Scholarlink

<p>Precipitation is the primary source of water in a river basin and plays a key role in water resource management by affecting water availability. However, large river basins like the São Francisco River Basin in Brazil show significant variability in rainfall patterns, leading to different water supply levels across sub-regions. To identify areas with similar rainfall patterns, the k-means clustering method was used on rainfall stations throughout the basin. The clusters were validated by analyzing their means, medians, and interquartile ranges. This analysis was supported by a multi-criteria assessment using two methods: one with a correlation matrix and the other with a covariance matrix. As a result, the stations were grouped into five homogeneous clusters based on precipitation. The two clusters in the southern basin were heavily influenced by the South Atlantic Convergence Zone (SACZ). In contrast, the northern clusters were mainly affected by the Intertropical Convergence Zone (ITCZ) and orographic features such as the Borborema Plateau. The coastal cluster was influenced by the sea and trade winds blowing from east to west. Multivariate analysis showed high correlation and seasonal similarity among stations within the same cluster. Clusters 4 and 5 experienced the highest average precipitation, while clusters 1 and 2 recorded the lowest, with cluster 2 being particularly notable.</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

Identification of homogeneous rainfall regions in the São Francisco River Basin, Brazil, through cluster analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月16日

查看全文
  • Anderson de Oliveira Pinheiro,
  • Demetrius David da Silva,
  • Michel Castro Moreira,
  • Clívia Dias Coelho,
  • Ricardo Santos Silva Amorim

摘要

Precipitation is the primary source of water in a river basin and plays a key role in water resource management by affecting water availability. However, large river basins like the São Francisco River Basin in Brazil show significant variability in rainfall patterns, leading to different water supply levels across sub-regions. To identify areas with similar rainfall patterns, the k-means clustering method was used on rainfall stations throughout the basin. The clusters were validated by analyzing their means, medians, and interquartile ranges. This analysis was supported by a multi-criteria assessment using two methods: one with a correlation matrix and the other with a covariance matrix. As a result, the stations were grouped into five homogeneous clusters based on precipitation. The two clusters in the southern basin were heavily influenced by the South Atlantic Convergence Zone (SACZ). In contrast, the northern clusters were mainly affected by the Intertropical Convergence Zone (ITCZ) and orographic features such as the Borborema Plateau. The coastal cluster was influenced by the sea and trade winds blowing from east to west. Multivariate analysis showed high correlation and seasonal similarity among stations within the same cluster. Clusters 4 and 5 experienced the highest average precipitation, while clusters 1 and 2 recorded the lowest, with cluster 2 being particularly notable.