A method to generalize stream flowlines in small-scale maps by a variable flow-based pruning threshold |
| |
| Authors: | Michael Tinker Peter Anthamatten Jeff Simley Michael P Finn |
| |
| Institution: | 1. United States Geological Survey , National Geospatial Technical Operations Center, Box 25046, MS 505 , Denver , CO , 80225-0046 , USA mdtinker@usgs.gov;3. Department of Geography and Environmental Sciences , University of Colorado Denver , Denver , CO , USA;4. United States Geological Survey , National Geospatial Technical Operations Center, Box 25046, MS 505 , Denver , CO , 80225-0046 , USA |
| |
| Abstract: | The aim of this paper is to explore and describe a method of automated generalization designed to produce a map which strikes a balance between cartographic and hydrologic representations. Following a discussion of scholarly literature on generalization, we describe a novel method for automated generalization of hydrographic stream data, using the National Hydrography Data Set (NHDPlus) as an example. Traditional hydrography shows a fairly uniform density of stream flowlines over space. While this is pleasing to the eye, traditional methods tend to under-represent rivers in humid areas and over-represent them in arid areas. We address this problem through a method in automated generalization to produce a high-quality presentation of hydrographic data, suitable for display as a wall map or in an atlas. Streams are pruned based on a variable flow threshold, derived from the local mean annual precipitation by a regression equation. After running the model using different parameters, we produce a more satisfactory portrayal of stream networks in the United States that communicates the flow of water through rivers and reflects the regional climate. Specific advantages in generalizing with variable flow threshold include (1) the method allows for fine gradations in output scale; (2) the output maps tend to minimize density variations in the raw data; (3) the subjective criteria are easily derived; and (4) the method can be performed rapidly on large data sets, as long as the stream data has been enriched with reliable flow rates. |
| |
| Keywords: | generalization hydrography National Hydrography Data Set NHDPlus |
|
|
