| 摘要: |
| ［目的］以湖南省油菜春季涝渍灾害为例，创建基于涝渍过程的逐日灾变等级指标、灾害影响量化评价与灾损量化评估模型，探索区域农作物涝渍灾变动态监测评估的天气学方法。［方法］基于湖南省油菜春季涝渍灾变过程解析，以过程灾变判别指标为基础，采用基于假设的滚动模拟寻优、实际灾情验证与个例分析等方法，厘定轻度涝渍与重度涝渍灾变的最佳阈值，构建油菜春季涝渍过程逐日灾变等级指标； 利用多元回归等方法，构建对应的灾害影响量化评价模型和减产率量化评估模型； 基于个例分析，验证指标及模型结果与历史灾情记录的吻合情况。［结果］湖南省轻度与重度涝渍灾变的最佳阈值为144； 不同县涝渍灾变等级阈值存在一定差异，平均受灾频率越低，洪涝脆弱性越低，防灾减灾能力越强，阈值越高； 基于受灾天数和重灾天数的灾害影响指数都表现为结荚期涝渍对油菜减产率影响更大，且两者的空间分布形势都与各县年平均减产率的空间分布形势基本一致； 个例中，基于指标的全省重灾站数百分比的时间演变与实际灾情记录一致，减产率量化评估模型的结果也与实际灾损相匹配。［结论］油菜涝渍灾变等级指标、灾害影响指数及减产率量化评估模型，实现了对涝渍灾变过程等级的动态监测、影响与灾损的量化评估，为基于天气学方法开展区域油菜涝渍灾变等级的动态监测提供了理论支持和方法支撑，同时为历史灾情资料的补充和量化及灾情记录的再分析提供了可行的思路。 |
| 关键词: 涝渍灾害灾变过程等级指标灾损评估油菜 |
| DOI： |
| 分类号: |
| 基金项目:中国气象科学研究院科技发展基金项目“气候变化背景下粮油作物气象灾害指标构建及演变规律研究”（2018KJ012）；“十二五”国家科技支撑计划课题“洪涝灾害监测预警与防控与应急关键技术研究与应用”（2012BAD20B02） |
|
| CATASTROPHE CLASSIFICATION INDICATOR AND YIELD LOSS ANALYSIS OF OILSEED RAPE VERNAL WATERLOGGING IN HUNAN PROVINCE |
|
Fan Yuxian1,2, Huo Zhiguo1,3※, Shang Ying1
|
|
1. Chinese Academy of Meteorological Sciences，Beijing 100081, China;2. Meteorological Bureau of Yiyang City, Yiyang, Hunan 413000, China;3. Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing，Jiangsu 210044, China
|
| Abstract: |
| Taking spring waterlogging disaster of oilseed rape in Hunan province as an example, this research constructs a waterlogging damage valuation model and a quantitative assessment model for the rape yield lose through the process based classification indicator for dynamic waterlogging damage monitor, in order to explore a meteorological method for regional crop waterlogging process monitoring and evaluation. Based on the existing disaster identification index of rape spring waterlogging process in Hunan province, the methods of rolling simulation optimization, actual verification and case analysis were adopted to determine the optimal threshold for light and severe waterlogging, establish the original index to a process based rapeseed spring waterlogging catastrophic grades index. At the same time, the relationship between the yield loss and different grades waterlogging frequency was explored by multiple regression analysis, and the yield loss quantification estimation model and waterlogging damage valuation model were constructed. Afterwards, the match degree of index and models with historical disaster records had been tested through a case study. The results were showed as follows.1.44 was the optimal threshold to distinguish the light and severe rape vernal waterlogging in Hunan. This threshold varied with counties, and higher threshold value was detected in areas with less average waterlogging frequency, lower flood vulnerability and better disaster prevention and mitigation capabilities. Waterlogging in rape pod stage had the greatest influence on the rape yield according to the disaster impact indexes, including index based on duration of damage and index based on duration of severe damage. Similar spatial distribution of rape yield reduction rate had been detected base on the two indexes at the county scale. The temporal characteristic of calculated severe damage percentage based on grades index was consistent with the actual records in cases, and the result of quantification estimation model of yield reduction rate was also matched with the actual losses. Generally, the process based classification indicator for dynamic waterlogging monitor, the waterlogging damage valuation model and the quantitative assessment model for the rape yield loss performed well in the implementation of regional dynamic monitoring and quantitative assessment of rape waterlogging damage. The results can provide theoretical support for regional monitoring and assessment of oilseed rape vernal waterlogging dynamically. Additionally, it also provides a feasible approach for the supplement and reanalysis of historical disaster records. |
| Key words: agricultural waterlogging catastrophe process classification indicator damage assessment oilseed rape |
