Evaluation of vector coastline features extracted from 'structure from motion'-derived elevation data

By: , and 

Links

Abstract

For extensive and remote coastlines, the absence of high-quality elevation models—for example, those produced with lidar—leaves some coastal populations lacking one of the essential elements for mapping shoreline positions or flood extents. Here, we compare seven different elevation products in a lowlying area in western Alaska to establish their appropriateness for coastal mapping applications that require the delineation of elevation-based vectors. We further investigate the effective use of a Structure from Motion (SfM)-derived surface model (vertical RMSE<20 cm) by generating a tidal datum-based shoreline and an inundation extent map for a 2011 flood event. Our results suggest that SfM-derived elevation products can yield elevation-based vector features that have horizontal positional uncertainties comparable to those derived from other techniques. We also provide a rule-of-thumb equation to aid in the selection of minimum elevation model specifications based on terrain slope, vertical uncertainties, and desired horizontal accuracy.

Additional publication details

Publication type Conference Paper
Publication Subtype Conference Paper
Title Evaluation of vector coastline features extracted from 'structure from motion'-derived elevation data
DOI 10.1142/9789814689977_0251
Year Published 2015
Language English
Publisher World Scientific
Contributing office(s) Pacific Coastal and Marine Science Center
Larger Work Type Book
Larger Work Subtype Conference publication
Larger Work Title Proceedings of the Coastal Sediments 2015