Putting farmers first in chickpea selection in Bangladesh

In Bangladesh, new crops are already tested on farms as well as on research stations. Even so, farmers often don't get what they want or need. Now, farmers themselves decide what works best for them on their farms under normal farming conditions. Involving farmers in selecting and testing improved varieties, known as 'participatory varietal selection', has proven successful with chickpea in Bangladesh's High Barind Tract. Here chickpea is becoming popular but yields are often less than a quarter of what they could be. Farmers who tested and adopted new varieties already harvest more grain. Plus, they can get an extra crop from land that previously just lay fallow after the annual rice crop. Involving farmers in selecting varieties has great potential, so efforts need to be focused on making this approach more widely accepted in Bangladesh.

Project Ref: PSP11:
Topic: 1. Improving Farmers Livelihoods: Better Crops, Systems & Pest Management
Lead Organisation: CAZS-NR, UK
Source: Plant Sciences Programme

Contents:		Description
		Validation
		Current Situation
		Lessons Learned
		Impacts On Poverty
		Environmental Impact
		Annex

Description

Research Programmes:

R8269

• UK
  • CAZS Natural Resources, Bangor UK (Dr Dave Harris)
• Bangladesh
  • PROVA, Rajshahi, Bangladesh (Mr A. M. Musa)
  • Pulses Research Centre (PRC), BARI, Ishurdi
• India
  • ICRISAT, Patancheru, AP, India (Dr J.V.D.K. Kumar Rao)

Chickpea is becoming a popular and profitable crop to replace fallow in the High Barind Tract (HBT) of Bangladesh following the harvest of rainfed aman rice. The potential yield of chickpea, with minimal biotic and abiotic constraints, in the HBT is >2.5 t ha^-1 but farmers' yields are usually in the range of 0.5-1.0 t ha^-1. A series of Mother Trials and Baby Trials was conducted in 2003-04 and 2004-05 to compare the performance of, and obtain farmer feedback on, a range of chickpea genotypes under farmer-managed conditions in farmers' fields and to assess whether varietal and trait preferences varied across the HBT.

Using feedback from farmers, it was possible to assemble a farmer-researcher ideotype for chickpea improvement in the HBT. Ideotypes, or ideal plant types, have previously been used by plant breeders and physiologists to summarize breeding objectives, and to specify the traits that need to incorporated into existing varieties in order to breed superior varieties. Use of Participatory Varietal Selection (PVS) broadens the ideotype concept by also incorporating the ideas of the major clients, the farmers. Thus, to improve upon BARI chola 5 (the current most popular improved variety grown in the HBT), the following characteristics are needed:

• Shorter duration, to escape terminal drought and heat stress that occurs from early March, but without reduced yield potential;
• Cold tolerance to permit earlier flowering and pod set, and hence earlier maturity;
• Improved ability for seedlings to establish at low seedbed moisture;
• Deep rooting to capture moisture;
• Greater resistance to collar rot (Sclerotium rolfsii) and fusarium wilt (Fusarium oxysporum);
• Greater resistance to Helicoverpa pod borer; and
• Grain characteristics resembling the old variety Nabin (BARI chola 1).

Types of Research Output:

Chickpea is the main commodity in this document, but the PVS methodology is applicable to all crops.

Production Systems:

Farming Systems:

This output could be clustered with any crop genetic improvement projects, and projects involving farmer participation in improvement of agronomy, such as:

• CPP, Chickpea ICM, R8427, R8366, R7885
• CPP, Cost effective weed management packages for lowland rice in Bangladesh, R8412, R8234, R7471
• PSP, Chickpea varieties for Western India, Prog. Dev.
• PSP, Rice fallow rabi cropping systems, R8098, R8221

A client oriented breeding (COB) approach (e.g. PSP, Concepts and approaches of COB; PSP, COB horsegram and chickpea, Eastern India,) is recommended for further improvement of chickpea varieties for the HBT with:

• farmer involvement in selection of entries and segregants;
• on-farm, rather than research station, evaluation of entries and progeny;
• environmentally targeted to variations within HBT, including geographical (north, south) and location on the toposequence;
• continuous farmer interaction and feedback from parent selection to varietal release.

Validation

It has been argued, and demonstrated for several crops, that adequate consideration of client-orientation is necessary for efficient and effective identification of improved crop varieties that will be adopted by resource poor farmers (Stirling and Witcombe, 2004; Witcombe et al, 2005). We thus used participatory varietal selection (PVS) techniques to guide chickpea introduction and breeding programmes appropriate for the HBT. The prime aim was to determine farmers' preferences for traits and varieties to ensure client orientation.

PVS of chickpea in the HBT of Bangladesh was validated by an NGO, PROVA, with collaboration of BARI and DAE. In Mother Trials, sets of varieties were sown in farmers' fields in 8-12 dispersed replications. Measurements included farmer assessment of varietal and trait preferences both pre- and post-harvest, and recording of yields. Baby Trials (77 in 2003-2004 and 50 in 2004-2005) involved disbursement of seed lots (2.5 kg) of improved chickpea varieties for farmers to test against the "local" variety they would normally use, in operational scale plots using farmers' own inputs. Data collected included  post harvest assessment of farmer preferences for traits and varieties, and recording of farmer-measured yields. Data obtained over space and time were assembled by PROVA to formulate "ideal" plant types to guide future plant introduction and breeding programmes that would target all chickpea growers of the HBT.

In 2003-04 (Table 1) BARI chola 4, 5 and 7 were most preferred for pre-harvest traits, followed by BARI chola 2 and Annigeri, with BARI chola 8 and Local least preferred. For post-harvest traits, BARI chola 2, 4, 5, and 7 were equally preferred, with Annigeri intermediate and BARI chola 8 and Local least preferred. Similar information was obtained in 2004-05 and the farmers' expectations of yield in both seasons corresponded with the actual plot yields measured

Table 1. Pre-harvest and post-harvest assessment of chickpea Mother Trials in the HBT, 2003-04 ("1" = worst; "7" = best).

[1] Differences between varieties significant at P<0.001 for each trait

In Baby Trials, farmers showed an overwhelming preference for the improved variety over "Local" or Nabin for most traits. Table 2 presents an example, for the comparison between BARI chola 2 and farmers' 'local' in 2003-04. In 2004-05, when a test variety (BARI chola 7 or Annigeri) was compared with BARI chola 5, farmers ascertained little difference in ranking between varieties.

Table 2. Preferences of 21 farmers who evaluated BARI chola 2 in Baby Trials in the HBT in 2003-04.

The output was validated in the HBT of Bangladesh over two rabi seasons, from 2003-04 to 2004-05, in the rainfed rice-chickpea cropping system (semi-arid, wetland rice based, smallholder dry).

Current Situation

Although improved varieties of chickpea (e.g. Bari chola 2, 5) are being grown, and seed saved, by farmers in the HBT as part of an ongoing initiative to promote double cropping (see PSP, Rice fallow rabi cropping systems), the ideotype itself is not currently being used because PROVA is not proceeding with chickpea breeding and introduction programmes and the Pulses Research Centre (PRC), BARI, does not appear ready or willing to adopt the methodology.

Neither the output nor the PVS methodology is currently being used for chickpea in the HBT due to lack of resources of the NGO. More exposure of PRC personnel to PVS is necessary before GOs in Bangladesh are likely to adopt the approach.

Use of the PVS methodology for chickpea has been suspended until a project that will support follow-up chickpea introduction and breeding using PVS techniques is identified.

In Bangladesh, on-farm trials are called farmer field variety trials and demonstrations and minikit trials. In all these programmes, run at the district and village level, farmers are given seed of new varieties to test under a package of practices. All these activities can be modified to accomodate PVS if capacity is strengthened in more farmer-oriented techniques. We have found in the linear research to extension system that it is extensionists i.e. the DAE who have assisted the most. The key factors in success of PVS elsewhere have been in demonstratijng that PVS works and communication of these results in carefully targeted workshops.

Policies are unhelpful for the adoption of PVS as release proposals give such a high emphasis to research station trials.  Active lobbying with policymakers for changes in policy is required so that on-farm participatory trials have equal status to research trials.

Lessons Learned and Uptake Pathways

Promotion is currently under suspension.

The plant breeders currently responsible for producing new varieties of chickpea in the HBT are not sufficiently convinced that PVS is a valid approach for a) promoting new varieties by increasing farmers' choices and b) generating relevant data to guide the breeding of potentially better varieties in the future.

It must also be recognised that legume breeding and agronomy has a low priority for GoB, so few resources are available. However, given the recent success in promoting double cropping, largely based on chickpea, in the area it may be that new varieties of chickpea and other 'orphan' crops will have a higher profile and that PVS will become more attractive.

Chickpea is an 'orphan' crop and seed can only be produced in the rabi season for the following rabi season so expensive seed, that is vulnerable to storage pests, has to be stored throughout the rainy season. It also has a low seed multiplication rate and the high value and high volume (bulk) of chickpea seed increases the investment and costs of seed storage. Nevertheless, it is a very popular and profitable crop to grow after rice when few other crops will grow.

Consequently, there is a need to strengthen the capacity of existing networks for raising awareness of the benefits of new chickpea varieties (and the PVS technology) in the following areas:

• Capacity building by training of GOs, NGOs and farmer groups.
• Creating awareness of the role of legumes in human and animal nutrition and cropping sequences for maintaining soil fertility.
• Creating awareness of the new varieties through meetings with stakeholders, demonstrations and publication of literature.

Finally, convincing data are required that can be used in sensitization and training programmes. This will entail an ongoing field-based PVS programme.

Successful case studies and their widespread and attractive presentation to stakeholders is the key to achieving impact with large numbers of poor people.

Using Rogers' (2003) five categories to characterise diffusion of information as a framework for the lessons learnt:

1. The relative advantage of a technology compared to what it is replacing; This is high. Replacement of chickpea landraces and indigenous varieties can increase grain yield and profitability and provide options for cultivating previously-fallow land (see PSP,  Rice-fallow rabi cropping systems).
   
2. The compatibility of the technology with existing systems and ways of doing things, which is closely related to culture; In the HBT, cultivation of chickpea is a relatively recent phenomenon, so compatibility of new varieties is medium/high, but becoming higher as farmers become more familiar with the crop. It is worth noting that PVS involves farmers actually growing the crop, thus automatically increasing familiarity.
   
3. The complexity of the technology in terms of what people need to learn to make it work; The complexity is low/medium. For current growers, only replacement seed is necessary, without other management changes. Chickpea is becoming widespread in the HBT, partly because it needs few inputs and is thus relatively easy to grow, so new growers are likely to have been exposed to some degree and may be particularly receptive to additional extension efforts.
   
4. The observability of a technology in terms of how easy it is to demonstrate and observe performance; The observability of the benefits of new varieties is very high when PVS is used  because side-by-side comparisons are easily made in farmers' own fields.
   
5. The trialability of a technology in terms of how easy it is to test it before deciding to adopt. The trialability is very easy as long as seed is available.

Impacts On Poverty

Poverty Impact Studies: 

It is too early for impact analysis of the benefits of new chickpea varieties in relation to those currently grown in the HBT. However, assuming that input costs are approximately similar for old and new varieties, the fact that e.g., Bari chola 5 outyields 'local' by almost 35% means that appropriate new varieties are likely to be more profitable. Even where the cost of seed of the new variety is more than the local, extra yield and more market-friendly characteristics (resulting in a higher sale price) can result in better net returns from the new one (Table 3).

Table 3. Comparison of returns from a short-duration chickpea variety ICCV 2 and a local variety (Data from CRS, Satna, M.P., India).

(1 US$ = Rs. 45.5 approximately)

Two studies of the impact of chickpea as a new crop, but without explicit consideration of varietal effects, in the HBT are available:

Saha, A.K. (2002). Impact assessment study for the DFID-funded project R7540 'Promotion of Chickpea following Rainfed Rice in the Barind Area of Bangladesh'. CAZS Natural Resources, University of Wales, Bangor, UK.

Socioconsult (2006). Report on Impact Assessment Study of Chickpea in the High Barind Tract (HBT). Socioconsult Ltd., SEL Centre, 29 West Panthapath, Dhanmondi, Dhaka.

How the Poor have Benefited (including gender and other poverty groups):

Chickpea itself is the most consistently profitable crop that can be grown without irrigation after rice in the HBT (Saha, 2002; Socioconsult, 2006; Table 4 below). Evidence from elsewhere (Joshi and Witcombe, 1996) has shown that there can be large increases in total productivity and other advantages for resource-poor chickpea farmers who adopt new varieties through PVS. Participatory identification of ideotypes that inform successful breeding of additional new varieties can lead to a virtuous cycle of PVS-COB-PVS that can impact positively on the livelihoods of poor people.

Table 4. Input costs and profitability (Taka/ha) calculated for mean and maximum yields of rainfed rabi crops in the HBT in the 2003-04 and 2004-05 seasons.

Environmental Impact

Direct and Indirect Environmental Benefits:

Use of PVS techniques that lead to higher and more stable chickpea yields would encourage further chickpea cultivation and the environmental benefits thereof, through increased "legume effects" such as longer-duration ground cover to minimise soil erosion, better soil health and increased fertility. Extra, nutritious  fodder from chickpea could reduce pressure on common grazing areas.

No adverse environmental impacts envisaged.

Lack of choice in life is a defining characteristic of poverty. The direct (higher yield, increased food security and income) and indirect (e.g., earlier harvest) benefits of better, more appropriate chickpea varieties give farmers and their families more choices and lead to more resilient livelihood strategies. In adition, regular use of PVS techniques would maximize chances of varietal adaptation to any climate change, thus permitting continued cultivation of chickpea and the benefits derived from that.

Annex

References

Joshi, A. & Witcombe, J.R. 1996. Farmer participatory crop improvement. II. Farmer participatory varietal selection in India. Experimental Agriculture 32:461-477.

Rogers, E.M. (2003). Diffusion of innovations. 5th Edition. New York: Free Press.

Stirling CM, Witcombe JR. 2004. Farmers and plant breeders in partnership. Second edition. Bangor (UK): Department for International Development (DFID), Plant Sciences Research Programme (PSP), Centre for Arid Zone Studies (CAZS), University of Wales. 38 pp.

Witcombe, J.R., Joshi, K.D., Gyawali, S., Musa, A. M., Johansen, C., Virk, D.S. & Sthapit B.R. (2005). Participatory Plant Breeding is Better Described as Highly Client-Oriented Plant Breeding. I. Four Indicators of Client-Orientation in Plant Breeding. Experimental Agriculture 41: 299-319.

Relevant Research Projects, with links to the
Research for Development (R4D) web site
and Technical Reports:

R4D	Project Title	Technical Report
R7471	Developing weed management strategies for rice based cropping systems in Bangladesh
R7885	Promoting the adoption of improved disease and pest management technologies in chickpea by poor farmers in mid hills and hillside cropping systems in Nepal
R8098	Promotion of rainfed rabi cropping in rice fallows of India and Nepal: pilot phase
R8221	Promotion of rainfed rabi cropping in rice fallows of eastern India and Nepal: Phase 2
R8234	Promotion of cost-effective weed management practices for lowland rice in Bangladesh
R8269	Improvement of rainfed cropping systems in the High Barind Tract of Bangladesh
R8366	Policy and strategy for increasing income and food security for poor farmers in Nepal and South Asia through improved crop management of high yielding chickpea in rice fallows.
R8412	Decision support frameworks for weed management in lowland rice in Bangladesh
R8427	Ensuring the sustainability of an integrated crop management approach to chickpea production for poor farmers through up-scaling and far-reaching adoption in Nepal.

For relevant research projects, with links to further information

Geographical regions included:

Bangladesh,

Target Audiences for this content:

Crop farmers,