Using remotely sensed data for census surveys and population estimation in developing countries: Examples from Nigeria

The conduct of conventional census surveys in most developing countries and the rational use of the results obtained have been impaired largely by the lack of basic infrastructures. These infrastructures include up‐to‐date administrative and topographic maps, settlement and land use information and geographically referenced enumeration areas (EAs). Using examples from Nigeria, this paper demonstrates how remotely sensed data can be used to acquire some of the basic data requirements for census surveys and to estimate population. The result obtained shows that visual identification of settlements on Landsat MSS and TM is more accurate and economical than equivalent digital classification techniques. Black and white aerial photographs were used to estimate the population of a model town and to establish EAs. The population estimation method employed can be used to obtain intercensal population estimates for the rapidly growing central places, while the established EAs for the study area have created a permanent base for future census surveys and census cross‐validation, population estimation and other social surveys.