华北地区农业干旱预测模型及其应用研究

对华北地区冬小麦进行了水分胁迫实验，确定了冬小麦光合作用速率对水分胁迫的响应曲线，提出了农业干旱指数和农业干旱预警指数两个基本概念，在此基础上建立了具有明确生物学机理的华北农业干旱预测数值模式。对北京、石家庄、郑州和济南1998年至2000年432旬的农业干旱模拟结果表明:农业干旱预警定性准确率为90.7%，定量准确率在87.5%左右；此外，模拟表明模式也可以对区域农业干旱进行准确有效的预测。利用1961~2000年气象资料对北京等地区历年农业干旱进行数值分析，结果表明:不同于大气干旱，在自然气象条件下，北京等地区作物生长期内几乎每年都存在农业干旱现象，特别是冬小麦灌浆至成熟期每年均存在一次较为严重的农业干旱胁迫过程，但农业干旱胁迫程度年际间存在一定波动，功率谱分析表明其具有3~6年的周期变化规律。对平均气候状况下华北地区农业干旱进行了时空动态分析，结果发现在自然条件下，华北大部分地区冬小麦4月下旬至5月下旬，即大约在冬小麦开花、灌浆至成熟期，农业干旱胁迫指数存在一种自然的逐渐加强的动态过程，这与华北地区的农业生产实践是基本一致的。

A Drought Prediction Model in North China Plain and Its Application

A drought prediction model in North China is developed based on the response curve of photosynthesis rate of winter wheat to water stress and the concept of agricultural drought stress index and of agricultural drought early warning index. The results of simulated agricultural drought for 432 dekads in 1998-2000 for Beijing, Shijiazhuang, Zhengzhou and Jinan show that the accuracy of agricultural drought trend prediction is 90.7% and the accuracy of agricultural drought quantitative prediction is 87.5%. In addition, the model is able to predict drought in a region. The model is applied to a numerical analysis for agricultural drought in 1961-2000 in Beijing. The results show that agricultural drought occurs nearly every year under normal climatic condition in Beijing, especially the severe drought occurs in the period from grain filling to maturity of winter wheat. But the severe level of drought stress in each year is different, there is a 3-6 cycle according to the power spectrum analysis. According to the temporal and spatial analysis for agricultural drought in North China under normal climatic condition, it is found that the agricultural drought stress indexes strengthen gradually from flowering, grain filling to maturity of winter wheat in North China. It conforms to the fact of agricultural production in North China.