Nearshore Wave Dynamics at Mangalia Beach Simulated by Spectral Models

Round  1

Reviewer 1 Report

The paper presents the wave condition acting the target beach by employing a spectral model. The spectral model is Mike 21 SW developed by the Danish Hydration Institute but the application and modification for the target beach have been done by the authors. The computation results were validated by the observed wave data and the seasonal variation of wave action was discussed. A few papers have studied the west coasts in the Black Sea and the calculation results may become useful for future planning beach nourishment. Therefore the paper is suitable as a contribution to JSME after the following revisions.

 1)     The wind prediction is inevitable in order to forecast the wave transformation. What type wind data is employed to determine the offshore condition of SWAN?

2)     You mentioned about the buoy data at Page 3. Please make a short comments about buoy location, depth etc.

3)     Figure 2(b) is not clear. Please replace the figure with larger characters.

4)     Location P1, P2 and P3 in Figure 3 is not expressed in the other figure. Please indicate the locations of each point.

5)     In Figures 4 and 5, the note in upper left is too small.

6)     At Line 263, please use the number for reference.

7)     At Line 287, “Y-shaped dike” is suddenly appearing. A short explanation is necessary.

8)     At Line 315, “sediment accumulation” is discussed, but the sediment profile change is not computed in the paper. Please re-consider the conclusions. The simple facts obtained in the observation and calculation should be included in the conclusion.

Author Response

ANSWERS TO THE REVIEWERS' COMMENTS

 Manuscript ID: jmse_525240   Title: Nearshore wave dynamics at Mangalia Beach simulated by spectral models

GENERAL COMMENTS

A major revision has been carried out following carefully all the indications, suggestions and observations formulated by the reviewers.

The main changes operated are outlined next:

 1) As suggested, an overall revision of the text was carried out in order to correct some small mistakes all along the work. A spell check and editing of the English language and style was also carefully performed. Moreover, many phrases and paragraphs have been reformulated in order to provide a better picture of the objectives and the results of the proposed work.

2) New details have been considered.

3) A new figure (Figure 1) was added.

Finally, the authors would like also to sincerely express their gratitude to the reviewers for their valuable comments that obviously helped in improving the quality of the present work.

At this point, it has to be highlighted that the authors tried in a very honest way to follow all the suggestions and observations formulated by the reviewers and also to operate (as much as it was possible) all the corrections indicated by them. In order to follow the corrections operated in the paper, the option track changes was used.

The specific corrections operated according to the suggestions of the reviewers are given next together with detailed explanations.

Reviewer 1

Comment 1: The wind prediction is inevitable in order to forecast the wave transformation. What type wind data is employed to determine the offshore condition of SWAN?

The type of wind data used to force the SWAN model is reanalysis provided by NCEP-CFSRv2 (U.S. National Centres for Environmental Prediction - Climate Forecast System Reanalysis Version2). These wind fields were used for both levels of the wave modelling system. The information about the type of wind data is now provided in the revised manuscript.

Comment 2: You mentioned about the buoy data at Page 3. Please make a short comments about buoy location, depth etc.

Following the observation formulated by the reviewer, a more detailed description of the buoy was conducted. The following text was added in the manuscript:

Field data are essential to evaluate the accuracy of the model results. In this work, the Coastal Gauge (MEDA) Mangalia buoy maintained by GeoEcoMar (National Research and Development Institute for Marine Geology and Geo-ecology) was used for the model calibration and validation. The buoy is located at latitude 43.802176° N, and longitude 29.602483° E [WGS84] at a depth of 15 m (see Figure 1). The measurements were made in the framework of the project "Marine Pollution Oriented Marine Policies - PERSEUS" from OCEAN 2011-3, "Assessing and predicting the combined effects of natural and human pressures in the Mediterranean and the Black Sea for better their administration ". The project has been finished, but the wave measurement instruments installed continue to operate the data series recorded can be viewed using a specific FlowQuest 1000 + Wave program, both online and offline [22].

Comment 3: Figure 2(b) is not clear. Please replace the figure with larger characters.

Figure 2 (b) was changed in accordance with the requirements.

Comment 4: Location P1, P2 and P3 in Figure 3 is not expressed in the other figure. Please indicate the locations of each point.

Information on the location of points P1, P2 and P3 can be found in the next paragraph, representing 3 of the 4 boundaries of the field of study. The fourth limit is the shoreline.

The boundary conditions, used in the simulated model in Mike 21, were obtained from SWAN simulations in three points used as limits:

·          P1 - x=28.602 [deg]; y=43.816 [deg]; depth= -11m, North boundary;

·          P2 - x=28.598 [deg]; y=43.801 [deg]; depth= -11.95m, South boundary;

·          P3 - x=28.614 [deg]; y=43.807 [deg]; depth= -15.53m, East boundary.

 In the revised manuscript also a new figure was added (Figure 1) where the position of the points are indicated.

Comment 5: In Figures 4 and 5, the note in upper left is too small.

Figures 4 and 5 (in the revised manuscript Figures 5 and 6) have been modified according to the requirements of the reviewer.

Figure 5. Comparison between the MEDA buoy records for significant wave height (in red) and the results of the MIKE 21 SW model (in blue), January 2016.

The second wave parameter investigated in this study is the mean wave period and as can be seen from Figure 6, the fluctuations of the simulated mean wave periods are not very different from the observed data.

Figure 6. Comparison between the MEDA buoy records for mean wave periods (in red) and the results of the MIKE 21 SW model (in blue), January 2016.

Comment 6: At Line 263, please use the number for reference.

At line 263 the reference number used was mentioned.

Kabiling et al [22] presents the definition of uncertainty or “coefficient of variation” as:

Equation (12)

where  represents the uncertainty for the wave height,  represents the standard deviation (or ‘error’) and  is the mean value of the  wave height [26].

Comment 7: At Line 287, “Y-shaped dike” is suddenly appearing. A short explanation is necessary.

We removed this paragraph because it is not relevant for the sediment transport.

Comment 8: At Line 315, “sediment accumulation” is discussed, but the sediment profile change is not computed in the paper. Please re-consider the conclusions. The simple facts obtained in the observation and calculation should be included in the conclusion.

Thank you for this suggestion. Following this observation formulated by the reviewer, the new conclusions were included in the text:

Based on the results presented in this study it can be concluded that using the Mike 21 SW model, forced with boundary conditions provided by SWAN model, reliable simulations of the wave conditions near to Mangalia harbor have been obtained after the calibration of the model. The wave parameters simulated by Mike 21 SW have been compared with the measurements obtained from MEDA buoy maintained by GeoEcoMar. The statistical parameters show a good agreement between both data.

The Black Sea being an enclosed sea with limited fetches, the predominant type of waves is seas that are affected by the local wind. It was observed that in winter, when the wind speeds are higher and with a high variability of the mean direction, the beach is affected by greater erosion and result in two different beach profiles.

During the winter months (especially in January) the storms are more frequent  and the wave height increases (being influenced by the changes in wind direction and wind speed) to near mid spring. This leads to the formation of beach berms, shaping the beach in a more concave shape. At the same time, offshore sandbanks are formed during the winter, which subsequently helps protect the beach by breaking waves off the coast.

Also during the year 2016, towards the end of the spring and during the summer and autumn months, the smaller and calmer waves dominated, the sand slowly turning back to the beach. Beaches and dunes are usually recovered if sediments are not lost at sea.

Author Response File: Author Response.docx

Reviewer 2 Report

See the attached file.

Comments for author File: Comments.pdf

Author Response

ANSWERS TO THE REVIEWERS' COMMENTS

Manuscript ID: jmse_525240   Title: Nearshore wave dynamics at Mangalia Beach simulated by spectral models

GENERAL COMMENTS  

A major revision has been carried out following carefully all the indications, suggestions and observations formulated by the reviewers.

The main changes operated are outlined next:

1) As suggested, an overall revision of the text was carried out in order to correct some small mistakes all along the work. A spell check and editing of the English language and style was also carefully performed. Moreover, many phrases and paragraphs have been reformulated in order to provide a better picture of the objectives and the results of the proposed work.

2) New details have been considered.

3) A new figure (Figure 1) was added.

Finally, the authors would like also to sincerely express their gratitude to the reviewers for their valuable comments that obviously helped in improving the quality of the present work.

At this point, it has to be highlighted that the authors tried in a very honest way to follow all the suggestions and observations formulated by the reviewers and also to operate (as much as it was possible) all the corrections indicated by them. In order to follow the corrections operated in the paper, the option track changes was used.

The specific corrections operated according to the suggestions of the reviewers are given next together with detailed explanations.

Reviewer 2

Comment 1: The abstract is too short and it is not representative of the manuscript content.

Taking into account the observations of the reviewer, the abstract was modified, the current text being as follows:

Abstract: The objective of this paper is to present an integrated picture of the relationship between the waves and the modifications induced by them in the Romanian shoreline. Thus, the hydrodynamic processes at the Mangalia beaches, located in the southern side of the Romanian nearshore, are simulated using the modeling system Mike 21 SW developed by DHI. This is one of the newest spectral wave models, which can be used for regional and local scale simulations. The model has been calibrated and validated using buoy measurements. The analysis of the statistical parameters shows a good match between the model and the observed data. Furthermore, a model to compare the differences that occur between the cold and warm seasons on the beach profiles was developed. The results obtained indicated a reinforce of the coastal erosion in the winter when the waves are stronger (especially in January, February).

Comment 2: The introduction does not introduce the topic, does not create context and background and it contains details of the case study that should be in an other section.

In the section Introduction some changes were made and new information representative for the targeted area were added. Also, a new figure (Figure 1)  of the area was added. 

1. Introduction

Hydrodynamic processes occurring in the marine environment are a consequence of the complex interaction between sea and atmosphere, the waves being generated by the wind interaction with the sea surface [1]. For this reason, the system evolution and phenomena occurring along the coastline are difficult to be simulated with numerical models and a proper implementation of the wave models is very important [2]. Prediction of the sea state conditions in the proximity of the coastline is quite difficult due to various processes of the wave transformation, which are influenced among others by the bathymetry of the area, the location of the hydrotechnical structures and the regime of wind and currents [3].

Numerical models are essential tools to simulate the wave conditions in coastal areas, providing information on the wave climate that are helping the design of the coastal structures, which can protect the coastal areas during the extreme events. During the time, the coastal environments have been often the subject of several spectral wave models for various studies related to wave hindcast or forecast. The model MIKE 21 SW (MIKE 21 Spectral Waves) is one of the newest spectral wave models used to simulate the physical processes related to the generation and transformation of the wind waves [4]. A calibration and validation of the wave model results is necessary to ensure that reliable information is obtained and at this step, some obstacles occur due to difficulties to obtain the field data. The measurements are also important to obtain information regarding the wave climate in a location/area and about the evolution of the coastal erosion phenomenon that intensifies and threatens the inhabited areas.

MIKE 21 SW model has been implemented in different marine environments of the World showing to be an efficient tool. For example, the Indian National Ocean Service Information Center implemented the MIKE21 SW model in the Indian Ocean and it has been used for wave predictions during a cyclone. The comparisons performed against buoy measurements indicate a good agreement between them [5]. The effects of the storm waves on some beaches located in Brazil and Gulf of Mexico have been studied using also the results of the MIKE21 SW model [6, 7]. In another study, the results provided in Persian Gulf by two wave models, namely MIKE21 SW and WAVEWATCH III, were compared. The comparisons have shown that in deep water better results are obtained with WAVEWATCH-III, while in shallow water with MIKE-21 SW [8]. The sea state conditions in the Black Sea basin have been also simulated with MIKE-21 SW model in order to study the spatiotemporal variability of the wave climate over an extended time period covering 37 years [9].

Harvesting the ocean energy represents a promising topic and the ocean energy potential can be assessed based on the results provided by the numerical models that can also provide guidance regarding the best places with wave energy resources. For this purpose, MIKE 21 SW was used to calculate the wave energy potential in the Black Sea [10] and Aegean Sea [11]. The model results in the Black Sea at various locations show that better results can be obtained with MIKE21 SW for the wavelengths than for the wave periods.

The Romanian coastline has a length of about 243 km, in the western part of the Black Sea. The limits are given by the border with Ukraine (in the bay of Musura) to the North, and the border with Bulgaria (near Vama Veche) to the South. At present, the Romanian seaside does not benefit from many field measurements of the wave parameters.

The coastal geomorphology of the Romanian shore can be divided in two large units/areas with different sediment balance that differently reacts to the action of the main environmental factors [12]. Thus, the first unit (the Northern unit) stretches from Musura Bay to Midia Harbor (including the Danube Delta, with the lake complex Razim – Sinoe), it covers about 160 km long and is characterized by low beaches in the delta/lagoon and slope submarine lines. Over the time, the Danube Delta has suffered numerous accretion processes (the Sulina Lob about 12 km wide) and erosion (Delta St. George II and Chilia) [13].

The Southern unit has a different shape, the soft cliffs with small pocket-sized beaches in front, separated by small sandy shores are present. These beaches have steeper slopes than in the Northern unit. The Southern unit is also characterized by various deposits of sediments (where the rocks in the basement appear on the surface of the earth's crust due to erosion), predominantly those of the Sarmatian limestone age. Over these limestones, there are layers of Pleistocene clay covered with deposits of loess and paleosols [14]. These types of sediment lead to the formation of unstable cliffs with deposits of siltic sediments and very fine sands. The beaches in front of the cliffs are rockier with reduced quantities of medium-sized granular sediments. This is due to the construction of the Midia harbor that has modified the sediment transport patterns, which has led to the implementation of some beach nourishment projects [15].

The target area of the present work is a coastal sector near to Mangalia port located in the Southern part of the Romanian seashore. The implementation and calibration of the model MIKE21 SW in the target area represents the main objective of the work. An analysis of the wave climate, wave seasonality and how the sea state conditions affect the Mangalia beach is also carried out, based on the simulation results.

 
Comment 3: The Materials and method section contains information that are very common in the user manual of the most known wave spectral software like SWAN or MIKE 21, this is not of interest for the readers.

The Materials and methods section was shortened, according to the requirements of the reviewer.

2. Materials and Methods

2.1. Target area

Mangalia port, the target area of this study (Figure 1), is extremely important for the national, regional and local economy and is also important from a social perspective, being the major source of jobs in the coastal area. This harbor is located on the Southern part of the Romanian coastline, being protected by dikes built during the period 1970-1980 that they have altered over time the coastal dynamics of the Southern Unit. Coastal alluvial transport direction is to south, but the existence of levees and other coastal development projects induce different local patterns. The construction of the port of Mangalia also affected the southern beaches and cliffs from 2^nd May to Vama Veche (two small villages) [12]. For this case study, where the erosion rate is visible, fact proved by the research carried out for the elaboration of the Coastal Master Plan [12], the MIKE 21 SW was implemented in order to analyze the phenomena occurring in the nearshore.

                 Figure 1.  Google Earth view, Mangalia area

2.2. Wave Model

MIKE 21 SW modeling system [4], developed by the Danish Hydration Institute (DHI), is a numerical tool with capabilities to simulate and analyze the wave climate in offshore and coastal areas. The wave model uses unstructured mesh technique to define the geographical domain.

The governing equation of this model is based on wave action density spectrum, where  is the independent phase parameter (the relative angular frequency), and  is the direction of wave propagation. The action density,  is dependent on energy density [4]

Equation (1)

Mike 21 SW includes two types of formulations: directional decoupling and full spectral. The first formulation is based on the parameterization of the wave action conservation equation and is more appropriate for nearshore processes while the second form is also suitable for the offshore area. Parameterization is performed in a frequency range by inputting the zero moment and the first momentum of the wave action spectrum as parameters. The wave action balance equation formulated in Cartesian or spherical coordinates.

Equation (2)

where  is the action density, t is the time,  is the Cartesian coordinates,  is the propagation velocity of a wave group in the four-dimensional phase space and S is the source term for energy balance equation,  is the four-dimensional differential operator in the and  space.

Fully spectral formulation is also based on the wave action conservation equation, where the waveform spectrum of directional frequency is a dependent variable [16, 17]. For the present study, the full spectral formulation was chosen.

Mike 21 SW module simulates variable flow, taking into account the bathymetry, sources and external forces. Bathymetric data was obtained, retrieved and processed in XYZ format from old bathymetric maps and then it was added in the Mesh Generator module. The bathymetric data resolution was 0.001˚. While the field and boundaries are defined in the Mike Zero Mesh Generator and the interpolation is running, the boundary conditions are excluded. In addition, the points were interpolated by the adjacent value and the value for those points was set.

2.2.1. Unstructured Mesh Approach

The unstructured mesh approach is becoming increasingly important for coastal applications. The coastal areas require a better prediction of the wave characteristics for the safe design of hydrodynamic structures. This approach makes the representation of the offshore, coastal and harbor structures more accurate in the computational domain.

Bathymetric data was obtained, retrieved and processed in XYZ format from old bathymetric maps and then added in the Mesh Generator module. The bathymetric data resolution was 0.001˚. While the field and boundaries are defined in the Mike Zero Mesh Generator and the interpolation is running, the boundary conditions are excluded. In addition, the points were interpolated by the adjacent value and the value for those points was set.

Figure 2a shows the details of this unstructured mesh, which contains the values of the nodes and mesh attributes corresponding to the bathymetric data and boundary points in the Mangalia region. Figure 2b shows the general view of detailed bathymetry in MIKE ZERO View.

Figure 2. (a) Detailed view of bathymetry in the Mangalia area, with the Southern limit of Mangalia harbour; (b) General view of the workspace’s bathymetry.

2.2.2 Initial Conditions

As initial conditions the spectra from empirical formulations are used. The parameters that influence the wave regime in the coastal zone are also considered. Among the parameters available for breaking waves, the breaking parameter,  was specified.  For the mean size of the sediments in the Mangalia area was considered to be D₅₀ = 0.48 mm

The JONSWAP spectrum with the fetch value of 100 km was considered, while the parameters describing the whitecapping (C_dis) and dissipation coefficient (δ) are C_dis = 3.5 and δ = 0.8, respectively.

2.2.3 Boundary Conditions

The boundary conditions for the MIKE 21 SW model are provided by the SWAN (acronym from Simulating Waves Nearshore [18]) model, a state-of-the-art wave model based on the spectrum concept. The wave modelling system based on SWAN model has two computational levels. This wave modelling system was extensively validated by comparisons with various measurements [19, 20]. The first computational level covers the entire Black Sea basin and has a resolution of 0.08 degrees in both longitude and latitude (176 points in the x-direction and 76 points in the y-direction). The second level is focused on the Romanian nearshore and the spatial resolution is 0.02 degrees (101 points) in both directions (see more information in [20, 21]). The computational grids are structured grids and coincide with the bathymetric grids. The SWAN model was forced with the reanalysis wind fields provided by the NCEP-CFSRv2 (U.S. National Centres for Environmental Prediction - Climate Forecast System Reanalysis Version2) data base. These wind fields were used for both levels of the wave modelling system. The simulations with SWAN model were carried out for the entire year 2016. At the second level the values of the wave parameters were provided in order to be used as boundary conditions for the local scale where MIKE 21 SW was implemented and calibrated.

The calibration process was performed using a waveform spectral formula (significant wave height H_s, mean wave period T_m and mean wave direction) for the three open boundaries. The boundary conditions were obtained from SWAN simulations in three points used as limits:

·          P1 - x=28.602 [deg]; y=43.816 [deg]; depth= -11m, North boundary;

·          P2 - x=28.598 [deg]; y=43.801 [deg]; depth= -11.95m, South boundary;

·          P3 - x=28.614 [deg]; y=43.807 [deg]; depth= -15.53m, East boundary.

Figure 3 show the significant wave height_, fields simulated throughout the Black Sea basin and Romanian coastal environment at different time frames.

Figure 3. (a) The significant wave height fields in the Black Sea basin on January 1, 2016; (b) The significant wave height fields in the Romanian coastal environment on January 17, 2016.

In Figure 4 the wave roses in the three locations established as limits are presented. It can be noticed that the Northern boundary (P1) has significantly lower wave heights than the Southern (P2) and the Eastern (P3) boundaries. This can be explained by the existence of protection systems in the area, which have the role to dissipate the wave energy.

Figure 4. Wave rose diagrams for January 2016 at the three points considered: (a) P1, (b) P2, (c) P3.

2.2.3. Outputs Parameters

The output parameters considered are those obtained from the integration of the wave spectrum and can be provided over the computational grid or for specific locations previously defined. The relationships of the parameters are:

1) Significant wave height,

Equation (3)

2) Mean wave period,

Equation (4)

3) Mean wave direction, considering nautical convention

Equation (5)

where,  Equation (6)

The difference between wind-waves/seas and swells can be calculated using either a frequency threshold constant or a dynamic frequency with a higher frequency limit. For the present study, a constant frequency equal to 0.1 Hz was used, as in the case of data recorded from the buoy [4].

Comment 4: The figures are of low quality and don’t communicate primary finding. The resolution is very low.

Thanks for your suggestion. We reviewed the images with low resolution.

Comment 5: The authors present only one year of wave climate analysis, this is not a sufficient time interval to describe wave conditions of the studied area.

 In this work, the simulations were performed only for 2016 because this year was characterized by strong storms that have brought many negative effects on the Romanian seaside. Moreover, for this year measurements were available to validate the simulation results. Subsequently, as the data will be available, we will take a longer-term analysis, but at this time, we wanted to investigate the differences that occur throughout the four seasons in 2016.

Comment 6: The section “discussion” contains comments that lack of any evidence for example: “Due to the Y-shaped dike located between the two beaches, the sediments in the North of the area are maintained when the waves come from the N-E direction. The changes appear when the 288 wave direction is S-E, but still in this case the hydrotechnical(??) works that are near harbor protects a portion of the beach”. Did the authors make field measurements? Did the authors perform a sediment transport analysis?

 Taking into account the reviewer's observations, we have reformulated these paragraphs as follows:

4.Discussion

Taking into consideration that the Black Sea basin has a limited fetch the predominant waves are the seas, with mean periods lower than 10 s [25]. In these conditions, the influence of the local winds is high, even in winter when high wind speeds are over the Black Sea basin and the possibility to find swell is higher. Thus, from Figure 11 it can be observed that Hs variation of the waves coming near to the shore in January 2016 are consistent with the  wind speed variation. The wind speeds are

Figure 11. Wind speed in relation to wave height.

The differences between the waves encountered in cold and warm seasons can also be understood from the calculation of the coefficient of variation. Kabiling et al [22] presents the definition of uncertainty or “coefficient of variation” as:

Equation (12)

where  represents the uncertainty for the wave height,  represents the standard deviation (or ‘error’) and  is the mean value of the  wave height [26].

From the relation (12) we can see a dependence of the direct proportionality between the mean value of the significant wave height () and the standard error for a given coefficient of variation (, so that as the height of the waves increases, its error,, also increase. This also applies to other wave parameters. Errors, therefore, increase with the magnitude of the waveform and decrease as their magnitude decreases.

In the Black Sea, based on the data used in this study, the coefficient of variation is 0.69, which gives us information about the high variability of the Black Sea rather than data errors. In this area, there are large seasonal variations, where the difference between the significant summer and winter wave heights is high and the coefficient of variation tends to be closer to 1.

Subsequently, the spectral model described above was used and some conclusions were drawn regarding the Mangalia shoreline. At the same time, the spatial and temporal changes of the wave characteristics were taken into account. These changes are morphodynamically important.

In winter, the magnitude increases for significant wave height, reaching a maximum of 2.47 m (February 19, 2016), unlike the warm season reaching a maximum of 0.89 m (July 13, 2016). These events prove and reinforce worrying coastal erosion cases in the winter.

Comment 7: The conclusion section is very poor, the statements are obvious and unclear, for example: “the waves that reach the shore decrease in height as they enter the depths of the sea”. It gives to readers only general information, it is not effective.

Thank you for this suggestion. Following this observation formulated by the reviewer, the new conclusions were included in the text:

Based on the results presented in this study it can be concluded that using the Mike 21 SW model, forced with boundary conditions provided by SWAN model, reliable simulations of the wave conditions near to Mangalia harbor have been obtained after the calibration of the model. The wave parameters simulated by Mike 21 SW have been compared with the measurements obtained from MEDA buoy maintained by GeoEcoMar. The statistical parameters show a good agreement between both data.

The Black Sea being an enclosed sea with limited fetches, the predominant type of waves is seas that are affected by the local wind. It was observed that in winter, when the wind speeds are higher and with a high variability of the mean direction, the beach is affected by greater erosion and result in two different beach profiles.

During the winter months (especially in January) the storms are more frequent  and the wave height increases (being influenced by the changes in wind direction and wind speed) to near mid spring. This leads to the formation of beach berms, shaping the beach in a more concave shape. At the same time, offshore sandbanks are formed during the winter, which subsequently helps protect the beach by breaking waves off the coast.

Also during the year 2016, towards the end of the spring and during the summer and autumn months, the smaller and calmer waves dominated, the sand slowly turning back to the beach. Beaches and dunes are usually recovered if sediments are not lost at sea.

The following changes were made regarding list of some of the minor issues:

 Comment 1: line [39] - “acreage” what it means?

 In the 39th line, the term "acreage" was incorrectly wrriten. The correction was operated, the correct word is "accretion", from "accretion processes" meaning the process where beaches show sediment accumulations modifying the sedimentary balance of the beaches.

 Thus, the first unit (the Northern unit) stretches from Musura Bay to Midia Harbor (including the Danube Delta, with the lake complex Razim – Sinoe), it covers about 160 km long and is characterized by low beaches in the delta/lagoon and slope submarine lines. Over the time, the Danube Delta has suffered numerous accretion processes (the Sulina Lob about 12 km wide) and erosion (Delta St. George II and Chilia) [13]

Comment 2: line [41] - “fragments of shells and shells of other molluscs” This sentence is unclear.

We removed this paragraph because the information provided does not add value to the present paper.

Comment 3: line [46] - “Sarmatian limestone” Is it defined by the International Commission on Stratigraphy?

 Sarmatian Stage is major division of Miocene rocks and time (23.7 to 5.3 million years ago). The Sarmatian Stage occurs between the Pontian and Tortonian stages. The lagoonal-type fossil fauna from this stage is represented by peculiar species of clams, gastropods, and bryozoans. Thus, with the passing of time, the sediments originating from this era were formed. To be more specific, we modified “Sarmatian limestone” in “Sarmation limestone age”. (https://www.britannica.com/science/Sarmatian-Stage)

The Southern unit is also characterized by various deposits of sediments (where the rocks in the basement appear on the surface of the earth's crust due to erosion), predominantly those of the Sarmatian limestone age.

Comment 4: line [51] - “sediment transport”. Should be “the sediment transport”.

Thank you for the observation of the reviewer. We made this change.

The beaches in front of the cliffs are rockier with reduced quantities of medium-sized granular sediments. This is due to the construction of the Midia harbor that has modified the sediment transport patterns, which has led to the implementation of some beach nourishment projects [15].

Comment 5: line [115] - the equation (7) is composed of two lines with different character size.

 Thank you for the observation of the reviewer. We made this change.

Comment 6: line [126] - “where the erosion rate is visible”. How the authors prove this statement?

Regarding the details of the visible erosion rates, we did not make a sediment transport or field measurement simulation on morphodynamics, but we used the data described by the Coastal Masterplan (2014). We have added the Coastal Masterplan as a reference.

For this case study, where the erosion rate is visible, fact proved by the research carried out for the elaboration of the Coastal Master Plan [3], the MIKE 21 SW was implemented in order to analyze the phenomena occurring in the nearshore.

Comment 7: line [142] - “It can be clearly seen from this figure that the mesh is much smoother near the shoreline as opposed to the offshore area”. Smoother? maybe smaller than... The triangle sizes abruptly change from large to small this can produce errors in the numerical simulations.

We removed this paragraph because the information provided does not add value to the present paper. The following text was added in the manuscript:

Figure 2a shows the details of this unstructured mesh, which contains the values of the nodes and mesh attributes corresponding to the bathymetric data and boundary points in the Mangalia region. Figure 2b shows the general view of detailed bathymetry in MIKE ZERO View.

Comment 8: line [176] - In the Figure caption are cited P1, P2, P3 points. Where are these points inside the numerical domain?

Information on the location of points P1, P2 and P3 can be found in the next paragraph, representing 3 of the 4 boundaries of the field of study. The fourth limit is the shoreline. The positions of the points are also indicated also in the new Figure 1.

The boundary conditions, used in the simulated model in Mike 21, were obtained from SWAN simulations in three points used as limits:

·          P1 - x=28.602 [deg]; y=43.816 [deg]; depth= -11m, North boundary;

·          P2 - x=28.598 [deg]; y=43.801 [deg]; depth= -11.95m, South boundary;

·          P3 - x=28.614 [deg]; y=43.807 [deg]; depth= -15.53m, East boundary.

 In the revised manuscript also a new figure was added (Figure 1) where the position of the points are indicated.

Comment 9: line [308] - “beach profiles”. These profiles are not defined inside the manuscript, there is no graph that shows their shapes.

 Taking into account the reviewer's observations, we have reformulated these paragraphs as follows:

Based on the results presented in this study it can be concluded that using the Mike 21 SW model, forced with boundary conditions provided by SWAN model, reliable simulations of the wave conditions near to Mangalia harbor have been obtained after the calibration of the model. The wave parameters simulated by Mike 21 SW have been compared with the measurements obtained from MEDA buoy maintained by GeoEcoMar. The statistical parameters show a good agreement between both data.

The Black Sea being an enclosed sea with limited fetches, the predominant type of waves is seas that are affected by the local wind. It was observed that in winter, when the wind speeds are higher and with a high variability of the mean direction, the beach is affected by greater erosion and result in two different beach profiles.

During the winter months (especially in January) the storms are more frequent  and the wave height increases (being influenced by the changes in wind direction and wind speed) to near mid spring. This leads to the formation of beach berms, shaping the beach in a more concave shape. At the same time, offshore sandbanks are formed during the winter, which subsequently helps protect the beach by breaking waves off the coast.

Also during the year 2016, towards the end of the spring and during the summer and autumn months, the smaller and calmer waves dominated, the sand slowly turning back to the beach. Beaches and dunes are usually recovered if sediments are not lost at sea.

Author Response File: Author Response.pdf

Round  2

Reviewer 2 Report

See attached file.

Comments for author File: Comments.pdf

Author Response

ANSWERS TO THE REVIEWERS' COMMENTS

 Manuscript ID: jmse_525240  Title: Nearshore wave dynamics at Mangalia Beach simulated by spectral models

 GENERAL COMMENTS  

A minor revision has been carried out following carefully all the indications, suggestions and observations formulated by the reviewer.

 The main changes operated are outlined next:

 1) New details have been considered.

 2) A new figure (Figure 11 a and b) was added.

 3) New references were added.

Finally, the authors would like also to sincerely express their gratitude to the reviewers for their valuable comments that obviously helped in improving the quality of the present work.

At this point, it has to be highlighted that the authors tried in a very honest way to follow all the suggestions and observations formulated by the reviewers and also to operate (as much as it was possible) all the corrections indicated by them. In order to follow the corrections operated in the paper, the option track changes was used.

The specific corrections operated according to the suggestions of the reviewers are given next together with detailed explanations.

Comment 1: The paper is centered on numerical simulations in a small area of the Black Sea. The simulations were performed for only one year (2016). Due to the quickly varying nature of the coastline, the authors with only one year cannot conclude anything about coastal erosion. They proved only that the adopted model correctly works. After a validation performed using one year, the authors should simulate at least 10 years of wave parameters. Furthermore, the authors should perform a diachronic analysis of shoreline and/or perform a numerical simulation of the local solid transport. The readers could be interested to generalize the given results but one year in a small region is not sufficient time.

The results presented in this paper represents only the first implementation of the MIKE21 SW model in a coastal area of the Black Sea, near to the port of Mangalia. For this reason the main objective was to calibrate and validate the wave model. Moreover, we wanted to show that the SWAN + MIKE21 SW system can be applied with good results to simulate the wave conditions at a local scale of the Black Sea. Unfortunately, only few measurements were available! Close to ports there are hydrotechnical structures that alter the dynamics of the coastal area. This required a thorough analysis to highlight the main changes that occur in port areas. After this first step, of course, our attention will be focused to simulate the sea state conditions in the target area over an extended time period (minim 10 years) in order to evaluate the wave climate and the changes occurred. 

Comment 2: It is well known, from the scientific literature, that exists a winter beach profile and a summer beach profile. Thus to affirm this circumstance is trivial.

Following the observation formulated by the reviewer, the following text was changed in accordance with the requirements:

The Black Sea being an enclosed sea with limited fetches, the predominant type of waves is seas that are affected by the local wind. It was observed that in winter, when the wind speeds are higher and with a high variability of the mean direction, the beach is affected by greater erosion.

Comment 3: The authors didn’t provide a figure with the numerical results of MIKE 21 SW.

Thanks for your suggestion. A new figure (Figure 11 a and b) was added.

Comment 4: The authors cite literature about wave energy harvesting but didn’t cite a sufficient number of it, as an example they should cite: - Goncalves, Marta, Paulo Martinho, and C. Guedes Soares. ”Wave energy conditions in the western French coast.” Renewable Energy 62 (2014): 155-163. - Wave Energy Assessment around the Aegadian Islands (Sicily) C Lo Re, G Manno, G Ciraolo, G Besio - Energies, 2019

Taking into account the reviewer's observation, the articles suggested were cited in the manuscript and included in the references list [10, 11].

Harvesting the ocean energy represents a promising topic and the ocean energy potential can be assessed based on the results provided by the numerical models that can also provide guidance regarding the best places with wave energy resource.[10,11]

Author Response File: Author Response.docx

Round  3

Reviewer 2 Report

The authors addressed all the reviewer comments and significantly improved the overall quality of their manuscript. I suggest a double check in order to avoid typos. 

Line 337 page 14, please correct "Goncalves" with "Gonçalves".

Line 339 page 14please correct "Dicily" with "Sicily".

The paper can be accepted in the present form.