Dear authors,

Referee #3 had some further comment on your paper, I'd like to ask you to address (see below). After this, I will be happy to accept your manuscript for final publication in ACP.
Best regards
Barbara

----------------------
I recommend publication of this revised manuscript after the following minor corrections are addressed.

General Comment:

The revised manuscript is well written, with a thoughtful and thorough documentation of the model, methods and results. I would like to see the authors taking a bit more care with some of their statements related to the connection between the HadGEM3-UKCA process sensitivity simulations and the AeroCom model ensemble as noted in the comments below.

1) In the response to comment 7) of referee 2, the authors state ‘We only claim that these processes are sufficient to produce this diversity, not that they are necessarily its cause amongst the AeroCom models.’ This is an important point to keep in mind – the study is not able to show that the processes studied in the sensitivity studies are the cause of the inter-model diversity. As a result, I suggest that the authors consider changing the word ‘relationships’ in the manuscript title to ‘comparisons’ as the former suggests a connection between the process sensitivity in HadGEM3-UKCA that might be interpreted as causal of the differences.

2) P1, L3: Consider moving ‘in one particular model’ to immediately follow the words ‘the controlling factors’. Then perhaps start a new sentence with ‘We compare the resulting diversity …’. This revision avoids the suggestion that you are able to investigate the controlling factors for the AeroCom diversity.

3) P2, L12: Consider starting the paragraph with an identification that this paragraph refers to HadGEM3-UKCA model, perhaps add ‘In the HadGEM3-UKCA mode; convective transport…’

4) P2, L21: Consider quantifying the size range meant by ‘smallest’ and ‘larger’ particles.

5) P3, L51: Consider adding a reference at the end of this statement regarding previous publications that show this large diversity.

6) P3, L59: Please consider providing an alternative for the terminology ‘explain this diversity’ since this statement seems to suggest that the processes in the sensitivity studies are at least partly explaining (i.e. partly a cause of) this diversity.

7) P3, L62: Consider changing the words ‘are responsible for’ as this implies that the process sensitivity study is able to explain causes of the inter-model diversity. If the processes studied are not necessarily the cause of the inter-model diversity, then the authors can only state that the sensitivity studies are able to produce diversity similar to the AEROCOM inter-model diversity, but the authors are not able to establish what is responsible for this diversity.

8) P12, L363: This line says ‘many different processes show significant effects’ – does this contradict P2, L1-2 that says ‘controlled by a relatively small number of processes’?

9) P19, L630: Please consider changing the terminology ‘these processes can account for’ – this may imply that these processes are the cause of the diversity. Perhaps ‘can produce a spread similar to the overall spread’ might be better here.
10) P19, L637: Consider changing the terminology ‘beyond those captured by the processes considered here’ – this suggests that the processes considered are a cause of the intermodal diversity.

11) The words ‘reproduce’ and ‘replicate’ are used at a number of places throughout the text and I suggest that the authors reconsider the terminology as this might suggest a causal relationship between the processes in the sensitivity studies and the AeroCOM inter-model diversity. It is possible that these processes may contribute to the AEROCOM inter-model diversity, but the methodology of this study does not allow us to draw conclusions about what causes the inter-model diversity. Perhaps stating that sensitivity studies with HadGEM3-UKCA ‘can produce a similar…’ as opposed to reproduce or replicate might be better terminology.