Seasonality in an empirically derived Markov model of tropical Pacific seasurface temperature anomalies

An empirically derived linear dynamical model is constructed using the Comp rehensive Ocean-Atmosphere Data Set enhanced sea surface temperature data i n the tropical Pacific during the period 1956-95. Annual variation in the M arkov model is sought using various tests. A comparison of Nino-3.4 forecas t skill using a seasonally varying Markov model to forecast skill in which the seasonal transition matrices are applied during opposite rimes of the y ear from which they were derived is made. As a result, it is determined tha t the seasonal transition matrices are probably not interchangeable, indica ting chat the Markov model is nor annually constant. Stochastic forcing, wh ich has been hypothesized to exhibit seasonality, is therefore not the sole source of the annual variation of El Nino-Southern Oscillation (ENSO) dyna mics and the phase locking of ENSO events to peak during November.