Decision on bg-2018-4

Dear authors
The three referees have made many critical and constructive comments on your MS. Because they all recognise the high quality of your data I believe it can be reconsidered after major revisions. I recommend you take a lot of care in addressing all the reviewers’ comments and particularly those that have potential strong impact on the conclusions of your paper. Note that the revised version must be considerably improved before your MS can be accepted for publication and that a second round of review will probably be necessary. Indeed, the reviewers have pointed at the very poor quality of your English language, which prevented a complete evaluation of your study during this first round of review. Consequently, I ask you make the necessary to write your MS in fully understandable English.

Looking forward to reading this soon.

With best regards

Gwenaël Abril BG associate editor

Decision#2 on bg-2018-4

Dear authors,

Because I wanted to move forward the review process of your paper and the edition of the special issue on carbon in Asian rivers, I made my own detailed evaluation of your revised MS. This new version that was supposed to be intensively revised has disappointed me and gave me hard work. As already pointed out by the two referees during the first round of review, I also had difficulties to read and follow this second version of your MS, which is still very confusing and would need profound revision before it can reach the quality standards for biogeosciences. My critics (detailed below) concern the form (the presentation of figures and text, English language, structure and length of the text) as well as the scientific content with some important missing information, and some hazardous and repetitive statements concerning for example the 13C signature of the DIC. In addition, the method used to calculate DIC fluxes to the coastal ocean based on estuarine data and freshwater discharge is not appropriate and in contradiction with classical literature on estuarine biogeochemistry. At that step, these critics would be largely sufficient to simply reject you paper. However, because of the important sampling effort, the intrinsic quality of your dataset and, the general lack of data in this region of the world, I believe your paper should be published in BG as a significant contribution to the special issue on C in Asian Rivers. However this can be done only if you seriously take into account the numerous comments listed below and you profoundly re-work your MS, change text, figures and tables and provide all necessary information, and make your interpretation of the data much less speculative. Again, major revisions are necessary and I may send the next version of your MS for more external review. When revising your MS, please keep in mind that reviewing is a volunteer time-consuming work and that referees and editors are not supposed to re-write papers. English editing is also necessary.

Major comments

Method for DIC fluxes calculation
L200 “Total export flux of DIC… multiplying the mean concentration of DIC in an estuary with the annual discharge. Spatial variability of DIC in estuaries was minimized to a large extend by collecting samples from head to mouth of the estuary” Estuarine saline samples contain a fraction of seawater; their DIC is thus a mixture of riverine DIC and marine DIC, with proportions depending to the value of salinity. The concentrations of DIC in these estuarine samples are affected by dilution with seawater and cannot be multiplied by the freshwater discharge to derive a flux to the ocean. According to all textbooks on estuarine hydrology and chemistry, the DIC concentration at the zero salinity end-member must be used. Concentration versus salinity plots can be used to obtain an “apparent zero end-member” (AZE). The method used here is underestimating the real DIC flux when freshwater DIC cc is higher than seawater DIC cc (most cases in these indian rivers), and, inversely, overestimating the flux when river DIC < ocean DIC.

The use of the term “estuary” all through the MS in often inappropriate. At many places in the MS, in the abstract, introduction and discussion, you mix the two notions of river and estuary. You use the word “estuary” when it should be “river”. See below

Interpretation of 13C DIC
In general, the discussion of the 13C DIC signal is confusing and too superficial. It must be strengthen by more quantitative analysis that allows to evaluate the respective contribution of processes under some specific conditions. You mention weathering of silicate/carbonate and “storage” in dams as major processes. However the relationship between DIC cc and 13C and lithology is only verbal. You must relate DIC concentrations and 13C-DIC signature with lithology and the proportion of carbonate versus silicate rocks on the respective watershed.

Include lithology in the map in figure 1 and provide in a table the proportion of carbonate and silicate rocks in each watershed. The table should also include classical information for each watershed (discharge, drainage, presence of dams, eventually population density). All information necessary for the quantitative information of the data should be included. Sections 4.1.2 and 4.1.3 on the origin of DIC is extremely speculative and sometimes repetitive. We must know what are major processes and minor processes. Pollution by megacities is ignored although it is probably a major driving force on DIC in these rivers.



Detailed comments

Abstract
L14: spatial and/or temporal variability?
L17: “exchange of groundwater with low DIC” this is speculation because groundwater are supposed to have high DIC concentrations
L21: “intrusion of marine waters”: intrusion of seawater occurs in all estuaries (not only east-flowing estuaries), and are more or less important depending of sampling strategy and salinity of the samples. Comparison between systems can be done only on the basis of the zero salinity end-member.
L27-28: “9.4% of the Asian rivers” unintelligible sentence

Intro
L44 “DIC is the major constituent of carbon species” where, please rephrase
L46 “33-400 Tg of DIC” do you mean 330-400 Tg? Please proofread you MS to remove such typos before re-submitting
L50: should be “rivers” instead of “estuary” here
L51-55: rephrase and specify what process is more important quantitatively
L62-65: this is too general, how does human alterations of riverine DIC relate with findings of your study?
L70-88: this is too general, and not focussed in relation with the present study. Please rewrite
L88 “their sources” of DIC?
L90 aquatic systemS
L91-93 “due to distinct isotopic composition of different sources” not only as 13C-DIC is also affected by fractionation by internal processes such as aquatic primary production and gas exchange. The figure 5 should be used here (as new figure 2) to support this statement
L100: “despite distinct isotopic composition of DIC is expected for different sources, the identification of DIC sources is still challenging due to isotopic fractionation associated with complex mixture of sources and processes such as photosynthesis” awkward sentence. Mixture does not fractionate. All through the MS you must prioritize what is more important among the different processes that impact 13C-DIC

Study area
We need a table with all necessary information about each watershed
L121_124 Not sure information on upwelling in the Arabian sea is necessary here
L138 rivers instead of “estuaries”?
L147 when talking about discharge, better refer to “rivers” rather than “estuaries”
Sample collection
We need to know about salinity of your samples. Comparing data with high salinity from an estuary with data with low salinity in another estuary is meaningless. Salinity must appear somewhere in the MS, and what should be compared is DIC and 13C-DIC at the zero salinity end-member.
L200: use the AZE DIC concentration to calculate fluxes

L214: here you mix different information (river composition and dilution with seawater) and the fact that salinity is high in an estuary is due to your sampling strategy. All interpretation should refer to river end-member, and should not include different proportion of dilution with seawater
L233-234 unintelligible language
Section 3.3: re-calculate DIC fluxes using the zero salinity end member
L258 “no vertical salinity stratification was observed in all estuaries sampled… Mandovi estuary” awkward sentence
L273 “among the estuaries sampled along the indian coast, the SW estuaries are characterized by lower mean concentrations of DIC than the SE, NE…” awkward sentence, please rephrase
L279 “the SW region of India receive the highest amount of precipitation during the SW moosoon than the SE, NE and NW regions of India” awkward sentence, please rephrase
L286 “About three times higher catchment area normalized discharge might have diluted DIC concentration in the SW estuaries” not clear, please rewrite
L289 rivers instead of “estuaries”
L290-293 not sure dilution will alter the 13C-DIC. You mention “residence time of soil water” but this is not clear: what process will deplete 13C DIC ? Be more precise when referring to Amiotte Suchet 1999
L296: you mention here “lower catchment area” but the info on catchment areas is missing in the MS
L300-304: rewrite and provide more quantitative evidence for the occurrence of in stream processes. Here the interpretation of the data is only verbal. What instream process are you referring to?
L309-311. Comparing low salinity 13C signature in a given estuary with high salinity 13C-DIC signature in another estuary makes no sense. The fact that 13C-DIC correlates with salinity in a given estuary is a truism. These types of statements must be avoided in all the MS
L320 “exchange of SW estuaries with…” unintelligible sentence: how can you exchange en estuary?
All the literature converges to the conclusion that groundwater have high DIC concentrations, so your statement here is pure speculation
L328 “DIC input from the dissolution of atmospheric CO2 can be ruled out”. Isotopic equilibration of DIC with the atmosphere occurs even if waters are oversaturated with respect to the atmosphere.
L331-336 this is speculation : the fact that 13C DIC correlates positively with DOC does not necessarily mean that oxidation of organic matter predominates. High 13C DIC can be due to fractionation by phytoplankton and high DOC to the exudation/hydrolysis of phytoplanktonic cells. The correlation can also be indirect, High DOC and high 13C DIC coming from the nature of soils and rocks (or pollution) on the watersheds, or can also be due to the dilution of seawater in the samples. The highest value of DOC in fig 4e at about 13 mg L-1 may also come from contamination by sewage. See also below my comments on figure 4.

L338 “is distinctly enriched than that…” unintelligible language
L339-342: the contribution of C4 plant material is speculative. Many other processes may enrich DIC in 13C such as carbonate weathering (which also increases alkalinity), gas exchange and mixing with seawater DIC.
We need some quantitative analysis here: what are the contributions of silicate and carbonate rocks on the respective watersheds, and does 13C DIC correlate with alkalinity? High alkalinity is an indication of carbonate weathering

L326-377. All the section 4.1.2 is extremely speculative, discussion on processes are only verbal and not quantitative. All the possible processes are mentioned but without giving more emphasis on one process compared to another. Entire section must be rethought and re-written.
L360 “we could not able to evaluate” please proofread your MS before submitting!
L362 “the same is correlated with DO saturation and found significant…” revise language
“confirming that biological processes enriched 13C DIC in the Indian estuaries. CO2 outgassing due to heterotrophic decomposition…. In reservoirs/dams…” This is extremely confusing and speculative. In fact you mention all possible processes but you don’t know from your analysis which one is most important. In addition “biological activity” is too general and what most probably makes the 13C become enriched is primary production (autotrophy) and not heterotrophy.
Pollution by megacities is ignored although it is probably a major driving force on DIC in these rivers.

L388 your are repeating the same statements as before
L395 “…consumption of atmospheric/soil CO2 through silicate weathering is lower by 2 times…” is this a rule that applies everywhere? Rephrase

End of section 4.1.3 is too long and repetitive

L427 80% of DIC and 20% of DOC (what about POC?) is different from other rivers in the world and not “consistent with earlier reports elsewhere”

L435 “similar those found…” revise language

L451: you repeating the same statements
L465: “significant correlation between soil organic carbon and DIC yield … confirms that strong influence of soil and DIC yield…” a correlation does not necessarily prove a process, correlation can be indirect. However, this is probably one interesting finding that deserves more discussion: soil OC oxidation will release DIC in the form of excess CO2, but it wil also enhance weathering and the export of DIC in the form of alkalinity. Deserve more discussion

L489 (and abstract) “storage of water in dams” you must consider the probable occurrence of phytoplanktonic production that would decrease DIC and increase 13C DIC

Owing to the population density in India, DIC is probably largely affected by the release of sewage waters

Fig1: please show lithology and calculate the respective proportion of carbonate and silicate in each watershed

Fig2 and 3 can be combined. Please place the west on the left and the east on the right as in figure 1. Reading would be improve by showing DIC yields on a seprated panel

Figs 4 and 6 are very confusing, number of data points are different from one panel to another, avoid mixing different information in such figure. Prepare one figure with a limited number of variable on the X axis. It is not clear why fig 4a, 6g and 6 h have only 4 data points: insert these ones in a different figure.
What do ovals represent is not mentioned in the captions
In fig 4 you show 13C DIC versus salinity, and 13C dic versus DOC. What about DOC and DIC versus salinity? Salinity plots should appear in a separate figure in order to calculate zero salinity end-members

Fig5 could appear before when mentioning the processes controlling 13C DIC in rivers in the introduction

Table 1 is useless; what is the meaning of comparing DIC concentrations in world rivers with Indian estuaries that include saline samples?

Table 2 should show the information for individual rivers, not only regions. Quantitative information on the lithology in the watershed should be included and used for quantitative analysis of the data

Dear authors,
Before your paper can be published, please revise the following points and submit a revised MS together with a letter that explain how you have addressed the comments
Thanks you in advance
G Abril, BG associate editor


L18-22 make only one sentence, no need to repeat twice “keeling plote & miller Tans..” in the abstract
L21: “the similar 13C of DIC source”: what DIC sources and where, all rivers? rephrase
L22-23 not clear what you mean by “pH and DIC speciation on measured 13C-DIC”; the sentence about 13C DIC in the abstract can be more focused on main result
L24-25 it seems strange that 13C DIC trends are not mentioned here: depleted when heterotrophic respiration and enriched in autotrophic sites?
L38-40 you basically repeat what was said before in the abstract without giving more detailed interesting information such as : where is autotrophy favoured (and why), where is heterotrophy favoured and why?
Abstract could be improved to provide more precise information, avoiding too general statements, and repetitions

L58 and L68 Raymond et al 2013 is not the appropriate reference for DIC export fluxes to coastal regions
L105 it is not clear here if “equilibrium” is the right word, rather than “isotopic equilibration”. If “Isotopic equilibration with the atmosphere” refers to the changes in 13C-DIC due to selective 12CO2 and 13CO2 fluxes at the water-air interface, then the process is very significant when water pCO2 is high such as in small streams. (Deirmendjan and Abril 2018 JoH, and refs therein from J. Barth group). Please update the statements here L105-110

L154 remove “rather than headwaters” no need
L225 and L382under-saturation, not “unsaturation”

L300 exchange of CO2 with the atmosphere, and not “exchange with atmospheric CO2”
L323 “Bouillon” not “bouvillion”
L354 “interpreted “as” not “to”
L372 “DIC … is largely contributed by” please rephrase
L394 “the remaining into the Arabian Sea” please rephrase
L397 not sure “far less” here is correct English


What does shaded area mean in table 1?
Fig 1 is of poor graphical quality, please improve
Fig2 I could not understand why degassing at pH <6.4 and at pH > 6.4 have different effect on 13C-DIC. Degassing will make 13C-DIC converge to about zero per mil
Weathering by “carbonic acid derived from dissolution of atmospheric CO2” is something strange, and likely not significant in soils where most of the CO2 comes from soil respiration. Where do these statements come from? Please revise this figure and improve the legend

Dear authors
thank you very much for the detailed revision of your MS, which is now acceptable for publication in Biogeosciences. I would like to apology for the duration of the review process, due to delay in the editing of the special issue on asian rivers, and also to the numerous iterations necessary for your paper to reach the quality standards for BG. Thank you for submitting your work to BG and to contribute to the special issue with this important work.

Before you MS can be published, Figure 1 must be improved:

This figure is of poor quality and some letters appear in the background (something like "mapsof"). Please redraw to remove these letters. We also need a reference for the lithological map, it can be an published report or an official web site.

with best regards
Gwenaël Abril, BG associate Editor