This study generates a decade-long (2011-2021) high-resolution aerosol dataset for East Asia using GOCI satellite observations and radiative transfer modeling. The dataset provides robust hourly aerosol optical properties, including aerosol optical depth (AOD), fine-mode fraction (FMF), single-scattering albedo (SSA), Ångström exponent (AE), and aerosol classification. These products are particularly valuable for climate and environmental research communities in improving weather forecasting and air pollution monitoring. While the dataset meets ESSD standards for long-term aerosol records, the manuscript requires major revisions to address some comments before publication.

General comments:

The GOCI satellite’s native 500 m spatial resolution provides unique advantages over other geostationary satellites (e.g., Himawari-8/9, MSG-R, GOES-R) for aerosol monitoring at finer scales. However, the decision to upscale to 2 km resolution - while improving retrieval robustness through pixel grouping - potentially diminishes this competitive advantage. The comparative analysis presented in the study actually demonstrates the superior capability of higher-resolution observations, making the 2 km resolution choice appear scientifically questionable.

While AOD validation is thoroughly presented, the derived products (FMF, SSA, AE) lack equivalent validation despite their scientific importance. These parameters should receive proper quantitative evaluation given their utility in aerosol characterization.

The figures and tables require significant improvement.

Specific comments:

1. Line 95, the advantage of GOCI, including its visible spectrum, long-term observation records with stability, and the presence of a successor satellite, is not fully shown. The developed retrieval algorithm uses only limited bands rather than the full spectrum. Compared to GOCI, the MSG satellite series have longer observation records and multiple successor satellites.
2. Line 125, how is the cloud threshold determined? Only three tests are used—is this too few? And how is the water detection threshold used? Confirm whether they align with the operational algorithm’s implementation.
3. Lines 140-145, the minimum reflectance technique often leads to underestimation of surface reflectance since no day can truly be aerosol-free, even if it is very clean.
4. Line 155, define and explain the look-up table (LUT) methodology for readers who may be unfamiliar with this technique, despite its common use in the field.
5. Figure 4, why are the gaps different between the two?
6. Line 255, the uncertainty of AERONET AOD should be updated based on the latest studies.
7. Figure 6, why are only R, EE, and MBE shown?
8. Line 320, the time-series AOD changes (Fig. 9) should be explained in more detail, as they appear very interesting, particularly the intra-annual variations.
9. I encountered difficulties accessing the dataset (https://doi.org/10.7910/DVN/WWLI4W) during my review. Could this be due to regional access restrictions in China, or is there another technical issue with the data repository?

Technical corrections:

1. Table 1, suggest to improve by maintaining consistent line spacing throughout all paragraphs in the table.
2. Figure 1, the city legend is missing and should be included.
3. Figure 3 requires descriptive titles and proper labels for each subplot.
4. Figure 4, a legend identifying the sites should be provided, and the figure appears excessively long and could benefit from resizing or reorganization.
5. Line 250, “to” -> “against”.
6. Line 260, first “;”-> “:”.

This manuscript presents a valuable decadal, hourly, high-resolution (2 km) aerosol optical properties (AOPs) dataset for East Asia, derived from GOCI satellite observations and the Yonsei Aerosol Retrieval Algorithm (YAER). The dataset covers 2011–2021 and includes AOD, fine mode fraction (FMF), single scattering albedo (SSA), Ångström exponent (AE), and aerosol type. The paper highlights improvements in spatial/temporal coverage and retrievals over challenging environments (clouds, coastlines) and provides validation against AERONET. Such a dataset will undoubtedly benefit climate, air quality, and health research in the region.

The manuscript is generally well-structured, and the methods and results are clearly described. However, several issues require attention before publication.

Major Comments:

1. The manuscript notes that FMF, SSA, AE, and aerosol type are ancillary and “recommended for qualitative or interpretive use” due to uncertainty. However, the practical scientific utility of the dataset would be significantly enhanced by providing quantitative validation and uncertainty characterization for these variables (not just for AOD). Please include additional validation results for FMF, SSA, AE, and aerosol type where possible (even if only for selected periods/sites with ground truth), and discuss sources of error and their implications for users.
2. The authors describe an advanced cloud detection and removal scheme but do not provide a systematic assessment of residual cloud contamination, which is known to bias AOD retrievals, especially at high resolution. It would be valuable to compare the performance of the new cloud screening to that of standard (operational) products.
3. While the validation of AOD against AERONET is comprehensive, the data show persistent underestimation at high AOD (>1.0) and overestimation over turbid water and sparsely vegetated land. The current discussion is brief. Expand the discussion on possible causes for these retrieval biases, such as surface reflectance a priori selection, aerosol model assumptions, or radiative transfer LUT limitations. Suggest potential mitigation strategies, or at least clarify limitations for high-AOD and complex surface regimes.
4. While the manuscript references related studies and datasets (e.g., MODIS, VIIRS, GEMS), a direct comparison with other existing satellite AOD products over the same region and period (where possible) would further contextualize the strengths and weaknesses of the new GOCI dataset. It should provide a comparative assessment (statistical metrics, spatial patterns, or time series) between your dataset and other satellite or reanalysis products (e.g., MODIS MAIAC, VIIRS DB, GEMS) to demonstrate added value and highlight specific improvements or tradeoffs.

Minor Comments:

1. Please clarify how gaps due to clouds, sun-glint, or other data removal are handled (e.g., are missing values flagged, interpolated, or left as NaN?).
2. Improve the clarity of several figures (e.g., Figures 2–4): add scale bars, colorbars, and clearer labeling for readers less familiar with the region.