines qui a été le plus appréciées et qui
continue à l’être à Kabamba ;
- Il existe des outils de technologie appropriée pour semer rapidement en ligne, par
exemple le rayonneur. Il pourrait être opportun d’inventorier ces outils et d’utiliser un
pour le semis en saison B-2008 ;
- Dès qu’il y a des assurances que l’engrais est efficace dans l’amélioration des rendements
des cultures à Kabamba, une campagne de sensibilisation pourrait être amorcée dans le
but de favoriser l’adoption ;
- Ce n’est qu’à l’évaluation des récoltes qu’on peut se faire une opinion sur les préférences
des technologies essayées.
- Nous n’avons pas enregistré de reproche lié à l’essai CAS-2 lors de l’évaluation.
Cependant lors de l’évaluation de l’essai SYS-2 à Luhihi, l’association Rhusimane a
reproché l’utilisation des monocultures dans les essais : la monoculture, ont dit les
membres, constitue une perte car après la récolte des légumineuses et des céréales il ne
reste plus rien dans le champ alors que si ces cultures étaient associées au manioc, celui-ci
leur permettrait de gagner plus ;
- Les cultures semées en association avec le manioc au champ en saison B à Kabamba
sont surtout le haricot et les céréales (maïs et/ou sorgho).
163
A
AN
NN
NE
EX
X 1166:: F
FE
ER
R--11:: IID
DE
EN
NT
TIIFFIIC
CA
AT
TIIO
ON
NO
OFF IIN
NPPU
UT
TSS
R
RE
EQ
QU
UIIR
RE
ED
D FFO
OR
R SSO
OIIL
L FFE
ER
RT
TIIL
LIIT
TY
YA
AM
ME
EN
ND
DM
ME
EN
NT
T
Objectives:
(i) to investigate yield limiting soil constraints for bean and maize and examine yield
improvements by amendments with combinations of lime, manure, NPK and mavuno;
(ii) to explore variability in soil fertility status within the study area;
Sites:
8 sites: 2 fields (one of medium and one of poor soil fertility status) at each of the 2 associations
at Lurhala (CINAMULA and ALEMALU) and Burhale (APACOV and ABAGWASINYE).
3m
Treatment structure
Each trial consists of 10 plots (treatments), in 2 separate blocks (blocked per species):
- main plots: soil amendments
Maize
Climbing bean
1. control;
2. NPK only;
3. mavuno (NPK + micronutrients);
4. FYM only;
5. FYM + NPK;
6. FYM + mavuno;
7. lime only;
0.75m
0.5m
8. lime + NPK;
9. lime + mavuno;
10. Tithonia residues (take subsample per trial, determine FW and DW and store sample).
- blocks: species (maize, climbing bean);
Trial management
Plant varieties and spacing:
Variety: Katsumani cv.;
Maize:
Plant spacing: 0.75m between rows, 0.25m between plants within the row;
Plot size: 6 rows of 3m = 4.5m × 3m = 13.5m2.
Climbing bean: Variety: AND10;
Plant spacing: 0.50m between rows, 0.10m between plants within the row;
Plot size: 6 rows of 3m = 3m × 3m = 9m2.
Area required per trial = 10 × (9 + 13.5) = 225 m2 ≈ 3 are.
Soil amendments and mode of application:
Lime: 4 tonnes ha-1, broadcasted and incorporated;
(=5.4 kg per maize plot of 13.5 m2 and 3.6 kg per climbing bean plot of 9 m2)
FYM (from goats): 10 tonnes fresh matter ha-1, broadcasted and incorporated;
(=13.5 kg per maize plot of 13.5 m2 and 9 kg per climbing bean plot of 9 m2)
NPK: 20 kg P ha-1, banded application, in the planting line (but deep enough not to hamper
germination).
(=26.5 g per maize row of 3 m and 17.7 g per climbing bean row of 3 m)
Mavuno: 20 kg P ha-1, banded application, in the planting line (but deep enough not to
hamper germination).
(=40.9 g per maize row of 3 m and 27.3 g per climbing bean row of 3 m)
For maize, additional urea (top-dressing) is applied 6 weeks after planting at 60 kg N ha-1 (for
treatments with NPK or mavuno only);
Tithonia: 5 tonnes DM ha-1, (≈ 25 tonnes fresh matter ha-1), broadcasted and incorporated.
(=33.75 kg per maize plot of 13.5 m2 and 22.5 kg per climbing bean plot of 9 m2)
164
Observations
DAY 0
- Initial soil description and soil fertility evaluation.
Per trial, at least 8 soil samples are taken at 0-15cm prior to trial installation and application of
soil amendments, using a ‘W’ design. Soil samples are bulked, air-dried and stored pending
shipment and analysis (in total, 8 soil samples must be collected: 2 action sites × 2 associations ×
2 trials per association).
DAY 28 - 32
- Soil sampling at 4 weeks after planting.
Per treatment and species, 6 soil cores (0-15cm) are collected diagonally across the plot from
between the planting rows and are bulked and air-dried. A subsample of 200 g is send to Nairobi
for analysis; the remaining soil is stored at INERA.
- Maize leaf sampling at 4 weeks after planting.
Cut off (using a thoroughly cleaned knife) the youngest fully-developed leaf (i.e. the leaf collar
emerged) from its collar for 1 plant in the 3rd and 1 plant in the 4th line (at least 0.5 m away from
the plot ends). Choose representative plants of average height (not the largest or the smallest
plants). Avoid plants with diseased or damaged leaves, or plants having leaves stained with soil.
Place the leaves flat into large brown paper bags. Do not fold the leaves – if the leaves are too
long then cut the leaves in half, using a thoroughly cleaned knife. Label the bags with the trial
name, date, mandate area, action site, association name, field identification or farmer name,
“MAIZE LEAVES at 4WAP” and treatment. The bags with samples need to be laid out in the
sun and dried as quickly as possible. Leave samples should not be placed in plastic bags as this
may cause “respiration”. Neither should the leaf samples be laid bare to dry. These leave samples
will be analysed for micronutrient concentration, so all contamination should be avoided. If sundrying is not possible, leaf samples can be directly dried in an oven (65ºC) – care should however
be taken and not too many samples should be dried at once, to allow fluid air circulation and
swift drying. At all times the samples need to remain in the closed paper bags and should be kept
in a dry, well-aired place, protected from dust.
DAY 55 - 70
- Maize biomass and ear leave sampling at tasseling.
At tasseling, once the ear leaf is completely visible (silking to early milk stage, commonly between
55 and 70 days after planting), 3 maize plants are harvested randomly within the 3rd row or 4th
row (at least 0.5 m away from the plot ends). Avoid plants sampled for leaves at 4 weeks after
planting, and damaged, diseased or abnormally developed plants. The plants are separated in
leaves (cut off at the collar), stem + tassel, ear leaves and the immature kernel + silk + cob. Use a
thoroughly cleaned knife for cutting up, to avoid contamination. Do not fold up any plant parts,
but cut up in pieces (keep pieces as large as possible) to fit the paper bags. Label the bags with
the trial name (FER-1), date, mandate area (SK), action site (Lurhala or Burhale), association
name, field identification or farmer name, plant part (“MAIZE LEAVES at TASSELING”,
STEM + TASSEL”, “EAR LEAVES” or “KERNEL + SILK + COB”) and treatment. The bags
with samples need to be laid out in the sun and dried as quickly as possible. Samples should not
be placed in plastic bags as this may cause “respiration”. Neither should any samples be laid bare
to dry. These samples will be analysed for micronutrient concentration, so all contamination
should be avoided. If sun-drying is not possible, leaf samples can be directly dried in an oven
(65ºC) – care should however be taken and not too many samples should be dried at once, to
allow fluid air circulation and swift drying. At all times the samples need to remain in closed
paper bags and should be kept in a dry, well-aired place, protected from dust.
BEANS 50% FLOWERING
- Bean biomass and leaf sampling at 50 % flowering.
At 50 % flowering, 4 bean plants are harvested: 2 plants in the 3rd row and 2 plants in the 4th row
(at least 0.5 m away from the plot ends). Plants are cut at least 5 cm above the soil surface to
avoid contamination by soil particles. For each of the plants cut, the growing tips are identified
165
(several per plant) and starting from each growing tip, the youngest fully developed leaves (trifoliates) are separated and stored in a separate paper bag. The leave sample should not be folded,
but placed neatly inside the bag. Bags are labelled with the trial name (FER-1), date, mandate area
(SK), action site (Lurhala or Burhale), association name, field identification or farmer name, plant
part (“BEAN LEAVE SAMPLE at 50% FLOWERING” or “BEAN BIOMASS at 50%
FLOWERING”) and treatment. The bags with samples need to be laid out in the sun and dried
as quickly as possible. Samples should not be placed in plastic bags as this may cause
“respiration”. Neither should any samples be laid bare to dry. These samples will be analysed for
micronutrient concentration, so all contamination should be avoided. If sun-drying is not
possible, leaf samples can be directly dried in an oven (65ºC) – care should however be taken and
not too many samples should be dried at once, to allow fluid air circulation and swift drying. At
all times the samples need to remain in the closed paper bags and should be kept in a dry, wellaired place, protected from dust.
BEANS 50% PODDING
- Climbing bean biomass sampling at 50 % podding.
At 50 % podding, plants within a 50 cm section in the net plot are cut (note the number of plants
cut). Sections missing plants due to sampling at 50 % flowering need to be avoided. At the same
time, 3 maize plants per treatment are cut as reference for BNF assessment (in the 2nd, 3rd, 4th of
5th row, at least 50 cm away from the plot ends; avoid plants sampled for leaves at 4 weeks after
planting). If this coincides (less than 5 days difference) with the maize biomass sampling at
tasseling, harvesting of maize reference plants for BNF assessment can be omitted. Biomass
samples are air-dried, weighed, ground and stored pending shipment and analysis.
HARVEST
- Maize grain and biomass yield at harvest. Bean grain yield at harvest.
Samples are air-dried, weighed, ground and stored pending shipment and analysis.
166
A
AN
NN
NE
EX
X 1177:: P
PRRO
OT
TO
OC
CO
OL
L FFO
OR
RO
ON
N--SST
TA
AT
TIIO
ON
NA
AN
ND
DO
ON
N--FFA
AR
RM
M
M
MU
UL
LC
CH
HT
TR
RIIA
AL
LSS
Introduction
Deux types de dispositifs expérimentaux sont retenus :
- essais menés en milieu contrôlé, soit en station expérimentale, soit en milieu agricole,
mais dans les deux cas gérés par les chercheurs (étudiants doctorants) : ces essais sont
destinés à comprendre les processus (process oriented research) et sont assortis de mesures et
déterminations choisies et formatées dans ce but ;
- essais menés milieu paysan, dans des parcelles situées en milieu réel, dans un triple but
d’extrapolation, de validation et démonstration.
Ces deux dispositifs sont assortis de matériaux partagés, de suivis communs et de suivis distincts :
- Matériaux communs : sites (1), protocoles (2)
- Suivi commun : paramètres agronomiques (3)
- Suivis distincts :
- cycle des éléments minéraux (Syldie Bizimana)
- propriétés physiques (Tony Muliele)
1. Sites
1.1. Essais gérés par les chercheurs
Burundi
Gitega
Ferralsol
(FAO Acrisol / « latérite »)
Cibitoke
Vertisol (FAO Vertisol / alluvions)
Kirundo
Ferrisol (FAO Nitisol sur schiste)
Congo
Mulungu
“sol brun s/ basalte tertiaire”
(FAO Nitisol-Ferralsol / basalte)
Walungu
“sol rouge s/ basalte tertiaire”
(FAO Ferralsol / basalte)
Rwanda
Butare-Gitarama Ferralsol
(FAO Acrisol sur granite)
Kibungo
Ferrisol (FAO Nitisol sur schiste)
Ruhengeri
sol brun (FAO Andosol sur cendres volcaniques)
1.2. Essais en milieu paysan
Pour chaque site/région, nous sélectionnerons des exploitations dont le système local est
dominant, afin de réduire la variabilité du « témoin » (systèmes similaires).
Par système local, nous sélectionnerons au moins trois exploitations. Le nombre minimum
d’essais en milieu paysan s’élèvera donc à 24.
2. Protocoles
Traitements : témoin + 3 traitements
Pailles importées
Pailles exportées
T0
Non
Oui
T1
Non
Non1
T2
Oui3
Non2
T3
Oui4
Non
1 autopaillage
2 autopaillage + importation
3 paillage graminéen (Hyparrhenia diplandraà
4 paillage graminéen (Guatemala grass : Tripsacum laxum)
167
Labour
Oui
Non
Non
Non
Haricot
Oui
Oui
Oui
Oui
Répétitions : 4
Disposition : en fonction du terrain (topographie, …)
Carré 6x6 (36 bananiers), ‘net plot’ 4x4 plants (16 bananiers)
Rectangle 7x5 (35 bananiers), ‘net plot’ 3x5 plants (15 bananiers)
Densité 2x2 m
Epaisseur du paillage :
~ 5cm (~25T/ha –matière sèche–)
2.2. Essais en milieu paysan
Traitements
Pailles importées
Ta
Non
Tb
Non
Tc
Oui1
1 paillage graminée
Labour
Oui
Non
Non
Le protocole dépend du système local appliqué par l’agriculteur, l’objectif étant de valider, en
milieu paysan, les résultats obtenus par les essais gérés par les chercheurs.
Dans ces conditions :
- le témoin varie en fonction du système local ;
- l’exportation du paillage n’est pas imposée, mais sera enregistrée ;
- si Ta/Tb intègre le haricot, cette plante doit aussi être cultivée dans Tb/Tc ;
- la densité sera enregistrée, en tentant de retenir des systèmes locaux adoptant des
densités proches de 2500 p/ha (2x2m).
Répétitions : aucune au niveau de l’exploitation ; par système local, nous sélectionnerons au
minimum trois exploitations.
Disposition :
‘net plot’ 3x3 plants (9 bananiers), ‘total plot’ 5x5 plants (25 bananiers)
3. Matériaux et caractérisation des sites
Matériel de plantation
rejets baïonnettes à sélectionner sur pieds ; parage très soigneux ; désinfection à l’eau
bouillante (<30 sec) ?
Caractérisation des sols
Description morphologique des profils (00-160 cm)
Analyses :
- granulométrie, densité apparente
- propriétés physico-chimiques (CEC, pH, cations échangeables, C, N, P)
- analyses totales
- analyses minéralogiques
- analyses spécifiques
4. Suivi de la croissance et des paramètres agronomiques
Tous les bananiers mesurés seront soigneusement numérotés. A chaque bananier enregistré,
correspondra un « carnet de croissance » (fichier) reprenant toutes les mesures liées à la
croissance et aux paramètres agronomiques.
De la plantation à la récolte (rythme mensuel) :
- circonférence (base du plant, 100cm dès que possible) ;
- nombre de feuilles mortes (retirées après mesure) et vivantes, feuille morte <50% de
limbe vert ;
- hauteur du plant.
168
Floraison :
- circonférence (base du plant, 100cm) ;
- nombre de mains à fleurs femelles
- nombre de doigts de la rangée inférieure de l’avant-dernière main (complète) ;
- hauteur du plant.
Récolte :
- poids du régime.
Appréciation des contraintes biotiques (selon protocoles standards existants) :
- dégâts foliaires dus aux maladies fongiques ;
- dégâts au rhizome et à la base du pseudo-tronc (charançon) ;
- dégâts racinaires (nématodes, …).
4.2. Essais en milieu paysans
Avant la floraison, rythme bimensuel à trimestriel (de la plantation à la récolte) :
- circonférence (base du plant, 100cm dès que possible) ;
- nombre de vivantes, feuille vivante >50% de limbe vert ;
- hauteur du plant.
Entre floraison et récolte :
- circonférence (base du plant, 100cm) ;
- nombre de mains à fleurs femelles ;
- nombre de doigts de la rangée inférieure de l’avant-dernière main (complète) ;
- hauteur du plant.
5. Paramètres relatifs aux propriétés physiques des sols
Note :
les déterminations réalisées dans les essais gérés par les chercheurs et en milieu paysan sont affectées par
les symboles respectifs ec et mp.
5.1. Caractérisation de base (ec) :
- granulométrie ;
- courbe teneur en eau succion ;
- densité apparente (en fonction du volume de la motte) ;
- constitution minéralogique et stock organique.
5.2. Suivi (plantation–récolte) selon un rythme saisonnier :
- densité racinaire par ‘core sampling’ : ec + mp
- résistance mécanique à la pénétration : ec + mp
- densité apparente : ec + mp
- teneur en eau (EM probes in PVC tubes) : ec
- température (sondes) : ec
- infiltration (double anneau) : ec
- rainfall simulation (bananiers de bordure (1ère récolte), net plot (2ème récolte)) : ec
- pluviométrie : ec + mp
5.3. Caractérisation en fin d’essai (près de la récolte du 2ème cycle)
Ces caractérisations seront réservées aux observations et mesures destructives près de la
récolte du 2ème cycle. Des profils racinaires1 seront ouverts au pied des bananiers significatifs :
- comptages racinaires réalisés à l’aide d’une grille : ec + mp
- résistance mécanique à la pénétration : ec + mp
- densité apparente: ec + mp
- infiltrométrie à disque par horizon : ec + mp
En fonction des possibilités et uniquement en scénario ec, des profils racinaires pourraient
être étudiés à la récolte du 1er cycle au pied des bananiers de bordures.
1
Delvaux et Guyot (1989) Caractérisation de l'enracinement du bananier au champ. Incidences sur les relations
sol-plante dans les bananeraies intensives de Martinique. Fruits, 44 (12): 633-647.
169
6. Suivi des paramètres relatifs au cycle des éléments minéraux
6.1. Caractérisation de base des sols (ec) :
- granulométrie ;
- densité apparente ;
- constitution minéralogique et stock organique ;
- analyses chimiques totales et stock des éléments minéraux ;
- stocks minéraux et organiques des éléments nutritifs ;
- stocks des éléments échangeables et non échangeables ;
- N, P minéralisables.
6.2. Suivi mensuel des paramètres pédologiques (ec, échantillon composite) :
- C, N totaux ; C oxydable, N minéralisable ;
- P total et P disponible ;
- pH, cations échangeables.
6.3. Suivi cyclique des paramètres pédologiques (mp, échantillon composite, 1er+2ème cycle à
la floraison cycle saisonnier à préciser) :
- C, N totaux ; C oxydable ;
- P disponible ;
- pH, cations échangeables.
6.4. Diagnostic foliaire à la floraison (ec + mp, 1er+2ème cycle, échantillon composite) :
- éléments majeurs N, P, K, Ca, Mg ;
- éléments mineurs.
6.5. Minéralomasse des bananiers (ec, 1er+2ème cycle, stade récolte) :
- éléments mineurs ;
- échantillonnage et analyse des parties aériennes des bananiers (cfr essais sol-plante,
Martin-Prével et al.) : pseudo-tronc, feuilles, régime ;
- détermination de la minéralomasse et des exportations minérales.
6.6. Minéralomasse des pailles importées :
- éléments mineurs
170
A
AN
NN
NE
EX
X 1188:: R
REELLEEVVAAN
NT
TR
RE
EPPO
OR
RT
TSS,, PPR
RE
ESSE
EN
NT
TA
AT
TIIO
ON
NSS,,
A
AN
ND
D PPU
UB
BL
LIIC
CA
AT
TIIO
ON
NSS
Articles in international refereed journals
Bairagi, R., and Ahsan, R.I., 1998. Inconsistencies in the findings of child nutrition surveys in
Bangladesh. American Journal of Clinical Nutrition 68:1267-1271.
Carpentier S., Witters E., Laukens K., Deckers P., Swennen R. and Panis B., 2005. Preparation of
protein extracts from recalcitrant plant tissues: An evaluation of different methods for twodimensional gel electrophoresis analysis. Proteomics 5 (10):2497-2507.
Carpentier S., Witters E., Laukens K., Panis B. and Swennen R., 2005. Proteome analysis: a
successful approach for functional research in recalcitrant non-model crops. Proceedings
11th symposium on Applied Biological Sciences. Leuven, Belgium, 6 October 2005.
Communications in Agricultural and Applied Biological Science 70 (2):3-4.
Carpentier S., Witters E., Laukens K., Van Onckelen H. , Swennen R. and Panis B., (in press).
Banana (Musa spp.) as a model to study the meristem proteome: acclimation to osmotic
stress. Proteomics.
Chun X. Xu, Panis B., Strosse H., Hua P. Li, Huo G. Xiao, Huai Z. Fan and Swennen R., 2005.
Establisment of embryogenic cell suspension and plant regeneration of the dessert banana
"Williams" (Musa AAA group). Journal of Horticultural Science & Biotechnology 80 (5):523528.
Coemans B., Matsumura H., Terauchi R., Remy S., Swennen R. and Sági L., 2005. SuperSAGE
combined with PCR walking allows global gene expression profiling of banana (Musa
acuminata), a non-model organism. Theoretical and Applied Genetics 111 (6):1118-1126.
Coemans B., Takahashi Y., Berberich T., Ito A., Kanzaki H., Matsumura H., Saitoh H., Tsuda S.,
Kamoun S., Sági L., Swennen R. and Terauchi R., (submitted). High-throughput in planta
expression screening identifies an ADP-ribosylation factor (ARF1) that is involved in nonhost resistance and R gene mediated resistance. Plant Physiology.
Dolezelova M., Dolezel J., Van den houwe I., Roux N. and Swennen R., 2005. Ploidy levels
revealed. Infomusa 14 (1):34-36.
Dussert S., Davey M. W., Laffargue A., Doulbeau S., Swennen R. and Etienne H., 2006.
Oxidative stress, phospholipid loss and lipid hydrolysis during drying and storage of
intermediate seeds. Physiologia Plantarum 127:192-204.
Fermont, A.M., P.J.A. van Asten, K.E. Giller. Increasing land pressure in farming systems in the
mid-altitude zone of East Africa: The changing role of cassava and consequences for system
sustainability. Submitted to Agriculture, Ecosystems and Environment.
Geuns J., Ledo Orriach M., Swennen R., Zhu G. Y., Panis B., Compernolle F. and Van der
Auweraer M., 2006. Simultaneous Liquid Chromatography Determination of Polyamines and
Arylalkyl Monoamines. Analytical Biochemistry 354: 127-131.
Henriet C., Draye X., Oppitz I., Swennen R. and Delvaux B., 2006. Effects, distribution and
uptake of silicon in banana (Musa spp.) under controlled conditions. Plant and Soil 287:359374.
Mtambanengwe F, Mapfumo P, Vanlauwe B 2006 Comparative short-term effects of different
quality organic resources on maize productivity under two different environments in
Zimbabwe. Nutrient Cycling in Agroecosystems, In Press.
Ohl T. J., Cullis M. A., Kunert K., Engelborghs I., Swennen R. and Cullis. C.A., (submitted).
Genomic changes associated with somaclonal variation in banana (Musa spp.). Physiologia
Plantarum.
Panis B., Helliot B., Strosse H., Remy S., Lepoivre P. and Swennen R., 2005. Germplasm
conservation, virus eradication and safe storage of transformation competent cultures in
banana: The importance of cryopreservation. Chang W.-C., Drew R. (ed.). Proceedings IInd
IS on Biotech. of Trop. & Subtrop. Species. Acta Horticulturae 692. ISHS, 51-59.
Panis B., Piette B. and Swennen R., 2005. Droplet vitrification of apical meristems: a
cryopreservation protocol applicable to all Musaceae. Plant Science 168:45-55.
171
Pérez Hernández J. B., Galán Saúco V., Swennen R. and Sági L., (submitted). Agrobacteriummediated transformation is an efficient tool to generate transgenic plants from embryogenic
cell suspension cultures of banana (Musa spp.), a tropical monocot species.
Pérez Hernández J. B., Swennen R. and Sági L., 2006. Number and accuracy of T-DNA
insertions in transgenic banana (Musa spp.) plants characterized by an improved anchored
PCR technique. Transgenic Research 15:139-150.
Pypers, P, J Delrue, J Diels, E Smolders, R Merckx 2006 Phosphorus intensity determines shortterm P uptake by pigeon pea (Cajanus cajan L.) grown in soils with differing P buffering
capacity Plant Soil 284:217-227.
Pypers, P, L Van Loon, J Diels, R Abaidoo, E Smolders & R Merckx 2006 Plant-available P for
maize and cowpea in P-deficient soils from the Nigerian Northern Guinea savanna Comparison of E- and L-values Plant and Soil 283:251-264.
Pypers, P, M Huybrighs, J Diels, R Abaidoo, E Smolders & R Merckx 2006 Does the enhanced
P acquisition by maize following legumes in a rotation result from improved soil P
availability? Soil Biology and Biochemistry, submitted.
Pypers, P., S. Verstraete, Cong Phan Thi, R. Merckx 2005 Changes in mineral nitrogen,
phosphorus availability and salt-extractable aluminium following the application of green
manure residues in two weathered soils of South Vietnam. Soil Biology & Biochemistry 37,
163-172.
Remy S., Thiry E., Coemans B., Windelinckx S., Swennen R. and Sági L., 2005. Improved TDNA vector for tagging plant promoters via high-throughput luciferase screening.
BioTechniques 38:763-770.
Samyn B., Sergeant K., Carpentier S., Debyser G., Panis B., Swennen R. and Van Beeumen J., (in
press). Homology-based functional proteome analysis: a successful approach for the nonmodel plant Musa spp. Jounal of Proteome Research.
Shakir, A., and Morley, D., 1974. Measuring malnutrition. Lancet 1(7860):758-759.
Strosse H., Schoofs H., Panis B., André E., Reyniers K. and Swennen R., 2006. Development of
embryogenic cell suspensions from shoot meristematic tissue in bananas and plantains (Musa
spp.). Plant Science 170 (1):104-112.
Talwana H., Speijer P. R., Gold C. S., Swennen R. and De Waele D., 2006. Effect of nematode
infection and damage on the root system and plant growth of three Musa cultivars
commonly grown in Uganda. Nematology 8(2):177-189.
Tittonell P, Leffelaar PA, Vanlauwe B, van Wijk MT and Giller KE 2006 Exploring diversity of
crop and soil management within smallholder African farms: a dynamic model for simulation
of nutrient (N) balances and use efficiencies at field scale. Agricultural Systems, 91, 71-101.
van Asten, P.J.A., S. Kaaria, A.M. Fermont, R.J. Delve. Challenges of conducting rigorous farmer
participatory research in Africa. Submitted to Experimental Agriculture.
Vanlauwe, B, P Tittonell and J Mukalama 2006 Within-farm soil fertility gradients affect response
of maize to fertilizer application in western Kenya. Nutrient Cycling in Agroecosystems, In
Press.
Zhu G. Y., Geuns J., Dussert S., Swennen R. and Panis B., 2006Change in sugar, sterol and fatty
acid composition in banana meristems caused by sucrose-induced acclimation and its effects
on cryopreservation . Physiologia Plantarum 128:80-94.
Papers presented at international congresses and symposia
Aert R., Hribova E., Delezel J., Swennen R. and Sági L., (in press). Cot filtration of the banana
(Musa acuminata) genome. Plant Genomics European Meeting V. Venice, Italy, 11-14
October 2006. Poster Abstract.
Arinaitwe G., Remy S., Thiry E., Sági L. and Swennen R., 2005. Integration of rice chitinase
genes in banana. International Consortium on Agricultural Biotechnology Research
(ICABR). Ravello, Italy, 6-10 July 2005. Poster abstract.
Bationo A, J Kihara, B Vanlauwe, J Kimetu and K L Sahrawat. 2006 Integrated nutrient
management - Concepts and experience from SSA (in Press).
Bationo A, Vanlauwe B, Kimetu J, Kihara J, Abdoulaye MS, Adamou A, Tabo R, Koala S 2005
Increasing land sustainability and productivity through soil fertility management in the West
African Sudano-Sahelian zone. In: Nutrient and Water Management Practices for Increasing
172
Crop Production in Rainfed Arid/Semi-arid Areas. International Atomic Energy Agency,
TECDOC 1468, pp. 53-75, IAEA, Vienna, Austria.
Birabwa, R., P.J.A. van Asten, I.N. Alou, G. Taulya. Got matooke (Musa AAA-EA) for
Christmas. Accepted for African banana conference, Mombasa, October 2008.
Carpentier S., Swennen R. and Panis B., 2006. The use of uni- and multivariate statistics in data
analysis: a case study of the banana meristem proteome. Proteom'Lux 2006: Bridging the gap
between gene expression and biological function. Luxembourg, Luxembourg, 11-14 October
2006. Abstract.
Carpentier S., Witters E., Laukens K., Swennen R. and Panis B., 2005. Two-dimensional gel
electrophoresis and subsequent protein identification via MALDI-MS/MS: a successful
approach to unravel the abiotic stress responses in a non-model organism (Musa spp.).
HUPO 4th Annual world congress: "From defining the proteome to understanding function.
Technical University Munich, Germany, 28 August - 1 September 2005. Molecular & Cellular
Proteomics 4 (8): S257.
Carpentier S., Witters E., Laukens K., Swennen R. and Panis B., 2006. Proteome research of
banana meristems to study cryoprotection. 43rd Meeting of the Society for Cryobiology in
association with the Society for Low Temperature Biology . Hamburg, Germany, 24-27 July
2006. Poster abstract.
Coemans B., Takahashi Y., Berberich T., Saitoh H., Matsumura H., Kanzaki H., Ito A., Remy S.,
Swennen R. , Sági L. and Terauchi R., 2006. High-throughput in planta expression screening
identifies an ADP-ribosylation factor (ARF1) that is involved in plant cell death. 3rd EPSO
conference. Plant dynamics: from molecules to ecosystems. Viségrad, Hungary, 28 May - 1
June 2006.
Gaidashova, S.V., G. Germeau, P. van Asten, B. Delvaux. 2007. Identification of constraints to
banana production in three eco-regions of Rwanda. Proceedings of ISAR Conference, March
2007, Rwanda.
Gaidashova, S., M. Bagabe, A. Nsabimana, P. van Asten. Fusarium wilt disease in Rwanda: A real
threat to apple banana market. Accepted for African banana conference, Mombasa, October
2008.
Hauser, S., P. van Asten. Methodological considerations on banana and plantain yield
determinations. Accepted for African banana conference, Mombasa, October 2008.
Jagwe, J., S. Abele, L. Niyangabo. Banana Marketing in Burundi. Accepted for African banana
conference, Mombasa, October 2008.
Jefwa, J.M., B. Vanlauwe, N. Sanginga, A. Elsen, E. Kahangi, L. Ruto, P. van Asten, T. Losentge,
M. Mwajita, J. Mungatu.. Abuscular Mycorrhizal Fungi (AMF) and root nematode damage in
banana plantains of maragua district in Central Kenya. Accepted for African banana
conference, Mombasa, October 2008.
Kahangi, E., B. Vanlauwe, N Sanginga, P van Asten, Mbugua, L Ruto, Kimenju, J Jefwa.
Characteristics of banana farms and causes of decline in production in maragua district in
Central Kenya. Accepted for African banana conference, Mombasa, October 2008.
Muchunguzi, P, M Batte, M Nyine, M Pillay, C Kiwanuka, P van Asten, L Wairegi, J Lorezenzen.
Innovations for producing clean banana planting materials by macropropagation in Uganda.
Accepted for African banana conference, Mombasa, October 2008.
Murekezi, C., P.J.A. van Asten. Farm banana constraints in Rwanda: A farmer’s perspective.
Mwangi, M., Vigheri, N., Fiaboe, K., Bandyopadhyay, R., 2006. Emergence and Spread of Banana
Xanthomonas Wilt in East D.R. Congo and Strategies to Halt its Spread Towards Central
and West Africa. Presentation given by P. van Asten at the MUSACO meeting, 18-22
September 2006, Limbé, Cameroon.
Nyombi, K., van Asten, P.J.A., Taulya, G., K. Giller, 2006. The effect of soil quality parameters
on growth rates of Kisansa TC banana plants in Senge, Uganda. Paper presented at the Soil
Science Society of East Africa Conference, Masaka, November 2006.
Nyombi, K., P.J.A. van Asten, P.A. Leffelaar, M. Corbeels, C. K. Kaizzi and K.E. Giller.
Developing a simulation model to understand and improve growth of East African highland
banana (AAA-EAHB, cv. Kisansa). Accepted for African banana conference, Mombasa,
October 2008.
173
Ochieno, D.M.W., T. Dubois, D. Coyne, M. Dicke, A. van Huis, P. van Asten. Benefits of non
pathogenic Fusarium oxysporum on bananas as influenced by nutrients. Accepted for
African banana conference, Mombasa, October 2008.
Panis B. and Lambardi M., 2005. Status of cryopreservation technologies in plants (crops and
forest trees). International Workshop on "The role of biotechnology for the characterisation
and conservation of crop, forestry, animal and fishery genetic resources". Turin, Italy, 5-7
March 2005. 43-54.
Panis B. and Swennen R., 2005. Applications of cryopreservation in banana. COST 843 final
conference / COST 843 and COST 851 joint meeting. Stara Lesna, Slovakije, 28 June - 2 July
2005. 17-19. Abstract.
Sági L., Remy S., Coemans B., Thiry E., Santos E., Matsumura H., Terauchi R. and Swennen R.,
2005. Functional Analysis of the Banana Genome by Gene Tagging and Sage. Plant &
Animal Genomes XIII Conference. San Diego, CA, USA, 15-19 January 2005.
Santos E., Remy S., Thiry E., Windelinckx S., Swennen R. and Sági L., (submitted). Tagging
Novel Promoters in Banana Using the Luciferase Reporter Gene. 27th International
Horticultural Congress on "Global Horticulture: Diversity and Harmony". Seoul, Korea, 1319 Augustus 2006. Acta Horticulturae.
Santos E., Remy S., Thiry E., Windelinckx S., Swennen R. and Sági L., 2006. Tagging LowTemperature Responsive Promoters in Banana Using the Luciferase Reporter Gene. 27th
International Horticultural Congress on "Global Horticulture: Diversity and Harmony".
Seoul, Korea, 13-19 Augustus 2006. 218-219. Abstract.
Taulya, G., van Asten, P.J.A., Okech, S.H.O, Gold, C.S., 2006. Effect of drought on the yields of
an East African highland banana (cv. Kisansa) in Uganda. Paper presented at the Soil Science
Society of East Africa Conference, Masaka, November 2006.
van Asten, P.J.A., A. Twagirayezu, S. Gaidashova. 2008. Effect of Guatamala grass (Tripsacum
laxum) mulch applications on soil moisture conservation and soil fertility status. Proceedings
of ISAR Conference, March 2007, Rwanda
van Asten, P.J.A., L.W.I. Wairegi, F. Bagamba, and C. Drew. Factors driving fertilizer adoption
in banana systems in Uganda. Accepted for African banana conference, Mombasa, October
2008.
van Asten, P.J.A., D. Florent, M.S. Apio. Opportunities and constraints for dried dessert banana
export in Uganda. Accepted for African banana conference, Mombasa, October 2008.
van Asten, P.J.A., D. Mukasa, N.O. Uringi. Farmers earn more money when banana and coffee
are intercropped. Accepted for African banana conference, Mombasa, October 2008.
van Asten, P.J.A., C. Murekezi, L. Wairegi, S. Gaidashova, T. Muliele, S. Hakizamana, S.
Bizimana, N. Vigheri, J. Walangululu, B. Delvaux. Commonly perceived production levels
and constraints in East African highland banana systems do not correspond to field realities.
van Asten, P.J.A., I.N. Alou, B. Vanlauwe, D. Mubiru, T.K.W Mugaaga. Resource availability and
yield differences between farms in a banana producing valley in Southwest Uganda. Accepted
for African banana conference, Mombasa, October 2008.
Van Asten, P.J.A., 2006. Soil quality constraints in EA highland bananas. Poster presented at the
Farmer Field School Experiences'Workshop on Land and Water Management in Africa, Jinja
April 24-29 2006.
Van Asten, P.J.A., Abele, Blomme, G., S. Sanginga, P., Vanlauwe, B., Vigheri, N., Lunze, L.,
Walangululu J., Farrow, A., 2006. The role of bananas in Eastern DRCongo. Presentation at
the MUSACO meeting, 18-22 September 2006, Limbé, Cameroon.
Van Asten, P.J.A., Fermont, A.M. 2006. Is farmer knowledge guiding us in the right direction?
Paper presented at the Farmer Field School Experiences'Workshop on Land and Water
Management in Africa, Jinja April 24-29 2006. 11 pp.
Van den houwe I., Panis B., Arnaud E., Markham R. and Swennen R., 2005. Banana (Musa spp.)
genetic resources maanged in the INIBAP-IPGRI gene bank: conservation and
documentation status. 3rd GBIF Science Symposium "Tropical Biodiversity: Science, Data,
Conservation". Brussels, Belgium, 18-19 April 2005. Abstract.
Van den houwe I., Panis B., Arnaud E., Markham R. and Swennen R., 2006. The management of
banana (Musa spp.) genetic resources at the IPGRI/INIBAP-IPGRI gene bank: the
174
conservation and documentation status. Segers H., Desmet P. , Baus E. (ed.). Tropical
Biodiversity: Science, Data, Conservation. Proceedings of the 3rd GBIF Science Symposium.
Brussels, Belgium, 18-19 April 2005. 143-152.
Vanlauwe, B., P. van Asten, E. Kahangi, N. Sanginga, T. Losenge. J.M. Jefwa J.M., Ruto. Banana
fertilizer application and their effects on growth, flowering and production in an on-station
trial in Maragua district of Central Kenya. Accepted for African banana conference,
Mombasa, October 2008.
Vigheri, L.N., L. Lukalango, K.Sikyolo, N. Dowiya, P.J.A. van Asten, G. Blomme. Etude du
germoplasme des bananiers et bananiers plantains en territoire de Beni et de Lubero.
Vigheri, L.N., L.K. Mbogho, K.Sikyolo, G.Blomme, P.J.A. van Asten. Contribution à l’etude du
maladies et ravageurs du bananiers et bananier plantain en territoire de Beni et de Lubero.
Vigheri, L.N., S. Saghasa, K.Sikyolo, G. Blomme, P..J.A. van Asten. La place du bananier et
bananier plantain dans les systèmes de cultures en territoire de Beni et Lubero, Province du
Nord Kivu, R.D. Congo. Accepted for African banana conference, Mombasa, October 2008.
Wairegi, L.W.I., P.J.A. van Asten, C. Kiwanuka, M. Tenywa, M. Bekunda. 2007. Assessment of
soil management practices in East African highland cooking banana (Musa spp. AAA-EA)
systems in Uganda. Paper presented in ‘International Symposium on Innovations for a Green
Revolution in Africa: Exploring the Scientific Facts’, 17-21 September, 2007.
Wopereis, M, K E Giller, A Maatman, B Vanlauwe, A Mando, A Bationo 2006 Innovations for
increasing productivity through improved nutrient use in Africa. Proceedings of the World
Congress of Soil Science, July 2006, Philadelphia, USA.
Articles in books
Panis B. and Lambardi M., 2006. Status of cryopreservation technologies in plants (crops and
forest trees). Chapter 6. Ruane J., Sonnino A. (ed.). The role of biotechnology in exploring
and protecting agricultural genetic resources. Food and Agriculture Organization of the
United Nations, Rome, Italy: 61-78.
Pérez Hernández J. B., Remy S., Swennen R. and Sági L., 2006. Banana (Musa sp.). Wang K. (ed.).
Methods in Molecular Biology, vol. 344: Agrobacterium Protocols 2/e, volume 2. Humana
Press Inc., Totowa, NJ:167-175.
Roux N., Strosse H., Toloza A., Panis B. and Dolezel J., 2005. Potential of flow cytometry for
monitoring genetic stability of banana embryogenic cell suspension cultures. Chapter 25.
Hvoslef-Eide A. K., Preil W. (ed.). Liquid culture Systems for in vitro Plant Propagation.
Springer, 337-344.
Theses and other reports
David, S., 1998. La production de semences de haricot: Manuels pour les producteurs de
semence de haricot sur une petite échelle. Manuel 1. Réseau de la Recherche sur le Haricot
en Afrique, Séries de publications occasionnelles, N°29. CIAT, Kampala, Ouganda.
Germeau, G., 2006. Identification des contraintes en culture bananière traditionnelle dans trois
régions du Rwanda par enquête diagnostic. MSc thesis, Université Catholique de Louvain,
113pp.
Nibasumba, A. 2007. Garniture Cationique des sols et des racines dan des systèmes de culture
bananière du Burundi et du Rwanda. MSc thesis, UCL, Belgium.
Rishirumuhirwa, T., 2006. The role and management of bananas in Burundian farming systems.
Consultancy report for the DGDC-funded IITA-led project on bananas in the Great Lakes
region. 52pp.
175

Progress Report

Transcription

Documents pareils

piekkick st-piekke

fiche informative - Ville de Kirkland

les vacances de mr bean les vacances de mr bean

tourin a la tomate cote de porc haricots blancs fromage

Brunch - Experience Old Montreal

Sosie Mylene farmer

achierpointcom

Serge Reggiani / Mylène Farmer

Mylène Farmer - avant que l`ombre - Bercy 2006

Les vacances de Mr. Bean