Summary/qualitative assessment: This manuscript seeks to explain the offset in timing between the observed maxima in cloud top height and rainfall in a simulated tropical cyclone (TC). The authors demonstrate that this difference in timing arises due to the bimodal nature of the TC response to the diurnal cycle: the peak in cloud top height occurs during the day, due to heating-induced upward motion from the absorption of solar radiation by ice crystals; the peak in rainfall occurs at night in response to latent heating and increased convection, due to both radiative destabilization at cloud top as well as low-level convergence induced by cloudy vs. clear sky differential radiation.

This is a clear, well-written paper which elegantly describes the relevant mechanisms and highlights the important processes. The figures are dense but clearly labeled, which makes for ease in interpreting. I recommend acceptance after a few clarifications and minor revisions.

It may be useful to also include a 'no radiation" case in Fig 1, to show both how the presence (and absence) of radiation affects the timing of cyclogenesis, as well as for clear comparison with previous studies (e.g., Melhauzer and Zhang 2014).

One major take-away from this work is that the outflow-layer circulation (daytime circulation) is sensitive to the concentration of ice, and therefore the model microphysics. How sensitive are these results to the chosen microphysics scheme? I do not suggest that the authors perform such sensitivity calculations for this study, but I do think this is an important aspect to consider (similar to the discussion of domain size in lines 367-374).

Are these simulates computed for a constant Julian day? If so, I would suggest stating that here for clarity and for possible reproduction of the work. If not, how might temporal variations in solar zenith angle affect the results for these long simulations? For example, after 90 days of integration, how might the change in incoming radiation affect the results?

With a stretched vertical grid, what is the resolution near the outflow layer in the simulated TC? How might this resolution affect the magnitude of the vertical circulation (w') at this level (e.g., Fig. 2a.-b.)?

Suggest elaborating on "such mechanisms" for better understanding. For example, explicitly stating that "understanding this offset phasing in TCs" is critical.

Add "the" after of

Suggest rewording of this phrase to, e.g., "that has long been considered in the literature."