Analysing the spatial heterogeneity of poverty using remote sensing : typology of poverty areas using selected RS based indicators

In this paper we discuss methods and tools to spatially locate areas of urban poverty. Much work on poverty analysis uses city wide statistics that hides the spatial heterogeneity of urban poverty. For spatially targeting poverty alleviation interventions, disaggregated spatial information is needed on various dimensions of the phenomenon. An important aspect in this regard is how poverty is conceptualized, measured and analyzed within a specific local context. The starting point for our work is an analysis of urban poverty in Delhi using a livelihood approach and based on 2001 census data at ward level. We use high resolution imagery (Ikonos 2001/2002) to examine the physical development of a selection of these wards to develop a methodological approach for combining additional indicators of physical development to the poverty index. Spatial indicators of poverty areas extracted from classified imagery on location characteristics of poverty areas (such as roof coverage densities, lack of proper road network characterised by irregular layout of settlements) are used. Spatial indicators are used to understand the heterogeneity within several deprived wards. We also differentiate between types of poverty areas (e.g. high density central slum areas versus developing poverty areas in the fringe of the municipal corporation) aiming eventually at the development of a typology of poverty areas. The results indicate that poverty is not spatially uniform within the selected deprived wards, and hence demonstrate how data aggregation can hide spatial variation of poverty. By comparing between poverty areas an initial typology is extracted, highlighting the multidimensional of the phenomenon. Information on the spatial heterogeneity as well as the typology are important information for local authorities to improve the targeting of poverty alleviation interventions.