Characterizing the spatial distribution of field-scale snowpack using unpiloted aerial system (UAS) lidar and structure-from-motion (SfM) photogrammetry

Abstract

Unpiloted aerial system (UAS) light detection and ranging (lidar) and structure-from-motion (SfM) photogrammetry have emerged as viable methods to map high-resolution snow depths (∼1 m). These technologies enable a better understanding of snowpack spatial distribution and its evolution over time, advancing hydrological and ecological applications. This is particularly critical in mixed vegetation environments, where both forest canopy and open areas influence snow accumulation and melt patterns. In this study, a series of UAS lidar/SfM snow depth maps were collected during the 2020/2021 winter season in Durham, New Hampshire, USA, with three objectives: (1) quantifying UAS lidar/SfM snow depth retrieval performance using in situ magnaprobe measurements, (2) conducting a quantitative comparison of lidar and SfM retrievals of shallow snow depths (<35 cm) throughout the winter, and (3) understanding the spatial distribution of snow depth and its relationship with terrain features. Eight UAS surveys were conducted over approximately 0.35 km2 including both open fields and a mixed forest. In the field, lidar had a slightly lower error than SfM, compared with in situ observations, with a mean absolute difference (MAD) of 3.5 cm for lidar and 4.0 cm for SfM. Snow depth maps from SfM and lidar were fairly consistent in the field, with only marginal differences on most dates. In the forest, SfM greatly overestimated in situ snow depths compared with lidar (lidar MAD = 6.3 cm, SfM MAD = 31.4 cm). There was no clear agreement between SfM and lidar snow depth values for individual 1 m2 pixels in the forest (MAD = 55.7 cm). Using the concept of temporal stability, we found that the spatial distribution of snow depth captured by lidar was generally consistent throughout the period, indicating a strong influence from static land characteristics. Considering both areas (forest and field), the spatial distribution of snow depth was primarily influenced by vegetation type while also reflecting the effects of soil variables (e.g., soil organic matter). When the field and forest areas were analyzed separately, the spatial distribution was distinctly affected by slope and the shadowing effects of the forest canopy.