Raw data of PhD research: The feasibility of high yielding triple rice systems in the polders of the low salinity coastal zone of Bangladesh

The agricultural productivity of the polders of the coastal zone of Bangladesh is very low, despite large investments in embankments and sluice gates to protect the lands from tidal flooding and saline water intrusion. Excessive waterlogging as a result of high rainfall and poor drainage, soil salinity, severe cyclonic storms and tidal surges are major constraints to increasing cropping productivity and intensity in the polders. In the rainy season, most of the cultivable lands grow photoperiod sensitive, tall, local aman varieties that are well adapted to waterlogging (prolonged periods of 30-50 cm water depth), but with low yield potential (2-3.5 t/ha) and late maturity (December). During the dry season, low yielding local rabi crops are often grown, but large areas remain fallow for 3-7 months each year. However, the south-central coastal zone (Barisal division) is supplied by large rivers and abundant fresh water throughout the dry season. Furthermore, the rivers surrounding the polders are tidal, and the dense networks of small canals inside the polders create the opportunity to distribute fresh river water across the landscape in the dry season. The same networks provide the opportunity to drain the lands by gravity at low tide during the rainy season, by strategic opening of the sluice gates connecting the canals to the rivers at low tide. Good drainage management provides the entry point for the production of modern, high yielding aman varieties, while irrigation provides the entry point for the production of boro crops, and early establishment of aus crops. Thus, the main hypothesis of this thesis is that, with good water management and modern high yielding rice varieties, it is possible to grow three rice crops and produce 16-20 t/ha of grain per year, in the low salinity coastal zone. To test this hypothesis, a cropping system field experiment was conducted at Patuakhali for two years (2012-14) with 3 aus sowing dates (10 April, 25 April, 1 May), 4 aman varieties (BRRI dhan33, BRRI dhan49, BRRI dhan52, BRRI dhan53), and 5 boro sowing dates (15, 20, 30 November and 5, 15 December). The results showed the feasibility of producing three high yielding rice crops within a 12-month period, with adequate time for turnaround between crops. Over the two years, aus-aman-boro system yields of 13-17 t/ha/yr were achieved. Significantly, all the aman and aus varieties tested in this study were able to withstand submergence for periods of up to 10 days when the water was drained rapidly, and went on to produce high yields (aman ~6 t/ha; aus ~4 t/ha). This was true of varieties both with and without SUB1. System yields in the first year (13-16 t/ha) were generally lower than in the second year (15-17 t/ha). This was mainly due to lower boro yield (by 1-2 t/ha) which was associated with brown spot infestation under the unusually low temperatures in December 2012 and January 2013. In the second year, yields were highest in the systems with 25 April sowing of the aus crop (BRRI dhan65), followed by BRRI dhan49 or BRRI dhan52, and 30 November or 15 December sowing of the boro crop (BRRI dhan28). These were also the systems with the best turnaround times between all crops and highest profitability. However, all combinations of aus establishment method, aman variety and boro sowing date were profitable in terms of gross margin (44.6-113 kTk./ha), net return (18.7-87.0 kTk./ha) and benefit cost ratio (1.08-1.38). The results suggest that there is an opportunity to greatly increase rice crop production and improve livelihoods by investment in improved water management and the adoption of modern high yielding rice varieties and production methods in the polders of the south central coastal zone of Bangladesh