摘要
为了减少由农业干旱引起的相关损失,基于总初级生产力(GPP)制定了干旱恢复标准,并以此量化了不同时间尺度下的农业干旱恢复时间.通过随机森林算法分析了农业干旱恢复时间的影响因素,明确了不同时间尺度下影响农业干旱恢复时间的最主要因素.结果表明,长江流域在1、6、12、24个月时间尺度下的平均干旱恢复时间分别为5.7、3.9、3.8和3.0个月,随着时间尺度的增加干旱恢复时间会逐渐降低.从空间分布状况来看,干旱恢复时间较长的地区主要分布于长江流域的西部和东南部地区.对影响因素的分析结果表明,在短时间尺度下气象变量是影响干旱恢复时间长短的最主要因素,但对较长的时间尺度而言,干旱恢复时间的长短则更多受干旱发生季节的影响.
To mitigate losses caused by agricultural drought,drought recovery standards based on gross primary productivity(GPP)are established,quantifying the recovery time of agricultural drought at different temporal scales.The random forest algorithm is employed to analyze the influencing factors of drought recovery time,determining the most significant factors affecting drought recovery duration at different time scales.The result reveals that the average drought recovery times at the 1,6,12,and 24-month time scales are 5.7,3.9,3.8,and 3.0 months,respectively.As the time scale increases,there is a gradual decrease in drought recovery time.In terms of spatial distribution,regions characterized by longer drought recovery times are mainly distributed in the western and southeastern regions of the Yangtze River Basin.The analysis on influencing factors indicates that meteorological variables are the primary determinants affecting drought recovery time in the short time scales.However,over the longer time scales,the season of drought occurrence exerts greater influence on drought recovery time.
作者
朱仟
魏庆
白郅杰
闵馨童
Zhu Qian;Wei Qing;Bai Zhijie;Min Xintong(School of Civil Engineering,Southeast University,Nanjing 211189,China)
出处
《东南大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2024年第3期675-683,共9页
Journal of Southeast University:Natural Science Edition
基金
国家自然科学基金资助项目(52009020)。
关键词
干旱恢复时间
随机森林
影响因素
长江流域
drought recovery time
random forest
influencing factor
Yangtze River Basin