New paper shows no 'average' change in El Ninos due to CO2

A paper published today in Geophysical Research Letters reports "the overall response to CO2 increases is determined using 27 [climate] models, and the ENSO [the El Nino Southern Oscillation] amplitude change based on the multi-model mean is indistinguishable from zero." Alarmists, such as Kevin Trenberth, claim that increased CO2 causes an increase in the frequency and intensity of El Ninos, but this paper finds that the mean response from climate models does not support such claims.

GEOPHYSICAL RESEARCH LETTERS, doi:10.1029/2012GL052759

Significant Changes to ENSO Strength and Impacts in the Twenty-First Century: Results from CMIP5

Key Points

• ENSO amplitude is insignificant in the majority of IPCC-class models.
• ENSO amplitude change is not due to mean state or seasonal cycle changes.
• The teleconnection response is sensitive to the ENSO amplitude change.

Author:

Samantha Stevenson

Changes to the El Ni\~{n}o/Southern Oscillation (ENSO) and its atmospheric teleconnections under climate change are investigated using simulations conducted for the Coupled Model Intercomparison Project (CMIP5). The overall response to CO2 increases is determined using 27 models, and the ENSO amplitude change based on the multi-model mean is indistinguishable from zero. However, changes between ensembles run with a given model are sometimes significant: for four of the eleven models having ensemble sizes larger than three, the 21st century change to ENSO amplitude is statistically significant. In these four models, changes to SST and wind stress do not differ substantially from those in the models with no ENSO response, indicating that mean changes are not predictive of the ENSO sensitivity to climate change. Also, ocean vertical stratification is less (more) sensitive to CO2 in models where ENSO strengthens (weakens), likely due to a regulation of the subsurface temperature structure by ENSO-related poleward heat transport. Atmospheric teleconnections also show differences between models where ENSO amplitude does and does not respond to climate change; in the former case El Ni\~{n}o/La Ni\~{n}a-related sea level pressure anomalies strengthen with CO2, and in the latter they weaken and shift polewards and eastwards. These results illustrate the need for large ensembles to isolate significant ENSO climate change responses, and for future work on diagnosing the dynamical causes of inter-model teleconnection differences.