Two-Channel System Dynamics of the Outer Weser Estuary—A Modeling Study

They could expand their vision and confront their work with other results found in other systems, showing in this way their influence in a more general international context.

A new paragraph has been added to the discussion section where the findings of this study are compared to the findings of three other studies: Van der Wegen et al. (2012) and Dam et al. (2016) for the Western Scheldt Estuary and Yu et al. (2012) for the Qiangtangjiang Estuary. The comparison is structured in similarities, disparities and unique aspects.

Put more information in Figure 1 (ex: north symbol, coordinates, …)

Figure 1 now contains additional features: A scale bar, the coordinates of the areal overview, a north arrow as well as the constructed training walls and groynes. Additionally, Figure 1 is enlarged to allow a clearer visualization.

The authors used model Delt3D-flow. Please insert more details about the implementation.

Additional information on the Delft3D model are provided. The basic equations of the Delft3D Flow-module are stated in combination with the main assumptions linked to it. Further reference to more detailed descriptions of the underlying physics and their mathematical representation are added. The authors beware the recap and description of the equations itself, as these have been stated and described comprehensively in other studies. Thus, a detailed discussion of these equations is beyond the scope of this study.

What is the source of tidal data?

In the lines 152-157 on page 5 the extraction of the tidal components from a larger model (Maintained by the Federal Waterway Engineering and Research Institute, BAW) is described. Now, a proper reference is added to this section, to make the origin of the applied data more transparent.

The caption and Figure is very confuse

The confusion on the Figure and caption is assumed to be linked to the comments from Reviewer #2 and related to Figure 2 and 4.

Figure 2 is adapted and expanded. Now two plots describe the validity of a MorFac 400 including a more descriptive labeling of the axes and a more detailed caption. Additionally, further and explanatory references to Figure 2 are included in the text.

Figure 4 is revised for clear identification of the individual steps within the CASE method. Furthermore, the caption has been shorten and more precise explanations given.

In section 1.1, when there is information about the groyne construction, presenting the location of those groynes as well as the time of construction, can assist the readers in understanding the arguments better.

The constructed training walls and groynes are now included in Figure 1, visualizing the number of constructions, their extent and location.  Additionally, more details are given regarding the period of construction and reference.

In section 1.1, one of the missing information about the area is the type and distribution of the bed sediment.

A brief summary of the bed material in the channels and tidal flats in the Weser Estuary is given in line 65-67. In order to provide more and clear information on the local sediment distribution, additional information are included (e.g. relative quantity of silt and clay on the tidal flats). Furthermore, an additional reference is made to the sedimentology maps of “Geopotenzial Deutsche Nordsee”.

Can the authors justify their approach in keeping the "bed-slope factor in local flow direction" as constant? Compared to the other factor, can you provide references to the literature to justify your assumption?

Walstra et al. 2004 showed that the bed-slope factor in local flow direction, Alfabs, has “very limited effect, when applied within realistic ranges”. Comparable studies (e.g. van der Wegen et al.) state the default value of 1.0. To make this aspect clear to the reader, a comment and reference has been added to the manuscript in order to fill the gap pointed out.

In lines 117-119, the differences between the “Engelund & Hansen” and “van Rijn” transport formulas are mentioned. This appears to be in contrast with what is stated in lines 115-116 as one chosen value of 7.5 for the bed-slope factor in local flow direction. Did you calibrate for two or one method/value?

Alpha bn, the transverse bed-slope factor, has been calibrated for both transport formulations resulting in final values that differed in a magnitude in the order of 10. As the Engelund & Hansen transport formula is applied for the final simulations, only the corresponding Alpha bn value has been given. For clarification, the statement of the applied Alpha bn value has been moved to the end of the section which deals with the bed-slope factors. Additionally, an explicit statement is added, that connects the given Alpha bn value to the Engelund & Hansen formula.

Figure 2 requires more explanation. In its current format, and a short mention in lines 140-141, it is not clear what the different elements are in the plot, and what is meant by the quantity in the y axis. The unit for the x-axis should also be the year, but not the morphological year.

Figure 2 is extended by subplot containing the final hypsometric curve. The labels of the axes have been adapted in order to establish a more descriptive and clearer character. Furthermore, it is described and explained more detailed in the text. The caption has been rewritten, recycling parts of the old caption in combination with additional clarifications.

The cited model of Jade-Weser-Elbe in line 156, maintained by the Federal Waterway Engineering and Research Institute, requires proper referencing.

A reference has been added, that reports the set-up of the Jade-Weser-Elbe model. ( BAW. Erzeugung naturähnlicher Randwerte für den seeseitigen Rand von Ästuarmodellen an der Nordsee (BAW-Bericht A 395 502 10059) 2014.)

What is concluded in lines 165-175 appears to be that the tide is important. This is then discussed more in lines 176-195. However, when describing the use of MorFac, tide schematization was not considered and explained. I'd suggest discussing some literature such as Bernardes et al. (2006), Latteux (1995) and Lesser et al. (2004), regarding tide schematization when using MorFac.

A paragraph is added that addresses the tide schematization in combination with a MorFac. The mentioned literature (in addition with further publications) has thankfully been adapted to discuss this issue and state the most important assumptions made.

Figure 4 has some issues. 1) The caption is too long, and contains some descriptions that are not found in the text. 2) The colours are too similar in many of the graphs. 3) The unit of time is mentioned ‘my’ rather than the year. 4) The graph on the middle right is incorrectly referred to as the middle left. 5) The axes on the lower-left plot are not defined.

In accordance with the comments on Figure 4 both, Figure 4 and its caption are revised. The subplots of Figure 4 are indicated by the letters a) to e). The Hovmöller plots now contain of a gray scaled bathymetry as base layer with pronounced solid lines for the three exemplary cases on top. This should improve its readability. Additionally, the caption has been shortened and additional references to Figure 4 and subfigures are included in the text. Furthermore, details that have been missing in the text are added and the time is now given in years instead of morphological years.

In line 231, and three other places in the text, the term morphological year is mentioned but incorrectly. In fact, the hydrodynamic time gives the morphological year and using MorFac the actual number of years would be achieved. For instance, running for 1.625 years, with MorFac=400, 650 years of results would be achieved

The authors follow the clear argumentation of the relation between hydrodynamic time, morphological year and the actual time/number of years. Accordingly, the temporal description in Figures, captions and manuscript are adjusted to “years” in order to prevent misconceptions.

In lines 162-163, could the author clarify for which location(s) the hypsometric curves are plotted in Figure 2?

Also, more explanation is required about the parameter in the y-axis of the plots in Figure 2. How was it calculated? For which locations? What is the justification for the cumulative area being reduced or increased for a particular depth range?

The authors mentioned that over the years, there were variations in the tidal range due to the construction of the groynes and training walls. What effects had the tidal range variations on the change in the cumulative relative incidence of a particular depth range?

We reworked parts of the description of the hypsometric curve calculation and aimed for a clear explanation of the location/area (for Figure 2 this is the whole model domain, as the MorFac is affecting the whole model) and the way of calculation. We are not sure, if we get your second and third aspect of the first point correctly, but we would like to respond as comprehensive as possible. The parameter in the y-axis is based on the percentage of appearance of depths in the model domain shown in depth contour lines accumulated ascending from the deep channel areas (-28m) to the tidal shoals (+/-0m). A description of its development and how this justifies the results is also added (lines 182-186).

The effect of the changes in the tidal range, as a result of the constructions, made in the estuary, is very complex. Multiple measures were taken at several points in time, including capital dredging, changing the width and depth of channels. Furthermore, as adaptions were started over a century ago, long term effects such as sea level rise have affected the hypsometry of the Weser Estuary as well. As the combination of these changes would require an extensive analysis of different bathymetries and measures in time, it is considered to be out of the scope of this study, especially, as the training walls and groynes are not included in the long-term simulations carried out, as they would have prevented a natural development of the channel-shoal-system.

As for the statement in lines 171-172, could the authors also present and discuss MorFac values more than 400?

If there is any chance to achieve the results faster with a larger MorFac, it is arguably more beneficial to do so. Conversely, if there would be any issue (e.g., in model convergence) with larger MorFac values, then those would definitely be out of the scope for comparison. So, it is worth mentioning them in the manuscript.

In any case, this reviewer believes that the statement made in lines 171-172 does not provide strong evidence for the authors’ argument.

As you indicated correctly, larger MorFacs may cause model instabilities, especially, when combined with extreme initial conditions (such as a flat-bed). We included this in the MorFac description section and further reworked our argumentation on the statement in line 171-172. Your comment inspired several aspects we changed in that section to give a clear description of the criteria applied for the MorFac validation.

This reviewer can now better understand the plot in Figure 2a. I wonder if it would be possible to remove the black polylines around each of the colored areas. The reason is for them not being confused with what the authors called "thick lines".

We gladly notice that the adjustments from the first review makes Figure 2 better understandable. Your further suggestion of removing the black lines in Figure 2 a) has a very positive effect on the readability of the thick lines and is included thankfully.

This reviewer still believes that the caption for Figure 2 is too long, while there are repeated contents in the caption that are already presented in the text.

We have reworked the caption and removed the descriptive content. Now the caption is considerably shorter (< 3 lines). Comparing the red and the blue version of the caption makes your comment valid.

Following the similar comment above, the caption for Figure 4 is too long.

Also, this reviewer suggests adding the parameter name for each of the axis, rather than relying on the readers to guess or read the caption.

If the authors do not see the possibility of shortening the caption, then they may make separate figures with understandable content.

Choosing red and blue colors to mimic the track change feature did not completely work for this figure caption, as it is not clear now what will remain and what has been removed from the caption.

While the authors aimed at making the colors understandable in the plots, they only changed the colors for panels a and c in Figure 4. The main issue is the similarity of colors in the chosen green-yellow-red pattern, which is repeated on all the figures, as well as the panels for Figure 4. Hence, changing the depth colors for panels a and c has not fixed the issues in panels b and d.

As for Figure 2, the caption of Figure 4 is reworked and drastically shortened (~3.5 lines). It now includes only the indications of the subfigures. Furthermore, the axis of the subplots have been labeled rather than described. Additionally, thank you for specifying the conflict between the applied colors in Figure 4, 7, 8, 9 and 10. The colormap has been adapted and the “vik” colormap (from Crameri scientific colorbars) has been applied for the red and blue color, making it less difficult for readers with colorblindness. The yellowish color has been kept (instead of the white middle part of the vik colormap) for having a contrast compared to the background.

Furthermore, by clustering the changes in the caption in a blue colored text and a red colored text should increase readability.

In Figure 6, which location is considered for the cumulative bed load output?

Also, what is the difference between Figures 2 and 6? In one there are white areas between each depth range. On the other, the depth ranges are alongside each other. Though, the latter makes more sense.

The hypsometry in Figure 6 is based on the area of interest (Outer Weser Estuary). This is now specified in the text as well, by an additional statement.

The difference between Figure 2 and Figure 6 is that Figure 2 only shows the depth contour lines and their variations within the MorFac-validation-simulations, while Figure 6 shows the development of the hypsometry for one simulations, where there is only one dividing contour line. In Figure 2 the filled area indicates the variation of depth contour lines, while the depth ranges itself are filled with color in Figure 6. We hope that this clarifies the white spaces in Figure 2.