闽东粮食产量气象风险评估与分析

加强粮食气象灾害风险的分析和评估工作,对农业种植结构调整、防灾减灾和可持续发展具有重要意义。文章采用1949~2013年闽东9县市区粮食产量资料,应用线性拟合方法计算出粮食气象产量增长率时间序列作为分析对象,通过经验正交分解(EOF)、线性倾向估计、概率分布函数拟合及检验、聚类分析等气候统计学方法,分析闽东粮食气象产量增长率空间与时间变化特点,减产率等级分布特征及风险区划。结果表明,闽东山区粮食增长率总体要高过沿海县市,增产趋势更明显些。粮食增长率的空间特征值与年平均气温、日照时数为显著负相关,山区与沿海的气候差异与粮食增长率密切相关。9县市粮食增长率具有斜"L"型年际变化特征,可分为下降、上升、缓降、持平等4个阶段。聚类分析表明闽东粮食具有显著增产、一般增产、一般减产和显著减产等4类年型,增产与减产年份发生概率分别为54.3%、45.7%。沿海县市轻、中度减产发生概率要大于山区县,而重度减产发生概率则相反,山区县减产风险普遍高于沿海县市区。闽东总体增产年份比减产年份略多,但减产年份发生概率也较大,农业气象灾害造成减产危害不可忽视,因此加强粮食生产的农业气象防灾减灾工作十分重要。

THE METEOROLOGICAL RISK ASSESSMENT OF GRAIN YIELD IN NORTHEASTERN FUJIAN PROVINCE OF CHINA

To strengthen the assessment and analysis of meteorological disaster risk is significant for structural adjustment of agricultural plantation, disaster prevention and reduction, and sustainable development of agriculture in an area. Based on the grain yield data from 1949 to 2013 year in 9 counties of northeastern Fujian province in China, This paper used linear regression fitting method to calculate the time series of the grain meteorological yield increasing rate object, and then used the meteorological statistical methods of empirical orthogonal decomposition (EOF), linear trend estimation, the probability distribution function fitting and testing and cluster analysis to analyze the changing characteristics of spatial and temporal distribution of the grain yield increasing rate, yield reduction rate and the risk zoning in northeastern Fujian province. The results showed that the overall increasing rate of grain yield in mountain counties was higher than that in coastal counties. The increasing trend of grain yield was more obvious in mountain counties. The spatial characteristic value of the grain yield increasing rate was significantly negative correlation with the annual average temperature and sunshine hours, the climate differences between the mountain and coastal area were closely related to grain growth rate. The temporal variation trend of the grain increasing rate showed the inclined "L" type in 9 counties and can be divided into four stages, i.e., fall, rise, slow down and flat, from 1949 to 2013. The cluster analysis showed that there were 4 types of grain yield variation year, such as the significant increasing, general increasing, general reduction and significant reduction, the occurrence probabilities of the years of increasing yield or reduction yield were 54.3%, 45.7% respectively. The occurrence probabilities of mild, moderate grade of yield reduction in coastal counties were higher than that in mountain counties, while the occurrence probabilities of severe grade of yield reduction were opposite, the risk of yield reduction in mountain counties was generally higher than that in coastal counties. The yield increasing years were slightly more than yield reduction years in northeastern Fujian province, but occurrence probability of yield reduction year was high due to the agro-meteorological disasters. So it was very important to strengthen the agricultural meteorological disaster prevention.