Addressing constraints to pulses in cereals-based cropping systems, with particular reference to poverty alleviation in north-western Bangladesh

The 60 on-farm demonstrations conducted in the 2006-07 season confirmed that the existing chickpea technology is generally suitable for the central and northern parts of the High Barind Tract (HBT). Major biophysical constraints contributing to risk of chickpea cultivation in the HBT continue to be low and unpredictable soil moisture status and Helicoverpa pod borer attack. The market price is clearly very attractive, and farmers appreciate that costs of inputs are very low compared to boro rice, wheat or maize. The sub-sector mapping arising from the Business Development Services (BDS) workshop held in February will help to identify the key bottlenecks on which the Project should focus to more widespread adoption of chickpea in the HBT.

Generally, lentil cultivation under the project has been successful in northern Bangladesh, with yields exceeding the national average being achieved by the novice growers in most of the 120 demonstration plots sown. Plant populations established were mostly satisfactory except when sown into excessively wet soil, where nodulation was inhibited or delayed resulting in slow and/or uneven greening of the canopy. The other major constraint to lentil was stemphylium blight but this could be managed by fungicidal spraying and minimizing canopy density. It appears that there is a narrow range of optimum population density for lentil in this environment: this range will need to be established for PTOS-sown crops.

Although soils of the region are acid (pH 4.5-5.5), lentil appeared to grow satisfactorily in many plots. Poor growth could mainly be attributed to initial waterlogging events, which prevented or delayed N2 fixation. However, in the on-farm trial at Tupamari, Nilphamari, there was a large lime response, even though growth in general was very poor at this location. Further experimentation is required to understand the nature of the lime response.

Many lentil fields had two weedings, but this had not controlled weeds sufficiently. Further attention to weed control will be needed for the moister conditions of northern Bangladesh, especially with PTOS-seeding in rows. BARI developed a dryland hand weeder (or 3-kata) that may be used for weeding in row-planted crops. Applying herbicide is one of the options for weed control. Typical B deficiency symptoms of lentil were observed in only a few of the fields seen. An uneven hand broadcasting of the B fertilizer may have resulted in the sporadic appearance of symptoms.

Novice farmers (50) in the northern Bangladesh also generally achieved yields of chickpea higher than the national average. An on-farm trial confirmed the need to apply boron, molybdenum and Rhizobium, but there was also an additional response to lime. Other major constraints to chickpea were botrytis grey mould, collar rot (Sclerotium rolfsii) and pod borer. Biophysical constraints to lentil and chickpea in the northern districts were prioritized in a project report and overall constraints to production and consumption of these crops in north-western Bangladesh will be analysed through a BDS approach.

After various modifications with the rotary tiller, the 4-blade strip tiller was found to be the best option for minimum tillage, row planting. Bent blades throw too much soil out of the strip, leaving an open slot to about 2-3 cm depth. Straight 'C' type blades work better. However, on the heavier soils such as the Barind soil the slot remains open to 2-4 cm, and makes it difficult for the press wheel to press the soil down on the seed. This may require laddering to drag soil back into the strip. Chains behind the rotary tiller may be able to pull the soil back into the strip. Perhaps a different shape or thickness of the press wheel may be required. It would be desirable that further testing of some of these options occurs before the 2007 rabi season.

A 16 HP power tiller (PT) pulled 2 tynes through the medium textured HBT soils. However, very few 16 HP PT exist and hence the development of no-till planters should not be reliant upon 16 HP PT, but rather developed for 12 HP machines as these are cheap and commonly available.

The project organized a series of training programmes on lentil and chickpea cultivation and followed up with field days aimed at disseminating optimum cultivation methods. The project held an inception workshop on 5 November 2006, a BDS workshop, on 4-5 February 2007 and a review workshop and coordination meeting on 24 May 2007. Two PhD scholars were identified to work on the project and arrangements made for a training visit of two Bangladeshi Project members to Australia in August-September, 2007. The Australian component of the project began in June 2007 with the sowing of a field trial examining interactions between row spacing, fertilizer placement, soil moisture and seed priming of chickpea at Merredin in Western Australia.

Overall, most activities of the project are on schedule. The economic environment in which pulses are produced has changed markedly in the last 11 months due to the sharp rise in prices of internationally traded grains, fertilisers and fuel. There are opportunities and threats for pulses in Bangladesh as a consequence of these changes. Pulses remain highly profitable, but the rising price of maize, wheat and Boro rice has reinforced the view that our research should specifically target those areas with limited or no irrigation so that the Rabi season pulses, chickpea and lentil, are not grown in direct competition with cereals.

Two years of demonstrations have shown that the ICM package developed for chickpea is applicable in the central and northern High Barind Tract (HBT), but Mo and Rhizobium are essential because of the surface soil acidity. A replicated field experiment at one location showed no direct response to lime application, suggesting that acidity primarily lowers Mo availability. However, this finding needs confirmation. Sowing into a seedbed of adequate moisture content and controlling pod borer remain critically important to achieving reasonable chickpea yields and extension advisers and farmers need thorough training in the management of these constraints. Extensive evaluation of lentil in the HBT, where it was previously considered to be an unsuitable crop, has identified niches where lentil would be a viable alternative to chickpea. A technique of lab rearing of larvae, to allow year-round production of quality HNPV at low cost, was confirmed. Commercial production and sale of HNPV is now being attempted. Surveys suggest a 25 % increase in chickpea planting in the areas where demonstrations were held in 2006-07. In addition PROVA responded to farmer interest by selling them 580 kg of seed plus Mo and Rhizobium, thus commercializing technology transfer. The practice of priming with Mo + Rhizobium increased among farmers exposed to ICM training from 13 to 66 %. This plus the increase in farmers practicing proper seed storage, from 41 to 56 %, bodes well for continued spread of chickpea in the HBT through the use of the ICM package.

Two seasons of evaluation of both chickpea and lentil in northern Bangladesh, have shown that reasonable yields can be obtained provided the ICM package is followed. Mo and Rhizobium were also required here but additional responses to lime were obtained, suggesting that in northern districts additional soil acidity constraints exist apart from Mo deficiency. However, it was observed that the present priming procedure for inoculating seed with Rhizobium was not fully effective, particularly for lentil, and attempts are being made to improve the inoculation procedure. A time of sowing study indicates that the sowing window for lentil is much wider than previously considered, extending to near the end of November. In many of the demonstrations control of foliar diseases and pod borer was not satisfactory. This suggests that further, more intensive, training is required for farmers and extension personnel in pest and disease management. In addition, site selection was sometimes inadequate, with some plots located in areas of intensive irrigation, and attempts are being made to more precisely delineate potential areas for lentil and chickpea, where irrigation facilities are restricted or absent. RDRS has recorded technology adoption at specific locations, mainly in Dinajpur District, but not in areas of predominantly irrigated crop production.

A sub-sector analysis for lentil and chickpea in NW Bangladesh is nearing completion. This will identify the key bottlenecks for expansion of chickpea and lentil production and provide a framework for their further promotion.

The second version of the zero tillage planters were tested in farmers' fields and in field experiments on farmers' fields in the 2007 rabi season. Only the strip tillage option is presently a workable option for the heavy clay soils of the HBT. However, the strip tillage operation often throws too much soil out of the tilled slot which results in ineffective pressing of soil around seed. Further investigation is underway to improve the strip tillage planters. Tyned no-till planters are able to sow precisely on the lighter textured soils of the northern Bangladesh districts, but are not currently satisfactory for the heavy clay soils of the HBT. The priority for 2008-09 is to manufacture a number of planters for strip tillage and have them tested in farmers' fields to determine their effectiveness in planting under a variety of field conditions, the robustness of the planters, and their acceptability by farmers.

Two PhD scholars have commenced studies under the project on "Managing risk to crop establishment of chickpeas in rainfed environments of Western Australia and Bangladesh" and "Fertiliser management in chickpea in the High Barind Tract soils under reduced tillage".