The paper represents the second part of the description of the AMmonia-CLIMate (AMCLIM) model for quantifying ammonia emissions specifically from livestock production responding to environmental conditions and agricultural practices. The model which incorporates a high level of detail is well described in general and seems to perform well compared to observations and also to previous global-scale estimates. A serious and appropriate sensitivity analysis is also conducted confirming the important role of pH in the emissions.

The strength of the approach lies in the diversity of agricultural practices data and a detailed representation of ventilation scheme in the housing that are considered for ammonia emissions. 

The study will likely provide a major advancement for the community especially for designing precise mitigation strategies for NH₃ losses but I have some comments before it can be considered for publication.

• I believe it is crucial in the introduction to put the AMCLIM approach in the state-of-the-art modeling context for ammonia emissions from agriculture. From l.51, this part can be more developed/precise to highlight the strength of AMCLIM. For instance, global processed-based models included in Land Surface Models (LSM) such as FAN v2 in CESM (Vira et al., 2020) and CAMEO in the IPSL ESM (Beaudor et al., 2023) in which detailed soil C/N cycles are implemented with multiple interactions (vegetation, soil BGC, water, and energy budget) also incorporate agricultural practices (by livestock type) and run with dynamic environmental conditions at the global scale. However one of their weaknesses is to rely on Emission Factors (EF) for the manure management part. I think it would be helpful to mention it in the introduction because AMCLIM is an interesting trade-off between process-based models/ Earth System Models and more socio-economic models such as the Integrated Assessment Models (IAMs).
  
  
  
  
• It is hard to get the global picture of the results of the different NH₃ volatilizations; the "per grid" unit for the emission flux distributions (Fig 7-11) is not common and it is hard to compare with other literature results (which are usually given by /m2). Since the spatial resolution of the model is also not clearly stated, it is almost impossible to compare it with any other results. I know that the Part 1 paper also presented the ammonia flux in that unit and maybe for consistency reasons, the authors would like to stick to that unit but a clear justification should be included. 
• On the same topic, to ease the comparison with other global results, one suggestion would be to include a global map of the combined synthetic fertilizer and livestock emissions (i.e. total agricultural sources) in the common units: gN/m2/yr. This map could be shown as a final result?
  
  
  
  
• The 2-year (2010 vs. 2018) comparison needs to be justified earlier in the manuscript and I am missing a description of the input that has been used for 2018. The setup for this experiment is not properly described in the Methods. Is the livestock distribution dataset for 2010 vs. 2018 the same? How the 2010 year inputs have been extrapolated?
  
  If the meteorological impact is the major reason behind the 2-year comparison maybe the title of section 4.4 could be more explicit.
  
  Please consider that  2018 results already appear in the previous sections when describing the maps of the volatilization rates.
  
  The maps related to 2018 are in the Appendix, this makes the manuscript somehow very heavy for the reader to do this back and forth (especially for 6 maps x 3 animal types). Is there any way to combine these results maybe in a histoplot by region instead of multiple maps?

Minor comments:

• Fig 1: I think representing the NH₃ losses on that schematic would be useful
• L. 174 : ​​Is there any difference between F_{NH3 volatilization} and F_NH3 throughout the equations?
• L.190 : I think the pH variation with the different processes would be interesting to describe. Perhaps a little paragraph explaining its drivers would be beneficial since it is a critical parameter admitted by the authors and known by the community.
• L.343 : I am not sure this reference is well placed in the context
• Section 2.52 : I think a table gathering information on the different inputs used for the global scale (for both 2010 and 2018) and site simulations would be useful
• L. 408: What are the spatial and temporal resolutions of the model?
• L. 486: What could be a reasonable explanation for these discrepancies?
• Section 3.2: Is there any logic behind the order of the animal types that are presented? I wonder why cattle are not presented first since they are responsible for the highest emissions.
• Section 3.2.2: I would add a reference to the section where the new poultry emissions are compared to the older version. This would help the reader to understand that there is a dedicated section for that comparison.
• The title of Section 3.3.2 could be a bit more precise: “Comparison of grazing NH₃ emissions with observations” ?
• Section 3.4: For the N budget, I wonder if other N species were also taken into account. N₂O, N₂ and NO₂ emissions can also originate from manure. Even though these EFs are relatively small compared to NH₃, I think it is important to mention them for a consistent N budget. I invite the authors to have a look at studies from Sommer et al., 2019 and EMEP/EEA 2019 for more information.
• L. 806: Please explain what is the GUANO model.
• Section 4.2: The GLEAM model and estimated EF from Yang et al would benefit from a little description. I wonder how the AMCLIM EF also compares with EFs from EMEP/EEA or Sommer et al., 2019 both specific to Europe and sometimes applied to the whole globe in several process-based models (Vira et al., Beaudor et al.,).  In the study from Sommer et al., 2019, they also analyzed the EFs of the ALFAM2 process-based model  (given as % of TAN content in the manure applied) by season and livestock type which could be interesting to compare.
• L.833: For the animal EFs taken from Yang et al., study, would you have any references for the data presented? I don’t see any precise reference in their paper but instead this description: “NH₃ EFs from different animal types were collected by keyword searches of above several databases, including “animal ammonia/NH₃ emission”, “cattle”, “buffaloes”, “chickens”, “ducks”, “goats”, “sheep”, “livestock production” and “animal husbandry operations”. “ 
• L.918: Please define FAN v2. In addition, a little description would be beneficial for the reader maybe in the introduction as suggested earlier.
• L. 978: Please correct for "database". 
• L.982: I believe the sentence in parenthesis should not be there.

References :

Vira, J., Hess, P., Melkonian, J., and Wieder, W. R.: An improved mechanistic model for ammonia volatilization in Earth system models: Flow of Agricultural Nitrogen version 2 (FANv2), Geosci. Model Dev., 13, 4459–4490, https://doi.org/10.5194/gmd-13-4459-2020, 2020.

Beaudor, M., Vuichard, N., Lathière, J., Evangeliou, N., Van Damme, M., Clarisse, L., and Hauglustaine, D.: Global agricultural ammonia emissions simulated with the ORCHIDEE land surface model, Geosci. Model Dev., 16, 1053–1081, https://doi.org/10.5194/gmd-16-1053-2023, 2023.

Sommer SG, Webb J and Hutchings ND (2019) New Emission Factors for Calculation of Ammonia Volatilization From European Livestock Manure Management Systems. Front. Sustain. Food Syst. 3:101. doi: 10.3389/fsufs.2019.00101

The manuscript presents AMCLIM, a process-based, dynamic model for estimating NH₃ emissions from livestock. The model comprehensively accounts for NH₃ emission processes and nitrogen flow across housing, manure management, and land application. Figures 1–3 clearly show these processes, and the model simulations successfully capture NH₃ emissions observed from individual animal houses. AMCLIM represents a significant improvement in the estimation of NH₃ emissions from livestock. I believe the manuscript should be published, with only a few comments for the authors to address:

1. In the Introduction, the authors highlight challenges in previous process-based models when representing various management practices. I recommend a direct comparison between the processes considered in AMCLIM and those in previous studies to better illustrate the differences.
2. Equations (1)–(19) provide a detailed description of how the model calculates NH3 concentrations inside animal houses and nitrogen flow. However, it remains unclear how these nitrogen pools or NH₃ concentrations are ultimately converted into NH3 emissions. Could the authors clarify?
3. In Figure 2, I suggest including the explicit equations for f(Tgnd, pH, RH) and f(Tgnd, pH) since the influence of pH on the calculations is discussed in the manuscript.
4. In Section 3.1, the indoor temperature, airflow, and RH for individual animal houses are derived from observations, which likely contribute to the good agreement between modeled and observed NH3 emissions. How are these indoor parameters determined when estimating NH3 emissions at the global scale?
5. The global livestock population data is based on FAOSTAT for 2010. How is this dataset extended to other years? Are the calculations for 2010 and 2018 based on the same livestock population and distribution, or do they incorporate changes over time?
6. Line 574: Do the volatilization rates refer to the ratio of NH3 emissions to total nitrogen excreted by livestock (Nr)? Please clarify.
7. In the comparison of NH₃ emissions between 2010 and 2018, could the authors directly show the spatial distribution of the difference? Current spatial maps make it difficult to distinguish the changes clearly.
8. Line 575: It would be helpful if the authors could explain the spatial variations in Pv.