中国农业气象 ›› 2024, Vol. 45 ›› Issue (6): 629-642.doi: 10.3969/j.issn.1000-6362.2024.06.006

• 农业气象灾害 栏目 • 上一篇    下一篇

四川盆区油菜花期低温阴雨灾损评估及风险区划

赵艺,郭翔,王鑫,杨德胜,王明田   

  1. 1.四川省农业气象中心,成都 610072;2.南方丘区节水农业研究四川省重点实验室,成都 610066;3.四川省气象台,成都 610072
  • 收稿日期:2023-06-29 出版日期:2024-06-20 发布日期:2024-06-17
  • 作者简介:赵艺,E-mail:524968359@qq.com
  • 基金资助:
    高原与盆地暴雨旱涝灾害四川省重点实验室科技发展基金项目(SCQXKJQN202124;省重实验室−2018−重点−05−03)

Loss Assessment and Risk Zoning of Low-temperature and Rainy Weather Disasters on Rape Yield during the Flowering Period in the Sichuan Basin

ZHAO Yi, GUO Xiang, WANG Xin, YANG De-sheng, WANG Ming-tian   

  1. 1.Agro-meteorological Center of Sichuan Province, Chengdu 610072, China; 2.Key Laboratory of Water-Saving Agriculture Research in Southern Hilly Areas, Chengdu 610066; 3.Sichuan Meteorological Observatory, Chengdu 610072
  • Received:2023-06-29 Online:2024-06-20 Published:2024-06-17

摘要: 花期低温阴雨是四川盆区油菜生产的主要气象灾害之一。本研究利用四川盆区1961-2020年101个气象站的逐日气象资料,1983-2000年油菜产量资料和干旱、低温冷害、连阴雨、大风、冰雹灾情资料,筛选花期低温阴雨灾害年份,采用数理统计方法获取花期低温阴雨产量灾损率和致灾因子,通过相关性分析确定油菜花期低温阴雨致灾指标,并利用2001-2020年灾情资料进行检验;基于筛选出的低温、连阴雨关键致灾因子,采用回归分析法建立油菜花期低温阴雨灾害损失评估模型,并进行回代和预测检验,分析灾害风险区划和变化趋势。结果表明:(1)日平均气温≤7℃、过程持续天数≥1d是四川盆区油菜花期低温致灾指标,日平均气温≤7℃的负积温和≥3d过程持续降水量共同组成低温阴雨灾害的关键致灾因子;选用2001-2020年油菜花期低温冷害发生情况对指标进行检验,与实际情况相符。(2)选用1983-2000年的气象、产量、灾情数据建立四川盆区油菜花期低温阴雨灾损评估模型,模型对轻度、中度灾害损失评估等级与实际等级相同或相差1级的评估准确率在96%以上;重度、特重评估等级与实际等级相同的准确率为0,与实际等级相差1级的准确率为75%和0。选用2001-2020年油菜花期低温阴雨灾害发生情况对模型检验,与实际情况基本相符。(3)1961-2020年四川盆区油菜花期低温阴雨灾害风险偏高区域主要分布于盆区西南部、南部及东北部,以中-高风险为主;灾害风险偏低的区域集中于盆区西北部及中部,以低-次低风险为主。(4)气候变暖背景下,四川盆区油菜花期低温阴雨灾害高风险区域呈减少趋势,低风险区域呈增加趋势。综上分析,四川盆区油菜花期低温冷害指标结果可靠,低温阴雨灾损评估模型能够较好地评估灾害损失,可应用于农业气象业务服务;灾害风险区划结果可为四川油菜生产布局提供参考依据。

关键词: 四川盆区, 油菜花期, 低温阴雨, 灾损评估, 风险区划

Abstract: Low-temperature and rainy weather during the rape flowering period is one of the meteorological disasters affecting the rape-yield in the Sichuan Basin region. In this study, the daily meteorological data from 101 weather stations in the Sichuan Basin from 1961 to 2020, as well as the rape yield data and the disaster information (drought, low-temperature cold damage, consecutive rainy weather, gale and hail) from 1983 to 2000, are utilized. The years with low-temperature and rainy weather disasters during the rape flowering period are selected, and then the yield loss rate and disaster-inducing factors of low-temperature rainy weather during the rape flowering period are obtained by utilizing mathematical-statistical methods. Furthermore, the disaster-inducing indicators of low-temperature and rainy weather disasters during the rape flowering period are determined through correlation analysis, and are validated by using the disaster information data from 2001 to 2020. Based on the selected key disaster-inducing factors of low-temperature and continuous rainy weather, an assessment model on the losses caused by low-temperature and rainy disasters during the rape flowering period by using the regression analysis method was built. Then, the back substitution and prediction test are conducted to analyze the disaster risk zoning and variation trends. The results indicate that the daily average temperature ≤7°C and the duration of disaster events ≥1d are the disaster-inducing indicators of low-temperature during the rape flowering period in the Sichuan Basin. The ≤7°C negative accumulated temperature and the accumulated precipitation of the persistent rainfall event with ≥3d duration are the key disaster-inducing factors of low-temperature and rainy weather disasters. The indicators are examined with the low-temperature cold damage data during the rape flowering period from 2001 to 2020, and the results align with the actual situation. Based on the meteorological data, rape yield and disaster information from 1983 to 2000, an assessment model on the losses caused by low-temperature and rainy weather during the rape flowering period in the Sichuan Basin is established. The back substitution results indicate that for the light and moderate assessed levels, the assessment accuracy (the difference between assessed level and actual level is one or zero) is over 96%. For the severe and extremely severe assessed levels, the assessment accuracies with zero level difference are both 0, and the assessment accuracies with one level difference are 75% and 0, respectively. Then, by using the assessment model the low-temperature and rainy weather disasters during the rape flowering period in each year of 2001–2020 are predicted, and the results are consistent with the actual situation. From 1961 to 2020, in the southwestern, southern and northeastern parts of the basin, the risk levels of low-temperature and rainy weather disasters are mainly medium or high. The regions with lower risk level (low or sub-low) are mainly concentrated in the northwestern and central parts of the Sichuan Basin. Under the background of climate warming, the high-risk areas for low-temperature and rainy weather disasters during the rape flowering period in the Sichuan Basin show a decreasing trend, while low-risk areas show an increasing trend. In summary, the indicators for low-temperature cold damage during the rape flowering period in the Sichuan Basin are reliable. The loss assessment model of low-temperature and rainy weather disasters can effectively evaluate disaster losses and can be applied in agricultural meteorological service operations. The results of disaster risk zoning can provide a reference for the rape production layout in Sichuan Province.

Key words: Sichuan Basin, Rape flowering period, Low-temperature and rainy weather, Hazard loss assessment, Risk zoning