Exotic invasive insects are a mounting threat to native biodiversity and resources. Upon arrival, alien species exist at low abundance, complicating detection. Furthermore, the analysis of abundance data can be difficult because they are not robust enough to make inference regarding establishment. Yet, if eradication is to be a goal, managers require knowledge of the infestation extent. One approach in data analysis is to consider epidemiological methods, in which disease incidence can likewise be rare. In this context, the focus is to quickly detect space-time patterns of disease incidence to minimize outbreaks. When dealing with alien species, there is a similar desire for an early warning system so that new invaders can be targeted prior to expansion. In both cases, detection data often involve extremely low levels of incidence, and are representative of limited temporal and spatial scales. After all, if the space-time extent is large enough to conclusively determine the extent of the infestation, then eradication may no longer be feasible. However, when testing the applicability of epidemiological methods in studies of invasive species, benchmarking studies should utilize robust sets of data. I used historical data on gypsy moth defoliation in Pennsylvania, 1975-2000, to develop a paradigm for investigating space-time dynamics. I then applied this approach in an analysis of detection data from Minnesota, 2000-2004. This approach yields new information regarding gypsy moth space-time dynamics as it migrates into new areas of the U.S., and also provides potential methods for the analysis of other exotic invaders.