Vulnerability in Delhi, India: Identification and Integrative Assessment of Informal Settlements using Remote Sensing and GIS

In order to carry out the urban planning and development tasks necessary to improve the living conditions of urban dwellers, a detailed spatial data basis is required. Due to the high dynamics of megacities, traditional methods such as statistical analyses or fieldwork are limited to capture the urban process. Remote sensing provides the opportunity to monitor spatial patterns of urban structures with high spatial and temporal resolution. The analysis of heterogeneous, high-fragmented and dynamic urban environments requires the application of very high-resolution (VHR) satellite data.

The present study investigates the potential to use VHR QuickBird data to identify urban structures and dynamics within Delhi, India. The paper presents a semi-automated, object-oriented classification approach which allows for the identification of informal settlements within the urban area. An object-based image processing approach makes use of homogeneous spatial units instead of single pixels. Compared to conventional pixel-based image processing, utilizing only the spectral response, image objects contain additional information like object texture, shape or relations to neighbours. Also due to the complexity of urban environment and changeable understanding of individuals an (semi-) automatic classification approach based on the application of object features seems to be very promising. Another advantage of object-oriented image analysis lies in the fact, that the obtained classification results can be exported in form of thematic layers, which enables a close connection to GIS. The final outcome is a land cover map, where the identification of informal settlements - regarding the building size and building density – is of special importance.

Focal point of this contribution is not the classification method but rather an integrative data analysis concept. For this purpose the obtained image classification results are embedded in a GIS environment in order to derive information on population and water related parameters. Summarizing it can be stated: it shall be evaluated, whether the image data can present strong indicators together with socio-economic data to assess and classify the living conditions of the local dwellers in informal settlements within a megacity. Therefore appropriate, up-to-date, city wide information and new techniques has to be provided to the persons in charge in a very timely manner. This could support a more proactive and sustainable urban planning and land management – which increases the importance of remote sensing techniques. This is understood to be a first step to the development of a transferable methodology for the identification and analysis of urban structures within megacities like Delhi.