The atmosphere intrinsically generates chaotic fluctuations. The thermal and pressure anomalies created in the atmosphere are damped by radiative loss and surface drag on timescales less than a month. The intrinsic excitation of NAO pattern is limited to a period of time less than a few days.
 

If the NAO pattern is coupled to an ocean which retains a simple memory of the previous year's NAO state, the large one-year lag correlation in NAO indices are due to red noise processes. Stephenson et. al. (2000) show that the autocorrelations of the NAO SLP index retain small magnitudes even up to larger lags. This behaviour differs from the fast exponential decay expected for short­range processes such as red noise. Rodwell et. all (1999) obtained a statistically significant correlation between model simulated and observed NAO index only in the last 50-year period of the interval 1897-1997.
 

If the NAO couples to oceanic processes involving resonant oscillatory modes, spectral peaks or other spectral features inconsistent with a red or white spectrum should be observed. The NAO indices contain a broad spectrum of variations with significant variance at biennial and 6-10 years periods (Hurrell and Van Loon, 1997; Pozo-Vazquez et al., 2001). The contributions of the biennial component of variability have clearly marked teleconnections (Stephenson et al., 2000). Much of the variance at biennial periods comes from the early period (1875-1939), while the variability between 6 and 10 years is present throught the record but has become most pronounced over the latter half-century (Hurrell and Van Loon, 1997; Pozo-Vazquez et al., 2001).

Upward and downward decadal trends are also present in the historical record of the NAO. In recent years, some observational evidence is found for interdecadal climate oscillations over the North-Atlantic area (Deser and Blackmon 1993, Sutton and Allen 1997). This climate oscillation is characterized by changes in the strength of the westerlies and large scale propagating sea surface temperature anomalies. The timescale of the oscillation is about 10-15 years. One may argue that even complete understanding of a hypothetical feedback mechanism producing enhanced interdecadal variability would not lead to much predictability of the NAO signal.
 

Paleoclimate evidence suggests that NAO variability is highly intermittent and does not exhibit a preferred time scale (Appenzeller et al. 1998a). Appenzeller et al. (1998) showed by means of wavelet analysis, that in a 1400-year simulation of the ECHAM3 General Circulation Model (GCM) developed at the Max-Planck-Institute in Hamburg, as well as in ice-core data, the dominant frequencies of the NAO-index changes in time. Another indication that the NAO may change its regime is the strong positive trend of the index since the late 1960s. During this latter part of the record, an 8-year oscillation may be observed.