The solution is simple in my view. Just make it dangerous enough that they can't get it no matter how much they would want to loot it. And if they can outsmart the dangers we face them with they must be smart enough to figure out how to deal with the radio activity.
> We have all become very marker-prone, but shouldn't we nevertheless admit that, in the end, despite all we try to do, the most effective "marker" for any intruders will be a relatively limited amount of sickness and death caused by the radioactive waste? In other words, it is largely a self-correcting process if anyone intrudes without appropriate precautions, and it seems unlikely that intrusion on such buried waste would lead to large-scale disasters. An analysis of the likely number of deaths over 10,000 years due to inadvertent intrusion should be conducted. This cost should be weighted against that of the marker system.
I'm not convinced. I guess there are levels of radiation poisoning that have effects immediate enough to make it obvious from whence they came. There are also many levels which are not, but will still end or ruin lives.
Above all, engineering and use of the site is to be prevented - nobody should dynamite it and release radioactive dust, nobody should lead a river through it or flood the site, nobody should use parts of it to build anything. Immediate proof that the site is dangerous, by people coming in contact with it dieing while not being contaminated with any kind of radioactive dust, may be a safe and universal isolation of it. Even animals may learn to avoid it.
The only problem is that it's probably impossible to have sufficiently high levels of radiation on the outside of the 'Keep' while still keeping the isotopes safe from any kind of disaster or environmental effects.