Beyond the deterministic limit where the initial value sensitive predictability can hardly be found, the boundary condition dependent potential predictability is examined. The tropical anomalies of the opposite phases of El Nino/Southern Oscillation (ENSO) can significantly change the extratropical natural variability on a wide range of spatial and temporal scales. We address the impact on the low-frequency variability, such as the persistent blocking flows, and the high-frequency variability, represented primarily by the storm tracks. The NCEP/NCAR reanalyses are used for this investigation. Several diagnostics tools help to reveal the dynamical processes leading to the large change of the natural variability and the potential predictability in the extratropical latitudes between these two phases of the ENSO cycle. During El Nino winters, the principal storm tracks are steered more into the southern and Baja California region, by the much eastward extended subtropical jets. On the other hand, the storms are diverted more into the higher latitudes (Aleutians and Gulf of Alaska) during La Nina winters, when jetstreams are much weaker east of 160 degrees W. Although being passively steered to widely different regions, the high-frequency transients do feed back actively to strengthen and maintain the subtropical jet across the central North Pacific and also act to slow down the equatorward flank of the jet. The feedback by the transients is stronger during the El Nino than the La Nina winters, helping in maintaining stronger signals from the tropics for the El Nino winters. There is also a large change of low-frequency variability: much larger magnitude of kinetic energy and height variance during La Nina than El Nino winters. The local barotropic energy diagnosis reveals that, on average, the low-frequency components extract more energy from time-mean flows during La Nina than El Nino winters, helping in explaining the presence of much larger low-frequency variability during the La Nina winters. With stronger ENSO signals and weaker natural variability during El Nino winters, the potential predictability in the north Pacific sector is significantly higher, on these two counts, than during the La Nina winters.
