Extraction of Urban Information for Seismic Hazard and Risk Assessment in Lima, Peru Using Satellite Imagery

Abstract

Lima City, Peru, is, like Japan, on the verge of a strike by a massive earthquake. Building inventory data for the city need to be created for earthquake damage estimation, so the city was subjected to the extraction of spatial distribution of building age from Landsat satellite time-series images and an assessing building height from ALOS/PRISM images. Interband calculation of Landsat time-series images gives various indices relevant to land covering. The transition of indices was evaluated to clarify urban sprawl taking place in the northern, southern, and eastern parts of Lima City. Built-up area data were created for buildings by age. The height of large-scale mid-to-highrise buildings was extracted by applying spatial filtering for a DSM (Digital Surface Model) generated from stereovision PRISM images. As a result, buildings with a small square measure, color similar to that of their surroundings, or complicated shapes turned out to be difficult to detect.