Abstract
Urban flooding in arid regions poses severe risks to infrastructure and communities. Accurate, fine-scale mapping of flood extents and recovery trajectories is therefore essential for improving emergency response and resilience planning. However, arid environments often exhibit limited spectral contrast between water and adjacent surfaces, rapid hydrological dynamics, and highly heterogeneous urban land covers, which challenge traditional flood-mapping approaches. High-resolution, daily PlanetScope imagery provides the temporal and spatial detail needed. In this work, we introduce FM-LC, a hierarchical framework for Flood Mapping by Land Cover identification, for this challenging task. Through a three-stage process, it first uses an initial multi-class U-Net to segment imagery into water, vegetation, built area, and bare ground classes. We identify that this method has confusion between spectrally similar categories (e.g., water vs. vegetation). Second, by early checking, the class with the major misclassified area is flagged, and a lightweight binary expert segmentation model is trained to distinguish the flagged class from the rest. Third, a Bayesian smoothing step refines boundaries and removes spurious noise by leveraging nearby pixel information. We validate the framework on the April 2024 Dubai storm event, using pre- and post-rainfall PlanetScope composites. Experimental results demonstrate average F1-score improvements of up to 29% across all land-cover classes and notably sharper flood delineations, significantly outperforming conventional single-stage U-Net baselines.
Introduction
In arid environments like Dubai, urban flood mapping faces a significant "spectral ambiguity" problem. Standard satellite imagery often fails to distinguish between shallow floodwater, moist vegetation, and dark urban surfaces because they reflect light in very similar ways. While high-resolution data from providers like PlanetScope (3m resolution) offers the necessary detail, traditional deep learning models, such as a standard U-Net, frequently misclassify these similar-looking areas. This leads to "noisy" maps with blurred boundaries, making it difficult for emergency responders to identify exact flood extents or track recovery trajectories.
The framework was validated using a real-world case study: the historic April 2024 Dubai storm. The study area covered approximately 1,535 $km^2$ along the southeastern coast of the Persian Gulf, utilizing 8-band PlanetScope multispectral imagery at 3m resolution.
Methodology
The FM-LC (Flood Mapping by Land Cover identification) framework employs a hierarchical approach in three sequential stages:
- Initial Multi-class Segmentation: A U-Net with a ResNet-34 encoder performs a broad classification of the scene into four categories: water, vegetation, built area, and bare ground.
- Expert Refinement: The system identifies which classes the first model is most "confused" about (e.g., mistaking water for vegetation). It then deploys a lightweight, binary "expert" model trained specifically to distinguish between those two problematic classes, overriding the initial coarse labels.
- Bayesian Smoothing: To clean up the final output, the framework applies a post-processing step based on Markov Random Fields. This step looks at a pixel's neighbors to ensure spatial consistency, which sharpens boundaries and removes spurious noise.
By calling in an "expert" model for the most confusing pairs, the framework achieves much higher precision without needing a massive, complex architecture. Furthermore, incorporating local neighborhood information via Bayesian smoothing is critical for delineating narrow water bodies and scattered urban vegetation.
Experimental Results
The FM-LC framework significantly outperformed standard single-stage U-Net baselines across multiple metrics. The final model reached an Overall Accuracy of 96.09% and a Kappa statistic of 0.939.
Key performance highlights include:
- Accuracy Boost: An average F1-score improvement of 12% overall, with specific class improvements (like vegetation) jumping by as much as 29%.
- Qualitative Sharpness: Visual results show much cleaner delineations of flood boundaries, successfully identifying narrow channels and small patches of water that baseline models missed or blurred.
As shown in the time-series analysis, the model effectively tracks the flood extent and subsequent recession over a 10-day period, providing a clear picture of the recovery trajectory.
Conclusion and Future Work
The FM-LC framework demonstrates that a hierarchical approach combining general segmentation with expert refinement is highly effective for flood mapping in challenging arid environments. However, some limitations remain. Currently, the "vulnerable class" requiring an expert model is identified through manual inspection; future work aims to automate this "early checking" process. Additionally, while the model tracks floods over several days, incorporating multi-temporal analysis could further improve the consistency of recovery trajectories. Future research will also explore the generalizability of this approach to other arid cities.
Summary updated