Review comments on “Aerosol single scattering albedo derived by merging OMI/POLDER satellite products and AERONET ground observations” by Dong et al. submitted to the journal Earth System Science Data.

##################################################

General comments:

This study by Dong et al. uses Ensemble Kalman Filter (EnKF) data synergy technique to construct Merged-OMI and Merged-POLDER datasets based on SSA products of AERONET, OMI and POLDER. At the same time, the results verify the effectiveness of EnKF technology in extending the information obtained from the ground station to a larger area. The constructed SSA global dataset can provide data support for global aerosol radiative forcing assessment and climate change detection. Overall, this manuscript is clearly written, and the topic is suitable for the journal. I would support for publication in the journal of Earth System Science Data after some corrections and clarifications.

1. Following the ESSD requirement, the author need mention the generated dataset with register doi in your abstract. And in the main text, some general description about the data structure is mandatory.

1. General question about how you make averaging to get monthly/annual mean SSA from OMI, POLDER and AERONET? SSA is a relative value and always provided with certain AOD threshold, or its uncertainty strongly depends on the AOD levels. Therefore, a direct value average seems meaningless to me, you need average aerosol absorption and extinction, then covert to averaged SSA.

1. Why not merge OMI, POLDER and AERONET into one single global SSA datasets?

##################################################

Specific comments:

Abstract: in order to follow the ESSD requirement, you should mention the generated dataset with register doi in your abstract.

P2 L55: I would draw your attention about two recent and studies related to your discussions by Chen et al. (2020) and Schutgens et al. (2021) about aerosol absorption evaluation and intercomparison between satellite products.

Chen, C., Dubovik, O., Schuster, G.L. et al. Multi-angular polarimetric remote sensing to pinpoint global aerosol absorption and direct radiative forcing. Nat Commun 13, 7459 (2022). https://doi.org/10.1038/s41467-022-35147-y

Schutgens, N., Dubovik, O., Hasekamp, O., Torres, O., Jethva, H., Leonard, P. J. T., Litvinov, P., Redemann, J., Shinozuka, Y., de Leeuw, G., Kinne, S., Popp, T., Schulz, M., and Stier, P.: AEROCOM and AEROSAT AAOD and SSA study – Part 1: Evaluation and intercomparison of satellite measurements, Atmos. Chem. Phys., 21, 6895–6917, https://doi.org/10.5194/acp-21-6895-2021, 2021.

P3 L71: why not merge OMI, POLDER and AERONET into one single global SSA datasets?

P4 L95: could you please explicit how you do average SSA? SSA is a relative value, therefor a direct value average seems meaningless. In addition, SSA is usually provided with certain AOD threshold.

L105: could you please provide proper reference to it?

Section 2.3: if I understood, is it 1D data assimilation? You correct only the grids with ground-based measurements, right?

Line 225: Figure 3 shows that the RMSE of SSA/original-OMI and SSA/AERONET is 0.036, while the RMSE on this figure is 0.035.

Figure 4: there are less effects on Sahara Desert w/wo data assimilation, both Original and Merged-OMI seem underestimate SSA there, could you elaborate more about it?

Figure 6: in most cases/sites, R is improved after merging, while Bias and RMSE could become worse in few sites (6h and 6i), could you explain why is that?

L308: there is a known issue for POLDER measurements at 440 nm.

Section 3.4: since you treat 4 channels (440, 670, 865, and 1020 nm) SSA for POLDER, it would be interesting you check or evaluate the SSA spectral dependence w/wo assimilation.

Figure 17: could you provide the same figure for merged product?

Aerosol single scattering albedo (SSA) is a critical parameter for characterizing aerosol scattering/absorption properties. However, obtaining accurate long-term global SSA datasets remains challenging due to limitations and uncertainties in satellite-based measurements. This manuscript constructs two long-term monthly mean SSA datasets over land by synergizing satellite observations (from OMI and POLDER) and ground-based AERONET measurements through the Ensemble Kalman Filter (EnKF) technique. The datasets provided in this study could improve the estimation of aerosol radiative effects and contribute meaningfully to climate change assessments. I would like to recommend the publication of this manuscript as the following issues are addressed.

General comments:

1. Could you please provide the corresponding AOD data as a quality control reference in your dataset? As you are aware, the uncertainty in the merged SSA product is closely related to AOD. Providing AOD data can help users implement effective quality control when utilizing the SSA dataset.
2. Did you derive the monthly mean SSA from the original daily satellite products? If so, how was the monthly average calculated, and were any quality control measures applied during the process? These details should be clearly described in the data section.
3. Would it be possible to include the merged SSA results at 550 nm based on POLDER? It is commonly used in both satellite retrievals and model applications, which would enhance the usability and comparability of the dataset.
4. Why is the spatial resolution of the merged dataset set to 1°? The SSA background field can vary significantly within a 1° grid cell, which is also a key reason for the limited representativeness of ground-based observations. The dataset at a finer resolution, such as 0.5°, could potentially improve the effectiveness and applicability of ground-based observations.

Specific comments:

1. Please specify the start and end months for 231/106 members of the ensemble in Section 2.1.
2. For the global ensemble dataset, each grid cell may contain different numbers of effective members. Could it be that some grid cells have too few members? Could you provide the distribution of effective member counts across the global grid cells?
3. Figure 6: The title appears to be incorrect, which should refer to the analysis based on OMI.