It's basically an umbrella term for studying systems that are complex in the sense that they have strange nonlinear effects and may not be "well-behaved" in the way that classical, mathematically well-understood systems are. A large portion of it grew out of chaos theory, though some has gone in other directions since then.
Physics PhD who did his research in complex systems (how sand supports weight)...
Usually these complex systems consist of parts where the individual interactions are thought to be well understood but the aggregate behavior is not (aka non-linear).
In the case of sand, you can pretty much model the forces acting on a single grain/particle using intro physics but you wouldn't expect the long-range force chains you see in static piles or avalanches when more sand is added.
