Relations Between Field-surveyed Channel Networks and Map-based Geomorphometric Measures, Inez, Kentucky zyx David M. Mark Department of Geography, State University of New York at Buffalo, Buffalo, NY 14260 Abstract. Channel networks near Inez, Kentucky, were surveyed in the field within 29 small drainage basins that appeared on 1 :24,000-scalemaps to be unbranched (first order). The basins contained from 0 to 5 channel sources, with an average of 2.15 per basin. Many channel char- acteristics measured in the field are well correlated with morphometric variables measured from maps, but a substantial proportion of the variance remains unexplained. Although simple cor- relations are fairly strong for geometric characteristics, they are relatively weak (yet statistically significant) for topological measures. The spatial and temporal persistence of the empirical predictive relations is as yet unknown. Key Words: geomorphology, channel networks, cartographic generalization, map accuracy, topographic maps, Kentucky. RAINAGE network studies have formed an D important part of geomorphology for al- most four decades. The topology and geom- etry of the drainage networks themselves have been fruitful areas of study, and the drainage basins related to these networks have formed the basis of drainage basin morphometry studies. The results of these studies have in turn contributed substantially to hydrology- being included in flood-prediction and other runoff models-and to modern concepts of landscape evolution. Because of the scale of most drainage basin studies, it is impossible to examine all chan- nels in the field. Although studies of badlands or microtopography (for example, Schumm 1956, Abrahams and Campbell 1976) are im- portant exceptions, drainage network geo- morphology has come to rely heavily on to- pographic maps as data sources, with only occasional recourse to aerial photographs or to field studies. Despite this reliance on maps, few geomorphologists seem to know exactly how the maps they use are constructed; al- zyxwv Annals of the Association of American Geographers. zyxwvutsrqponm 73(3) 1983 pp 358-372 c Copyright 1983 zyxwvutsrqponmlk by Association of American Geographers though most are probably aware of the dif- ferences between channels found in the field and those interpreted from maps, they rarely mention this difference especially in North American drainage network studies. Evi- dently, either they consider the differences to be unimportant or, because fieldwork is not a viable alternative to map studies, they feel that the difficulties are best left unstated. The present paper reports on results of field surveys of drainage networks within drainage basins that on 1 :24,000-scale topographic maps appear to contain unbranched, or first- order, streams. The study basins are located in northeastern Kentucky, in an area that has provided map-based data for more than a dozen papers in quantitative geomorphology. First, methods for extraction of the drainage network from contour maps are reviewed. Second, an extensive exposition of guide- lines for the preparation of topographic maps is presented, with emphasis on those aspects that are relevant to geomorphology, and ex- amples of how these methods may influence 358