Multi-scale assessment of population distribution utilizing remotely sensed data - The case study Padang, West Sumatra, Indonesia

Abstract

Timely and accurate population information is crucial for decision-making in assessing risk and vulnerability before an expected disastrous event, coordinating emergency actions during an event or in managing disasters after the event. This research presents the capabilities of multi-sensoral remotely sensed data to provide spatial population distributions on different levels of geometric resolution supported by ancillary population data. On the one hand broad-scale assessment is carried out based on medium resolution Landsat data. Utilizing total population data available from a census a distribution algorithm is presented to allocate population from generalized administrative units at finer scale. On the other hand a small-scale analysis of urban morphology derived from high resolution Ikonos data is presented. Physical urban parameters like building size or building heights and in addition the land use enable to assess population distribution with respect to the time of day.