Enhancement - focal mechanism uncertainty

From Dave Shelly:

I’ll suggest a possible route that I think might both simplify the approach and put it more solidly in a probabilistic framework:

One approach would be to specify location uncertainties (horizontal and vertical) and focal mechanism uncertainties (Kagan angle) uncertainties, assuming a simple Gaussian distribution. Then, based on the slab2 model, you could make upper plate, lower plate, and interface regions (you could also perhaps include slab2 model uncertainties if these are known). Because the interface is a surface, you would still be required to make some choice about how far from the surface you want to consider as “interface.” Then, simply based on the location uncertainties, you could rigorously determine the probabilities of upper plate, lower plate, or interface slip (or outside of any of these zones). Finally, if a mechanism is available, you could determine its likelihood of consistency with interface slip. Again, you’d have to make a choice here about how to weight this in relation to location uncertainties, but you would seem to want to adopt a scheme whereby the interface probability was increased when the mechanism was highly likely to be consistent and decreased when it was unlikely to be consistent.