One-dimensional deep learning driven geospatial analysis for flash flood susceptibility mapping: a case study in North Central Vietnam

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 17(5), P. 4419 - 4440

Published: July 6, 2024

Abstract Flash floods rank among the most catastrophic natural disasters worldwide, inflicting severe socio-economic, environmental, and human impacts. Consequently, accurately identifying areas at potential risk is of paramount importance. This study investigates efficacy Deep 1D-Convolutional Neural Networks (Deep 1D-CNN) in spatially predicting flash floods, with a specific focus on frequent tropical cyclone-induced Thanh Hoa province, North Central Vietnam. The 1D-CNN was structured four convolutional layers, two pooling one flattened layer, fully connected employing ADAM algorithm for optimization Mean Squared Error (MSE) loss calculation. A geodatabase containing 2540 flood locations 12 influencing factors compiled using multi-source geospatial data. database used to train check model. results indicate that model achieved high predictive accuracy (90.2%), along Kappa value 0.804 an AUC (Area Under Curve) 0.969, surpassing benchmark models such as SVM (Support Vector Machine) LR (Logistic Regression). concludes highly effective tool modeling floods.

Language: Английский

---

Improving the explainability of CNN-LSTM-based flood prediction with integrating SHAP technique

Ecological Informatics, Journal Year: 2024, Volume and Issue: 84, P. 102904 - 102904

Published: Nov. 17, 2024

Language: Английский

---

Assessing future changes in flood susceptibility under projections from the sixth coupled model intercomparison project: case study of Algiers City (Algeria)

Natural Hazards, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 2, 2024

Language: Английский

---

A novel flood conditioning factor based on topography for flood susceptibility modeling

Geoscience Frontiers, Journal Year: 2024, Volume and Issue: 16(1), P. 101960 - 101960

Published: Nov. 1, 2024

Language: Английский

---

Revisiting the historical tritium levels in precipitation in Greece – Preliminary assessment of groundwater transit times

Journal of Environmental Radioactivity, Journal Year: 2025, Volume and Issue: 282, P. 107619 - 107619

Published: Jan. 17, 2025

Language: Английский

---

Recent advances and future challenges in urban pluvial flood modelling

Urban Water Journal, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 25

Published: Jan. 20, 2025

Urban pluvial floods are characterised by a number of features such as the high spatial and temporal resolution needed to capture their dynamics, complexity dual drainage systems lack sewer data availability, which make them very different from other types floods, like coastal or fluvial increase difficulty modelling them. As consequence, most flood management plans do not include rigorous evaluation urban risk. In this paper, we give comprehensive view current state modelling, restricted mere description mathematical approaches, but also including relevant that should be considered in validation studies have been performed date our vision on challenges addressed near future.

Language: Английский

---

Distinguishing Lithofacies of Flysch Formations Using Deep Learning Models: Integrating Remote Sensing Data with Morphological Indexes

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(3), P. 422 - 422

Published: Jan. 26, 2025

The main objective of the present study was to develop an integrated approach combining remote sensing techniques and U-Net-based deep learning models for lithology mapping. methodology incorporates Landsat 8 imagery, ALOS PALSAR data, field surveys, complemented by derived products such as False Color Composites (FCCs), Minimum Noise Fraction (MNF), Principal Component Analysis (PCA). Dissection Index, a morphological index, calculated characterize geomorphological variability region. Three variations U-Net architecture, Dense U-Net, Residual Attention were implemented evaluate performance in lithological classification. Validation conducted using metrics accuracy, precision, recall, F1-score, mean intersection over union (mIoU). results highlight effectiveness model, which provided highest mapping accuracy superior feature extraction delineating flysch formations associated units. This demonstrates potential integrating data with advanced machine enhance geological challenging terrains.

Language: Английский

---

A flash flood susceptibility prediction and partitioning method based on GeoDetector and random forest

Stochastic Environmental Research and Risk Assessment, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 8, 2025

Language: Английский

---

A real-time prediction model for instantaneous dam-break flood evolution of concrete gravity dams based on attention mechanism and spatiotemporal multiple features

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 150, P. 110616 - 110616

Published: March 23, 2025

Language: Английский

---

Spatial Prediction of Fluvial Flood in High-Frequency Tropical Cyclone Area Using TensorFlow 1D-Convolution Neural Networks and Geospatial Data

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(22), P. 5429 - 5429

Published: Nov. 20, 2023

Fluvial floods endure as one of the most catastrophic weather-induced disasters worldwide, leading to numerous fatalities each year and significantly impacting socio-economic development environment. Hence, research new methods algorithms focused on improving fluvial flood prediction devising robust management strategies are essential. This study explores assesses potential application 1D-Convolution Neural Networks (1D-CNN) for spatial in Quang Nam province, a high-frequency tropical cyclone area central Vietnam. To this end, geospatial database with 4156 locations 12 indicators was considered. The ADAM algorithm MSE loss function were used train 1D-CNN model, whereas popular performance metrics, such Accuracy (Acc), Kappa, AUC, measure performance. results indicated remarkable by achieving high accuracy metrics Acc = 90.7%, Kappa 0.814, AUC 0.963. Notably, proposed model outperformed benchmark models, including DeepNN, SVM, LR. achievement underscores promise innovation brought realm susceptibility mapping floods.

Language: Английский

---