Effects of Extreme La Niña Events on the Asymmetry of ENSO Impacts

The positive and negative phases of extreme El Niño-Southern Oscillation (ENSO) events are marked by asymmetrical atmospheric responses and coupled feedbacks. For example, nonlinear feedbacks within the tropical ocean-atmosphere system, including the thermocline feedback and convection response to ENSO phase, allow for extreme El Niño events that are greater in magnitude than the strongest La Niña events in our current climate. The strongest El Niño events, such as the 2015-2016 event with a peak amplitude SST anomalies of 2.5°C during the boreal winter (DJF), can yield different local responses, including the related precipitation and wind stress anomalies in the central Pacific, compared to weak and moderate El Niño events or La Niña. Atypical effects are also seen in their teleconnections such as strong eastward shifted temperature and pressure anomalies over North America. This contrasts with the local and remote impacts of the strongest La Niña events such as the 1983-1984 event with a DJF peak of -1.8°C. Current observational records provide a relatively small sample size of observed ENSO events, with no recorded La Niña events approaching amplitudes similar to the strongest El Niño events, limiting our capacity for understanding such events. Some future climate projections, however, suggest the possible occurrence of high amplitude La Niña events. We propose an idealized coupled modeling approach to produce large sample sizes of varied ENSO amplitudes of both phases to study their nonlinear effects in the tropics and extratropics.

To isolate and better understand the asymmetric impacts of the ENSO phases, we perform a suite of experiments using the Community Earth System Model 2 (CESM2) that systematically “fix” the strength and frequency of the ENSO cycle. To do so, we force the tropical Pacific Ocean with climatological wind stress to disengage the Bjerknes feedback and then prescribe anomalous ENSO wind stresses with a period of 2 years. This yields a perfectly periodic ENSO cycle, while the model freely evolves and remains thermodynamically coupled outside of the tropical Pacific. We perform three experiments within this perfectly periodic framework: high (2.5°C), moderate (1.5°C), and low (0.5°C) amplitude ENSO, each yielding 100 ENSO cycles. Comparison across the three experiments allows us to identify the nonlinear impacts of ENSO amplitude on the tropical Pacific and the extratropics. Notably, this framework allows for extreme La Niña of equal amplitude to realistic extreme El Niño, thus allowing us to determine whether extreme ENSO of opposite phases drive similar or asymmetric impacts. Our results show that extreme La Niña events of comparable strength to realistic extreme El Niño events do in fact drive similar teleconnection patterns, thus the asymmetrical temperature and precipitation impacts between El Nino and La Nina often mentioned in the literature vanishes with the presence of extreme La Niña events. Our results suggest that if in the future extreme La Niña events do become more common, we may observe a reduction in the asymmetry of ENSO impacts as well.