A Geometry-Informed Neural Operator surrogate of LISFLOOD-FP for rapid pluvial-fluvial flood forecasting: the case of Forlì, Italy
Abstract
Physically based two-dimensional hydraulic models are too slow for real-time inundation forecasting and large ensembles. We present a time-dependent geometry-informed neural operator (GINO) surrogate of LISFLOOD-FP for the municipality of Forlì (Italy), severely affected by the May 2023 Emilia-Romagna floods. The surrogate is designed for an operational setting in which only river stage gauges and rain gauges are available. A reproducible pipeline turns public regional observations into a physically based training set: each river inflow is driven by the first stage gauge downstream of it through a calibrated water-surface boundary condition, and rain enters as SCS-CN effective rainfall. The surrogate is an autoregressive GINO trained on 216 simulations of 24 observed events with scaled forcing. It predicts volume-conserving polygon-mean depths on an exposure-based Voronoi mesh of 659 nodes, with a 20 min step over 72 h. The surrogate is tested on 38 independent observed events that it never saw during training. On these events it reaches a critical success index (CSI) of 0.934 at 0.05m and 0.844 at 0.3 m, a depth RMSE of 0.008m and an R2 of 0.984 for maximum-depth maps. On one NVIDIA A10G GPU a 72-h forecast takes 1.7 s instead of 55 s with LISFLOOD-FP, and a 50-member ensemble, computed in a single batched rollout, takes 6.2 s instead of about 46 min, which makes probabilistic flood forecasting practical. These results show that neural-operator surrogates can deliver accurate pluvial-fluvial flood forecasts driven by river stage for operational early warning.
Related
- Dynamical-Model-Informed Stochastic Flow Matching for Global Probabilistic Subseasonal Forecasting of Means and Extremes
- Spherical Physics-informed Neural Operator with Multi-scale Coupling for Meteorological Downscaling
- Conditional-Risk Optimization and KKT Geometry of a Regime- and Horizon-Adaptive Probabilistic Ensemble for Local Weather Forecasting
- GeONet: Geometry-Encoded Operator Network for Predicting Flow in Heterogeneous Porous Media
- Covariance-Informed Ensemble Score Filtering: Embedding Localized Covariance Geometry into Score-Based Transport
BibTeX
@misc{garofalo2026geometryinformed,
title = {{A Geometry-Informed Neural Operator surrogate of LISFLOOD-FP for rapid pluvial-fluvial flood forecasting: the case of Forlì, Italy}},
author = {Samuele Maria Garofalo and Fabio Limardo and Simonetta Bodojra and Mohammad Amin Nabian and Mehdi Taghizadeh},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.03381},
eprint = {2611.03381},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.03381},
}
RIS
TY - UNPB
TI - A Geometry-Informed Neural Operator surrogate of LISFLOOD-FP for rapid pluvial-fluvial flood forecasting: the case of Forlì, Italy
AU - Samuele Maria Garofalo
AU - Fabio Limardo
AU - Simonetta Bodojra
AU - Mohammad Amin Nabian
AU - Mehdi Taghizadeh
PY - 2026
DA - 2026/11/02
PB - arXiv
DO - 10.48550/arxiv.2611.03381
UR - https://arxiv.org/abs/2611.03381
AB - Physically based two-dimensional hydraulic models are too slow for real-time inundation forecasting and large ensembles. We present a time-dependent geometry-informed neural operator (GINO) surrogate of LISFLOOD-FP for the municipality of Forlì (Italy), severely affected by the May 2023 Emilia-Romagna floods. The surrogate is designed for an operational setting in which only river stage gauges and rain gauges are available. A reproducible pipeline turns public regional observations into a physically based training set: each river inflow is driven by the first stage gauge downstream of it through a calibrated water-surface boundary condition, and rain enters as SCS-CN effective rainfall. The surrogate is an autoregressive GINO trained on 216 simulations of 24 observed events with scaled forcing. It predicts volume-conserving polygon-mean depths on an exposure-based Voronoi mesh of 659 nodes, with a 20 min step over 72 h. The surrogate is tested on 38 independent observed events that it never saw during training. On these events it reaches a critical success index (CSI) of 0.934 at 0.05m and 0.844 at 0.3 m, a depth RMSE of 0.008m and an R^2 of 0.984 for maximum-depth maps. On one NVIDIA A10G GPU a 72-h forecast takes 1.7 s instead of 55 s with LISFLOOD-FP, and a 50-member ensemble, computed in a single batched rollout, takes 6.2 s instead of about 46 min, which makes probabilistic flood forecasting practical. These results show that neural-operator surrogates can deliver accurate pluvial-fluvial flood forecasts driven by river stage for operational early warning.
ER -