Erosion and Sedimentation Processes in a Semi-Arid Basin of the Brazilian Savanna under Different Land Use, Climate Change, and Conservation Scenarios
Round 1
Reviewer 1 Report
Major comments:
I never understood what you mean by "off-site" impacts. You state that on-site impacts are erosion that happens in the pixel (line 209), but I don't see any discussion about what off-site erosion is. Table 3 is the only table or figure that specifically mentions off-site erosion results and they were negligible. I am not sure what the purpose of the off-site part of the analysis was for. Either clarify it or only focus on the on-site part of the analysis.
Minor comments:
Lines 32-35: This sentence is awkwardly worded.
Table 1: Factor is spelled incorrectly in the header.
Lines 251-253: I don’t understand why the baseline value of Rm (4944.4) was lower than both the dry year (Rm = 7336.3) and the wet year (Rm = 9874.7). You would think baseline would be in the middle since it the mean of the precipitation dataset.
Figure 5: Use a black box in the BL map to show the areas indicated in a), b), and c).
Lines 309-314: This section is just a single sentence and then the figure 8. I can see that the total sediment retention increases as rainfall increases. But why are all of the scenarios the same within each climate scenario? Why include this figure if you aren’t going to talk about it? I recommend removing the figure.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear authors and editors.
The article is devoted to the assessment of soil erosion losses and sediment yield within the Brazilian savanna. In the article, an assessment is made under the conditions of the baseline scenario, in which land use remains unchanged, and the values of the erosion potential of precipitation will be long-term averages. In addition, erosion, sediment deposition and sediment yield were assessed under the optimistic and pessimistic scenarios of land use change, as well as under climate change towards greater aridity and greater moisture.
The article is fully suitable in its subject matter for publication in the journal.
The main problem of this article, in my opinion, is the presentation of the methodology:
1) Not everywhere it is indicated in which units of measurement the variable is given.
2) Some formulas refer to sources in which I did not find them. Need to find out where they come from.
3) It is not always indicated why these initial data were used, and not others.
More detailed comments below.
Line 15 «108» = 108?
Line 15 «215.3 to 44.1» I didn't find these numbers anywhere in the text?
3 Line 15 1227.2 to 52223.3 Mg.yr-1 from the entire watershed?
4 Line 17-19 I don't think this needs to be give in the "abstract", since this has been known for a long time and the factors that determine erosion in your watershed are the same as elsewhere.
5 Line 19-21 this is also obvious.
6 Line 28 better write «climate» but «climate changes»
7 Lines 45-46 “Bra Borrelli zilian Cerrado” What does it mean??
8 «However, many models are complex, requiring large amounts of data, which hinder their application in data-scarce areas [3]». Since your study is conducted using simulations, here nesesary to give a brief overview of existing models that allow estimation of soil erosion and sediment runoff from catchment areas to rivers. Give a brief summary of their pros and cons, as well as the reasons why you did not use them..
9 Line 77-78 The basin is a relatively “circular” – I don't understand what you want to say here?
1 Figure 1 I would add the names of the reserves.
1 Figure 2 Twice Jul, Jul
1 Line 98-100 Incomprehensible, as if unfinished sentence? Please check.
1 Line 103 here and in formula 1 - "st" in different registers.
1 Mn.d-1 – what is the unit of measure?
1 Line 106 Css – this value is measured once a day and how? Give here briefly the measurement procedure , what are the measurement errors?
1 Formula 4 and line 113 - there is no “b” in the formula, but it is in the explanations, besides the source {22} is not in English, give here in more detail how formula 4 was obtained, this is important.
1 Formula 5 USLEi - in what units is it obtained?
1 Line 118 “sediment retention potential” –do you men here sediment accumulation/deposition?
1 I did not find the exact formula 7 in the source 32, you changed it a little, please justify it.
2 Formula 7 SDRi in what units is it calculated? There are no explanations.
2 Formula 11 USLEi , SDRi – what units of measurement are given here please specify.
2 Line 144 in the original method, Ei has different units than yours.
2 Formula 13, did not find such a formula in source 19, please explain where you got it from and why it can be used.
2 154-157 I do not think that it is necessary to give a definition of the C factor here - it is known to everyone.
2 166 it is not clear why the SRTM DEM was used (it is not clear which version of the model was used), since at present there are several more modern models on which errors associated with vegetation have been eliminated, which is especially important for areas predominantly covered by native vegetation.
2 171-173 the hydrographic network and the river network are not the same thing, not every ravine or gully on the terrain model is a river. Therefore, a justification is needed here, why you can use the model you received as a model of a river network.
2 179-187 there are currently global R factor models, do you know about them and why didn't you use them?
2 201 «by comparing the synthetic and map drainages» I don't understand what it's compared to.
2 202 which of the above equations involve parameters Lmax and kb? What are they needed for?
3 Line 206 How was the value of Eobs estimated?? This variable appears here for the first time.
3 228-232 not clear, please rephrase.
3 233-234 where did the numbers come from 30, 50, 100 м ?
3 238-242 not clear please rephrase.
3 243-247 here you write that the value of the P factor was set to 0.5, and above =1 («Since these types or practices were not analyzed in 160 the present study, the P factor was taken as 1.0»)
3 Section 3 not titled in the text.
3 Figure 4 is difficult to understand, please comment on it a little more in the text.
3 Line 296 How and why the sediment threshold was set in 1.0 Mg. ha-1. yr-1?
3 467-469 I don't think this is the original result. These conclusions are obvious and have been obtained from hundreds of other similar studies.
3 469-470 «for scenario for the scenario» duplication!!
4 475-478 similar results have already been obtained many times by other researchers.
4 In my opinion, in the conclusions, it is necessary to add how the results obtained correlate with the results of other researchers within territories with similar natural conditions, as well as within territories with different natural conditions.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 3 Report
The authors assess the soil loss and sediment yield in the basin considering changing climate, land-cover and management practices. It is meaningful to the control of the soil erosion in the Pardo-FB. In general, the manuscript is well written and easy to understand. However, there are some grammar mistakes need to be corrected before the publication. Therefore, minor revision is recommended.
1. The repetition can be observed in the Abstract and the Conclusion. The authors are recommended to re-write the Abstract and the Conclusion. The results should be removed from the Abstract and the Conclusion should be concise.
2. L10,16: off-side or off-site? Please confirm.
3. The variables in Equations and their definitions are inconsistent. The authors are advised to double check and correct all those variables. (Some of symbols should be subscript)
4. It seems that the Literature review is not updated. Reviewing of some literature which were published recently were recommended. The authors should explain why SDR-INVEST is chosen in the analysis.
5. In the part of conclusion, the authors are recommended to summarize the advantages and limitations of this method for future research works.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Dear authors and editors. Re-evaluating this manuscript, I can say that the article contains quite a few methodological issues that require more justification and clarification. Without this justification, it is impossible to fully guarantee the correctness of the results obtained. The manuscript is somewhat sloppy, with many punctuation errors. All of the above allows me to recommend to the editors of the journal to reject this article in this form.
Подробности ниже in Revision comments2 sections.
- Comment 4
Lines 17-19: I don't think this needs to be give in the "abstract", since this has been known for a long time and the factors that determine erosion in your watershed are the same as elsewhere. Line 19-21 this is also obvious.
Authors' reply
As requested, we deleted lines 17-21 and rewrote the abstract.
Revision comments 2.
You do not explore “landscape connectivity” in your article, so you cannot write about it in the abstract.
Even though you have rewritten "abstract" part, there are still obvious statements that have been made in many studies before you. For example, «The results indicate that anthropic areas and rainfall erosivity increase landscape connectivity, generating higher erosion and sedimentation rates».
- Comment 7
«However, many models are complex, requiring large amounts of data, which hinder their application in data-scarce areas [3]». Since your study is conducted using simulations, here nesesary to give a brief overview of existing models that allow estimation of soil erosion and sediment runoff from catchment areas to rivers. Give a brief summary of their pros and cons, as well as the reasons why you did not use them.
Authors' reply
We corrected the text, as requested.
Revision comments 2. I think this is not an adequate overview. In addition to the SWAT model, there are many models that make it possible to estimate the erosion-accumulation balance in the catchment area and the amount of suspended sediment yield from the catchment area to rivers.
- Comment 11
- Line 98-100 Incomprehensible, as if unfinished sentence? Please check.
- Line 106 Css – this value is measured once a day and how? Give here briefly the measurement procedure, what are the measurement errors?
Authors' reply
-Sentence in lines 98-100 were corrected, as requested.
- Css was better explained and described in corrected text, as requested.
Revision comments 2.
“Daily sediment concentration” and “Css was collected once every 3 months, 4 times a year” here is a contradiction? What instruments measured sediment concentration and what is the error of the method?
- Comment 16
Line 118 “sediment retention potential” –do you men here sediment accumulation/deposition?
Authors' reply
Corrected the text.
Revision comments 2.
The question is the same “sediment retention potential” – do you men here sediment accumulation/deposition?
- Comment 17
I did not find the exact formula 7 in the source 32, you changed it a little, please justify it. SDRi in what units is it calculated? There are no explanations.
Authors' reply
- Formula 7 is in the SDR-InVEST model manual (Sharp.et al., 2020) and was based on the study by Vigiak et al. (2012). We improved the text to represent this fact.
- ‘Sdr’ is the sediment delivery ratio (unitless), as well as all variables in its formula. We correct the text.
Revision comments 2.
In the text, you reference to Vigiak et al. (2012) where there is no such formula, this formula is given in the documentation for SDR-inVest!
- Comment 18
Formula 11 USLEi , SDRi – what units of measurement are given here please specify. Line 144 in the original method, Ei has different units than yours.
Authors' reply
- USLE and SDR units have been corrected in the text, as requested.
- In the latest version of SDR-InVEST’ documentation, the Ei (international) unit is Mg. yr-1, which is different (updated) from the older document and model versions.
Revision comments 2. According to SDR-InVEST' documentation Ei - measured in Mg.ha-1 .yr-1, and in your article in Mg.yr-1.
- Comment 19
Formula 13, did not find such a formula in source 19, please explain where you got it from and why it can be used.
Authors' reply
In source 19 the formula is described as follows:
However is equal to , as shown in equation 11.
Revision comments 2.
I did not find this formula in the indicated source, another formula is given in the SDR-InVEST documentation for calculating sediment retention.
- Comment 21
166: it is not clear why the SRTM DEM was used (it is not clear which version of the model was used), since at present there are several more modern models on which errors associated with vegetation have been eliminated, which is especially important for areas predominantly covered by native vegetation.
Authors' reply
We used version 3 of the SRTM, which has improvements to reduce land-cover bias, and we added this information in the corrected text.
Revision comments 2.
Here I would like to see a little justification why exactly the SRTM v.3 model was used here? What method of correction of errors due to vegetation DEM was used (provide a reference).
- Comment 22
171-173 the hydrographic network and the river network are not the same thing, not every ravine or gully on the terrain model is a river. Therefore, a justification is needed here, why you can use the model you received as a model of a river network.
Authors' reply
The Invest model generates a synthetic drainage network (via GIS flow accumulation) to calculate concentrated flow erosion and sediment transport, using the river hydrography as an input, for the channel threshold (initiation) process. Hence, small order rills and ravines which are not mapped in the hydrography are generated by the model in the basin slopes.
Revision comments 2.
I do not agree with this approach, since the gullies and ravines here will be presented as rivers, which is not true, because they have completely different patterns of transport and sediment accumulation! At a minimum, this approach should be well substantiated in the article.
- Comment 24
201: «by comparing the synthetic and map drainages» I don't understand what it's compared to.
Authors' reply
The ‘total flow accumulation’-TFA is the number of cells necessary for the formation concentrated flow channels, and its definition determines the density of the channel network, based on the DEM. According to Sharp et al (2020), the TFA should be enough for the modeled flows to come as close as possible to the mapped hydrography channel initiation threshold (2500 pixels, in our study), We changed the text to represent this context.
Revision comments 2. I do not agree with this approach, since the model of the river network can be very different from the model of the hydrographic network. It's not the same thing. If you have a small catchment area, you can digitize the river network from topographic maps at a scale of 1:100,000 or 1:50,000.
- Comment 25
202: which of the above equations involve parameters Lmax and kb? What are they needed for?
Authors' reply
‘Lmax’ is the maximum value of the slope length parameter (L) in the LS factor (constant parameter), used in Equation 6. Values of that exceed this are limited to this value. The kb parameter is used in Equation 7. We changed the text to include these improvements.
Revision comments 2.
In the text of the article in formulas 6 and 7, as well as in their descriptions, I did not find these parameters!
- Comment 26
Line 206: How was the value of Eobs estimated?? This variable appears here for the first time.
Authors' reply
Eobs is the Qst (observed sediment yield) obtained with the sediment rating curve, as explained in text.
Revision comments 2.
This explanation must be included in the text of the manuscript. In addition, a good rationale should be included in the text as to why it is possible to compare and calibrate the river catchment sediment yield with the suspended sediment yield measured in the river. After all, everyone knows that the sediment yield consists of many factors, where the yield from the basin is only one of the parts. Another major part of sediment yield is channel erosion, which is especially important for poorly plowed watersheds. There are also other parts of the erosion-accumulation balance.
- Comment 29
Figure 4 is difficult to understand, please comment on it a little more in the text.
Authors' reply
We corrected the equation, as requested.
Revision comments 2. I asked for comments on Figure 4 , not on the equation. Is it measured data or simulation data, how many measurements, etc.?
- Comment 30
Line 296 How and why the sediment threshold was set in 1.0 Mg. ha-1. yr-1?
Authors' reply
The limit of 1.0 Mg. ha-1. yr-1 was established as the off-site tolerance limit was taken from the literature (Morgan, 2005; Moldenhauer and Onstad, 1975; Verheijen et al., 2009), as described in item 2.6
Revision comments 2. Is these figures (10 and 1.0 Mg. ha-1. yr-1) are relevant for the study area, please justify.
- Comment 31
467-469: I don't think this is the original result. These conclusions are obvious and have been obtained from hundreds of other similar studies.
475-478: similar results have already been obtained many times by other researchers.
In my opinion, in the conclusions, it is necessary to add how the results obtained correlate with the results of other researchers within territories with similar natural conditions, as well as within territories with different natural conditions.
Authors' reply
Accepted and conclusions rewritten
Revision comments 2.
Line 405 "27.0 Mg ha-1yr-1", I did not see this value in the results section.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 3
Reviewer 2 Report
Dear authors and editors. I carefully evaluated the responses of the authors and the new version of the manuscript. I can say that the authors answered my main questions and comments in great detail, so I can recommend the article for publication. However, there are several points in the article that need to be corrected. The list below.It is necessary to check all units of measurement further in the text in the methodology section;
11 In lines 191-195 of the new version of the manuscript, there is no explanation of what sediment retention potential is!
22) Kb- still in the text!!!
33) Despite the explanations in the coverletter document, I did not see an explanation of the possibility of using observed sediment loading at the outlet of the watershed for comparative analysis in the new text of the manuscript!!!
44) In lines 250-261 of the new version of the manuscript, I did not see any explanations about on-site and off-site tolerance values.
55) Line 456 19/10=1.9, but not 2.7!!!
Author Response
"Please see the attachment
Author Response File: Author Response.docx
