Mapping urban areas by fusing multiple sources of coarse resolution remotely sensed data

In recent decades, rapid rates of population growth and urban expansion have led to widespread conversion of natural ecosystems and agricultural lands to urban land cover. The amount and rate of this land conversion affects local and regional ecosystems, climate, biogeochemistry, as well as food production. The main objective of the research described in this paper is to improve understanding of the methodological and validation requirements for mapping urban land cover over large areas from coarse resolution remotely sensed data. A technique called boosting is used to improve supervised classification accuracy and provides a means to integrate MODIS data with the DMSP nighttime lights data set and gridded population data. Results for North America indicate that fusion of these three data types improves urban classification results by resolving confusion between urban and other classes that occurs when any one of the data sets is used by itself. Traditional measures of accuracy assessment as well as new, maplet-based methods demonstrate the effectiveness of the methodology for creating maps of cities at continental scales.