Added 2015 Illapel earthquake example

[apadaki]

No description provided.

[mandli]

@apadaki This looks good to me. @rjleveque do you want to take a look at this to make sure this looks good and does not duplicate anything that might be somewhere I do not know about?

[rjleveque]

@apadaki, thanks for doing this! It's nice that you download the USGS fault parameters and the gauge observations directly in the maketopo.py script.

As set up, the code takes quite a while to run and doesn't seem to be refining optimally for capturing the signal at the two gauges, e.g. the refinement of the main wave is mostly to the south of gauge 32412. I think it would be better if you could add some flagregions (preferably in the new style) to guide the refinement more, allowing high levels only on the path towards the gauges for this example.

It might also run faster if you use fewer levels with larger refinement ratios, I notice some levels only refine by 2, you could have fewer levels always refining by 4, for example. And maybe you don't need such fine levels in this example? People can add more if they want to, but unless such fine grids are needed to illustrate something, it might be nicer if it runs faster by default.

[apadaki]

@rjleveque Thank you for your feedback! I'll make sure to implement fewer refinement levels to improve the run time and specify manual regions in a commit sometime today or tomorrow.

[apadaki]

@rjleveque I simplified the AMR ratios and added flagregions around each gauge, and the simulation seems to run much faster and give similar results on both gauges. If anything else needs to be changed, please let me know!