This manuscript describes the modernization and refactoring of the widely-used state-of-art Noah-MP LSM, which was released as Noah-MP v5.0. The mordent Fortran code and data structures and standards are adopted in the refactoring. Five key features, including re-organized model physics inro individual process-level Fortran module files, enhanced data structure, enhanced code structure, self-explanatory variable naming standard, and enhanced interface structure to couple with the host models are introduced. It is introduced that the latest released Noah-MP has been coupled with NCAR/HRLDAS system. Some benchmark simulation results over the CONUS are presented. The work of coupling the Noah-MP v5.0 with the latest NASA/LIS system and the WRF-Hydro/NWM system is on-going. In the future, it is also plan to couple to other weather and climate models.

The original code of the model is length single Fortran file. One has to read through the whole file and locate places interested. The refactoring procedures provide a modernized and interoperated model system, with which users could easily understand, modify or utilize in wide applications. As an open-sourced model, the simulation results should be reproducible with given forcing datasets. It is a substantial contribution to modelling science within the scope of Geoscientific Model Development. The manuscript is logical clearly, concisely, and well-structurally interpreted. In addition, model refactoring is a burdensome but less productive task. Not much people have willingness to do this work. This manuscript describes the refactoring conception and processes of the widely-used land surface model. For this kind fundamental work, encourage and circulate should be deserved. I think it is acceptable with minor revisions.

The manuscript detailed describes the model structures, and what have been done with the model coupling to HRLDAS. My only concern is about the future plans. It is stated that the Noah-MP v5,0 will be coupled with more host models. It is encouraging. However, every model has its discrepancies. What are the advantages and disadvantages of this model? how about the future plans for model developments and applications? What is the next step in next several years to promote the advantages and makeup the weakness?    

This paper describes the refactored version of the NoahMP land surface model. As the authors acknowledge, work described here is a software exercise and I do not find much utility in this manuscript, particularly given that a technical report (He et al. 2023) has already been developed. In my opinion, while the software exercise mentioned here sounds like a good step, it doesn’t rise to the level of a paper for a number of reasons, listed below:

MAJOR COMMENTS

From what I can make out based on the description and the expertise of the authors, I think they missed a major opportunity in this refactorization exercise, as there is no mention or adherence to any formal software engineering principles. In many places (e.g. line 92), the mention of ‘modern Fortran code standard’ has been highlighted. Is there a formal standard that you are referring to? Be specific.

Page 3: The five key features mentioned seem quite interrelated and qualitative. What does enhanced data structure, enhanced code structure, optimized variable declaration etc mean? What formal software engineering principles were followed? Did this effort involve software engineers? How do you know if this new version is ‘enhanced’? Did you solicit the input of the community in designing this new framework?

Section 3: Other than separating the code into different modules, has the calling structure changed? From a debugging point of view, having the source code in different modules is no different from having them in a single file. For example, are there explicit code locations of extensibility defined in the new structure?

One of the major impediments to using land surface models is the specification of input datasets. How does the new structure provide an improvement in this regard?

Finally, this refactorization (despite the wide use of Noah-MP) seems like an insular effort. Did you solicit the requirements of different communities that use NoahMP (NWP, Hydrology, Data Assimilation, Crop modelers, etc.)?  Did you ensure that this new infrastructure is actually more efficient for including in systems like NWM, WRF, LIS, etc.?

MINOR COMMENTS

Line 26 and 28: The modularization, interoperability mentions are repeated. Additionally, it is mentioned again in the text, and in Conclusions. Please reduce the redundancy of such descriptions.

Line 46: ‘Moden LSMs have been …. As indispensable components …’

Page 2, last para: This is an impressive set of references of the use of NoahMP. I still think it could use some improvements; Liu et al. 2017 is not provided. There are several use of NoahMP for data assimilation than mentioned here, please include them.

Line 97: This is not a ‘study’, rather a description of the software reorganization of the code.

Page 4, paragraph beginning on line 140: I suggest rewriting this para to improve the readability. In this para, there are four sentences that begin with the style of  ‘Noah-MP does this’.

Line 263: What is the point of describing new physics that is NOT included in the community release? Is it listing different physics options being worked on? Please explain the significance or remove this para.

Section 4: What is the definition of flux, state, and parameters?

Figure 11: 2m temperature is an input to the model. Why is that being benchmarked?