Review of “Earth System Model Aerosol-Cloud Diagnostics Package (ESMAC Diags) Version 2: Assessments of Aerosols, Clouds and Aerosol-Cloud Interactions Through Field Campaign and Long-Term Observations” by Tang et al. 2023, GMD

General comment:

This paper introduces a climate model diagnosis package that focuses on evaluation of aerosol and cloud related quantities, and applies it to U.S. DOE’s E3SM. They find that the recent version of E3SM, v2, reasonably reproduces observed aerosol and cloud properties such as number concentrations of aerosol particles and cloud droplets. They also further investigated potential causes of differences in some simulated quantities comparing to that observed and suggest a direction to improve cloud microphysics schemes.

The aerosol and cloud are known for the biggest uncertainty in the climate modeling thus it is very important to establish standardized methods and comprehensive toolkits for the evaluation, which this study is aiming for and documented about. With that, this paper’s scope fits well with that of the GMD journal. The manuscript is also well-written and well-organized. The code and data well prepared for being reused via code sharing (e.g., github) and archiving (i.e., DOI from zenodo), contributing to the open-science effort of the community. I have only a few technical suggestions as listed below and recommend accepting the paper after the suggestions being considered.

Specific comments:

Figure 1: What is the units for the color shading? Can it be included in the plot for clarity?

Line 142-144: Were the NetCDF files generated following the CF-convention? If so, describing that would be plus.

Figures 6 and 7: Some of whiskers in the box-and-whisker plots are off the chart, I am curious if it would be worth to adjust y-axis range to include the top whiskers. In particular, Figs. 6a and 7c.

Figures 9, 10, and 12: What are the units for the color shadings? Can it be included in the plot for clarity?

I wonder if the authors could describe little bit more details about how the model simulation data was prepared and fed to the package. This is important because it could improve applicability of the tool for diverse models beyond E3SM.