general comments

This study, by integrating multiple existing high-resolution remote sensing products, achieved a high spatiotemporal accuracy accounting of the water footprint for major crops such as wheat, maize, rice, soybeans, and potatoes from 2001 to 2020. The overall writing logic of the paper is clear, and the scientific question is well-articulated. The research results hold certain value for data application in fields like agricultural water resource management. However, there are still deficiencies in the study's innovation, methodological explanation, and result description, which require further revision and improvement.

specific comments

1. Introduction: Are the daily-scale accounting results based entirely on daily-scale remote sensing products? And given that there are many studies on crop water footprint accounting, what are the other novel aspects of ChinaCropWF besides its daily resolution?
2. Methods: please check the equations in Figure 1, which should be consistent with those in 2.2.1
3. Methods: for wheat, if both spring and winter wheat were considered at the same time?
4. Methods: How were evaporation and transpiration amounts calculated? They are not presented in the uploaded dataset.
5. Results: "3.1 Spatiotemporal evolution patterns of crop water footprint", why does potato show large interannual variations, with consecutive years being relatively similar?
6. Results: Can the production water footprint go beyond the county scale to obtain higher-precision spatial data?
7. Discussion: "4.2.1 Phenology impacts on water footprint" compares different phenological period product datasets. However, this section merely provides a simple comparison of accounting results and does not demonstrate the advantages of daily-scale accounting. Further analysis is recommended.

technical corrections

1. "2.1.4 Crop planting area" should refer to the crop phenology period.
2. “Author contributions: XW and EH the research” means “EH and XW wrote the original draft”?

This manuscript presents ChinaCropWF, a novel, high-resolution (1 km, daily) gridded dataset quantifying the blue and green water footprints of China's five major food crops (rice, maize, wheat, soybean, and potato) over two decades (2001–2020). The dataset is generated by integrating multi-source remote sensing products (precipitation, evapotranspiration, crop distribution, and phenology) within a soil water balance framework. The topic is highly relevant to the scope of Earth System Science Data, addressing critical gaps in agricultural water-use data. The construction of a daily-scale, 1-km resolution dataset represents a significant methodological advancement over existing monthly or growing-season products. The manuscript is well-structured, the methodology is sound, and the initial validation against field data and comparison with existing datasets demonstrate the product's value. However, several aspects require clarification and improvement before final publication to enhance the manuscript's clarity, robustness, and impact.

Specific Comments

1. The manuscript mentions using a "soil water balance method" to account for soil water variations (ΔS) and improve accuracy. However, the description is somewhat brief. Please provide a more detailed explanation of how ΔS (both blue and green components) is calculated and integrated daily. How is the soil profile characterized? What are the assumptions regarding runoff and deep percolation? A schematic diagram of the daily soil water balance model would be beneficial for readers.
2. The authors acknowledge a key limitation: assuming a fixed relative soil moisture of 75% for non-rice crops to estimate initial soil water content. This is a significant simplification that could impact the accuracy of the water footprint, especially in water-stressed or highly variable environments. It is better to add a simple sensitivity analysis in the supplementary materials to show how varying this initial moisture assumption might affect the final water footprint estimates for a sample region or crop? This would provide valuable context for users of the dataset.
3. In Figure 4(f), there is an abrupt change from 2015 to 2016. It is necessary to explain the causes, i.e. is it an actual occurrence, or is it caused by data uncertainty?
4. The authors stated “As of 2024, these crops were cultivated on a total of 119.3 Mha in China, with maize, rice, wheat, soybean, and potato accounting for 44.7, 29.0, 23.6, 10.3, and 3.2 Mha, respectively, collectively representing 92.9% of the total sown area.” When calculating the area proportion (92.9%), it is necessary to consider the multiple cropping scenarios of these five major crops.
5. Lines 242-244: a reference should be given for the explanation “evaporation peaked during 2002-2005, coinciding with a weakened East Asian summer monsoon”.
6. Line 15: The ranking of total water footprints is given as "rice > maize > wheat > soybean > potato". Consider adding a brief note on why potato's range is so wide (0.15-11.31 Gm³) – is it due to data limitations or actual variability?
7. Line 350-355: In addition to remote sensing uncertainties, consider briefly mentioning the uncertainty introduced by using different data sources for crop planting areas for different crops and how this might affect the consistency of the final product.
8. The meanings of some abbreviations are unclear, such as Rice-LR, Rice-SR&ER. In addition, figures need to be self-explanatory; some figures require more detailed explanations, such as Figures 3 and 4.