Generally speaking, the stability of a fluid system is increased if one adds a `stable' or bottom-heavy density gradient. But in the late 1970s a number of authors, including Paul Roberts, Keith Stewartson and myself, started coming across systems where a `stable' density stratification has precisely the opposite effect, and acts as a catalyst for instability.
The systems in question all had one thing in common: the instability mechanism involved diffusive - or even doubly-diffusive - oscillations in a subtle way, and by studying a variety of examples I was eventually able to offer a simple physical explanation for this strange effect.
J. Fluid Mech. Vol 96, pp 723-733, 1980
Back to Research Page