Using GIS and LIDAR to Map Headwaters Stream Networks in the Piedmont Ecoregion of North Carolina

Abstract

A large percentage of nonpoint source pollution found in our Nation's waterbodies is suspected to occur through first- and second-order (headwaters) streams. Such streams drain a much greater proportion of watershed area and have a much greater length of riparian zone interaction with the land than the higher-order streams typically studied for nonpoint source water quality problems.

The State of North Carolina and the U.S. EPA are interested in examining the contribution of lowerorder streams to the overall nonpoint source pollution problem; however, the mapping of first- and second-order streams is extremely poor. The recent availability of fine resolution Light Detection and Ranging (LIDAR) data for portions of the State of North Carolina provides the opportunity for developing improved methods of mapping lower-order streams using Geographic Information System (GIS) approaches. In this study, I investigated the state-of-science for mapping topography and extracting headwaters stream networks using LIDAR data and GIS approaches. I applied these techniques to map headwaters streams at a study site in the Piedmont Ecoregion of North Carolina. I found that LIDAR produced more accurate elevation maps (elevation accuracy within 1.2') than currently available maps, such as the USGS 7.5 minute Digital Elevation Models (elevation accuracy within 49'). The Triangulated Irregular Network (TIN) produced the best topographic maps, but the Digital Elevation Model (DEM) was better for automatically extracting headwaters streams. The best headwaters stream maps were derived by using a hydro-enforced TIN for generating the base DEM,and extracting the stream network from this base DEM using ArcHydro and the AGREE algorithm. These improved headwaters stream maps will enable decision-makers to assess and mitigate nonpoint source water quality problems.