Shallow snow depth mapping with unmanned aerial systems lidar observations: A case study in Durham, New Hampshire, United States

Abstract

Shallow snowpack conditions, which occur throughout the year in many regions as well as during accumulation and ablation periods in all regions, are important in water resources, agriculture, ecosystems, and winter recreation.Terrestrial and airborne (manned and unmanned) laser scanning and structure from motion (SfM) techniques have emerged as viable methods to map snow depths. Lidar on an unmanned aerial vehicle is also a potential method to observe field and 15 slope scale variations of shallow snowpacks. This paper describes an unmanned aerial lidar system, which uses commercially available components, for snow depth mapping on the landscape scale. The system was assessed in a mixed deciduous and coniferous forest and open field for a shallow snowpack (< 20 cm). The lidar ground point clouds yielded an average of 90 and 364 points/m 2 in the forest and field, respectively. Comparisons of snow probe and lidar mean snow depths in the field, at 0.4 m resolution, had a mean absolute difference of 0.96 cm and a root mean squared difference of 1.22 20 cm. In the forest, the in situ mean snow depth was nearly twice that from the lidar from mean absolute difference of 9.6 cm and root mean squared difference of 10.5 cm. These differences in forests are likely due, in part, to limitations of sampling using a snow probe. At 1 m resolution, the field snow depth precision was consistently less than 1 cm. The forest and heavily vegetated areas had modestly reduced performance with typical values within 4 cm precision. Performance depends on both the point cloud density, which can be increased or decreased by changing the flight plan, and the within cell variability that 25 depends on site surface conditions.

Characterizing Spatial Structures of Field-Scale Snowpack using Unpiloted Aerial System (UAS) Lidar and SfM Photogrammetry

“…This study was conducted at the University of New Hampshire Thompson Farm Research Station in southeast New Hampshire, United States (N 43.10892°, W 70.94853°, 35 m above sea level), which was chosen for its mixed hardwood forest and open field land covers (Perron et al 2004;Burakowski et al, 2015;Jacobs et al, 2021) that are characteristic of the region (Error! R eference source not found.).…”

“…UAS lidar and Structure-from-Motion (SfM) photogrammetry have emerged as viable methods for mapping high-resolution snow depths (~1 m), enabling a better understanding of snowpack spatial structure and its evolution over time at the field scale (Feng et al, 2023;Harder et al, 2019;Jacobs et al, 2021;Koutantou., 2022). As the use of UAS-based high-resolution snow depth mapping becomes more prevalent, there is a growing need for a comprehensive understanding of their strengths and weaknesses for capturing snowpack evolution throughout the entire snow period for various landscape features (e.g., forest and fields).…”

Sixth Oregon climate assessment

“…Remotely sensed data on snow can come from sources including airborne light detection and ranging (LIDAR) (Painter et al 2016), unpiloted aerial vehicle (UAV) LIDAR (Jacobs et al 2020), satellite LIDAR (Abdalati et al 2010, visible-range imagery (Painter et al 2009), and radar (Gusmeroli et al 2014). The spatial resolution of these data can be quite high, e.g., on the order of 1 m (3 ft) for airborne LIDAR.…”

Factors Influencing the Accuracy of Shallow Snow Depth Measured Using UAV-Based Photogrammetry

“…Unmanned Aerial Vehicles (UAVs), which recently went through dramatic technological advancement, emerged as the spotlight of the field of surveying in the last decades. Accordingly, snow depth surveying utilizing UAV was also a part of the spotlight [31]. The techniques of photogrammetry [32,33] can construct three-dimensional models of snow fields from multiple photographs taken from UAV, so it has been recently applied to estimate the snow depth [34][35][36].…”