DETECTING PREHISTORIC LANDSCAPE USE THROUGH REMOTE SENSING IN NW ARIZONA, USA

Remote sensing data is being used to estimate soil type, current vegetation cover, soil moisture, surface roughness, and proximity to known habitation sites to determine the potential of landscape features to sustain productive prehistoric maize agriculture in the Mt Trumbull region of NW Arizona, an arid region of the Colorado Plateau. In some areas prehistoric site density is ~120 sites per square mile. Existing images being examined include 18 ASTER images covering the period 2000-2008; one meter digital orthophoto quad data for the 4 central USGS 7.5’ quads; 10 m digital elevation data for the same quads; and NRCS soil maps. New ASTER images along with complementary ground based IR camera images were acquired on May 2010. The hydrologic model FLO-2D was used to evaluate the hydrology of two watersheds in the study area. Data used for the final models considers both rainfall and soil variation. In the field we measured depth of soil and slope, aspect, and soil moisture content. Adequate soil moisture during the germination and tasselling phases of maize growth is essential for a productive harvest. Data loggers have recorded air temperature at several locations allowing us to model length of growing seasons. Ancient maize is thought to have required about 120 frost free days to produce adequate yields. Using two areas where archaeological survey is complete, a crop/habitation model is being constructed to show optimal (and suboptimal) areas for maize agriculture. Important factors include high soil moisture to support germination and tasseling; specific soil associations; sufficient precipitation during the summer; and a plant community dominated by rabbitbrush. The results of the hydrologic modeling indicate areas where topographic and soil conditions provide high flow depth and high infiltration, likely resulting in more soil moisture available for maize. Our model will provide some predictive capability in the region for prioritizing subsequent survey intensity. The research proposed here may also allow planners of the new Grand Canyon-Parashant National Monument (where the Mt Trumbull study area is located) to avoid particularly sensitive areas when designing recreation and other facilities for the Monument.