Using Neural Networks to Predict Hurricane Storm Surge and to Assess the Sensitivity of Surge to Storm Characteristics. Issue 24 (23rd December 2022)

• Record Type: Journal Article
• Authors: Lockwood, Joseph W.; Lin, Ning; Oppenheimer, Michael; Lai, Ching‐Yao
• Is Part Of: Journal of geophysical research. Volume 127:Issue 24(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-23
• Subjects: Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1029/2022JD037617
• Languages: English
• ISSNs: 2169-897X

Abstract

Abstract: Hurricane storm surge represents a significant threat to coastal communities around the world. Here, we use artificial neural network (ANN) models to predict storm surge levels using hurricane characteristics along the US Gulf and East Coasts. The ANN models are trained with storm surge levels from a hydrodynamic model and physical characteristics of synthetic hurricanes which are downscaled from National Centers for Environmental Prediction (NCEP) reanalysis using a statistical‐deterministic hurricane model. The ANN models are able to accurately predict storm surge levels with root‐mean‐square errors (RMSE) below 0.2 m and correlation coefficients > 0.85. The ANN models trained with the NCEP data also show satisfactory accuracy (RMSE below 0.7 m; correlation > 0.7) in predicting storm surge levels for hurricanes downscaled from future climate projections. Once trained, we use the ANN models to assess the sensitivity of storm surge levels to variations in hurricane characteristics and local geophysical features. Progressively stronger maximum wind speeds and larger outer radius sizes independently increase storm surge levels at all locations along the US East and Gulf Coasts. The response of storm surge levels to changes in hurricane translation speed, however, is found to be sensitive to coastal configuration, with increases in hurricane translation speed amplifying (reducing) storm surge levels in open ocean (semi‐enclosed) regions. Plain Language Summary: Hurricane storm surge represents a significant threat to coastal communities around the world. Here, we use a type of machine learning model known as artificial neural network (ANN) models to predict storm surge levels using hurricane characteristics along the US Gulf and East Coasts. The machine learning models are trained with storm surge levels from a hydrodynamic model and physical characteristics of synthetic hurricanes which are downscaled from historical data. The machine learning models are able to accurately predict storm surge levels. The ANN models trained with the NCEP data also show satisfactory accuracy in predicting storm surge levels for hurricanes downscaled from future climate projections. Once trained, we use the ANN models to explore the sensitivity of storm surge levels to variations in hurricane characteristics and local geophysical features. Progressively stronger maximum wind speeds and larger outer radius sizes independently increase storm surge levels at all locations along the US East and Gulf Coasts. The response of storm surge levels to changes in hurricane translation speed, however, is found to be sensitive to coastal configuration, with increases in hurricane translation speed amplifying (reducing) storm surge levels in open ocean (semi‐enclosed) regions. Key Points: Artificial Neural network (ANN) models trained with synthetic datasets are able to predict storm surge levels along the US East and Gulf Coasts ANN models accurately predict storm surge levels and unlike earlier studies, do not underestimate extreme levels Increases in hurricane translation speeds amplify (reduce) storm surge levels in open ocean (semi‐enclosed) regions