Modelling of population dynamics: GIS versus Remote sensing – a case study for Istanbul

Abstract

Over the last decades, the rapid growth of the world population has led to a large number of emerging megacities. The 1999 Izmit (Turkey) earthquake is a striking example of the impact of natural hazards on megacities. On August 17 1999, a magnitude 7.6 earthquake struck the area of Izmit in Turkey, causing about 20.000 fatalities and US$6.5 billion economic loss. The probability of a magnitude 7 earthquake striking Istanbul within the next 30 years ranges between 30% to 70%. In order to reduce the impact of natural hazards on human lives, emergency management plans are essential. The development of these plans strongly relies on up-to-date population and inventory data. However, existing techniques for population data generation do not meet the requirements of today’s dynamic cities. In this context remote sensing has become an important source of information in the last years. However, a rational discourse on the suitability of remote sensing for urban applications is still missing. In this study a quantitative evaluation of the suitability of IKONOS imagery for population modelling using the district of Zeytinburnu (Istanbul, Turkey) is conducted. The results reveal that IKONOS images can be used for complementing existing inventory data set. The automated extraction of single buildings was identified as the major source of error in the population estimation. Further advantages and limitations such as the associated costs are discussed in this present paper.