Empirical patterns of variability in atmospheric and oceanic excitation of polar motion

Abstract

Excitation of polar motion is related in large measure to the redistribution of atmospheric and oceanic mass and to circulation changes. The atmosphere and ocean exhibit spatial patterns that may be isolated by a principal-component type of analysis as fundamental modes explaining their variability. These patterns contribute to polar motion excitation. Here atmospheric excitation functions χ A are computed in equal area sectors (equivalent to 5°×10° boxes at the equator) for the period 1983–2000, based on the reanalyses of the US National Centers for Environmental Prediction–National Center for Atmospheric Research. Oceanic excitation functions χ O are calculated in sectors of comparable resolution, based on the near-global ocean model of Ponte and Stammer (1999) [Geophys. Res 104 23,293,409] for the period 1985–1997. To find patterns of variability in χ A and χ O we examined modes obtained using complex Empirical Orthogonal Function analysis designed to mathematically isolate independent types of variability. EOF analysis, commonly used in studies of climate, separates types of variability that often are caused by an underlying physical mechanism. The first mode is clearly significant in both χ A and χ O and their associated time series have strong annual oscillations with temporally varying amplitudes. The first mode of χ A has strong signals over central Asia, Greenland, Australia, and the southern tips of South America and Africa, and those of χ O are over the mid-latitude North Pacific and North Atlantic as well as areas of the Southern Ocean. Others modes, though only marginally significant, have elements of noted climate signals such as the North Atlantic Oscillation or Pacific-North American patterns.