Collaborative Research: El Nino/Southern Oscillation (ENSO) Predictability--Initial Condition Signal versus Uncoupled Atmospheric Noise

The equatorial Pacific has an outsized influence on weather and climate as it is home to the El Nino/Southern Oscillation (ENSO) phenomenon, featuring an alternation between warm equatorial sea surface temperatures (SSTs) during El Nino events and cold SSTs during La Nina events. These SST changes affect the global atmospheric circulation and thus induce long-range impacts including changes in the severity of winters in Canada, the distribution of rainfall in the Asian monsoon, and Atlantic hurricane activity. These worldwide impacts have motivated intensive efforts to develop ENSO prediction systems, which show some promise but so far have had mixed results.Here the PIs hypothesize that the mixed success of ENSO predictions is intrinsic to ENSO, meaning that some ENSO events are in fact more predictable than others. In that case a "forecast of the forecast skill", meaning a method to determine in advance the likely accuracy of an ENSO prediction, would have great practical value. The Principal Investigators (PIs) are particularly concerned with the skill of ENSO predictions made before the onset of the event, which are skillful to the extent that ENSO events develop from some form of preconditioning such as the buildup of heat in the subsurface equatorial ocean. Model simulations conducted previously by the PIs suggest that preconditioning is not a necessary condition for ENSO events, which can initiate rapidly starting from neutral conditions. Such events, arising purely from the chaotic growth of very small initial perturbations of the atmosphere-ocean system, would be inherently unpredictable. But the PIs argue that some forms of preconditioning can render the system less susceptible to noise-level perturbations and instead drive the evolution of an El Nino or La Nina event in a deterministic manner, allowing skillful predictions several months ahead of the onset.The PIs test their hypothesis through numerical simulations performed with the Community Earth System Model (CESM). The work takes advantage of procedures they have developed for strictly controlling the state of the equatorial Pacific in CESM simulations, so that various forms of preconditioning can be imposed or excluded. Additional work uses the PI's Interactive Ensemble (IE) technique (see for example AGS-1558837) to consider the role of random weather fluctuations unrelated to ENSO in limiting the predictability of ENSO events.The work is of societal as well as scientific interest given the global impacts of ENSO events, and results of this research have direct bearing on attempts to improve ENSO forecasts. The software developed for the project is made available to the research community, along with relevant datasets, to facilitate adoption of methods and results by the research and operational communities. In addition, the PIs participate in summer internship programs targeting high school and college students through Miami-Dade College and the American Indian Higher Education Consortium.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.