U.S. Department of the InteriorU.S. Geological Survey

April, 2012

High-Resolution Hydrographic Mapping Using Lidar-Derived Digital Elevation ModelsBy Sandra K. Poppenga1, Ryan F. Thompson2, and Daniel G. Driscoll3

1U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center, Sioux Falls, SD, USGS South Dakota Water Science Centers, Huron2 and Rapid City3, SD

Abstract

For many decades, U.S. Geological Survey (USGS) topographic maps have been widely used as a primary basis for hydrographic mapping. How-ever, capabilities for higher-resolution hydrographic mapping are improving rapidly with advances in the availability of high-resolution, light detec-tion and ranging (lidar) digital elevation models (DEMs) and the capabilities for processing these data. High-resolution hydrographic mapping would be especially beneficial in many parts of eastern South Dakota where accurate definition of drainage features can be very difficult, especially in areas with low relief or numerous depressions.

Introduction

Lidar point cloud data have become prevalent as source data for generating high quality lidar DEMs that are incorporated into the National Elevation Dataset (NED), the primary elevation product of the USGS (Gesch, 2007). Although it was only a decade ago when the NED consisted of 30-meter DEMs, today the NED has evolved into a multi-resolution dataset that is increasingly comprised of high-resolution lidar DEMs. This evolution from 30-meter DEM data (fig. 1) to highly detailed 3-meter or better lidar DEMs (fig. 2) is important for those who need more accurate elevation informa-tion for hydrologic or hydrographic purposes.

Results

An example of high-resolution hydrographic mapping is provided by a recent pilot project that was con-ducted by USGS Earth Resources and Observation and Science (EROS) Center and USGS South Dakota Water Science Center (SDWSC) (figs. 5-7). In figures 5 and 6, there are comparisons of elevation-derived watershed boundaries and a 12-digit Watershed Boundary Dataset (WBD) Hydrologic Unit (HU) from a pilot area in Roberts County, South Dakota.

The red polygon in figures 5-7 represents the WBD 12-digit HU # 090201010204 (U.S. Geological Survey, 2011). This HU was delineated using the USGS 1:24,000-scale topographic base maps as the source data-set. The black polygon in figure 5 was automatically delineated from a 30-meter flow direction grid derived from a NED DEM. The surface channels shown in figure 5 are also derived from the 30-meter DEM, which is a different topographic coverage than what was mandated for use in development of the 12-digit WBD. Notice the substantial difference in drainage divides in Figure 5 (yellow arrows).

Figure 6 shows the improvement in watershed delineation (brown polygon) that is achieved through incor-poration of a 3-meter lidar-derived DEM for the area. Notice the ground surface detail visible in the lidar data (fig. 6) compared to the 30-meter DEM (fig. 5). Substantial improvements in the depiction of drainage channels also are apparent in figure 6.

Lidar-derived hydrologic derivatives also provide the flexibility to automatically generate drainage densities at various thresholds without re-digitizing an entire network. In Figure 7, the denser surface channels (com-pared to figs. 5 and 6) were derived from the lidar flow direction grid. This flow direction grid provides the capability to automatically delineate sub-basins from any point within the watershed boundary. This capabil-ity is important for automating high-resolution hydrographic mapping to derive sub-basin watershed bound-aries.

Background/Methods

Watershed boundaries were originally digitized from older topographic map contours. Where present, blue lines denoting stream networks were used to interpret surface drainage (fig. 3). Automated methods are now available to delineate watershed boundaries using pertinent high-resolution lidar-derived DEMs, and aerial imagery is helpful for visualizing watershed boundaries relative to features not present in older maps (fig. 4). The methods used to delineate lidar-derived watershed boundaries begin with defining the direction water would flow (flow direction grid), based upon gravity, across the lidar DEM ground surface. In the past, generating a 30-meter flow direction grid was a fully automated process (Jenson and Domingue, 1988; Franken, 2004), however the methods that are now used to generate lidar-derived flow direction grids require additional techniques to define downstream surface flow through conduits, such as a culverts or bridges (Poppenga et al., 2009, 2010, 2011).

Using these additional techniques, lidar-derived hydrologic derivatives, such as flow direction, flow accumulation, hydrologically-enforced DEMs and shaded relief, and surface channels were generated for a pilot project in northeast South Dakota. These data were used to automatically generate watershed boundaries that can be useful for high-resolution hydrographic mapping.

Conclusion

The use of lidar DEMs to automatically refine drainage divides is beneficial for hydrographic mapping in that watershed boundaries can become both horizontally and vertically integrated (conflated) with topo-graphic data to provide a new level of hydrologic and hydrographic geospatial detail.

The topographic detail that is inherent in lidar DEMs provides the capability to automatically improve and sub-divide drainage divides into smaller watersheds that are useful and important for specific management practices. This is advantageous because by digitally automating these processes, the laborious task of re-digitizing a new level of watershed boundaries from topographic contour maps has been replaced by an effi-cient automated process.

References Cited

Franken, S.K., 2004, The USGS/EROS Data Center produces seamless hydrologic derivatives with GIS: ArcNews Online, no. Fall, p. 8, 13–14.

Gesch, D.B., 2007, The National Elevation Dataset, chap. 4 of Maune, D., ed., Digital elevation model tech-nologies and applications—the DEM users manual, 2nd ed.: Bethesda, Md., American Society for Photo-grammetry and Remote Sensing, p. 99–118.

Jenson, S.K., and Domingue, J.O., 1988, Extracting topographic structure from digital elevation data for geographic information-system analysis: Photogrammetric Engineering and Remote Sensing, v. 54, no. 11, p. 1593–1600.

Poppenga, S.K., Worstell, B.B., Stoker, J.M., and Greenlee, S.K., 2009, Comparison of surface flow features from lidar-derived digital elevation models with historical elevation and hydrography data for Minnehaha County, South Dakota: U.S. Geological Survey Scientific Investigations Report, 2009-5065, 24 p.

Poppenga, S.K., Worstell, B.B., Stoker, J.M., and Greenlee, S.K., 2010, Using selective drainage methods to extract continuous surface flow from 1-meter lidar-derived digital elevation data: U.S. Geological Survey Scientific Investigations Report, 2010-5059, 12 p.

Poppenga, S.K., Worstell, B.B., Stoker, J.M., and Greenlee, S.K., 2011, Using selective drainage methods to hydrologically-condition and hydrologically-enforce lidar-derived surface flow, in Remote Sensing and Hydrology 2010, Jackson Hole, Wyo., 27-30 September 2010, IAHS Publication 3XX: International Com-mission on Remote Sensing.

U.S. Geological Survey and U.S. Department of Agriculture, Natural Resources Conservation Service, 2011, Federal standards and procedures for the National Watershed Boundary Dataset (WBD) (2d ed.): U.S. Geo-logical Survey Techniques and Methods 11–A3, 62 p.

Figure 1. NED 30-meter shaded relief of an area in Roberts County, South Dakota Figure 2. NED 3-meter lidar-derived shaded relief of an area in Roberts County, South Dakota

Figure 3. In the Watershed Boundary Dataset, watershed boundaries (red polygon) were digitized by interpreting topographic contours.

Figure 4. Lidar-derived watershed boundaries displayed on National Agri-cultural Imagery Program aerial photography from 2008.

Figure 6. Watershed boundary (brown polygon) and elevation-derived stream network (blue lines) produced from the 3-meter lidar NED con-firm similar differences from the WBD (red polygon).

Figure 7. Lidar-derived hydrologic derivatives can be used to correct existing WBD boundaries, further subdivide existing 12-digit hydro-logic units, and produce surface drainage networks of differing densities.

Figure 5. Watershed boundary (black polygon) and elevation-derived stream network (blue lines) produced from the 30-meter NED with high-lighted differences from the WBD (red polygon).