Justin Ruths

Climate-Controlled Earth

Imagine that I could tell you how we can make a difference on the global warming effect. And I don't mean that you should start recycling, or use less hairspray, or plant more trees. What if we knew just how to massage the quantities of compounds in our atmosphere to regulate our temperature - not unlike how you might regulate your household temperature with an a/c unit and thermostat? Granted these notions are far off and somewhat fanciful, but the research I've proposed would let us know if this is possible.

Over the last few decades awareness of a global warming effect has been growing. Scientists have since been struggling to model and characterize the enormously complex system of Earth. The Earth system has so many intertwined processes that sorting out what is actually happening is impossible without picking apart each piece. There are many academic teams working on very select problems - such as modeling how vegetation has a role in the carbon cycle or how ocean currents change based on temperature. Recently, these models are becoming more and more accurate which in turn makes them more and more useful.

A control approach to this problem will help us in three main ways:

Stability: Using mathematical tools, we can determine first off if the earth is a stable system. This roughly says that if we stop doing whatever we're doing to make global warming worse all together, will the earth return to the previous state? A ball at the bottom of a valley is stable because it will roll back to its starting spot if it is nudged. A ball at the top of a hill is unstable because it will not return to its starting spot. If the system is stable, then we ask how big is the hill - how far can we push the ball until it rolls over the other side and won't come back. There are some useful diagrams of these ideas in the poster below.

Controllability: Systems analysis will also let us know if there is a variable or set of variables we can use to steer, or control, the system. Consider two blocks on the ground. You're able to move the first one but not the second. Unless the blocks are coupled (string, rubber band, touching) in some manner there is no way you will be able to move the second block. If you scale that idea up and apply it to the earth, are the "blocks" or variables of the system coupled in such a way that we would be able to change other "blocks" or variables? In particular, we would like to determine how we can induce a change in the carbon content of the atmosphere which has been shown to have a direct effect on temperature.

Observability: In systems so large it is often difficult or impossible to measure all of the variables. Here our tools will determine if we can reconstruct the variables we don't know with the ones we do. This sounds a lot less impressive and dramatic than the first two topipcs, but it is an essential step towards accomplishing the others.

A formal write-up of this plan of research (PDF)
A poster presenting the proposal (PDF 2mb)