What Causes Ice Ages?
A Serbian scientist named Milutin Milankovitch quantified long-cycle variations of the earth's orbit around the sun. He wrote a book in 1930 that documented his calculations of the amount of sunlight that each latitude received during every variation of the earth's orbit. It turns out that the earth's orbit has three different types of "perturbations." From the excellent NASA article, the orbit variations are of three types:
The circle around the sun becomes more oval and less circular (and back to circular). This cycle is about 100,000 years long. This means the Earth moves closer to the Sun at times when it's orbit is more "elliptical."
The seasonal tilt of the Earth varies over 41,000 years between 22 deg and 24 degrees. The less the tilt, the less the seasonal variation. The more the tilt, the more solar radiation at the higher latitudes.
The Earth wobbles on it's axis (like a wobbly top) with a cycle lasting 19,000 to 23,000 years.
Based on the orbital variations, Milankovitch predicted that the ice ages would peak every 100,000 and 41,000 years, with additional “blips” every 19,000 to 23,000 years. This is all verified by ice core, tree rings, coral reef and geological records, sea levels, glacial extent, etc. From Nasa, here's variations over the past 2 1/2 million years.
What this means is that there are regular periods of reduced solar radiation and cooler temperatures in higher latitudes (especially at 65 degrees North) that allow winter snow pack to persist through the summer months. Once snow pack persists over the summer, it increases in extent and forms a "micro-climate" that reinforces the stability of the growing snow fields. Snow fields become glaciers. There you go, that's why there are ice ages: it's the effect of Milankovitch cycles!
The past 100 years of global warming and CO2 accumulation is an entirely different topic than this blog--since the past 60 years of CO2 accumulation is has gone from 300 ppm to 420 ppm in a relatively short time. But the Earth’s average temperatures haven’t risen that much; except for Arctic areas (but not Antarctica) but are still -50 F in winter.
Here's good graph of the record of Antarctic ice from Vostok showing 420,000 year record of CO2 levels, CH4 levels and insolation at 65 degrees latititude (higher insolation means more sunshine and less glaciation):
My knowledge of this is highly superficial. Another excellent article on this topic on the Milan is at the excellent Wikipedia. It says that we are in a warming cycle for the next 40,000 years, so unfortunately we can expect no help from the Milankovitch cycles to offset the effect of our rising CO2.