observational data sets. Instrumental records also are critical in characterizing patterns of climate variability that might have contributed to paleoclimatic abrupt change, and might contribute to abrupt climate change in the future. It is important to the understanding of abrupt climate change that these patterns or “modes” of circulation and its variability be understood, particularly on the time-scale of decades to centuries. The abrupt changes surveyed here are smaller in strength than the extreme events of the paleoclimate record, yet they are nonetheless significant as human populations press the capacity of the environment, locally and globally.

Atmospheric instrumental data include surface values and vertical profiles of numerous physical variables, including temperature, pressure, radiation, and winds. Surface observations, satellite radiometric observations, and the global network of regularly launched radiosonde profilers are assimilated into computer models of the atmosphere to analyze weather and climate. They capture both the conditions that cause atmospheric circulation and the resulting atmospheric motions. Much of our current understanding of climate comes from the relatively accurately observed period since 1950. More subtle are the measurements of trace chemicals, which both affect the physical state of the atmosphere, and can be used to infer its motions. The longest atmospheric time series, dating back several hundred years, are surface temperature and pressure.

The ocean, like the atmosphere, is a thin fluid envelope covering much of the earth. Satellites are now collecting global observations of the temperature, elevation and roughness of the sea surface, which tell us the surface currents and winds fairly accurately. Crucial climate variables, such as sea-ice cover and movement (and to a lesser accuracy, ice thickness), have been measured by satellites beginning in the 1970s. Yet, oceanic data are still more restricted in coverage and duration than atmospheric data, for it is still difficult to penetrate the depths of the ocean with instruments in sufficient numbers.

In addition to the purely instrumental problem, ocean currents and eddies are smaller in size than major atmospheric wind fields, making the mapping of ocean circulation more difficult (weather patterns are well matched in size to the spacing of major cities, which historically made their discovery possible, using simple barometers). Another contrasting property is the time for fluid to adjust fully to a change in external forcing: in the atmosphere this time is a month or two, while in the ocean it is measured in millennia. The ocean dominates the global storage of heat, carbon, and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement