Resolve "Add a Finite Volume local operator for the Richards equation" (!132) · Merge Requests · dorie / dorie

Resolve "Add a Finite Volume local operator for the Richards equation"

What does this MR do?

Implements the finite volume scheme for Richards equation in a local operator, and adds it to the regular DORiE workflow by using the keyword richards.FEorder=0.

If the Richards FV solver is selected, the regular flux output is omitted. This is also mentioned in the docs.

Is there something that needs to be double checked?

This MR cannot be accepted before !105 (merged). The reason is that gradients in FV are equal to zero, and hence, fluxes are meaningless here. !105 (merged) solves this reconstructing the fluxes of the FV scheme.
I didn't implement source because is not currently utilized.

Can this MR be accepted?

Related issues

Closes #137 (closed)

### What does this MR do?

Implements the finite volume scheme for Richards equation in a local operator, and adds it to the regular DORiE workflow by using the keyword `richards.FEorder=0`.

If the Richards FV solver is selected, the regular flux output is omitted. This is also mentioned in the docs.

### Is there something that needs to be double checked?
* This MR cannot be accepted before !105. The reason is that gradients in FV are equal to zero, and hence, fluxes are meaningless here. !105 solves this reconstructing the fluxes of the FV scheme.
* I didn't implement source because is not currently utilized.

### Can this MR be accepted?
- [x] Implemented
- [x] Documented (Doxygen)
- [x] Documented (Sphinix)
- [x] Added test
- [x] Pipeline passing
- [x] Added entry to `CHANGELOG.md`

### Related issues

Closes #137

<!--
PLEASE READ THIS!

A Merge Request should be associated to a certain task or issue.
Its changes are supposed to be merged into the master branch.

Briefly explain __how__ you achieved the proposal of the task.

IMPORTANT: Make sure to set the merge request WIP if you are not finished yet.
-->

Edited May 08, 2019 by Lukas Riedel