Seasonal nitrous oxide emissions outweigh the effect of higher nitrogen rate in flooded triple rice systems.

Intensive rice cultivation with multiple aeration in both irrigated and rainfed farming leads to increased nitrous oxide (NO) emissions. However, the spatio-temporal variations in NO emissions has been largely neglected in the policy guidelines of seasonal rice cultivation. NO emissions were quantified in two rice regions in Bangladesh: Bogura and Cumilla in annual triple rice systems for two consecutive years. The treatments were two residue levels (NR, no residue and CR, crop was cut at 30 cm height); and four nitrogen (N) rates: control (no N), farmers' practice (FP), current recommended N rate (RD) and 125 % of RD, 1.25RD. The yield target and N application rate was different for the pre-monsoon rice (T. Aus), monsoon (T. Aman) and winter rice (Boro). The NO emissions within the season increased with the N application rate. Despite the different N rates, NO emissions were similar between seasons, suggesting that the estimation of NO emission factors (EFs) needs to be disintegrated into the different seasons in multiple rice systems. Integration of crop residues coupled with RD of N fertilizer (CR-RD) increased NO emissions over the NR with the same N rate (NR-RD). Mean NO emissions ranged from 0.68 to 0.88 kg N ha in Boro, 0.75 to 0.82 kg N ha in T. Aus and 0.69 to 0.77 in T. Aman, indicating that even the lower N rate in the warmer seasons can emit similar NO to the cooler season resulting in higher EFs. The NO EFs, being higher in warmer seasons, ranged from 0.0039 to 0.0074, was slightly above the IPCC default EF (0.0033 for flooded and 0.0050 for intermittent draingage), but still within the global rice EF range. While crop residue increased rice NO emissions, its integration with optimum N rates will minimize the climate impacts of rice through reduced NO emissions.