Enhancing predictability by increasing nonlinearity in ENSO and Lorenz systems

Abstract. The presence of nonlinear terms in the governing equations of a dynamical system usually leads to the loss of predictability, e.g. in numerical weather prediction. However, for the El Niño-Southern Oscillation (ENSO) phenomenon, in an intermediate coupled equatorial Pacific model run under the 1961–1975 and the 1981–1995 climatologies, the latter climatology led to longer-period oscillations, thus greater predictability. In the Lorenz (1963) 3-component chaos system, by adjusting the model parameters to increase the nonlinearity of the system, a similar increase in predictability was found. Thus in the ENSO and Lorenz systems, enhanced nonlinearity from changes in the governing equations could produce longer period oscillations with increased predictability.