Brubaker and Boyer, 2012

Talk/Poster

Multi-scale lidar greatly improve characterization of forested headwater streams in central Pennsylvania. (poster)

Kristen M. Brubaker and Elizabeth W. Boyer (2012)
AGU Chapman Conference on Remote Sensing of the Terrestrial Water Cycle, Kona, HI, February 2012  

Abstract

Most current hydrographic data used in Geographic Information Systems (GIS) have been derived by digitizing blue line streams from USGS topographic maps or by modeling streams using traditional digital elevations models (DEMs) in GIS. Both methods produce stream models that lack detail and accuracy, particularly in headwater streams. In addition to channel network delineation, another hydrologic attribute that is of interest to hydrologists, modelers, and ecologists, is topographic index (TI) as measured by the formula ln(a/tanβ). This metric and its distribution is an important component to the hydrologic model TOPMODEL and other hydrologic models, but is also used extensively to represent soil moisture in fields of ecology, forestry, and soil science.

Newly available lidar data available statewide in Pennsylvania can produce DEMs with an accuracy and resolution that far exceed previously available elevation data. In this study, streams were modeled using lidar-derived DEMs of 1 m, 3 m, and 10 m resolutions using existing GIS software programs and compared to both actual streams and streams modeled using a 10 meter National Elevation Dataset (NED) DEM. Results showed that the most accurate stream locations could be modeled using a lidar-derived DEM thinned to 3m resolution or smoothed using a mean smoothing filter. Also, when a 10 m resolution lidar-derived DEM was compared to the NED 10 m resolution DEM, the streams delineated with the 10 m lidar data were significantly better than those modeled with the 10 m NED data, showing that significant improvement in accuracy can be achieved with no increase in data storage. When topographic index was modeled with multiple resolutions of lidar-derived DEMs, the spatial and statistical distributions were both very different, with finer resolution DEMs not accurately modeling areas of high TI. Additionally, depending on the flow accumulation algorithm used, there were differences in the change in statistical resolution with response to initial DEM resolution.

Citation

Kristen M. Brubaker and Elizabeth W. Boyer (2012): Multi-scale lidar greatly improve characterization of forested headwater streams in central Pennsylvania (poster). AGU Chapman Conference on Remote Sensing of the Terrestrial Water Cycle, Kona, HI, February 2012.

This Paper/Book acknowledges NSF CZO grant support.