The Milankovitch Theory of Ice Ages explains the rhythmic pattern of glacial-interglacial cycles. Here, I mention a few of the things that cannot be explained by Milanktovitch’s theory (and that’s OK), and offer alternative causes.
Milankovitch Theory relates the change in Earth’s orbit around the Sun to the timing of Ice Ages. Cool summers (less sunlight) in the high Northern latitudes (65 deg N) allows snow and ice to persist year-round and thus you can begin to form an ice sheet. The Southern Hemisphere has more ocean at this key latitude, so we focus on the North. Glaciers are sensitive to temperature, especially summer melt temperature, so we focus on the intensity of that season.
Below is a problem/possible explanation list.
Problem #1: The variations in Earth’s orbital parameters (obliquity, precession and eccentricity) change insolation (incoming solar radiation) and are way too small to have any significant impact on Earth’s temperature. Only variations in eccentricity (shape of Earth’s orbit around the Sun) can produce net changes in annual insolation, but those variations are very small.
Possible explanation #1: Yes, eccentricity is the only cycle that can change the total amount of incoming radiation, but the distribution of that radiation at middle or higher latitudes is vitally important to climate feedbacks. Milankovitch cycles can still be the ‘pacemaker’, or initiator and then the climate system takes over making feedbacks that change the temperature of the planet, a little change in the Sun matters a lot.
Problem #2: Milanktovitch Theory cannot explain why the peak glaciations occurred every 100,000 years for the last 8 ice age cycles, and every 41,000 years before that. Also, the insolation changes are not large at the 100,000 year periodicity.
Possible explanation #2: First, I don’t think Milankovitch needs to explain this change, I’ve read ideas about changes in the jet stream shifting because of the Rockies uplifting, which would allow more of Canada to be in the polar cell. It is possible that changes in tectonics (mountain building) could influence the ocean and wind circulation, making ice ages last longer now than they were a million years ago. Second, the 100,000 cycle is an average, so perhaps the peak glaciations are either 80,000 or 120,000 years apart which could be 2 or 3 obliquity cycles.
Problem #3: The skewed response of glacial-interglacial transitions – cooling periods last ~90,000 years and the interglacials (like today) normally last 10-12,000 years.
Possible explanation #3: Glaciers respond to temperature in a non-linear way. It is far easier to melt ice sheets than it is to grow them, and again, feedbacks help to speed the melting process.
Problem #4: Does the Milankovitch theory of ice ages need to be retired?
Possible explanation #4: Milankovitch theory is a huge step toward the answer, feedbacks (Croll’s work) is another huge step. Figuring out the timing of temperature change, CO2 change, ocean circulation change, cryospheric change, etc, will need to happen for the full picture, perhaps greater collaborations between these fields and models will help move us forward.