[CIG-CS] A new mantle convection code: ASPECT

[Walter Landry]

Wolfgang Bangerth <bangerth at math.tamu.edu> wrote:
> 
> All,
> before we all take off for AGU, I'd like to let everyone know about a new 
> mantle convection code we've been building with CIG support. We call it  
> ASPECT (short for Advanced Solver for Problems in Earth's ConvecTion) and it 
> is based on the following principles:
> 
> - Compressible or incompressible, linear or nonlinear material laws
> - Arbitrary geometries, including cartesian and spherical
> - Modular and easily extendible code
> - Adaptive mesh refinement
> - Efficient linear and nonlinear solvers
> - Highly scalable to hundreds or thousands of processors
> - Accurate discretizations: Inherently stable Stokes discretization,
>    modern stabilization techniques for the temperature equation

Did you add something to handle sharp material interfaces that are not
aligned with the mesh, or does the error in the pressure still
diverge?

Cheers,
Walter Landry