Rate and duration of grain filling of aerobic rice HD297 and their influence on grain yield under different growing conditions

Abstract

Aerobic rice is grown in non-puddled soil, and it typically shows low yield, harvest index, and percentage of filled grains (PFG). In lowland rice, PFG is closely related to the grain filling patterns of superior and inferior grains. A synchronous pattern leads to high PFG. Since no studies have been done on the grain filling of aerobic rice, its effects on yield are unknown. Two field experiments were conducted simultaneously to investigate the grain filling and yield formation of aerobic rice HD297 at two contrasting sites in northern China with and without nitrogen urea (150 kg/ha) as fertilizer. Nitrogen urea (fertilizer-N) significantly improved the yield of HD297 at a low soil fertility site, but not so at a high soil fertility site. Among yield components, the number of productive tillers and PFG contributed most to the yield. Fertilizer-N and high fertility improved total dry matter and tillering, but significantly decreased PFG. Grain weight of superior grains showed a typical 'S' curve with faster filling rate, while that of inferior grains continuously increased during the filling stage with low rates. The filling pattern of HD297 was asynchronous, and was not significantly affected by the fertilizer-N or the site. Fertilizer-N had little effect on the non-structural carbohydrates (NSCs) accumulation or contribution to grain. Postanthesis, the NSC contribution was only around 70%, suggesting an insufficient carbohydrate supply to the spikelets during the filling stage. The yield of HD297 could be increased by delaying N-dressing for longer photosynthesis, increasing plant density for more superior grains, or improving the variety for the ideal filling pattern.