The Southern Hemisphere has less instrumental coverage, in space and time, than the Northern Hemisphere. While many of the earlier papers in the section concentrate on the latter, D.J. Karoly describes the observed variability in the atmospheric circulation south of the equator.
Atmosphere-ocean interaction is involved in the last two papers of this section. C. Deser and M.L. Blackmon review atmospheric climate variations at the surface of the North Atlantic, while D.R. Cayan and his associates study a general-circulation model simulation of the Pacific Ocean, driven by observed surface fluxes.
Atmospheric models have a relatively long tradition in the climate community, going back to the 1950s. They span a spectrum, from simple radiative-convective models in one vertical dimension, through energy-balance models in one and two horizontal dimensions, to fully three-dimensional general-circulation models. These models are of interest in their own right as important tools for investigating climate change, either by themselves or coupled to models of other climate subsystems. They also provide an instructive example of the creation of a full suite of models for intercomparison and validation; their development is only now being followed, more or less closely, by the modeling enterprise in oceanography, hydrology, and other disciplines contributing to the climate-change enterprise.
Modeling and observations are closely linked by the reciprocal problems of simulating the observed variability and validating the existing models. The Atmospheric Modeling section is thus appropriately begun by T.M.L. Wigley's and S.C.B. Raper's paper on modeling and interpreting paleoclimate data, with an emphasis on the greenhouse effect. G.R. North and K.-Y. Kim apply classic time-series analysis techniques to climate-signal detection. R.S. Lindzen then considers a few themes in these two papers from complementary viewpoints.
A number of causes for climate variability on time scales of decades to millennia are reviewed by D. Rind and J.T. Overpeck. They emphasize the modeling approach to a study of these causes, while J.M. Wallace addresses similar issues by analyzing the climate record.
The two sections of this chapter are introduced by essays, one by T.R. Karl and the other by M. Ghil. They provide an overview of the current state of the fields of atmospheric observations and atmospheric modeling, reviewing those topics not covered by the workshop and offering a perspective on the papers included. Following each paper, a commentary by the discussion leader and a condensation of the spirited discussion that took place at the workshop shed additional light on our current knowledge in both areas.
The ocean is examined in Chapter 3, and atmosphere-ocean interaction is explored further in Chapter 4. Also of interest are the proxy records of climate variability, such as tree rings and coral reefs, which are covered in Chapter 5. The sponsoring committee's conclusions, which draw on the material in this chapter and in the other three, appear in Chapter 6.