Assessing the Ability of Structure From Motion to Map High-Resolution Snow Surface Elevations in Complex Terrain: A Case Study From Senator Beck Basin, CO

摘要

The utility of differentially mapping snow depth to assess snow water resources at the watershed scale has been demonstrated using snow-free and snow-on lidar surface elevations. On more limited spatial and temporal scales, the same principle has been successfully applied with the relatively new photogrammetric technique Structure from Motion (SfM). Given the low cost of cameras relative to lidar technology, early studies are promising, yet it is well known that reconstructing elevations over bright snow surfaces in complex terrain has been a limitation for traditional photogrammetric methods. Therefore, before progressing to snow depth, it is worthwhile to constrain how well snow surface elevations are mapped with SfM. The lidar-based Airborne Snow Observatory, which also has an RGB camera, provides a unique opportunity to assess SfM against coincidentally collected lidar. Here we present a lidar-SfM snow surface elevation comparison from the 21 February 2017 flight, which took place in Senator Beck Basin, San Juan Mountains, CO, during the National Aeronautics and Space Administration (NASA) SnowEx campaign (Year 1). After coregistration of the two surface models, the normalized median absolute deviation was 0.17 m with a mean relative elevation difference of 0.014 m at identical spatial resolution of 1 m. The digital surface model was created without the use of ground control points, shows a promising potential to apply SfM for watershed-scale surface elevation and snow depth mapping, and warrants further investigation of SfM as a supplement or alternative to lidar.