Your question intrigued me so I decided to read up on the research statistics and found some interesting documents:
http://www.mun.ca/psychology/brl/publications/Snook11.pdf
http://www.gis.smumn.edu/GradProjects/NeldnerR.pdf
I came to the conclusion that ~20 miles was a good boundary (assuming the murderer also lived in Mason City) and compared that with a map of the area. Clear Lake (
خرائط Google) caught my eye instantly. Has it been checked? There are definitely some other locations with potential (forests, other bodies of water), but this is by far the largest, and also quite remote.
Side note: Interestingly I put in her birthplace first mistaking it for where she lived and noticed it was quite a distance away (~250 miles). I wonder why and when she moved, since the murderer may have been an acquaintance made before she moved.