Identifying spatial concentrations of surnames

Surnames (family names) have been overlooked as a valuable source of spatially referenced population data. Presented here is a methodology, based on kernel density estimation, which is used to identify the areas of Great Britain where any surname is most concentrated. This not only indicates a surname's geographic origin in the country but also its current spatial extent and spatial relationship with other surnames and place names. We argue that such analysis can provide baseline and change measures, and an empirical basis to change forecasting. Such analysis offers valuable insights into national, regional and local changes in population structure, and testimony to the relevance of GIScience to population genetics, historical geography and genealogy.