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

Abstract

The agronomic implications of growing rice under sprinkler irrigation on a duplex soil in inland south east Australia were examined by comparing 3 sprinkler irrigation regimes applied once, twice and three times a week (S1W, S2W and S3W) with continuous flood irrigation (CF). Each sprinkler irrigation treatment was managed to replace evaporative loss since the previous irrigation. Each irrigation main plot was split into subplots receiving 3 levels of nitrogen (N) fertilizer −0, 80 or 120 kg N ha−1 (0N, 80N and 120N). Grain yield on all sprinkler-irrigated treatments was reduced by 50% or more when compared with CF. There was a slight decline in grain yield with lower frequency of sprinkler irrigation. The main factors contributing to the lower yields were reductions in the number of spikelets per panicle and in floret fertility. Panicle density was not significantly influenced by the irrigation treatments. Sprinkler irrigation delayed anthesis by at least 8 days, and the duration of anthesis was extended by 5 to 7 days. The lower yields under sprinkler irrigation did not appear to be due to greater N deficiency and it is uncertain whether any or all irrigation treatments suffered from P deficiency. Irrigation treatment had little effect on plant N concentration, but P concentration in the plant tops was reduced by sprinkler irrigation during the vegetative and reproductive stages. Low night temperatures throughout the reproductive phase reduced floret fertility, but all irrigation treatments at the same N rate (except SlW/120N) were affected to a similar extent. Water stress, as evidenced by leaf rolling in response to high evapotranspiration rates in this semi-arid environment, was considered to be the main factor contributing to the decline in yield on the sprinkler-irrigated treatments. The irrigation and N rate treatments did not interact significantly in their effect on any important parameter measured. Increasing the rate of applied N from 0 to 80 kg N ha−1 increased yield by about 1 t ha−1 by producing more panicles, but 120N was of no further benefit and was associated with increased floret sterility. Nitrogen fertilizer increased both the N uptake and N concentration in the plant material. We conclude that sprinkler irrigation for rice production is unlikely to be a viable strategy for water management on duplex soils in inland south east Australia. Conventional dry-seeded culture with permanent flood delayed until panicle initiation is probably a preferable alternative to sprinkler irrigation for increasing water-use efficiency.