Influence of Field Environment and Fertilizer Management on Ammonia Loss from Flooded Rice

Abstract The importance of ammonia volatilization after urea application to flooded rice ( Oryza sativa L.) was assessed in six experiments at two field locations in the Philippines, using a nondisturbing micrometeorological technique. Earlier studies had produced conflicting results probably because the techniques used did not conserve the balance of physical, chemical, and biological processes that affect ammonia loss in this ecosystem. Urea was either broadcast and incorporated immediately before transplanting of rice seedlings (BI treatments) or broadcast into the floodwater 14 or 21 d after transplanting (AT treatments) and 5 to 7 d before panicle initiation (PI treatments). Ammonia volatilization proceeded rapidly and continued for 6 to 10 d after urea was applied. In the AT treatments, NH 3 loss accounted for 47% and 27% of the urea‐N applied. Ammonia loss showed pronounced diurnal fluctuations that were synchronized with fluctuations in floodwater pH, between maxima of 8.6 to 9.0 at 1200 to 1400 h and minima of 7.8 to 8.0 at 0500 to 0600 h in the AT experiments. Floodwater pH in nonfertilized areas showed similar trends, indicating that fertilizer addition was not the prime cause of the pH fluctuations. Ammoniacal N in the floodwater accumulated to ≃ 14 g m −3 in the AT studies, but the concentrations decreased during daytime and increased again at night, with a net loss of ammonia during the day. Differences in wind speed appeared to account for most of the difference in NH 3 loss observed between the AT treatments. Rates of ammonia loss (10–15%) were lower when urea was applied at a later growth stage (PI), even though floodwater ammoniacal N again rose to 13 g m −3 . In contrast to the AT studies, floodwater pH at PI did not exceed 8.0 to 8.3. Apparently, shading of the floodwater by the rice canopy lowered the photosynthetic activity of the aquatic biomass and thus reduced the degree of CO 2 depletion and the potential for NH 3 loss. Ammonia volatilization loss after urea incorporation (BI treatment) accounted for only 13% of the N applied at Los Baños. This sharp reduction in loss, compared with those after AT applications, was largely due to lower floodwater urea and ammoniacal N concentrations from 2 days after application in a system where urea was largely incorporated into the soil.