An outsider's review of the astronomical theory of the climate: is the eccentricity-driven insolation the main driver of the ice ages?

Although the astronomical theory of the climate (also known as the Milankovitch theory) explains a number of key features of the paleoclimate time series (especially frequency and phase distribution) in terms of orbital influence on the waxing and waning of ice sheets, it still faces several important unresolved challenges. In this review, we discuss five main challenges we believe the theory must resolve in order to survive: (1) The large amplitude of the similar to 100 ky ice age cycles; (2) the switch from 41-ky- to 100-ky-long glaciations around 0.9 Ma BP; (3) the absence of significant spectral power at 413 ky over the past 1.2 Ma; (4) the variation in glacial cycle duration in the last 800 ky; and (5) the presence of non-orbital spectral peaks in the climate record. But in spite of these problems, we conclude that the Milankovitch theory may still be a viable one; at least in the original sense of Hays et al. [Science 194 (1976) 1121], that orbital changes somehow influenced the ice ages. Many investigators, ourselves included, appear to see the impact of orbital forcing in the data, clear evidence of astronomical forcing, as well as evidence which suggests that the climate system responds nonlinearly to all Milankovitch (orbital) frequencies. (C) 2001 Elsevier Science B.V. All rights reserved.