Review Reports - Effects of Combined Application of Nitrogen Fertilizer and Multiple Soil Amendments on Soil Properties and Bacterial Community Structure in Arid-Zone Jujube Orchards

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study presents a well-structured three-year field experiment evaluating combined soil amendment effects in an arid jujube orchard. The NBP treatment shows consistent superiority across physicochemical, biological, and yield metrics, and the SEM framework is conceptually appropriate. However, the manuscript suffers from an overly descriptive analytical approach, and a fragmented narrative diluted by excessive figures. Both the Introduction and Discussion require targeted strengthening to better contextualize findings within current soil microbiome and arid agroecosystem literature. Substantial revision is recommended before acceptance.

Major comments

1. The manuscript applies a standard microbiome pipeline (alpha diversity, PCoA, phylum composition, FAPROTAX, RDA, Pearson correlation) without clear analytical focus. This descriptive approach fails to directly address the central mechanistic question. The authors should replace peripheral analyses with targeted approaches to generate mechanistic insights. Consolidating or moving non-essential community analyses to supplementary materials would substantially sharpen the manuscript's scientific focus. The Discussion section should reflect this mechanistic reframing accordingly.

2. The manuscript contains 14 main figures reflecting a data-reporting rather than hypothesis-driven presentation. Figures addressing overlapping content and those contributing minimally to the central thesis should be consolidated or moved to supplementary materials. The Introduction should more explicitly frame the study's core hypotheses to guide readers toward the most relevant results.

3. FAPROTAX-predicted functional gene abundances are presented as corroborating evidence for enhanced nutrient cycling, but the tool's limitations are inadequately acknowledged. The reported 2.7-fold increase in "ureolysis genes" cannot be equated to actual ureolytic activity. Authors should either validate predictions using targeted qPCR of functional genes (e.g., ureC) or substantially reframe these results as hypothesis-generating.

4. The SEM explains only 26% of urease variance and 18% of Shannon diversity variance, indicating substantial omitted variables. Excluding soil pH, water content, and available P likely introduces omitted variable bias into path coefficients. The exceptionally strong UA>yield path (β = 0.908) may partially reflect collinearity with unmeasured covariates rather than direct causation. Did you compare alternative model specifications using AIC? And the Introduction should also better justify the theoretical basis for the hypothesized causal chain in SEM results.

Minor comments

1. Treatment group colors (CK, NB, NS, NP, NBP, NW) differ between figures throughout the manuscript. A single, fixed color palette applied consistently across all figures would significantly reduce cognitive burden and improve visual coherence.

2. There are typo errors (e.g., Keywords propertiess, L628 may reflects). A thorough typo error check is recommended throughout.

3. L784-786: It looks MDPI template placeholder text, not your own data availability texts. Please, provide accession number in a public repository.
4. Figure 13 presents numerous pearson correlations without multiple testing corrections. Without correction, a substantial proportion of reported significant correlations may represent Type I errors. Please apply correction and revise accordingly.
5. Figure 8 is unclear. strongly recommend to use a more informative and alternative for multi-group OTU intersection analysis.
6. Chloroflexi abundance increased by 62% under NBP, yet receives no mention in the Discussion. Given this phylum's roles in carbon cycling, aggregate formation, and anaerobic metabolism, its enrichment warrants mechanistic interpretation.
7. L28-29: The Abstract states that "urease activity served as the primary pathway linking soil organic matter improvements to enhanced crop productivity." Given that the SEM explains only 26% of urease variance, this causal claim is overstated.
8. L211-213: Please verify and standardize the units for all enzyme activities to ensure consistency and reproducibility.
9. L396: The x-axis order of treatments in Figure 6 differs from the order used in all other figures
10. References: Several references do not follow a consistent citation format. All references must be reformatted to comply with the journal's author-year citation style before resubmission.

Author Response

Dear Editor,

We would like to submit our revised manuscript entitled “Effects of Combined Application of Nitrogen Fertilizer and Multiple Soil Amendments on Soil Properties and Bacterial Community Structure in AridZone Jujube Orchards” to agronomy. The concerns raised by reviewer(s) have been properly addressed in the revision (see Author’s response to reviewer’s comments). We hope now the manuscript can be accepted for publication. If you have any further questions, please don’t hesitate to contact us at the address below.

Thank you and best regards,

Yuxuan Wei and Yuyang Zhang (Email: wyx19982026@163.com;)

On behalf of all co-authors

Authors’ responses to Reviewers’ comments

Reviewer 1

Comment: The manuscript applies a standard microbiome pipeline (alpha diversity, PCoA, phylum composition, FAPROTAX, RDA, Pearson correlation) without clear analytical focus. This descriptive approach fails to directly address the central mechanistic question. The authors should replace peripheral analyses with targeted approaches to generate mechanistic insights. Consolidating or moving non-essential community analyses to supplementary materials would substantially sharpen the manuscript's scientific focus. The Discussion section should reflect this mechanistic reframing accordingly。

Response: Thanks for this observation. You're right that the first draft was basically running through the standard microbiome checklist without a clear story. We've fixed this by focusing everything on the SOM → urease → yield pathway that our SEM revealed. Here is exactly what we changed:

Consolidation of peripheral analyses to supplementary materials.As suggested, we have moved the descriptive OTU composition analysis (formerly Fig. 8) and the detailed PCoA clustering patterns (formerly Fig. 9) to Supplementary Figures S1 and S2, respectively. In the main text, these are now condensed into a single mechanistic paragraph (Section 3.3.2) emphasizing that NBP and NW treatments, despite differing amendment strategies, converged toward similar community configurations that align with their shared enrichment of N-cycling functional genes and comparable yield responses (lines 431-441).
Mechanistic reframing of phylum-level analysis.We have completely restructured Section 3.3.3to focus on functional guilds rather than taxonomic enumeration. The revised text now explicitly links specific phyla to biogeochemical functions: Actinobacteriota and Bacteroidota are discussed as carbon-decomposing consortia corresponding with elevated invertase activity; Proteobacteria as N-cycling guilds aligning with urease activity; and Gemmatimonadota as P-solubilizing groups matching available P increases (Fig. S3). We have removed peripheral descriptions of phyla with unclear mechanistic relevance (e.g., detailed Acidobacteriota ecology) to sharpen the focus on nutrient cycling mechanisms (lines 442-455).
Integration of microbial data with soil functional indicators.We have established explicit connections between community shifts and ecosystem function by correlating taxonomic changes with measured enzyme activities and nutrient pools. For instance, the 43% enrichment of Actinobacteriota under NBP is now directly tied to the 178% increase in invertase activity (Fig. 5), and the enrichment of Proteobacteria is linked to enhanced urease activity and nitrogen availability (Figs. 4 and 5).(lines 447).
Clear delineation between exploratory and confirmatory analyses. To avoid analytical redundancy, we have repositioned the RDA (Fig. 9) as an exploratory approach to identify environmental gradients, explicitly stating that these patterns provided foundational hypotheses subsequently validated by the SEM framework (Fig. 11). This hierarchical distinction clarifies that RDA serves to generate mechanistic hypotheses (pH and SOM as drivers), while SEM tests the causal pathways (SOM → urease → yield)(Section 3.4.1; lines 488-498).
Discussion reframed around mechanistic pathways.In the Discussion (Section4.2), we have deleted the peripheral descriptions of Acidobacteriota subdivisions and FAPROTAX methodological limitations. Instead, we now emphasize the functional specialization hypothesis, explicitly discussing how specific recruitment of decomposer guilds (Actinobacteriota, Bacteroidota) and nutrient- cycling taxa (Proteobacteria, Gemmatimonadota) mediates the relationship between organic matter input and crop productivity, consistent with the SEM-identified pathway (lines 608-621；lines 644-656).

Comment: The manuscript contains 14 main figures reflecting a data-reporting rather than hypothesis-driven presentation. Figures addressing overlapping content and those contributing minimally to the central thesis should be consolidated or moved to supplementary materials. The Introduction should more explicitly frame the study's core hypotheses to guide readers toward the most relevant results.

Response: Thanks for this incisive comment. We have addressed this comprehensively through the following revisions:

Streamlining the figure panel (14→11 figures): Beyond the changes made in response to Comment1(moving the OTU Venn diagram and detailed PCoA clustering to Supplementary Figures S2 and S3), we have further relocated the phylum-level composition chart (formerly Fig. 10) to supplementary materials (now Fig. S3).
We have completely rewritten the final paragraph of the Introduction (lines 100–115). The previous "laundry list" of five objectives—(1) evaluate soil properties, (2) characterize bacterial communities, etc.—has been replaced with a hypothesis-driven framework that explicitly posits the three causal statements quoted above. This revision frames the study as a rigorous test of which microbes mediate the amendment effect (functional guild recruitment) and how they do it (enzyme-mediated nutrient mineralization), guiding readers toward the most relevant results from the outset.

Comment: FAPROTAX-predicted functional gene abundances are presented as corroborating evidence for enhanced nutrient cycling, but the tool's limitations are inadequately acknowledged. The reported 2.7-fold increase in "ureolysis genes" cannot be equated to actual ureolytic activity. Authors should either validate predictions using targeted qPCR of functional genes (e.g., ureC) or substantially reframe these results as hypothesis-generating.

Response: We appreciate the reviewer’s critical assessment of the FAPROTAX results. We have addressed this concern through targeted revisions in the Results section (Section 3.3.5). We have added a disclaimer at the end of Section 3.3.5 stating: "It is important to note that FAPROTAX provides predictions of metabolic potential based on taxonomic assignments rather than direct measurements of functional genes... these predictions should be interpreted as indicative of functional potential rather than confirmation of actual gene expression or enzymatic rates." Also in Section 3.3.5, we have revised the terminology to reflect the predictive nature of these data:

Changed "exhibited the highest abundance" to "was predicted to harbor"

Changed "representing a 2.7-fold increase" to "representing a putative 2.7-fold increase"

Changed "were most abundant" to "were predicted to be most abundant"

Added "predicted" before all gene abundance descriptions. (lines 454-478).

Given that we do not have targeted qPCR data (e.g., ureC) in the current study, we have followed the reviewer’s alternative suggestion to substantially reframe these results as predictive, hypothesis-generating evidence within the Results section, avoiding any implication that FAPROTAX predictions equate to actual gene expression.

Comment: The SEM explains only 26% of urease variance and 18% of Shannon diversity variance, indicating substantial omitted variables. Excluding soil pH, water content, and available P likely introduces omitted variable bias into path coefficients. The exceptionally strong UA>yield path (β = 0.908) may partially reflect collinearity with unmeasured covariates rather than direct causation. Did you compare alternative model specifications using AIC? And the Introduction should also better justify the theoretical basis for the hypothesized causal chain in SEM results.

Response: Thank you for these important statistical concerns regarding the SEM specification and interpretation. We have carefully addressed each point through the following revisions:

On the low explained variance (26% for urease, 18% for Shannon):You are right that substantial variability remains unexplained, which is typical for field experiments subject to unmeasured microclimatic fluctuations, root exudate chemistry, and legacy management effects. We now explicitly acknowledge this limitation in Section 3.4.3 while noting that the model captures 88% of yield variance, suggesting the specified SOM→UA→Yield pathway represents the dominant productivity driver in this system, even if peripheral biological processes (diversity, precise enzyme regulation) involve additional unmeasured factors.
On omitted variables (pH, water content, available P):We acknowledge these variables likely influence the system to some degree. However, the high yield variance explained (88%) indicates our core mechanism captures the primary causal chain. The influence of additional edaphic factors is now acknowledged as a limitation in Section 3.4.3: "While the moderate R² values... indicate substantial influence of unmeasured factors... the high yield variance explained suggests the specified pathways capture the dominant mechanisms." (lines 564-567).
On the strong UA→yield path (β = 0.908) and collinearity concerns:We calculated variance inflation factors (VIFs) for all predictors to rigorously test this: SOM (1.28), AHN (1.28), UA (1.34), and Shannon (1.09)—all well below the collinearity threshold of 5.0 (and even the conservative 2.0 threshold). This statistically excludes multicollinearity artifacts. Ecologically, we explain this strong coefficient in Section 4.3 by referencing the severe nitrogen limitation characteristic of desert brown soils: in such N-limited systems, urease-mediated mineralization constitutes the primary bottleneck, naturally exhibiting disproportionate influence on yield. Thus, β = 0.908 likely reflects genuine process dominance rather than statistical artifact.(lines 596-715).
On AIC comparison and model selection:We compared four alternative specifications (detailed in Table S3). While Models 1 (AIC = 11.215) and 2 (AIC = 31.431) exhibited lower AIC, they showed significant misspecification (χ² p < 0.001), indicating overparameterization. Our final model (AIC = 122.425) achieved the best balance—acceptable fit (χ² = 7.508, p = 0.057) without overfitting. We have added this comparison to Section 3.4.3.
On theoretical justification in the Introduction:We have added theoretical grounding in the third paragraph: "Soil ecological theory emphasizes that nutrient mineralization is governed by specific extracellular enzymes catalyzing rate-limiting steps... In N-limited desert soils, urease activity represents the primary bottleneck... thus functioning as the pivotal mediator between soil organic matter pools and crop productivity." (lines 90-97).

Summary of key statistical evidence now provided:

Model fit: χ² = 7.508, df = 3, p = 0.057; Fisher's C = 12.128, p = 0.059
VIF values: 1.09–1.34 (all < 2.0, well below 5.0 threshold)
R²: Yield = 0.88; UA = 0.26; Shannon = 0.18
AIC comparison: 4 models tested (Table S3)

Comment: Treatment group colors (CK, NB, NS, NP, NBP, NW) differ between figures throughout the manuscript. A single, fixed color palette applied consistently across all figures would significantly reduce cognitive burden and improve visual coherence.

Response: We sincerely appreciate this valuable suggestion regarding visual consistency. To address this concern, we have implemented the following revisions:

We have standardized the color scheme for the yield figures to match the palette used in the soil physicochemical property figures, ensuring visual coherence throughout the manuscript.
We have relocated the microbial community analyses (OTU composition, PCoA, and phylum-level composition) to the Supplementary Materials as Figure S1, S2, and S3, respectively. This modification serves dual purposes: (i) enhancing the focus of the main text on core findings, and (ii) reducing the cognitive burden associated with processing multiple complex figures.We sincerely hope that you can understand.

Comment: There are typo errors (e.g., Keywords propertiess, L628 may reflects). A thorough typo error check is recommended throughout.

Response: Thank you for catching these errors. We have corrected the typo in the Keywords ("propertiess" → "properties") and the grammatical error in the Discussion ("may reflects" → "may reflect"). We have also conducted a thorough proofreading of the entire manuscript to identify and correct any additional typographical or grammatical errors to ensure linguistic accuracy throughout.

Comment: L784-786: It looks MDPI template placeholder text, not your own data availability texts. Please, provide accession number in a public repository.

Response: We apologize for the oversight regarding the placeholder text. We have removed the MDPI template statement and replaced it with the following specific Data Availability Statement: "The data presented in this study are available on request from the corresponding author."

Comment: Figure 13 presents numerous pearson correlations without multiple testing corrections. Without correction, a substantial proportion of reported significant correlations may represent Type I errors. Please apply correction and revise accordingly.

Response: We have applied Benjamini-Hochberg FDR correction to all 253 Pearson correlations in Figure 10. Following this stringent correction, only nutrient availability-related correlations (e.g., Avail-P/Yield, UA/Yield) remained significant (q < 0.05), while phylum-level correlations with soil properties did not survive multiple testing correction. We have revised Section 3.4.2 to report only FDR-corrected significant results and updated the figure caption accordingly.

Comment: Figure 8 is unclear. strongly recommend to use a more informative and alternative for multi-group OTU intersection analysis.

Response: Thank you for this suggestion regarding the clarity of Figure 8. We have moved the OTU composition analysis to Supplementary Figure S1, where it is presented in higher resolution for readers requiring detailed taxonomic information. Specifically, we have replaced the original Figure 8 with a high-resolution version, now presented as Supplementary Figure S1 (Bacterial community OTU composition across treatments). We acknowledge that Venn diagrams have inherent limitations for multi-group comparisons. However, given that our revised manuscript now focuses the main text on the mechanistic SEM pathway (SOM → urease → yield) and functional guild analysis, we have retained the OTU intersection data in supplementary materials primarily for reference purposes, rather than as a central analytical figure.

Comment: Chloroflexi abundance increased by 62% under NBP, yet receives no mention in the Discussion. Given this phylum's roles in carbon cycling, aggregate formation, and anaerobic metabolism, its enrichment warrants mechanistic interpretation.

Response: Thank you for this astute observation. We have now added a dedicated mechanistic interpretation in Section 4.2 (lines 646-657) that addresses this phylum's functional roles.

Comment: L28-29: The Abstract states that "urease activity served as the primary pathway linking soil organic matter improvements to enhanced crop productivity." Given that the SEM explains only 26% of urease variance, this causal claim is overstated.

Response: Thank you for this important clarification. We have revised the Abstract to state that urease activity served as a "significant mechanistic pathway" rather than "the primary pathway."

Comment: L211-213: Please verify and standardize the units for all enzyme activities to ensure consistency and reproducibility.

Response: Thank you for this important suggestion regarding unit standardization. We have carefully revised the enzyme activity units in lines 373-379 as follows:

Phosphatase: Corrected "u g⁻¹" to "μg g⁻¹ h⁻¹" (micrograms per gram dry soil per hour), explicitly specifying the time dimension.

Catalase: Standardized as "mL g⁻¹ h⁻¹" (milliliters per gram dry soil per hour).

Regarding the standardization of units across different enzymes, we respectfully note that these units cannot be uniformly converted (e.g., to mg kg⁻¹ d⁻¹) without violating standard soil enzymology protocols. This is because different enzymes require different incubation periods based on their reaction kinetics:

Invertase/Urease: Measured over 24 hours (d⁻¹) due to slower reaction rates

Phosphatase: Measured over 1 hour (h⁻¹) as per standard p-nitrophenyl phosphate protocols

Catalase: Measured over 20 minutes (converted to h⁻¹) via permanganate titration

Converting all to "per day" would introduce artificial 24-fold errors for phosphatase and catalase, misrepresenting the actual enzymatic reaction rates measured in the laboratory. We have now ensured formatting consistency while retaining the methodologically appropriate units for each enzyme assay.

Comment: L396: The x-axis order of treatments in Figure 6 differs from the order used in all other figures.

Response: Thank you for catching this inconsistency. We have revised the Figure 6.

Comment: References: Several references do not follow a consistent citation format. All references must be reformatted to comply with the journal's author-year citation style before resubmission.

Response: Thank you for pointing out this formatting inconsistency. We have thoroughly reviewed and reformatted all references throughout the manuscript to ensure strict compliance with the journal's author-year citation style. Each reference has been carefully checked for uniformity in author name formatting, publication year placement, journal abbreviations, and volume/page numbering to meet the journal's specific requirements.

Reviewer 2 Report

Comments and Suggestions for Authors

Manuscript Agronomy 4180314

“ Effects of Combined Application of Nitrogen Fertilizer and Multiple

Soil Amendments on Soil Properties and Bacterial Community Structure”

The experiment was conducted in southern Xinjiang, a region in western China. The area is characterized by agricultural land bordering the Taklamakan Desert. The region's sandy soils are used to cultivate crops such as cotton, fruit and wheat, which require adequate irrigation. Consequently, the region has become the focus of studies on the management of sandy soils with high alkalinity, low fertility, poor water retention and a high risk of salinisation. The authors of the reviewed manuscript conducted a three-year experiment in an area where jujube orchards are cultivated. Six nitrogen fertiliser strategies were implemented, with the study focusing on changes in soil structure, enzyme activity, and microbial populations, as well as their impact on crop productivity. Structural equation modelling (SEM) was employed to analyse the relationships between soil amendments and crop productivity. The researchers concluded that the most effective strategy is to apply a mixture of biochar, bentonite and urea, increasing jujube orchard production.

I would like to highlight the following points in the manuscript:

1) The introduction is too long and could be shortened, as it contains facts that have already been extensively covered in published articles.

2) The keywords should be listed in alphabetical order.

3) A list of abbreviations should be included at the beginning of the manuscript.

4) Lines 115–120 are overly subjective and present observations that should be included in the 'Conclusions' section.

5) The description of the doses applied to the soil (lines 165–171) repeats information presented in Table 1. I suggest deleting this passage.

6) The choice of doses applied was not properly justified.

7) The appropriate references for the programmes used were not cited (lines 236–245).

8) The term soil should be presented with a lowercase initial letter (line 248).

9) In general, the English writing is good, However, there are passages that are overly long, which makes the reading experience less enjoyable. I cite the following excerpts as examples: lines 202 – 214, 225 – 234, 236 – 245, 249 – 264, 277 – 283, 289 – 298, 313 – 330, 338 – 360, 403 – 414, 443 – 458, 489 – 508, 517 – 532, 548 – 563, 572 – 590, 631 – 646.

10) Although the soil analysis results have been included in the supplementary material, I suggest attaching them to the text, as this would make easier to interpret the results.

11) The word “are” should be corrected (line 487).

12) I suggest that the analytical methods be presented in tabular form.

13) The sentence needs to be rewritten due to a subject-verb agreement error (lines 709 – 712).

14) What happened to the presence of heavy metals after the additives were applied?

15) What are the environmental consequences of adding the materials?

16) Although the authors obtained a considerable amount of data, they did not discuss the possible limitations of the experiment.

17) The manuscript does not consider the limitations to the applicability of the results in terms of both the cost of materials and the conditions of the region.

18) The conclusions are well-formulated and in line with the five objectives set out in the Introduction.

19) References 3, 4, 5 and 6 are unnecessary, as they contain content that is already well known.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

The English in the manuscript is fine, but as I said in the file above, it needs to undergo the improvements I pointed out.

Author Response

Dear Editor,

Thank you and best regards,

Yuxuan Wei and Yuyang Zhang (Email: wyx19982026@163.com;)

On behalf of all co-authors

Authors’ responses to Reviewers’ comments

Reviewer 2

Comment: The introduction is too long and could be shortened, as it contains facts that have already been extensively covered in published articles.

Response: Thank you for this suggestion. We agree that the original Introduction contained extensive background details on individual amendments that are well-documented in existing literature. We have shortened these sections by removing generic descriptive facts (e.g., specific percentage ranges for carbon sequestration or bulk density reduction) and consolidating the description of amendment mechanisms to focus on their complementary limitations and synergistic potential.(Lines 54–84)

Comment: The keywords should be listed in alphabetical order.

Response: Thank you for this formatting suggestion. We have reordered the keywords alphabetically as follows: arid zone agriculture; jujube yield; soil amendment; soil physical and chemical properties.(Lines 35–36)

Comment: A list of abbreviations should be included at the beginning of the manuscript.

Response: Thank you for this suggestion. We have added a comprehensive list of abbreviations at the beginning of the Materials and Methods section. (Line 123)

Comment: Lines 115–120 are overly subjective and present observations that should be included in the 'Conclusions' section.

Response: Thank you for this observation regarding the subjective tone of lines 115–120. We agree that interpretative observations should be reserved for the Conclusions section. In response to Reviewer #1's comments regarding mechanistic focus, we have already substantially revised this paragraph to remove subjective interpretations and maintain an objective, descriptive tone appropriate for the Results section. (Lines 84–102)

Comment: The description of the doses applied to the soil (lines 165–171) repeats information presented in Table 1. I suggest deleting this passage.

Response: Thank you for this suggestion to improve conciseness. We have deleted the repetitive description of amendment doses previously presented in lines 165–171. This information is now presented exclusively in Table 2, avoiding redundancy between the text and tabular data while ensuring all experimental details remain clearly accessible to readers.

Comment: The appropriate references for the programmes used were not cited (lines 236–245).

Response: Thank you for pointing out this omission. We have now added the appropriate references for all statistical software and computational packages used in the analysis.

Comment: The term soil should be presented with a lowercase initial letter (line 248).

Response: Revised. (line 227)

Comment: In general, the English writing is good, However, there are passages that are overly long, which makes the reading experience less enjoyable. I cite the following excerpts as examples: lines 202–214, 225–234, 236–245, 249–264, 277–283, 289–298, 313–330, 338–360, 403–414, 443–458, 489–508, 517–532, 548–563, 572–590, 631–646.

Response: Thank you for pointing out these specific passages that were overly long and affected readability. We have thoroughly revised all the cited sections to improve conciseness and flow:

For lines 202–214, 225–234, 236–245, 249–264, 277–283, 289–298, 313–330, 338–360, 403–414, and 631–646, we have streamlined the text by removing redundant transitional phrases, consolidating repetitive numerical descriptions, and focusing on key patterns rather than exhaustive data enumeration. Regarding lines 443–458, 489–508, 517–532, 548–563, and 572–590, these sections were already substantially condensed during our revisions addressing Reviewer #1's comments regarding mechanistic focus and peripheral analyses. The current versions in the revised manuscript reflect these prior streamlining efforts.

Comment: Although the soil analysis results have been included in the supplementary material, I suggest attaching them to the text, as this would make easier to interpret the results.

Response: Thank you for this suggestion regarding the placement of soil analysis data. After careful consideration, we have decided to retain the detailed soil physicochemical data in Supplementary Table S2 rather than moving it to the main text.

Our reasoning is as follows: The table contains a comprehensive matrix of 8 indicators across six treatment groups, which would occupy substantial manuscript space if included in the main text. This could disrupt the narrative flow of the Results section, where we aim to maintain focus on the core mechanistic pathway (SOM → urease → yield) without distracting readers with extensive tabular data.

We have ensured that key trends and significant findings from these data are explicitly described in the main text, allowing readers to grasp the primary conclusions efficiently. Meanwhile, the complete dataset remains fully accessible in the supplementary materials for those requiring detailed values.

We are so sorry and we hope this arrangement balances data transparency with readability, keeping the main text concise while ensuring all supporting data are available for verification.

Comment: The word “are” should be corrected (line 487).

Response: Thank you for catching this grammatical error. In response to Reviewer #1's suggestion to consolidate peripheral analyses in the supplementary materials, we have moved the relevant content to Supplementary Figure S3 (previously line 487 in the main text) and have corrected the word "are" to the appropriate form within the supplementary file to ensure grammatical accuracy.

Comment: I suggest that the analytical methods be presented in tabular form.

Response: Thank you for this formatting suggestion. After careful consideration, we have retained the narrative format for the analytical methods section rather than converting it to tabular form.

Our reasoning is that the text-based presentation allows for clear sequential description of methodological details (e.g., specific extraction conditions, instrument parameters, and quality control steps) that are essential for ensuring reproducibility. A tabular format might compress these critical details or require extensive footnotes, potentially reducing clarity. We are so sorry and truly hoping you can understand.

Comment: The sentence needs to be rewritten due to a subject-verb agreement error (lines 709 – 712).

Response: Thank you for pointing out this grammatical issue. We have corrected the subject-verb agreement error in lines 709–712. (lines 634– 636)

Comment: What happened to the presence of heavy metals after the additives were applied?

Response: Thank you for raising this important environmental safety concern. We clarify that heavy metal concentrations were not analyzed in this study, as our research focused specifically on soil nutrient cycling, microbial functionality, and crop productivity responses to amendments.

However, we acknowledge the relevance of heavy metal monitoring when applying soil amendments. The materials used in this study (bamboo-derived biochar, natural bentonite, and microbial inoculants) were agricultural-grade products with negligible heavy metal contents according to supplier specifications and previous safety assessments.

The total elemental analysis we performed (Table S2) focused on macronutrients and micronutrients (P, K, Na, Ca, Mg, Fe, Mn, Zn) rather than toxic heavy metals. Fe, Mn, and Zn detected in our analysis are essential plant micronutrients, not contaminants.

We agree that long-term monitoring of heavy metal dynamics would be valuable for comprehensive environmental risk assessment. We have now added a statement in the Discussion (Section 4.1) acknowledging this limitation: "While this study focused on fertility and productivity outcomes, future research should evaluate the long-term fate of potential contaminants, including heavy metals, when applying carbonaceous and mineral amendments in arid agroecosystems." (lines 574– 577)

Comment: What are the environmental consequences of adding the materials?

Response: Thank you for raising this important question regarding the broader environmental implications. While our study focused primarily on soil fertility mechanisms and crop productivity, the data allow us to infer several environmental consequences of the amendment applications:

Positive Environmental Outcomes:

Enhanced Water Conservation: The NBP treatment increased soil water retention capacity by 283% (Section 3.1.1), which significantly reduces irrigation requirements—a critical benefit for water-scarce desert agroecosystems where evaporation (2,824 mm) far exceeds precipitation (33.4 mm).
Salinity Alleviation: NB treatment reduced soil electrical conductivity by 67.3% (from 223 to 73 μS cm⁻¹), mitigating salt stress in inherently saline-alkali desert soils (Section 3.1.2). This improves soil health without chemical amendments.
Carbon Sequestration: Biochar-bentonite co-application increased soil organic matter by 30% (Section 3.1.4) and total carbon stocks by 22.9%, contributing to long-term carbon storage in marginal lands.
Improved Soil Physical Structure: Significant reductions in bulk density (8.2%) and increased porosity (54.21%) enhance soil aeration and reduce compaction (Section 3.1.1).

Potential Trade-offs and Risks Identified:

Nitrogen Management: WhileNB elevated the C/N ratio to 7.14 (Section 3.1.3), suggesting potential temporary N immobilization, this was mitigated in the NBP combination which maintained balanced nutrient stoichiometry.
Anaerobic Microsites: We acknowledge that enhanced water retention from bentonite may create localized anaerobic conditions potentially promoting denitrification and N₂O emissions (Section 4.2). This represents a trade-off requiring optimization of irrigation management to balance water conservation with greenhouse gas considerations.
Sodium Enrichment: Bentonite-containing treatments increased total sodium concentrations (up to 162% in NBP, Section 3.1.5). However, this did not translate to increased salinity (EC remained reduced) due to concurrent improved leaching and water retention, suggesting the sodium remained in non-toxic forms or was effectively managed by the biochar-bentonite matrix.

Our three-year study focused on fertility mechanisms rather than comprehensive environmental risk assessment. We did not directly quantify greenhouse gas emissions (N₂O, CH₄) or heavy metal dynamics. We agree that long-term monitoring of these parameters would be essential for a complete environmental impact assessment, particularly regarding the denitrification potential under enhanced water retention conditions.

Comment: Although the authors obtained a considerable amount of data, they did not discuss the possible limitations of the experiment.

Response: Thank you for this suggestion. We have added a concise acknowledgment of limitations in the final sentence of the Conclusion.(lines 699–703)

Comment: The manuscript does not consider the limitations to the applicability of the results in terms of both the cost of materials and the conditions of the region.

Response: Thank you for raising these important practical constraints. We acknowledge that material costs (particularly for biochar and bentonite at field-scale application rates) and regional specificity (Southern Xinjiang's unique desert brown soils and extreme continental climate) represent significant limitations to the immediate commercial applicability of our findings. However, this study was designed primarily as a mechanistic investigation to elucidate the causal pathway (SOM→urease activity→yield) underlying amendment effects, rather than an economic feasibility assessment. While the specific cost-benefit ratios will indeed vary based on local material availability and labor costs, the identified mechanism—enzyme-mediated nitrogen mineralization as the rate-limiting bottleneck in N-limited desert soils—provides a theoretical framework applicable to similar arid agroecosystems globally.

We respectfully sorry and suggest that detailed economic analysis and multi-regional validation, while valuable, fall beyond the scope of this mechanistic proof-of-concept study, though we recognize these as critical next steps toward practical implementation.

Comment: The conclusions are well-formulated and in line with the five objectives set out in the Introduction.

Response: Thank you for this positive feedback.

Comment: References 3, 4, 5 and 6 are unnecessary, as they contain content that is already well known.

Response: Revised.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The author has adequately addressed my concerns. I support acceptance of this manuscript.

Author Response

Comment 1:The author has adequately addressed my concerns. I support acceptance of this manuscript.

Response: Revised.

Reviewer 2 Report

Comments and Suggestions for Authors

Please note that the words “field” and “yeld” in lines 144 and 167 should be written with lowercase initial letters.

Author Response

Comment1 :Please note that the words “field” and “yeld” in lines 144 and 167 should be written with lowercase initial letters.

Response: Revised. (Lines: 144 and Lines: 167 )