This study uses NCEP–NCAR reanalysis data covering the boreal winters of 1958–97 to examine the power spectral, timescale, and climate noise properties of the dominant atmospheric teleconnection patterns. The patterns examined include the North Atlantic oscillation (NAO), the Pacific–North American (PNA), and west Pacific (WP) teleconnections, and a spatial pattern associated with ENSO. The teleconnection patterns are identified by applying a rotated principal component analysis to the daily unfiltered 300-mb geopotential height field. The NAO and PNA were found to be the two dominant patterns on all timescales. The main finding is that the temporal evolution of the NAO, PNA, and WP teleconnections can be interpreted as being a stochastic (Markov) process with an e-folding timescale between 7.4 and 9.5 days. The time series corresponding to the ENSO spatial pattern did not match that of a Markov process, and thus a well-defined timescale could not be specified. The shortness of the above timescal... Abstract This study uses NCEP–NCAR reanalysis data covering the boreal winters of 1958–97 to examine the power spectral, timescale, and climate noise properties of the dominant atmospheric teleconnection patterns. The patterns examined include the North Atlantic oscillation (NAO), the Pacific–North American (PNA), and west Pacific (WP) teleconnections, and a spatial pattern associated with ENSO. The teleconnection patterns are identified by applying a rotated principal component analysis to the daily unfiltered 300-mb geopotential height field. The NAO and PNA were found to be the two dominant patterns on all timescales. The main finding is that the temporal evolution of the NAO, PNA, and WP teleconnections can be interpreted as being a stochastic (Markov) process with an e-folding timescale between 7.4 and 9.5 days. The time series corresponding to the ENSO spatial pattern did not match that of a Markov process, and thus a well-defined timescale could not be specified. The shortness of the above timescal...