SUPERVISED CLASSIFICATION OF TYPES OF GLACIATED LANDSCAPES USING DIGITAL ELEVATION DATA

Automated approaches for identifying different types of glaciated land scapes using digitally processed elevation data were evaluated. We tes ted the ability of geomorphic measures (e.g. elevation, relative relie f, roughness, and slope gradient) derived from digital elevation model s (DEMs) to differentiate glaciated landscapes using maximum likelihoo d classification and artificial neural networks (ANN). The automated m ethods were trained and validated using an existing Quaternary geology map and a manual interpretation of the contour data portrayed on topo graphic quadrangles. The need for such methods arises from efforts to classify types of landscapes (e.g. ecoregions) in Michigan. One fundam ental control of the landscape structure in Michigan, including soil t ype and vegetation, is the underlying sedimentary and landform assembl ages produced by an array of glacial processes during the waning phase of the Pleistocene. Traditional methods for identifying different lan dscapes (e.g. ice-contact landscapes, stagnation landscapes) have reli ed on printed topographic maps and have been very effective, but time consuming. The maps resulting from the four supervised classification trials had between 51% and 61% agreement with the original Quaternary geology map. The output from the maximum likelihood classification had slightly higher agreements than the output from the neural net, which is attributed to the generalization inherent in the Quaternary geolog y map compared with the nature of the classifier for the neural net. T he neural net, however, identifies significant detail and non-linear r elationships between classification inputs and output classes. Future work should incorporate a map of soils into the classification. (C) 19 98 Elsevier Science B.V.