Growth and CO2 assimilation of lowland rice in response to timing and method of N fertilization

In order to understand more clearly the dynamics of rice (Oryza sativa L.) yield development in relation to N fertilization, a field experiment was conducted in Laguna, Philippines. The soil, a Maahas clay (Aquic Hapludalf), was flooded, puddled, and then planted with 20-day-old seedlings of IR64 rice. Treatments were six combinations of N fertilizer timing and method: (1) 0 N control; (2) prilled urea broadcast at 15 and 60 days after transplanting (DAT) (BR-LATE); (3) prilled urea injected with a spring auger applicator at 15 DAT and broadcast at 60 DAT (INJ-LATE); (4) prilled urea broadcast and incorporated at 0 DAT and broadcast at 40 DAT (BR-EARLY); (5) urea super granules (USG) manually deep-placed at 3 DAT and prilled urea broadcast at 40 DAT (DP-EARLY); and (6) USG manually deep-placed at 3 DAT (DP). Except for the control, all treatments received a total of 58 kg N ha^-1. Yield results were consistent with those of other experiments, namely, DP had the highest yields, the early-split treatments (BR-EARLY and DP-EARLY) were second best, followed by the late-split treatments (BR-LATE and INJ-LATE), with the control having the lowest yield. Sequential harvest results showed that the advantages of DP in terms of dry matter assimilation, tillering, and leaf area index (LAI) were expressed later in the season. For all treatments, midday net CO₂ assimilation (Ac) peaked around 48 DAT, approximately panicle initiation. Grain yield was highly correlated with midday Ac at panicle initiation and flowering but not at other growth periods. Rates of midday Ac and night respiration depended strongly on shoot N content. We conclude that N application method and timing should be designed to give high shoot N content at panicle initiation and flowering, and that DP satisfied this requirement best among the treatments tested.