Hi Alyx,
Maybe I misunderstood you, but it sounds like you are mixing two different fluid descriptions. The initVEFluidHForm is used for a quite old but still functional incompressible vertical equilibrium solver. The compositional mixtures are used for a 3D compositional
simulator. These are very different and cannot be used together in any way. Compositional solvers require a fair bit of experience to get correct results for CO2 mixtures and can be expensive to run with sufficient resolution. VE models make simplifying assumptions,
but are much faster when vertical segregation is fast. I do not think the VE solvers are the right choice for a H2 storage scenario.
The best way to get a understanding of how the different solvers in MRST are set up is to read the free MRST books while simultaneously looking at the corresponding examples.
Best regards,
Olav