Dissolved Iron and Organic Matter in Boreal Rivers across a South–North Transect

Round 1

Reviewer 1 Report

See attached pdf

Comments for author File: Comments.pdf

The manuscript is well written.

Author Response

Overview

This paper presents data on iron and DOC from a suite of high latitude rivers. The authors compare and contrast the results of the different chemical techniques for each river and also include a thermodynamic model. While the data looks overall of high quality, there are still some key aspects of the methodology that are not well explained which disadvantages the reader when they come to evaluate the evidence to support the findings of the authors. In particular there is little information on the parameters that are used in the modelling work and so it is not possible to understand how the measurements inform the model nor how the model improves our understanding of the data. I would urge the authors to include more details on how the work was carried out and to place more emphasis on what measurements are going into the model to inform it.

We thank the reviewer for general appreciation of our work, and we carefully revised the manuscript in response to his/hercomments.

General Comments

There is no indication in the methods regarding the MINTEQ model as to what inputs where in place for the solubility limits for iron. This is important as it is a strong control on total iron concentrations in the soluble phase and a limiting factor for colloidal iron also. It should also be noted that MINTEQ does not include colloids and so it needs to be explained how the data relating to physical separation of iron and DOC was included in the model, if at all. Including details on the parameterization of the model inputs would be a start for this, as the references cited don’t include this information either. This is particularly relevant for Figure 3 – as there is no indication in the text as to what are the species included in this plot.

We added relevant information to the caption of this figure

For the determination of cationic and anionic iron there is simply not enough information on the methodology provided for the reader to be able to understand how this measurement was made. This should be simple enough to include, but currently Figure 6 could be replotted as the % colloidal iron vs total Fe with the soluble Fe level vs latitude, as there are other controls that are likely at play here, such as temperature, pH and overall suspended load. At present it does not appear that the data is being adequately evaluated and more could be done in this manuscript to look at what is controlling iron speciation in these rivers.

We agree with this insightful remark. The information was added. We also added a big deal of discussion on possible factors controlling Fe distribution within colloids and LMW organic complexes.

Specific Comments

Line 119: Was the water temperature recorded at the time of sampling?

Yes; the data added to Supplementary Material.

Line 123: This reference is not appropriate here as it not publicly available (see comment for line 446) nor does it deal directly with the topic for which it is cited here as the thesis is focused on CDOM. There are more appropriate references that show that some organisms do pass through filters (Hahn, 2004; Liu et al., 2019; Nakai, 2020).

Corrected.

Line 144: The reference cited (Drozdova et al., 2021) provides insufficient information on this method, as it does not include any description regards what type of detector was employed for this work and how the system was otherwise quantified or calibrated. It’s clear that this employs ion exchange chromatography for the separation of species by charge but beyond that little else – for instance how where neutral complexes quantified using this scheme?

In response to this comment, missing information was added: “The experimental determination of the content of anionic metal compounds in the dis-solved fraction was conducted using ion exchange chromatography with the anion-exchange resin DEAE-cellulose (Sigma Aldrich). The sample was passed through glass columns filled with cellulose ionite in Cl-form (at a rate of ~1 cm³/min). The columns were 90 mm in length and 25 mm in diameter, the height of sorbent was 40 mm. The content of sum of cationic and neutral forms of metals was determined in the filtrate.”

Line 152: Please provide more information as to the solubility controls on iron used in the model runs. As noted above MINTEQ does not consider colloids, so was an assumption in the modelling that the dissolved and colloidal ligands behaved the same?

Unfortunately, in this study we could not perform vMinteq modeling of metal-colloidal complexation because this model does not consider explicitly colloids. Therefore, we assumed that the dissolved and colloidal ligands behaved in a similar way.

Line 153: It would be helpful for the reader, if the authors could justify the use of the Stockholm Humic model here, as compared to a Nica-Donnan model with both fulvic and humic components (e.g. (Drozdova et al., 2020; Manasypov et al., 2023)).

We added a comparison of two models in the revise text (Line 162-168). From our extensive experience with these models, the Nica-Donnan often overestimates the amount of metals, especially Fe, which is bound to DOM.

Line 161: What temperature are the pH and conductivity measurements made at? Conductivity and pH are temperature dependent measurements (See also the comment for line 119).

The data were added to Supplementary Material. The temperature difference was only 5 degrees, which should not have had a significant effect on pH and electrical conductivity.

Line 209: Please provide more information on the outputs from the MINTEQ modelling. For instance, what species predominates in each river.

The information was added to the Table S2.

Line 216: There is no HPLC method referred to in the methods section, do the authors mean instead refer to the ion chromatography data?

Yes, corrected accordingly.

Line 216: What is the L referred to here? Is it an organic ligand with charge 4-? It could also be in this case Fe(OH)₄^- for the samples from rivers with higher pH (e.g. Ilexa).

In this case, “L” is any ligand, it can be either organic or inorganic.

Line 226: It is not clearly explained why the authors believe that anionic species are allochthonous and cationic/neutral are autochthonous – the authors need to include some text with more details on this.

We agree with this insightful remark. We corrected the text accordingly.

It would also be useful in this context to plot pH vs percentage of anionic as at higher pH you would expect more deprotonation of ligands. Similarly how does the percentage anionic compare to the spectral characteristics?

The information was added in the Figure S5, and we commented these trends in the revised Discussion.

Line 446 – This reference is to a PhD thesis which is apparently not downloadable via the internet, therefore strongly suggest this reference be updated to something more appropriate as indicated above I note that the same reference was also cited previously by Drozdova et al. 2020 so while there is a precedent, it is still incorrect and misleading

Corrected.

Reviewer 2 Report

Line 102 - word from is redundant.

Line 138 - missing word after Major... Do you mean ions, metals, cations or something else?

Line 250 - Is the year 2000 correct?

Table 1 - indescription of table explain MAAT, MAP.

Table 2 - use greek letter kappa for conductivity.

I realized some missing or redundant words. Those sentences are difficult for reader to understand.

Author Response

Comments and Suggestions for Authors:

Line 102 - word from is redundant.

Corrected.

Line 138 - missing word after Major... Do you mean ions, metals, cations or something else?

Corrected.

Line 250 - Is the year 2000 correct?

Corrected.

Table 1 - indescription of table explain MAAT, MAP.

The information was added.

Table 2 - use greek letter kappa for conductivity.

Corrected.

Comments on the Quality of English Language. I realized some missing or redundant words. Those sentences are difficult for reader to understand.

We performed extensive editing of the English, corrected/replaced redundant words and provided missing words wherever necessary

Reviewer 3 Report

this manuscript deals with a current environmental problem, i.e. the darkening of waters. The background and consequences of the problem are well explained, and the results provide more light on the issue. The biggest problem with the manuscript is that calculation methods and statistical analyzes are not described at all. They should be added to the manuscript so that the reader can really evaluate it.

Author Response

This manuscript deals with a current environmental problem, i.e. the darkening of waters. The background and consequences of the problem are well explained, and the results provide more light on the issue. The biggest problem with the manuscript is that calculation methods and statistical analyzes are not described at all. They should be added to the manuscript so that the reader can really evaluate it.

We thank the reviewer for appreciation of our work. We agree with this insightful remark. The information was added in the methodological part of the text (Section 2.2).

Reviewer 4 Report

The manuscript is very interesting and rich in content, but some parts need to be deepened the work to be complete. Below some major corrections are reported.

L75-78 – these are conclusions, not the purpose of the research.

L87 – what does the term "baseflow" mean?

Table 1 – explain the abbreviations MAAT, MAP

L124 – why were samples tested only after 3 days? Did this storage time (at 4oC) affect the results? the iron may have oxidized or reduced during this time.

L160 – pH 8 is alkaline, not neutral (change to ... from neutral to slightly alkaline ...)

L161 delete…in the study…

Section 3.1. compare the Fe results to those recorded in rivers in other areas of the Earth

What were the total Fe concentrations?

Sections 3.2., 3.3. – lack of discussion of the results (the main weakness of the article)

Standardize the units (e.g. L169 Fe mg/L and L198 Fe mg/L)

L238-243 Explain these differences

Conclusions – rephrase (e.g. L273-275 is a re-discussion of the results)

What is the permissible concentration of Fe in water under Russian legislation? this should be discussed in section 3.1.

Based on these comments, I recommend a major revision of this manuscript before final decision about its acceptance.

Author Response

Comments and Suggestions for Authors

The manuscript is very interesting and rich in content, but some parts need to be deepened the
work to be complete. Below some major corrections are reported.

L75-78 – these are conclusions, not the purpose of the research.

The excessive sentence was removed.

L87 – what does the term "baseflow" mean?

We changed by: “summer low-water season”

Table 1 – explain the abbreviations MAAT, MAP

We added this information in the text.

L124 – why were samples tested only after 3 days? Did this storage time (at 4^oC) affect the results? the iron may have oxidized or reduced during this time.

Immediately after sampling and filtration, the samples were acidified with bidistilled HNO₃ and stored in the refrigerator for metal analysis by inductively coupled plasma mass spectrometry. Therefore, storage had no effect on the results. We added explicatory sentences in the revised version.

L160 – pH 8 is alkaline, not neutral (change to ... from neutral to slightly alkaline ...)

Corrected.

L161 delete…in the study…

Deleted.

Section 3.1. compare the Fe results to those recorded in rivers in other areas of the Earth What were the total Fe concentrations?

The information was added in the text (Line 237).

Sections 3.2., 3.3. – lack of discussion of the results (the main weakness of the article)

We added more discussion of the results in these sections

Standardize the units (e.g. L169 Fe mg/L and L198 Fe mg/L)

Corrected.

L238-243 Explain these differences

We suppose that differences in values of pH, DOC and iron content are most likely due to different climatic conditions during the years of sampling (precipitation and temperature). We added the following to section 3.4:

We hypothesize that variations in pH, DOC, and iron content can be attributed to differ-ences in climatic conditions during the sampling years (precipitation and temperature). These parameters may, in turn, influence 1) the degree of river water feeding by surround-ing bogs, providing high molecular weight aromatic OM, 2) connectivity to the groundwa-ter, which provides input of Fe(II) necessary for formation of organo-ferric colloids in the riparian/hyporheic zone, and 3) in-stream aquatic primary productivity, providing au-tochthonous, low molecular weight organic ligands. The complexity and simultaneous operation of these often counteracting factors could explain why there was no correlation between the contents of Fediss and Fe< 1 kDa and pH value and the geographical location (lati-tude) of the sampled water bodies.

Conclusions – rephrase (e.g. L273-275 is a re-discussion of the results)

We removed excessive, repetitive phrases from the Conclusion and revised overall presentation

What is the permissible concentration of Fe in water under Russian legislation? this should be discussed in section 3.1.

The information was added in the text (Line 199).

Round 2

Reviewer 4 Report

The authors addressed my concerns. The manuscript can be accepted for publication in its current form. Thanks!