Abstract

ABSTRACTSea-level forecasting in Venice is a fundamental task for safeguarding the ancient city, its cultural heritage and natural environment. In the last 10 years the finite-element numerical Shallow Water Hydrodynamic Model (SHYFEM), developed at the Institute of Marine Sciences (ISMAR) of the National Research Council (CNR) of Venice, has been operationally implemented at the Istituzione Centro Previsioni e Segnalazioni Maree (ICPSM) forecast centre of the Venice Municipality. The model calculates sea level and currents in the whole Mediterranean Sea and the Venice Lagoon under the action of atmospheric forcing fields (pressure and winds) operationally supplied by the European Centre for Medium-Range Weather Forecasts (ECMWF). A neural-network module was implemented to correct model results with observed sea-level data in the Adriatic Sea. Every 6 h, the operational model produces a sea-level forecast for the next six days in the Northern Adriatic Sea in the Venice Lagoon. A statistical analysis was performed on the operational results for 2008–2009, to investigate the model goodness-of-fit and its reliability in the operational context. The main statistical quantities (mean error Ē, standard deviation σ, maximum overestimation Emax and maximum underestimation Emin) were computed. The model predicts the sea level both in Venice and in the open sea with good accuracy, with Ē and σ of the order of 1 cm and 5 cm, respectively, in the first 24 h of anticipation during normal weather conditions. Model errors increase when only considering periods of storm surge events.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call