Ammonia volatilization from the surface of a Japanese paddy field during rice cultivation

Ammonia (NH3) volatilization from the surface of a Japanese paddy field during rice cultivation was measured using the dynamic chamber method with a dry NH3 collector. A preliminary investigation showed that a dry NH3 collector with phosphoric-acid-impregnated filters could collect volatilized NH3 with sufficient efficiency. The experimental field included six lysimeter plots of Gray Lowland soil with a soil pH (H2O) of 5.7. Urea was applied as nitrogen fertilizer at a rate of 50 kg N ha(-1) by incorporation with puddling into the plowed layer as the basal fertilization (BF) and at rates of 30 and 10 kg N ha(-1) by top-dressing as the first (AF1) and second (AF2) additional fertilizations, respectively. Relatively strong NH3 volatilization occurred immediately after AF1 with a maximum flux of 45 g N ha(-1) h(-1). In contrast, the NH3 volatilization fluxes following BF or AF2 were weaker. The ammonium (NH4+) concentration and the flooded water table were found to be major factors influencing NH3 volatilization; higher NH3 volatilization fluxes were often observed with a higher NH4+ concentration in floodwater and a lower water table. Incorporation of urea with puddling resulted in lower NH4+ concentrations in floodwater than in the case of top-dressing application, which likely resulted in fewer NH3 volatilization fluxes after BF than after AF1 and AF2. In contrast, relatively strong NH3 volatilization occurred in the plots immediately drained after AF1, which suggested that a top-dressing application under nearly drained conditions enhanced NH3 volatilization. The ratio of NH3 volatilization loss to applied nitrogen for each application of fertilization was 0.2 +/- 0.1, 3.8 +/- 2.2 and 0.7 +/- 0.5% for BF, AF1 and AF2, respectively. In addition, the total ratio of NH3 volatilization loss to total applied nitrogen throughout rice cultivation was estimated to be 1.4 +/- 0.8%. These values were smaller than those reported from other Asian paddy fields, with the exception of the maximum NH3 loss after AF1, 9.0%, which was comparable to the minimum NH3 loss reported in Asian paddy fields. The application rates of nitrogen fertilizer in the present study were smaller than those in Asian paddy fields, although they are conventional for Japan. In conclusion, the very small values of NH3 volatilization recorded in the present were ascribed to the small rates of urea application per fertilization, which restrained increases in NH4+ concentrations in floodwater, and to the relatively low soil pH, which resulted in prevention of NH4+ dissociation in floodwater.