PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 8, 2026Geophysical Research Letters0 citationsOpen Access

Understanding CMIP6 Multi‐Model Ensemble Projected Pacific Warming Pattern Variability

View Full Paper
SMShayne McGregorEOEllie Q. Y. OngYPYann Planton

Key Points

  • This research aims to investigate the variability in projected Pacific warming patterns from CMIP6 climate models.
  • Analyzed multi-model ensemble outputs from CMIP6 climate models.
  • Compared observed SST trends with model projections.
  • Evaluated model representations of the mean state in the Pacific Ocean.
  • Most models show an 'El Niño-like' warming pattern, while observations reflect a 'La Niña-like' trend.
  • Dominant variability patterns linked to hemispheric warming and equatorial gradients are identified.
  • Models with realistic mean states better capture observed trends and project future patterns.

Abstract

Abstract Most CMIP6 climate models simulate a Pacific SST warming pattern through the 20th and 21st centuries that is “El Niño‐like,” with a weakened zonal equatorial gradient. However, observed trends are “La Niña‐like,” displaying a strengthened zonal equatorial gradient, raising concerns about the accuracy of these projections. Here we explore multi‐model ensemble variability in the projected Pacific warming pattern. This pattern variability is largely explained by two dominant patterns: one linked to hemispheric warming asymmetries, and the other to zonal equatorial gradients. Crucially, model differences in projected zonal equatorial gradients are strongly tied to the model representation of the mean state in the off‐equatorial east and west Pacific. This implies that mean state biases partly control the projected warming pattern. Moreover, models with more realistic mean states tend to produce: (a) historical trends with stronger gradients, aligning better with observations, and (b) more “El Niño‐like” future projections.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

McGregor et al. (2026) studied this question.

synapsesocial.com/papers/698829410fc35cd7a8849632https://doi.org/10.1029/2025gl118815
Ask AI
Helpful
Bookmark
Share
View Full Paper