Impacts of a large extra-tropical cyclonic system in Southern Brazilian Continental Shelf using the COAWST model

Abstract

Abstract. The Southern Brazilian Continental Shelf (SBCS) is an area with great ecological and economic importance to Brazil. In this region can be observed the recurrent passage of frontal systems and extra-tropical cyclones, which are more frequent during the winter months of the southern hemisphere. These systems act on the ocean surface layers as direct driving forces, which may change the thermohaline structure of the water column and induce sea level perturbations. This study used the coupled ocean-atmosphere regional model (COAWST) to evaluate the effect of the passage of a frontal system associated with an extra-tropical cyclone. The ROMS oceanic model was configured with two nested grids, in order to solve the hydrodynamic processes at different scales. The parent (20–40° S/40–60° W) and child (25–29.3° S/46.3–50° W) grid comprise the coastal region, with a horizontal resolution of 1/9° and 1/27°, respectively, with 32 vertical levels. The initial conditions are the Global Analysis Forecast from CMEMS and forcing files used the Climate Forecast System v.2 (CFSv2) data, from NCEP. This event took place on the continental shelf of the State of Santa Catarina, in September of 2016. The model results were compared to remote sensing data and to the tide gauges from the City of Imbituba (State of Santa Catarina, Brazil). The comparison showed a correlation higher than 78 % between sea level rise data and the model results. The filtering of sea level data made it possible to identify the meteorological component in the model results. The comparison between the tidal-gauge and the model output presented values under 25 cm. The model was capable of representing the sea level anomalies propagation associated with the passage of the atmospheric frontal system. The model output showed the presence of a sea level anomaly propagating northward along the continental shelf at 480 km day−1, probably associated with the presence of a coastal-trapped wave.

Reply on RC2

“…The comparison between satellite and modeled SSH data showed mean differences between -0.2 and 0.2m on the SBCS and offshore, which were close to those found by Hermes & Reason (2005). The typical high frequency periodicity in tidal data may combine with the lower sampling frequency of the altimeter and cause distortions or associated errors, which can lead to the presence of low-frequency artificial signals in the sampled time series (Strub et al, 2015). This concern with tidal errors is greater where tides are large (Palma et al, 2004(Palma et al, , 2008, although it is possible to occur significant errors in conditions similar to the area of this study.…”

“…According to this characterization, it is possible to observe that the numerical model was able to represent the main water masses in the study region: PRP, SASW, Subtropical Shelf Water (STSW), Tropical Water (TW), and South Atlantic Central Water (SACW). It should be pointed out that, even during the spring, the presence of the SASW in latitudes higher than 35°S indicates a slow retraction of the Brazilian Coastal Current (BCC, not shown) as described by Souza & Robinson (2004), Piola et al (2005Piola et al ( , 2008, Möller et al (2008), and Matano et al (2010). The T/S diagram data points that corresponded to the SASW and STSW were concentrated at the edges of the characterization boxes, showing that the mixing occurs gradually over the continental shelf as demonstrated by Möller et al (2008) and Mendonça et al (2017).…”

“…The model setup and the validation methodology are similar to those used by Mendonça et al (2017), since the authors used the same parameterization of the ocean and atmosphere models, essential to guarantee the quality of the simulation. The current understanding of coastal processes is largely based on numerical models, which mathematically reproduce the dynamic conditions of a region of interest.…”

“…Then mass conservation generates a compensating offshore flow in the lower layer. As this flow crosses the isobaths, it leads to changes in the local relative vorticity, which is expressed in terms of velocity along the coast and, in the presence of variability along the coast, it generates the propagation of the waves (Brink, 1991). In the southern hemisphere, topographic Rossby waves propagate with the coast at its left, thus progressing to the north along the western continental margin of South America.…”

“…The numerical model generated a coefficient of determination in the range of 78%, slope of the regression line close to the ideal value (1.00), with a standard deviation of less than 10 cm. According to Brink (1991), the existing models tend to underestimate the amplitude of observed fluctuations in the current along the coast and at sea level, usually by around 10-50%. The wave propagation velocity analyzes showed higher velocity values (~ 10.6 m.s -1 ) over the continental shelf of Uruguay and Rio Grande do Sul.…”