C-STRIDE: An Observation-Driven AI Digital Twin for Predicting Basin-Wide Flood Fields from Sparse Stream-Gauge Histories

2026-09-30

AI Summary

This study addresses the high computational cost and slow update rates of basin-wide flood forecasting under sparse hydrological station conditions by proposing an AI digital twin framework that integrates topographic and rainfall information. The approach trains a deep learning network on simulation data from a calibrated two-dimensional hydrodynamic model, generating full-domain water depth maps from limited historical station observations and extrapolating them up to 24 hours ahead. Furthermore, it eliminates the need for independent data assimilation steps by directly incorporating multi-source observations to enable rapid, continuous updates. Compared with baseline methods relying solely on station data, the proposed framework reduces prediction error by 40% and improves computational efficiency by 150-fold, thereby achieving efficient and accurate real-time flood forecasting.

Abstract

Emergency managers need to know where floodwater is, how deep it is, and how it will change over the coming hours across an entire river basin. During a flood, however, real-time measurements come from only a handful of stream gauges, and high-resolution hydrodynamic models are too costly to rerun each time new data arrive or to run as large ensembles. We present C-STRIDE, an observation-driven AI digital twin that turns short records from a few stream gauges, together with terrain and rainfall, into basin-wide maps of water depth and extends these predictions up to a day ahead. It is trained on simulations from a calibrated two-dimensional hydrodynamic model and needs no separate data-assimilation step. In the Des Plaines River basin near Chicago, six gauges inform predictions over 4.2 million 30-m grid cells. Terrain improves the predictions most, rainfall keeps errors from growing over longer horizons, and together they reduce errors by about 40% compared with gauge records alone. When future rainfall is known, errors remain near 15% one day ahead, compared with nearly 40% without rainfall. Given real instead of simulated gauge records, the model shifts its predictions toward the observed hydrographs at three of six gauges without retraining, and it runs about 150 times faster than the hydrodynamic model. These results show how sparse gauges, terrain, and rainfall can be combined into fast, continuously updated flood predictions, a step toward operational flood digital twins that still requires testing with real-time data and rainfall forecasts.

Problem

• flood prediction
• digital twin
• sparse stream gauges
• hydrodynamic modeling
• basin-wide flooding

Innovation

• AI digital twin
• flood prediction
• sparse stream gauges
• hydrodynamic model
• observation-driven