Cai, W., G. Wang, A. Santoso, M.J. McPhaden, L. Wu, F.-F. Jin, A. Timmermann, M. Collins, G. Vecchi, M. Lengaigne, M.H. England, D. Dommenget, K. Takahasi, and E. Guilyardi (2015), Increased frequency of extreme La Nina events under greenhouse warming, Nature Clim. Change, 5, 132-137. doi:10.1038/nclimate2492
The El Niño-Southern Oscillation (ENSO) is a naturally occurring climate phenomenon that originates in the tropical Pacific Ocean. It is neither strictly atmospheric nor oceanic, but rather a coupled oscillation between warm (El Niño) and cold (La Niña) events alternating about every 5 years (figure 1). Media and news agencies worldwide have devoted much of their attention to ENSO due to its far-reaching influences on severe weather patterns and effects on marine and terrestrial ecosystems. For many, the uncertainty of how ENSO will fair under future climate conditions is troublesome. However, a new study, led by Dr. Wenju Cai of the Australian Commonwealth Scientific and Industrial Research Organization (CSIRO), reveals that continued present-day greenhouse gas emissions will significantly double the frequency of extreme La Niña events well into the 21st-century. This study comes after the discovery of more frequent El Niño events due to human-induced changes in tropical atmospheric circulation.
One might ask, how can both extreme El Niño and La Niña events become more frequent if they are opposite? This study examines a suite of climate models to understand the future frequency of extreme La Niña events and how big El Niño events may be conducive towards the development of subsequent extreme La Niña events.
The authors examine the spatial and temporal variability of sea surface temperature (SST) in the tropical Pacific Ocean during the modern satellite-era and during the seasonal period when La Niña events peak (Dec-Feb). The dominant pattern of SST variability appears canonical with cold SST in the central-eastern equatorial Pacific and warm SST anomalies in a horseshoe-like pattern in the western equatorial Pacific (figure 2a). Extreme La Niña events are defined when central Pacific SSTs are below average conditions by at least 1.75ºC (figure 2b, black contour line).
This study also performed climate model simulations with historical human-caused and natural forcings, as well as future greenhouse gas emission projections. Twenty-one Coupled Model Intercomparison Project Phase 5 (CMIP5) models captured the dynamics required to produce extreme ENSO events. Of these 21 models, the authors compared the frequency of extreme La Niña events in the Climate Control period (1900-1999) with that of the Climate Change period (2000-2099).
From the 21 models analyzed, 17 suggested an increase in the frequency of extreme La Niña events in the Climate Change period. Of those 17 models, approximately 75% of the extreme La Niña events directly followed an extreme El Niño event. These results were supported by strong inter-model agreement. If greenhouse gas emissions continue like business as usual, it is projected that an extreme La Niña event will occur once every 13 years instead of one event every 23 years. Furthermore, modeling experiments show a 73% increase in extreme La Niña events in the 21st-century compared to the 20th-century.
Previous studies have shown that the projected slowdown in Walker Circulation (circulation of the lower atmosphere in the tropics) underpins the projected increase in frequency of extreme El Niño events under greenhouse gas forcing. La Niña, however, is characteristic of enhanced Walker Circulation, so this makes more frequent La Niña events in a warmer climate seem counterintuitive. The authors point out that the mechanism for extreme La Niña is not opposite and equal to extreme El Niño. This is an important conclusion from the study and suggests that more frequent extreme El Niño events may be conducive to the conditions that support the development of extreme La Niña. Supporting evidence of this argument is represented by the succession of the 1998-99 La Niña after the 1997-98 El Niño. Additional processes that are independent of Walker Circulation may also make extreme La Niña events more likely under climate change and these are primarily attributed to the increase in ocean surface temperatures due to greenhouse warming.
Severe weather conditions associated with extreme La Niña events in the 20th century will occur more frequently in the 21st century. These events include devastating floods in the western Pacific and central Americas, drought over the southwestern United States and eastern Pacific, and increased Atlantic hurricane and west Pacific cyclone activity (figure 3). More occurrences of extreme weather events and frequent swings of opposite El Niño and La Niña states will make future predictions of ENSO a growing demand and lingering challenge.
Hillary received her MS in oceanography from the University of Maine in 2014 and works in the Ecosystem Modeling Lab at the Gulf of Maine Research Institute in Portland, ME.