the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 02 Jan 2026
The role of individual forcings in driving the wave–like recent trends in northern hemisphere summer atmospheric circulation
Abstract. The summer climate in the Northern Hemisphere during recent decades has shown distinct trend patterns, with warming hotspots that spatially match with the ridges of a circumpolar atmospheric wave-5 pattern in the upper troposphere. The drivers behind this wave–like trend and warming pattern are not yet well understood. Through the use of the Large Ensemble Single Forcing Model Intercomparison Project (LESFMIP) and the Atmospheric Model Intercomparison Project (AMIP) simulations, we study the contributions of different forcing components as well as the role of oceanic temperature variability to the observed changes. Analysis of the single-forcing experiments shows that in particular historical anthropogenic aerosol forcing leads to responses that have some pattern similarity with the observed changes in atmospheric circulation, diagnosed from the 200 hPa geopotential height (Z200) after removing the zonal mean. However, despite high pattern agreement, the magnitude of the trends is underestimated in models. Our results suggest that the observed spatial structure of trends in Z200 is at least partially caused by anthropogenic aerosol emissions and is not the result of global warming caused by greenhouse gas emissions, and highlight important inconsistencies between models and observations.
- Preprint
(1139 KB) - Metadata XML
-
Supplement
(660 KB) - BibTeX
- EndNote
Status: open (until 13 Feb 2026)
- RC1: 'Comment on egusphere-2025-6277', Anonymous Referee #1, 26 Jan 2026 reply
-
RC2: 'Comment on egusphere-2025-6277', Anonymous Referee #2, 02 Feb 2026
reply
When I first finished reading this manuscript, my initial inclination was to recommend rejection. However, after further consideration, I decided to give the authors an opportunity to revise the manuscript. That said, extensive revision and substantial improvement will be necessary before the manuscript can be considered for acceptance. See my attached document "Reviewer_comments.docx".
-
RC3: 'Comment on egusphere-2025-6277', Anonymous Referee #3, 08 Feb 2026
reply
review of Marcet-Carbonell et al
I had my first read of this paper before the other reviewers posted their comments, but it looks like I am the last to finally organize my thoughts. I am not going to repeat things requested by the other reviewers (including a summary of results), and rather focus on those aspects not already covered.
My main general comment is that the authors seem to prematurely dismiss volcanos as being important, even though figure 3 indicates a pattern correlation as strong in the multi-model mean as the other runs. The authors dismiss this high correlation because the amplitude of the signal is too-weak, but then again all of the amplitudes are too weak (though not quite as much as hist-volc) and issues with signal-to-noise are raised as a possible factor behind generically too-weak signals. While I can accept that the models likely do have S2N issues, this just begs the question of why they then rule out hist-volc as being important. If we are to inflate variances or bias correct, then perhaps hist-volc will turn out to be important after all?
I think the piece of analysis that is most missing from the paper in its current form isn't anything that reviewer 2 requested, but rather a proper quantification of whether these models have S2N issues in summer. Most of the S2N paradox literature concerns winter, though there is some on summer too (Dunstone et al 2023) but with a focus on the NAO and not on the wave-5 pattern. Please do a formal analysis of the RPC in these models! I suggest following the methodology of Hardimann et al 2022 for the RPC definition, though there are others out there that would also work ok.
Assuming there are RPC issues, the next step would be to pick individual ensemble members that most closely the ensemble mean response in ~10-20 years subsamples of the record, and then plot the Z200asym maps for those individual ensemble members. [Note that the chosen members will change in each 10 year subset, and so you will be computing a trend over a varying set of members (See Smith et al 2020). Assuming the models have a reasonable amount of variance [which isn't guaranteed, but at least the bias likely won't be as big as possible issues with the signal], the signal in this timeseries and the accompanying trend should more closely match the amplitude of that in observations. This signal can be computed for both hist-vol and hist-aer, and I think this is the proper way to assess the relatively importance of the two forcings for the observed signal while also clarifying whether S2N issues are causing a too-weak response in the grand ensemble mean. Because you have large ensembles the "matching method" of Smith et al 2020 **should** work.
Overall, I am positively incline towards this paper, and I think that with some revisions, it will be suitable for publication in WCD. I think that adding all of the analyses requested by reviewer 2 might be too extreme, however the results could stand to be expanded upon as described above.
Dunstone, and Coauthors, 2023: Skilful predictions of the summer North Atlantic Oscillation. Commun. Earth Environ., 4, 409, https://doi.org/10.1038/s43247- 023-01063-2.
Hardiman, S.C., Dunstone, N.J., Scaife, A.A. et al. Missing eddy feedback may explain weak signal-to-noise ratios in climate predictions. npj Clim Atmos Sci 5, 57 (2022). https://doi.org/10.1038/s41612-022-00280-4
Smith, Doug M., Adam A. Scaife, Rosie Eade, P. Athanasiadis, Alessio Bellucci, I. Bethke, Roberto Bilbao et al. "North Atlantic climate far more predictable than models imply." Nature 583, no. 7818 (2020): 796-800.
Citation: https://doi.org/10.5194/egusphere-2025-6277-RC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 221 | 116 | 16 | 353 | 27 | 12 | 14 |
- HTML: 221
- PDF: 116
- XML: 16
- Total: 353
- Supplement: 27
- BibTeX: 12
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review’s comments for the manuscript egusphere-2025-6277, entitled “The role of individual forcings in driving the wave–like recent trends in northern hemisphere summer atmospheric circulation"
General comments
By using the Large Ensemble Single Forcing Model Intercomparison Project (LESFMIP) and the Atmospheric Model Intercomparison Project (AMIP) simulations, this study investigates drivers and physical mechanisms of the summer circulation trends in the Northern Hemisphere during recent decades. Analysis of the single-forcing experiments shows that historical anthropogenic aerosol forcing leads to responses that have some pattern similarity with the observed changes in atmospheric circulation although with an underestimated magnitude of the trends in models. These results suggest that the observed spatial structure of trends is at least partially caused by anthropogenic aerosol emissions and is not the result of global warming caused by greenhouse gas emissions and highlight important inconsistencies between models and observations. Another important aspect of the study is the assessment of roles in solar, ozone, and volcanic forcings. The study is a good contribution in understanding recent summer circulations changes in the northern hemisphere and the paper is well written. Therefore, the paper is acceptable for publication after minor revisions by addressing the following specific comments.
Specific comments