Scholarlink

<p>Clustering of Intensity–Duration–Frequency (IDF) curves is a key tool in hydrology and water engineering, essential for analyzing and predicting extreme precipitation for various durations and return periods. These curves assist engineers in designing drainage systems and flood control measures tailored to precipitation intensities, playing a critical role in water and flood resource management. This study analyzed 24-h precipitation data from 1982 to 2023 for 72 meteorological stations across Iran, using a 6-h time step. Precipitation intensities for Short (2 and 5&#xa0;years), Medium (25 and 50&#xa0;years), and Long (100 and 200&#xa0;years) return periods were calculated using the most suitable probability distributions. Subsequently, clustering algorithms, including Fuzzy C-Means (FCM), K-Means, and Hierarchical methods, were applied to classify Iran into five homogeneous regions based on precipitation intensities. The results revealed that Iran could be divided into five areas with distinct precipitation characteristics, with Region 3 having the highest and Region 1 having the lowest precipitation intensities. Generalized Extreme Value (GEV) and Log-Normal distributions were also identified as the dominant probability distributions for precipitation analysis. Among the clustering methods, the Hierarchical algorithm demonstrated the best performance. The findings of this study have significant applications in flood management and engineering design, such as dams and drainage networks. Analyzing precipitation intensities and probability distributions can enhance flood risk prediction and contribute to reducing the human and economic impacts of floods.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Classification of design precipitation intensities for different return periods in Iran

  • 期刊论文

  • 发表日期：

    2025年3月3日

查看全文
  • Alireza Abdolmanafi,
  • Bahram Saghafian,
  • Saleh Aminyavari

摘要

Clustering of Intensity–Duration–Frequency (IDF) curves is a key tool in hydrology and water engineering, essential for analyzing and predicting extreme precipitation for various durations and return periods. These curves assist engineers in designing drainage systems and flood control measures tailored to precipitation intensities, playing a critical role in water and flood resource management. This study analyzed 24-h precipitation data from 1982 to 2023 for 72 meteorological stations across Iran, using a 6-h time step. Precipitation intensities for Short (2 and 5 years), Medium (25 and 50 years), and Long (100 and 200 years) return periods were calculated using the most suitable probability distributions. Subsequently, clustering algorithms, including Fuzzy C-Means (FCM), K-Means, and Hierarchical methods, were applied to classify Iran into five homogeneous regions based on precipitation intensities. The results revealed that Iran could be divided into five areas with distinct precipitation characteristics, with Region 3 having the highest and Region 1 having the lowest precipitation intensities. Generalized Extreme Value (GEV) and Log-Normal distributions were also identified as the dominant probability distributions for precipitation analysis. Among the clustering methods, the Hierarchical algorithm demonstrated the best performance. The findings of this study have significant applications in flood management and engineering design, such as dams and drainage networks. Analyzing precipitation intensities and probability distributions can enhance flood risk prediction and contribute to reducing the human and economic impacts of floods.