Automated Terrace Locator Toolbar

Python - 3D Analyst - Lidar - Remote Sensing

Insert Caption.

Fluvial Geographers often refer to fluvial terraces as examples of the tectonic and climate processes that control the location, shape, and landmarks of rivers. These processes are related to the identification and analysis of terrace location, size, shape, and temporal processes measuring rate of change. Both processes have been studied in their own respects, but have remained separate even though they often look at the same area and features for study. These processes used to be conducted through the analysis of topographic maps with hand-drawn and visual analyze to identify terrace and fluvial features. With the standardization of LiDAR implementations and GIS Analysis, researchers have now begun to automate the processes of terrace extraction.

Early models of terrace extractions (Demoulin et all 2007) have been greatly improved for faster data analysis with the use of high-resolution data (del Val et all 2014). Both models still have problems with false positive terrace identification due to the method of analysis that identifies slopes, for location of flat surface features, making the need for the river system knowledge to be extensive before running analysis. This leaves incredible room for error when reproducing these results with the identification of features such as hill tops and the modern floodplains for the unexperienced researcher that could produce maps with error. While automation is presented in these studies, it requires an expertise on the study area before analysis. However, with the introduction of high-resolution data, 1-meter Digital Elevation Maps (DEM), these methods prove to be outdated and simplistic without relying on more computational analysis that could be processed to have high accurate terrace identification from Stack Profile Analysis at cross-sections of the river. The purpose of this method is to identify escarpment and rapid elevation change in order to create polygon shapefiles with detailed statistical analysis of each terrace formation.

In addition to identifying terrace locations, this study will also expand on the idea presented by Shuniji Ouchi on the analysis of offset channels across the area. In the first implementation of the study longitudinal profile data will be collected, but not analyzed yet. By collecting these data points further manipulation can be done in order to find strike-dip fault displacement along the profile that spans down the valley (Ouchi 2005). While these strike-dip displacements only are apparent when first developed as new gully’s and alluvial fans quickly develop in their place.

The end result of this study aims to produce a highly user-friendly toolbar that can be used to extract terraces, with controls of key variables, without the requirements of intensive knowledge of the study area before-hand. The model building in ArcMap has severe limitation on data manipulations that require the addition of Python programming for calculations which promotes the use of entirely new concepts that cannot be used for analysis natively in the program. This study will prove very useful for researchers by the potential time and money saving that comes from using an optimized analysis.