Climate models and ocean observations diverge primarily on the pattern of hemispheric ocean warming—specifically, the interhemispheric thermal contrast (IHTC), or the difference in average sea surface temperature (SST) between the Northern and Southern Hemispheres.

This is not a failure of global temperature trends (models and observations align closely on overall ~1.5°C warming since ~1900), but on regional and hemispheric details that matter enormously for atmospheric circulation.

Since 1950, CMIP6 climate models (the current generation used in IPCC assessments) simulate a positive trend in IHTC: Northern Hemisphere (NH) oceans warm faster than Southern Hemisphere (SH) oceans on average. This produces enhanced warming in parts of the North Atlantic and North Pacific in the model ensemble mean.

Observations (from multiple SST datasets) show the opposite: a negative IHTC trend, with relatively faster warming in the SH (or less warming in the NH). The observed trend sits at roughly the 1st percentile of the model distribution—meaning nearly all models get the direction and/or magnitude wrong.

Global-mean ocean temperatures can still align reasonably between models and observations because the hemispheric differences partially cancel out in the average. But the pattern (the contrast) matters critically for large-scale atmospheric circulation.

2026 peer-reviewed research in Nature Communications by Chengfei He and colleagues (including Amy Clement and Mark Cane) is titled exactly: “Climate models exaggerate greenhouse gas impact on recent interhemispheric temperature patterns and tropical climate.”

The paper demonstrates that CMIP6 climate models systematically overestimate the influence of greenhouse gases (GHGs) on recent (since ~1950) ocean temperature patterns between the Northern and Southern Hemispheres. This bias arises from an exaggerated positive wind-evaporation-sea surface temperature (WES) feedback in the tropics and subtropics, and it has direct consequences for projections of tropical rainfall, the position of the rain belt, and related extremes like drought and hurricanes.

The IHTC is simply the average sea-surface temperature (SST) difference between the Northern Hemisphere (NH) oceans and Southern Hemisphere (SH) oceans. It is a key driver of large-scale atmospheric circulation.

Models (CMIP6 ensemble): Simulate a clear positive IHTC trend—NH oceans warm faster than SH oceans. This produces enhanced warming patterns in regions like the North Atlantic and North Pacific.

Observations (multiple SST datasets such as ERSSTv5, HadISST, COBE2): Show the opposite—a negative IHTC trend (relative SH warming or NH cooling). The observed trend falls at roughly the 1st percentile of the model distribution.

Global-mean ocean warming still matches reasonably well between models and observations (both show ~1.5 °C since the early 20th century), but the pattern (the hemispheric contrast) diverges sharply. Pre-industrial control runs confirm this is not just random internal variability.

Models attribute the long-term positive IHTC trend overwhelmingly to GHGs. In reality, the trend is driven more by anthropogenic aerosols (mostly NH industrial emissions that reflect sunlight and cool the surface) plus natural forcings (volcanoes, solar variability).

The mechanism in models is an amplified WES feedback:

1. GHGs cause initial warming (stronger in NH due to more land).
2. This weakens tropical/subtropical trade winds.
3. Weaker winds reduce evaporation from the ocean surface (less cooling).
4. Reduced evaporation allows even more SST warming—especially in the NH—amplifying the IHTC (positive feedback loop).

High-equilibrium climate sensitivity (ECS) models show the largest positive IHTC trends and biggest mismatches with data. Models do a good job simulating multidecadal variability in IHTC (tied to aerosol and natural forcings), but their long-term trend is artificially boosted by this overstated GHG-WES response.

Lead author Chengde He: “The climate models are too sensitive to greenhouse gases.”

“The tropical rain belt, hurricanes, drought, wildfires—they’re all related.”

Because high-ECS models exaggerate the GHG-driven northward ITCZ shift, future northward shifts of the tropical rain belt are likely less pronounced than those models project (under scenarios like SSP5-8.5). Low-ECS models, which better match the observed IHTC trend, show muted shifts.

A bonus from the study: Models simulate multidecadal IHTC variability realistically. Using this as an “emergent constraint,” the authors narrow the effective radiative forcing from aerosol-cloud interactions to −0.6 ± 0.3 W/m²—a “likely” range 57% narrower than the latest IPCC estimate. This helps reduce uncertainty in historical climate forcing without changing the overall picture of GHG-driven warming.

The paper is open access and available at Nature Communications. It does not question overall anthropogenic warming or the reality of GHG forcing—it refines our understanding of regional patterns and tropical feedbacks, which are critical for trustworthy projections of rain, drought, and extremes.

Ongoing improvements in model resolution, aerosol physics, and ocean-atmosphere coupling should help close this gap.

Climate models exaggerate greenhouse gas impact on recent interhemispheric temperature patterns and tropical climate

Published: February 27, 2026, in Nature Communications

Nature Communications volume 17, Article number: 3265 (2026) 

Lead author: Chengde He (Northeastern University)

Colleagues: Amy C. Clement, Mark A. Cane, and others

Abstract

The interhemispheric thermal contrast, defined as the mean sea surface temperature difference between the northern and southern hemispheres, crucially influences tropical climate. Climate models show a positive interhemispheric thermal contrast trend since 1950, with more warming in the northern hemisphere compared to the southern hemisphere, contradicting the observed negative trend. Here we show this discrepancy stems from models overestimating greenhouse gas responses via wind-evaporation-sea surface temperature feedback, while anthropogenic and natural aerosols combine to produce the negative trend in observations. Consequently, models with high equilibrium climate sensitivity exhibit larger discrepancies with observations. Despite model failure to reproduce the trend, the modeled multidecadal interhemispheric thermal contrast variability aligns with observations, enabling a constrained estimate of effective radiative forcing due to aerosol-cloud interactions of, with a “likely” range 57% narrower than the latest IPCC report. Our study further suggests that future northward shifts of the tropical rain belt are likely to be less pronounced than predicted by climate models with high equilibrium climate sensitivity.