Authors: Mark Groom*, University of South Florida St. Petersburg, Barnali Dixon, Co-Author, Professor, Leo Meirose, Co-Author, Graduate Student, Kyle Flanagan, Co-Author, Graduate Student
Topics: Geographic Information Science and Systems, Spatial Analysis & Modeling
Keywords: GIS, SWAT, delineation, DEM, water
Session Type: Paper
Start / End Time: 8:00 AM / 9:40 AM
Room: Galvez, , Marriott, 5th Floor
Presentation File: No File Uploaded
The focus of this study aims to evaluate the relationships between DEM resolution and the ability to accurately delineate watershed networks, utilizing the SWAT model. Due to the sensitivity of hydrological modeling to spatial resolution, it is important to understand the effects of DEM resolution on modeling output. DEMs with finer resolution characterize topographic landscapes with greater detail, but generally cost more to develop. DEMs when resampled to coarser resolution may not provide the same topographic information present in the original DEM. It is important to understand original and resampled DEM resolutions may not provide the same watershed delineation results. Using a SWFWMD watershed as study area, stream outlets, subbasins, and reaches were delineated at multiple resolutions, increasing in DEM grid size (10x10m, 30x30m, 90x90m, 10m resampled to 30m, and 10m resampled to 90m). After completing the SWAT watershed network delineation scenarios, analyses such as fractal dimension, shape index, and longitudinal root-mean-square-error (LRMSE) were conducted to evaluate trends and spatial statistics to quantify spatial differences in streamlines generated from original versus resampled DEM resolutions. Preliminary results suggest increasing DEM cell size did not significantly change configuration of SWAT modeled streamline networks in terms of reach count, total delineated stream length, and fractal complexity among the 10m, 30m, 10m resampled to 30m, 90m, and 10m resampled to 90m DEM resolutions. Future steps include delineation of a separate watershed with a physical, real-world stream network to evaluate SWAT modeling accuracy of stream networks at different DEM resolution.