See attached PDF file for peer review comments.

It is very valuable to estimate the crop-specific fertilizer in China. However, there are some serious problems in the process of accounting for crop-specific fertilizer, so I question the quality of the final data product. 

1. For the fragmented crop distribution in most parts of southern China, this paper calculated the crop-specific fertilizer application rate from 1952 to 2018 based on the cropland area data with a resolution of 5 km produced by provincial data and just five years' crop rotation data. I have serious doubts about the accuracy of the data.

2.   When calculating crop-specific fertilizer application, the article mentioned "The N fertilizers use rate for each major crop types (except other crops) was intermittently reported in the Cost-benefit Report of the National Agricultural Products (CBR) covering the period of 2004-2018 (Table 1)". However, there is no corresponding crop fertilizer allocation table in the text or supporting materials, nor is there a link to the data source. I entered to the CBR website to check, but did not get the corresponding data. And this part of data is very critical, which directly affects the accuracy of the final product. Moreover, so-called high-resolution data, based only on provincial rates of crop fertiliser allocation, are crude.

All in all, I don't think this article has enough innovation or contribution in terms of data source and data production method for publishing in the ESSD. The final data product was only qualified in terms of the average cropland fertilization, but the crop-specific fertilizer data was very crude, which greatly weakened the use value of the data. To sum up, I suggest rejecting the manuscript.

This work reconstructed a historical, annual N fertilizers use dataset in China at 5 km × 5 km resolution covering the period of 1952 to 2018 by integrating improved cropland maps. The dataset is useful for many purposes. Generally the paper is well-written, I have several concerns for authors to improve it before accepting for pubilication.

1) The Figure 1. Methodology flowchart of nitrogen (N) fertilizer map reconstruction  is not detailed enough to illustrate the methods used. Need to be improved by inclsuign the major methods used.

2)  Are the spatial pattern of N fertilizers use in each of the major crop types planted in China available? Why not resported in the main text? Are the data are availbale?

3) It is good that the authors compare the datasets wiith previous datasets. It is ture that the newly constrcued datasets are different but may need evidence that this data is more robust.