Skip to main content

Coral reconstructions of central tropical Pacific climate suggest a 25% increase in recent El Niño Southern Oscillation intensity

Recent observations showed that intensities of the El Niño Southern Oscillation (ENSO) were highly variable, thus raising the question whether or not this apparent intensification is significant and related to human-induced global warming. ENSO cycles are one of the most studied climate phenomenon, but yet our understanding of how they are changing in a world with enhanced greenhouse gas concentrations is not clear. CMIP5 climate models predict a robust increase in ENSO variability in a world with higher carbon dioxide concentrations, meaning that extreme El Niño events, such as those in 1997/98 and 2015/16, could occur more frequently. Paleo reconstructions of historical ENSO events, however, provide a critical baseline to compare with contemporary observations.

Estimates of interannual variability from new and published coral records from the northern Line Islands
Estimates of interannual variability from new and published coral δ18O records from the northern Line Islands. a) Relative ENSO variance changes in fossil coral δ18O calculated from the standard deviation of sliding 20-year windows of monthly coral δ18O data that was first 10-year high-passed, then 13-month running averaged, plotted as the average standard deviation of each coral timeseries relative to the 1987-2007 CE intervals of corresponding modern spliced coral δ18O timeseries from each site (the dashed ‘zero’ line). Coral data originate from Palmyra (red) (Cobb et al. 2003), Fanning (blue) (Cobb et al. 2013), and Kiritimati (green) (Cobb et al. 2013; McGregor et al. 2013a; and this study). The bars represent the full range of interannual variability in 20-year windows of each coral sequence. Coral timeseries of <20 years are plotted with open circles. Coral timeseries from 20 years < length < 320 years are plotted linearly with size where smaller dots are shorter sequences. The gray box denotes the full range of ENSO variability from the modern spliced coral records in 20-year windows, as reflected in panel (b). b) Same as in (a) but for the modern spliced coral δ18O timeseries for each island (Palmyra: 1886-2007 CE; Fanning 1949-2005 CE; Kiritimati Island: 1939-2016 CE). Colors as in (a).

In a recent article published in Geophysical Research Letters, Grothe and others compiled monthly records of ocean surface temperature over the last several millennia using fossil coral oxygen isotopes from the Line Islands, central tropical Pacific. This is the most comprehensive Line Islands fossil coral dataset to date, adding over 200 years of new data for a total of almost 1,500 years of regional ocean surface temperature records spanning over the last 7,000 years. The authors measured the changes in ENSO intensity by first removing any long-term trend that may bias the records, as well as the seasonal cycle, and then measured the resulting spread relative to the mean in 20-year windows. Comparison of the fossil corals samples with modern corals showed that the recent intensification was quite apparent – the most recent 20 years was stronger than almost all 20-year periods measured over the entire preindustrial fossil coral dataset. In all, ENSO intensity has strengthened by 25% over the last 30 years compared to the preindustrial period.

The authors assessed the significance of their results by replicating the exact coral dataset using state-of-the-art time-series of ENSO cycles. Using a Monte Carlo statistical method, the exact number and length of corals used in the reconstruction were extracted from long unforced time-series and similarly compared as the preindustrial and industrial coral dataset. After running the analysis 10,000 times, the simulation results indicated that the observed 25% increase in ENSO intensity was significant. More work is needed to identify the underlying response of ENSO to global warming, but this study highlights some of the first evidence that ENSO variability may already be intensifying due to greenhouse gas emissions. The authors do note that there is an observable freshening trend in the central tropical Pacific, but that this change is too small to account for the large increase in ENSO intensity.

Written by
Pamela R. Grothe, Department of Earth and Environmental Sciences, University of Mary Washington

Pamela R. Grothe1†, Kim M. Cobb1, Giovanni Liguori1, Emanuele Di Lorenzo1, Antonietta Capotondi2, Yanbin Lu3, Hai Cheng3,4, R. Lawrence Edwards3, John R. Southon5, Guaciara M. Santos5, Daniel M. Deocampo6, Jean Lynch-Stieglitz1, Tianran Chen7, Hussein R. Sayani1, Diane M. Thompson8, Jessica L. Conroy9, Andrea L. Moore10, Kayla Townsend1, Melat Hagos1, Gemma O’Connor1, Lauren T. Toth11

1School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, Georgia, USA
2NOAA, Earth System Research Laboratory, Physical Sciences Division, Boulder, Colorado, USA
3Department of Earth Sciences, University of Minnesota, Minneapolis, Minnesota, USA
4Institute of Global Environmental Change, Xi’an Jiaotong University, Xi’an, China
5Department of Earth System Science, University of California, Irvine, California, USA
6Department of Geosciences, Georgia State University, Atlanta, Georgia, USA
7South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou China
8Geosciences Department, University of Arizona, Tucson, Arizona, USA
9Department of Geology and Department of Plant Biology, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA
10Department of Earth and Environmental Sciences, University of Mary Washington, Fredericksburg, Virginia, USA
11USGS, St. Petersburg Coastal and Marine Science Center, St. Petersburg, Florida, USA.
Currently at Department of Earth and Environmental Sciences, University of Mary Washington, Fredericksburg, Virginia, USA