陕西省农业干旱风险评估方法研究

发布时间：2018-08-09 18:33

【摘要】：干旱灾害严重制约陕西省农业经济发展,素有“十年九旱”之称,农业干旱风险评估是科学制定抗旱减灾策略措施的基础工作。本文在分析陕西省干旱特征基础之上,采用主成分分析法和专家咨询法相结合的方式确定评价指标,运用层次分析法确定指标权重,基于灾害风险理论,构建农业干旱风险综合指标评估模型,开展陕西省农业干旱风险评估研究,主要研究结果如下:(1) 1994-2013年陕西省冬小麦需水关键期水分亏缺指数呈线性上升趋势,冬小麦生长面临着越来越严重的干旱威胁。近20年来,陕西省降水距平百分率指数正负距平呈波动变化,三大自然区域(陕北、关中、陕南)变化趋势基本一致,在2003年和2011年出现2次较大正距平,但总体来看,三大区域出现负距平频率比正距平高,分别为陕北52.5%、关中65%、陕南61.7%,说明研究期间陕西省降水总量存在下降趋势,降水量下降增加了农业干旱灾害发生的风险。(2)从主成分分析的结果可以看出,前4个主成分方差累计贡献率为91.80% (大于85%),表明前4个主成分包含了全部测量指标所具有的主要信息。提取特征向量分量值大于0.2的指标,得出第一、第二、第三、第四主成分分别反映了 9个、6个、4个和5个指标的信息。另外,主成分分析法用于指标筛选,能减少评价指标个数,体现出一定的指标筛选优势,但容易忽略指标间的相关性。而专家咨询法依靠专家丰富的理论知识和实践经验,可以准确筛选出适合当地风险评估的指标,但易受专家知识层面和个人爱好等主观因素影响。以主成分分析法和专家咨询法相结合的方式,能更为科学的选取评价指标。(3)运用自然灾害风险理论构建的综合指标农业干旱风险评估模型,能较为准确评估陕西省农业干旱风险,结果表明陕西省农业干旱风险大体呈现从南向北逐渐递增空间分布趋势。并且,2009-2013年期间,陕北地区农业干旱风险略呈下降趋势、关中地区基本稳定不变、陕南地区呈急剧升高态势。原因主要是受降水量、粮食播种面积、产水模数和经济水平4个因素共同影响。
[Abstract]:Drought disaster seriously restricts the development of agricultural economy in Shaanxi Province, known as "ten years and nine years drought", agricultural drought risk assessment is the basic work of scientific formulation of drought and disaster reduction strategies and measures. Based on the analysis of drought characteristics in Shaanxi Province, this paper uses the method of principal component analysis and expert consultation to determine the evaluation index, and uses the analytic hierarchy process to determine the index weight, which is based on the theory of disaster risk. The main results are as follows: (1) the water deficit index of winter wheat in Shaanxi Province increased linearly during the critical period of winter wheat water demand from 1994 to 2013. Winter wheat growth is facing more and more serious drought threat. In the past 20 years, the positive and negative anomalies of precipitation anomaly percentage index in Shaanxi Province have fluctuated, and the three natural regions (North Shaanxi, Guanzhong and South Shaanxi) have basically the same trend of change. There were two large positive anomalies in 2003 and 2011, but generally speaking, The frequency of negative anomaly in the three regions is higher than that of positive anomaly, which is 52.5 in Northern Shaanxi, 65in Guanzhong and 61.7 in Southern Shaanxi, indicating that the total precipitation in Shaanxi Province decreased during the study period. The decrease of precipitation increases the risk of agricultural drought disaster. (2) from the results of principal component analysis, we can see, The cumulative contribution rate of the first four principal components is 91.80% (> 85%), which indicates that the first four principal components contain the main information of all the measurement indexes. When the component value of the feature vector is greater than 0.2, the first, second, third and fourth principal components reflect the information of 9, 6, 4 and 5 indexes respectively. In addition, principal component analysis (PCA) can reduce the number of evaluation indexes and reflect the advantages of index selection, but it is easy to ignore the correlation between indicators. Depending on the experts' abundant theoretical knowledge and practical experience, the expert consultation method can accurately screen out the suitable local risk assessment index, but it is vulnerable to subjective factors such as expert knowledge level and personal preference. With the combination of principal component analysis and expert consultation, the evaluation index can be selected more scientifically. (3) A comprehensive index agricultural drought risk assessment model based on natural disaster risk theory is established. The results show that the agricultural drought risk in Shaanxi Province is increasing gradually from south to north. During 2009-2013, the risk of agricultural drought in northern Shaanxi decreased slightly, and remained stable in Guanzhong, and increased sharply in southern Shaanxi. The reasons are mainly affected by precipitation, grain sowing area, water yield modulus and economic level.
【学位授予单位】：西安理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：S423

【参考文献】