Quantitative Analysis of Aeolian Sand Provenance: A Comprehensive Analysis in the Otindag Dune Field, Central Inner Mongolia, China

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

See attached review.  I learned a lot; thank you.

 

JUST IN CASE -- Here it is copy-pasted:

Review of “Aeolian Sand Provenance in the Otindag Dune Field” by Y.Y. Cui, Y.L. Zhou, et al..;  LAND 3104392

Submitted June 2024

 

Summary and Recommendations – Moderate Revision (= Can be much more concise; but explain the logic of the methods)

            On the surface, the fact that the winds in this region blow mainly from the west and northwest, and other loose material is carried seasonally from the mountains in the south into this arid basin, seems to logically suggest that the sand composition across this large sandy desert will reflect those physical parameters.  But this detailed analysis definitely supports that logical view with definitive data – they did a lot of sampling for this.  Albeit the automated mineral-analysis and R-package for crunching that data removed a lot of the laboratory and statistical workload.

            Scientifically, their analysis and conclusions seem very solid.  Perhaps partly because it is “as expected”; so no great surprises.  But the methodology used can be useful for other studies by other geologists – hence the main reason to publish their analytical process and interpretations..

Hence, my main recommendations are to explain the methods, logic of the statistical package parameters, and other ratios for a general geologist – see below.  Plus, make the text more concise, because it seems to often repeat that the winds come from the west and northwest, that the heavy grain distribution fits the expected sources, etc.  However, the final statistics need to rounded (62.84% is overly precise when the uncertainty value is nearly 25% => just say 60% +/- 25%?), and the uncertainties made realistic -- e.g., a “(16.05±27.17%)”  would indicate a high possibility of a Negative percentage, which definitely implies something wrong with the statistical calculation method.

 

Some general suggestions or concerns on current manuscript

 

1.  The LOGIC of the “fingerprinting” CI-CR-CTS from the Provenance “R” program

            Much of the analysis was using an “R” package; which is fine.  But saying that the methods are described in that documentation or other publications doesn’t help the reader of this paper.  Please clearly explain what each of these abbreviated parameters imply with simple examples.  For example, what does a “Counting-tHMC” result of 25 imply?

            This is especially true when saying: “In the initial step, the CI and CR methods were employed to remove those tracers exhibiting non-conservative and non-consensual behavior.”, because the reader can’t visualize what is meant by CI and CR, or how they relate to something called “non-consensual” (“consensual” usually means “giving consent to some action”).

 

2.  Correct your “precision” and adjust the impossible standard deviations

The final statistics (and reported values in the tables) need to rounded (62.84% is overly precise when the uncertainty value is nearly 25% => just say 60% +/- 25%?), and the uncertainties made realistic -- e.g., a “(16.05±27.17%)”  would indicate a high possibility of a Negative percentage, which definitely implies something wrong with the statistical calculation method.   Maybe do some of these statistics “by hand”, because obviously the computer version can’t be correct.

And, yes, it is easy to have a 97% goodness-of-fit when fitting to values with such high uncertainties!  The “GoF” isn’t very informative in these cases.

The over-use of precision is throughout the text – probably remove all “decimals” at a minimum:  “dominated by Hornblende, chlorite and garnet, which accounted for 26.51%, 9.4% and 18.12% respectively.” Should be “27%, 9% and 18%”  (Note, it is not clear what is the other 45% of those heavy minerals, if the “dominant” only are 55% of the total).  And similar, beginning with the Abstract.

 

3.  Maps need to show the main places of the text

The introduction talks about the Greater Khingan Range and Yinshan mountains bordering the Otindag dune region.  But neither of those, or “IMDOB” are shown/labeled on the initial location maps.  Also, it would help to show the approximate position of this important “source of Beijing sand storms” on a map that includes Beijing. 

The lower Map of Figure 1 had “UCM, EHOB, etc.” without any explanation in the caption.

 

4.  Use less abbreviations

It rapidly gets confusing when an “O” in one abbreviation means something totally different than the “O” in another abbreviation (e.g., NOP vs PAO).  As a rule, only use abbreviations in which the letters are unique.  I would prefer to OMIT most of the abbreviations are occur only a couple of times – such as “NOB”, “IMDOB”, “CAOB”, “PAO”, “RDD”, “GOF”, etc. – Just write out “Northern Orogenic Belt”, etc.

 

5.  Source Control Phases (for Beijing sandstorms)

            Can you briefly explain what type of controls are being done; and what you would recommend based upon your detailed analysis?  Where is Beijing relative to the different sand movement directions?

 

 

Other items (in order of appearance in the manuscript)

 

1.  Title – Give where in China (e.g., central Inner Mongolia)

 

2.  Line 192 – “VU” – What is it?

 

3.  Line 197 – Is there a reason to think that today’s (2017-2022) wind directions are the same as during the rest of the Holocene for those sand deposits?  Probably YES, but indicate that assumption?

 

4.  Line 278 – “Counting-[delta]tHM” – probably needs “C” = Counting-[delta]tHMC

 

5.  Lines 327 and 350 and 655  – Obviously, one accidently forgot to remove these “Author Instruction” sentences!

 

6.  Line 380 – “OP” – What is it?

 

7.  Figure 8 – mainly lots of dot clouds, too small and scattered to tell the reader much.  Omit?

 

8.  Line 538 – “to be west” – Should be “from the west?

 

9.  Line 596 – “covered by basalt fields” – Do you know the approximate age?

 

Author Response

Responses to Reviewers’ Comments

Manuscript number: LAND 3104392

Title: Quantitative analysis of aeolian sand provenance: A Comprehensive Analysis in the Otindag Dune field, Central Inner Mongolia, China

Authors: Yingying Cui, Yali Zhou, Ivan Lizaga, Zhibao Dong, Jin Zhang, Aimin Liang, Ping Lü and Tong Feng

Dear editor and reviewers,

Enclosed please find our revised manuscript “Quantitative analysis of aeolian sand provenance: A Comprehensive Analysis in the Otindag Dune field, Central Inner Mongolia, China” for publication in Land. Thank you for the comments and constructive suggestions from you and the reviewers about our manuscript (LAND 3104392). We have carefully considered these comments and suggestions in our revision and responded to each of the main comments. In any case, we tried our best to answer questions carefully. We hope that our responses and the resulting changes will be satisfactory, and we will be happy to work with you and the reviewers to resolve any remaining issues.

The revised sections were marked in red color in the revised manuscript. The point-by-point responses to your comments/questions are detailed below. If you have any question about our manuscript, please don’t hesitate to inform us as soon as possible. Thank you again for giving us the opportunity to improve our manuscript.

 

Yours sincerely,

Yingying Cui

29 July. 2024

 

 

Responses to Reviewer #1

Summary and Recommendations – Moderate Revision (= Can be much more concise; but explain the logic of the methods)

On the surface, the fact that the winds in this region blow mainly from the west and northwest, and other loose material is carried seasonally from the mountains in the south into this arid basin, seems to logically suggest that the sand composition across this large sandy desert will reflect those physical parameters. But this detailed analysis definitely supports that logical view with definitive data – they did a lot of sampling for this. Albeit the automated mineral-analysis and R-package for crunching that data removed a lot of the laboratory and statistical workload.

Scientifically, their analysis and conclusions seem very solid. Perhaps partly because it is “as expected”; so no great surprises. But the methodology used can be useful for other studies by other geologists – hence the main reason to publish their analytical process and interpretations.

Hence, my main recommendations are to explain the methods, logic of the statistical package parameters, and other ratios for a general geologist – see below. Plus, make the text more concise, because it seems to often repeat that the winds come from the west and northwest, that the heavy grain distribution fits the expected sources, etc.

However, the final statistics need to rounded (62.84% is overly precise when the uncertainty value is nearly 25% => just say 60% +/- 25%?), and the uncertainties made realistic -- e.g., a “(16.05±27.17%)” would indicate a high possibility of a Negative percentage, which definitely implies something wrong with the statistical calculation method.

Response: Thanks for your suggestion.

(1) For the methods, logic of the statistical package parameters： We carefully considered the logical relationships between the quantitative analysis methods CI, CR, and CTS used in the paper. These three methods progressively propose specific criteria to filter and select the final tracers, which are then used to obtain the final quantitative results within the framework of the Fingerpro model. The specific relationships among them can be found in the section 'Some General Suggestions or Concerns on the Current Manuscript'

(2) For other ratios： We further clarified the role of each parameter by referring to the original references of the mineral parameters used in the paper. Specific Response can be found below.

(3) For the issue of lack of conciseness: We have removed some repetitive expression, in order to make the sentences more concise and clearer.

 

Some general suggestions or concerns on current manuscript

1.The LOGIC of the “fingerprinting” CI-CR-CTS from the Provenance “R” program

Much of the analysis was using an “R” package; which is fine. But saying that the methods are described in that documentation or other publications doesn’t help the reader of this paper. Please clearly explain what each of these abbreviated parameters imply with simple examples. For example, what does a “Counting-tHMC” result of 25 imply?

This is especially true when saying: “In the initial step, the CI and CR methods were employed to remove those tracers exhibiting non-conservative and non-consensual behavior.”, because the reader can’t visualize what is meant by CI and CR, or how they relate to something called “non-consensual” (“consensual” usually means “giving consent to some action”).

Response: Thanks for your suggestion.

(1) For the LOGIC of the “fingerprinting” CI-CR-CTS from the Provenance “R” program：

Since this paper primarily aims to validate the model's application in analyzing heavy mineral data, we strive to explain the logical relationships among the CI, CR, and CTS models as concisely as possible. Their specific definitions and principles are as follows:

1)CI (Conservativeness Index):

Definition: CI is a non-parametric testing method used to analyze whether a tracer exhibits conservativeness (i.e., whether the tracer remains stable and unchanged in the study area, excluding special non-representative tracers).

Principle: It creates an index by integrating source area data and sink data, which is more complex than traditional range testing.

2)CR (Consensus Ranking):

Definition: CR is a scoring function based on multiple random debates among tracers.

Principle: During the debate, tracers that make consensus difficult are assigned lower scores. The consensus process is conducted through a multi-stage method similar to expert opinion integration. Imagine several friends (tracers) discussing an issue (pollution source); if some friends always disagree with others, they will receive lower scores. In fingerprint analysis, the tracers in the dataset are akin to experts presenting differing opinions, helping to identify more reliable tracers.

3)CTS (Consistent Tracer Selection):

Definition: The CTS method is similar to Discriminant Function Analysis (DFA), but it not only identifies the most distinctive tracers but also examines their mathematical properties. For example, if there are three sources, three or more tracers must be selected to ensure that the analytical results are mathematically consistent.

Principle: It can provide a scoring ranking based on the resolving power of the tracers, ultimately yielding combinations of multiple different tracers.

4)Unmixing:

The selected tracers are mixed again to obtain multiple solutions. If the solutions are consistent (reflecting the same viewpoint), we choose the consistent result as the final outcome. If there are multiple solutions (possibly two or three different viewpoints), we select the solution based on expert knowledge to determine which one can be considered the final result.

In summary, we use CI, CR, and CTS to select the most suitable tracers that can effectively distinguish among the three sources for final quantitative analysis by unmixing.

(2) For mineral ratios: We further clarified the role of each parameter by referring to the original references used in the paper regarding the mineral parameters.

1)Counting-HMC: The "Heavy Mineral Concentration index" (HMC) defines the abundance of heavy minerals (transparent, opaque, and unidentifiable turbid grains denser than 2.90 g/cm³). When Counting-HMC < 1, heavy minerals in the sediment are likely influenced by diagenesis, requiring heightened attention during data interpretation.

2)Counting-tHMC: The "transparent Heavy Mineral Concentration index" (tHMC) is calculated from the HMC index. It also reflects diagenetic dissolution. There is currently no strict criterion for what value indicates significant influence from diagenetic dissolution. However, it can be confirmed that the value of Counting-tHMC in the study area is greater than 4, which must not be affected by diagenetic dissolution.

3)Counting-SRD: Counting-SRD is the weighted average density of all clastic particles in the sediment. If the SRD value deviates from the normal density of the sediment source rock, it indicates that the sediment has undergone enrichment of heavy or light minerals, thus being influenced by hydraulic sorting. When SRD > 3.5, it indicates that hydraulic sorting has occurred. In this study, the SRD is around 2.7, indicating that the samples in the study area are not influenced by hydraulic sorting.

4)%OP: Represents the proportion of opaque heavy minerals to the total heavy minerals.

5)%ZR = 100 × (Zircon + Rutile) / (Zircon + Tourmaline + Rutile)

6)Counting-△tHM: Represents the weighted average density of transparent heavy minerals.

Since %OP and %ZR involve opaque minerals, zircon, and rutile, which have relatively high densities, they serve as good indicators for testing hydraulic sorting. Therefore, the linear correlation between %OP, %ZR, and Counting-△tHM can be used to examine whether the sediment has been influenced by hydraulic sorting.

The main function of the other indicators mentioned in the paper, such as ATi, GZi, POS, and MZi, is to determine whether the provenance has changed. Their values vary relatively across different study areas, and there are no specific numerical standards to indicate which type of rock they originate from. The revisions were mainly based on the specific values mentioned by Morton (2005) in "Provenance of Late Cretaceous to Paleocene submarine fan sandstones in the Norwegian Sea."

In the paper, it has been revised as follows:

• “For all the collected samples (Table 2), the Counting-HMC values ranged from 4.14 to 51.36, the Counting-tHMC values ranged from 4.01 to 50.97, and both values were greater than 1.” Please see lines 387-489 in the revised manuscript.
• “The rang ATi of the Otindag is from 3.24 to 85.63 (mostly ＞20) and ATi values are higher in the north than in the south (Fig.6c). It indicates significant influence of granite on the Otindag and the northern part of Otindag is most affected by late Paleozoic granite.” Please see lines 420-423 in the revised manuscript.

The remaining revisions focus on the lines 418-434 in the revised manuscript t of the manuscript.

 

2. Correct your “precision” and adjust the impossible standard deviations

The final statistics (and reported values in the tables) need to rounded (62.84% is overly precise when the uncertainty value is nearly 25% => just say 60% +/- 25%?), and the uncertainties made realistic -- e.g., a “(16.05±27.17%)” would indicate a high possibility of a Negative percentage, which definitely implies something wrong with the statistical calculation method. Maybe do some of these statistics “by hand”, because obviously the computer version can’t be correct. And, yes, it is easy to have a 97% goodness-of-fit when fitting to values with such high uncertainties! The “GoF” isn’t very informative in these cases.

The over-use of precision is throughout the text – probably remove all “decimals” at a minimum: “dominated by Hornblende, chlorite and garnet, which accounted for 26.51%, 9.4% and 18.12% respectively.” Should be “27%, 9% and 18%” (Note, it is not clear what is the other 45% of those heavy minerals, if the “dominant” only are 55% of the total). And similar, beginning with the Abstract.

Response: Thank you very much for your constructive suggestions.

• For the impossible standard deviations:

In the previous model used, the allowable parameters ranged from -5 to 105, which resulted in values such as 16.05 ± 27.17%. We have revised the parameters in our code to be between 0 and 100. The new results differ slightly from the previous ones due to this modification, but the overall trend remains consistent.

Regarding the issue of high goodness of fit (GOF), your statement is correct. First, by modifying the model parameters, we have minimized the high uncertainty in the final results as much as possible. Second, despite the inherent uncertainty in the data, our model is still able to provide valuable insights into the provenance issues in the study area. The model demonstrates high practicality in quantitatively distinguishing different sources, representing an attempt to adapt measured heavy mineral data to a provenance quantitative analysis model, which contributes to advancing traditional regional geological research. Figure 1 shows the result obtained after modifying the parameters.

 

Figure 1. Spatial pattern of provenance contributions of each source type to the Otindag. GOF, goodness of fit; N, sample number. Sediment wind transport potential (red arrow), hydrodynamic transport (green arrow).

• For the “precision”:

All mineral contents, mineral indices, and quantitative results in this paper retain two decimal places for two main reasons. 1) Indicators like the SRD must preserve two decimal places. For example, if the SRD value is 2.73, rounding it to 3 changes its meaning entirely. The former indicates that the sediment is not influenced by hydraulic sorting and that the parent rock source is primarily granite or similar rocks with a density around 2.73. In contrast, a value of 3 suggests that the parent rock source may be of basic composition. This could lead us to draw incorrect conclusions. Therefore, to ensure data consistency, we prefer to retain two decimal places. 2) The mineral content and the final quantitative model of material provenance also need to maintain two decimal places. Rounding could result in a final sum that is not equal to 1 or greater than 1, which is clearly incorrect. Thus, we believe it is necessary to retain two decimal places. The remaining 45% are described by mineral species, focusing on lines 495-304 in the revised manuscript.

3. Maps need to show the main places of the text

The introduction talks about the Greater Khingan Range and Yinshan mountains bordering the Otindag dune region. But neither of those, or “IMDOB” are shown/labeled on the initial location maps. Also, it would help to show the approximate position of this important “source of Beijing sand storms” on a map that includes Beijing.

The lower Map of Figure 2 had “UCM, EHOB, etc.” without any explanation in the caption.

Response: Thank you for your comments. The Greater Khingan Range (Daxinganling Mountain) and the Yinshan Mountains have been marked on the overview map of the study area, and the structural map of IMDOB has been redrawn with the locations indicated in Figure.2c. The abbreviations in Figure 2 have also been revised and explained.

Figure 2. (a)Atmospheric circulation pattern of China including the study region. (b) Locations of the Otindag and sampling points. The sand drift potential was calculated using wind data from the US Meteorological Data Center (https://www.ncei.noaa.gov/maps/hourly/) following the method of Fryberger and Dean [34].) (c)Structural framework of Central Asian Orogenic Belt including the study region (modified after Wang Zhigang[35])(d) Schematic map of the geological characteristics of the Otindag and its surrounding areas, with modifications based on previous studies [35,36]. Abbreviations: SOB, the southern orogenic belt; SZ, the Solonker suture zone; NOB, the northern orogenic belt; EHOB, the Erenhot–Hegenshan ophiolite belt; UCM, the Uliastai continental margin; NCC: North China Craton; IMDOB: Mongolia–Daxing'an Orogenic Belt.

4. Use less abbreviations

It rapidly gets confusing when an “O” in one abbreviation means something totally different than the “O” in another abbreviation (e.g., NOP vs PAO). As a rule, only use abbreviations in which the letters are unique. I would prefer to OMIT most of the abbreviations are occur only a couple of times – such as “NOB”, “IMDOB”, “CAOB”, “PAO”, “RDD”, “GOF”, etc. – Just write out “Northern Orogenic Belt”, etc.

Response: Thank you for your comments. The use of abbreviations in the paper has been revised. Since SOB, SZ, NOB, EHOB, UCM, NCC, and IMDOB are already abbreviated in Figure 1d, their abbreviations will be used directly in the main text. Other terms that appear no more than three times will have their abbreviations removed and will be presented in full. Please see lines 152-166 in the revised manuscript.

 

5. Source Control Phases (for Beijing sandstorms)

Can you briefly explain what type of controls are being done; and what you would recommend based upon your detailed analysis? Where is Beijing relative to the different sand movement directions?

Response: Thank you for your comments. The current desertification prevention and control measures in the Otindag employ a comprehensive approach that combines various methods, including physical and biological techniques. The main strategies involve planting suitable shrubs and grasses or establishing grass grids to control the movement of sand. Relying on major ecological projects such as the Beijing-Tianjin Sandstorm Source Control Project and other national and regional ecological initiatives, a total of 1.394 million acres of forestry tasks for sand source management have been completed, increasing the forest coverage rate from 0.26% in 2000 to the current 1.9%. Focusing on the management of desertified land and comprehensive prevention of desertification, a total of 3.7003 million acres of grassland affected by the "three types of degradation" have been treated, achieving a restoration rate of 22.4%. The overall situation of grassland desertification has been effectively curbed, with noticeable improvements in local ecosystems and significant strengthening of ecological protection efforts. The management of the southern part of the sandy area has shown remarkable results, while some areas of shifting sand remain in the western and northwestern parts. We recommend continuing to strengthen more scientifically based zonal management based on the achievements already made. In the eastern part of the Otindag, measures such as returning farmland to forest are primarily adopted; in the western and northern parts of the Otindag, a combination of measures including rotational grazing and grazing bans is used to increase the coverage of shrubs and grasses, reduce the hazards of desertification, and implement engineering measures for sand fixation. In the southern edge of the sandy land, an ecological protection system is established, mainly focusing on shrubs. Please see lines 567-574 in the revised manuscript.

 

Other items (in order of appearance in the manuscript)

1. Title – Give where in China (e.g., central Inner Mongolia)

Response: According to your comments, we have revised the title to: “Quantitative Analysis of Aeolian Sand Provenance: A Comprehensive Study in the Otindag Dune Field, Central Inner Mongolia, China”.

2. Line 192 – “VU” – What is it?

Response: Thank you for your comments. “VU” is defined in the study of wind and sand physics as the vector unit of sediment transport potential (DP) and composite sediment transport potential (RDP). Edwin D. McKee first proposed the definition of “VU” in 1979 based on the calculation formula for sediment transport potential. The formula is: DP = V²(V - Vt)t, where DP represents sediment transport potential, measured in (VU), V is the wind speed exceeding the critical initiation value, Vt is the critical initiation wind speed, both measured in m/s, and t is the duration of wind action that initiates sand movement, expressed as a percentage of the total time over the year.

3. Line 197 – Is there a reason to think that today’s (2017-2022) wind directions are the same as during the rest of the Holocene for those sand deposits? Probably YES, but indicate that assumption?

Response: Thank you for your comments. This article uses the sediment transport dynamics of the modern surface to provide a clearer and more intuitive understanding of the main transport directions of the modern sand dunes formed by wind in the Otindag. It is evident that the sand in the Otindag has been continuously renewed and repeatedly deposited since the Holocene. However, according to Zhou (2008), the wind direction in the Otindag has not changed significantly compared to the modern monsoon pattern. Therefore, we use the sediment transport dynamics of the modern surface to represent the movement direction of the sand.

4. Line 278 – “Counting-[delta]tHM” – probably needs “C” = Counting-[delta]tHMC

Response: Thank you for your comments. The name in the original reference was without “C”, and we want to be consistent with the references.

5. Lines 327 and 350 and 655 – Obviously, one accidently forgot to remove these “Author Instruction” sentences!

Response: According to your comments, we have been deleted from the manuscript.

6. Line 380 – “OP” – What is it?

Response: Thank you for your comments. %OP is a parameter of minerals that represents the proportion of opaque heavy minerals to the total heavy minerals.

7. Figure 8 – mainly lots of dot clouds, too small and scattered to tell the reader much. Omit?

Response: Thank you for your comments. Figure 8 shows a traditional method of mineral analysis, which reveals the similarity between the source area and the sink area. By comparison, we can observe that the overlap between the source area and the sink area is very high, indicating that the three source areas we selected are appropriate and can all be used for the final quantitative analysis. We hope to retain this figure.

8. Line 538 – “to be west” – Should be “from the west?

Response: Thank you for your comments. It has been modified as required. “By analyzing the spatial pattern of annual sand transport potential in the Otindag, the predominant sand transport directions were found from west and northwest.”

9. Line 596 – “covered by basalt fields” – Do you know the approximate age?

Response: Thank you for your comments. The age of the basalt overlying the lacustrine sand layers varies depending on the sampling location. We selected two ancient lacustrine profiles under basalt pressure, named ERGT and TBTALG. The ERGT sampling site is located on the lava plateau in the northern part of Abaga. The eruptions of basalt in the northern plateau primarily occurred during the Miocene to Pliocene epochs. These can be preliminarily divided into four eruption periods: around 11 Ma during the Miocene (first-level plateau), around 7 Ma during the Miocene (second-level plateau), around 6 Ma during the Miocene (third-level plateau), and around 4 Ma during the Pliocene. The basalt overlying the samples we collected is dated to approximately 4 Ma in the Pliocene. The southern basalt plateau of the Abaga rock area has formed a three-tiered terrace landscape, with ages ranging from 0.47 Ma, 2.08 Ma, to 3.27 Ma from the first, second, to third levels, respectively. This sequence of younger over older ages may indicate that after the late Pliocene, due to the uplift caused by neotectonic movements, some basalt rose to form plateaus, while the sunken areas on both sides accumulated younger basalt. The TBTLG profile is located on the second-level lava plateau, which is approximately 2 Ma old. In summary, the age of the lacustrine sand layers under basalt pressure that we collected is roughly in the Pliocene, predating the modern surface sand layers. This suggests that he could provide sand for the modern surface.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This is a comprehensive and detailed study of the sources of sand for the Otindag dunefield in northern China.  The authors use modern techniques of mineral analysis using Qemscan methodology and instrumentation to identify heavy minerals in source areas and sinks.  Various methods of statistical analysis are used to trace dispersal of mineral components.   

 

In general, the manuscript is well-written, but  is too long and needs to be shortened especially in the introduction and methods section.  There is a lot of data presented here and I suggest that many figures and tables tend to be duplicative.  Some of the figures are not needed.  

 

The authors would be better served by removing the discussion of the complex indices and concentrating on the straightforward comparisons of source and sink using classical approaches. You focus on heavy minerals, but what is the light mineral composition of the sands (Quartz, feldspar)? 

 

 

Specific comments:

 

Figure 3 not needed

 

Section 2.4 – if you can’t use the Garzanti and Ando indices, then why describe them in detail

 

Section 3.2 belong in the discussion 

 

Lines 325-332 and 350-353 need to be deleted

 

Fig 4 – sediment is mis-spelled

 

Figure 6 is incomprehensible – delete

 

Move Tables 2 and 4 to a supplementary data section

 

 

 

Comments on the Quality of English Language

Generally good language use age, but check of errors - see fig 4 as an example

Author Response

1.This is a comprehensive and detailed study of the sources of sand for the Otindag dunefield in northern China. The authors use modern techniques of mineral analysis using Qemscan methodology and instrumentation to identify heavy minerals in source areas and sinks. Various methods of statistical analysis are used to trace dispersal of mineral components.

In general, the manuscript is well-written, but is too long and needs to be shortened especially in the introduction and methods section. There is a lot of data presented here and I suggest that many figures and tables tend to be duplicative. Some of the figures are not needed.

Response: Thank you for your comments. The methods and introduction sections of this paper have been condensed, primarily by reducing the introduction and background sections.

(1) The research status section of the introduction and the description of sand sediment types have been merged to eliminate redundant content. The merged content is as follows: "The Otindag dune field (Otindag) is located in eastern China, within the semi-arid climate zone, on the edge of the East Asian monsoon region. This area serves as a natural laboratory for studying the formation and evolution of sand dunes in transitional zones between agricultural farming and animal grazing. Since 2000, the Ecological Protection and Construction Project in Beijing-Tianjin area has achieved remarkable results. The western part of the Otindag still contains large areas of semifixed dunes, which can easily become the source of sandstorms that negatively impact the health of nearby residents. Therefore, studying the provenance of aeolian sand in Otindag is not only contribute to understanding the formation mechanism of dune field, but also plays a crucial role in ecological restoration and the well-being of surrounding residents. Several researchers have studied the origin of the sand in this dune field. By analyzing rare earth elements, Liu and Yang proposed that the aeolian sand in the dune field originates from nearby sources rather than distant source [8]. Xie and Ding analyzed the U-Pb ages and Hf isotopes of detrital zircons from the Otindag since the Last Glacial Maximum and concluded that the primary sources of the dune field are the Central Asian Orogenic Belt (CAOB) and the North China Craton (NCC) [9]. Previous research has indicated that the CAOB and the NCC are the parent source areas of aeolian sand in the Otindag. However, the specific immediate source areas and quantitative contributions have not been determined. It is crucial to determine the immediate source area of aeolian sand in order to effectively implement control measures and ecological restoration in this region.” Please see lines 51-70 in the revised manuscript.

 

(2) The introduction to the CI.CR and CTS methods in the fingerprint quantitative analysis model has been simplified.

“The revised content is as follows:In order to determine different source contributions, fingerprinting methods have been proposed since the early 1980s as a means of provenance tracing [20]. The procedure estimates the relative contribution of each potential source, using a variety of selected tracer properties. Initial studies were performed based on a single tracer [21]. However, the inclusion of quantitative mixing models and the use of multiple tracers enabled researchers to discriminate more than two sources [22,23]. Nowadays, numerous studies use fingerprinting techniques to examine specific catchment management problems [24,25], to evaluate processes [26] and contamination in the river and coastal waters [27,28]. In recent times, the technique is gaining popularity regarding the understanding of aeolian sand provenance in dessert environments [29-31]. The crucial steps to implement the technique involves the selection of conservative tracers for differentiating potential sources. Lizaga discussed the main methods of tracer selection and found that when traditional ap-proaches such as the mixing polygon or three-step method (i.e., range test, Kruskal-Wallis test and discriminant function analysis test) are used for selecting tracers, erroneous trac-ers could be included while informative tracers removed, leading to errors in both Fre-quentist and Bayesian models [32]. In this regard, the Conservativeness Index (CI), Con-sensus Ranking (CR), and the Consistent Tracer Selection (CTS) methods offer crucial in-formation on the relationship between sources and mixtures and are applied before un-mixing as they are model-independent. These methods also facilitate the detection of non-consensual and non-consistent tracers, as well as the extraction of multiple partial solutions from a dataset.” Please see lines 85-105 in the revised manuscript.

 

2.The authors would be better served by removing the discussion of the complex indices and concentrating on the straightforward comparisons of source and sink using classical approaches.

Response: Thank you for the reviewers’ comments. I have reduced the discussion section on mineral indicators and added research and discussion on direct comparisons between minerals. Please see lines 405-436 in the revised manuscript.

3.You focus on heavy minerals, but what is the light mineral composition of the sands (Quartz, feldspar)?

Response: Thank you for the reviewers’ comments. The light minerals in sand mainly consist of quartz, feldspar, mica, and small amounts of kaolinite, talc, calcite, and dolomite (Table 1). This paper primarily focuses on the heavy mineral content of the samples from the study area for the following two reasons: 1. Heavy minerals tend to preserve their parent rock characteristics during transportation and migration, and the heavy minerals from clastic sediments originating from different sources will form unique compositional features. Therefore, compared to light minerals, heavy minerals are more suitable for source analysis of the Otindag. 2. Table 1 indicates that the light mineral content of the samples from the Otindag. and the three potential source areas cannot effectively distinguish between different sedimentary regions.

Table1 The content of light minerals in Otindag

	Quartz	Feldspar	Muscovite	Other
Otindag eolian sand	35(18-71)	50(38-57)	5(0-12)	0(0-2)
Yinshan sediments	34(20-54)	49(38-49)	6(2-13)	1(0-4)
Lake sediments	35(17-58)	49(30-61)	6(1-15)	1(0-4)
Upwind sediments	35(14-59)	47(35-58)	8(2-21)	1(0-4)

Note: Other: The total of kaolinite, talc, calcite, and dolomite.

 

Specific comments:

1.Figure 3 not needed

Response: Thank you for the reviewers’ comments. Figure 3 have been deleted.

 

2.Section 2.4 – if you can’t use the Garzanti and Ando indices, then why describe them in detail

Response: Thank you for the reviewers’ comments. We have removed the discussion of the Garzanti and Ando indices in section 2.4. Please see lines 231-236 in the revised manuscript.

 

3.Section 3.2 belong in the discussion

Response: Thank you for your valuable opinion. We have placed Section 3.2 in the discussion 4.1. Please see lines 374-401 in the revised manuscript.

 

4.Lines 325-332 and 350-353 need to be deleted

Response: Thank you for the reviewers’ comments. Lines 325-332 and 350-353 have been deleted from the manuscript.

 

5.Fig 4 – sediment is mis-spelled

Response: Thank you for your valuable opinion. We have corrected the mis-spelled of the sediment in Figure 1 (previous Figure 4).

Figure 1. The fan-chart of heavy mineral content of all samples from the Otindag and its potential source area. Amp: Hornblende+ actinolite+ tremolite, Gr: andradite+ spessartite+ almandine+ pyrope, Ep: Epidote+ Zoisite+ Allanite, Chl: Chlorite, Bt: biotite, ZTR: zircon+ tourmaline+ rutile, Oth: monazite+ spinel+ Sphene, Opa: ilmenite+ magnetite+ goethite, Ap: apatite, Pyx: orthopyroxene+ plagiopyroxene, Fa: fayalite.

 

6.Figure 6 is incomprehensible – delete

Response: Thank you for the reviewers’ comments. Figure 6 shows the specific information for each tracer obtained using the CI and CR methods through coding. In the figure, we can see the values of CI and CR. We selected the tracers to be excluded from the CTS method based on the criteria CR > 85 and CI > 15. Retaining Figure 6 can help readers better understand the specific implementation of the model. We hope to keep it.

 

7.Move Tables 2 and 4 to a supplementary data section

Response: Thank you for your valuable opinion. We have moved Tables 2 and 4 to a supplementary data section.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The paper is interesting and presents a methodology with different indexes to characterize the provenance of wind materials.

I only suggest that figure 1 include a small map of the area's position in China, which only occurs in map 2. For readers outside of China it would be interesting.

Author Response

Responses to Reviewers’ Comments

Manuscript number: LAND 3104392

Title: Quantitative analysis of aeolian sand provenance: A Comprehensive Analysis in the Otindag Dune field, Central Inner Mongolia, China

Authors: Yingying Cui, Yali Zhou, Ivan Lizaga, Zhibao Dong, Jin Zhang, Aimin Liang, Ping Lü and Tong Feng

Dear editor and reviewers,

Enclosed please find our revised manuscript “Quantitative analysis of aeolian sand provenance: A Comprehensive Analysis in the Otindag Dune field, Central Inner Mongolia, China” for publication in Land. Thank you for the comments and constructive suggestions from you and the reviewers about our manuscript (LAND 3104392). We have carefully considered these comments and suggestions in our revision and responded to each of the main comments. In any case, we tried our best to answer questions carefully. We hope that our responses and the resulting changes will be satisfactory, and we will be happy to work with you and the reviewers to resolve any remaining issues.

The revised sections were marked in red color in the revised manuscript. The point-by-point responses to your comments/questions are detailed below. If you have any question about our manuscript, please don’t hesitate to inform us as soon as possible. Thank you again for giving us the opportunity to improve our manuscript.

 

Yours sincerely,

Yingying Cui

29 July. 2024

Responses to Reviewer #3

I only suggest that figure 1 include a small map of the area's position in China, which only occurs in map 2. For readers outside of China it would be interesting.

Response: Thank you for the reviewers’ comments. We have redrawn Figure 1 and highlighted area's position in China.

Figure 1. (a)Atmospheric circulation pattern of China including the study region. (b) Locations of the Otindag and sampling points. The sand drift potential was calculated using wind data from the US Meteorological Data Center (https://www.ncei.noaa.gov/maps/hourly/) following the method of Fryberger and Dean [34].) (c)Structural framework of Central Asian Orogenic Belt including the study region (modified after Wang Zhigang[35])(b) Schematic map of the geological characteristics of the Otindag and its surrounding areas, with modifications based on previous studies [35,36].

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for shortening the paper slightly, and changing the title and a better location figure.

I recommend that you use the Suppl. to put in a copy of your nice detailed 2-page reply to me on "For the logic of the fingerprinting", and also the "VU" definition ("VU" is still not included in the manuscript), etc.   Also the fascinating item about the Basalt dating, because that is also relevant, as you indicate in your reply that "the age of the lacustrine sand layers under basalt pressure that we collected is roughly in the Pliocene, predating the modern surface sand layers. This suggests that [those beds] could provide sand for the modern surface."

PLUS, please remove the excess decimal precision throughout.  For percentages, merely state "the total may not equal exactly 100% due to rounding of the analytical results" at the bottom of the relevant tables, or in the text.  We teach our geophysics and geochemistry students to never show more precision in an answer than the precision of the lowest-accuracy input data.  Therefore, a statement like your "The average POS index of aeolian sand in the Otindag dune field was 6.17%" is definitely not what one should be claiming, especially for an "average".

I still think that your ternary diagram suite (Fig. 4 in current version) is not very useful and definitely not very readable to most readers of this journal; and therefore could also go into the Suppl. 

Author Response

Responses to Reviewers’ Comments

Manuscript number: LAND 3104392

Title: Quantitative analysis of aeolian sand provenance: A Comprehensive Analysis in the Otindag Dune field, Central Inner Mongolia, China

Authors: Yingying Cui, Yali Zhou, Ivan Lizaga, Zhibao Dong, Jin Zhang, Aimin Liang, Ping Lü and Tong Feng

Dear editor and reviewers,

Enclosed please find our revised manuscript “Quantitative analysis of aeolian sand provenance: A Comprehensive Analysis in the Otindag Dune field, Central Inner Mongolia, China” for publication in Land. Thank you for the comments and constructive suggestions from you and the reviewers about our manuscript (LAND 3104392). We have carefully considered these comments and suggestions in our revision and responded to each of the main comments. In any case, we tried our best to answer questions carefully. We hope that our responses and the resulting changes will be satisfactory, and we will be happy to work with you and the reviewers to resolve any remaining issues.

The revised sections were marked in red color in the revised manuscript. The point-by-point responses to your comments/questions are detailed below. If you have any question about our manuscript, please don’t hesitate to inform us as soon as possible. Thank you again for giving us the opportunity to improve our manuscript.

 

Yours sincerely,

Yingying Cui

30 July. 2024

Responses to Reviewer #1 Round2

I recommend that you use the Suppl. to put in a copy of your nice detailed 2-page reply to me on "For the logic of the fingerprinting", and also the "VU" definition ("VU" is still not included in the manuscript), etc. Also the fascinating item about the Basalt dating, because that is also relevant, as you indicate in your reply that "the age of the lacustrine sand layers under basalt pressure that we collected is roughly in the Pliocene, predating the modern surface sand layers. This suggests that [those beds] could provide sand for the modern surface."

PLUS, please remove the excess decimal precision throughout.  For percentages, merely state "the total may not equal exactly 100% due to rounding of the analytical results" at the bottom of the relevant tables, or in the text.  We teach our geophysics and geochemistry students to never show more precision in an answer than the precision of the lowest-accuracy input data. Therefore, a statement like your "The average POS index of aeolian sand in the Otindag dune field was 6.17%" is definitely not what one should be claiming, especially for an "average".

I still think that your ternary diagram suite (Fig. 4 in current version) is not very useful and definitely not very readable to most readers of this journal; and therefore could also go into the Suppl. 

Response: Thank you very much for your constructive suggestions.

(1) Add supplementary materials. We agree with your comment. In the supplementary materials, we have provided a detailed description of the logic of the fingerprinting. Additionally, we have included Fig. 4 in the supplementary materials.

(2) Add content in the manuscript. We have added an explanation of "VU" in the manuscript. Please see lines 170-172 in the revised manuscript. We have also included the age of the basalt in the manuscript. Please see lines 205-208 in the revised manuscript.

(3) Remove the excess decimal precision throughout. We have adjusted the precision throughout the manuscript. Except for the longitude of SRD and Counting-△tHM, which must be retained to one decimal place, all other longitudes have been rounded to the nearest whole number.

Author Response File: Author Response.pdf