Nonlinear Autoregressive Exogenous (NARX) Neural Network Models for Storm Tide Forecasting in the Venice Lagoon

A breakthrough in storm tide prediction for Venice and its lagoon has been achieved using Nonlinear Autoregressive Exogenous (NARX) neural network models. This advanced approach significantly enhances the accuracy of forecasting extreme high tides, which pose a persistent threat to this UNESCO World Heritage Site due to astronomical, meteorological, and basin-specific factors. The model's reliability makes it a valuable tool for managing the MOSE flood defense system, offering proactive protection against the damaging effects of storm surges that have historically plagued Venice. By leveraging neural network techniques, researchers have developed a robust framework capable of capturing the complex dynamics influencing tide levels, thereby improving preparedness and mitigating risks to this culturally significant urban landscape.

Di Nunno, F., Granata, F., Gargano, R., & De Marinis, G. (2023). Nonlinear Autoregressive Exogenous (NARX) Neural Network Models for Storm Tide Forecasting in the Venice Lagoon. EWaS5, 93. https://doi.org/10.3390/environsciproc2022021093