Geophysical site characterization using geomorphology, geology and integrated digital imaging methods on satellite data

This project explores the use of remote-sensing imagery and integrated digital imaging analysis methods to improve the estimation of seismic hazards in the Islamabad, Pakistan region. The principal utility of the resultant map is to characterize the local landform features in tenns of their potential to amplify ground motions generated by large-magnitude earthquakes. Because these landform features dramatically affect ground shaking during an eai1hquake, precise maps are critical for the development of urban hazard maps that identify areas of highest hazard within a greater metropolitan region. This project validates that remote-sensing imagery, along with an integrated pixel-based and object-oriented digital imaging analysis methods, has significant potential to characterize the detailed geomorphology and geology. To demonstrate this potential, the project produces a first-order site conditions map of the Islamabad study area. Information about local geomorphology and geology from ASTER (Advanced Space-borne Thermal Emissions and Reflectance) satellite imagery are extracted through the use of the eCognition (object-oriented) multi-resolution segmentation software program on the ASTER relative DEM imagery and the use of the ENVI (pixel-based) software program on the ASTER spectral imagery. The resulting geomorphic units in the study area are classified as mountain, including the Margala Hills and the n01iheast trending linear ridges, piedmont, and basin terrain units. The local geologic units are classified as limestone, in the Margala Hills and the linear ridges, and sandstone rock-types for the piedmonts and basins, which correlate with the local geology. Then, the geomorphology and geology are correlated with the expected compositional rigidity for correlation with the average seismic shear-wave velocity through the upper 30 m of the subsurface (Vs30). In the resultant map image, the high relief and limestone regions of the Margala Hills are assigned the highest velocity range (Vs30 500 m/s), the Islamabad and Rawalpindi regions (piedmont bench), just south of the Margala Hills, are assigned with the inte1mediate velocity range (Vs30 = 200-600 m/s), and the basin, to the south, is assigned with the lowest velocity range (Vs30 300 m/s). The resultant map is determined to be a first-order site conditions map and a preliminary map of potential site amplification in the study area.