Jemma Jeffree: Asymmetry in ENSO prediction is independent of ENSO skewness

ESM/Land/CMIP7 Workshop 2026 ESM/Land/CMIP7 Workshop Abstracts 2026
1

Jemma Jeffree: Asymmetry in ENSO prediction is independent of ENSO skewness

  1. The El Niño Southern Oscillation (ENSO) is a mode of climate variability in the equatorial Pacific, with impacts on seasonal climate both in Australia and around the world. During the observational record, the transition between an El Niño (positive ENSO phase) and La Niña (negative ENSO phase) has been particularly predictable compared to other transitions.

  2. The ENSO in ACCESS-ESM1-5 is negatively skewed (El Niños smaller than La Niñas), in contrast to the positively skewed observational record (El Niños bigger than La Niñas). We wanted to ask whether the skewness of the observational record was related to the prediction asymmetry. I.e. we hypothesised that perhaps it was the size of observed El Niños rather than the fact that they were El Niños which lead to better prediction skill in the year afterwards. We ran an initialised ensemble of ACCESS-ESM1-5 with pre-industrial forcing, to calculate predictability in this alternate world with negative ENSO skewness. We also made some statistical forecasts with the ACCESS-ESM1-5 large ensemble to compare a larger spread of internal variability with observations.

  3. We found the same predictability asymmetry in ACCESS-ESM1-5 as in observational forecasts from a positively skewed model. This suggests that the processes contributing to asymmetry in forecasts are more nuanced than just “big events lead to predictability.”

  4. This talk is pitched to be accessible to any climate researcher, and may be of particular interest to those who research ENSO, other internal variability, seasonal predictions, or coupled model biases.

  5. ENSO, ACCESS-ESM1-5, coupled model, seasonal forecast