ABSTRACT
Without detailed ground surveys, information collected from remotely sensed satellite images loses much of its potential meaning. Studies can be conducted with reference to the application of specific techniques with great success, only if the archaeological situation is known in detail. For example, if 90 percent of the settlement sites from a particular region are known to date to the Roman period, and satellite remote sensing specialists locate 100 additional possible archaeological sites, it is likely that the majority of the new sites would also date to the Roman period. One would not normally, question such a supposition in an archaeological publication. When publishing reports, archaeologists risk conjecture if their studies have not implemented ground-truthing or previous survey work risk due to the error rates typical in all remote sensing analysis. It is hoped that, when attempting different applied remote sensing analytical methods, some will work, while it is anticipated that others will not. Techniques of analysis that do not work are still very useful to remote sensing archaeologists, as it is through understanding why particular techniques do not work that one can identify techniques for future work that could help to locate additional archaeological features of interest. Understanding this from a ground-based perspective is invaluable.
