On a planet where axial tilt causes ice ages, how could yearly seasons exist?

Question

A hypothetical planet experiences both regular yearly seasons and irregular miniature ice ages. The ice ages last for several years at a time and cease for several years at a time. Bizarrely, the ice ages are caused by axial tilt and the seasons are not. To survive, most life on the planet have developed various arctic adaptations such as producing antifreeze in their blood and entering suspended animation.

Answer 1

Obviously, it would be impossible for the planet to slow down and speed up it's rotation once per rotation period....

Assuming you have a localised population in your story, the days can be made shorter simply by tipping your world on it's axis, in exactly the same way as our seasons are governed here on earth.

Or, you could have the population be forced into living in valleys, which would limit the amount of useable sunlight they have in each day.

Answer 2

Changing length of the day every few years is hard. Making it chaotic is next to impossible - most astronomical phenomens are periodic.

You could go with volcanic activity. Dust and ash in the air can't make day shorter, but they can make it dimmer, and useful part of the day would get shorter. And, contrary to astronomical phenomens, volcanos are chaotic thing.

Answer 3

Instead of changing the length of the day, you could have the planet on an elliptical orbit. Pluto, for example, moves from 29.7 to 49.3 AU from the sun with every orbit. If the orbit lasted many years (Mars's orbit is 1.9 years, Jupiter's is 11.8 years), then orbital eccentricity would cause solar radiation to vary over the course of a few years, while the seasonal cycle caused by the planet's tilt on axis would be unchanged.

Here's some baseline numbers: lets say earth's average solar radiation is 1 unit. The Earth's orbit shifts from .983 to 1.016 AU. Since solar radiation follows an inverse square law, this mean solar intensity varies from $\frac{1}{.983^2}=1.035$ units at perihelion to $\frac{1}{1.016^2}=.969$ units at aphelion, a difference of about 7% of higher over low. On the other hand, you are looking for 'ice age' changes equivalent in magnitude to the difference between summer and winter on the US east coast. For a nice wintery locale like Cleveland, summer irradiance is about 5.5 kWh/m^2/day while winter is about 2.0 kWh/m^2/day. This is a difference of about 275% of high over low.

Lets take the example of a planet in ~Jupiter's orbit, with a sun bright enough to provide the same solar radiation (of 1 unit) at 5 AU and an orbital period of 10 years. In order for the planet to receive 275% more radiation in the 'interglacial' versus the 'ice-age', we want to calculate e such that the the ratio between radiation at perihelion $5-e$ and aphelion at $5+e$ is 2.75, or $$\frac{\frac{1}{(5-e)^2}}{\frac{1}{(5+e)^2}}=2.75,$$ $$\frac{5+e}{5-e}=\sqrt{2.75},$$ so that $e=1.24$. So your planet must move from 3.76AU to 6.24AU every 10 years to get the desired Ice Age cycle.

Is an orbit that eccentric feasible? A tougher question, but most things can be feasible somehow. Some migration of gas giants into resonances could have set the elliptical orbit. If there were no other terrestrial planets, very little asteroid matter, and large gas giants at 15+ AU to protect your main planet, then your planet would probably have cleared its orbit of all other objects and and it should work. A sun powerful enough to give you earth-like insolation at 5AU would probably burn out long before life evolved, but I will dismiss that as details.