Mapping Fluvial Terraces Using Geospatial Analysis: A Case Study on the Piedmont of Georgia, USA

Abstract

Mapping of alluvial soils and geomorphic units can be useful for purposes as varied as land management, cultural resource planning, flood risk assessment, and modeling stream system response to environmental change. This research project develops a method that combines USDA-NRCS soil map units with a geospatial workflow processing 1-m LiDAR imagery to delineate stream terraces in the Upper Little Tallapoosa basin of western Georgia (USA). Existing soil surveys distinguish between high and low terrace alluvium by different soil series. “Old alluvium” is represented by the Masada (map unit abbreviations MoB, MpB, MpC2), Augusta (Asl), and Worsham (Wsl) series. These are further distinguished as low vs. high terraces: Masada is typically found in high terraces, and Augusta and Worsham in lower terraces. We use various geospatial tools to de-trend a LiDAR-based DEM in order to optimize for identifying stream terrace surfaces. This approach results in planimetric terrace surface polygons that differ somewhat from the soil survey map units. We then create gradient profiles to identify any correlated terrace surfaces along the stream valley.  Hypsometric curves and histograms of select stream reaches are used to visualize elevation distributions and to identify any terrace surfaces. Field checks at several sites, based on backhoe trenches and hand augered cores, indicate that our new integrated approach is an effective method for mapping terraces.