The growth and nitrogen economy of rice under sprinkler and flood irrigation in South East Australia

Summary Dry-seeded rice (Oryza sativa L., cv. Calrose) was subjected to 4 irrigation treatments — continuous flood (CF) and sprinkler irrigation at frequencies of one (S1 W), two (S2W) and three (S3W) applications per week — commencing 37 d after 50% emergence (DAE). The amount of water applied was calculated to replace water lost by pan evaporation. Urea (120 kg N ha^–1) was applied in a 1:1 split 36 and 84 DAE, and there were also unfertilized controls for each irrigation treatment. Amounts of nitrate (NO₃^–) in the soil were very low throughout the growing season in all treatments, despite regular periods of draining which lasted for up to 7 d in SlW. In all irrigation treatments, the majority of the fertilizer nitrogen (N) was located in the top 20 mm of soil. After each application of fertilizer, levels of mineral N in CF declined rapidly, while levels in S3W and S1W remained high for 1–2 weeks longer. The poor growth of sprinkler-irrigated rice was not due to lower amounts of mineral N in the soil. The greater persistence of fertilizer N in the sprinkler-irrigated treatments was probably due to reduced root activity near the soil surface because of frequent periods of soil drying in between irrigations. Net mineralization of soil N in the unfertilized sprinkler-irrigated treatments was reduced by about half compared with CF.On average, the quantity of water applied (1.2–1.4 × EP) to the sprinkler-irrigated treatments appeared to be sufficient to meet the evapotranspiration demands of the crop, except possibly around flowering time. However, the plants may have suffered from moisture stress in between irrigations. Soil matric potential data at 100 mm suggested little water stress in the sprinkler-irrigated treatments during the vegetative stage, consistent with the similar tiller and panicle densities in all irrigation treatments. However, the crop was stunted and yellow and leaf rolling was observed in the sprinkler-irrigated treatments during this period. Soil matric potential data at 100 mm indicated considerable water stress in S1W beyond the commencement of anthesis, and in S2W during grain filling, consistent with the reduced floret fertility and grain weight in those treatments.