Comparison of high-yield rice in tropical and subtropical environments: I. Determinants of grain and dry matter yields

Abstract

Yields over 13 t ha −1 have been reported for irrigated rice in subtropical environments while maximum yield of only 10 t ha −1 has been achieved in the tropical lowlands. While it is generally accepted that the longer growth duration in subtropical environments mainly contributes to the greater yield potential, comparisons of adapted cultivars in tropical and subtropical climates are lacking and other factors that might be responsible for differences in yield potential have not been identified. Field experiments were conducted in a tropical environment at the International Rice Research Institute (IRRI) in the Philippines and in a subtropical environment at Taoyuan Township, Yunnan, China in 1995 and 1996. Three to five high-yielding rice cultivars were grown in each experiment under optimum crop management to achieve maximum attainable yields. Yield, yield components, plant dry matter and harvest index (HI) were determined at maturity. Growth analyses were conducted at key growth stages to determine crop growth rate (CGR), leaf area index (LAI), and leaf area duration (LAD). Daily radiation and air temperature were monitored. The highest yield of 15.2 t ha −1 was produced at Yunnan by Shanyou 63, a Chinese indica F 1 hybrid, whereas maximum yield at IRRI was 9.3 t ha −1. On the average across cultivars, Yunnan produced 33 and 62% greater yields than IRRI in 1995 and 1996, respectively. Sink size (spikelets per m 2) was responsible for these yield differences. Larger panicles (spikelets per panicle) contributed mostly to the greater sink size at Yunnan. Biomass production was 42% and 58% greater at Yunnan than at IRRI in 1995 and 1996, respectively, while differences in HI were relatively small. Dry matter accumulation and CGR were significantly greater at Yunnan than at IRRI during vegetative and grain-filling stages. These differences were relatively small and inconsistent across cultivars and years during the reproductive phase. Yunnan had greater LAI and LAD than at IRRI, which might be responsible for greater CGR. These results indicate that further improvement in rice yield potential in the tropics will depend mainly on the ability to increase sink size and biomass production.