Positive Storm Surges in the Río de la Plata Estuary: forcings, long-term variability, trends and linkage with Southwestern Atlantic Continental Shelf dynamics

Abstract

The Río de la Plata Estuary (RdP), one of the most populated and developed areas ofSouthern South America, often experiences positive storm surges (PSS). These episodicrises of sea level due to meteorological forcing drive floods that endanger human livesand cause property damage. In this work, PSS are studied and contextualized in both thedynamics of the adjacent Southwestern Atlantic Continental Shelf (SWACS) and the syn-optic dynamics of the atmosphere. The study is based on statistical analyses of tide gaugeobservations gathered at the upper RdP (period 1934–2020), numerical simulations of sealevel in the SWACS and atmospheric reanalysis products. Results reveal hitherto unknownaspects of ocean dynamics in the SWACS forced by the atmosphere in which PSS eventsimpacting the RdP are embedded, showing that: (1) Strong PSS in the RdP can be locallyforced by cyclogenesis; nevertheless, most of PSS events affecting the estuary are remotelyforced at the southern SWACS and reach the estuary as free propagating coastal waves,producing a surge that can be as strong as the locally forced ones; (2) the spatial patternof the sea level anomalies in the SWACS during the PSS in the RdP can be described interms of three distinctive modes (or spatial structures), each one of them related to char-acteristic regional atmospheric synoptic processes (in particular traveling Rossby wavesand cyclogenesis); this provides clues to both better understand the surges and to extendtheir predictability; and (3) the PSS extension in the SWACS and the RdP depends on thespeed at which atmospheric systems pass through the region. These findings provide valu-able insights that can aid in anticipating extreme situations several days in advance to thenumerical ocean forecast systems. The analysis of the gauge observations at Buenos Airesshows that (1) the number of cases of PSS per year presents large multidecadal and interan-nual (pseudo-cycle at 5 years) variability; (2) in this frame, it is not reasonable to estimatelong-term lineal trends and/or extrapolate tendencies; therefore, our results refute the sug-gestion that the number of PSS cases in the RdP has been increasing over time, as reportedin previous works.