Comment on acp-2021-139

This study investigates the impact of black carbon (BC) aerosols on the dynamics of the planetary boundary layer (PBL) and aerosol-PBL feedback mechanism in wintertime Beijing. The study applies a large-eddy simulation (LES) model coupled with a radiative transfer and an aerosol model (SALSA). The influence of BC concentrations is investigated from different angles and the results of different sensitivity studies support each other. The method and research questions are novel and overall, the study has been designed and conducted well. However, some points require further attention before publication.

This study investigates the impact of black carbon (BC) aerosols on the dynamics of the planetary boundary layer (PBL) and aerosol-PBL feedback mechanism in wintertime Beijing. The study applies a large-eddy simulation (LES) model coupled with a radiative transfer and an aerosol model (SALSA). The influence of BC concentrations is investigated from different angles and the results of different sensitivity studies support each other. The method and research questions are novel and overall, the study has been designed and conducted well. However, some points require further attention before publication.

General comments:
As mentioned above, the study is well planned and conducted. However, it seems that finalizing the manuscript lacks some effort. For instance, sometimes sections and figures are incorrectly referred to and the language does not always sound professional. I would also double-check the usage of articles (a/an and the). In this study, you investigated three different kinds of model setups described in Sections 2.3-2.5. Each setup investigates different model sensitivities. You end up using the word "case" a lot. For instance, you use the word "case" for different setups (Sections 2.3-2.5), but you also use it e.g., in Table 2 ("BC case and No BC case"). Maybe you could try to come up with some more indicate words to make it easier to follow the text? For example, scenario, sensitivity, simulation etc. Different "cases" (i.e., setups described in Sections 2.3-2.5) could also have some more indicative names, e.g., case_aerosol_loading, case_met and case_BC_loading. The Discussion section now contains conclusions and most of the content in Conclusions should be moved to Discussions. Please review the content of these sections. For instance, Conclusions should not introduce any new arguments, while now the novel mechanism is presented for the first time there. Regarding Conclusions, the research questions should be restated there. Limitations of the study have not been discussed anywhere. Applying LES in this type of study is very novel and I think you should stress more in the text. Furthermore, visualising the simulation setup would be very useful for the readers. Now you are only showing one-dimensional vertical profiles while LES resolves the three-dimensional flow and concentration fields. P2 L59-60: This phrase is partly a repetition of the first phrase of the previous paragraph. P2 L72: "through synoptic scale winds" --> "by synoptic-scale winds"? P3 L83: remove "a" before "polluted" P4 L90: The dome effect is being discussed a lot in the introduction but not later in the manuscript.
Section 2.1: I would mention somewhere that LES resolves the three-dimensional turbulent field of wind and scalar concentrations and that it directly resolves most of the energy and parametrises only the smallest scales.
P4 L102: I would not use the expression "is based on". These are just numerical methods applied in UCLALES to resolve the flow and dispersion.
P4 104: I would open "doubly periodic" to "periodic in both horizontal directions" P4 L104-105: the surface scheme for moisture and heat, right? P4 L104-107: I think the urban surface scheme I rather essential, so I suggest you give some more information about it here. Especially when Slater et al. (2020) is not an openaccess publication.