Season, nitrogen rate, and plant type affect nitrogen uptake and nitrogen use efficiency in rice

Studies were undertaken in the Burdekin River Irrigation Area of northern Australia to improve the efficiency of nitrogen (N) use for rice (Oryza sativa L.) production. The aim was to maximise grain yield by optimising its functional components: N uptake, efficiency of N use for dry matter production (NUEdm), and harvest index (HI). The effiects of season (wet and dry), N rate (0, 70, 140, 210, and 280 kg/ha), and plant type (maturity and stature) on N uptake, NUEdm, and HI were examined in 2 wet and 2 dry seasons. Leaf area development was closely related to N uptake. In the wet season, genotypes had similar rates of increase in leaf area index (LAI) with N uptake but differed in the level of LAI (curves were parallel). In the dry season, the relationship between N uptake and LAI was different for each genotype (curves not parallel). In both seasons cv. Newbonnet generally had a lower LAI per unit N uptake (i.e. leaf area production was not excessive) than cvv. Lemont and Starbonnet. Dry matter production and grain yield were also closely related to N uptake. At low levels of N availability (N uptake <100 kg/ha) tissue N concentrations were low and both total above-ground dry mass (AGDM) and grain yield were linearly related to N uptake. As N availability increased, N uptake and tissue N concentration increased, resulting in a deviation of the AGDM and grain yield curves from the linear at about 100 kg/haN uptake. This finally resulted in AGDM and grain yield plateauing at around 200 kg/ha N uptake. Above this level some factors other than N availability limited yield. Seasonal differences in N uptake, NUEdm, and HI were observed. Seasonal variation in the response of grain yield to N uptake was found. There was a trend for higher N uptake in the absence of fertiliser application in the wet than the dry season, and the recovery fraction was less for N rates >140 kg/ha in all seasons, i.e. fertiliser N uptake efficiency declined with increasing N rate. Nitrogen was used more effectively by the rice crop to produce grain compared with non-grain parts when average daily mean temperatures were lower during the period between panicle initiation and anthesis. Genotypic variation was found in N uptake, NUEdm, and HI. The ability to capture these components in crop improvement programs depends on the extent to which genetic linkages between N uptake and both NUEdm and HI can be broken. While our data suggest that N uptake is generally negatively correlated with both NUEdm and HI, there is some evidence that these linkages can be broken. For example, the fact that HIdid not change with increasing N uptake in Lemont and, to a lesser extent, in Newbonnet suggests that HI does not always decline with increasing N uptake. The example of Newbonnet suggests that, to some extent, it is possible to increase yield by increasing each of the functional components independently within a specific genotype.