The manuscript showcases an integration between the MPAS model with the JEDI data assimilation system. The authors focus on the evaluation of EnVar performance when assimilating various subsets of observations. By incrementally increasing the number of satellite radiance observing systems, the authors establish the robustness and correctness of the integrated JEDI-MPAS system. A highlighted component in the manuscript is the JEDI-MPAS’s capability of assimilating all-sky microwave radiances to update the hydrometeor analyses.

In general, the manuscript is well-written and properly supported. The implementation of JEDI-MPAS is of significant importance to the community and is suitable for the scope of Geoscientific Model Development.

Minor comments:

1. A more detailed caption for Figure 1 is required for standalone comprehensibility.
2. Figure 2 has quality issues.
3. Please refer to the GMD style guide
   1. Avoid the use of hyphenate modifiers containing abbreviated units. This includes “6-h” (e.g., line 240, 275 ), “30-km” (e.g., line 237, 241,), etc
   2. Numbers that begin a sentence should be spelled out or reformulated (e.g., line 243, 412, 416).
   3. The units of physical quantities in the denominator should be formatted with negative exponents (e.g., m/s should be $m s^{-1}$)
   4. Equations should be referred to by the abbreviation "Eq." and the respective number in parentheses (e.g., line 118, 161, 190, ).

Summary:

This paper summarizes the first implementation of the JCSDA JEDI-enabled data assimilation (DA) framework for MPAS (JEDI-MPAS 1.0.0). The paper begins with a brief description of the formulation MPAS model, and includes specific details on the implementation of dual resolution (30 and 60 km) 3DEnVar, including a detailed description of the background error covariances as implemented using JEDI.    The paper also discussed the choice of analysis variables and variable transforms, observation processing and QC. Notably, it gives a detailed description of the all-sky microwave AMSU-A radiance assimilation for both non-precipitating and precipitating cloudy situations to update all five hydrometeors analysis variables.

The validity of the JEDI-MPAS 1.0.0 system is demonstrated through a series of one-month cycling DA and long forecast experiments comparing the baseline system (conventional plus GPS observations) with the progressive addition of microwave radiances from AMSU-A, AMSU-A all-sky and MHS. Skill comparisons with GFS cold start forecasts are also made.

Overall, the paper is well-written, complete and contains appropriate references and citations. It will provide a useful reference for both developers and users of JEDI-MPAS, as well as general JEDI developers and users.  

Specific Comments:

Line 286: This line “Overall the mean and RMS give the same indication regarding the four experiments’ relative performances.” Is not clear. What is meant by “give the same indication”?

Line 305: What is the relevance of this comment?  “Note that the majority of radiosonde temperature observations in GSI’s ncdiag files is expressed in terms of Tv.” Also, “is” should be replaced with “are”.  

Lines 313 and 357:  An experiment “cldmhs” is mentioned in two locations in the text.  Do you mean “clrmhs”?

Line 437: Suggested wording changes: less-advanced and fewer observations.

Figure 1: The caption for this figure is insufficient.

This manuscript introduces the JEDI implementation with MPAS of an EnVar system on a coarse scale.  The manuscript is well written but does need a few tweaks to finish it.  My specific comments are below

1) The abstract is too long, it needs to be a concise summary of the manuscript, not read like the introduction.

2) Please use standard notation from Ide et al 1997 where it should be h(x) for nonlinear observation operators and Hx for linear.

3) The caption for figure 1 is unacceptable and needs to concisely explain what is in the figure