Temporal and Spatial Variability Scales of Salinity at a Large Microtidal Estuary
Round 1
Reviewer 1 Report
A 3D numerical model was applied to simulate the characteristics of a micro-tidal large estuary. The authors analyzed the salinity stratification under fluvial flow and storm influence and concluded annual and seasonal tendency using the 10 years results. They also summarized how the salinity vertical distribution would change under different wind direction during storm events. Results show some interesting findings and are well presented. However, the experiment design is not clear enough and the model’s performance is not convincible, which means a significant part of a numerical simulation is missing. I’d like to recommend a minor revision. Please find my comments below.
Line 57: May also refer to Huang et al (2020) and other related works which have similar research interests.
Huang, W. C. Li, J. White, S. Bargu, B. Milan, and S. Bentley, 2020. Numerical Experiments on variation of freshwater plume from Mississippi River diversion in Lake Pontchartrain estuary, Journal of Geophysical Research: Oceans,125(2), C2019JC015282, doi:10.1029/2019JC015282.
Line 58-59: What is “the little available information”? Is this referring to any references?
Line 125-128: May indicate in the figure the boundaries of the inner zone, intermediate zone, and the outer zone.
Line 129: please consider using a color image with higher resolution. Font size of the text in the figure is hard to see. It would be great if the authors can show the bathymetry loaded onto the model grid, so that we can know if the channels and other features are well resolved.
Line 200: it would be better to show the whole mesh domain and indicate the open boundaries.
Line 214: 0.02 degree equals to ~2km already. Is it too coarse to represent the bathymetry of the lake?
Line 187: what is the simulation period?
Line 259: Figure2: the text on the figure is too small, this reviewer couldn’t see, other figures have the same issue.
Line 256: Figure2: the model didn’t do well for the stratified vertical distribution: the subplot in row 2 column 3, row 3 column3, row 4, column1, and all other stratified distribution. Seems that the model was only able to reproduce the stratified distribution for row 7 column2 and the end row column 1.
Line 231: why there is no validation for water level and velocity, which is also very important aspects for a model skill assessment?
Line 292 – 357: The paragraphs are discussing about the influence from the river discharge, does that mean the model was run without any other forcing like atmospheric forcing or tidal forcing? How the simulation was designed? Also, the abstract may need to briefly summarize the major findings from these paragraphs.
Line 360: the authors may need to justify why they choose a year with high annual mean flow under storm influence.
Author Response
Line 57: May also refer to Huang et al (2020) and other related works which have similar research interests.
Huang, W. C. Li, J. White, S. Bargu, B. Milan, and S. Bentley, 2020. Numerical Experiments on variation of freshwater plume from Mississippi River diversion in Lake Pontchartrain estuary, Journal of Geophysical Research: Oceans, 125(2), C2019JC015282, doi:10.1029/2019JC015282.
The reference was very interesting and was added.
Line 58-59: What is “the little available information”? Is this referring to any references?
The little available information refers to the previous sentence ‘There is not much information about salinity distribution in the Río de la Plata’. The references are listed later in the parragraph (Guerrero et al (1997), Framiñan et al. (2008), Sepúlveda et al (2013), Simionato et al (2007)). The sentence was rephrased as follows: ‘This information indicates that it is an estuary that can be characterized as salt-wedge, partially mixed estuary, and even well mixed, according to the temporal scale or the location inside the estuary.’.
Line 125-128: May indicate in the figure the boundaries of the inner zone, intermediate zone, and the outer zone.
Actually the boundaries among the different zones were indicated in the Figure 1. There are lines between the coastal cities and the name of the each zone. Maybe the quality of the Figure 1 in the pdf manuscript doesn’t allow to see them, but it is clear in the docx manuscript version. In order to emphasize this information we improved the figure moving the names to the respective zone and using a larger font size.
Line 129: please consider using a color image with higher resolution. Font size of the text in the figure is hard to see. It would be great if the authors can show the bathymetry loaded onto the model grid, so that we can know if the channels and other features are well resolved.
There was a problem with the figures quality in the pdf version. Nevertheless, we consider the reviewer suggestions. The font size of the texts in the figure have been increased. The bathymetry loaded onto the model grid is presented in the Appendix Figure A.
Line 200: it would be better to show the whole mesh domain and indicate the open boundaries.
We agree with the reviewer. The whole mesh domain is described in line 211-213 and is showed in Figure A of the Appendix. Nevertheless, we indicated the open boundaries in Figure A.
Line 214: 0.02 degree equals to ~2km already. Is it too coarse to represent the bathymetry of the lake?
The resolution is enough to represent the main morphological features of the Río de la Plata bathymetry displayed in Figure 1. There are other implementations of numerical models with even lower resolution (Fossati et al, 2013; Fossati et al, 2014a) which are able to represent properly the estuarine hydrodynamics.
Line 187: what is the simulation period?
This was clarified with the sentence ‘In this work an extensive period of ten years was simulated (2006 - 2015)’ in line 222.
Line 259: Figure2: the text on the figure is too small, this reviewer couldn’t see, other figures have the same issue.
There was a problem with the figures in the PDF version, where they are very poor quality unlike de Word version. We talk to MDPI editors to fix the problem and we sent them the figures in good quality through wetransfer. They should be in better quality for the next review.
Line 256: Figure2: the model didn’t do well for the stratified vertical distribution: the subplot in row 2 column 3, row 3 column3, row 4, column1, and all other stratified distribution. Seems that the model was only able to reproduce the stratified distribution for row 7 column2 and the end row column 1.
We partially agree with the reviewer comment. Using the same nomenclature we think the stratified distribution is well represented in row 2 column 1, row 7 column 2, row 11 column 1 and 2 but in several more cases the model represent the vertical stratification (row 2 column 2, row 3 column 1, row 5 column 1, row 6 column 1) or the tendency in the vertical profile (row 3 column 2, row 4 column 4, row 7 column 3, row 10 columns 1, 2 and 4, row 11 column 4).
Knowing that the comparison between model and CTD salinity vertical profiles collected instantaneously in several locations is very exigent for the model, we decide to show that the model generally reproduce both mixing conditions and stratification condition that are observed in in-situ data for the same station despite large difference between the salinity values.
Line 231: why there is no validation for water level and velocity, which is also very important aspects for a model skill assessment?
We agree with the reviewer about the importance of hydrodynamic validation. The water level, velocity and salinity validation is presented in the appendix.
Line 292 – 357: The paragraphs are discussing about the influence from the river discharge, does that mean the model was run without any other forcing like atmospheric forcing or tidal forcing? How the simulation was designed? Also, the abstract may need to briefly summarize the major findings from these paragraphs.
In all the simulations the model was run with all forcings described in Model Description (river discharge, atmospheric forcing and tidal forcing). There are not idealized simulations. The modeling methodology consist in including the main forcings of the system (fluvial flow, astronomical and meteorological tides, and winds) in order to obtain an approximation of the ‘more real conditions as possible’ with the model. In this way we include for example a reliable representation of the temporal variations of the storms and the conditions of tides when the storms arrives, or the effect of the previous hydrodynamics and salinity conditions. Simulating several years we can improve our knowledge of the system by analyzing the response of the system in a wide range of forcings combinations (i.e. different storm conditions, fluvial flow rates, tidal conditions) and when we analyzed them.
The river discharge influence was analyzed in an annual temporal scale. For that the model results (stratification in an hourly frequency) were average during each year obtaining the annual mean stratification.
In the abstract we mentioned the main results (The correlation between fluvial flow and the salinity field showed that high annual fluvial flows generate an extension of the freshwater area, with larger longitudinal salinity gradients and the shift of the salinity front towards the ocean). Nevertheless considering the reviewer suggestion we included further information in the abstract:
“The temporal variability of the salinity stratification was characterized at different temporal scales: annual, monthly and storm. At the same time, the influence of fluvial flow and winds was determined. The correlation between fluvial flow and the salinity field showed that high annual fluvial flows generate an extension of the freshwater area, with larger longitudinal salinity gradients and the shift of the salinity front towards the ocean. The tendency at the monthly scale is not as clear as the observed at the annual scale.”
Line 360: the authors may need to justify why they choose a year with high annual mean flow under storm influence.
The storm influence was studied considering two years with very different annual mean flow in order to represent a wide range of fluvial flow conditions (in Figure 3 the simulated years are order according to increasing annual mean river flow). The obtained results are the same for both years showing the fluvial flow is not relevant in a shortly temporal scale (i.e. the storm scale) of about 1 or 2 days. This was clarified in the text with the paragraph:
“The analysis of the salinity variability during storms was based on the numerical results for the years 2012 and 2014; one year with low annual mean flow and another with high annual mean flow, respectively. These years were selected in order to represent a wide range of fluvial flow conditions. “
Reviewer 2 Report
Figure 1: This image has very low quality. It would be better if it is replaced with a colored, high-quality version.
Line 200: Replace "Martinez et," with "Martinez et al.,"
Figure 2: This image has very low quality as well. It is impossible to see the lines, the legend, etc. It is difficult to see, but the model does not seem to be in good agreement with the observations. Can the authors comment on that?
Lines 265-274: Can the authors please provide references for the stratification calculation method they used? I am also wondering why the author did not use the buoyancy frequency (N2).
Figure 3 to 13: All the figures in this paper have very poor quality. It is not possible to read the axis, numbers, etc. Please provide better quality figures. The axes are not properly labeled (x and y labels). Colorbars should be labeled as well.
I think the paper has good merit overall, but it needs to be improved before publication. I would also recommend using an editor/native speaker to edit the language of the manuscript.
I would like to see the improved/modified figures before I can provide further feedback.
Author Response
Figure 1: This image has very low quality. It would be better if it is replaced with a colored, high-quality version.
We don’t know why the figures in the pdf version of the manuscript were in low quality resolution. In the docx version of the manuscript we see the figures correctly. We asked MDPI for help about this issue and we sent them the figures in good quality through wetransfer. They should be in better quality for the next review.
Line 200: Replace "Martinez et," with "Martinez et al.,"
Corrected.
Figure 2: This image has very low quality as well. It is impossible to see the lines, the legend, etc. It is difficult to see, but the model does not seem to be in good agreement with the observations. Can the authors comment on that?
About the quality of the figure we already answered in the first comment.
Knowing that the comparison between model and CTD salinity vertical profiles collected instantaneously in several locations is very exigent for the model, we decide to show that the model generally reproduce both mixing conditions and stratification condition that are observed in in-situ data for the same station despite large difference between the salinity values.
Lines 265-274: Can the authors please provide references for the stratification calculation method they used? I am also wondering why the author did not use the buoyancy frequency (N2).
The stratification was calculated on a simple way using salinity values at bottom and surface. Using the buoyant frequency would have been also suitable for the work. Since the water temperature in the model is constant the density only depends on the water salinity. The buoyant frequency (N2) is another simple way to analyze the stratification but we prefer to work with the water salinity, not using the water density.
Figure 3 to 13: All the figures in this paper have very poor quality. It is not possible to read the axis, numbers, etc. Please provide better quality figures. The axes are not properly labeled (x and y labels). Colorbars should be labeled as well.
Already answered in the first comment.
I think the paper has good merit overall, but it needs to be improved before publication. I would also recommend using an editor/native speaker to edit the language of the manuscript.
We agree with the reviewer. The manuscript was sent to the MDPI English editing service.
I would like to see the improved/modified figures before I can provide further feedback.
We are really sorry for the figures quality problem. They were sent with a good quality (in the docs manuscript it is possible to see them correctly) but in the pdf the quality of the figures is very bad. We asked MDPI for help about this issue and we sent them the figures in good quality through wetransfer. They should be in better quality for the next review.
Round 2
Reviewer 2 Report
The authors answered most of my questions. However, the figure quality still remains as a problem. They claimed the original figures are high resolution, so maybe published version will better quality. On the other hand, the authors did not address my comments regarding the figure axis labels. What are the x- and y-axis of the figures? The colorbars also need a label.
Author Response
Sorry for not addressing those suggestions. We labeled all the colorbars. Regarding the figure axis labels we chose to add the limits of the maps in the figure caption. We made this decision because labeling the axis in each figure lost the attention of what we really wanted to show.
