From the Cornell International Institute for Food, Agriculture and Development (CIIFAD) Annual Report 1997-98

 

South Asia

Sustainability of the Rice-Wheat Cropping System

 

 

 

Through its participation in the Rice-Wheat Consortium for the Indo-Gangetic Plains, CIIFAD is trying to ensure the sustainability of the rice-wheat cropping system on which nearly 1 billion people depend for their staple food. CIIFAD, together with partner institutions in Bangladesh, India, Nepal and Pakistan, is working to identify and overcome soil-related constraints to crop production and increase the nutrient output of the cropping system to better meet human nutritional needs. A multidisciplinary approach, integrating technical and socioeconomic knowledge, is being taken to develop appropriate, acceptable interventions.

 

In 1997, the South Asia program completed the first year of a five-year program of collaborative research to mitigate soil-related constraints affecting South Asia's major cropping system. The rice-wheat system (RWS) involves growing paddy rice and wheat in alternating wet and dry seasons. It occupies 22.5 million hectares of land in the Indo-Gangetic Plains of South Asia and in China, and it supplies staple grains to about 20 percent of the world's population.

The rice-wheat program has support both from CIIFAD and the U.S. Agency for International Development (USAID) through its Collaborative Research Support Program (CRSP) on Soil Management under a five-year, US$3 million grant. During the past year, the CRSP program was constrained by sanctions against India and Pakistan. Our activities in those countries could only be continued at a low level with resources from CIIFAD. Fortunately, our program started it activities first in Bangladesh and Nepal, although they have been developing rapidly in India as well. We do not presently have activities in Pakistan as that country has had difficulties in maintaining its participation in the Rice-Wheat Consortium.

The Rice-Wheat Consortium, the network through which our program activities are carried out in the Indo-Gangetic Plains, also underwent changes during the year. The International Maize and Wheat Improvement Center (CIMMYT) replaced the International Crops Research Institute in the Semi-Arid Tropics (ICRISAT) as the international agricultural research center managing the facilitation unit responsible for program development and implementation. This unit maintains the structure necessary for a multi-country, multi-institution collaborative program.

Activities of the CIIFAD/CRSP rice-wheat program are reported here under the main headings of diagnostic activities, strategic research, technology adoption, and human resource development.

 

Diagnostic Activities

 

Surveys

 

Farm-level surveys previously carried out in the region addressed rice-wheat system constraints from a general agronomic perspective. Often a "laundry list" of problems was indicated, yet few data were available to quantify the magnitude, geographical extent, or connection with management practices. Several surveys were undertaken in Bangladesh and Nepal to address the lack of data on:

 

Additional surveys will measure the various constraints in rice production. All this information will be used to prioritize future research efforts.

 

Participatory Rural Appraisal

 

Until recently, little site-specific information was available on existing production practices, problems and potentials for improvement in Chuadanga and Dinajpur districts, even though they are significant areas of rice-wheat production in Bangladesh. Participatory rural appraisals (PRAs) were carried out in the villages of Sankarchandra (Chuadanga) and Jagdal (Dinajpur) with a multidisciplinary team of scientists from the Bangladesh Rice Research Institute (BRRI) and Bangladesh Agricultural Research Institute (BARI).

Although there were some limitations to the process (see pages 78-79), the PRAs offered the national agricultural scientists an opportunity for more direct contact with farmers as well as inter-institutional, multidisciplinary interactions. These represent advances in a system that tends to emphasize single-discipline, single-commodity research on agricultural experiment stations.

 

Solarization Follow-Up

 

Solarization involves heating the soil by covering it with clear plastic. Temperatures above 50øC generally provide good control of soil-borne pathogens. Solarization can also affect weed and insect populations and nutrient availability. Our purpose in using solarization is to determine the extent to which soil-borne pathogens limit yields of rice and wheat, or conversely, to see what yield gains could be realized if these biological, and possibly nutrient, problems in the soil can be overcome.

Solarization trials initiated last year on experiment stations at Nashipur and Joydebpur, Bangladesh showed positive effects on growth of wheat and summer (T. Aman) rice. In addition, residual effects of soil solarization prior to rice on the following wheat crop were observed at Nashipur (yields increased by 0.5 tons per hectare), but not at Joydebpur station. Improved growth of rice and wheat plants following solarization was attributed to observed increases in root health and decreases in root knot nematodes. Also soil inorganic nitrogen levels were consistently higher as was extractable manganese. Levels of other macro- and micronutrients did not appear to change with solarization.

In 1998, solarization trials with rice were initiated on 11 sites (nine on farms, two on station) in Bangladesh, at five sites (three on farms and two on station) in India, and on one farm in Nepal. In almost all cases, solarization improved plant growth. However, this does not always translate into increased crop yields because of greater pest pressures and greater susceptibility to lodging and loss of grain to rats in solarized plots.

 

 

Rice on a farmer's field in Joydebpur, Bangladesh showing differences between solarized (right row) and non-solarized (left row) treatments. Variation within each row represents seed fungicide treatments.

 

In Bangladesh, spring (Aus) rice yields on two farms at Joydebpur doubled with solarization, while at Chuadanga, yields were not significantly different. Trials with solarization prior to summer (monsoon) rice are mostly still underway, but yields were increased by 25 to 50 percent in three harvested trials in India.

Dramatic treatment effects between solarized and unsolarized main plots as well as with a split-plot treatment of ñ vitavax (a systemic fungicide) were also observed on a farmer's field at Joydebpur, Bangladesh. Further analysis is necessary to understand the causal mechanisms for observed plant growth differences. This information will help us to direct research to identify practical alternatives for using solarization to deal with specific constraints.

 

Strategic Research

 

Rice-Wheat System Diversification

 

Efforts toward diversifying the rice-wheat system currently involve: screening green manure species for non-nitrogen benefits, and relay planting mungbean into wheat.

 

Green manure screening

A legume screening trial was initiated in December 1997 on the Nepal Agricultural Research Council (NARC) experiment station in Khumaltar, Nepal. Fourteen legumes were grown during the winter months to compare biomass, nutrient extraction, and associated pest complexes (root pathogens and insects). These data will be used to select a subset of legumes to try within a rice-wheat rotation.

A second year of data was collected from the green manure screening trial being carried out at Ludhiana, India. Cowpea, mungbean, sunnhemp (Crotalaria juncea), guar (cluster bean), Sesbania, and pearl millet, were grown along with soybean and lablab (new for 1997) during the 60-day period between wheat and rice crops. The accumulated biomass material was incorporated into the soil, which was directly transplanted with rice.

Data collected to date from this screening experiment on agronomic factors, soil physical properties, root health, and nutrient extraction show significant differences among the species, especially in terms of nutrient extraction and subsequent crop root health. This kind of information should prove beneficial to farmers trying to solve site-specific problems with nutrient deficiencies (other than, or in addition to, nitrogen) or with soil pests. Future efforts will focus on soils with known nutrient deficiencies or root pathogen problems to evaluate both the growth of the green manure species themselves and subsequent benefits for crop production.

 

Relay planting mungbean into wheat

Mungbean is a popular high-value cash crop in Shankarchandra, a village in Chuadanga district, Bangladesh. Mungbean is normally cultivated after wheat is harvested; however, sowing after March 15 usually leads to enhanced vegetative production with reduced seed yield. Relay seeding mungbean into standing wheat could achieve more timely establishment. Experiments were conducted on farm fields in Sankarchandra and at the Joydebpur research station to evaluate the productivity of wheat and mungbean when mungbean is relay planted into wheat at 15 and 7 days before wheat harvest and at wheat harvest.

Results from Joydebpur indicate that mungbean relayed into wheat had no adverse effect on wheat yields (2.6 tons per hectare). As expected, mungbean yields decreased with a later relay-planting date, while above-ground biomass showed a reverse trend with planting date. Unfortunately a hail storm destroyed the mungbean crop at Shank-archandra, so productivity could not be assessed.

 

John Duxbury with Nur-E-Elahi (BRRI), and Craig Meisner (far left) evaluating mungbean relayed into wheat at the BRRI experiment station in Joydebpur.

 

While this experiment demonstrated the potential for mungbean relayed into wheat, the highest mungbean yields were still very low (177 kilograms per hectare). Possible explanations for the poor yields are soil nutrient deficiencies (see Wheat and Legume Sterility section) and inappropriate mungbean variety.

The current indeterminant, long-duration variety tested may not be the most appropriate for this application. Shifting to determinant, shorter-duration varieties might help to increase productivity in this system. A good possibility is the 60-day variety, with 2 tons per hectare yield potential, developed by the Asian Vegetable Research and Development Center (AVRDC), now that resistance to yellow mosaic virus has been incorporated into it.

 

Micronutrient Deficiencies

 

Multiple micronutrient deficiencies (zinc, manganese, copper, boron, molybdenum) occur in soils of the Indo-Gangetic Plains and are becoming more prevalent as cropping intensity increases. The deficiencies are exacerbated by low soil organic matter (levels less than 1 percent) and little return of crop residues or other organic materials to soils.

 

Micronutrient enriched seeds

The micronutrient content of seeds may be important to crop establishment on micronutrient-deficient soils because increased levels of needed nutrients can be provided by the seeds themselves, and because micronutrients such as manganese and boron have been linked to resistance or tolerance to diseases and environmental stresses.

Evaluation of micronutrient-enriched wheat seed (generated by foliar applications of micronutrients) showed a yield response on six of 15 farms tested for a second year in the Dinajpur district of Bangladesh. The geographical pattern of these positive responses was also consistent with the previous year's results, indicating areas where soil micronutrients (especially zinc and molybdenum) are limiting wheat yields.

 

Widespread micronutrient deficiencies in Bangladesh?

In addition to the indications of zinc and molybdenum deficiencies with wheat and the new rice variety BR-32 (see page 76), we also have found available boron levels in soils of the Chuadanga region that are ten times lower than required for sufficiency, and nickel levels in farmer wheat seed from Dinajpur district that were below the critical level for normal seedling development.

In order to better understand the situation with regard to micronutrient status of Bangladesh soils, we have initiated interactions with the Soil Resource Development Institute (SRDI), which has very good geographic information systems (GIS) capacity. It is presently preparing countrywide maps for zinc, boron and copper status of soils. We will use these maps and other databases to do larger-scale, GIS-based evaluations of nutrient constraints to rice, wheat and grain legume productivity in Bangladesh.

An important institutional constraint is a lack of information flow and utilization between SRDI and research scientists in its sister organizations BARI and BRRI. By developing collaborative projects between these units we hope to encourage broader interactions in the future.

 

Wheat and Legume Sterility

 

Grain head sterility, caused by a failure to properly fertilize florets, may affect up to 10 percent of the area cultivated with wheat in Nepal and Bangladesh. Sterility is erratic and often affects a complete wheat crop on a farm when it does occur, imposing a severe burden on farm families. The underlying cause of sterility is thought to be boron deficiency; however, other phloem-immobile nutrients could also be involved.

Sterility is often associated with conditions that reduce transpiration (waterlogged soils, cloudy days) at the time of pollen tube development, and certain modern varieties appear to be genetically susceptible. Probably there is also an interaction between plant nutrient status, which is related to soil fertility, and environmental conditions.

An on-farm field experiment carried out in 1997-98 at Sipaghat village, a mid-hill region of Nepal, demonstrated the following:

 

 

Future work in Nepal, in collaboration with scientists from the Nepal Agricultural Resource Council and CIMMYT, will involve multi-location germplasm screening trials with and without boron addition to assess the contribution of different factors (boron supply, germplasm, and environment) to sterility. In Bangladesh, information from wheat sterility surveys will be compared with SRDI data on soil boron status to determine whether the soil data set is a good predictor of boron deficiency.

Unreliable grain formation is also a problem with many grain legumes and is the biggest technical deterrent to farmer inclusion of these crop species in the rice-wheat cropping system. The situation is similar to wheat in that crops may exhibit good vegetative growth (e.g., mungbean relay trial), but develop few pods and sometimes have sterile pods.

While environmental conditions also play a role in grain development in legumes, we suspect that micronutrient deficiencies, especially boron, are also a major causal factor for sterility in these crops. Trials to evaluate this hypothesis are planned for the coming crop year.

 

Improved Plant Breeding Strategies

 

Most plant breeding programs in South Asia make their selections of preferred cultivars under optimal soil fertility conditions. This approach does not allow lines that have the best capacity to extract nutrients from soil nutrient pools to be identified. If nutrient-efficiency traits are not incorporated into new cultivars, yields may be unnecessarily reduced when crops are grown under conditions of suboptimal nutrient supply, which are widespread in the region. Unfortunately plants do not always exhibit visible deficiency symptoms so the existence of a nutrient deficiency may be unrecognized. An example is the response of BR-32 rice to foliar application of micronutrients (see box on next page).

In Bangladesh, a screening program is underway to identify phosphorus-efficient wheat genotypes. In the rice-wheat system, additions of phosphorus fertilizer are usually needed to correct phosphorus deficiency with wheat but not with rice because phosphorus availability increases under flooded conditions. This breeding strategy is expected to lead to reduced need for inputs of phosphorus fertilizer, which is an expensive import for Bangladesh and which is not always available or affordable for farmers.

Results of screening trials located on phosphorus-deficient soils at two locations where phosphorus treatments were 0 and 80 kilograms of phosphate (P2O5) per hectare, identified several lines that appear to be more phosphorus-efficient and/or responsive to phosphorus addition than are the widely used variety Kanchan or a newly released variety, BAW898. One line yielded about 1 ton per hectare higher than Kanchan and BAW898 without phosphorus addition at both sites, but was not the most responsive to phosphorus addition.

Possibly there may be an induced zinc deficiency at the high phosphorus level. Selections from this trial need to be rigorously evaluated in different environments, and the possibility of interactions with zinc needs to be explored. This trial is breaking new ground in Bangladesh, and its extension to other nutrients, especially micronutrients, could lead to more appropriate germplasm selection for both rice and wheat than can be obtained from traditional breeding approaches.

 

 Micronutrient Deficiencies in New Varieties

 

Efforts to enrich rice seeds with micronutrients revealed that BR-32, a new summer rice variety (T. Aman) developed by the Bangladesh Rice Research Institute for the rice-wheat system, is susceptible to micronutrient deficiencies, particularly zinc and probably also molybdenum. Although no visible deficiency symptoms were observed, foliar application of miconutrients increased the yield of BR-32 by 1 ton per hectare compared with no micronutrient application.

BR-11, the most widely used variety at present, gave no response to micronutrient addition. BR-32 yielded less than BR-11 without micronutrient addition but 0.5 tons per hectare more with micronutrient addition. BR-32 helps with timely planting of wheat because it has a 15-day shorter growth duration than BR-11, and it is also more resistant to pests and diseases.

It appears that plant breeders have unwittingly introduced a "micronutrient flaw" that could be extremely important to the productivity of BR-32 given the widespread prevalence of micronutrient deficiencies in Bangladesh soils. Fortuitously, we will have a better sense of how BR-32 performs after the current cropping season, because Bangladesh colleagues selected this variety for their on-farm solarization trials, causing us to modify the experimental design to include micronutrient treatments.

 

Alternative Tillage and Crop Establishment Practices

 

The soil physical condition that is best for paddy rice is totally different from that which is best for wheat. This creates a management conflict in the rice-wheat rotation. Current farmer practice and research experiments in the Indo-Gangetic Plains optimize soil conditions for paddy rice, which limits wheat productivity by restricting root development and hence access to water and nutrients. Wheat production on heavy textured soils may not be attempted because of difficulties in establishing that crop, and diversification of the cropping system is also discouraged.

Conceptually, disaggregated puddled soil and friable well-aggregated soil are at opposite ends of the possible range of soil physical conditions. We are encouraging scientists in the region to include research at the friable end of this spectrum as well as at in-between points. Needless to say, convincing scientists to try growing rice like a wheat crop is not easy! However, declining water tables in the highest yielding areas of northwest India mandate some change in water management practices.

An experiment designed to evaluate a range of tillage and crop establishment practices has been implemented in several locations in the Indo-Gangetic plains. For rice, this experiment includes comparisons of deep ripping (to fracture tillage pans) and not puddling with conventional paddy as the main treatment, and direct seeding versus transplanting as a secondary treatment. All plots are then split in the wheat phase of the rotation to allow establishment by two different methods: surface seeding (a no-tillage method); and by drill seeding following conventional tillage.

After one cycle of this rotation in a heavy textured soil at Bhairahawa, Nepal, we found no effect of the various treatments on rice yield (mean 5.6 tons per hectare), but a large effect of establishment method on wheat yield (4.2 and 2.2 tons per hectare with surface seeding and the Chinese drill, respectively). The higher yield of wheat with surface seeding was due to earlier (more timely) planting of wheat which was possible because no soil preparation was needed. There was no effect of the rice tillage treatments on wheat yield; but such effects may take time to develop and may vary with soil texture and rainfall amounts and pattern.

 

Technology Adoption

 

The wide geographic scope and large variability of natural and human resources in South Asia makes it imperative to organize research and extension in geographical terms. The rapidly expanding technology of GIS can be used to determine extrapolation areas, to evaluate relationships between variables and, when coupled with crop modeling, to evaluate alternative management or production scenarios. However, use of both GIS and crop modeling by South Asian agricultural scientists is currently rather limited. Several workshops were cosponsored by CIIFAD and ICRISAT to train national scientists and to plan and coordinate a regional information management program:

 

 

Surface seeding

This method allows farmers to reduce the turn-around time between rice and wheat crops by avoiding the time-consuming practice of tillage before wheat seeding. Wheat seed is broadcast into the field roughly one week before rice harvest. To be successful, the method requires that soil must be wet enough "to show footprints." Also seeds need to be coated with cow manure as a deterrent to birds. Yields using this method can be 4.5 tons per hectare as compared with 2.5-3 tons per hectare by conventional practices. Demonstrations of surface seeding versus conventional practice were carried out on five farms in the Bhairahawa, Nepal area, and on one farm each in Nashipur and Chuadunga, Bangladesh.

 

Minimum tillage with Chinese seed drill

This technology significantly reduces land preparation costs and time for wheat planting through reduced tillage using a two-wheeled Chinese tractor. A rotavator-type implement tills the top 5-10 cm of soil and a drill plants the seed, all in one operation. This approach is particularly advantageous on lighter textured soils and when there is insufficient soil moisture at wheat planting. The technology was demonstrated at Chuadunga, Bangladesh.

 

Chinese reaper

A reaper attachment on the Chinese tractor can significantly reduce labor and time at rice harvest, thereby reducing the turn-around time between rice and wheat. On-farm demonstrations of the reaper were conducted in Chuadunga, Bangladesh. An instructional video was made in conjunction with the reaper demonstration entitled "Mechanical Revolution in South Asia: The Growing Use of the Chinese Hand Tractor." A Bangladeshi NGO, Netrokona Agriculture Production and Water Management, will be using this video in its work with farmers.

 

The plants on the left and in the background were surface seeded a month earlier than the plants in the foreground, which were sown by Chinese drill.

 

Human Resource Development

 

A scientist exchange program and several specific training activities were provided to enhance the capability of national scientists to address rice-wheat sustainability issues. The South Asian scientists and their activities are:

 

 

Several Cornell graduate student programs are also being funded through the South Asia program. Multidisciplinary team building and systems approaches are emphasized in all these activities.

As discussed on pages 85-87, we have come across an important threat to health and well-being of people in Bangladesh, a previously overlooked spread of rickets in a particular locality there. This is an opportunity to develop with rural communities, assisted by NGO, university and government partners, a food-based approach to countering this malady. Working with the Food Systems for Improved Health (FSIH) group at Cornell and other partners, we are trying to diagnose the causes of this disease and to find remedies within the framework of our program on rice-wheat systems.

 

 

John Duxbury
Soil, Crop and Atmospheric Sciences

 

Julie Lauren
Soil, Crop and Atmospheric Sciences
 

Participatory Rural Appraisal: Experiences from the Field

 

Participatory research has become a rapidly expanding strategy among agricultural and social scientists who wish to learn more from farmers and be sensitive to their particular needs and interests. The use of participatory techniques often is complemented by greater awareness of regional variation in farm household practices, and growing appreciation for local and indigenous knowledge. There is recognition that as agricultural and social scientists and practitioners, we are relatively ignorant about both the organization of farm labor within households and the ways in which household decision-making unfolds, especially with regard to the adoption of new farming practices and technologies.

One form of participatory research, participatory rural appraisal (PRA), promotes greater interaction between researchers and farm household members and is assumed to use techniques that provide quality data to researchers while facilitating exchange among researchers and community members. Robert Chambers, the founder and promoter of PRA, identifies eight different techniques found useful for PRA researchers. These include participatory mapping, transect walks, matrix scoring, well-being grouping and ranking, seasonal calendars, institutional diagramming, trend and change analysis, and analytical diagramming.

Each technique is assumed to enhance scientific understanding of the constraints inhibiting improved agricultural production, the use of sustainable practices, and the interactions between production and consumption. Accomplished over the course of approximately one week, the process is used to verify and confirm a shared appreciation of the experience by all participants. It is envisioned as one way to draw participants into the project or goal of the research.

Time constraints and tight budgets provide the rationale underlying the use of PRA as a research strategy. Supporters argue that comparisons reveal little difference between the quality, reliability, and understanding of PRA data and the information gathered through more costly surveys and field methodologies. Supporters also assume that traditional research designs promote didactic rather than experiential learning, focus on learning new approaches to knowledge production rather than on changing behavior, and that they respond to long-term rather than short-term investments.

These characteristics of PRA are useful for thinking about the process of understanding and working toward change. But, short-term interaction with farm family members falls short of the insights coming from qualitative methodological approaches and gender-aware research. These approaches stress the significance of time, sensitivity to subject, and the building of trust and exchange between the researcher and the subjects of study. Moreover, these approaches recognize that to understand the complexities of social life, including processes of decision-making and changes in long-term patterns of production, exchange, and consumption, it is critical to observe, develop trust, and share information over a relatively long period.

National scientists working with the Bangladesh Rice Research Institute and Bangladesh Agricultural Research Institute chose to use the PRA approach to: 1) increase interaction and exchange among various disciplines; 2) identify farm practices and constraints as perceived and experienced by farmers; 3) explore the needs of farmers to facilitate in the likely adoption of new practices and technologies; and 4) refine an agenda for future research. What has the PRA field experience revealed?

Visits to the two Bangladesh rice-wheat project sites and a seminar organized to explore research scientists' understanding of PRA revealed a curiosity, excitement, and enthusiasm for the approach. This enthusiasm indicates a willingness among scientists to engage directly with farm communities that were previously seen merely as recipients of experiments and field trials, and a curiosity about what can be learned from this short-term immersion.

However, two broad problems were also revealed from seminar discussions. When asked what they learned from their village visit or what information was garnered from farmer exchanges that they did not previously know, only a few of the scientists were able to provide substantive answers. To the question, "How do you interpret the varied sources of information that you were able to record during your field visit?" the problem was even more troubling, since it revealed that scientists seemed unable to make sense of information that did not fit within their existing frameworks and failed to draw connections between different types of data.

In sharing their findings, scientists noted the same list of constraints to increased yields that they assumed prior to their field visit. Also, despite the increased participation of women in the PRA, there was the assumption that women knew relatively little about farm production. Thus, while scientists expressed excitement about the approach and were keen to continue using it, when queried if their field visit altered how they thought about farm practices, few thought that the PRA had made a notable difference.

Perhaps most revealing about the PRA experience is the absence of a dialogue among diverse groups of scientists and their inability to employ PRA other than as a package of techniques that they pursue ritualistically. As each group employed the approach in a similar way, it became evident that PRA was a methodology that was unable to assist scientists in using their field visits to generate new forms of exchange among themselves as well as with farmers. Likewise scientists were unable to engage in the sort of exploration of themes across disciplinary boundaries that would help illuminate farmer practices and needs.

In short, in the tradition of the Green Revolution package where the system was assumed to work best when all of the techniques were used together, the PRA formula did not fulfill the possibility that local community members would be fully represented in the research process, including in the data analysis stage. Moreover, there is little indication that the PRA would generate long-term collaboration between the research and farmer community.

These problems are structural features of the PRA approach that stem from the way it is taught, from desire to keep costs low and to share results quickly. The consequence is limited attention to the research process and limited opportunity for sustained long-term relations between scientists and producers. Thus, while the participatory impulse is critically important for all research, its packaging under the PRA rubric suggests that new forms of training, a new appreciation for self-reflection, and more critical engagement across sites, disciplines and among researchers with diverse skills and expertise will be necessary to make it an effective and innovative tool for scholarly endeavor.

 

Shelley Feldman
Rural Sociology

.

Community map drawn on the ground by villagers during a PRA
exercise in Jagdal, Bangladesh, February 1998.
The map delineates boundaries of fields and villages and indicates land use.