This manuscript aims to evaluate hectometric-scale modeling of the urban mixing layer using an extensive ceilometer network. Specifically, it seeks to assess the model’s ability to determine the urban ABL, focusing on the mixing layer height (MLH), by comparing simulated aerosol vertical profiles with ceilometer measurements using the same retrieval algorithm. The model’s approach to determining ABL/MLH turbulent mixing and its relationship to aerosol vertical distribution is not adequately discussed. Therefore, it remains uncertain whether this evaluation is truly useful in evaluating the urban mixed layer. The study then assesses the model’s ability to capture the spatial distribution of MLH and evaluates the urban smoke plume for two case study days. These results suggest that the model still does not accurately represent land characteristics, which is an important and interesting finding, supported by previous research. Additionally, the impact of this misrepresentation on urban plume transport is an important result, and further insights on this should be included. The manuscript would benefit from a more thorough discussion of the implications of these findings, explaining how these models are to be implemented and how the paper's findings may influence that implementation. Overall, the reviewer recommends publication after addressing major revisions.

Major Comments: 

• The manuscript demonstrates an inconsistent use of terminology regarding the mixing layer height (MLH) and its relationship to the turbulent or thermodynamic ABL and aerosol distribution. While acknowledging that aerosol-based MLH may differ from thermodynamic MLH, the manuscript frequently equates them or uses them interchangeably. It is crucial to consistently differentiate between these terms.
• Similarly, the authors should carefully consider the distinction between "mixed" (e.g. L70) and "mixing" (e.g. L84) layers. The term "mixed layer" hints at a final, fully-mixed state, while the "mixing layer" generally describes the convective daytime ABL in the context of continuous mixing. Ensure clear and consistent usage of these terms throughout the manuscript. 
• The term “aerosol layer height” may be more accurate than “mixed layer height” in this context, especially considering that, as shown in Figure 4 for instance (afternoon decay period), turbulence is no longer the dominant driver in the model. This suggests that the term “mixed layer” may not be appropriate when discussing the ZM or ceilometer-based ZO retrievals. When addressing the links to a thermodynamic ABL, then using the BLD would be more appropriate.  
• The manuscript mentions the use of a turbulence scheme but does not explain how it influences the simulation of ABL nor how a BLD is estimated. It is important to clarify how the model defines the vertical extent of surface aerosols and how the turbulence scheme interacts with this vertical distribution in order to better assess and evaluate the model, which is a stated objective of this study. 
• The manuscript lacks a detailed explanation of the aerosol prediction methods and aerosol scheme. A more thorough description of these processes is essential for a comprehensive assessment of the model's capabilities in simulating the urban boundary layer and urban plume.

• The manuscript uses both the CABAM and STRATfinder algorithms for ALC retrievals. While both algorithms aim to retrieve the ZO, the differences between them and the implications of using two separate algorithms should be discussed. It is important to address how these differences may affect the comparison to the ZM, as this could influence the results and interpretation. 

• Section 4 would benefit from specific quantitative values when discussing over- and underestimations. This would provide a clearer understanding of the magnitude of these errors and allow for a more precise evaluation of the model’s performance. 

• How much of the bias can be attributed to differences in vertical resolution between ceilometers and models? This should be addressed earlier in the manuscript to assess the extent of the biases presented. 

• Section 5 should be revised in light of the points above. If the authors can better support the rationale for using simulated aerosols for MLH, and ensure that the definitions of thermodynamic and aerosol-based layers are consistent, the manuscript will be clearer and more coherent. 
• The manuscript's title, 'Hectometric-scale modelling of the urban mixed layer evaluated with a dense LiDAR-ceilometer network,' suggests a primary focus on the urban mixed layer. However, a substantial part of the manuscript is dedicated to evaluating the urban plume. A revision of the manuscript's stated goals to ensure better alignment between the title and the content is needed. 
• Furthermore, if a central goal is to investigate the simulation of the plume, it is crucial for the authors to explain why the analysis predominantly focuses on the top of the aerosol layer (identified as ZM/ZO) rather than providing an examination of the full vertical and spatial distribution of aerosols. A detailed analysis of the entire plume's structure would offer a more complete understanding of the model's performance in simulating urban aerosol transport.
• The misrepresentation of soil moisture in the model is repeated throughout the manuscript, it would be helpful to see the extent of this misrepresentation and how it may relate to the findings in the study. 

Minor  Comments: 

• There is a discrepancy between the text and figure captions for Figures 4, 6, 7, and 8, where it appears that the images for 300m and 100m simulations were swapped. 
• Figures 7, 8, 9, 10, 11, and 12 are difficult to interpret. Increasing their size and resolution, especially for the ALC circle markers, would improve clarity, as it is currently hard to see the bias shading. 
• Section 2.1 would benefit from a more detailed description of the region under study. 
• Regarding Figure 1, was the synoptic setup validated using observational data or reanalysis? 
• The caption for Figure 2 is unclear and should be revised for clarity. 
• It would be helpful to label parts A, B, and C in Figure 2 for better clarity. 
• L168 requires a reference for the statement. The claim that CL31 and CL51 sensors “reach sufficient overlap at lower heights” is questionable, as these systems typically achieve overlap at approximately 100m. These sensors also have known artifacts and signal issues, as noted by Kotthaus et al. (2016) - “Recommendations for processing atmospheric attenuated backscatter profiles from Vaisala CL31 ceilometers”. 
• L214: Should this refer to Figure 4b? 
• L221: how is the BLD defined or simulated? 
• For Figure 5, it would be useful to describe the image in order of a, b, c, and d. 
• In L281-282, the references to Figures 7c and 7g are unclear regarding the 'shallow downwind' and 'high upwind' values. The visibility of markers in these figures is poor, making it difficult to discern the discussed trends. To improve clarity, it would be beneficial to include specific quantitative values in the text to support these observations.

The authors address the very relevant topic of the recent development of atmosperic numerical models towards the hectometric scale. The continuous increase of computing power meanwhile allows several Meteorological Services to run their mesoscale models at an hectometric scale, or they are close to it. One obvious application at that scale is to simulate the weather and climate phenomena over urban areas. In the recent years, several Weather Services have developed and integrated urban schemes in their NWP models.

A remaining issue for activating these urban schemes in operational NWP is the availability of observational datasets covering rural and in particular urban areas. The authors present and utilize a dense LiDAR-ceilometer network over Berlin.

General comments:

- The strategy in this manuscript is focussed on the MURK aerosol scheme in the UM. Could this method also be transferred to a sophisticated aerosol model, if becoming available in UM, or to other atmospheric models in general. There is one sentence like this in section 5, maybe this could be a bit elaborated?

- Likely also MURK has errors or shortcomings. How much do they influence or limit the general findings of this study?

- I find Fig. 4 very convincing. But it stays a bit unclear, why the UM boundary layer diagnostic BLD is so very different from the aerosol-based method, particularly in the afternoon and evening io 4 August. Some literature is quoted confirming this result. Why does the parcel-based BLD not capture the effect that the aerosols are actually further transported verically upward? By which physical process does this happen? Maybe a figure of the vertical motion (in a cross section) could shed more light on this?

- Section 4.2 has a very descriptive character. Would it possible to also find some reasoning or explanations for the things described?

- Figs. 9-12 look very convincing. Anyway, are there ideas about the (remaining) differences between model and obs in Figs.10 and 12?

Minor comments:

- Fig. 1: The rings are not explained well. The explanation comes in the caption of Fig. 2. Maybe move this to caption of Fig. 1, or refer to Fig. 2?

- Fig. 2: The descriptions of rings A to C are not so easy to find. Maybe different colour?

- L. 172: … observed MLH (from now on referred to as Z_O) …

- L. 174: Similar with model MLH Z_M.

- L. 209: WEDD: Maybe add the name „Wedding“ of the quarter (for the readers who know Berlin 🙂).