Climate models · Extremes
CMIP6 runs hotter — but its regional extremes scale like CMIP5's
Save this development or follow its topic to track what changes.
Short answer
At matched global warming levels, CMIP6 and CMIP5 agree closely on how regional temperature and precipitation extremes scale, despite differing global sensitivity.
What happened
Seneviratne and Hauser compared ETCCDI extreme indices — TXx, TNn and Rx1day — across CMIP5 RCP and CMIP6 SSP ensembles at +1.5, +2 and +4 °C of global warming, mapped the response by AR6 region, and split projection uncertainty into global versus regional transient climate sensitivity. Multimodel-mean regional sensitivity was very similar between generations. At +1.5 °C, regional-sensitivity uncertainty often exceeded global-sensitivity uncertainty, particularly in CMIP6. Local differences included hotter South American monsoon extremes and a wetter West Africa and Sahel.
Why it matters
CMIP6's higher equilibrium climate sensitivity prompted the worry that regional impact projections would have to be redrawn. Scaling to warming level instead of forcing shows the regional patterns are largely unchanged — and shifts attention to constraining the land processes that set regional sensitivity, not only global sensitivity.
Evidence
- Study type
- Multimodel comparison of ETCCDI extreme indices scaled to shared global warming levels
- Sample
- CMIP5 RCP and CMIP6 SSP ensembles evaluated at +1.5, +2 and +4 °C
- Journal
- Earth's Future · peer reviewed
- Replication
- Consistent with subsequent warming-level analyses of the fuller CMIP6 archive
- Limitations
- The CMIP6 archive was incomplete as of late March 2020. The uncertainty decomposition at +1.5 °C does not carry over to +4 °C, and the paper does not diagnose why the South American monsoon or Sahel responses differ.
What this connects to
Sources
The 2 studies this explanation is built from, by the role each plays. Every source links to PaperFren’s explanation of it and to the original paper.
Primary study
- Regional extreme sensitivity similar in CMIP5 and 6
CMIP6 and CMIP5 agree on how hot days, cold nights, and heavy rain scale with global warming, even though global climate sensitivity differs.
What it does not showLimitations
The CMIP6 archive was incomplete (late March 2020). The +1.5 °C uncertainty split does not simply carry to +4 °C because the global-warming range among models widens. The analysis does not diagnose why SAM or Sahel differ.
PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by
Supporting evidence
- Europe’s first continent-wide 2.2 km rain projections
A 2.2 km convection-permitting model projects wetter northern winters and drier summers, with summer hourly extremes unlike coarser GCMs.
What it does not showLimitations
Only 10-year time slices and a single GCM/RCP8.5 path; no intermediate nest; E-OBS under-samples hourly extremes; Central European summer-extreme decline is not significant.
PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by
Before
Higher climate sensitivity in CMIP6 raised the expectation that regional extreme projections would differ substantially from CMIP5.
Now
Regional scaling with warming level is robust across generations, and at +1.5 °C the regional term dominates the uncertainty. The CMIP6 archive was still incomplete when this ran, and the analysis does not explain the regional differences it finds.