306 Monitoring Threatened Species and Ecological Communities
be the case (e.g. species living within dense forest canopy). Second, beyond the
obvious cost factor, one must decide which type of drone is appropriate: fixed-wing
(shaped like a small plane) or multi-rotor (based on the working principles of a
helicopter, but with several rotors). Compared with multi-rotor designs, fixed-wing
drones typically f ly longer and are less noisy, but are unable to hover or f ly at very
slow speeds, require open spaces for take-off and landing, and require more
training for the operator. Some applications, particularly related to use of imagery
(e.g. stitching orthorectified pictures or creating 3D terrain models based on
pictures) require substantial expertise.
Another important consideration is legislation, which often restricts the type
and use of drones. In Australia, the Civil Aviation Safety Authority specifies that
the only drones that can be f lown for research without a pilot licence must be
under 2 kg, and even then with restrictions regarding where and when – these
conditions will often be easier to meet in remote areas. Finally, social and ethical
issues of privacy (what is the impact of capturing video of humans living in the
area?) and animal welfare concerns (what is the impact of the drone on wildlife?)
should be kept in mind.
A National Environmental Science Programme Threatened Species Recovery
Hub project led by Dr Lahoz-Monfort from the University of Melbourne is trialling
the use of drones to monitor the elusive Lumholtz’s tree-kangaroo Dendrolagus
lumholtzi in the Atherton Tablelands, Queensland, in collaboration with wildlife
biologist and tree-kangaroo expert Roger Martin from James Cook University.
Lumholtz’s tree-kangaroo (LTK) lives in rainforests in far north Queensland,
occupying both larger forest tracts but also small remnant patches scattered among
a largely agricultural landscape (Martin 2005). Tree-kangaroos are arboreal, and
thus difficult to detect in the high dense tree canopy. Traditionally they have been
monitored by spotlighting at night (looking for eye shine) or searching for pellets
on the ground (which may be confused with other macropod species). The
population abundance and trends of LTK are poorly known, due to the difficulty of
reliably detecting, and hence monitoring, the species. Although currently classified
as Near Threatened by the International Union for Conservation of Nature, their
true conservation status remains uncertain.
Drones that can carry different types of measuring devices and thermal sensors
are a promising option (Gonzalez et al. 2016). LTKs bask high in the canopy in early
morning, particularly in colder winter months. To improve LTK detectability, a
low-cost thermal detection system was developed that can be attached to a
commercial drone and f lown over the canopy. The custom-made box includes a
low-resolution (80 × 60 pixels; ~A$350) thermal infrared sensor and a mid-
resolution (8 megapixels) colour camera, both controlled by a cheap single-board
computer (Fig. 23.1). The thermal ‘hotspots’ found in the images (either manually or
using automated computer algorithms; Dell et al. 2014) will indicate where to check