Human societies interact daily with their environment. Archaeologists have always taken this into consideration when trying to understand the paths chosen by past cultures. The development of modern analytical tools such as Geographic Information Systems now make it possible to study the relationship between humans and their environment in ways that were not possible even ten years ago.

As
part of a large research project near Lake Roosevelt in Tonto Basin, Arizona,
research, archaeologists assembled a database of environmental information
to study the distribution of settlements on the landscape and to assess
the economic potential for Salado populations. The image at the left is
a 3-D model of the topography of Tonto Basin, constructed from elevation
data obtained from the United States Geographical Survey. Clicking on these
images will bring up higher resolution copies to view. Remote sensing data
from Landsat imagery were also used - the scene at the right shows the
green spectrum band ( in false color) with Roosevelt Lake visible in black.

GIS
allows environmental data such as topography and streams to be combined
with site survey data, much in the way transparent overlays were used in
the past. This can even be done in 3D such as this drape of a Landsat image
over a wire-frame terrain model. However, unlike mylar overlays, the layers
of a GIS database may be analytically compared as well as visually
compared. Each element in an image can be linked to a record in a data
table, allowing objects to be hidden, shaded, added, multiplied, etc, according
to their linked attributes.
To illustrate this process, let's consider using GIS to solve what normally would be a rather complicated problem. Archaeologists have suggested a number of environmental variables will influence where the Salado would have chosen to live. Based on their ideas, we could propose a hypothesis that sites would have been placed in locations that (1) were reasonably flat, (2) were close to water and, (3) had good soils for farming. We can test this hypothesis in a portion of the Tonto Basin using a method called Predictive Modeling. If you would like to see how, then click Continue.