To determine the influence of agricultural meteorological disasters on agriculture in Heilongjiang Province,the disaster areas associated with different types of disasters and their variation characteristics were analyzed based on the statistical data of agricultural disasters from 1983 to 2013 in Heilongjiang Province,China.The moving average and the Mann-Kendall test were applied to identify the variation trends of drought,flooding,hailstorms and freezing(based on the disaster ratio and the disaster intensity index).Then,the Morlet wavelet analysis method was used to identify the periodicity of these four kinds of agricultural meteorological disasters.Finally,a fuzzy comprehensive evaluation method was adopted to analyze the degrees of agricultural loss induced by these disasters.The following results were obtained:1)The disaster ratio and disaster intensity index for drought exhibited increasing trends;the disaster ratio and disaster intensity index for flooding exhibited decreasing trends;for hailstorms,the disaster ratio exhibited no obvious trend of change,whereas the disaster intensity index exhibited an increasing trend;and for freezing,the disaster ratio also exhibited no obvious trend of change,whereas the disaster intensity index exhibited a decreasing trend.2)Mutation points were observed in the disaster ratio series for drought,flooding and hailstorms,whereas no mutation point was evident in the disaster ratio series for freezing.3)Multiple time-scale characteristics were observed in the disaster ratio series for all four types of agricultural meteorological disasters.Furthermore,the disaster ratio series for the different types of disasters had different main periodicities.4)From the perspective of the degree of agricultural loss induced by each type of disaster,drought was identified as the most severe type of agricultural meteorological disaster,followed by flooding,freezing,and hailstorms.The degree of agricultural loss caused by each type of disaster was different during different periods.Finall
针对降水随机性较强、影响因素复杂、单一模型预测精度低的特点,采用集对分析法,研究降水量与气象影响因子的关系。将基于密度参数的径向基函数人工神经网络模型与灰色模型相结合,利用信息熵权重法计算2个单一模型的权重,构建基于信息熵的集合模型(Combing model based on information entropy,IE-CM),用于三江平原友谊农场的降水量预测。研究结果表明,与单一模型相比,IE-CM模型预测结果的决定系数、平均相对误差及均方根误差较单一模型均有所提高,分别为0.99、10.655%和3.03 mm,预测结果的合格率为83.3%,均满足水文预测要求。
