Archaeological site potential modeling can trace its origins to a variety of sources, including
human geography, settlement archaeology, ecological archaeology, and paleoecology. The basic assumption
is that pre—contact land use was limited or shaped by ecological factors, such as the locations of
the natural resources upon which people depended for their livelihood. Potential modeling therefore
represents the attempt to reconstruct past land use patterns through multi-disciplinary research.
There are two basic approaches to predictive modeling. The first employs known site locations, derived
from either extant inventories or through sample surveys, as a guide for predicting additional site
locations. The second approach predicts site locations on the basis of expected behavioural patterns, as
identified from suitable ethnographic, historical, geographical, ecological, and archaeological analogues.
Every modeling exercise should incorporate elements of both approaches.
Three major factors limit the resolution of our images of the past and hence our ability to predict
pre-contact site locations with absolute certainty: the inadequacies of the existing archaeological
database; our knowledge of the pre-contact natural environment is limited by both the inadequacies
of the existing paleoenvironmental database and the inherent difficulties in interpreting extinct
ecosystems; and pre‑contact aboriginal people possessed a world view that was substantively different
from our own. There are certain classes of sites, used for burials or vision quests, for example, which
were situated primarily for ideological or aesthetic reasons and are, therefore, impossible to understand
using economically based methods of spatial analysis.
In spite of these limitations, predictive modeling efforts to date have proven successful to the extent
that they can permit site potential assessments at a level of probability that is useful in the context of
heritage resource assessment and planning.

GIS map showing pre-contact and historic feature potential for a section of the Humber River in Toronto.
In developing an archaeological potential model, a Geographic Information System (GIS) is used to map
sets of information as separate, but complementary layers of spatial data on 1:10,000 scale digital base
maps. First, various criteria are mapped and evaluated in order to develop the archaeological potential
zone. These include: soils classified by drainage, texture, and capability for agriculture; hydrography;
and areas of early Euro-Canadian settlement and development as documented in historical maps and atlases
and evaluated in terms of historical themes considered to have been most significant in the region. The
archaeological potential zone is then refined by eliminating areas where previous land development had
severely disturbed the landscape. Created as a discrete layer of archaeological land integrity, this map
is based on the "built-up" layer of the National Topographic database, together with interpolation from
road network mapping, land use mapping, and detailed aerial photography. The resulting zones of precontact
and historic archaeological potential can then be determined.
