|BSill 3D Final5.mph||2.4 MB|
|BSill 3D with particles.mph||7.3 MB|
|BSill_3 particles.mph||3.9 MB|
|Download all files (Zip-archive)||~ 10.8 MB|
Modelling Magma Intrusion in Sills
Nick Petford, University of Northampton
We used COMSOL Multi physics (Pipe and Particle Tracing Modules) to simulate the lateral, pressure driven flow of viscous magma in 2 and 3D. The model uses field (digital) images of the magmatic intrusion as the basis for the FE mesh. All calculations are solved in that geometry. As well as recovering average flow rates for a range of viscosities typical of natural magmas we were able to identify the presence of (non-thermal) eddies at undulations in the upper and lower contact surfaces. Eddies scale with the fluid flow properties and dampen out as viscosity increases (Re < 1000). Particle tracing allowed us to map out the orbital dynamics of particles (assumed to be crystals in magma) 'tuned' the density range common in naturally occurring minerals. These provisional results suggest that in 3D the eddy structures are roll-like and extend normal to the mean flow direction. Their presence, along with the potential for mixing of particles (crystals) trapped for a period of time from the main flow is a hitherto undocumented process in magma fluid dynamics.
Mar 16, 2017 at 9:25am UTC
I ran 'BSill 3D with particles.mpg' and after 30 minutes, it stopped by saying that the solution is not converged and maximum number of Newton iterations have reached. However, I deactivated 'Particle Tracking for Fluid Flow' module as I did not have license for that.
Mar 16, 2017 at 6:04pm UTC
BSill 3D with particles.mph didn't converge for me either, and I have the particle tracking.
Login to comment on this entry.