"Safety should not be sacrificed for cost."
You know, it is hard for me to imagine that "settlement" could actually be controlled and predicted in an Earth Fill Dam of this monstrous size. I would be surprised if an Earth Fill Dam of this size wouldn't have settlement disparity over time.
So one wonders, if a dam of such huge proportions is required, why an earthen dam versus some kind of appropriate concrete dam? One wonders that if cost is the reason even if not the safest alternative, that such a decision would violate (b) "Safety should not be sacrificed for cost."
Earth Fill Dams, if monitored & taken care of, are a useful design. The original designers placed 100 monuments over the surfaces of the dam just for this purpose - to take care of it & monitor by measurements.
Settlement is more consistent in a sloped rock canyon with sides that allow conformal compaction. In the case of a steeper slope transition zone in the sloped canyon, care is required to monitor for differential compaction, as the lesser angle slope to the greater angle slope changes the compaction forces.
There are a number of remedial measures & technology in fixing differential compaction related seepage issues (injection of mix of clayey & Bentonite material for example). The safety factor for the thickness of the core at Oroville plus the extra attention to the internal slope angled design of the Zone2 and Zone 3 vertical layers provides excellent support to the plastic Zone 1 vertical core layer. Models of the design were tested in actual seismic acceleration conditions & soils/compaction was confirmed to have a good margin of "Factor of Safety".
What surprises me is that this "historic" or "long standing" issue hasn't been figured out. The last thing you want to happen to an earthen dam is to have a decades old puzzle to become an unwanted surprise.