The Ecuadorian Andes feature a hotspot of biodiversity. At the same time Ecuador has the highest deforestation rate of South America. In order to develop sustainable counteractive measures, it is important to locate and quantify the ongoing changes on a regional scale. In this study we present a cost-effective remote sensing based change detection algorithm suitable for application in a tropical mountain rainforest landscape. We show that the deforestation in the study area is closely linked to the existence of roads but topography also influences the spatial distribution of land cover change.