Rice varieties and cultural management practices for high and sustained productivity in the coastal wetlands of southern Bangladesh.

Abstract

Abstract The importance of including farmers in targeted breeding and participatory variety selection (PVS) to ensure the adoption of high-yielding varieties was demonstrated in the coastal wetlands of the Ganges-Brahamputra Delta of southern Bangladesh. Five sites were selected for conducting 'mother and baby' trials. About 245 salt-tolerant genotypes from the International Rice Research Institute (IRRI) and Bangladesh Rice Research Institute (BRRI) were divided into three mother trial sets of 72 genotypes for the wet ('T. aman') season, 76 for the irrigated dry ('boro') season and 85 for the dry direct-seeded ('aus') season. During the PVS trials, farmers chose the best one or two genotypes from mother trials to grow in their fields as baby trials. Through local non-government organizations (NGOs: Proshika, Gurpukur, Uttaran and Sushilan), a total of 199 resource-poor farmers, including 15 women, selected 34 genotypes for the three seasons. Farmers were given 500 g of seeds of each selection for baby trials under their own management. For the 'boro' season, IR64401-2B-14-1-1, IR60483-2B-17-2-1-2 and BR5777-4-2-1-HR2 were selected. For the 'T. aman' season, IR66401-2B-14-1-1 was preferred for its short growth duration and high yield. For the direct-seeded 'aus' season, three genotypes, IR72593-B19-2-3-1, IR64419-3B-12-2 and IR64419-3B-4-3, were identified because of their short growth duration. IR63307-4B-4-3 was identified through PVS trials and was released as BRRI dhan 47 for commercial use in coastal Bangladesh in 2006. Out of 53 landraces collected from these coastal areas, four genotypes - Capsule, Ashfal, Ashfal balam and Chikiram Patnai - were identified as new salt-tolerant donors. Moreover, 16 modern genotypes, including four varieties from Vietnam, were identified as tolerant of salt stress at the seedling stage (EC of 12 dS/m). OM1490 had short duration similar to that of BRRI dhan 28, a popular dry-season variety, but produced more than 1 t/ha extra grain yield. In addition, OM2718, AS996 and BR7109-5R-4 had growth duration similar to BRRI dhan 47 and a yield advantage of more than 1 t/ha. In contrast to BRRI dhan 47, these new lines were non-shattering. Adding ash along with the recommended N, P, K and Zn at 120, 80, 30 and 0.5 kg/ha, respectively, did not increase the grain yield of all varieties in saline areas because of the high soil potassium saturation (>20%). Moreover, farmers in these areas had been growing salt-sensitive rice varieties successfully by practising: (i) high transplanting rates of 5-6 seedlings/hill; (ii) draining the field, followed by irrigation with fresh water from shallow tube wells; and (iii) shifting the shallow tube-well installation after 3-5 years to tap less salt water. Monitoring of the salinity in nearby rivers (Retna) suggested the possibility of their use for rice irrigation from June to the middle of February, when salinity was mostly below 4 dS/m, with potential for an additional dry-season crop.