This is an ambitious and important study that, with revisions, should be published.

As a soil ecologist focusing on soil carbon–nutrient dynamics and their interactions under environmental and management changes, I found the work both impressive and valuable. Using comprehensive, high-resolution global datasets, the authors employ spatiotemporal machine learning approaches to map soil organic carbon (SOC), soil pH, soil type, and other variables at a remarkable 30 m resolution worldwide. They also estimate SOC changes over the past two decades. The harmonized legacy soil observations are particularly noteworthy, and the resulting maps will be valuable for diverse applications, such as driving soil carbon models, informing land management, and guiding policy decisions.

That said, there are several areas—both scientific and presentational—where improvements would substantially enhance the manuscript’s rigor, clarity, and impact.

The manuscript uses an excessive number of abbreviations, which interrupts reading flow. Some abbreviations are not defined at first mention—for example, CCC and RMSE in the abstract. The authors should not assume all readers will be familiar with these terms. I recommend carefully reviewing the manuscript to ensure that all abbreviations are defined upon first appearance and that only essential abbreviations are retained. Given the already considerable length of the manuscript, the use of numerous abbreviations does not meaningfully reduce length and may hinder comprehension.

The manuscript is long enough to deter some potential readers. If journal policy allows, I recommend moving certain sections—such as extended details on data collection, preparation, modeling approach, and mapping techniques—into the Supplementary Information. This would allow the main text to focus more on:

• Implications and potential applications of the dataset
• Interpretation of key findings (e.g., variable importance, spatial patterns)
• Broader impacts and future directions

While complexity is sometimes unavoidable, figures should be as simple, clear, and self-explanatory as possible. Specific suggestions:

Figure 1: Reorganize to emphasize the workflow; remove unnecessary logos. Use a consistent layout style (either top-down or left-right) with clearly separated blocks for each step.

Figure 2: The content can be succinctly described in a few sentences; consider removing the figure or replacing it with more impactful visuals.

Figure 5: Overly complex and difficult to interpret; despite repeated attempts, I could not fully understand it.

Figure 6: Contains too much information without adequate caption detail. Consider showing only the left panel with block diagrams; integrate textual explanations, interpretations, and distribution plots into the main text or Methods section.

Scientific Comments:

Motivation for 30 m resolution. The rationale for mapping at 30 m resolution should be articulated more clearly. Higher resolution should serve a clear theoretical or practical purpose—for example, improving global SOC stock estimates, supporting fine-scale land management, or providing critical input to Earth system models. The current introduction touches on these but could better synthesize them into a concise, logically connected research objective. The four research questions presented are somewhat disjointed; consider distilling them into a single, coherent framework.

Multicollinearity of predictor variables. The manuscript does not address multicollinearity among predictors—a significant concern for high-dimensional datasets, especially with overlapping climatic and remote sensing variables (e.g., SAVI, NDVI, GPP, and climate metrics). Multicollinearity can cause overfitting and obscure variable importance. The authors should clarify whether collinearity was assessed or controlled, and if not, explain the rationale. Reducing redundancy could also decrease computation time and improve model interpretability.

Inclusion of bedrock depth. Bedrock depth is a critical factor influencing SOC stocks, particularly in mountainous regions where bedrock often occurs at shallow depths (<1 m). While the authors mention future inclusion, I suggest considering it now—global bedrock depth maps do exist and could be integrated relatively easily. Bedrock depth affects root distribution and carbon inputs to the soil, potentially altering model performance and the relative importance of predictors.

Overall, this is an excellent and timely study with strong potential impact. Addressing the concerns outlined above—particularly improving structure, clarifying motivations, simplifying figures, and addressing certain methodological points—will greatly strengthen the manuscript’s readability and scientific contribution.

This manuscript represents an important contribution to the field of fine-resolution global digital soil mapping. The methodology and its application potential are commendable, and the work is generally of publishable quality. The paper can be significantly strengthened by addressing several key issues related to the methodological description, which currently lacks sufficient detail, and by improving the consistency of writing and terminology throughout the text.

Specific Comments:

Please add continuous line numbers throughout the manuscript. This will greatly facilitate referencing specific locations during the review and revision process.

Abstract needs better clarification. Please provide the abbreviation of SOC in P1 Line 5. It is not clear about the spatiotemporal Machine learning. How time is incorporated into this model? Is it a 3D+T model? Why using 68% probability for quantifying the prediction uncertainty? Soil carbon density (P1 Line 9) and soil carbon (P1 Line 11) should be corrected as soil organic carbon density and soil organic carbon due to the presence of soil inorganic carbon in many soil samples. It is reasonable that authors did not consider the temporal changes of soil texture fractions and soil types, while bulk density is highly correlated to soil organic carbon and therefore its temporal changes should be considered if you taking 5-year time intervals for soil organic carbon. Please provide the full names of RMSE and CCC for the first time. It is not necessary to indicate the RMSE in log-scale since this information is useless. It is not clear why authors only present the most important variables for soil organic carbon density and pH.

P3 Lines 5-7: This work did not solve this issue during predictive modelling.

P3 Lines 21-23: Authors overlooked the maps from Australia. It would be better to include the work by Grundy et al. (2015). Grundy M.J., Rossel, R.V., Searle, R.D., Wilson, P.L., Chen, C. and Gregory, L.J., 2015. Soil and landscape grid of Australia. Soil Research, 53(8), pp.835-844.

P4 Line 29: This work covers the period from 2000 to 2022+, so how to evaluate the impact of land use conversion for SOC and pH on a scale of 25+ years?

P5 Line 20: the difference between observations and measurements should be better clarified.

P5 Line 21: Once you match the O&M with covariate layers by year, it means that you overlook the legacy effect from environmental covariates (e.g., land use change), which can be quite important for soil properties, such as soil organic carbon. I understand that this consideration would pose a heavy load for computing, but at least you should address this limitation in the discussion.

P5 Lines 25-26: It is not clear that whether authors included the profile issue in performance evaluation to avoid data leakage.

P7 Line 26: Please correct SOC as SOCD or SOCd. Indeed, the full name should be specified in P7 Line 22.

P8 Line 4: in t m^-2?

P10 Line 5: A recent released dataset from Chen et al. (2025) may be helpful for your future work. Chen, Z., Chen, L., Lu, R. et al. A national soil organic carbon density dataset (2010–2024) in China. Sci Data 12, 1480 (2025). https://doi.org/10.1038/s41597-025-05863-3

Figure 3: Please use either SOCD or SOCd in the manuscript.

P15 Lines 28-30: What is the advantage for estimating SOC density directly from SOC content? SOC together with other variable control the variability of bulk density, only take SOC to estimate SOC will limit the prediction accuracy. SOC [kg m^-3] should be corrected as SOC density [kg m^-3].

P16 Lines 11-12: variable resolutions in 1 kilometer resolution? Scale is not an appropriate term here.

P17 Line 15: Since it is a 3D+T model, it is important to demonstrate the time span of soil data to support spatiotemporal modelling.

P24 Lines 14-25: Please also include R² in the accuracy evaluation. Why not report the accuracy of silt content here?

Figure 11: It would be also interesting to demonstrate the difference of model performance across different continents, which would be helpful for the design of future direction.

P53 Line 11: assessment indicates that (R2) the best achievable. R2 is a typo here?