The paper deals with a significant topic of anthropogenic heat from buildings. Unfortunately the authors were not aware of recent developments (e.g., the capability of EnergyPlus to calculate heat emissions from buildings in version 9.1released in 2019 ) and directly relevent publications which are not mentioned at all in the manuscript, such as:

Modeling and Analysis of Heat Emissions from Buildings to Ambient Air. Applied Energy, 2020. https://doi.org/10.1016/j.apenergy.2020.115566

A Simulation-Based Assessment of Technologies to Reduce Heat Emissions from Buildings. Building and Environment, 2021. https://doi.org/10.1016/j.buildenv.2021.107772

EnergyPlus already can calculate in detail (at sub-hourly timestep) heat emissions from buildings to ambient air by components. There is no need to re-invent the wheel (a new definition or formulas).

The use of the shoebox model to simulate and analyze the building heat emission is questionable as the shoebox model is not a real building. In ASHRAE Standard 140 context, the building model is used as a common simplified benchmark model to test and compare results from various building energy modeling tools. Why use this model? If the analysis is to study how human activities influence building operation and thus heat emissions, a real residential or commercial office building model should be used.

The manuscript has some typos, e.g., "and time lag and are poorly quantified"

In this manuscript authors analyze the difference between Building Energy consumption and heat flux ejected to the atmosphere by buildings due to human activities. They base their analysis on the comparison between a no occupied building and one occupied. They first show, using conservation equations, why these two quantities are different, and then they use a series of EnergyPlus simulations to quantify the differences.

I like this manuscript. I think it highlights a very important point for urban atmospheric modelling. My few comments below are oriented to improve the clarity and strength the message.

Comments:

1. I strongly suggest authors to include a list of symbols. This will improve the clarity and help the readers to follow the equations.
2. Authors define Q _F,B as “…the heat released from buildings into the atmosphere as a result of human activities inside the building (including human metabolism)” (lines 53-54). Then in Q _F,B they include Q_waste,o , the heat emitted to the atmosphere by HVAC. Indeed, running the HVAC is a result of a human activity, so it is understandable to include Q_waste,o in Q _F,B , but I think it would be worth reminding here that not all the Q_waste,o is coming from heat generated inside the building by human activities (e. g. lighting, powered appliances and metabolism). A significant part of this heat is energy that has entered in the building from outside in form of radiation through the windows, or heat diffusion through the walls. In other words, an empty building with HVAC functioning, would still have Q_waste,o different than zero. This energy, that is not from anthropogenic origin, would have been stored in the building without HVAC. This message (to me one of the most important of the study), is already implicit in the manuscript, but I believe it should be made explicit in the text.
3. Lines 57-58 “This energy modifies the internal building volume…”. Energy cannot modify the internal building volume, Please clarify.
4. Lines 65-66 “Some of this additional energy is transported out of buildings through indoor-outdoor ventilation exchange and immediately contributes to Q _F,B “ . HVAC also would immediately eject the additional heat.
5. Line 79. Just to improve clarity, and avoid misunderstanding, I would remind here that Q_H is not the sensible heat flux that could be measured on a mast in the inertial sublayer. This should include also Q_Waste and Q_BAE .
6. Figure 1. If ΔH is the difference in sensible heat flux from building surfaces between an occupied and an unoccupied building, I am a bit surprised to see such small variability between night and day. I would expect that the change in storage would affect more the heat flux, in particular during night.
7. Based on this study – Can authors derive some recommendations to users and developers of urban canopy parametrizations that include simplified building energy models?
8. As suggestion for future studies: it would be interesting to investigate how the use of HVAC systems modulate the temporal behavior of the storage term also for other types of buildings (e. g. commercial), that have a different human behavior (occupancy, schedule of operation, etc.). Given that storage is key in the development of the nocturnal UHI – how does HVAC affect the strength of the nocturnal UHI?

This paper has provided an innovative perspective on the definition of anthropogenic heat flux QF,B, which distinguishes it from the widely used proxy building energy consumption QEC. The key connection between them lies in the building storage heat flux. The paper started with a conceptual discussion of their distinctions and then conducted building energy simulation using EnergyPlus for a typical building in Beijing, China. Based on a one-year simulation, the building storage heat flux and other relevant surface energy budget terms are compared and discussed for different scenarios of building energy use and differences between QEC and QF,B. The results caution the effects of such differences in future studies using the inventory approach to quantify anthropogenic heat flux, especially in atmospheric modeling. The paper is well written and organized. I have only minor comments.

1. line 43: The last sentence of this paragraph seems to only discuss the building energy used for heating. It can be rephrased in a more general heating and cooling context.
2. line 53 – 57: The first two sentences appear a bit contradicting. The first sentence says that QF,B is the heat released from buildings into the atmosphere and is zero when building is unoccupied. The second sentence says that ‘heat released from the unoccupied building is non-zero’. I understand that heat released from the unoccupied building is not QF,B. Perhaps it is less confusing and more consistent with introduction of Eq. 1 and Eq. 5 to mention first that regardless of whether human activities are present, buildings exchange (as you showed later QF,B can be positive or negative, so not just releasing heat) heat with the atmosphere.
3. Eq. 5 has subscripts ‘o’ to distinguish some of the common terms from those in Eq. 1. However, to emphasize, it may be helpful to re-iterate that introducing of Qinteral,o on the LHS modifies the entire energy balance in your chosen control volume.
4. Eq. 15: It appears to the reviewer that a long time limit exists, such that QF,B = QEC, as the long time average of storage heat flux tends to zero. Also, does the simulation data support this speculation? Maybe this is helpful to comment on, especially regarding using QEC from the inventory as a proxy for some annual/seasonal averaged quantities.
5. Fig.3: it may help visually to put a R=1 line.
6. Line 365: does ΔB refer to the definition in Eq. 9?
7. Does the ambient temperature considered in the building energy simulation kept as an externally imposed forcing? i.e., there is no feedback between heat released from the buildings and the ambient air.