Figure 1.1: Lag Time Results (Anderson, 2000)
Discussions with the TxDOT project supervisor, David Stolpa, in March 2001 pertaining to the use of methods for hydrologic modeling led to other uncertainties and doubts about the automated process. Questions arose regarding scale effects of digital elevation models; as well as mechanical processes behind calculating slope, area, and longest flow path for a watershed. This uncertainty in the parameter values led TxDOT to question how variations in hydrologic parameters would ultimately affect watershed lag time values and discharge.
Following these conversations, the scope of the research project was redefined to explore the variation in lag time and discharge resulting from paper map and automated methods of calculating hydrologic parameters. In order to evaluate parameter uncertainty, four levels of extracting hydrologic parameters are explored: (1) measurement from paper maps; (2) on-screen extraction from raster maps; (3) using GIS and two different resolutions of grid-based digital elevation models (DEMs); and (4) using a triangulated irregular network (TIN).
This report attempts to quantify the errors associated from parameter variation in a two-step process. The first step is to quantify the error in lag time