The original Milankovitch theory specifies orbital forcing of northern ice sheets. Broecker and Denton (1989) challenged subsequent extensions of this theory to explain global climatic responses. They noted that mountain glaciers and pollen sequences in the southern hemisphere indicate synchronous or earlier warming after the last glacial maximum, even though summer insolation at the precession cycle is out of phase between the hemispheres. The Milankovitch mechanism cannot explain near-synchroneity of global climatic responses at orbital time scales.

Greenhouse gases offer another way to synchronize global responses, as confirmed by new gas time scales for Vostok ice. At the 41,000-year tilt cycle, CO2 has the same delayed phase as northern hemisphere ice, so CO2 acts as positive feedback to changes in ice volume. Changes in CO2 affect the southern and northern hemispheres synchronously and in phase with northern ice.

At the 23,000-year precession cycle, both CO2 and CH4 have the same ‘early’ phase as July insolation. The CH4 variations result from summer insolation forcing of north-tropical monsoons and boreal wetlands, while the CO2 signal originates from winter (also July) oceanic processes in the southern hemisphere. Both the southern hemisphere and northern tropics respond to the synchronized early forcing from July insolation and the two greenhouse gases. Only the northern ice sheets and immediately proximal regions have the delayed Milankovitch phasing at 23,000 years.

All 100,000-year responses have similar phases because all are paced by orbital eccentricity. The specific sequence of climatic changes on a given termination reflects the superposition of effects at all three orbital periods. In general, the hemispheres respond nearly in phase, but with a small lead in the tropics and southern hemisphere.