Structure-from-Motion Production and Analysis of Digital Surface Models of NOAA Coastal Airborne Imagery from Alaska's North Slope
Abstract
Alaska's northern coastline from the Canadian border to Kotzebue Sound is notably lacking in elevation data that is needed to estimate volumetric change along the coastal bluff, identify areas of permafrost degradation, and update shoreline positions. In 2016 and 2017, NOAA's Remote Sensing Division collected 50 cm resolution, semi-oblique and nadir georeferenced RGB imagery along Alaska's southeastern, western and northern coastlines as baseline datasets to aid in navigation, determine pre-storm conditions and facilitate in coastal-zone management. The 2017 imagery extends along approximately 800 km of Alaska's remote North Slope coastline, includes an additional near-infrared (NIR) band and was collected with overlap sufficient for the application of Structure-from-Motion (SfM) photogrammetric techniques. SfM techniques have the ability to produce Digital Surface Models (DSMs) of complex topography from overlapping imagery and minimal ground control, with resultant ground sampling distances that are comparable to LiDAR. Although photogrammetric methods such as SfM do not offer the multi-return capabilities of LiDAR that allow for the discrimination between vegetation and bare ground, SfM methods are well suited to the coastlines of the North Slope of Alaska where vegetation is sparse or low-lying. DSM production from this image collection fill a vital gap in elevation data for this region and, when combined with near-infrared (NIR) imagery, present opportunities to better understand vegetation phenology and health in Arctic areas undergoing substantial change. Here we present a protocol for Structure-from-Motion processing into DSM datasets from a subset of the 2017 NOAA airborne imaging data collected in the Arctic National Wildlife Refuge and its comparison to other elevation data sources. We also present coastal volumetric change estimates and assess the potential for these DSMs to be used in extraction of vector coastline features.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMEP23D2365E
- Keywords:
-
- 1621 Cryospheric change;
- GLOBAL CHANGEDE: 1630 Impacts of global change;
- GLOBAL CHANGEDE: 4315 Monitoring;
- forecasting;
- prediction;
- NATURAL HAZARDSDE: 4217 Coastal processes;
- OCEANOGRAPHY: GENERAL