在模型验证的基础上，利用ORYZA2000模型分析表明，北京地区旱稻产量潜力30年平均为8 573 kg hm-2，变化范围为6 747~11 278 kg hm-2，年际间变异系数为16.3%。而雨养产量的多年平均值为4 084 kg hm-2，变化范围223~8 018 kg hm-2，年际间的变异系数高达51.1%。产量潜力与雨养产量之差表明北京地区旱稻尚有一定的增产潜力。旱稻全生育期需水量的多年平均值为713 mm，年际间变异为8.5%。其中，出苗—穗分化阶段由于持续时间长，需水量最大，占全生育期的49.7%。全生育期的平均需水强度为5.3 mm d-1，需水强度最大的时期为穗分化—开花阶段，平均为5.5 mm d-1。北京地区生育期内的降水量不能完全满足旱稻的需水要求，50%的年份水分亏缺量在250 mm以上，水分亏缺量多年平均值为226 mm，年际间变异较大，变异系数高达81.4%。
With the decreasing of available water in agriculture, water consumption in rice production has to be reduced and water utilization efficiency increased. Irrigated aerobic rice is a new system being developed in lowland areas with water shortage and in favorable upland areas with access to supplementary irrigation. Domestic and foreign researchers have started studying this new aerobic rice system. In this study, yield potential and water requirement of aerobic rice in Beijing region were obtained by using ORYZA2000 model as well as historical weather data from 1971 to 2000. ORYZA2000 model for simulating the growth of aerobic rice were validated by using field experiment results in Beijing. In this study, ORYZA2000 model was evaluated once more for simulating water requirement of aerobic rice. Results showed that ORYZA2000 model could present water requirements of aerobic rice for the whole growth season satisfactorily with RE (relative error) ranging from 13.0% to 21.5%, which mainly resulted from simulation inaccuracy at the stage of emergence to PI. For the 30 years, averaged yield potential was 8 573 kg hm-2 shifting from 6747 to 11278 kg ha-1, interannual variation coefficient of yield potential was 16.3%, while averaged rainfed yield was only 4 048 kg ha-1 varying from a very low level of 223 kg ha-1 to even 8 018 kg ha-1. Therefore, the variation among years was very large with the coefficient of 51.1%. The variations of rainfed yield were mostly related to the rainfall among years. The gap between rainfed yield and yield potential indicated that yield of aerobic rice in Beijing region still could be increased to some extent by supplementary irrigation. Averaged rainfall of the 30 years during the growth season was 486 mm, and it was not enough to meet the water requirement of aerobic rice, which is 712.7 mm with 8.5% variance calculated from ORYZA2000 model. Water shortage was serious in most years, and there were 15 years from 1971 to 2000 in which water shortage was higher than 250 mm, indicating the yield of aerobic rice could be increased by supplementary irrigation. Most of water was consumed at the stage from emergence to panicle initiation, which is the longest duration in the whole life of rice; so much water could be saved if some measures were taken to reduce soil evaporation at this stage. Mean water requirement density for the whole season was 5.3 mm d-1, while the highest value, 5.5 mm d-1, occurred at the stage from panicle initiation to flowering.
全 文 ：