Review of "River effects on sea-level rise in the Río de la Plata during the past century" by C. Piecuch (OS 2022)

Summary:

This manuscript investigates the dynamic link between the variability of Rio de la Plata discharge and coastal sea level variability observed in the immediate vicinity of its outlet, primarily focusing on the long timescales. It starts with an in-depth analysis of the few long tide-gauge and streamwflow datasets available there, which reveal increasing trend of both the discharge and the coastal sea level over the past century. Then it proceeds with two idealized modeling frameworks that are derived in order to explain the observed co-existence of discharge and sea level trends, in a causal fashion. The first framework is essentially a one-dimensional barotropic frictional model, where the water-air interfacial slope is balanced by the bottom friction in the along-estuary direction. This framework is applied to the inner estuary, around Buenos Aires. The second framework is a more complex two-dimensional baroclinic framework, where the plume would induce a coastal jet of brackish water, in thermal wind balance with a sea level higher at the coast than further offshore, off the plume offshore edge. It is applied further downstream, around Montevideo. It is concluded that both these modelings results, although based on highly idealized assumptions, stand in very good agreement with the observed relationship between discharge trends and sea level trends, for both regions. Hence it is concluded that the link between Rio de la Plata discharge trend and sea level trend in these two regions is causal in nature.

General comments:

This study nicely tackles the long-lasting issue of long-term trends of sea level in estuarine ambients. It does so in a very relevant region, home of virtually the sole long-term observational records of the east coast of South America, itself in a very poorly observed basin as far as sea level is concerned: the southern Atlantic. The author makes clever use of the few observational records available, be it of streamflow or of coastal sea level. The approach is sound, and the results are convincing, in that the manuscript consistently backs its findings with statistical analyses. The merit of the study is to end up with a very simple conceptual framework that manages to explain the potentially complex and non-linear dynamics underlying the observed relationship between discharge and coastal sea level. This said, I am somewhat doubtful about the practical strategy of the author when it comes to test the validity of his two idealized frameworks against his observational findings, through the choice of numerical parameters of the required quantities (see my specific comment hereafter). I strongly encourage him to assess the relevance of his idealized models against one (or ideally several) OGCM outputs, typically considering the latest class of eddy-admitting CMIP6 historical runs. This would add considerable strength to the present manuscript. There are very concrete and practical implications of the findings reported here, as regards to the general understanding of the sea level budget and its closure over the southern Atlantic basin, as the observational databases  available therein are deeply influenced by the handful of  stations analyzed in the present manuscript.

Specific comments:

-l. 90: " To (...) reduce dimensionality": isn't there a more fundamental reason than just reducing dimensionality? Reducing the impact of observational errors, typically?

-l.  195: "As a mental model": an actual schematic would help the reader here.

-Table 4: shouldn't U be simply dictated by the geometry of the estuary at x=0, given the value of q observed? It is unclear what relevance the tidal current has in this steady-state model. Is the value chosen for H₀ some sort of optimum resulting from a tuning, so as to achieve best consistency of the prediction wrt to the result of the regression analysis based on observations?

-l. 254-256: " These theoretical estimates agree with the coefficient of (7.3±1.8)*10−6  m m−3  s and the streamflow-driven 255  sea-level trend of 0.71±0.35  mm yr−1  found earlier from regression analysis of observed streamflow and sea level at Buenos Aires": indeed, the values do agree very, very well. Hence it is needed here to get a feel of the extent of ad-hoc tuning, implicit  in the choice of parameters listed in table 4, so as to ensure this quasi-perfect match.

-l. 257-260: I have the same concern for the results of the Montevideo idealized model.

-l. 322: " Theories developed here may be helpful in this regard": indeed, nowadays there is a whole batch of centennial model outputs that became open for public use, several of which resolve -at least partly- the baroclinic Rossby radius of deformation at the latitude of Rio de la Plata. As part of CMIP6 for instance, multi-centennial historical simulations of the present climate as well as century-long projections have recently become commonly available (see e.g. Held et al 2019 https://doi.org/10.1029/2019MS001829, among many others). If indeed the two idealized dynamical balances proposed in the present study successfully explain the observed relationship between discharge variability and coastal sea level variability in the estuary at long timescales, this relationship should be captured by these long model simulations that have the full physics required to  capture these, and much more (in particular that have realistic mixing schemes, and do not impose the idealized frontal structure of the plume density nor its linearly stratified density profile). These OGCMs have, with their 1/4° typical resolution in the ocean, the capability to resolve to a fairly large extent the thermal-wind balance invoked in the present idealized framework. I strongly encourage the author to consider at least one of this class of state-of-the-art model historical simulations, and to assess the relationship between Rio de la Plata discharge trend, along-shore equatorward coastal current trend slightly downstream of the outlet, and cross-shore sea level slope trend, at the long timescales of interest here. If the idealized framework presented in the present manuscript holds there as well, this would add considerable strength to the results reported here (as the conclusion would not depend in any fshion on the potentially subjective choices of parameters listed in Table 4). The encouraging results observed from altimetry call for such an independent assessment of the idealized framework.

Technical corrections:

-l. 48: centred

-fig3 caption: "thick black" line is not seen

-l. 149-150: it should be Sections 4.1, 4.2 and 4.3

-l. 249: should be sections 4.1 and 4.2

General comments

The manuscript displays the role of the streamflow in the sea level variability especially at long-term trends in the Río de la Plata estuary. To fulfil the objective, annual data from tide gauges and stream gauges are analyzed. The main results indicate that the streamflow is not negligible in the sea level variability and in the long-term trend, except in the south of the river mouth. The river effect increases from the lower estuary to the upper estuary, explaining almost the 60% of the sea level variance. To corroborate that the streamflow is responsible of a percentage of the sea level trend, the author developed a theorical model finding a coherence between the simulated/predicted data and the observations.

The work presented is a hot topic from the climate change point of view. To understand the forcings of the sea level rate in coastal and regional areas is extremely important to prevent and mitigate the consequences. The work also contributes to the analysis of unexplored region compared with other part of the world. Most of the studies in the Río de la Plata estuary were focused on the analysis of the plume dynamics from synoptic to interannual temporal scales using models (e.g: Meccia et al., 2009; Dinapoli et al., 2021; Bodnariuk et al., 2021) and satellite data (e.g., Saraceno et al. 2014). Only a few works showed the sea level rate, however, the causes of the trends were not fully investigated.   

Regarding the presentation quality, the manuscript is well-written and well organized. The figures and tables represent the results written.

Specific comments

Title: I suggest adding “Estuary” after “Plata”

3. Results: Taking advantage of a long sea level record, I suggest studying the acceleration of the sea level rate and the possible relationship with the streamflow, especially in Buenos Aires. The bibliography cited in the manuscript indicates that the sea level is increasing, however, the analysis of a possible acceleration has not been published in the study region.

Pag. 4, line 119: see comment on Conclusions

5.Conclusions

Pag. 11, line 290: The author mentioned that the river effects on sea level are apparent at multidecadal and centennial periods. However, I did not find convincing evidence on the paper. There is a discussion based on bibliography about the ENSO signal, the author calculated the correlation between ENSO index and in situ data, and the standard deviation of the streamflow but I was expected a spectral analysis (e.g., wavelet) to asseverate that other signals are also important. For example, it would be interesting to analyze the cross wavelet transform between streamflow and sea level measurements. Regarding the ENSO as an interannual variability, Bodnariuk et al. (2021b) analyzed the effect of SAM (Southern Annular Mode) on the Río de la Plata using a reanalysis model (35-years). The influence of SAM on the sea level was also studied in a wider region including the Mar del Plata tide gauge location (Bodnariuk et al., 2021a; Lago et al., 2021).

Technical corrections:

Replace “Section 4.a” and “4.b” with “4.2” and “4.3”

Figure 3 caption: the colors of the thick lines of Río de la Plata, Río Paraná and Río Uruguay do not match with the legend of the time series.

Figure 4 caption: the line styles of the time series do not match with the legend.

References:

Bodnariuk, N., Simionato, C. G., & Saraceno, M. (2021a). SAM-driven variability of the southwestern Atlantic shelf sea circulation. Continental Shelf Research, 212, 104313.

Bodnariuk, N., Simionato, C. G., Osman, M., & Saraceno, M. (2021b). The Río de la Plata plume dynamics over the southwestern Atlantic continental shelf and its link with the large scale atmospheric variability on interannual timescales. Continental Shelf Research, 212, 104296.

Lago, L. S., Saraceno, M., Piola, A. R., & Ruiz‐Etcheverry, L. A. (2021). Volume transport variability on the northern argentine continental shelf from in situ and satellite altimetry data. Journal of Geophysical Research: Oceans, 126(2), e2020JC016813.

Saraceno, M., Simionato, C.G., Ruiz-Etcheverry, L.A., 2014. Sea surface height trend and variability at seasonal and interannual time scales in the Southeastern South American continental shelf between 27◦S and 40◦S. In: Continental Shelf Research, vol. 91, pp. 82–94. https://doi.org/10.1016/j.csr.2014.09.002.