Land use is an important explanatory variable in urban growth models, which explore the way various factors (e.g. geographic, economic, demographic etc.) interact to simulate growth dynamics. A serious and recurring problem for modelling urban systems has been the lack of spatially detailed data. Remote sensing and Geographical Information Systems have the potential to support such models, by providing data and analytical tools for the study of urban environments. High spatial resolution sensors such as the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) allow the estimation of land covers using either supervised or unsupervised classification techniques. Moreover, the accuracy of the classification can be improved by incorporating ancillary data in the classification scheme. In this study, spatial data from various sources are combined to develop statistical models relating land use to population density, distance from the center of the city, a land use mix index and monetary land values. The work emphasizes spatial relationships between various geographic, land-use, and demographic variables characterizing fine zones across and around regions. It derives and combines land use data for the Heraklion (Greece) region from ASTER images, cartographic maps and Greek National Statistical Service census of population data. The statistical techniques applied for explaining the variability of land use are ordinary and logistic regression. Land use mix appears to be a significant predictive factor whereas the explanatory power of population increases as the grid cell categorization with respect to land use becomes finer.