Custom Query (454 matches)

I rewrote scratch.UCERF3.inversion.InversionSolutionERF to use the new CompoundGriddedSurface? class, and also simplified it so each inversion rupture is a separate source (we can re-implement Kevin's aggregations of ruptures into sources later if that's what we want; not currently sure).

I need to add option for aleatory mag variability for each source later.

I also need to add a better naming convention for each source.

The previous version is now at scratch.UCERF3.oldStuff.OldInversionSolutionERF.

I added two fields to this class (with get and set methods):

dateOfLastEventMillis slipInLastEvent

I updated everything except the *XML* methods.

I don't want to touch these until I'm sure doing so won't create problems with any current XML files.

Thoughts?

A couple issues here.

1) There are two RELM testing regions, one a simple polygon with about 15 points, and one with several more that that, which was made from one of Danijel's testing files. The advantage of the latter was making a GriddedSurface? from it led to the exact same set of nodes as in the official test files (although I doubt this is still the case given changes to our code, but I don't know for sure because I can't get my hands on a test file). The more complex version (RELM_GRIDDED) has been deprecated (in favor of RELM_TESTING). The problem is the RELM_NOCAL and RELM_SOCAL are based on the more complex version, so simplified versions should be made and the originals deprecated.

2) All of these used BorderType?.MERCATOR_LINEAR rather than .GREAT_CIRCLE. A problem with this is 5% of Parkfied falls into No. Cal. using mercator, but doesn't using great circle. Also, SCEC VDO shows great circle, so what you see can be deceiving.

Peter - would there be any problem converting these to great circle?