Variability of the Earth’s orbit: The astronomical theory - Ice Ages and Interglacials: Measurements, Interpretation, and Models - page 241

ages. Conversely, time periods with peak solar intensity at higher northern
latitudes may trigger feedback processes that cause melting, leading to de-
glaciation. Thus, it is generally accepted that the variability of the Earth's orbit
about the Sun is a primary factor in determining the timing of ice age-interglacial
cycles.
M&M asserted that there are several persuasive reasons to think that
astronomy is responsible (at least to some extent) for the glacial-interglacial varia-
tions observed. One is the coincidence of astronomical frequencies with those
found in ice age data. In their spectral analysis of sediment data, M&M found
that the strongest frequencies that showed up typically had periods of 100,000,
41,000, and 23,000 years, which are readily identified with periods corresponding
to eccentricity, obliquity, and precession. A second reason pointed out by M&M is
that over long periods ( 800,000 years), these oscillations remain coherent
(i.e., they maintain a relatively constant phase). However, as F igure 7.1 shows, the
coherence is only approximate and there has been a systematic increase in the
spacing of ice ages over the past 800,000 years. Even more important is the fact
that coherence is even worse over a 2.7-million-year period. M&M further argued
that the narrowness of spectral peaks implies that glacial cycles are driven by an
astronomical force regardless of the details of the driving mechanism. The reason
given by M&M for this is that natural processes in astronomy virtually always
give rise to narrow spectral peaks, while natural processes in geology and climate
do not. Narrow peaks are characteristic of processes that have low loss of energy.
That implies a shortage of mechanisms capable of draining energy away. In
simpler terms, the regularity of the patterns of glaciation and deglaciation would
seem to suggest some astronomical pacemaker driving the process, rather than
random interactions between the atmosphere and oceans. For example, if the
glaciation-deglaciation patterns were caused by changes in thermohaline circula-
tion, why wouldn't the spectrum of data be broader? This seems to be a strong
point in favor of an astronomical factor being the cause of glaciation-deglaciation
cycles.
It is not immediately obvious which measure of solar intensity is of greatest
relevance in the astronomical theory. There is some reason to believe that ice ages
originate at high latitudes in the Northern Hemisphere (NH) because geological
evidence shows a great expansion of ice sheets in that region during ice ages and
because land (rather than water) occurs at high northerly latitudes, providing a
base for ice sheet formation. It also seems reasonable to guess that the onset of
widespread glaciation at high northern latitudes would be enhanced if a greater
preponderance of ice could survive the effects of higher solar irradiance in the
summer. Hence, most investigators have utilized midsummer solar irradiance in
the NH as a measure of solar variability from year to year. Alternatively, it has
been theorized that the key site for solar-induced climate change might be in the
SH causing variations in the oceanic transport of heat linking the NH to the SH,
but with a time delay. In particular, there is evidence that terminations of ice ages
might originate in the SH.
Aside from the question of north vs. south as the region of interest for solar