The scaling laws of human travel
A friend of mine recently pointed me to this fun Nature paper entitled "The scaling laws of human travel". Essentially, these guys took the Where's George? dataset and successfully characterized the movement of $1 bills across the United States.
For those not familiar with the Where's George project, the idea is to record the unique identifier of US dollar bills and stamp them all with the "WheresGeorge.com" logo. The project has set up a web site where people who find a Where's George bill log in and record the bill's unique identifier and the place they found it. There are millions of bills currently circulating freely across the US and beyond. (I saw a $1 bill stamped with "Where's George" in Addis Ababa, Ethiopia last fall!)
While it's an interesting standalone website, the subsequent data generated from this project seems to provide insight into the spatiotemporal dynamics of a (fairly) unbiased random sample of the US population. The authors point out that the distribution of the distance travelled decays as a power-law and show the trajectories of individual bills can be fit well using a two parameter variation of a continuous-time random walk (CTRW).
Although I would have liked the authors to discuss the implications associated with human traveled being governed by a CTRW rather than pure Levy flight dispersal (ie: so what?), their rigorous analysis of this dataset is a great example for other researchers with similar spatiotemporal data. The paper also is a good place to start when brainstorming about what other comparable datasets are out there that could help us get a better handle on the complex dynamics associated with the behavior of our society.