The inner splay is based upon the minimum V2 for the aircraft
there is one extra item in the ops engineering bag of tricks which can help out here - if the outer splay then doesn't create problems with rocky bits, one can restrict the normal (max) 15° bank OEI turn by specifying a lesser bank. This, for the min V2, opens up the turn radius. Obviously, one doesn't add to the pilot's woes at the time by specifying an unflyable intermediate angle like 12.5° ... but 5° and 10° are worth looking at, where relevant. As in everything ops engineering, life is a succession of compromises.
don't ever use OEISIDs developed for another aircraft type for your own operations
.. and we've all seen this sort of thing done.
I recall just this happening in a Regulator's navaid flight checking organisation (which ought to have known better) where the said organisation pinched some jet OEI escape procedures from another operator (one near and dear to OS's historical heart) and applied them to a much smaller (and VERY much slower) turboprop Type.
Our good colleague, Centaurus, has a range of tales along these lines to relate from a variety of operators (pilots for the flying for) over the years.
If it is expected that obstacle clearance throughout the whole of the SID will not be maintained in the event of an engine failure, the Airport Authority may publish an alternative SID that will ensure obstacle clearance is maintained.
.. and, if you REALLY believe that, I have a bridge to sell you. Had you suggested something like an alternative SID that may improve obstacle clearance I would have concurred.
SIDs, at best, have only a tenuous connection to obstacle clearance in the real world.
One might refer back to OS's wise words ..