Quantifying Slumness with Remote Sensing Data

Abstract

The presence of slums in a city is an indicator of poverty and its proper delimitation is a matter of interest for researchers and policy makers. Socio-economic data from surveys and censuses are the primary source of information to identify and quantify slumness within a city or a town. One problem of using survey data for quantifying slumness is that this type of data is usually collected every ten years and is an expensive and time consuming process. Based on the premise that the physical appearance of an urban settlement is a reflection of the society that created it and on the assumption that people living in urban areas with similar physical housing conditions will have similar social and demographic characteristics (Jain, 2008;Taubenböck et al., 2009b); this paper uses data from Medellin City, Colombia, to estimate slum index using solely remote sensing data from an orthorectified, pan-sharpened, natural color Quickbird scene. For Medellin city, the percentage of clay roofs cover and the mean swimming pool density at the analytical region level can explain up to 59% of the variability in the slum index. Structure and texture measures are useful to characterize the differences in the homogeneity of the spatial pattern of the urban layout and they improve the explanatory power of the statistical models when taken into account. When no other information is used, they can explain up to 30% of the variability of the slum index. The results of this research are encouraging and many researchers, urban planners and policy makers could benefit from this rapid and low cost approach to characterize the intra-urban variations of slumness in cities with sparse data or no data at all.

Informalidad laboral y en la vivienda : primeros indicios para las principales ciudades colombianas

“…En el ámbito internacional se han realizado múltiples estudios para varios países utilizando esta metodología (Mason y Fraser, 1998;Hofmann, 2001;Hofmann et al, 2008;Engstrom et al, 2015;Ravshty et al, 2015;Williams et al (2015)). En el caso de Colombia, uno de los más recientes es el trabajo para Medellín realizado por Duque et al (2013). Esta aproximación tiene la ventaja…”

Cityscape, poverty and crime: A quantitative assessment using VHR imagery

“…We extract information on land cover composition using per-pixel classification and on urban texture and structure using an automated tool for texture and structure feature extraction at object level (Ruiz et al, 2011). We use data from Medellin (Colombia), which is the second larger city in the country, and it has been one of the most violent cities in the world in past decades and is still one of the most socioeconomically divergent (Duque et al, 2013a). This city is a useful location for conducting intra-urban variability studies because it has experienced high population growth rates since the 1950s, and the unplanned urban growth in some parts of the city resulted in a high degree of spatial heterogeneity in both the socioeconomic and physical characteristics of its neighborhoods.…”