Graph transformer-based flood susceptibility mapping: Application to the French Riviera and railway infrastructure under climate change
6 affiliations
- 1Université Paris-Saclay, CentraleSupélec, ENS Paris-Saclay, CNRS, LMPS - Laboratoire de Mécanique Paris-Saclay, Gif-sur-Yvette 91190, France
- 2CNRS - Centre National de la Recherche Scientifique (France)
- 3CentraleSupélec (3, rue Joliot Curie, Plateau de Moulon, 91192 GIF-SUR-YVETTE Cedex - France)
- 4ENS Paris Saclay - Ecole Normale Supérieure Paris-Saclay (4 avenue des Sciences, 91190 Gif-sur-Yvette - France)
- 5LMPS - Laboratoire de Mécanique Paris-Saclay (4 avenue des sciences / 8-10 rue Joliot Curie, 91190 Gif-sur-Yvette - France)
- 6Université Paris-Saclay (Bâtiment Bréguet, 3 Rue Joliot Curie 2e ét, 91190 Gif-sur-Yvette - France)
Abstract
Study region The French Riviera watershed in southeastern France (≈2650 km²) extends from steep alpine headwaters to a densely urbanised Mediterranean coast traversed by the SNCF rail network, and has suffered repeated destructive floods, including the 1994 and 2010 Var floods, the 2015 Côte d’Azur flood, and the 2020 Storm Alex. Study focus Traditional machine-learning (ML) approaches achieve high point-wise metrics (AUC-ROC, accuracy) but fail to produce the spatially coherent predictions needed for reliable infrastructure-exposure estimation. We propose a graph transformer (GT) architecture incorporating a feature-similarity graph regularised by Laplacian positional encodings and spatial attention, and benchmark it against six traditional ML models, and show that these point-wise metrics alone are insufficient and should be complemented by spatial autocorrelation metrics (Moran's I, Geary's C) when evaluating multi-class flood predictions. Using topographic, hydrologic, and environmental features, we further project flood susceptibility and railway-track exposure for 2050 under RCP 4.5 and RCP 8.5 scenarios. New hydrological insights for the region The GT achieved AUC-ROC ≈ 0.974 (slightly below XGBoost ≈ 0.990) but substantially improved spatial coherence (Moran’s I ≈ 0.612, Geary’s C ≈ 0.473 versus XGBoost 0.409 and 0.589, and RF 0.474 and 0.523). Its smooth boundary delineation produced markedly different distributions: the GT classified 6.2% of the area as very-high with 35.6% railway exposure, versus 55–57% very-high area and 46–52% railway exposure for ML models; the GT very-high extent closely matches the 172.6 km² historical inundation inventory. Accumulated Local Effects analysis confirmed elevation, stream power, and precipitation as dominant monotonic drivers consistent with flood hydrology. Scenario-conditional projections for 2050 under RCP 4.5 (Q95) and RCP 8.5 (Q95) indicate very-high areas could expand to ≈ 28% and ≈ 36.3%, with railway exposure reaching ≈ 64% and ≈ 67.4%. The GT framework provides a replicable blueprint for regions seeking spatially accurate flood susceptibility mapping.
Related
- Landslide Susceptibility Modelling and Infrastructure Exposure in the Zillertal, Austrian Alps
- Reframing Mangroves as Socio-Ecological Coastal Infrastructure: Development and Application of a Six-Pillar Socio-Ecological Infrastructure Framework
- A geospatial machine-learning framework for regional electrification estimation mapping and trend analysis using demographic and infrastructure features
- Landslide Susceptibility Modeling and Mapping: A Comparison of Frequency Ratio and Decision Tree Models
- From an everyday device to professional instrumentation. A framework for Portable Geoscience Learning Infrastructure and its illustrative application to ITA ARANDU MS
BibTeX
@article{vemula2026graph,
title = {{Graph transformer-based flood susceptibility mapping: Application to the French Riviera and railway infrastructure under climate change}},
author = {Sreenath Vemula and Filippo Gatti and Pierre Jehel},
journal = {Journal of Hydrology Regional Studies},
year = {2026},
month = jun,
doi = {10.1016/j.ejrh.2026.103644},
url = {https://doi.org/10.1016/j.ejrh.2026.103644},
note = {CC BY},
}
RIS
TY - JOUR
TI - Graph transformer-based flood susceptibility mapping: Application to the French Riviera and railway infrastructure under climate change
AU - Sreenath Vemula
AU - Filippo Gatti
AU - Pierre Jehel
PY - 2026
DA - 2026/06/17
JO - Journal of Hydrology Regional Studies
DO - 10.1016/j.ejrh.2026.103644
UR - https://doi.org/10.1016/j.ejrh.2026.103644
AB - Study region The French Riviera watershed in southeastern France (≈2650 km²) extends from steep alpine headwaters to a densely urbanised Mediterranean coast traversed by the SNCF rail network, and has suffered repeated destructive floods, including the 1994 and 2010 Var floods, the 2015 Côte d’Azur flood, and the 2020 Storm Alex. Study focus Traditional machine-learning (ML) approaches achieve high point-wise metrics (AUC-ROC, accuracy) but fail to produce the spatially coherent predictions needed for reliable infrastructure-exposure estimation. We propose a graph transformer (GT) architecture incorporating a feature-similarity graph regularised by Laplacian positional encodings and spatial attention, and benchmark it against six traditional ML models, and show that these point-wise metrics alone are insufficient and should be complemented by spatial autocorrelation metrics (Moran's I, Geary's C) when evaluating multi-class flood predictions. Using topographic, hydrologic, and environmental features, we further project flood susceptibility and railway-track exposure for 2050 under RCP 4.5 and RCP 8.5 scenarios. New hydrological insights for the region The GT achieved AUC-ROC ≈ 0.974 (slightly below XGBoost ≈ 0.990) but substantially improved spatial coherence (Moran’s I ≈ 0.612, Geary’s C ≈ 0.473 versus XGBoost 0.409 and 0.589, and RF 0.474 and 0.523). Its smooth boundary delineation produced markedly different distributions: the GT classified 6.2% of the area as very-high with 35.6% railway exposure, versus 55–57% very-high area and 46–52% railway exposure for ML models; the GT very-high extent closely matches the 172.6 km² historical inundation inventory. Accumulated Local Effects analysis confirmed elevation, stream power, and precipitation as dominant monotonic drivers consistent with flood hydrology. Scenario-conditional projections for 2050 under RCP 4.5 (Q95) and RCP 8.5 (Q95) indicate very-high areas could expand to ≈ 28% and ≈ 36.3%, with railway exposure reaching ≈ 64% and ≈ 67.4%. The GT framework provides a replicable blueprint for regions seeking spatially accurate flood susceptibility mapping.
SN - 2214-5818
ER -