Despite advances in understanding the impacts of ARs on precipitation and river discharge, their influence on coastal ocean physical and biogeochemical processes remains poorly understood (e.g. Shinoda et al.39, Diaz et al.40, Garcia-Santos et al.38). We hypothesize that AR orientation not only modulates precipitation and river discharge distribution but also impacts river plume and shelf dynamics, influencing freshwater, sediment, and nutrient dispersal. Depending on prevailing wind patterns and synoptic conditions, ARs may either enhance or inhibit the offshore extent of river plumes, affecting water column stratification, salinity gradients, and nutrient availability. These interactions are critical for understanding how extreme weather events affect coastal marine ecosystems and economically important sectors such as fisheries and aquaculture. To address this hypothesis, we combine multiple data sources, including atmospheric reanalysis, satellite imagery, in situ oceanographic data, and long-term precipitation and discharge records from the region's major rivers.

This study focuses on central-southern Chile (33-40 S), where ten representative rivers were selected to analyze the freshwater discharge associated with ARs (Fig. 1a). A strong river discharge event following a ZAR is shown in Fig. 1b. This region includes Coliumo Bay, where field campaigns were carried out during the winters of 2022 and 2023 to monitor the oceanic response to AR landfalls (Fig. 1b). ARs affecting this region exhibit distinct orientations (Fig. 1c); however, this study focuses specifically on TARs and ZARs. In this region, all ARs occur mostly during austral winter (June-July-August) (Fig. 1d), consistent with the climatology described by Viale et al.. This seasonal pattern is driven by the equatorward displacement of the South Pacific Subtropical High (SPSH), which alters the storm track and increases the frequency of AR landfalls in the region. When restricting to intense events (Integrated Vapor Trasport (IVT) 500 ), both AR types remain more prevalent during winter. Given this seasonal dominance, the present analysis focuses exclusively on austral winter AR events.

TARs typically develop under more meridional flow regimes, associated with mid-latitude troughs and cyclonic activity that favor northwest-southeast moisture transport. On the day 0 (landfall day) (Fig. 2a), TARs are associated with IVT values averaging around 300 , along with enhanced northwesterly winds reaching up to 8 . The mean sea level pressure (SLP) field (contours) shows a trough southwest of the AR axis, responsible for steering the ARs in a clockwise path, allowing them to extend far north of their landfall latitude. In contrast, ZARs exhibit a pressure dipole, with high pressure to the north and low pressure to the south of the AR axis. This configuration generates a strong zonal (westerly) flow that constrains the AR trajectory and limits its latitudinal spread (Fig. 2a). On day 0, mean surface pressure north of the AR axis exceeds 1022 hPa, favoring southerly coastal winds north of approximately 30 S. At the point of landfall, both IVT and westerly winds intensify to about 350 and 8 , respectively.

TARs, due to their broader latitudinal extent, tend to produce precipitation over wider areas, including both inland valleys and coastal regions. They can result in accumulated precipitation of up to 100 mm near the Andes at 36 S, although most of the affected regions experience values between 40 and 80 mm (Fig. 2b). In contrast, ZARs are more latitudinally confined but often produce more intense precipitation, particularly over the Andean foothills between 36 and 40 S, where accumulated precipitation typically ranges from 70 to 100 mm. In valleys and coastal zones, ZAR-related precipitation tends to be lower, generally below 40 mm. The difference between TARs and ZARs reveals that precipitation associated with ZARs is stronger (up to 35 mm), predominantly impacting the Andes south of approximately 37 S. In comparison, TAR-related precipitation, although reaching similar magnitudes (up to 30 mm), is more broadly distributed, affecting both coastal and inland areas north of 38 S.

These precipitation patterns are reflected in river discharge responses (Fig. 2c). Both AR types can induce peak discharges up to 2000 , as observed in the Biobio River. TARs primarily affect rivers north of 37.5 S, such as the Maipo, Rapel, Mataquito, Maule, and Itata, with the Maule River showing the greatest sensitivity, exhibiting a discharge difference of approximately 50 compared to ZARs. In contrast, ZARs exert a stronger influence on rivers south of 37.5 S, including the Biobio (particularly near its Andean headwaters), Imperial, Toltén, and Valdivia, with the Toltén River displaying the largest response, around 150 greater than during TARs. Notably, the coastal section of the Biobio and Maipo rivers exhibits discharge anomalies close to zero, suggesting that both AR types have comparable impacts in these regions.

Satellite-derived surface reflectance following AR landfall reveals a significant increase in river plume extent relative to winter climatological values (Fig. 3). After TARs, northerly to northwesterly post-landfall winds drive river plumes southward along the coast (Fig. 3a), promoting plume convergence and forming extensive bands of freshwater attached to the shoreline. In contrast, post-ZAR southerly winds favor offshore Ekman transport and northwestward spreading of river plumes (Fig. 3b). Consistent with precipitation and discharge patterns (Fig. 2c,d), reflectance anomalies north of 37.5 S are more pronounced following TARs than ZARs, indicating stronger plume expansion for rivers such as the Biobio and those farther north. South of 37.5 S, reflectance anomalies are typically stronger after ZARs, reflecting the increased discharge from southern rivers, leading to an extensive band of Rrs(645).

To evaluate the effects of ARs on the coastal hydrography and circulation, a field campaign was carried out during July 2022 off Coliumo Bay (Fig. 1b). A detailed description of atmospheric forcings during this period is provided by Garreaud et al.. Six ARs (two TARs and four ZARs) made landfall in the study region during the field campaign. Among these, the events on July 9th (TAR 1), July 17th (ZAR 3), and July 26th (TAR 2) reached IVT values exceeding 500 , classifying them as moderate ARs (Fig. 4a), following the AR scale of Ralph et al.. All events were characterized by high precipitation and intensification of the meridional component of surface winds (Fig. 4b). In all cases, meridional winds were stronger than zonal winds. In both TAR and ZAR cases, orographic blocking induces a low-level northerly coastal flow, which enhances the meridional component of IVT; in contrast, ZARs are characterized by dominant upper-level westerlies that extend across the Andes.

Time series of atmospheric variables show reduced solar radiation and limited air temperature variability during all AR events (Fig. 4c). This refers to the reduced diurnal temperature range during ARs, caused by thick cloud cover that limits daytime solar heating and traps longwave radiation at night. The two closest gauged rivers, Itata and Biobio, exhibited increases in discharge 1-2 days after peak precipitation (Fig. 4d). These discharge responses exceeded the July climatological means of 665 for Itata and 1784 for Biobio, with the exception of the ZAR event on July 23rd, which primarily affected the Biobio river.

The vertical structure of the water column revealed a thermal inversion, typical of winter conditions in this region, with cooler surface waters (around 11.7 C) overlaying warmer subsurface layers (around 12.5 C) (Fig. 4e). Most ARs induced mixing in the surface layer (0-10 m), while the mixing of the entire water column was observed during the TARs on July 9th and 26th. Anomalously low temperatures and strong thermal stratification occurred from July 14th to 17th. This phenomenon likely resulted from a persistent low-pressure system along the central coast of Chile from July 12th to 16th, which reduced the meridional wind component, solar radiation, and air temperature (Fig. 4b,c). Additionally, the strong river discharge after ZAR 2 could contribute to the enhanced stratification observed in that period. The ocean currents intensified during all AR events, with velocities reaching up to 0.2 m/s, primarily near the surface (5 m) and the bottom (20 m) (Fig. 4f). Nonetheless, current direction did not exhibit a consistent or well-defined pattern in response to the ARs (arrows in Fig. 4f), likely due to the strong wind forcing over the shallow water column and the ADCP's proximity to the coast.

Following the passage of TAR 1, a substantial increase in nitrate concentrations was observed on July 11th, particularly in surface waters (2 m), reaching approximately 17 (Fig. 5a). In contrast, silicic acid exhibited a marked increase in the first 10 meters of the water column, reaching values up to 55 (Fig. 5c), highlighting the strong influence of riverine waters. Toward the end of the observation period, high chlorophyll-a concentrations were observed near the surface, linked to the diurnal pulse of PAR and likely supported by an increase in terrestrial nutrients, particularly silicic acid. An increase in chlorophyll-a concentration was also detected from the extracted samples, particularly at 2 m, where values peaked at 1.2-1.3 on July 11th and July 28th (Fig. 5d). The first peak coincided with an increase in Photosynthetically Active Radiation (PAR) (Fig. 6f), further supporting the role of light and nutrient availability in promoting phytoplankton growth during the post-AR phase.

Following TAR 2 (July 26-27), nutrient concentrations revealed similar trends, with a notable increase in nitrates and silicic acid concentrations (Fig. 5a,c), along with an increase in chlorophyll-a concentrations at the surface and the bottom (Fig. 5d). Unlike TARs, nutrient concentrations were generally lower following ZAR events. A noticeable increase in chlorophyll-a concentration was only observed after the passage of ZAR 3, reaching values up to 0.9 , likely due to the three-day delay between the event and the sampling, which allowed time for a phytoplankton response under optimal light conditions.

TAR 1 was associated with a synoptic-scale system consistent with an extratropical cyclone and cold front reaching south-central Chile during winter. This system produced an accumulated precipitation of approximately 32 mm/day at the Dichato station and was accompanied by strong north-northwesterly winds, averaging around 6 m/s, that prevailed throughout the event. The study by Garreaud et al. provides a more detailed analysis of this event, showing a frontal cloud band that extends over southern Chile, oriented nearly parallel to the Andes Cordillera, with a north-northwest IVT impacting the study area.

The Wirewalker profiler allowed a high-frequency vertical characterization of physical (temperature, salinity) and biological-chemical (fluorescence, dissolved oxygen) variables before, during, and after the TAR 1 landfall (Fig. 6). The temperature profiles confirmed the thermal inversion observed by the temperature sensor array, with cooler water at the surface (around 11.5 C) and warmer subsurface water (around 12.5 C) (Fig. 6a). The salinity profiles revealed fresher surface waters (32 PSU) and progressively saltier, denser waters near the bottom (34 PSU) (Fig. 6b). Stratification was evident in the Brunt-Väisälä Frequency (BVF) profiles, with surface values near 40 cph during July 8 (Fig. 6c). Relatively high concentrations of dissolved oxygen (9.5-10 mL/L) and chlorophyll-a (2-2.5 µg/L) were confined to the surface layers prior to the AR's arrival, while lower concentrations were observed in deeper waters (approximately 9 mL/L of oxygen and 1 µg/L of chlorophyll-a) (Fig. 6e,f). Turbidity remained low throughout the water column, consistently below 4 NTU (Fig. 6g). Nutrient concentrations (Fig. 5) were moderate before the event, with orthophosphate ranging between 1.4 and 1.6 µmol/L, nitrates between 10 and 12 µmol/L, and silicic acid concentration varying between 10 and 28 µmol/L in surface waters.

As the TAR impacted the area on July 9th, strong winds induced pronounced mixing of the water column (Fig. 6c). Notably, vertical mixing began a few hours prior to the official landfall of the AR, driven by the advance of strong winds associated with the approaching frontal system. This mixing disrupted the previously weak stratification, resulting in a homogenized water column with decreased temperature and salinity just below the surface, with values reaching 12 C and 32 PSU, respectively. The strong mixing was further reflected in low Potential Energy Anomaly (PEA) values, while an increase in Freshwater Content (FWC) confirmed a net freshwater input into the inner shelf. In addition, both oxygen and turbidity increased below the surface by approximately 0.5-1 mL/L and 2-4 NTU, respectively. Prior to the AR, high values (exceeding 200 ) of PAR had promoted chlorophyll growth. Following the AR-induced mixing, higher chlorophyll concentrations were also observed near the bottom (over 2.5 µg/L), though these levels declined in the subsequent hours due to reduced solar radiation during the AR passage.

Following the passage of the TAR, on July 10th, a pronounced decrease in surface salinity was observed, extending to a depth of about 10 m. This freshening confirms the arrival of runoff from nearby rivers, likely including the Itata River and adjacent tributaries, with minimum salinity values on July 11th. On July 10th, the Itata River discharge peaked at approximately 900 , about 235 above the winter climatological average (Fig. 4d). As this event was classified as a TAR, the resulting river plume was advected southward along the coast, influencing the hydrographic conditions in Coliumo Bay. The freshwater input was associated with increased surface oxygen concentrations (up to 10.5 mL/L) and turbidity (up to 10 NTU). The shallow, buoyant freshwater plume enhanced vertical stratification, with BVF values reaching 60 cph in the upper 10 m, indicating a sharp separation of salinity (and density) layers. As wind forcing diminished after the AR, there was a re-stratification which was further reinforced by the sustained arrival of the freshwater plume. This additional freshwater input led to a marked increase in FWC, surpassing the levels observed during the AR. Correspondingly, PEA values also increased, reflecting a progressive strengthening of water column stratification (Fig. 6d). Notably, the temporal evolution of FWC and PEA following the event showed a similar correspondence, suggesting that surface freshwater accumulation became the dominant driver of the vertical density structure during the post-AR period.

Oceanographic conditions following July 15th, including ZAR 3, ZAR 4, and TAR 2, were analyzed using a bottom-mounted SBE HydroCAT deployed at 20 m depth (Fig. 7, left panels). The instrument was deployed again in 2023 at the same location during additional AR events (TAR 3, ZAR 5 and TAR 4) (Fig. 7, right panels). During both periods, ARs triggered significant rainfall, particularly during ZAR 3 and ZAR 5, with precipitation rates reaching nearly 5 mm/h. Several additional rainfall events occurred in 2023 but were not classified as ARs. While the 2022 AR events were relatively short, the June-July 2023 period exhibited more persistent precipitation, including two consecutive ARs (TAR 3 and ZAR 5). This persistence likely contributed to cumulative hydrological effects, such as sustained freshwater input into the coastal system. Although meridional wind speeds reached up to 12 m/s even during non-AR events, AR periods were characterized by more prolonged and sustained winds, suggesting enhanced mechanical mixing over the inner shelf.

Temperature and salinity at 20 m depth exhibited similar responses to AR events, both decreasing as a result of vertical mixing induced by AR activity (Fig. 7b). These changes were more abrupt in 2023, likely due to the longer duration of the events during that period. Turbidity also displayed substantial variability (Fig. 7c). In 2022, three distinct turbidity peaks were observed, on July 17, 21, and 26, coinciding with episodes of intensified bottom currents (Fig. 4f). In 2023, turbidity responses were more directly associated with AR activity, showing a marked increase during ZAR 5, reaching up to 10 NTU (Fig. 7c). Dissolved oxygen at 20 m depth showed limited variability, with small increases during periods of intensified mixing. In contrast, chlorophyll-a concentrations exhibited more distinct changes, particularly during TAR 2 and TAR 4, when values surpassed 1.5 (Fig. 7d). These increases suggest a phytoplankton response likely driven by enhanced nutrient input and/or water column destabilization associated with AR-induced mixing.