the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Aug 2026
The CANARI HadGEM3 Large Ensemble: Design, and evaluation of historical experiments
Abstract. This paper introduces the CANARI HadGEM3 Large Ensemble (CANARI LE), a 40-member Single-Model Initial-condition Large Ensemble (SMILE) designed to advance the study of climate variability and change. It is based on HadGEM3-GC3.1, the physical climate model used by the UK Met Office and partners for submissions to CMIP6 (Climate Model Intercomparison Project Phase 6). The CANARI LE simulations span 1950–2100 and are driven by CMIP6 historical and SSP3-7.0 forcings. Compared to existing Large Ensembles, the CANARI LE offers several distinctive features: (i) its resolution (1/4° in the ocean, 60 km in the midlatitude atmosphere), (ii) sub-daily output at the surface and on atmospheric pressure levels, and (iii) the provision of boundary conditions for regional modelling. In this study, the design of the CANARI LE is documented and key aspects of historical model performance are evaluated, while applications of the SSP3-7.0 simulations are left for forthcoming studies. Model evaluations are carried out for global mean surface temperature, North Atlantic jets and storm tracks, Northern Hemisphere stratospheric circulation, Atlantic meridional overturning circulation and subpolar gyre, Arctic sea ice, and El Niño-Southern Oscillation teleconnections. In addition, an example is provided deriving analogues of the North Atlantic 2023 summer marine heatwave showing that the CANARI LE captures unprecedented events of this magnitude and spatial scale. The CANARI LE is a community resource that lends itself for the study of physical coupled climate processes and weather systems in a changing climate. The ability to drive regional models will potentially allow for risk assessments of unprecedented detail in climate impacts and adaptation studies.
- Preprint
(6939 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 12 Oct 2026)
- CEC1: 'Comment on egusphere-2026-2281', Juan Antonio Añel, 12 Aug 2026 reply
-
RC1: 'Comment on egusphere-2026-2281', Anonymous Referee #1, 16 Sep 2026
reply
This paper gives an overview of the new CANARI-LE and provides some nice analysis comparing its output to observations. Overall this is a useful and well written paper and I suggest it is published with some revisions outlined below.
Major points:
First while the output is available on the jasmin computer – is it accessible to the general science community outside of those with access to this system?
The introduction has some references that are not well places and references to some key studies on SMILEs are missing:
- Line 20 – Wood et al 2021 is a odd choice of reference as the only study cited here. I’d suggest adding a e.g. and adding Kay et al who really showed the internal variability in a large ensemble: https://journals.ametsoc.org/view/journals/bams/96/8/bams-d-13-00255.1.xml
-Line 25 Maher et al., 2019 should be referenced after more climate scenarios not for single forcing experiments as they do not do single forcing. Additionally CanESM2 was the first large ensemble with single forcing experiments and should be cited here: https://journals.ametsoc.org/view/journals/clim/30/2/jcli-d-16-0412.1.xml
- Line 25 Rogers et al, 2021 were not the first people to macro initialise and it is a odd choice of single reference ofr this statement
- Other references that are missing:
Overview of large ensemble experiments: https://esd.copernicus.org/articles/12/401/2021/
Multimodel large ensemble archive: https://www.nature.com/articles/s41558-020-0731-2.epdf?shared_access_token=_ThT0M1ilLK_nCG3M4d96tRgN0jAjWel9jnR3ZoTv0NnBpoO0B2TlXC-0PoMuTZLU9os0zosjl9RK_3dqMQAbxcvL4heRNcOyiYMd0p9Hw180TVwDMP6PU58fPRcKCPSqTZqzK_HXurN8KivkTZqGQ%3D%3D and https://gmd.copernicus.org/articles/18/6341/2025/
A specific initialisation using micro/macro and selecting climate states: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023MS003653
While the examples are really great there is a suite of literature on these topics and many have been investigated in large ensembles, can you add a few sentences comparing to previous work in each section?
Generally the summary statements could use some implications. It is clear what was found but not what this means and what the interpretation is.
Finally I was looking for some comparison to tell me the high-res is better or useful in some way and didn't really find it in the paper – can some of this be added? The GFDL-SPEAR LE that has a higher res atmosphere does this to some extent. This paper should also be cited in the introduction: https://www.gfdl.noaa.gov/spear_large_ensembles/
Minor points:
There is now a seasonal comparison of the Suarez-Gutierrez paper if of interest: https://www.nature.com/articles/s41467-025-67005-y no need to cite – just something that might be useful.
I suggest splitting figure 1 in to 2 periods to see the difference between internal variability and forced response. i.e. before the climate change signal and after
Figure 5 – suggest a difference plot between panels b and c
Figure 9 what is the reference in b? - this figure could use more explanation as it was difficult to understand what was going on
section 3.6 – I recommend a quick discussion of ENSO bises themselves before going into impacts, at a minium the nino3.4 strength mentioned can be compared to obs across members
typo -meaned on line 289
why choose JJAS Niño-3.4 I as this is not ENSO peak season??
Figure 10 – suggest using green brown standardised precip colour bar
line 294 =- where is this in figure 10 ‘The figure also shows the cross-ensemble variations in the strength of the regression pattern, as represented by the stan295 dard deviation of the regression coefficients at each location’ – this seems to not be in the figure
Figure 11 – many studies use quaractic dentrending – I suggest adding this
Section 4 needs a more detailed clear description from non analog expert on how these analogs are found – so this really useful and cool analysis can be replicated
352 – can you make more of a useful stronger statement than ‘In summary, the North Atlantic 2023 marine heatwave is a so-called record-shattering event (Fischer et al., 2021) at the scales of an entire season and ocean basin. The occurrence of this type of event is captured by the CANARI LE.’ this doesnt really tell the reader much. Is this event record shattering in CANERI LE, yes the model captures it why is this useful what did you learn?
Line 379 – ‘a little’ is not very specific can you be more precise
Line 385 – this implies model warms more than observations right? Can you say this in the summary
Line 381 – same with ‘is rather high’ this is not specific can you be clearer
386 – can you interpret this into a statement of what this means?
414 typo
Citation: https://doi.org/10.5194/egusphere-2026-2281-RC1
Data sets
CANARI-LE Priority Variables from the HadGEM3.1 large ensemble of climate model simulations for the CANARI project R. K. H. Schiemann et al. https://catalogue.ceda.ac.uk/uuid/4b62fafc68ae4c3784fb2819f652c537
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 180 | 87 | 10 | 277 | 19 | 9 |
- HTML: 180
- PDF: 87
- XML: 10
- Total: 277
- BibTeX: 19
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Dear authors,
I am writing here regarding the compliance with the code availability policy of the journal. We are aware of the legal problems to share the HadGEM3 code. However, it would be good if you could include in the Code and Data Availability section of the manuscript some information about to get access to it, and the adequate measures taken to properly store and preserve the exact version used here at the MetOffice computers.
Juan A. Añel
Geosci. Model Dev. Executive Editor