'Drone-Netting' for Sampling Live Insects.

Drones have become valuable tools for biodiversity studies by providing aerial photographs; however, for most entomological studies, images, in particular those taken remotely, are usually insufficient; rather sampling of specimens is required. We equipped a cheap off-the-shelf drone with a net bag, flew it over the ground, sweeping the vegetation, and sampled adult and larval insects as well as spiders. 'Drone-netting' proved to be a versatile method for general insect sampling, particularly in inaccessible terrains. It is time- and cost-effective, minimally invasive, and adaptable for many research tasks in entomofaunistics; it shows a degree of representativeness similar to hand-netting, and caught specimens stay alive and can be released if not needed.

For entomofaunistic studies, a wide variety of active and passive sampling methods are available (e.g., Grootaert et al. 2010, Samways et al. 2010, Häuser and Riede 2015), some quite selective, others able to collect a wide range of taxa. Frequently employed methods include Malaise, color pan/bowl (Moericke), and interception traps, light, and especially old-fashioned hand-sampling with a net bag (‘hand-netting’), the latter having the fewest limitations with regard to the range of taxa sampled (Samways et al. 2010:90f), but being time-consuming and leaving footprints sensu lato in the study area.

Often, researchers are faced with the problem that a potentially interesting study site is inaccessible. There are terrains where one cannot set up a trap or walk about with a net because of steep slopes, wetland areas, dangerous animals, etc.; traps, which need to remain exposed for some time, are vulnerable to disturbance by animals or people; many taxa can only be obtained by active sampling. Such circumstances limit α-taxonomomical studies that are particularly needed in tropical habitats where our understanding of insect diversity is insufficient and where plenty of species remain unknown. This led us to test drones carrying a net bag for sampling insects (‘drone-netting’). Here we share our most positive experience from a pilot study and discuss the pros and cons of the method, aiming to inspire colleagues to apply ‘drone-netting’, too.

Drones (Unmanned Aerial Systems, UASs; often called Unmanned Aerial Vehicles, UAVs) already serve entomology well in the context of detecting, monitoring (aerial digital imaging of insect tracks and signs; Lehmann et al. 2015, Näsi et al. 2015, Vanegas et al. 2018), and managing (precise application of insecticides; Iost Filho et al. 2020) insect pest populations or assessing habitat quality in a conservation context (Habel et al. 2016, Ivosevic et al. 2017). Shields and Testa (1999) and Kim et al. (2018) sampled insects in the atmosphere over crop fields with a remote piloted vehicle and a drone, respectively. Could ‘drone-netting’ be a supplementary method for surveying insect biodiversity?

For our pilot study aimed to gain proof-of-principle, we used an off-the-shelf, low-cost, and light (750 g) quadrocopter drone produced for hobbyists (Mavic Pro Platinum; DJI, Shenzhen, China); with a larger model, the downwash of rotors was too strong to catch small insects. Homemade net bags of various shapes and sizes and types of netting material were tested; the denser the netting the higher the drag, and the larger the bag, the heavier it is and the more it impacts maneuverability. Eventually, a 180 g net bag with an opening in the shape of a semi-circle (30 cm radius) that was 100 cm long and with mesh width of 1 mm, attached to the drone with three adjustable 2 m long ropes, was chosen for the pilot study. The drone with the net bag was flown repeatedly at various speeds and for various distances. The flying height was controlled by the pilot so that the net bag swept over the tips of the vegetation (Fig. 1) in order to catch not only flying insects but also those sitting on plant tips. Flying for about 10‒15 min (i.e., distances of 1,500‒3,000 m at a speed of approximately 3 m/s), always resulted in worthwhile catches; longer flights can overload the net, making the selection of desired specimens more difficult.

Fig. 1.

Drone with a net bag sweeping the tips of the vegetation and sampling adult as well as immature insects and spiders

All of our test flights sampled not only adult but also immature insects and spiders, clearly demonstrating that in addition to flying insects, vegetation dwelling organisms can be successfully captured using the ‘drone-netting’ method. In every flight, members of all insect orders expected in a given terrain were collected. The sampling appeared similar to that obtainable by hand-netting, but uncomparable to Malaise trapping, which samples mainly Diptera and Hymenoptera and rarely immatures.

Generally, very large differences between specific flights were found with respect to both the diversity of species as well as the numbers of individuals, heavily dependant on the 1) habitat and its vegetation, 2) date and time of day, 3) distance and time flown, 4) weather conditions (relevant for both insects and drone flights), 5) personal performance/skills of the pilot, and 6) shape and size of the net and its netting material. Overall, the impact of all these variables was trivial as the success of sweep-netting by hand also depends on all these variables, including the individual influence of the catcher.

Entomological sampling is generally characterized by high variation in success which makes it difficult―if not impossible―to compare catches qualitatively and quantitatively and to interpret them accurately from a community perspective. This problem remains when carrying out ‘drone-netting’. Deprecatory expressed: ‘Drone-netting’ is non-reproducible―it is thereby as good as other means in the entomological arsenal. Although ‘drone-netting’ is not the future method for insect biodiversity studies, it is a versatile addition to the existing qualitative sampling techniques. It is unique in its applicability in inaccessible habitats; it is easy, cheap, safe, environmentally friendly, relatively fast, and it provides quality specimens; it does not leave footprints in the habitat, and―as with hand-netting but in contrast to sampling with a Malaise trap and other passive sampling methods―specimens remain alive so that those not needed can be released; another advantage is that within a given time larger areas can be sampled.

We report ‘drone-netting’ in principle but deliberately exclude details. In addition to the variables discussed, the types of drones currently available are very diverse with respect to their technical specifications, and the market is rapidly expanding and further diversifying. As the type of drone changes, so does its weight-bearing capacity and the landscape/vegetation/meteorological conditions under which it can be flown with a net bag; technical aids available (e.g., autopilot) also differ greatly. Since ‘drone-netting’ depends on all of the aforementioned variables, plus the flying speed and time of the drone and also on the specific research question (the insects one aims to sample), there is no detailed reproducible protocol.

We have tested ‘drone-netting’ of insects mainly in inaccessible terrains, however, in readily accessible habitats it can also be very helpful. When adjusted appropriately it may serve as a supplement to conventional sampling tools. For example, costly and spacially limited tasks studied by sampling from a tower (e.g., Haddow et al. 1961) or from a canopy platform/walkway (Ozanne 2005) may be facilitated or supplemented with ‘drone-netting’. Sampling nocturnal insects in the air space is another potential application―what do we know about insects active at night apart from those that are attracted to artificial light sources? The field of aeroecology (Drake and Bruderer 2017) will likely benefit from employing UASs; ‘drone-netting’ in free space is devoid of some methodological obstacles, such as the maneuvering skills of the pilot. While future progress in drone and drone-flying technology will likely promote ‘drone-netting’ and also enable further applications, including flying a drone with a net in a forest, limitations related to sampling insects in general (see above) will, unfortunately, remain.