Please see the supplement.

General comments:

This study heuristically devised a simple model describing the behavior of liquid water path (L) as a function of cloud droplet number concentration (N) with its parameters adjusted to previously performed LES simulations and perturbed to explore the system sensitivity. The simple model is also used to analyze how the system behaves depending on external forcing of L and N. As the main framework of this study, these analyses are performed in terms of the logarithmic sensitivity parameter of L with respect to N where the precipitation and thermodynamics controls are identified and linked with each other. This is a very interesting study that provides a useful process-level insight into the cloud water adjustment to aerosol perturbations. I only have some relatively minor comments (listed below) that mainly require further clarifications of some model setup. I would recommend the manuscript be published after these comments are appropriately addressed.

Specific points:

Line 27, Line 166: What is “inverted v”? I cannot find the description of “v”.

Line 166: Likewise above, what is “regular v”?

Line 65: Insert “Based on (2)” prior to “The thermodynamic carrying capacity is...”.

Line 90: “The source S_N has been neglected for simplicity”: Does this mean that N is monotonically decreasing with time during the time integration according to (6)? If so, N should not reach the steady state. Please explain what happens with temporal evolution of N in this computational setup.

Line 106: “Without N dynamics”: Does this mean that only (3) is used without (6)? Please clarify.

Line 177: Would it be possible to write down the equation describing how the perturbation timescale (t_prt) comes into the perturbation added (Δln(N)_prt). It is unclear (at least for me) how the prescribed perturbation timescale is used in calculation of the perturbation.