|Re-review of ‘Analytical approach for determining the mean water level profile in an estuary with substantial fresh water discharge’ by Cai et al.|
Many thanks for the various clarifications in this revision. As it stands, the manuscript focusses too much on the analytical method and very little on the overall impact of the results. In the abstract and the introduction the main motivation is set as to provide new insight of river-tide interaction processes in tidal rivers, but this is not even put forward in the introduction. In fact, one of the main conclusions of the paper has to do with a methodological implication (see Lines 426-429) rather than, for instance, with how the new insight gained in this study can help in understanding other systems. I still have concerns that I will present in detail below.
Abstract (Page 1, L5). Several papers have dealt with this topic in the past and if there are issues not yet understood you should summarize the results from these studies and explain what is missing. In the introduction the authors mention many of these studies but fail at pointing out their outcome, instead, the authors seem to concentrate on the methods employed by these studies.
Abstract (Page 1, L10). The fact that river-tide interaction is relevant in the tide-dominated zone and the river flow in the river-dominated zone is already known from previous studies, however, the effect of the tide on the transitional zone is new to me, and not necessarily an obvious result. I think this should be made clearer.
Abstract (Page 1, L15). It is not clear if these results come from the application of the model to the specific case of the Yangtze or are generic from a parameter analysis. I think it is the former, since Figure 11, panel c and d, supports those results. If this is the case, you should mention that the model is applied to the Yangtze case prior to reporting the results.
Abstract (Page 1, L17). I think the word ‘demonstrate’ is misleading here, since there are no validation data available. Similarly, there is no ‘prediction’ (Line 19).
Introduction (Page 2, Lines 33-59). As stated above, a summary of the results by previous studies would be beneficial to understand these new results. I think the methodological discussion should stay, but the paper is clearly motivated by the physical implications of the processes discussed, and therefore, this should be made clear in the introduction and the conclusions.
Introduction (Page 3, Lines 76-82). This paragraph should be moved to Section 2.2 as it presents arguments for neglecting the effect of density in the momentum equation.
Section 2.2 (Page 4, Lines 119-120). These studies do not directly show the tidally averaged water level rises in landward direction. The first to recognize this rise in water level were Godin and Martinez (1994). Following observational studies are indicative of this effect.
Section 2.3 (Page 7, Lines 225-230). It is not yet clear how the Ft, Fr and Ftr terms from the present approach are different than those derived by Godin. Can you please include a plot comparing these two approaches so that the reader can see how Godin approach becomes inadequate?
Section 2.3 (Page 8, Line 241). I cannot find Section 5 in this paper (I think is the other, 2014b)
Section 2.3 (Page 8, Lines 247-248). I am puzzled about this statement. If the results are similar, why bother with a new approach? This is not explained in the manuscript.
Section 2.4 (Page 8, Eq. (22)). Can you please point which of these parameters are function of time. Figure 11 shows that small changes in v and Ur lead to large changes in F’s, so would be very useful to show the longitudinal profiles of these parameters (eta and delta) so as to assess model adequacy
Section 3.3 (Page 10, Lines 318-339). The calibration procedure is still unclear in the manuscript. Specifically, which stations are used to obtain the friction coefficient? What portion of the data is employed? Do you directly compare the resulting time series for the amplitudes (eta)? Is Figure 9 the result of the calibration? If so, you should include another figure (similar to Figure 8) where you show the results of the calibration, comparing time series for modelled and observed amplitudes for the stations where calibration is performed.
Section 3.4 (Page 11, Lines 353-355). In Figure 11 you are plotting the F’s for averaged values of tidal amplitudes and river discharge. In my opinion, you should evaluate those functions for each time step and then average the resulting values. These two quantities are different because the F’s are nonlinear. Figure 11 shows \overbar(F) when in reality you are computing F(\overbar(Q),\overbar(eta)). You should clarify this in the caption of Figure 11.
Section 3.5 (Page 12, Lines 373-376). I still think this section is out of place as it is in present form. Accurate prediction and reliable estimation are not consistent with reproducing the first-order dynamics of the system. I agree the latter has to do more with understanding processes and explaining the role of mechanisms, but is far from use in preventive water management. I think the authors are claiming too much in this section with very little support from the results, since there are no validation data. Also, in the response you state that asymmetry due to the river discharge is included, but is likely that asymmetry due to interaction of M2 with itself is more important, right? In fact, these two mechanisms can be out of phase along the river so one could expect anything from it. I think would be worth including a discussion about this.
Table 2. Parameters Ar and A0 should have units of squared meters.
Figure 3. The phase lag is not a dimensionless quantity.