The main topic of this contribution is an analysis of the visual interpretation process of remotely-sensed images for geoecological purposes. Several image interpretation steps, relations, and decisions are discussed, as well as background theories related to regional land cover/land use classes. As a case study and as an application of these methods a coloured Ecological Land Classification Map of Argolis and Arcadia (Peloponnesus, Greece) in the scale 1:100 000 was produced. In detail the relations between photo-pattern-areas and land use-/land cover-classes are pointed out. The interpretation key shows the rules for the matching process of these two feature categories. Bases for the set of land use-/land cover-classes are a regional model of the natural/semi-natural vegetation classes and a model of the agricultural land use in the study area. Due to perceptual and cartographic constraints, small land use patterns are aggregated to mixed land use classes or land use class mixtures. The results of visual-manual classification of multi-temporal satellite images are new regional information, which can be extended to the delineation of cultural-landscape regions. The use of remote sensing data, together with collateral data, is adequate and recommendable for such regional spatial-analytical purposes.