The Asymmetric Influence of Ocean Heat Content on ENSO Predictability

Abstract:

The El Niño-Southern Oscillation (ENSO) is the most energetic climate phenomenon at interannual timescales on Earth. It can deeply affect the global atmospheric circulation, causing droughts and floods, modulating tropical cyclones activity and impacting ecosystems and agriculture worldwide. Due to the large impacts of ENSO, a lot of effort has been made to improve its prediction and it is now possible to skillfully predict it up to 3 seasons in advance. In order to increase this lead-time and break the spring predictability barrier, we need to use the ocean memory associated with the western Pacific heat content. Indeed, recharge oscillator theory suggests that an unusually high heat content should lead to an El Niño and low heat content to La Niña, about 1 year later.

Here, we investigate if the predictability of ENSO depends on the initial state recharge, and discuss the underlying mechanisms. Observations and ensemble simulations computed with a global climate model indicate that discharged states evolve more systematically into La Niña events than recharged states into El Niño events. Model experiments, initialized in boreal fall from either recharged or discharged heat content and sampling the full range of corresponding ENSO phases, confirm that discharged states yield a more predictable ENSO outcome 1 year later than recharged states. We found that, as expected, recharged states evolve into positive central Pacific sea surface temperature anomalies in boreal spring. This induces stronger and more variable westerly wind event activity and a stronger wind stress feedback, causing a fast growth in the ensemble spread from boreal summer to the ENSO peak. These mechanisms are most likely the cause for the less predictable evolution of recharged heat content initial states in the model simulations.