Flood Susceptibility Zonation Using Geospatial Frequency Ratio and Artificial Neural Network Techniques within Himalayan Terai Region: A Comparative Exploration

Flood Susceptibility Zonation Using Geospatial Frequency Ratio and Artificial Neural Network Techniques within Himalayan Terai Region: A Comparative Exploration

---

연구 분야: Strategies

---

---

학회: International Conference on Computational Technologies and Electronics

---

초록

Flooding is a widespread natural disaster affecting environment, economy, infrastructure. This research involves implementing the Frequency Ratio (FR) and GIS-based Multilayer Feed Forward Artificial Neural Network (GMFFANN) models on the lower bank of the Teesta River in the Himalayan foothills Terai Region of West Bengal. Data from historical flood reports, databases, satellite imagery, and field surveys have been adopted to develop training and testing datasets for flood susceptibility based on 10 flood affecting factors. The GMFFANN model, trained using Conjugate Gradient Decent (CGD) algorithm, outperforms the FR model in Flood Susceptibility Zonation (FSZ) scenario with 91.5% and 79% accuracy assessment by Receiver Operating Characteristics (ROC) curve in both training and testing sets. The crucial findings of this study will undoubtedly aid local officials in developing appropriate long-term management plans to reduce future losses.

---