11.2
Design and Construction of Ice Roads Use of Uncrewed Aircraft Systems
11.4 Solid Ice
Once the ice road has been established, UAS can be flown in combination with on-ice visual inspections
to identify dry cracks, wet cracks, water on the ice surface, snow drifts, and other problems that may
compromise the ice road integrity and may not be visible from the inspector’s location on the ice.
Anomalous features identified during UAS flights over established ice roads can be further examined
using zoom features on the sensor being carried by the drone, or by investigating on foot if the
conditions are safe for foot traffic. UAS also can be flown after major storm events to ascertain the
condition of the road to support what maintenance or repair steps need to be undertaken to return the
ice road to safely passable conditions.
The roughness of the ice road surface can be measured using commercially available drones to create
3D models of the ice surface using commercial data processing programs. Ice roughness can be an
indicator of instability of ice depending on the time of year, recent weather, and the surrounding
conditions of the ice and landscape, and is a component of the ice that can be measured confidently by
using UAS as the only tool. A specific example of ice roughness that can be measured with drones are
pressure ridges on lakes, which can be large or small, but indicate areas of unstable ice. Unlike surface
roughness, commercially available sensors on small UAS cannot see through ice, thus cannot be the sole
method used to determine ice thickness or continued ice growth throughout the season. However,
investigations are still needed to determine if these commercial off the shelf UAS and sensor packages
can determine if the ice has grounded to the bottom of the lake or river channel based upon the color or
other components of the UAS-collected information.
11.5 Break-Up
Ice break-up, because of anomalous weather conditions or seasonality, is variable by the ice, the
underlying water, and the weather conditions of the year. As described in Chapter 8, when the air
temperature remains above 32°F for 48 hours or more, the ice road can decay significantly. Using UAS as
a regular monitoring tool for ice roads can help ice road managers identify early indicators of ice
changes that could lead to break-up conditions, thus reducing risk to operators on the ice roads.
Upwelling, wet cracks, and open water can be observed with UAS when snow is not obscuring the
surface and can be mapped relative to the shoreline to identify detours and other risk reduction
strategies. Other ice features that may indicate break-up is imminent that can be observed with UAS
include arched ice (indicating flow beneath), lifted ice (when ice breaks from the shoreline and is
floating on the river, but not moving), and different ice.
One of the most powerful applications of UAS to increase safety prior to and during break-up is change
detection analysis. To effectively monitor changes in ice roads using UAS, regular flights over a defined
area need to be performed. The images and videos can be reviewed manually by ice road managers, but
that process can be very time-consuming depending on how large the flight area is, and how much
corresponding imagery or video has been collected. A more efficient method of reviewing larger flight
areas is to use software designed to ingest UAS images to produce a single 2D map, or orthorectified
map image, which can be displayed on a computer or printed in large format. If the same flight plan is
used, and the same processing methods, subsequent 2D maps of the same area can be created and
systematically compared to each other to identify changes from one flight to the next. This is the
fundamental concept behind the structure from motion (SfM) processing, also known as
photogrammetry, which is becoming more popular for landscape level change detection analyses. The
repeated flights, maps, and analyses allow ice road managers to watch the dynamics of the ice over time
to support management decisions, which are critical during break-up conditions. Artificial intelligence is
ideally suited to support change detection analyses by identifying anomalous features from one ice road