WHAT IS A DUNE? EVALUATING CURRENT METHODS OF EXTRACTING COASTAL DUNES FROM DEMS

Coastal dunes are landscape features which can absorb and mitigate erosion caused by elevated water levels during a storm. In order to understand how a dune responds to and recovers from a storm, it is important that we can first identify and differentiate the beach and dune from the rest of the landscape. Current literature does not provide a consistent definition of what the dune features (e.g. dune toe, dune crest) are or how they can be extracted. Traditional methods for extracting the dune rely on subjective interpretation, profile smoothing, or arbitrary functions and parameters. The purpose of this research is to develop enhanced approaches to extracting dunes from a digital elevation model (DEM). The beach and dunes were extracted from DEMs using multiple techniques based on published literature. Manual delineation and the convergence index approach were used on the DEMs to extract the dune crest and dune toe. The manual delineation method was found to be very time intensive and subjective based on the user, while the convergence index approach was useful to easily identify the dune crest through maximum and minimum values. A slope-inflection point approach revealed general trends for the dune crest, toe and shoreline. The least cost path method proved to be time intensive due to data clipping; however, this approach resulted in continuous geomorphic landscape features (e.g. dune toe, dune crest). Relative relief is a relatively new approach used to extract dune features from DEMs. While this method shows the most features in multi resolution, it is difficult to assess the accuracy of the extracted features because extracted features appear as points that can vary widely in their location from one meter to the next. The vegetation approach was greatly impacted by the seasonal and annual fluctuations of growth but is advantageous in historical change studies because it can be used to extract consistent dune formation from historical aerial imagery. Preliminary results indicate that the method used to extract the dune greatly affects our interpretation of how the dune changes. Improving our ability to more accurately assess dune response and recovery to a storm will enable coastal managers to more accurately predict how dunes may respond to future climate change scenarios.