Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

A hybrid CNN-LSTM surrogate model for hyper-resolution spatiotemporal flood forecasting in Norfolk, Virginia

Journal Article · · Journal of Hydrology. Regional Studies
 [1];  [1];  [2];  [3];  [4];  [1];  [5];  [5]
  1. Univ. of Virginia, Charlottesville, VA (United States)
  2. Old Dominion Univ., Portsmouth, VA (United States); Thomas Jefferson National Accelerator Facility (TJNAF), Portsmouth, VA (United States)
  3. Thomas Jefferson National Accelerator Facility (TJNAF), Newport News, VA (United States)
  4. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
  5. Florida International Univ. (FIU), Miami, FL (United States)
+ Show Author Affiliations
Study region: Norfolk, Virginia, United States Study focus: Accurate and timely flood forecasting is essential for enhancing resilience in coastal urban areas in the context of increasing frequency and intensity of rainfall, sea level rise and urbanization. This study presents a hybrid deep learning-based surrogate model that integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to enable real-time spatiotemporal flood forecasting. The model leverages CNN to capture spatial features from inputs such as elevation and Topographic Wetness Index (TWI), while LSTM processes time-series inputs of rainfall and tide data to capture temporal features. New hydrologic insights for the region: The hybrid CNN-LSTM model was trained using the physics-based hydrodynamic model simulations obtained from the Two-dimensional Unsteady FLOW (TUFLOW) model for Norfolk, Virginia, and achieved high predictive accuracy across diverse flood-prone areas. The reduced computational time from four to six hours using TUFLOW to 3.2 min per event using CNN-LSTM enables rapid flood inundation mapping and early warning applications. The model effectively captured both spatial flood extents and their temporal evolution across different flooding scenarios, providing forecasts at a 2.5-m spatial resolution and 15-min temporal resolution and a one-hour-ahead prediction horizon. While challenges remain in terms of transferability to new regions and real-time data assimilation, this approach demonstrates strong potential for supporting operational flood risk management in coastal urban environments.
Research Organization:
Thomas Jefferson National Accelerator Facility (TJNAF), Newport News, VA (United States)
Sponsoring Organization:
National Science Foundation (NSF); USDOE Office of Science (SC)
Grant/Contract Number:
AC05-06OR23177
Other Award/Contract Number:
1951745
2209139
OSTI ID:
3019881
Report Number(s):
DOE/OR/23177--8090; JLAB-CST--26-4573
Journal Information:
Journal of Hydrology. Regional Studies, Journal Name: Journal of Hydrology. Regional Studies Vol. 64; ISSN 2214-5818
Publisher:
ElsevierCopyright Statement
Country of Publication:
United States
Language:
English

References (51)

Accelerated flooding along the U.S. East Coast: On the impact of sea-level rise, tides, storms, the Gulf Stream, and the North Atlantic Oscillations: EZER AND ATKINSON journal August 2014
Does the mid-Atlantic United States sea level acceleration hot spot reflect ocean dynamic variability?: SEA LEVEL ACCELERATION HOT SPOT journal August 2013
On strictly enforced mass conservation constraints for modelling the Rainfall‐Runoff process journal March 2023
Assessing the inundation risk of cultural heritages along the southwestern coast of Taiwan: present and future journal April 2024
A survey of the recent architectures of deep convolutional neural networks journal April 2020
Topography Wetness Index Application in Flood-Risk-Based Land Use Planning journal December 2014
Improving flood forecasting using time-distributed CNN-LSTM model: a time-distributed spatiotemporal method journal June 2024
Vulnerability to Coastal Flooding of Industrial Urban Areas in Greece journal June 2020
Assessment of inundation risk in urban floods using HEC RAS 2D journal September 2019
A deep-learning model for rapid spatiotemporal prediction of coastal water levels journal June 2024
Improving the explainability of CNN-LSTM-based flood prediction with integrating SHAP technique journal December 2024
Urban flood numerical simulation: Research, methods and future perspectives journal October 2022
Urban flood extent segmentation and evaluation from real-world surveillance camera images using deep convolutional neural network journal February 2024
Research progress and prospects of urban flooding simulation: From traditional numerical models to deep learning approaches journal January 2025
Review on Convolutional Neural Networks (CNN) in vegetation remote sensing journal March 2021
Dynamic risk of coastal flood and driving factors: Integrating local sea level rise and spatially explicit urban growth journal October 2021
Flood risk assessment and increased resilience for coastal urban watersheds under the combined impact of storm tide and heavy rainfall journal December 2019
Spatiotemporal deep learning rainfall-runoff forecasting combined with remote sensing precipitation products in large scale basins journal January 2023
Comparing a long short-term memory (LSTM) neural network with a physically-based hydrological model for streamflow forecasting over a Canadian catchment journal December 2023
Improving urban flood prediction using LSTM-DeepLabv3+ and Bayesian optimization with spatiotemporal feature fusion journal February 2024
Forecasting Multi-Step-Ahead Street-Scale Nuisance Flooding using a seq2seq LSTM Surrogate Model for Real-Time Application in a Coastal-Urban City journal August 2025
A critical review of RNN and LSTM variants in hydrological time series predictions journal December 2024
Interpretable spatio-temporal attention LSTM model for flood forecasting journal August 2020
Post-disaster flooded region segmentation using DeepLabv3+ and unmanned aerial system imagery journal June 2025
Physics-informed neural networks as surrogate models of hydrodynamic simulators journal February 2024
Exploring infiltration effects on coastal urban flooding: Insights from nuisance to extreme events using 2D/1D hydrodynamic modeling and crowdsourced flood reports journal March 2025
A Transdisciplinary Review of Deep Learning Research and Its Relevance for Water Resources Scientists journal November 2018
Training Machine Learning Surrogate Models From a High‐Fidelity Physics‐Based Model: Application for Real‐Time Street‐Scale Flood Prediction in an Urban Coastal Community journal October 2020
Physics‐Informed Neural Networks of the Saint‐Venant Equations for Downscaling a Large‐Scale River Model journal February 2023
Compound Flooding Hazards Due To Storm Surge and Pluvial Flow in a Low‐Gradient Coastal Region journal November 2024
Future flood losses in major coastal cities journal August 2013
Impact of Sea-Level Rise on Roadway Flooding in the Hampton Roads Region, Virginia journal December 2017
FV-DLV3+: a light-weight flooded vegetation extraction method with attention-based DeepLabv3+ journal October 2024
Designing Deep-Based Learning Flood Forecast Model With ConvLSTM Hybrid Algorithm journal March 2021
A rapid flood inundation model for hazard mapping based on least squares support vector machine regression journal January 2019
Data‐driven flood emulation: Speeding up urban flood predictions by deep convolutional neural networks journal December 2020
Interpretable physics-informed graph neural networks for flood forecasting journal July 2025
Long Short-Term Memory journal November 1997
Modeling of Coastal Inundation, Storm Surge, and Relative Sea-Level Rise at Naval Station Norfolk, Norfolk, Virginia, U.S.A. journal January 2013
Evidence of Sea Level Acceleration at U.S. and Canadian Tide Stations, Atlantic Coast, North America journal November 2012
Urban flood forecasting using a hybrid modeling approach based on a deep learning technique journal March 2023
Linking of 2D and Pipe hydraulic models at fine spatial scales journal June 2007
Dynamic Modeling of Inland Flooding and Storm Surge on Coastal Cities under Climate Change Scenarios: Transportation Infrastructure Impacts in Norfolk, Virginia USA as a Case Study journal May 2022
Flood Detection in Dual-Polarization SAR Images Based on Multi-Scale Deeplab Model journal October 2022
Flood Prediction Using Machine Learning Models: Literature Review journal October 2018
Application of Long Short-Term Memory (LSTM) Neural Network for Flood Forecasting journal July 2019
Developing a Discharge Estimation Model for Ungauged Watershed Using CNN and Hydrological Image journal December 2020
Fast Prediction of Urban Flooding Water Depth Based on CNN−LSTM journal April 2023
Rainfall–runoff modelling using Long Short-Term Memory (LSTM) networks journal January 2018
Deep learning rainfall–runoff predictions of extreme events journal July 2022
Bare-earth DEM generation from ArcticDEM and its use in flood simulation journal February 2023

Similar Records

Forecasting Multi-Step-Ahead Street-Scale Nuisance Flooding using a seq2seq LSTM Surrogate Model for Real-Time Application in a Coastal-Urban City
Journal Article · Sun Jan 12 19:00:00 EST 2025 · Journal of Hydrology · OSTI ID:2560298
A comparison of machine learning surrogate models of street-scale flooding in Norfolk, Virginia
Journal Article · Tue Nov 28 19:00:00 EST 2023 · Machine Learning with Applications · OSTI ID:2339786
Online thermal profile prediction for large format additive manufacturing: A hybrid CNN-LSTM based approach
Journal Article · Thu Jul 17 20:00:00 EDT 2025 · Additive Manufacturing · OSTI ID:3002014

Related Subjects

Coastal urban flooding
Deep learning
Deeplabv3+
Hybrid model
Spatiotemporal forecasting