* Old test used parabolic velocity profile as initital and boundary condition
  although Re was way over 1000. This led to some strange behaviour regarding
  the pressure gradient

* We now give the system some additional time so that the "right" pressure gradient
  can evolve before injecting the tracer

* Add output of deltaP

Kilian Weishaupt authored 8 years ago and Timo Koch committed 8 years ago

*Pure diffusion: no advective fluxes expected
*Advection
*Density driven flow with gravity