Author(s): Michelle Jackson; Monica Fossati; Sebastian Solari

Linked Author(s): ssolari

Keywords: Sea level; Climate change; Dynamical downscaling

Abstract: The Uruguayan coast is over 600 km long and comprises the Rio de la Plata estuary and the Atlantic Ocean. Along this coast meteorological residuals (positive and negative surges) are of the same order of magnitude as the tides, and both are an important component of the total sea level all along the coast. Changes in regional atmosphere circulation pattern (wind and sea level pressures) and in the mean sea level can affect sea level climate at the coast. However, as the estuary and the continental shelf are wide and shallow, non-linear interactions among the different sea level components cannot be neglected, and the effect of these changes on sea level climate must be analyzed together. This work quantifies changes in sea level climate (central and extreme values) along the Uruguayan coast produced by climate changes. To this end, a dynamical downscaling approach is used, based on two nested bidimensional hydrodynamic models that were previously implemented for obtaining a high-resolution (under 1 km) sea levels hindcast along the Uruguayan coast. The models are forced by fluvial flows from Parana and Uruguay rivers, astronomical tide at oceanic boundaries, and winds and pressure fields on the surface. In order to obtain the projected sea level climate, wind and pressure fields from seven global climate models (GCM) from the CMIP5 (Coupled model intercomparison project phase 5) were used, namely: ACCESS, CMCC-CM, CNRM-CM5, GFDL-ESM2G, MIROC5, IPSL-CM5A-MR and HadGEM2-ES. Moreover, mean sea level rise effect was also included in the downscaling, by imposing regional mean sea level changes corresponding to the different analyzed scenarios. Projections were obtained for two future scenarios: RCP4.5 and RCP8.5; and two temporal scales: short term (2027-2045) and long term (2082-2100). As a reference, the historical period 1986-2005 was considered. The historical simulations forced with CMIP5 were compared with the results of the high-resolution hindcast that was obtained with winds and sea level pressure from CFSR. The evaluation of the historical period shows that all GCM used are reasonably able to reproduce sea level climate along the coast, although with different skill. From this work, it is concluded that the main component of the sea level change is the regional mean sea level rise, followed by the effect that this rise has on the astronomical tide. The foregoing indicates that the dynamic downscaling methodology applied is correct and necessary to determine sea level change. The change in the long-term total sea level mean and for the most severe scenario RCP 8.5 reaches 61 cm in some coastal areas, while the increase in the long-term 99% quantile for the most severe scenario RCP 8.5 reaches 74 cm in the interior of the estuary and 71 cm in the coastal área of Montevideo.

DOI: https://doi.org/10.3850/IAHR-39WC252171192022384

Year: 2022