Temporal Variability of Sediments, Dissolved Solids and Dissolved Organic Matter Fluxes in the Congo River at Brazzaville/Kinshasa

Round 1

Reviewer 1 Report

The authors adequately adressed all comments of reviewers. The paper is suitable for publication.

I recommend English style and grammar polishing which can be performed at the MDPI office and constitutes the part of the APC.

Author Response

Please see the attachment, thank you

Author Response File: Author Response.pdf

Reviewer 2 Report

Reviewed version of MS by Moukandi N’kaya  is an improved version of a previous submission.

This study explores changes in DOC and TDDs in a sampling site along the Congo river during two monitoring long term study periods.

Although this MS is a simple description of study case, under my perspective, it is necessary to publish finding/observations from little studied sites especially ones from very important but still little known ecosystems such as the Congo river. Moreover is essential to make visible these studies and peoples that performed these efforts.

This revised version is almost ready for publication yet I found some points that need to be improved before the final acceptation.

• The title.

In the title are mentioned “sedimentary and biogeochemical fluxes”. It exists a growing interest about DOC in inland waters. To improve visibility of the present work it would be appropriate to mention DOC in the title. My suggestion is :  Temporal variability of sediments, dissolved solids and dissolved organic matter fluxes ….”

• At section “2. Sampling and analytical measurements” a description of protocol used to quantify TSS is missing. A short paragraph about this protocol is necessary.
• Figure 7. These plots need to be improved. My suggestion is to fit the “y” axes to the measured values in order to have a better visual description of the hysteresis curves. For instance the plot describing “TSS monthly mean    Discharge ”  the “y” axes should be limited between 15 and 35 aprox.
• Lines 388-390. Which “general model” do you refers”?..here some citation is necessary. See for instance Butturini, A., Alvarez, M., Bernal, S., Vazquez, E., & Sabater, F. (2008). Diversity and temporal sequences of forms of DOC and NO3‐discharge responses in an intermittent stream: Predictable or random succession?. Journal of Geophysical Research: Biogeosciences, 113(G3).
• Line 399. In my opinion this sentence is important. Could you provide an explanation about changes in rotational pattern of DOC between the two study periods?

All my best regards

Author Response

Please see the attachment, thank you

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Manuscript ID: geosciences-735323
Type of manuscript: Article
Title: Variability of sedimentary and biogeochemical fluxes in the Congo
River over the past 30 years
Authors: Guy Dieudonne Moukandi N'kaya *, Didier Orange, Sandra Murielle
Bayonne Padou, Alain Laraque

This study describes an interesting long term monitoring at the Congo river. Being conscious that the MS need a profound and very exhaustive revision and more depth data analysis, I believe that results might of great interest for hydrologists and biogeochemists (this river is one of the largest and flowing rivers in the world). In consequence in my opinion is a priority to disseminate this information as long as the authors provide a deep reorganization of their study.

In my opinion the MS needs a reorganization of its structure.

• To explain in more details the objective (end eventual hypothesis) of the study at the introduction.
• To describe in the M&M section all methods used in the research and remove these details from Results section.
• Remove from Result section all discussion sentences and just focus on their own results.

Regarding Results, I would suggest to analyze in mode details the C vs. hyteresis loops. In my opinion these are an interesting aspect the needs to be improved providing information about hysteresis of electrical conductivity..then compare these data with that of TSS, TDS and DOC.

I add all my comments and suggestion in the pdf in the form of cartoons.

If authors are open to improve their MS I am available to re-revise the text.

Comments for author File: Comments.pdf

Reviewer 2 Report

Review of Moukandi N'kaya et al. manuscript : " Variability of sedimentary and biogeochemical fluxes in the Congo River over the past 30 years” by reviewer 1.

This paper, by Moukandi N'kaya et al., focuses onto the variability of the discharge, the suspended matter, the total dissolved solids and the dissolved organic carbon exported along two periods (1987-1993 and 2006-2017) by the Congo River. The objective of this study is to evaluate any changes of the river delivery fluxes and dynamic recorded between these two periods. The inedited databases used in the study were produced by two monitoring programs PEGI (1987-1993) and SO HYBAM (2006-2017). This is an impressive database and the compilation and the extraction of this information is important for the geoscientists’ community especially for this major river at the global scale. Important results are presented. In particular, the Congo’s riverine water and matter annual fluxes are actualised and some differences are identified between the two periods both in term of net fluxes estimates and exportation dynamics. In this sense this data and work need to be communicate. However, I think that the organisation of the paper and part of the results/discussion need to be improved considering my main comments reported thereafter. In general, it is not easy to follow the main “home messages” of the study. For these reasons I recommend the paper for publication with major revision.

Major comments:
- My main major comment is that some differences are considered to be identified between the two periods but the explanations are not convincing. In particular, depending on the part of the manuscript TSS and TDS fluxes and concentration are considered to be different or not during the two periods. For example, at the end of the abstract, there is written that the solid fluxes and TDS fluxes did not radically changed between the two periods because of no high differences in discharge. However, earlier in the abstract and in the results part (e.g. table 1 and text related) differences in TDS concentration, conductivity and DOC concentration are reported. How can we explain no changes in fluxes when discharge is almost homogenous between the two periods and the concentrations are different? Moreover, in the conclusion (l. 414-417), differences in water discharge, TDS and TSS concentration are presented. A dilution effect is invoked to give an explanation but no clear demonstration is performed in this sense to validate this explanation. In general, this aspect, which is one of the scope of the paper, need to be reworked and more clearly exposed.
- Calculation: In most parts of the manuscript a comparison are done between two average values. First, as pH is a log scale value, it is not correct to calculate an average (give the mediana). Second, give the standard deviation of the average values to explore if the difference is significant.
It would be also important to associate to the annual flux a standard deviation of the interannual flux average to qualify if the differences observed between the two periods are significantly different or not (e.g. l. 256-261). For example is the difference of TSS fluxes of 30.2 and 33.6 x 106 t/yr between the two periods can already be considered as an increase considering standard deviation?
- Organisation: The paper mainly focused on observations and the discussion is not convincing because it does not give clear explanation. For example, one of the conclusion is that the “increase in flow rates led to a decrease in TDS concentrations (-16%) through a dilution effect. Conversely, TSS concentrations increased (+7%) because a higher flow favours their transport. It also allows more intense leaching and infusing of the Cuvette Centrale, the main producer of organic matter in this vast basin the low increase”. The decrease in flow rate is around 4% and, in the present version of the paper, there is no demonstration of the link between discharge evolution and matters concentration evolution. Moreover, the effect of the “cuvette Centrale” is not demonstrated. This last point could be (perahps) performed based on the evolution of rainfall distribution evolution throughout the basin. Another alternative would be to better explore the concentration-dischage relationship along the two periods (based on all the points and not only on interannual monthly averages).
More generally, there is a lot of results and I think that it is difficult to follow the logic of the text. I recommend to organise the results and the discussion according to the objectives of the paper, in the present form, the manuscript is confusing. I also think that part of the discussion part is already a “result” part. From my point of view, the discussion need to be more precise and more argued in the explanation of the mechanisms. The discussion part could be divided in sub-section to clearly structure the implication of the results and more references to current knowledge (based on current litterature).
In more details, discussion need to be more specified, reworked and better argued. For example, l. 343 – ‘…due to the recent human land-use…’ please give more precision and references. L. 350-352 : ok, what is the mechanism involved in this process? So a good relationship would therefore appear between DOC and pH. Is it the case (a figure DOC vs pH would be hepful)? L. 360-363 : “Moreover we observed that the hysteresis curve has been largely changed, even inversed during the recent period (Figure 6). If we consider a still slow evolution of human activities inside the Congo River Basin (CRB), all these clues bring to hypothesis that the slight increase of discharge have seriously affected recent DOC change in the Congo River’s fluxes.” I do not understand this last sentence and its relationship with the previous one. Moreover, I do not identify where this has been demonstrated in the manuscript. In particular a more specific analysis of the hydrology (and perhaps rainfall distribution?) could help to support this point. L. 373-376 ‘contrarily, by dilution effect, the TDS fluxes of the SOH period are lower (37 x106 t yr-1) than those of the PEGI period (42 x106 t yr-1).’ There is no evidence that this is due to dilution effect. L. 403 : is it significant? L. 433-434 : ‘The variations in flow rates between the two periods appear to be the predominant factors of the changes in concentrations and fluxes of the various studied components (TSS, TDS, DOC).’ The current results and discussion sections do not evidenced this.
- An important aspect which, from my point of view, need to be reworked, is the exploration of the concentration-discharge relationships. In part 3.3, this is slightly reported, but these relationships could be better explored particularly to support the affirmation that (l.414-417) the discharge variation control the differences of average TSS, DOC and TDS concentration observed (from my point of view and according to the data available it is not the case, but perhaps I’m wrong). More generally important information can be extracted from these relationships, in term of behavior and evolution of the matter export dynamic. Numerous papers treat this approach (including in large rivers) and can help the authors to interpret the behavior explored.
- The databases presented here (links for downloading or detailed data in supplementary file for example) need to be available to the reader.
- English need to be reworked.

Moderate comments:
Introduction : there is a synthesis of the history of the studies performed in the basin but from my point of view, a synthesis of their main finding (in link with the objectives of the present study) is missing. Moreover, It would be interesting, considering the scope of the present study, to also present climate changes and anthropogenic activities evolution along the last 30 years based on the litterature. The problematic could be better presented.
l. 208-211 : “the morphology of the river bed with the islands of sand seems has not changed during these last 40 years (no extension of islands and no reduction of anastamosed channels).” Please give a reference or present data (remote sensing data in supplementary material) to support this affirmation.
l. 210-211 : ‘The measured TSS concentration between the two periods has slightly increased from 25.3 to 27.2 mg L-1.’ Is this difference really significant considering the standard deviation and the weighted discharge TSS concentration mean?
Figure 4 : add all the data in background to visualise the variability of the concentration by comparison with the discharge. At least add the standard deviation on the graph to validate (or unvalidate) the fact that the river respond to an hystereis behavior for these parameters.
l. 262-264 : This signifies that the concentrations are almost constant by comparison with the discharge relative variability, so the discharge variability control the flux variability. This is an important point to discuss (which is not common for TSS for most rivers).
l. 305-306 and figure 8 : According to the figure 8 a significant relationship between surface TSS and section TSS can be extracted. It is important to also calculate an uncertainty from this relationship and to propagate it in the estimate of the TSS annual flux. This point is important for example to estimates if differences of the annual TSS flux between the two periods is significant or not.
l. 311 : refer to a citation or report the data table (potentially in a supplementary file).
l. 324-326 : “… This increase appears to be due to the increase in flows (> 4%), from 38,080 to of 39,660 m3 s-1 inter-annual average for respectively the two period studied.” No demonstration has been performed to affirm this. I recommend to develop this point.
l. 332. I recommend here to add a table or a figure to report previous values calculated in the Congo. I also recommend to compare these results with values recorded in the other large rivers in the world. Are these values high of low by comparison with the mean continental inputs to the oceans? Why?
l. 335-339 : this is more a result part than a discussion.
l. 379-383 : If this point is discussed, a section need to be added in the results part about the sources discrimination of the TDS solutes and this comment would need to be extended and calculated.
l. 384 : no hysteresis is observed for TSS and the significance of the hysteresis behaviors need to be demonstrated in the results part.
l. 392-395 : At least indication of this affirmation need to be added in the figure 8. Moreover, I do not understand why this part is in the discussion section, it would be more a result section.
l. 408-411 : This information has not been discussed in the paper, I recommend to eliminate it (moreover, l. 408, for which period?).
l. 413 : Is it significant?
L414-417 : In the results and discussion part this has not been clearly evidenced.
l. 442-443 : This is interesting but this is part of the results and it is not really a discussion/conclusion aspect.
What is the sense and the objective of the presentation of the DOC/TDM? There is not a lot of description of this index.
What is the relationship between TDS and conductivity along the two periods?

Minor comments :
l.14-15 “For 3 decades, the solid and dissolved fluxes of the Congo River have been regularly
monitored on a monthly basis, despite 12 years of deficiencies.” I recommend to reformulate. Suggestion : “Along the 3 last decades monthly solid and dissolved fluxes of the Congo River have been monitored for two periods : 1987-1993 and 2006-2017…” It is just a suggestion.
l. 20. As pH is a log scale value, it is not correct to calculate an average (give the mediana).
l. 21 and along the text, give the standard deviation of the average values.
l. 25-26 ‘chemical weathering still predominate over the solid fluxes’ replace by ‘…slightly predominate…’ Are these values really different? By comparison with other context I would qualify these values as ‘similar’ i.e. the chemical weathering is equivalent to the physical erosion flux. It is just my opinion.
l. 27-29 in the text and in the results the opposite is observed for COD or TDS for example.
l. 37-39 : be more specific. Be careful with the references links ‘[Error! Reference source not found. and Error! Reference source not found.]’.
l. 46 and 48: details acronyms.
l. 56 : what “river health” means? It is a term too generic. Be more specific.
l. 60 the reference [23] does not refer to the Amazon. Please check the references here and after.
l. 72 Large tropical rivers in general, no?
l.83 what is a ‘seasonal variation coefficient’ (how it is calculated)?
l. 84 : replace ‘…’ by ‘e.g.’ before the river names.
l.85-86, before the hydrological regime I suggest to add a sentence about rainfall regime and the rainfall distribution throughout the basin.
l.92 : While it is important to cite ‘Hughes et al., 2001’, this reference do not refer to a geological description. Please adapt the citation.
l. 150 “of the central …”
l.183-186 : The descriptive parameters of the water quality are quite similar, just we notice more acidic waters (from 7 to 6.8) with a weaker conductivity (from 36.6 µS cm-1 to 28.4 µS cm-1) and a higher annual concentration in DOC for the recent period.” So the descriptive parameters of the water quality are not “quite similar”.
l.186 ‘DOC has increased from 9.9 mg L-1 to 12.7 mg L-1, an increase of 28%.’ Add the standard deviation.
Figure 2 : I personally think that this figure is not very informative but it is just an opinion.
L.207, 229, 336, 343 and probably in other parts, the sentence is beginning by ‘And…’ this is not correct. Please check all the occurrences along the manuscript.
l. 212 what is the difference between dissolved concentration and TDS?
l. 216 What is ‘variation coefficient’? Please give the formula.
Figure 6 : Put in the same graphics the Pegi and the HYBAM period data to allow a direct comparison between these two periods (with standard deviation and, in background, the individual data).
Figure 7 : please add the depth of the river bottom.
l. 323-324. Is this increase significant?
l.328 : I would replace “specific erosion” by “specific suspended particulate matter flux”. It is just a suggestion.
l. 371 : ‘is marked by 4% increase’ please qualify this increase (significant? Low? High?)
‘over the last decades’ please replace by ‘ along the two studied periods’.

Reviewer 3 Report

General Comments

This article investigates the “Variability of sedimentary and biogeochemical fluxes in the Congo River over the past 30 years”. The subject treated is pertinent for publication since it deals with a very important topic about water resources addressing both the quantity and quality of river water.

In general, the article is good. However, several parts of the article need to be improved. In particular, the introduction needs to be organized in a more appropriate sequence. The importance and justification of conducting this research must be clear, including in the article itself. The goal should be the last part of the introduction and it must come before the scientific question and the hypothesis. Part of the paper's hypothesis is how climate change impacts water quality. I recommend removing this part of the hypothesis, because it was not approached, this issue was not discussed throughout the article or else if it is maintained, obligatorily, the article needs to be reformulated considering this issue. The methodology in turn is missing a lot of pertinent information that makes it possible to reproduce it. Add a table in the methodology with all the variables analyzed. Need to highlight which research contributions and what are the new aspects of this research compared to similar research cited in the first column, in the second, the methodology used to perform the analysis and in the third column, the reference of the method used for the two periods. Further detail the part of the water sample collections and measurements in the field. Still in the methodology, add precipitation data for all periods of the research, because it is important information to discuss the flow rates. Add the altitude map of the watershed under study. Add the soil map of the watershed under study. The authors discuss the results by citing this information that was not presented in the article and can enrich the discussion of the article. I recommend adding information about the main economic activities developed along the watershed, as well as population, city and sewer data. Another weakness of the article is the statistical part. In the descriptive analysis, it is necessary to do the data normality tests. If the data show a normal distribution, apply an average comparison test for parametric data. Otherwise, if the data are not normal, apply some average comparison test for non-parametric data. In the results, I recommend combining the information from tables 1 and 2, excluding similar information, leaving the manuscript more concise with just one table. In the discussion, needs a deepening of the authors' contribution, need to better explain why the results and the differences as well as the work contribution in the scientific environment. The conclusions need to be entirely reformulated, as the authors present part of the results and discussion. The conclusions need to be more objective and clearer. The conclusions need to respond clearly to the objectives proposed in the article. Therefore, the authors need to be clearer on what are the new aspects of this research. Unfortunately, I feel that the Introduction and Discussion do not adequately highlight / emphasize the new aspects of this research. 

Specific Comments

Title:

L3: Exclude from the title “in the last 30 years”, because throughout the text, data referring to only 17 years are presented (1987-1993 e 2006 a 2017). Therefore, I strongly recommend rewriting the title, because it gives the wrong idea that it is 30 years of monitoring.

Abstract:

L14: Start the abstract with a short introductory sentence.L14: The authors start talking about the last 3 decades of monitoring, however, data related to 17 years are presented throughout the text (1987-1993 e 2006 a 2017). Therefore, I strongly recommend rewriting, because it gives the wrong idea that it is 30 years of monitoring.L15: Cite the period with non-collected deficit data.

L16: When an acronym is cited for the first time, it must be preceded by its meaning, with the initial letters capitalized. Add meaning before acronyms “PEGI/GBF” and “SO HYBAM”.

L23: Add the reasons that led to the river discharge increase by 5%.

L23: When an acronym is cited for the first time, it must be preceded by its meaning, with the initial letters capitalized. Add meaning before acronym “TDS”.

L24: When an acronym is cited for the first time, it must be preceded by its meaning, with the initial letters capitalized. Add meaning before acronym “TSS”.

L24: It is more appropriate to replace the word "programs" with "period".

Introduction

L35: When an acronym is cited for the first time, it must be preceded by its meaning, with the initial letters capitalized. Add the meaning before acronym "ITCZ".

L38: Correct: Error! Reference source not found.

L39: Correct: Error! Reference source not found.

L46: When an acronym is cited for the first time, it must be preceded by its meaning, with the initial letters capitalized. Add the meaning before acronyms “ORSTOM” and “IRD”.

L48: When an acronym is cited for the first time, it must be preceded by its meaning, with the initial letters capitalized. Add the meaning before acronyms “PIRAT and PEGI”.

L54: When an acronym is cited for the first time, it must be preceded by its meaning, with the initial letters capitalized. Add the meaning before acronym “SOH”.

L56: Rewrite: in the last 30 years. Because throughout the text, data are presented for 17 years of monitoring (1987-1993 and 2006 to 2017). Therefore, I strongly recommend rewriting, because it gives the false idea that it is 30 years of monitoring.

Materials and Methods

L82: Right after the word “equator” refer to figure 1a.L86: Before referring to figure 1c, it is mandatory to refer to figure 1a and 1b, or else change the sequence of figure 1.

L86: Before referring to figure 1c, it is mandatory to refer to figure 1a and 1b, or else change the sequence of figure 1.

L144: Figure 1 must be moved to the previous page (page 3), the figure must be inserted right after its first citation in the text.

L144: Figure 1 needs several changes. Figure 1a: insert the north; change the color of the "Rain forest" to a more visible color; Highlight the course of the Congo River, from its source to its mouth and insert in the caption; Highlight the courses affluent to the Congo River and insert in the caption; Insert the altitude information for watershed of the congo river in Figure 1a or add that information to another figure throughout the material and methods. Figure 1b: you must insert the legend, including the identification of the data collection stations; Insert the geographic coordinates. Figure 1c: increase the font size of the figure (x-axis, y-axis and legend). Figure 1d: increase the font size of the figure (x-axis and y-axis); insert caption and in the description of the figure it describes the period from 1903-2017, however, in the figure it appears only until 2013.

Results

L192 e 220: Apply some statistical test of average comparison.

L202: Explain why the DOC was increased.L212: What statistical test was applied to confirm that the change is significant? It needs to be based on statistical analysis to find out if there was a significant difference.

L234: Figure 3, add the standard deviations for each variable.

L238: Explain why TDS concentrations are lower.

L247: Figure 4, add a bar with standard deviations for each variable.

L253: Refer to the figure and explain why the DOC was increased.

L265: Similar, based on which statistical test? It needs to be based on statistical analysis to see if they are similar.

L296: Figure 6, leave the scale equal on the X axis.

L317: Figure 7, the names of the months of January, February, March and April have been cut out.

L320: Figure 8, separate the database to develop the equation, around 70% of the data and for model validation leave 30%; compare the estimated data using the model and the measured data. It is necessary to do the model significance test and calculate statistical indicators to assess model performance such as: correlation coefficient; determination coefficient; F test; Nash-Sutcliffe coefficient; mean absolute error and root mean square error. As the model's performance was not evaluated, we do not know if the model is adequate and reliable. We don't know the estimation error if we use the model.

L333 e 334: The dissolved material had an opposite and significant change. Explain why it had the opposite change. Which statistical test was applied to confirm that the difference is significant? It needs to be based on statistical analysis to find out if there was a significant difference. 

Conclusion 

L404: The conclusions need to be more objective and clearer. The conclusions need to respond clearly to the objectives proposed in the article. Therefore, all conclusions need to be changed.

L405: It is not a conclusion.

L418: It cannot conclude about the flooded area because it was not the object of study of the article and was very superficial approach throughout the text, so you cannot conclude about it.

Reviewer 4 Report

The manuscript addresses important scientific questions, and it is generally suitable for the journal. It is very timing study, based on large number of measurements, adequate, high quality sampling, over long period of observation. The objectives are clear and of high actuality. The English still has to be polished, some phrases need revising.

Specific comments

L35 ITCZ, decipher. A better understanding…

L35, 41: “Then” is not needed.

L76 quantify the evolution of export rates

L108 of elliptical shape

Revise Fig 1d so that the X axis goes till 2017

L186: Is this increase significant? Provide a statistical criterion

Fig. 2. It will be worth adding water temperature here

L204-211: This should be at the beginning of Results or even in the Method section

L227 pattern, not ‘framework’

Fig. 3 is very good and useful. However, uncertainty bars are needed.

L240-241: There is no clear trend in Fig 3b

L252-254: This is very important result. It should be in the Abstract! What about other fluxes? How significant is this increase? The paper may be of high impact just for its DOC part.

Table 3: Add here Si fluxes if available. Again, this may increase the impact of the paper.

L306-313: Presumably, there is no remote sensing for aquatic part of the Congo watershed? If yes, cite it here.

L337 true for their interannual

L346-364: This is highly important part of the manuscript, worth per se of a paper in high-profile journal. The degree of forest flooding changed significantly. This may be one of the main message of the paper, which will attract the interest of C scientists. Fundamentally, it raises a question of effective watershed area relative to geographical area.

L380-382: Remove this. It is unlikely that suspended flux may originate from atmosphere. The ionic composition of dissolved flux is hard to discuss without distinguishing Cl, SO4, Na etc…

L382 : In the Congo River water

L390: This may provide the small differences…

Fig. 8: Temporal evolution of TSS sampled at different depths versus…

L402 particles, not colloids.

Any possibility of measuring POC in TSS?

L403 is repetitive, remove it.

L418 The increase in flooded area and in time of water residence in the flooded forest of left tributaries may explain the increase…

L440-441: Please perform this comparison with S. American and Asian rivers. This will increase the impact of the paper.