Comparison of dry seeded and puddled transplanted rainy season rice on the High Ganges River Floodplain of Bangladesh

In the High Ganges River Floodplain of Bangladesh, rice is normally established by puddling and transplanting. This is a costly practice in terms of tillage, labour and irrigation requirement. Dry seeding of rice has the potential to reduce these costs and facilitate timely crop establishment. However, the performance of dry seeded rice (DSR) in other parts of South Asia has been variable in comparison with that of puddled transplanted rice (PTR). Therefore, a four-year replicated experiment was conducted to compare the performance of PTR and DSR, grown during the rainy season, in the High Ganges River Floodplain. There were two tillage treatments for DSR − full tillage and strip tillage. Two levels of rice residue retention (removed; partial retention) were compared in sub-plots. Grain yield of DSR was significantly (by 5% or 0.2 t ha−1) lower than the yield of PTR, while DSR reduced irrigation input by 240-880 mm over the four years. The variation in the reduction in irrigation input was due to variation in the incidence and amount of rainfall across the four years. There were no significant differences between the use of full and strip tillage for DSR for any of the measured parameters, and any effects of rice residue retention were small. The cost of production of DSR was reduced by 9% (USD 87 ha−1) in comparison with PTR, which led to a small increase in the gross margin and BCR of DSR. The results suggest that replacement of puddled transplanted aman with dry seeded aman in the High Ganges River Floodplain can be done with only a small yield loss, but with significantly reduced irrigation input, lower costs of production, and similar profitability.