Dear authors

The two reviewers made a long list of sugegstions to improve the paper. I can see from your answers that you are able to address most of the raised concerns. Clearly NH3 emissions from animal production goes over several steps that will globally greatly vary. It is important to discuss whether compensating errors yields to a reasonable estimate of the global emissions, e.g. by applying a too high emission factor for spreading and neglecting the previous steps (housing and storage). In such a situation the response to climate change might not reflect what is happening in reality.
I am looking forward to get the revised manuscript. I might call additional reviews.
With kind regards and best wishes for the New Year
Albrecht Neftel, Editor BG

Dear authors

I went through the revised manuscript. You have addressed many of the comments by the reviewers and improved the manuscript. I am aware that in a global model with the aim to simulate the effect of climate change on the agricultural N-emission simplifications and shortcuts cannot be avoided.

Nevertheless I discovered two critical issues:

a): lines 404 and 405: It is assumed that a fraction (fu = 0.5) of nitrogen excreted is urine (urea), with the remaining 50 % excreted as faecal matter [Gusman and Marino, 1999].

To my knowledge the share of urea in urine is generally around 80%.

Urine:
For dairy cows: 50% of ingested N is excreted with urine and 29% with the feces, 19% ends up in in milk and 2% in tissue*
Content of urea in urine of ruminants: 59-89 % of total N in urine**

References:
*Tamminga, S. 1992. Nutrition management of dairy cows as a contribution to pollution control. J. Dairy Sci. 75(1): 345-357.
**Bristow, A.W., Whitehead, D.C., Cockburn, J.E. 1992. Nitrogenous constituents in the urine of cattle, sheep and goats. J. Sci. Food Agric. 59(3): 387-394.

For pigs I have very similar values in mind.

Consequently in your model you got overall a higher fraction of NH3 emission from application of mineral fertilizer than from organic fertilizer (19% versus 17%), because the loss rate of NH3 related to urea application is on the high side. I am aware that the database for this value is scarce and many of the reported value are of questionable quality.

b) Fig 4d and 5d show maps of fertilizer and manure application related NH3 loss. If I zoom into Europe I am surprised that no structure is seen in the manure related loss, especially as the Netherlands and also the Po valley regions are known as the world champions in NH3 emission due to the high animal density. Is this due to the rather coarse resolution? And if yes why other regions show more small scale variability.

In order that I can accept the paper for final publication you should address these questions.

With kind regards
Albrecht Neftel

Dear authors

I do accept the revised version for publication in BG. Please check carefully the proof reads as the copy editing process can introduce misunderstandings.

With kind regards
Albrecht Neftel, Editor for BG