This paper explores the use of generalized linear models (GLMs) for enhanci
ng standard methods of satellite-based land-cover change detection. It star
ts by generalizing satellite-based change-detection algorithms in a modelli
ng context and then gives an overview of GLMs. It goes onto describe how GL
Ms can fit into the context of existing change-detection methods. By way of
example, using a change detection over two locations in North Carolina, US
A, using Landsat Thematic Mapper data, it shows how the models provide a qu
antitative approach to image-based change detection. The application of GLM
s requires special consideration of the spatial correlation of geographical
data and how this effects the use of GLMs. The paper describes the use of
preliminary variogram analysis on the image data for initial sampling consi
derations. For the binary response (change/no-change) derived from the refe
rence data, a 'joint-count' test is used to assess their independence. Fina
lly, the model error term is checked through the empirical variogram of the
residuals. It is concluded that GLMs can be helpful in examining different
change metrics and useful by applying the resulting model throughout the i
mage to get a probability of change estimate as well as pixel-specific esti
mates of the variability of change estimate. However, as presented here, th
is application should respect the assumption of independent response data u
sed for the modelling.