Laboratory experiments are used to investigate the effects of injecting fluid at various flow rates and densities on a sloping boundary into a two-layer stratification in a rotating system.  The aim is to examine the effect of overflow waters into the Ionian Sea, and in particular how changes in the density or flow rate can affect the overall circulation within the basin.  This is a substantial experimental effort, conducted carefully, and the work is mostly described well, with an analysis of the partitioning between eddy and mean flow KE, the resulting flow within the basin, and comparisons with numerical model output for the Ionian Sea, assimilating in-situ data for 2012.  Overall this is a valuable piece of work.  There are some aspects of the presentation that should be improved before the paper is published.

1 While I realise details are included in other references, you need to give some more details about how the velocities are calculated from the laboratory experiments.  An image showing a velocity field earlier in the paper, to help explain the main features of the laboratory flow, would also be useful.

2 You mention viscous bottom draining (line 313) but there was no mention of this before – something on this should appear in the Introduction.

3 It would be useful to have a chart of the Ionian Sea sooner in the paper, and you should mark the locations of the main inflows and sketch the typical circulation(s).

4 You need to explain how you calculate MKE and EKE in more detail.

5 The English needs some attention – below I list a few corrections from the Abstract and Introduction by line number, but there are others, and throughout you often write “the experiment 24” or “the phase II” where “the” should be deleted.

30 Density records show

56 as happened in

69 these studies maintain that

80 of dense water in the

82 with observations

83 with a duration of

87 circulation of the open sea

91 of vorticity generation

Fig 1 (not to scale)

The manuscript present a reproduction of an important quasi-decadal mechanism that drives thermohaline oscillations in the Adriatic-Ionian region. I appreciate such a novel approach using tank experiments, which tries to provide the underlying physics that has (still) several possible explanations - this research is important to put proper weight on them (internal forcing vs. wind-stress curl) Therefore, I strongly recommend publication of the manuscript. Still, some issues should be cleared and corrected (also agreeing with comments of Anonymous Referee #1):

- Lines 94-106. It might be more appropriate to have this at the beginning of Section 2 (as an introduction, before Section 2.1), as justifying the applied methodology. 

- Fig. 1. It might be good to increase the font of the smallest labels, they can be hardly read in such a composite figure.

- Section 2.1 or elsewhere. I am wondering how the scaling between the tank simulation and the real Ionian Sea has been done for (turbulent) diffusion? I see no comments on that in the text, while I believe that it might be worth to discuss somewhere. Also, please add and discuss any other limitation or approximation of the tank experiments which are relevant for the presented experiments.

- Figure 2 caption. Please add the area or the location for which density and MLD is presented (at the very centre, Cp3 ?)

- Line 377. Densities of 2010 and 2015 kg/m3 (?)

- Fig. 9. There are two (b) marks in the figure - please correct.

- Lines 434-436. Why did you took 2200 m as the borderline between the slope and the flat central region? Just by looking in the topography 3000 m looks more appropriate for me (plus moving lower boundary line more to the south). Please justify your choice.

- Line 566. If you have doi, you don't need to provide the direct link to the reference.

- Line 572. Missing "doi:"

- References. Please unify: doi or DOI or https://doi.org/... and add missing doi numbers for all references (for these which have them).

- References. Please unify: short or full journal names.

Review of the paper entitle:

 “Impact of the dense water flow over the sloping bottom on the open-sea circulation: Laboratory experiments and the Ionian Sea (Mediterranean) example”

by

Miroslav Gačić, Laura Ursella, Vedrana Kovačević, Milena Menna, Vlado Malačič, Manuel Bensi, Maria-Eletta Negretti, Vanessa Cardin, Mirko Orlić, Joël Sommeria, Ricardo Viana Barreto, Samuel Viboud, Thomas Valran, Boris Petelin, Giuseppe Siena, Angelo Rubino.  

General comments. 

The scientific matter of the manuscript isn’t a really new; actually in the literature there are many example on this, either in the modelling field or in the analysis of the in situ observations and correctly in the manuscript there are a long list of key references. However, the manuscript tackle an important and new relevant scientific issue dealing with the analysis of the ocean processes related to the propagation, spreading and adjustment of the density anomaly in a complex topography like as it is the Ionian Mediterranean sub-basin. This study, specifically the tank-experiment, reveals the important role of the n-layer stratification in the vortex rotation within the framework quasi-geostrophic model on an f-plane. For all these reasons that the results of this paper are very interesting for the oceanographic communities and in particular for those scientists more implicated on the Mediterranean studies.  However the present version of the manuscript have a lot of a weak points and therefore is not ready to be published for the following reason.  

Major revision: Among the weak points the following ones is the most relevant:

 •                The manuscript encompassing a comprehensive introduction of the experimental apparatus and methodology followed by the authors and every thing is very well conducted, except the relative role (scale ratio) of the central part of the tank respect to the sloping part. In fact, looking the figure 7 in the manuscript seems that the dynamics driven by the slope domain dominates on those generate in the flat domain, making very difficult distinguishing the difference between the two dynamics. This isn’t irrelevant to make more realistic the comparison with the northern Ionian Sea circulation in section4.

•                Moreover is very confusing the theoretical and modelling equations that are used to analyse the experimental results: the equation 1 is not the same used by the cited paper of Lee-Lueng, F. and Davidson, R. A. (A note on the barotropic response of sea level to time-dependent wind forcing. J. Geophys. Res., 100, C2, 24955-24963, 1995) that use a classical linear barotropic vorticity equation, may be the authors have to use a different reference.

•                However, the most relevant matter is related to the stratification that, at the end, is the core problem of the manuscript. It is well know that a good representation of the ocean dynamics is a three-layer system and this is this is confirmed even in this case as is well evident in the figure 2c (experiment 27), specifically around the 75th day in which we see the respond of the pressure to the injection of the density anomaly and subsequence stratification in three layer (or a continuously stratification see references), is very interesting the impact of the redistribution of density and pressure within the water column in the figure 3 (and also figure 5) experiment 27 at the same day (around the 75th). These figures are the most interesting of the manuscript and at the same time are those that demonstrate the weakness of the theory presented in this manuscript: actually can’t demonstrate the opposite vorticity at the 75th day and the corresponding kinetic energy anomaly in the lower layers. However, despite this experimental evidence and the same recognition as the authors themselves that the dynamics follow at least a two-layers system, even so at the end the equation that the authors used is written in a one-layer formulation. This is not irrelevant for physical point of view. Is matter of fact that dealing with one, two or three layer formulation of the QG equation, produce a different vorticity relation between the several interfaces along the water column (see Sokolovskiy paper and all reference herein). This is true either in the flat or in the slope domain and finally on the comparison with the realistic example of the Ionian sea.  In conclusion the circular rotating tank experiment shows in an impressing way (this could be more impressing with a different scale ratio between the slope/flat domain), the adjustment of the vorticity along the continuously stratified water column (and its dependence from the layers-thickness distribution) when it is subjected to a density anomalies: first along the slope and afterwards during the spreading of the anomaly density flow along the flat bottom; finally is very arduous to do some comparison, in the present version of the manuscript, between the tank experiment and what was observed in the northern Ionian sea in 2012.  

Minor revision:

·       Line 326 “level 1” is referred to inclined laser sheet levels 1 of the Figure 1?

·       Line 451 at which model is referred? Please give more details;

·       figure 1 the word “cp3” is not clear;

·       figure 2 in the color tab 0-15 means that the range of density is between 1000-1015?

References.

1.     M. A. Sokolovskiy; Stability of an Axisymmetric Three-Layer Vortex, Izvestiya, Atmospheric and Oceanic Physics, Vol. 33, No. 1, 1997, pp. 16–26. Translated from Izvestiya AN. Fizika Atmosfery i Okeana, Vol. 33, No. 1, 1997, pp. 19–30. (and all references herein)