Evaluation of UKESM aerosol size and composition using ATom measurements indicates missing marine aerosol formation mechanisms

He, Xu-Cheng; Abraham, Nathan Luke; Ding, Han; Russo, Maria R.; Grosvenor, Daniel P.; Ge, Yao; Wang, Xuemei; Jones, Anthony C.; Campuzano-Jost, Pedro; Nault, Benjamin; Kupc, Agnieszka; Blake, Donald; Jimenez, Jose L.; Williamson, Christina J.; Carslaw, Kenneth S.; Weber, James; Archibald, Alexander T.; Gordon, Hamish

doi:https://doi.org/10.5194/egusphere-2025-3700

Preprints

https://doi.org/10.5194/egusphere-2025-3700

Preprints

27 Aug 2025

| 27 Aug 2025

Evaluation of UKESM aerosol size and composition using ATom measurements indicates missing marine aerosol formation mechanisms

Xu-Cheng He, Nathan Luke Abraham, Han Ding, Maria R. Russo, Daniel P. Grosvenor, Yao Ge, Xuemei Wang, Anthony C. Jones, Pedro Campuzano-Jost, Benjamin Nault, Agnieszka Kupc, Donald Blake, Jose L. Jimenez, Christina J. Williamson, Kenneth S. Carslaw, James Weber, Alexander T. Archibald, and Hamish Gordon

Abstract. Atmospheric aerosols influence climate through their interactions with radiation and clouds, yet large uncertainties remain in their simulation by global models. This study evaluates the United Kingdom Earth System Model version 1.1 (UKESM1.1) using global-scale aircraft observations from the. Atmospheric Tomography (ATom) mission, focusing on aerosol lifecycle processes in the remote marine atmosphere. We assess model performance in simulating aerosol precursor vapours, number size distributions, chemical composition, and environmental conditions. Several process improvements are tested, including sulfuric acid–ammonia nucleation, ammonium nitrate scheme, methanesulfonic acid condensation, and low-temperature isoprene-derived secondary organic aerosol formation.

Model biases differ significantly between the upper troposphere (UT) and the marine boundary layer (MBL). In the UT, UKESM1.1 overestimates nucleation and Aitken mode particles while underestimating accumulation mode, indicating insufficient growth. In the MBL, the model overestimates primary aerosols (e.g. seasalt) and precursor gases but underestimates nucleation and Aitken mode particles, even after incorporating updated nucleation and ammonium nitrate scheme. The persistence of low aerosol number concentrations, despite overestimated precursors, suggests missing formation pathways likely involving other species such as iodine, amines, or organic vapours.

These limitations result in an unbalanced cloud condensation nuclei budget that over-replies on primary emissions. Sensitivity tests reveal that model outputs are strongly influenced by dimethyl sulfide emissions and vapour condensation schemes. Our results highlight the need for future model development to prioritise mechanistic representation of currently missing aerosol sources, rather than relying on empirical tuning, to improve aerosol–climate interaction estimates.

Received: 30 Jul 2025 – Discussion started: 27 Aug 2025

Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Status: final response (author comments only)

RC1:
'Comment on egusphere-2025-3700', Anonymous Referee #1, 26 Sep 2025
The manuscript evaluates UKESM1.1 against ATom aircraft observations and implements several process upgrades (H₂SO₄–NH₃ nucleation based on Dunne et al., MSA condensation, ammonium nitrate scheme, alternative DMS chemistry/ climatology, isoprene SOA scheme). The paper identifies a persistent mismatch: MBL nucleation/Aitken aerosols are underpredicted while precursors are overpredicted, and UT nucleation is sometimes overpredicted but growth to accumulation mode is insufficient. The topic is important and the model-development tests are useful. Overall, the study is well-designed and easy to follow, with most conclusions supported by model results. Therefore, I think the paper may be accepted for publication although I'd like the authors to consider some options to improve their work and polish the manuscript.
What my concern is the text's length and narrative structure. For instance, the Result section's current variable category organization results in a long and detailed narrative that makes it difficult for the reader to quickly identify and connect the most important findings. Consequently, the significance of the paper's significant findings might be lessened. I suggest moving more technical or secondary information to the Supplement and reorganizing this section to highlight the most important findings.

L189 The authors stated that anthropogenic emissions are emitted into the lowest model. What is the height of the surface layer?

The authors should precisely specify model setups, including simulation periods, initial and boundary conditions, and details regarding the ERA5 dataset.

In section 2.3.1, the authors enumerated four nucleation schemes derived from CERN CLOUD (Cosmics Leaving Outdoor Droplets) studies, which encompassed binary and ternary ion-induced nucleation. Are the four schemes examined in this study? Please clarify. To prevent ambiguity, the authors should provide a detailed description of the nucleation methods employed in the model within the text or appendix, as the objective of the study is to investigate aerosol formation mechanisms.

It is preferable for readers to have a description of how the K_RH adapts the nucleation process. The authors modified the original formula (K_RH = 1 + c1(RH − 0.38) + c2(RH − 0.38)³(T − 208)², where c1 = 1.5 ± 1.3, c2 = 0.045 ± 0.003) to K_RH = (RH/0.38) × [1 + 0.02 × (RH − 0.38) × (T − 208)¹^.²]. The alteration is a significant adjustment that directly affects nucleation rates in the current study. To substantiate this, kindly furnish the precise algebraic expression, parameter values, and rationale for the selected coefficients (e.g., exponent 1.2, coefficient 0.02) inside the text or the appendix.

The authors emphasized that the CS2 scheme employed distinct emission inventories compared to the Strat-Trop chemistry scheme. What are the disparities in essential pollutants that regulate nucleation, such as SO₂ and precursors influencing OH radicals (e.g. NOx, VOCs)?
Citation: https://doi.org/10.5194/egusphere-2025-3700-RC1
RC2:
'Comment on egusphere-2025-3700', Anonymous Referee #2, 02 Oct 2025
The evaluation of the United Kingdom Earth System Model version 1.1 (UKESM1.1) using ATom observations highlights significant biases in model gas concentrations and aerosol size distributions. The study brings up the potential sources of these biases in a well-structured way, evaluating new approaches for UKESM aerosol and chemistry from incorporating new Ammonia-based nucleation pathways, different DMS chemistry and climatology, IPSOA and ammonium nitrate schemes, and MSA condensation. The claims made by the authors are displayed and answered sufficiently. In my assessment, the methodology, the data are clearly presented, and the conclusions are well-supported. Overall, the manuscript makes a meaningful contribution to the development of UKESM and ESM in general, and I believe it meets the standards for publication. Therefore, I think this paper should be accepted.
The review below is structured as direct responses to the questions outlined in ACP’s review criteria.
Does the paper address relevant scientific questions within the scope of ACP?

This paper examines how the aerosol lifecycle, including precursor vapors, number size distributions, and chemical composition, can be represented in global models, as well as how these representations can be evaluated, of which fall within the scope of ACP.
Does the paper present novel concepts, ideas, tools, or data?

The paper primarily focuses on evaluating and improving existing elements within the UKESM1.1 model, rather than presenting entirely novel concepts or tools. The paper implements and tests several updated or alternative process representations within UKESM1.1. The paper identifies missing parameterizations likely necessary to resolve existing model biases.
The study utilizes global-scale aircraft observations from the Atmospheric Tomography (ATom) mission as a comprehensive dataset for evaluation. The paper presents novel model outputs to compare and evaluate towards this ATom dataset.
Are substantial conclusions reached?

Substantial conclusions are reached regarding the need to include representation of currently missing aerosol sources and updated chemistry schemes in climate models like UKESM1.1 to accurately simulate the remote marine aerosol lifecycle.
Are the scientific methods and assumptions valid and clearly outlined?

Some weaknesses in the methodology are outlined below with questions. If no question is asked, these issues are adequately addressed in the paper.
The SOA formation is represented by scaling monoterpene oxidation yield (by a factor of two) to implicitly account for other sources like isoprene. This scaling factor (as suggested in the paper) is highly uncertain. Which aerosol size mode (nucleation, Aitken, accumulation, or coarse) is the SOA mass yield formed into?
The stated MSA condensation scheme is “preliminary” and stated to “not yet be fully validated” and act as “Initial proof-of-concept implementation”. The merging of MSA as H2SO4 aerosol mass also does not account for the optical properties of MSA in the particle phase in this new scheme.
Are the results sufficient to support the interpretations and conclusions?

In summary, the detailed comparison against the unique ATom dataset and the application of process-specific sensitivity tests provide sufficient evidence to back the paper’s primary conclusions concerning the need to implement new mechanisms. Adequately addressing the weaknesses of the implementations presented in this study.
I would like to see a short mention of the limitation imposed by the computational time required by the chemical solver, if the suggested pathways were implemented, given that most ESMs are constrained in this regard. It would also then be helpful to clarify whether these potential future model implementations are intended for the CMIP version or for UKESM.
Is the description of experiments and calculations sufficiently complete and precise to allow their reproduction by fellow scientists (traceability of results)?

Yes.
Do the authors give proper credit to related work and clearly indicate their own new/original contribution?

Yes.
Does the title clearly reflect the contents of the paper?

Yes.
Does the abstract provide a concise and complete summary?

Yes.
Is the overall presentation well structured and clear?

Yes.
Is the language fluent and precise?

Yes.
Are mathematical formulae, symbols, abbreviations, and units correctly defined and used?

Yes.
Should any parts of the paper (text, formulae, figures, tables) be clarified, reduced, combined, or eliminated?

Overall, the authors require the reader to frequently jump between figures to follow and support the scientific conclusions they present. Some figures could perhaps be moved to a supplementary section, or minor adjustments could be made to the various curtain plots of the UKESM experiments, as differences are difficult to discern in most figures. However, the overall structure of the figures is visually acceptable.
Are the number and quality of references appropriate?
In general, yes. The cited findings from previous literature identify candidate mechanisms for implementation. Moreover, claims regarding insufficient model oxidation products and the need for new mechanistic pathways could be further supported by results from the ESM study on EC-Earth3 by Svenhag et al. (2024), which used TM5 to generate ion-dependent NH₃–H₂SO₄ nucleation via a lookup-table approach.
Is the amount and quality of supplementary material appropriate?

Yes.
Citation: https://doi.org/10.5194/egusphere-2025-3700-RC2

Viewed

Total article views: 2,320 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
2,213	94	13	2,320	19	24

HTML: 2,213
PDF: 94
XML: 13
Total: 2,320
BibTeX: 19
EndNote: 24

Views and downloads (calculated since 27 Aug 2025)

Month	HTML	PDF	XML	Total
Aug 2025	714	21	3	738
Sep 2025	1,409	53	7	1,469
Oct 2025	90	20	3	113

Cumulative views and downloads (calculated since 27 Aug 2025)

Month	HTML	PDF	XML	Total
Aug 2025	714	21	3	738
Sep 2025	1,409	53	7	1,469
Oct 2025	90	20	3	113

Viewed (geographical distribution)

Total article views: 2,269 (including HTML, PDF, and XML) Thereof 2,269 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 26 Oct 2025

Short summary

Aerosols affect clouds and climate. However, current climate models still struggle to simulate them accurately. We used aircraft data from a global mission to evaluate how well the UK Earth System Model represents aerosols and their precursors. Our results show that the model misses key formation processes in clean ocean regions, suggesting that future improvements should focus on better representing how aerosols form naturally in the atmosphere.


Total:	0
HTML:	0
PDF:	0
XML:	0