88

IntroductionHambantota District is a major paddy (rice)producing area in the southern coastalregion of Sri Lanka. Communities here havebeen experiencing salt water intrusion intotheir rice fields, leading to reduced yields.This has been caused largely by seawatercontamination of irrigation systems.Farmers had been using saline-resistantvarieties developed at the rice researchstation, Ambalantota, but in badly affectedareas these varieties were still failing. Thisarticle focuses on work done to identifytraditional rice varieties suitable for culti-vation in the degraded paddies through aprocess of participatory research.

Background The activities described here were part of athree-year project (2005 to 2007) carriedout in four countries of south Asia by Prac-tical Action, a UK-based internationaldevelopment organisation. The work in SriLanka was implemented by Practical

Action’s south Asia office based inColombo. Practical Action had funding todevelop a pilot project. Its objective was toincrease the resilience of communities tocope with the impacts of increased climatevariability in their localities.1 In Sri Lanka,these were the likelihood of climate changeworsening existing problems of salinisationand flooding.

The project worked with 500 house-holds in Godawaya and Walawa GramaNiladhari divisions in Hambantota districtalong the Walava river estuary. Mostpeople’s assets comprise physical capital:crop harvests, livestock, boats, and otherequipment.

The villages were selected to participatefor several reasons:• The majority of people were relativelypoor – farmers owned typically less thanone hectare of paddy land.• The site lay close to an area where Practi-cal Action was involved in a post-tsunamireconstruction programme, so manage-

1 Funding was provided by the UK-based Allachy Trust.

by RACHEL BERGER, with ROHANA WEREGODA and VARUNA RATHNABHARATHIE

6Participatory rice varietyselection in Sri Lanka

89Participatory rice variety selection in Sri Lanka

ment and travel costs could be reduced. Theproject site was sufficiently far inland not tohave been directly impacted, though fisher-men’s livelihoods clearly were affected.• The problems faced by the community(coastal erosion, saline intrusion intopaddy fields) were typical of those likely tobe experienced by rural Sri Lankancommunities under climate change, even ifthe current problems cannot be directlyattributable to climate change.

Paddy cultivation in Sri Lanka is largelya smallholder activity. Rice growing isimportant culturally for smallholderfarmers. So even when faced with problemsof salinity, farmers were reluctant to switchto other crops.

Salinisation and soil erosion have led tothe abandonment of paddy land along theriver estuary. Salts drawn up from ground-water are deposited on the surface, and

spread as poorly maintained irrigationchannels become polluted from the salineestuary. Coastal erosion also reduces theavailability of land and forests that areincreasingly – and illegally – used by localpeople as the scarcity of productive farm-land increases across the district.

In future, storm surges and rising sealevels are likely to increase salt watercontamination. Combined with intenserainfall these problems are anticipated toincrease under current climate changepredictions. The main focus of the RiceResearch Institute in Sri Lanka is onlarger-scale irrigated paddy cultivation.However, it has developed a few varietiessuitable for saline conditions faced bysmall-scale farmers, although without theinvolvement of farmers. In the experienceof farmers in the project area, the varietiesonly tolerate limited levels of salinity.

Traditional paddy trial locations in Sri Lanka.

Sour

ce: h

ttp:

//im

ages

.goo

gle.

lk/im

ages

60 � Berger, Weregoda, and Rathnabharathie90

Methods and processesUnderstanding the situation faced by theHambantota communities was essentialfor designing project activities. Gatheringsufficient information was critical. In thefirst six months, we used a range of datagathering approaches: • Secondary data. To select project sites,we used data from e.g. the Sri Lankancensus and statistics and International

Union for Conservation and Nature(IUCN) reports on coastal recourses. Thisshowed that the Ambalantota division ofHambantota district was highly vulnerableto floods, sea water intrusion, and naturalresource depletion.• Focus group discussions. These wereused to initiate discussions in the villagesand collect background information to vali-date the secondary data. We held two

There were no records of any kind held at the local level. Inhabitants’ memories were the only source ofinformation. Groups were formed with participants of similar ages to enable discussions amongstindividuals with similar knowledge. Participants were grouped into those over 60 years old, between 45 and60 years, and the younger generation (below 45 years). Women were included, since their work of firewoodcollection, farming, and homestead vegetable gardening made them more aware of changes in resourceavailability and temperature than men.

While people had access to modern weather information, farmers still used a forecasting system known as‘Litha’. Litha is a table prepared by an astrologist, using the phases of the moon and the positions of thestars and planets. A full moon, for instance, is associated with rain. The table indicates times when seeds willgerminate, or the best time to plant crops so as to avoid pest attacks. However, whilst respondents reportedthe Litha as still in use, it was judged to have become less effective in recent years. This may suggest thatclimate change is undermining traditional techniques – or perhaps only that modern forecasts areincreasingly reliable.

A variety of other traditional forecasting indicators were used (see table below). Their effectiveness wasbeing debated within the community. It was also reported that traditional techniques are not being passedon to the younger generation, who prefer to rely on modern farming and forecasting practices.

Box 1: Community perceptions on changing climate

Observation

Large termites start breeding during a dry period..Ants appear with their eggs and move to a newnest.

Small termites start breeding during a rainy period.

A noise is heard emanating from the sea.

Off-season trees such as tamarind and wood applegive good yields.

Dogs and cattle make unusual sounds.

Prediction

Rain will come soon.

Rain will start within 24 hours.

Rain will stop soon.

Rain will come within seven hours and last for sevendays.

Good future rainy season – farmers cultivate largeareas.

Destructive rainy season leading to disasters isanticipated.

Table 1: Traditional forecasting indicators

2 The representatives were the Assistant Divisional Secretary (who reports to the DistrictSecretary, a government appointee, via the Divisional Secretary), local governmentofficials from the area, village farmer group leaders, and officers of the Rice ResearchInstitute in the area.

91Participatory rice variety selection in Sri Lanka

meetings with government representativesand non-governmental and community-based organisations to discuss problemsrelevant to livelihoods and to understandresource use and vulnerability.2 This led tothe selection of two paddy farmingcommunities (from Manajjawa andWalawa) for further vulnerability assess-ment. One key issue discussed with focusgroups in the community was knowledgerelating to weather and climate. This issummarised in Box 1. • Community resource mapping. Follow-ing the focus group discussions, we intro-duced resource and risk mapping toidentify the natural resource base andmanagement approaches. Selected farmerorganisations and other communitymembers drew maps of communityresources, illustrating natural resourcesimportant for livelihoods such as paddy

lands, home gardens, plantation lands,settlements, and water bodies.• Risk mapping. The groups involved inrisk mapping included community officebearers (such as the president of the villageagricultural society) and members offarmer organisations. Using the commu-nity resource maps, the groups identifiedareas at risk from salinisation and flooding,water-scarce areas, and irrigation canals.The main issues identified were flooding,salinity, conflicts between water resourceusers, and risks due to changing land usepatterns (the conversion of coastal foreststo paddy land). Participants were not askedto rank risks. • Field observations and transect walks.Practical Action project staff joinedcommunity members in village visits tocomplement the results of the focus groupdiscussions and mapping exercises.

Participatory method used

Key informant discussions

Focus group discussions

Community resourcemapping

Risk mapping

Field observation

Transect walks

Objective

To collect information on naturalresources, livelihoods, and social andinstitutional arrangements in the area.

To collect background information onresource use, vulnerabilities, andcapacities to validate secondaryinformation.

To identify the natural resource baseand natural resource managementmethods.

To identify risks to livelihoods withinthe area.

To further understand biophysicalresources, vulnerability, and existingproblems affecting farmers'livelihoods.

To further understand biophysicalresource vulnerability and existingproblems affecting farmers'livelihoods.

Information gathered

• Sources of livelihood• Use of natural resources• Existing socio-economic systems

• Vulnerabilities • Livelihood system problems

• Vulnerabilities• Availability and use of naturalresources • Area landscape

• Types and causes of risks andthreats • Extent of vulnerable areas

• Risks and threats in relation tobiophysical resources • Encroachments • Severity of problems identified

• Linking upstream anddownstream farmers• Variation in vulnerability

Table 2: Summary of the data-gathering methods

60 � Berger, Weregoda, and Rathnabharathie92

Most families were members of the fish-eries or farming society when the projectbegan. Farmer organisations in each villageact as the decision-making forum for issuessuch as the farming calendar, water distri-bution, and seed selection. In preparingpaddy land for sowing, all farmers need towork the land together, since channelsbetween fields are interconnected. Thewhole system will not operate until all fieldsand the dividing mud walls are ready.These farmer organisations are also linkedwith government agrarian services centres.Farmers also reported being involved ininformal institutions such as the Yaya(field) groups and Seettu (community-managed revolving fund) groups. In theproject villages, association members areprimarily small-scale farmers.

The participatory process sought toprioritise the main threats to people’s liveli-hoods that could be attributable to increas-ing climate variability. The communityreached a consensus during the participa-tory assessment process that floods(leading to crop losses) and salinisation(low yields and crop damage) are the mainthreats to livelihoods. They expressed aclear desire to reduce the risks of cropfailure and unprofitable yields.

After the survey, we developed a strat-egy to build capacity around resource useand management and, in particular, toestablish and replicate best practices incrop cultivation. The main vehicle for train-ing was the paddy farmer group. The groupwas mobilised to adopt and developsustainable farming practices, such ascompost making from rice husks. However,while farming activities are carried out bymen and women, the group trialling therice varieties was dominated by men. Insmall-scale paddy farming in Sri Lanka,women and men carry out separate tasks:men prepare the land, and the watercontrol mechanisms; women plant

seedlings and weed crops. Pesticides, whenused, are applied by men, and harvesting iswomen’s work. In most households, mentraditionally make many of the farmingdecisions, though there are women whotake on this role. Women’s time commit-ment to household activities was an obsta-cle to attendance at meetings. The projectteam made efforts to accommodate womenby finding alternative meeting times, butdid not challenge the cultural norm ofmale-dominated decision-making.

The paddy farmer group consisted of 16farm families. They volunteered to set asideland in their own fields for the trials. Thefarmers had some success with saline-resis-tant varieties developed at the Ambalan-tota Rice Research Institute. However, inbadly affected areas the resistant varietieswere still failing.

The farmer group agreed to trial sometraditional rice varieties alongside theresearch station varieties. The trial wassupported by the National Federation forConservation of Traditional Seeds andAgricultural Resources (NFCTSAR), anon-governmental organisation, whichsupplied farmers with traditional seeds.3

The Federation selected ten varieties thatit thought would tolerate the level of salin-ity present. The varieties had differentcharacteristics in terms of growth, yield,and flavour.

In the first year, both conventional andtraditional varieties were grown. Theconventional varieties were treated withchemical fertilisers and pesticides and thetraditional varieties with home-madeorganic compost and neem oil.4 A furthertrial, in the next rice growing season, wasof different varieties of traditional rice.Each variety received the same treatment– neem oil and organic compost. Thefarmers planted up to 5kg of each varietyin the saline affected areas of their paddyfields and continuously observed the

3 Local seed banks had fallen out of use following the introduction of high-yieldinghybrids in the 1960s and 1970s.4 Homemade neem oil is used as a pest-control.

93Participatory rice variety selection in Sri Lanka

growth and changes in the plants up toharvesting. Two farmers out of 16 culti-vated all 10 varieties while the rest culti-vated three or four varieties after seeinghow the 10 varieties performed in the firsttrials.

Which varieties to grow on a largerscale were chosen by scoring each varietyfrom one to 10, based on the farmer’s pref-erence, where one was the best score. Thefarmers chose the criteria e.g. plant height,duration of the crop, grain quality, graincolour, saline tolerance, and yield. Theselection process was supported by activi-ties to help improve soil quality (muchdegraded in the saline-affected areas) andsoil water retention, by using organicmulches. Mulches reduce the amount oftilling required, which in turn reduces soilsalinity caused by bringing salts up fromdeeper soil levels.

Outcome of the field trialsThe results clearly showed improved yieldsfrom the trial varieties. The traditional ricevarieties also carried a premium marketprice of up to 50% per kilogram compared

to hybrid varieties. After seeing that somevarieties performed well in saline soils thefarmers developed confidence in theirresearching and selecting varieties. Prior todeveloping this network the role of the insti-tute was to provide seed to the farmers withlittle or no interaction. During the project,the community were able to form good linksbetween the farmer organisations and thegovernment Rice Research Institute. Theproject also facilitated contact with localgovernment via the Grama Niladhari(village-level administration) and the irri-gation department. This prompted impor-tant actions such as sluice gate renovation(to prevent seawater flowing into the paddyfields) and support in improving freshwa-ter irrigation. The farmer groups alsoextended their network to other farmers inthe area. Farmer-to-farmer learning was amajor development. Farmers testing tradi-tional varieties shared their findings withfarmers from the adjacent village who facedsimilar problems. This stimulated discus-sions between farmers on variety selection,increased the knowledge of neighbouringfarmers on climate change issues, and

Meetings with the community (Bundala).

Phot

o: V

arun

a Ra

thna

bhar

athi

e

60 � Berger, Weregoda, and Rathnabharathie94

enabled the farmer groups to becomechange agents at the community level. Bythe end of the project a traditional paddycultivator group had been established andwas formally attached to the seed conserva-tion organisation, providing a widernetwork of contacts and support for theHambantota farmers and helping sustainthe project developments once PracticalAction withdrew.

Lessons learnt, critical reflections, andanalysis

Methodology for collecting informationThis article was written after the projectended and after a change in project staff,resulting in some loss of institutional knowl-edge. Writing it has highlighted some insti-tutional weaknesses within Practical Action:while there is certainly a high commitmentto participatory processes, there is no formalacross-organisation manual for training in

this area. Evaluations of the organisation’swork are generally very positive, but there isclearly room for improvement in the amountand depth of information gathered duringthe initial stages of project development. Forexample, very little information wascollected on social class or power issues inthe community, nor on land tenure – possi-bly because the situation was so typical of theregion, and so familiar to the projectmanagers that it did not seem worth record-ing. This was a small project, with limitedfunds. The initial activities, including thebaseline study and PRA were limited inscope. It can certainly be argued thatimproving rice yields under salinisationwould need to be part of a much broaderstrategy for adaptation, for which the projectfunding was inadequate to develop.

Value of the processImproved varieties are frequently intro-duced by research stations, but are not

‘Fertiliser pond’ in a paddy slot to provide liquid fertiliser (Bundala).

Phot

o: V

arun

a Ra

thna

bhar

athi

e

95Participatory rice variety selection in Sri Lanka

always suitable for use in different localcontexts. The formal research process failsto address farmer needs, does not buildtheir decision-making capacity, and ignoreslocal knowledge in the selection process.Participatory variety selection aims toaddress these shortcomings. In this smalltrial, the process was successful. It builtconnections between farmers’groups andthe Rice Research Institute. These connec-tions should be sustainable, and may leadto a more participatory approach withinthe Institute’s main research programme.

The trial was small-scale, due to limitedresources. The method was weakened bynot having control plots where conven-tional and traditional varieties bothreceived either chemical or organic treat-ment. This has since been rectified insubsequent replication in areas alsoaffected by salinity, where the results havealso been favourable. The methods andprocess would need wider testing before

concluding on the success of the approach,or the sustainability of the varieties selectedby the farmers. The yield of traditional vari-eties on average has proved slightly lowerthan conventional varieties. However,traditional varieties are more profitablethan conventional varieties. They are inshort supply, and are prized for consump-tion during festivals, for medicinal reasons,and for their taste and aroma. A simplecost-benefit analysis shows that despite theneed for greater labour input (to makecompost and neem oil), it is cost effective,because labour comes from family andneighbours. Women can also participatemore fully in the farming calendar whereorganic products are used.

A key contribution was in raisingawareness of local solutions to climatechange threats. Forgotten varieties ofindigenous rice were shown to offer a solu-tion to the increasing soil salinity. There arearound 2,000 traditional Sri Lankan rice

‘Pokkali’ a traditional rice variety grown in extremely saline conditions in Bundala.

Phot

o: V

arun

a Ra

thna

bhar

athi

e

60 � Berger, Weregoda, and Rathnabharathie96

varieties, many with high nutritional valueand medicinal properties, or resistant toparticular diseases and pests. Cultivatingthem has helped a group of marginalisedfarmers to increase their harvests, whilstthe project’s collaborative approach hashad a positive impact on the attitude oflocal agricultural institutions. The partici-patory approach to variety selection over-came the limitations of the conventionalresearch system in meeting the needs of

marginalised farmers and integrating localknowledge into the selection process.

ReplicabilityOther farmers who cultivated saline-affected lands nearby were invited to visitthe trials and discuss progress with thefarmers involved, and this led to wideradoption. Following the trials, at a meetingin Hambantota, officials from the agricul-tural department and government

A woman farmer shows her flourishing trial plot of traditional rice varieties.

Phot

o: V

arun

a Ra

thna

bhar

athi

e

97Participatory rice variety selection in Sri Lanka

researchers shared findings and discussedthese in the context of the impact of climatechange on farming in the region. Videodocumentaries have also been shown toagricultural extension officials at workshopswhere ministry officials and PracticalAction staff presented their differentapproaches. They have been shown tofarmer associations in each district, whohave also taken part in visits to successfulproject sites.

Practical Action has since expanded theapproach to many other communitiesaround Sri Lanka (see Map 1). Significantly,other communities have adopted itautonomously, through word of mouth andobservation. Traditional varieties are grownby one farmer in a corner of a large field,cultivated by many small-scale farmers. Thenext year, many other farmers are observedgrowing the crops. Farmer societies contactPractical Action to find out more, and arereferred to villages adopting the practice,and organise their own visits. Staff havebeen surprised by contact from people indistant places who have heard about thetrials. This autonomous replication of asuccessful practice is very encouraging,though of course difficult to assess withoutindependent evaluation.

Uptake is feasible, though farmers areprobably not undertaking rigorous trialsbefore adoption. The methodology ishowever replicable, provided farmers’groupsare adequately supported in understandingthe need for a rigorous approach. Supportcould be provided by the national RiceResearch Institute, but the mediation of anon-governmental organisation experiencedin participatory methods would increase thelikelihood of success.

A more farmer-focused agriculturalresearch agenda is crucial for farmerssuccessfully to adapt their food productionto climate change. At present, researchpriorities are defined by national govern-ment or even multi-national corporations.The likelihood for change on a large scale isremote at present.

Adaptation to climate change is highlycontext specific. At present, salinity may notaffect the majority of the area under ricecultivation. Nationally, there is no estimateof fields affected by salinity and salt waterintrusion and the likely causes. At a locallevel however, the number of communitiesthat have raised concerns with PracticalAction project staff suggests this issue maybe seriously underestimated. Communitiesthroughout a wide geographical area arekeen to experiment with saline-resistanttraditional varieties of rice. This attests tothe relevance of this approach and technol-ogy – particularly to communities inmarginal locations facing many social andeconomic challenges, in addition to theadded challenges of soil salinity and futureclimate change.

Conclusion The participatory research approachadopted during this project demonstratesthe importance of experimentation foradaptation. Participatory research providedthe farmers with a supportive environmentwith access to the resources necessary forexperimentation. It demonstrates the effi-ciency of locally informed and farmer-ledresearch. Farmers were able to assessthreats to their livelihoods and define theirown response. Their capacity to experimentin finding solutions to their own problemswas strengthened. The network of relation-ships built up between the farmers, formalresearch institutions, government extensionservices, and a national seed conservationbody will hopefully ensure continuity ofsupport for further research by the farmers.

This project underlines the importanceof biodiversity for adaptation. The farmersof Hambantota could assess and selectseeds able to survive in the degraded envi-ronment only through the preservation andfree availability of many seed varieties. Asclimate change pressures mount, it will becrucial to ensure that agricultural biodiver-sity is not sacrificed for short-term increasesin yield.

60 � Berger, Weregoda, and Rathnabharathie98

CONTACT DETAILSRachel BergerClimate Change Policy Adviser Practical Action Schumacher Centre for Technology and Development Bourton Hall Rugby CV23 9QZ UK Tel: + 44 1926 634400 Email: rachel.berger@practicalaction.org.ukSkype: rachelberger1 Website: www.practicalaction.org.uk

Rohana Weregoda3/44, Brewery LaneArmidaleNSW 2350AustraliaEmail: rohanlw@yahoo.com

Varuna RathnabharathieProject Manager, Organic AgriculturePractical Action 5 Lionel Edirisinghe MawathaKirulaponeColombo-5Sri LankaTel: +94 11 2829412 ext: 19 or 67Email: varuna.rathnabharathie@practicalaction.org.lkwww.practicalaction.orgwww.janathakshan.org

ACKNOWLEDGEMENTSThis article is based on material prepared by Rohana Weregoda,formerly Project Manager with Practical Action’s south Asia office inSri Lanka and Varuna Rathnabharathie, who is currently managingprojects including traditional rice cultivation for Practical Action.