Gupta et al. (2025) Assessment of Flood Potential Through Rainfall Pattern Analysis

Identification

• Journal: Lecture notes in networks and systems
• Year: 2025
• Date: 2025-11-23
• Authors: Aditya Gupta, Vibha Jain
• DOI: 10.1007/978-3-032-07735-6_21

Research Groups

Thapar Institute of Engineering and Technology, Patiala, India

Short Summary

This study analyzes over a century of rainfall data across all Indian states using a Bidirectional Long Short-Term Memory (Bi-LSTM) model to understand and forecast flood potential, demonstrating high accuracy in identifying high-risk areas.

Objective

• To better understand flood-prone conditions and forecast flood potential in all Indian states by analyzing historical rainfall patterns.

Study Configuration

• Spatial Scale: All Indian states, India.
• Temporal Scale: 124 years (from 1901 to 2024).

Methodology and Data

• Models used: Bidirectional Long Short-Term Memory (Bi-LSTM) model.
• Data sources: Over a century of historical rainfall data.

Main Results

• The Bi-LSTM model effectively captures complex spatio-temporal rainfall patterns.
• The model achieved an accuracy of 96.2% in forecasting flood potential.
• It demonstrated robust predictive performance in detecting high-risk areas for flooding.

Contributions

• The application of a Bi-LSTM model to a century-long, nationwide rainfall dataset for flood potential assessment.
• Provides a highly accurate and robust predictive tool for identifying flood-prone regions in India.

Funding

• Not explicitly mentioned in the provided text.

Citation

@article{Gupta2025Assessment,
  author = {Gupta, Aditya and Jain, Vibha},
  title = {Assessment of Flood Potential Through Rainfall Pattern Analysis},
  journal = {Lecture notes in networks and systems},
  year = {2025},
  doi = {10.1007/978-3-032-07735-6_21},
  url = {https://doi.org/10.1007/978-3-032-07735-6_21}
}