Wu et al. (2026) An ANN-Derived Model for Estimating Hourly Storm Patterns with Daily Precipitation Based on Climate Change-Induced Rainstorms

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Water
• Year: 2026
• Date: 2026-06-11
• Authors: Shiang‐Jen Wu, Yeh-Shiun Lu
• DOI: 10.3390/w18121432

Research Groups

Not specified

Short Summary

The study develops the SMESPHRDY model, utilizing Artificial Neural Networks (ANN) to estimate hourly storm patterns from daily rainfall data, achieving high accuracy particularly for 2-day and 3-day events.

Objective

• To propose and validate a framework (SMESPHRDY) for estimating hourly-based storm patterns (defined as sets of cumulative dimensionless rainfalls) based on daily storm patterns.

Study Configuration

• Spatial Scale: Miaoli City, Northern Taiwan (grid resolution)
• Temporal Scale: 1979–2099 (simulated period)

Methodology and Data

• Models used: SMESPHRDY (incorporating Artificial Neural Networks - ANN)
• Data sources: AR5 climate-induced simulated rainstorms

Main Results

• Simulated rainstorms were categorized into three distinct types based on duration: 1-day, 2-day, and 3-day rainy events.
• The ANN-derived relationship between hourly and daily cumulative rainfall demonstrated low bias.
• The model showed high validation accuracy and reliability, with the best performance observed in 2-day and 3-day rainy events.

Contributions

• Introduces a "smart model" approach to provide high-resolution hourly rainfall estimates from daily series, enhancing the ability to analyze storm patterns with finer temporal resolution.

Funding

Not specified

Citation

@article{Wu2026ANNDerived,
  author = {Wu, Shiang‐Jen and Lu, Yeh-Shiun},
  title = {An ANN-Derived Model for Estimating Hourly Storm Patterns with Daily Precipitation Based on Climate Change-Induced Rainstorms},
  journal = {Water},
  year = {2026},
  doi = {10.3390/w18121432},
  url = {https://doi.org/10.3390/w18121432}
}