Abstract:Forecasts of El Niño climate events are routinely provided and distributed, but the limits of El Niño predictability are still the subject of debate. Some recent studies suggest that the predictability is largely limited by the effects of high-frequency atmospheric ‘noise’, whereas others emphasize limitations arising from the growth of initial errors in model simulations. Here we present retrospective forecasts of the interannual climate fluctuations in the tropical Pacific Ocean for the period 1857 to 2003, using a coupled ocean–atmosphere model. The model successfully predicts all prominent El Niño events within this period at lead times of up to two years. Our analysis suggests that the evolution of El Niño is controlled to a larger degree by self-sustaining internal dynamics than by stochastic forcing. Model-based prediction of El Niño therefore depends more on the initial conditions than on unpredictable atmospheric noise. We conclude that throughout the past century, El Niño has been more predictable than previously envisaged. Present estimates of El Niño’s predictability are mostly based on retrospective predictions for the last two or three decades, encompassing a relatively small number of events. With so few degrees of freedom, the statistical significance of such estimates is questionable. In principle, predictability can also be estimated by perturbing initial conditions in numerical model experiments, but the answer is model dependent, and existing models have not been shown to be realistic enough for this purpose. El Niño is evident in instrumental observations dating back to the mid-nineteenth century and in proxy data sets over much longer periods, but no successful attempt to ‘hindcast’ the historic El Niño events before the mid-twentieth century has been reported. This is due partly to the lack of adequate data for model initialization and partly to the inability of present models to make effective use of available data. The study reported here represents the first (to our knowledge) retrospective forecast experiment spanning the past one-and-a-half centuries, using only reconstructed sea surface temperature (SST) data for model initialization. The intrinsic predictability of El Niño is surely limited, but there has been considerable debate about what the limitations really are. Classic theories consider El Niño and the Southern Oscillation (ENSO) as a self-sustaining interannual fluctuation in the tropical Pacific, being chaotic yet deterministic. Thus its predictability is largely limited by the growth of initial errors, and the potential forecast lead time is likely to be of the order of years. On the other hand, some recent studies emphasize the importance of atmospheric noise, particularly the so-called westerly wind bursts in the western equatorial Pacific. In such a scenario, ENSO is a damped oscillation sustained by stochastic forcing, and its predictability is more limited by noise than by initial errors. This implies that most El Niño events are essentially unpredictable at long lead times, because their development is often accompanied by high-frequency forcing. Such a view is not supported by the present findings. The observed and predicted SST anomalies averaged in the central equatorial Pacific are shown in Fig. 1a. (See Methods section

El Niño Modoki thus far can be mostly predicted more than 10 years ahead of time

El Niño Modoki can be mostly predicted more than 10 years ahead of time

Predictability of El Niño over the past 148 years

El Niño and Its Predictability : 1856-2003

Understanding the Low Predictability of the 2015/16 El Niño Event Based on a Deep Learning Model

El Niño prediction and predictability

Explainable El Niño predictability from climate mode interactions

A lower limit for atmospheric carbon dioxide levels 3 . 2 billion years ago

Explained predictions of strong eastern Pacific El Niño events using deep learning

Strong El Niño Events Lead to Robust Multi‐Year ENSO Predictability

Predicting the 2023/24 El Niño from a Multi-Scale and Global Perspective

The first kind of predictability problem of El Niño predictions in a multivariate coupled data‐driven model

Quantifying the Predictability of ENSO Complexity Using a Statistically Accurate Multiscale Stochastic Model and Information Theory

Predictability of the California Niño/Niña*

Quantifying the Relative Contributions of the Global Oceans to ENSO Predictability With Deep Learning

Deciphering Super El Niño: Development of a Novel Predictive Model Integrating Local and Global Climatic Signals

Increased Predictability of Extreme El Niño From Decadal Interbasin Interaction

The 2023/24 El Niño and the Feasibility of Long-Lead ENSO Forecasting

Keywords : Predictability of the California Nino / Nina *

Predicting El Niño Beyond 1-year Lead: Effect of the Western Hemisphere Warm Pool

Improved El Niño-Forecasting by Cooperativity Detection