If you run XTide with few locations specified on the command line, XTide will skip the harmonics file indexing step and just load those few locations as fast as possible. Because this process is substantially different from indexed access, there are differences in the resolution of location names that you must be aware of if there is more than one data set that could match your location name.
When the harmonics files are indexed, the names of all of the locations in all of the harmonics files are sorted into alphabetical order. A location name that could possibly match more than one data set always matches the one that is first in alphabetical order. (It is not case-sensitive.)
When fast loading is used, XTide checks each harmonics file in your HFILE_PATH in order, and scans each file from top to bottom. The first matching data set is immediately loaded. This MAY NOT be the one that is first in alphabetical order in all of the harmonics files.
There are steps that you can take to avoid getting the wrong data set:
Note that the reference stations needed by subordinate stations are also identified by name only, and therefore suffer from the same potential ambiguities. When adding subordinate stations to offsets.xml, always use a non-ambiguous name to specify the reference station.