Place names are signs of geographic entities, and the database of which, a digital gazetteer, is an increasingly important
form of geographic information. So the construction and the application of digital gazetteers are growing research areas.
And significant progress has been made in the development of which, but there are still some vital issues that require
further work: (1) Places and attributes related would inevitably change over time, but few gazetteer services model
temporal ranges; (2) Current gazetteers do not normally hold historical information; (3) The relationships between place
name entries are few considered in most existing digital gazetteers; (4) Geographic footprints currently used in gazetteers
are usually confined to simple representations. In this paper, we proposed a spatio-temporal data model for
administrative division place names, which are in a significant and large proportion of place names. We took Xiamen
City as a case, located in coastal areas of Fujian Province, Southeast of China, to describe our model. In the model, we
considered spatio-temporal changes and relationships between entries in gazetteers, and the footprints used are multiscale
patches adapting to hierarchical administrative system. Accordingly, our model could provide an important
reference for digitizing gazetteers and further for implementing digital earth.