The increasing prevalence of such footers implies that there must be some research supporting their use, but I haven't yet encountered any. Have you?

I think you're placing too much emphasis on the role of research as a driver of design principles. I would say that these things emerge from designer intuition, and proliferate if they don't cause serious problems, or interfere with other ideas.

The footer in particular is cheap real-estate. Nobody minds how long a page is (up to the point where it starts affecting loading time), so you might as well put something there.

Basically you can roughly subdivide the page into three sections from top to bottom: Navigation , Content and Footer.

The Navigation should cater to the primary use cases (Something is wrong, I want to get out of here, I want to go back, I want to try something else)

The Content to the secondary use cases (I'm where I want to be, I want to do something with this object/content, I want to see something related to this content).

Once you've dealt with those, you've eliminated the possibility that the user is unhappy and that she is happy, so you can cater to the few users who are:

  • Patient
  • Looking for something else

These are the users who are willing to read the instruction manual, the ones who'll go on a course to learn something, will ask someone for help with a ticket machine, etc. If you can give them an overview of information that costs some investment, but will yield good results in return, that's another 5% or so of users that find what they're looking for.

The only potential problem (as you say) is that the user mistakes the footer for content and keeps scrolling, but that's a pretty simple thing to communicate in design.