Numerical modelling of extreme wave events in the Southwestern South Atlantic

Severe ocean surface waves generated by wind (hereafter waves) have a strong impact on socio-economic activities such as navigation, harbours, oil exploitation, and coastal infrastructure. The South Atlantic monitoring remains behind regarding high-resolution wave products that can support the understanding and impacts of extreme wave events over the region. In this work, we present a high-resolution wave hindcast for the Southwestern South Atlantic (SWSA) evaluated under extreme conditions. Such a product can be used by several sectors to contribute to a more predictive and open data ocean, engaging the goals proposed by the UN Ocean Decade. The hindcast is produced using the WAM model forced by 1-hourly ERA5 surface winds. Three horizontal grids are used for downscaling, to keep a smooth resolution increase: a Global grid (0.25°), an intermediate grid that covers the Eastern coast of South America (0.1°), and a finer grid, focusing on the SWSA (0.05°). The spectral domain is discretized into 30 logarithmically spaced frequency bins and the wave propagating directions are set with a resolution of 24°. Sensibility runs are performed to obtain the more suitable configuration to represent the extreme wave climate in the region. The physics parameterization for the input and open ocean dissipation are tested between Jansen and Ardhuin formulations. The analyses showed that Ardhuin's parameterization (ST4) with Betamax of 1.60 performed better in comparison with buoys and satellite measurements during storm conditions. Moreover, the sea ice inclusion improved the wave height and wave direction in the coastal region, particularly on the southern Brazilian coast. Including depth refraction in both intermediate and finer grids also played an important role in the wave direction, improving the wave model performance against in situ data. We also present the wave hindcast evaluation against buoy and satellite data from 2017 to 2021, focusing on extreme wave events. Furthermore, significant wave height and wind speed are assimilated and the benefits of data assimilation in predicting extreme waves in the region are evaluated.