This paper investigates seasonal patterns of water particles accumulation in Pearl River Estuary subregions following a Lagrangian approach. The authors determined flow fields using an Eulerian model and then monitored particles trajectories for 30 days in the summer and 30 days in the winter by implementing a3D Lagrangian model. Their main conclusion is that plume fronts and velocity convergence are important factors in the determination of water accumulation dynamics while they also identify a negative correlation between divergence and accumulation probability. I reckon that the manuscript contains work that is worthy to be published but it also needs considerable improvements before this happens. Main concerns have to do with phrasing, figures quality and inadequate information supplied. I provide the following general and specific comments which hopefully the authors will find useful and constructive. 

General comments

1. The manuscript needs a thorough revision in terms of language. Specifically, the use of present tense throughout the manuscript is recommended instead of past tense. 
2. A more detailed description of the PRE dynamics is demanded so that the reader can appreciate the unique and complex physical processes in this ecosystem. For instance, the fact that the river discharge in the PRE is higher in the summer and lower in the winter is counterintuitive as in most systems, discharges are higher during winter due to increased rainfall and runoff. It is therefore recommended to the authors to mention this specifically in their introduction. In addition, the monsoonal cycles need to be mentioned as they are important for the water circulation. Relevant information could be included in a separate paragraph either at the end of the Introduction or at the start of section 2.
3. A deeper insight is demanded when discussing the results, particularly on the inter-play among the three basic physical mechanisms of gravitational circulation, tidal and shelf currents in the estuary. It is generally not a good practice to merge results with discussion. It is recommended to detach discussion from section 3 and put it in a separate paragraph. In this way, the core messages of this work can be better illustrated.
4. The results analysis and discussion focus mainly on the influence of river discharge, tide, and plume fronts. But the monsoonal cycles and consequently wind forcing, downwelling, upwelling etc. are also important parameters that affect water circulation and accumulation. These are only briefly mentioned in page 6 (lines 140-142) and somehow their effect is understated in the manuscript. The contribution of the monsoonal climate can be reevaluated and better highlighted. 
5. FontSizes and Figures sizes need to be enlarged.

Specific comments

Introduction:

Lines 43-46. It would be useful to expand and elaborate on why it is important to know and identify areas of water accumulation. Is this interesting only for the PRE or for any estuary? Is eutrophication the only problem or is it particularly concerning for the PRE? Note also that eutrophication is not an anthropogenic activity but one of its adverse consequences.

The introduction can be enriched with some further literature on the topic. Is this a topic that has been explored before and to what extent?

Methods

Model settings are either missing or insufficiently presented. Where are the ROMS model’s boundaries set? Which area do they cover? A figure with the model grid and bathymetry should be added. How much is the river discharge in the summer and how much in the winter? Please provide a sufficient summary of the model setup.

Line 78-79 ‘Data of shelf currents were obtained by the coarser – resolution simulation’. Is there a coarser model also used? Please explain.

Line 88 Has the model been validated?

Line 90 Please state the implemented diffusion coefficients.

Line 100 It is understood from Figure 3 that the calculations are in 3D, what is the vertical dimension of the grids?

Results and Discussion

Line 145-146 Hypoxia can be common in microtidal estuaries such as the PRE or the Mississippi (see for example Schiller et al.2011) and is usually associated with stratification induced by river flows in the absence of significant mixing. It would be interesting to see the density at the surface in Figure 6 for both summer and winter in addition to the bottom ones. I presume this would show strong stratification during the summer.

Line 148 and 200-212. The role of the gravitational circulation is a bit obscure from the figures. A comparison between Figure 3 b and d shows that the flow arrows at the bottom do not differ much. In fact, the flow field is very similar. In addition, the density field at the bottom is also very similar between summer and winter if we compare Figure 6 a and b. This is not unusual. See for example the paper by Wong et al. (2003) which also talks about the PRE. They mention that the bottom salinity front between summer and winter is similar which basically agrees with what we see in Figure 6. Figure 3d shows an increase of the accumulation in the upper estuary in the winter compared to the summer (Figure 3b). Regarding the similar density fronts, how certain is it that the gravitational circulation is responsible for this increase? Figure 5b indicates that the increase of accumulation in the upper estuary is caused by convergence (difference between Figure 5 a and b). Could this be related to the decrease of river discharges? 

Line 164 It would be helpful to add in Figure 4 a picture of the original (starting) distribution of the particles.

Line 180 Do u and v indicate mean over depth velocity or at the bottom layer?

Line 229 How is the negative anomaly defined?

Role of Tide and River

The main message I get from Figure 7 and 8 is a landward displacement of water accumulation when the tides are removed, and the discharge is reduced respectively. What is the cause for this? It is not very clear from the text.

Line 229 How is the negative anomaly defined?

Conclusions

The conclusions read more like a summary. It is suggested to rewrite it so that the core messages of this work are better highlighted.

Figures

Figure 1 Put panel names a and b on the left and right image. Change the colour of 50m isobath to something more visible.

Figure 3. Use a different colorscale for the vectors (e.g., blue) , differences can hardly be spotted with this one. Add an arrow size. Same for the colour of the areas names. Why do you show results at the surface only in this occasion and nowhere else? Is you research focusing on the bottom or the entire water column?

Figure 5. Keep the y axis scale equal between a and b for better comparison.

Figure 6. Add results for the surface layer.

Figure 7. Are these results given for the entire water column or just the bottom? Make y axis equal between c and d.

Figure 8. Same comments as in Figure 7.

This study uses Lagrangian particle tracking and Markov Chains to study the accumulation dynamics of water or passive/conservative mass in the PRE and connectivity among its subregions. Convergence and fronts were identified as major factors for high accumulation probability. The study is based on a validated model as the authors stated, yet more details on the model set up should be provided, particularly the sensitivity runs on modifying tide and river discharges. The identified seasonal accumulation dynamics in different subregions and their connectivity are interesting, yet more implications on the ecosystem in the PRE can be discussed. For example, based on the sensitivity run of reducing river discharge, should we worry about more pollutants will be accumulated in some regions in dry years? I also suggest the authors to carefully check the grammar and revise their writings to improve the clarity and readability of this manuscript. In addition, I am not familiar with the calculation of Markov Chains and I except other reviewers and the topic editors would have better judgements. Please find specific comments below.

Title: suggest revising it to specifically indicate that this study is about the accumulation dynamics of water (or passive material). Add ‘a’ before ‘Lagrangian Perspective’.

Abstract and Conclusion need to be more concise. These long paragraphs will distract readers from the main takeaways. It is also necessary to mention that results in this study are based on a model with climatological forcings and not specific to a certain year that may have large variations in the hydrodynamics.

Introduction: mass accumulation dynamics in different sub-regions in the PRE is an important part of the current study. If available, please supplement more previous findings on the different material accumulation features in these regions (e.g., total organic carbon, pollutants, suspended sediments, or other water quality parameter), to emphasize why we should care about dividing the PRE into the six subregions. Statements in lines 145-146 could fit in this part. It also needs to mention and explain why this study used a model producing climatological hydrodynamics rather than focusing on years with realistic forcings.

Line 13: ‘plume fronts’ only occurred in the abstract. Please use a consistent description on fronts with the main text.

Line 65: is the right panel of Figure 1 the model domain? If so, clearly state it in the caption. If not, please add a figure showing the model domain.

Line 77: climate --> climatology. Which period did the climatology forcing represent for?

Lines 78-79: please supplement more detailed descriptions of the coarse resolution model or add related references.

Line 79: clarify ‘statistics of atmosphere forcing…’?

Line 80: what is the accuracy of the monthly river discharge rate by ECMWF in this region? Why not using the locally observed river discharge?

Line 87: what does ‘other complicated hydrodynamic processes’ imply? What is ‘computations’? Computations of particle trajectory? Also, please add references to support the statement: ‘The results of the computations…in the estuary-shelf system’.

Line 90: clarify ‘different diffusivity coefficient’ in the vertical and horizontal turbulence.

Lines 91-93: the descriptions on particle tracking experiments are unclear. Were the 8386 particles released each day in January and July? How was the ‘uniform release’ achieved, i.e., what are the vertical and horizontal interval of each particle? How will the particle behave when they reach the boundaries, including open, land, surface, and bottom boundaries? time step of particle tracking? output frequency of the history files that were used to drive the particle tracking and the output frequency of particle location?

Line 105: it should be n_i^t0

Line 106: it is p^t in equation (1). Please use the same symbol.

Line 121: ‘transport accumulation’ seems to be an odd expression.

Lines 123-124: did particles come back after moving out of the seaside boundary?

Line 135: particles were only tracked for 30 days as stated in section 2.1. Why would the maximum day in x-axis 60 days?

Line 148: What leads to the further landward intrusion of bottom water in winter?

Line 155. Does probability have a unit? percentage? It is not easy to see the difference in the gray color for velocity arrows. Suggest using arrow length to represent the magnitude of velocity. In addition, there are some unnecessary gray dividing lines in Figure 3 and figures below. Please remove them.

Line 182: the accumulation probability was averaged over each sub-domain and not differentiate between surface and bottom in Figure 5? The div(V) in Figure 5c-d shows surface or bottom or depth-averaged results?

Line 215: the unit of density front should be kg m^-4 from the equation in line 205.

Section 3.3: please provide more details on the testing cases (or put them in the supplementary). For example, which options or files in roms were removed for the case removing tidal currents. How many rivers were included in the model domain? A time series of river discharge in base case and test case will be useful information.

Line 244: if the particle tracking was conducted for a longer time, will MO accumulate water from other regions?

Line 245: the arrows in Figure 7c-d seems to be confusing. Not every regions were noted by arrows? What does different arrow directions represent?

Line 281: suggest adding some descriptions on wind strength and magnitude of river discharge in the main text.

Technical corrections:

Line 13: add ‘of particles’ or other appropriate words after ‘bottom-layer accumulations’.

Line 15: add ‘a’ after ‘there is’

Line 16: local --> locally

Line 30: add ‘the’ after ‘Located in’

Line 35: change ‘Such as’ to ‘For example,’

Line 43: were --> are

Line 46: stronger --> strong

Line 51: remained --> remain

Line 60: explored --> explore

Line 159: subdivide --> divide

Line 164: where --> which

Line 203: accumulating --> accumulate; add ‘the’ between ‘in middle’

Line 219: contribution --> contributions