Petrology of inclusions in type II diamond indicates they form at 650km depth or deeper. Diamond bearing kimberlites are restricted to Archean cratons, far from recent or existing subduction zones, so the standard "ring of fire" model for volcanism doesn't apply to kimberlites. What is not known is how volcanism is triggered from such depths under stable cratons. Could an electrical (plasma) discharge from deep in the upper mantle be responsible for triggering the short lived phreatomagmatic volcanism that is characteristic of kimberlites?
Craters that are likely due to plasma events often have a second smaller crater that overlaps the initial larger crater. This is due to a second, weaker plasma strike. Kimberlites often have a second, smaller phase. Could this be due to a second electrical (plasma?) discharge shortly after the initial discharge?