Monitoring the trend of unplanned developments in an urban area

Although remotely sensed data has been used extensively for urban studies, additional collateral information is usually a pre-requisite for the correct automatic identification and localization of objects found within a particular scene. Hence the ability to categorize an urban area into informational classes namely formal or informal developments requires information about the town model which has in principle some legal backing. In this study, remotely sensed data are used as a source for current information. Specifically, spatially enhanced 1998 SPOT XS (10 m) resolution for Dar-es-Salaam, Tanzania is automatically interpreted and the classes aggregated into developed (built) and reserved (non-built). The 1992 land use map is used to mask out the newly developed areas. Weighted indicators based on compatibility of land uses, infrastructure network, and hydrological sources are used to establish their relationship on the new developed areas. Results show that existing land uses influence highly new developed areas. A combination of the weighted indicators is also employed to constrain the new developed areas thus resulting in a stratification of the new areas into fuzzy blocks ranging from those which are likely extensions of unplanned developments to those which are unlikely. Such information could facilitate planners in prioritizing areas that require urgent reaction planning