APPLICATION OF AIRBORNE RADAR, REMOTE SENSING, GIS AND MODELING TO SAN CLEMENTE ISLAND ARCHAEOLOGY

We are completing a technology evaluation of applicability of various remote sensing methods (concentrating on airborne radar), GIS, and modeling for archeological applications using San Clemente Island, CA as a test site. New approaches are desired for required archaeological surveys of DoD and DoE lands as present methods are too slow and expensive. Evaluation and mitigation are required before site disturbance. Presently over 19 million acres of DoD land alone are unsurveyed. In the western US surveys cost $30-$35 per acre, far more in the heavily vegetated eastern US. Delays and significant schedule impact can occur should archaeological sites inadvertently be discovered. We have used JPL AIRSAR multiparameter radar imagery and TOPSAR high resolution digital elevation data (http://airsar.jpl.nasa.gov/). The radar imagery is collected at three wavelengths (~68cm or "P-Band", ~24cm or "L-Band", and ~5.6cm or "C-Band") and full polarization diversity (HH, HV, VV, VH, where H=horizontal, V=vertical polarization, first letter is transmit polarization, second letter is receive polarization). Resolution is up to 1m per pixel (picture element). TOPSAR topographic data is achieved in an interferometric mode. Phase variations in radar echoes simultaneously received at two antennas are used to calculate topography. TOPSAR digital elevation models (DEMs) have vertical resolutions as high as 1.5 m. GeoSAR radar data (http://www.earthdata.com), now commercially available, have been integrated into the data set to test potential operational methodologies. GeoSAR data are P and X (3 cm) wavelength images and DEMs. Amongst the data, high resolution digital topography is key to allowing accurate registration of all various remote sensing and surface data sets in a GIS, including older, poorly controlled, data sets. Analysis and predictive modeling based on the integrated data sets indicates that areas likely to contain archaeological sites can be delineated, the landscape characterized in terms of potential use by native peoples, and some classes of archaeological sites directly detected.