The assessment of wave-induced overtopping in coastal areas is fundamental for the implementation of local safety measures, such as coastal defence structures. Coastal flooding forecast systems have gained importance in coastal areas in recent years to ensure the safety of people and goods and to reduce damage caused by wave overtopping during storm events. In Portugal, no fully operational early warning system exists. This dissertation is a contribution to the project To-SEAlert, which aims at increasing the efficiency and reliability of the wave overtopping and flooding forecast system HIDRALERTA, developed by the Portuguese National Laboratory of Civil Engineering. The main goal of this study was to test the capability of the numerical model SWASH to be implemented in the HIDRALERTA system for the Ericeira harbour prototype. In order to achieve this goal, the model was first applied to simulate a testcase and to conduct a sensitivity analysis. For Ericeira harbour, the model was calibrated for storms with different wave conditions and for two breakwater profiles with the approach currently implemented in HIDRALERTA, the neural network NN_OVERTOPPING2. The main work consisted in the calibration of the Manning coefficient for the two breakwater profiles with armour layers of Antifer cubes and Tetrpods. Five expressions for the determination of the Manning coefficient were developed: one equation for the profile with Antifer cubes under normal wave attack and four for the profile of Tetrapods under (i) normal wave attack, (ii) oblique wave attack with incident angles between 15 and 50°, (iii) oblique wave attack with incident angles between 15 and 30° and (iv) oblique wave attack with incident angles between 30 and 50°. The results showed that the SWASH model is sensitive to changes in grid size, the number of simulated waves and in bottom friction. SWASH is capable of matching the discharges estimated by NN_OVERTOPPING2 with a calibrated Manning coefficient. The developed expressions showed small errors between the calculated and calibrated Mannings and revealed that the incident wave angle has an influence on the Manning coefficient and must be included in the simulations. Despite an underestimation of the overtopping discharge in some cases (with respect to the ones estimated by NN_OVERTOPPING2) the SWASH model was found to deliver overall good results when applied with Manning coefficients calculated by the developed expressions and capable of being implemented in HIDRALERTA.