The phenomenon you are referring to is called situated learning.
Recall of information from memory is facilitated by retrieval cues; the more retrieval cues you have, the easier it is to recall an item from memory. Location (and pictorial information information in general) is perhaps such an effective cue because of the amount of information that it construes. You might say that a picture has higher entropy relative to, e.g., a sound, in that a picture can be easily recalled yet it conveys much more. A room contains numerous details-- the placement of furniture, the color of the walls, the books on the shelf-- all of which may act as retrieval cues.
The reason these seemingly meaningless details are such helpful retrieval cues can be linked to the idea of Hebbian learning, which is often summarized by the mnemonic, "cells that fire together, wire together". When you study in a room, neurons are firing in response to stimuli in the room, while other neurons fire in response to the content you are learning. Thus, when you are again tested in that same room, the stimuli in the room fire the same neurons, which act as retrieval cues. If you study in a different location, you would have no retrieval cues-- which is why Bjork suggests varying your study location.
I would say that this has an effect on both memorizing random facts as well as events that happened in a particular location. However, events tend to be more memorable for a number of other reasons. For instance, they are experienced (rather than simply read off a page). There is some evidence that autobiographical memories are treated differently than semantic memories, both at the cognitive and neural level.