An Oblique View of Climate.

One explanation for certain patterns of glaciation in the past invokes a large and comparatively swift decline in the tilt, or obliquity, of the Earth. A provocative hypothesis provides a mechanism by which such a decline could have occurred. How do variations in Earth's orbital and rotational geometry influence climate? Does the climate system, in turn, influence rotation? We all experience the radiative and thermal cycles of night and day, winter and summer. So we are familiar enough with the influence of Earth's rotational and orbital motions on the spatio-temporal pattern of light and temperature to make it easy to imagine how long-term variations in the orbit and rotation would affect climate. Much recent data and model­ling help confirm that principle.

Somewhat further removed from human experience is the notion hat climatic change itself could influence the rotational dynamics of the Earth. The basic idea is quite simple, and involves feedback between Earth's obliquity (the angular separation between the spin pole and or bit pole around the Sun) and its oblateness (departure from spherical symmetry).

First, however, it is useful to recall how orbital and rotational geometry influences climate. The main seasonal cycle is primarily determined by the orientations of the spin axis, and only secondarily by the eccen­tricity of the orbit. Currently, perihelion (Earth's closest approach to the Sun) occurs several weeks after winter solstice in the Northern Hemi­sphere (shortest daylight). However, neither the orbit nor the spin axis in fixed in space. Gravitational interaction with other planets (principally Venus) causes the shape and orientation of the orbit to change on a vari­ety of timescales, with the dominant period near 70, 000 years (70 kyr) and subsidiary oscillations at periods ranging from 50 kyr to 1.9 million years (Myr).

Gravitational torques exerted by the Moon and Sun on the oblate igure of the Earth cause the spin axis to precess with a period of 25.8 kyr. If the orbit plane were fixed, the path of the spin pole would be a circle centred on the orbit pole, keeping the obliquity fixed. However, because the orbit is also precessing, the obliquity oscillates by ± 1° about its present value of 23.5°, with a period of 41 kyr. These obliquity oscilla­tions modulate the seasonal and latitudinal pattern of incident radiation, and thus affect climatic variations.

How then does climate change influence rotation? One way is to change the spin precession rate by changing the oblateness of the Earth's! mass distribution. During major glacial cycles, mass transport between the oceans and ice sheets is sufficient to change the precession rate by about 1%. The net change includes accumulation of continental ice and partially compensating subsidence of the Earth's surface. If the obliquity and oblateness oscillations are exactly in phase, there is no long-term net effect. But if the oblateness lags behind the obliquity, there will be a secular change in obliquity, with a rate that depends on the amplitude and phase of the oblateness variations.

The long-term stability of Earth's climate system is an important question, but one that remains elusive. Despite progress in short-term weather prediction (based on improved quality and quantity of observa­tions, faster computers and better understanding of the system dynam­ics), our understanding of long-term climate dynamics is still quite primitive. Part of the problem, of course, is that the further back into the past we go, the more difficult it is to reconstruct which path the climate sys­tem has followed. When we still don't know what has happened, how can we reconstruct why? In this situation, the role of theoretical models is not so much to explain what actually happened as to broaden our perspectives on the types of behaviours that might have occurred. As always, more work is needed. In this case, distinguishing between the two competing climatic possibilities (equatorial versus global glaciation) should be easily resolvable by searching for contemporaneous high-latitude and low-latitude glacial deposits. Reconstructing an unambiguous obliquity history will be more of a challenge.