Application of LSTM models and recurrent neural networks in flood detection using satellite images: a short narrative review

Christopher Abraham Arevalo-Erosa Jose Emilio Ek-Chay Rodrigo Kalel Rodriguez-Chavez Máximo Arturo Guzmán-Ascencio Raul Humberto Rodriguez-Mendez Romell Gerardo Mejia-Ramón Misael Garcia
Vol. 4 No. 1 (2026)

Abstract

The detection of floods using satellite imagery is essential for strengthening monitoring systems and early-warning mechanisms in vulnerable regions. This Short Narrative Review examines the effectiveness of Long Short-Term Memory (LSTM) models and other recurrent neural networks (RNNs) in comparison with traditional methods and deep learning architectures based on convolutional neural networks (CNNs). The reviewed studies indicate that CNNs remain the most widely used and report strong performance in the spatial segmentation of flooded areas, particularly with architectures such as U-Net and DeepLab. Moreover, some articles highlight that LSTMs provide advantages by integrating temporal information and hydrological variations that CNNs do not capture, which may support the interpretation of events with temporal dynamics. The most common limitations include environmental variability, sensor dependence, and lack of methodological consistency across studies. Although the evidence is still limited, the reviewed literature suggests that LSTMs may serve as a complementary approach for flood detection based on satellite imagery.

Download PDF

References

  1. P. A. B. Andrade, P. E. C. Zamora, A. E. C. Lara, and J. P. P. Piedra, “A comprehensive evaluation of AI techniques for air quality index prediction: RNNs and transformers,” Ingenius, 2025. doi: 10.17163/ings.n33.2025.06.
  2. E. Heidari, V. Samadi, and A. A. Khan, “Leveraging recurrent neural networks for flood prediction and assessment,” Hydrology, vol. 12, no. 4, 2025. doi: 10.3390/hydrology12040090.
  3. A. M. Rugină, “The use of recurrent neural networks (S-RNN, LSTM, GRU) for flood forecasting based on data extracted from classical hydraulic modeling,” Modelling in Civil Environmental Engineering, vol. 18, no. 3, pp. 1–18, 2023. doi: 10.2478/mmce-2023-0011.
  4. C. Chen, D. Luan, S. Zhao, Z. Liao, Y. Zhou, J. Jiang, and Q. Pei, “Flood discharge prediction based on remote-sensed spatiotemporal features fusion and graph attention,” Remote Sensing, vol. 13, no. 24, 2021. doi: 10.3390/rs13245023.
  5. G. S. Nijaguna, D. R. Manjunath, M. Abouhawwash, S. S. Askar, D. K. Basha, and J. Sengupta, “Deep learning-based improved WCM technique for soil moisture retrieval with satellite images,” Remote Sensing, vol. 15, 2023. doi: 10.3390/rs15082005.
  6. H. S. Munawar, F. Ullah, S. Qayyum, S. I. Khan, and M. Mojtahedi, “UAVs in disaster management: Application of integrated aerial imagery and convolutional neural network for flood detection,” Sustainability, vol. 13, 2021. doi: 10.3390/su13147547.
  7. F. Pech-May, R. Aquino-Santos, and J. Delgadillo-Partida, “Sentinel-1 SAR images and deep learning for water body mapping,” Remote Sensing, vol. 15, pp. 1–22, 2023. doi: 10.3390/rs15123009.
  8. Z. Boulouard, M. Ouaissa, M. Ouaissa, F. Siddiqui, M. Almutiq, and M. Krichen, “An integrated Artificial Intelligence of Things environment for river flood prevention,” Sensors, vol. 22, pp. 1–20, 2022. doi: 10.3390/s22239485.
  9. X. Wu, Z. Zhang, S. Xiong, W. Zhang, J. Tang, Z. Li, B. An, and R. Li, “A near-real-time flood detection method based on deep learning and SAR images,” Remote Sensing, vol. 15, 2023. doi: 10.3390/rs15082046.
  10. H. Wu, H. Song, J. Huang, H. Zhong, R. Zhan, X. Teng, Z. Qiu, M. He, and J. Cao, “Flood detection in dual-polarization SAR images based on multi-scale DeepLab model,” Remote Sensing, vol. 14, pp. 1–22, 2022. doi: 10.3390/rs14205181.
  11. L. Zhang and J. Xia, “Flood detection using multiple Chinese satellite datasets during 2020 China summer floods,” Remote Sensing, vol. 14, p. 51, 2021. doi: 10.3390/rs14010051.
  12. S. Priscillia, C. Schillaci, and A. Lipani, “Flood susceptibility assessment using artificial neural networks in Indonesia,” Artificial Intelligence in Geosciences, vol. 2, pp. 215–222, 2021. doi: 10.1016/j.aiig.2022.03.002.
  13. R. Al-Ruzouq, A. Shanableh, R. Jena, M. B. A. Gibril, N. A. Hammouri, and F. Lamghari, “Flood susceptibility mapping using a novel integration of multi-temporal Sentinel-1 data and extreme deep learning model,” Geoscience Frontiers, vol. 15, 2024. doi: 10.1016/j.gsf.2024.101780.
  14. A. Cerbelaud, G. Blanchet, L. Roupioz, P. Breil, and X. Briottet, “Mapping pluvial flood-induced damages with multi-sensor optical remote sensing: A transferable approach,” Remote Sensing, vol. 15, pp. 1–27, 2023. doi: 10.3390/rs15092361.
  15. G. K. Wedajo, T. D. Lemma, T. Fufa, and P. Gamba, “Integrating satellite images and machine learning for flood prediction and susceptibility mapping for the case of Amibara, Awash Basin, Ethiopia,” Remote Sensing, vol. 16, p. 2163, 2024. doi: 10.3390/rs16122163.

License

This article is licensed under Creative Commons Attribution 4.0 International License (CC BY 4.0)

CC BY 4.0
← Back to articles