Scholarlink

Precipitation and drought patterns have become increasingly irregular due to climate change, posing challenges for local decision-making and water management. This paper presents PATPSL (Platform for the Analysis of Precipitation and Local Drought Trends), a web-based platform that automates the analysis of precipitation and drought trends in specific geographic coordinates in Mexico. PATPSL integrates data mining, geographic interpolation (IDW), drought indices (SPI, Palmer, CFE, Foley), and machine learning models (SARIMA, kNN, regression trees), alongside an algorithm (AEPCM) for monthly precipitation trend estimation. The platform offers a user-friendly interface, downloadable reports, and an interactive dashboard powered by Power BI. This paper details the system architecture, preprocessing pipeline, model evaluation, and platform deployment.

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

A Web-Based Decision Support Tool for Local Precipitation Trends in Mexico Using Data Science and Spatial Analysis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Rubén Calvario Pérez,
  • Carlos Enrique Mondragón Mendoza,
  • Adriana Berenice Celis Domínguez,
  • Roberto Eswart Zagal Flores,
  • Miriam Pescador Rojas,
  • Elizabeth Montero

摘要

Precipitation and drought patterns have become increasingly irregular due to climate change, posing challenges for local decision-making and water management. This paper presents PATPSL (Platform for the Analysis of Precipitation and Local Drought Trends), a web-based platform that automates the analysis of precipitation and drought trends in specific geographic coordinates in Mexico. PATPSL integrates data mining, geographic interpolation (IDW), drought indices (SPI, Palmer, CFE, Foley), and machine learning models (SARIMA, kNN, regression trees), alongside an algorithm (AEPCM) for monthly precipitation trend estimation. The platform offers a user-friendly interface, downloadable reports, and an interactive dashboard powered by Power BI. This paper details the system architecture, preprocessing pipeline, model evaluation, and platform deployment.