IEEE Geoscience and Remote Sensing Magazine - March 2023 - 81

technology can also be combined with lidar to analyze glacier
surface collapse events caused by surface melting and ice
dolines and fully utilize the cost, performance, and time efficiency
of UAV photogrammetry technology in large-scale
investigations as well as the ability of lidar systems to better
characterize ice morphology [18]. The subsidence formation
in the Dålk Glacier and its evolution were also described in
Boronina et al. [78].
UAV-based DOMs and DEMs are also utilized to detect
windblown snow phenomena [26], [79]; identify microtopographic
features, such as blue ice and melting ponds [64]; and
identify sorted-stone circles [80], providing planar and 3D information
about the ice surface. UAV images can also identify
fine spectral differences for monitoring surface water and
the sediment-rich glacier meltwater plume. The multispectral
sensor equipped on UAVs captures images that are used to
identify glacier outflows, estimate turbidity values of the sediment
plumes, and study the relationship among the different
quality parameters of surface water [28], [63], [81]. However,
because of the shadow phenomenon of glaciers that significantly
affects the spectral characteristics of meltwater by reducing
its reflected radiation, it is difficult to assess sediment
meltwater plumes in the polar marine environment. UAV images
determine spatial changes in the spectral characteristics
m
0 200
of shaded water, and the reflection is much higher in both
shaded and nonshaded areas than in satellite images [28].
UNMANNED AERIAL VEHICLES IN FLORA STUDIES
Changes in the Antarctic biome are influenced by rapid
and severe climate change in polar areas. Low-grade vegetation
that is extremely sensitive to temperature, moisture,
and ultraviolet radiation occurs only on the Antarctic coast
and peninsula region because of continual low temperatures.
Moss is an indicator of the regional impact of climate
change; therefore, it is critical to map and monitor its
range and health [56], [65]. UAVs have been used to map
the range and health of the Antarctic moss bed; UAV-based
and GPS-based DEMs were compared to calculate the topographic
humidity index and evaluate moss growth conditions
in the study area. To extract dense 3D point clouds
from overlapping UAV images, an algorithm was formulated
based on structure from motion (SfM). A 2-cm-resolution
DEM was generated by combining SfM and a patch-based
multiview stereo image vision method [36], [55], [56], [65],
[82]. In recent years, a variety of new sensors, including hyperspectral
and thermal infrared sensors, have improved
the registration method and ability of UAV-based images
to monitor Antarctic moss beds and analyze their health,
Elevation (m)
110
86
Curvature
-817986
792505
Landsat 8 Multispectral
20190128 (30 m)
Sentinel-2B
20190113 (10 m)
UAV DOM
20190114 (6.47 cm)
(a)
(b)
(c)
FIGURE 6. (a) Crevasses in a Landsat 8 multispectral image, Sentinel-2B image, and UAV orthomosaic. Their spatial resolutions are presented
in the small brackets; the red polygon is the region used for crevasse extraction. (b) The surface elevation of the red polygon based
on the UAV-DEM product. (c) The curvature map of the red polygon derived from the UAV-DSM product.
MARCH 2023 IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
81
Resolution
IEEE Geoscience and Remote Sensing Magazine - March 2023

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - March 2023
