Referee #1
The authors have modified the manuscript according to the comments from reviewers, I recommend publication after the following minor revisions.
Please inter-compare the MODIS observation with some results of active remote sensing or model simulation and add some other remote sounding results in China.
1. Huang, J., Minnis, P., Yi, Y., Tang, Q., Wang, X., Hu, Y., Liu, Z., Ayers, K., Trepte, C., and Winker, D., 2007. Summer dust aerosols detected from CALIPSO over the Tibetan Plateau. Geophys. Res. Lett. 34(18), L18805. doi:10.1029/2007gl029938.
2. Jia R., Y. Liu, B. Chen, Z. Zhang, J. Huang, 2015: Source and transportation of summer dust over the Tibetan Plateau. Atmospheric Environment, 123(2015), 210–219, doi:10.1016/j.atmosenv.2015.10.038.
3. Liu Y., Y. Sato, R. Jia, Y. Xie, J. Huang, and T. Nakajima, 2015: Modeling study on the transport of summer dust and anthropogenic aerosols over the Tibetan Plateau. Atmospheric Chemistry and Physics, 15(21), 12581–12594, doi:10.5194/acp-15-12581-2015.

Referee #2
---

Referee #3
The new MAIAC algorithm provided globally higher spatial resolution MODIS-based AODs at 1 km than DT and DB algorithms, thus the validation and comparison of these AODs from different algorithms are urgent to make at global and even regional scare such as South American, North American, and China, as listed by authors. In this paper, the authors also presented a comprehensive assessment and comparison of AOD from different algorithms in China using the measurements of AERONET. The updated version of the manuscript have been polished well in English language, and also responded to queries and fixed the existing errors from anonymous reviews. However, if the following suggestions or comments were specified in the context, it would be a comprehensive, objective and impartial evaluation and comparison to AOD from different algorithms. I suggest the manuscript should be accepted.

1. Page 1, Line 23, “Yunan Province” should be revised as “Yunnan Province”.

2. In the title of Figure 2, how did the AODs divide into 50 bins, please specify it for future readers.

3. In Figure 13, 14, and 16, the sudden change of AODs distribution related to MAIAC at the 30°N can be seen clearly. It doesn’t make sense. I guess that there seems to be something wrong in the data processing or whatever, please check the MAIAC dataset and specify it in the context.

4. The reference of Matins et al. (2017), mainly focused on validation of MAIAC product over South America, was cited frequently in this paper. Such as, the selection of spatial and temporal window, the definition of land cover type of AERONET sites, the strategy of comparison of retrieval accuracy and so on. Furthermore, All these seem to imply us that 1) the method of Matins et al. (2017) is superior or the most reasonably, 2) the difference between this paper and Matins et al. (2017) is only at location of the area of interest. So the authors should make a reasonable evaluation to the reference in the context. Furthermore, the reference of Zhang et al. (2019) in Page 2 Line 25 should be explained the relative simplicity of the assessment methods with more words, which will further demonstrate the comprehensive of the methodology used by the authors.

5. Page 10, Line 5, the different definition of EE envelope can be found in different reference, how did you consider it and please specify it in the context.