Nile tilapia (Oreochromis niloticus) seed production in irrigated rice-fields in Northwest Bangladesh—an approach appropriate for poorer farmers?

Participatory research was conducted with poorer farmers in two communities, Girai (G) and Bahagili (B) in NW Bangladesh to assess the production of Nile tilapia seed in irrigated spring rice-fields. All the selected households (G = 15; B = 4) had previous experience producing common carp (Cyprinus carpio) in the rice-fields allocated a separate plot in which a deeper area had been excavated for this trial. Mature GIFT strain Nile tilapia (12 female and 6 male; 121 ± 34 g and 158 ± 54 g size, respectively) were supplied to each household irrespective of the size of their trial plot (mean < 0.15 ha). The trial started in the spring rice season (boro) in February and ended at the end of main season (amon) in December 1999. Management practices, production and sales of fish were monitored weekly.The majority of farmers succeeded in producing fingerlings in their plots; 11% failed totally but around 70% produced more than 2000 fingerlings from a single plot. Production during boro and fallow period was much higher (> 90% total) than during the subsequent amon crop (< 10% total). Total production was highly variable among households but not different between the two study areas (G = 4092 ± 3277; B = 3730 ± 4232 fingerlings household^− 1). Daily production of fingerlings per unit area was relatively low (< 1 fish m^− 2 day^− 1) but efficiency of production was high, averaging 17.3 fingerlings. kg^− 1 female day^− 1. Mean individual harvest weight was 21 g.Most fingerlings were sold (43%) and/or stocked for further culture in their own grow-out system (39%), but some were used directly for household consumption (17%).