Wildlife tracking technology options and cost considerations

1. What they did

Researchers who want to follow animals over long distances can now choose from many satellite tracking systems, each with different strengths, limits and costs. The authors set out to help wildlife researchers pick the best option for their project.

They built a technology-choice decision guide, a flow chart that steps through the animal’s weight, its environment (dens, water, dense forest) and the project’s needs (accuracy, number of locations, speed, altitude, round-the-clock coverage, disturbance from people in the field). A second flow chart helps decide between solar-charged and ordinary batteries.

They tested the guide against four of their own satellite tracking studies:

• African elephants in Kruger National Park, South Africa: three elephants, GPS collars sending one location a day by geostationary satellite, over 5 years.
• New Zealand falcons in a plantation forest in New Zealand’s central North Island: five birds, light solar-powered Argos transmitters, over 3 years.
• An estuarine crocodile near Darwin, Australia: one animal, a battery Argos transmitter on the back of the neck.
• Northern royal albatrosses from Taiaroa Head, New Zealand: three juveniles followed from fledging through their first year at sea, solar GPS units sending four locations a day through Argos.

For each study they worked out the total cost per animal and the cost per data point (location), and compared them with what other technologies would have cost for the same number of locations over one year.

2. What they found

The decision guide

• Leaving cost aside, GPS is the method of choice because it gives accurate locations in quick succession, unless the animal or its environment rules it out.
• Weight sets the first limit. A transmitter should weigh no more than 3 to 5% of the animal’s body weight. With the smallest units then available (22 g for GPS with satellite download, 5 g for Argos, 0.26 g for VHF), an animal needed to weigh roughly 700 g for GPS, 170 g for Argos and 9 g for VHF. Animals under 9 g were too small to track.
• GPS and Argos signals cannot pass through water or solid surfaces such as dens and burrows, and GPS can struggle under dense forest canopy and in rugged terrain.

Looking back at the case studies

• Elephants: GPS with geostationary download worked well and they would choose it again. The battery only allowed one location a day, so they could not study movement within a day. Solar panels were unlikely to survive elephant behaviour and habitat.
• Falcons: the birds weigh 250 to 600 g, too light for GPS, so Argos was chosen. The solar units worked well even in dense forest.
• Crocodile: they used Argos, as other crocodile researchers had. In hindsight GPS would have been better: the crocodile could carry it and conditions suited it. Solar power was ruled out by the animal’s aggression and harsh surroundings.
• Albatrosses: the solar GPS units worked extremely well. The birds could have carried larger units with more locations a day, but the team chose to keep weight down.

Costs

• One year of tracking cost a similar amount per animal in all four studies, about US$7000 to US$9000 (Table 3). This included the transmitter, capture and attachment, satellite fees and analysis.
• The cost per data point varied widely: US$5 for the albatross (1460 locations a year), US$25 for the elephant, and US$51 to US$55 for the Argos-tracked crocodile and falcon.
• Longer studies lowered the cost per data point: for the falcon from US$53 in the first year to US$38 in the third, and for the elephant from US$25 to US$12 by Year 5.
• Compared with the alternatives (Table 4): Argos was cheapest per data point for the falcon and crocodile; GPS with cellular (GSM) download for the elephant, with GPS/GEO, GPS/Argos and Argos only slightly dearer; and light-level geolocation for the albatross, followed by GPS/Argos and Argos.
• VHF, as either the location or download method, was much more expensive per data point in every case study. VHF transmitters are cheap, but VHF needs a person in the field to find the animal, which brings high labour costs. For the elephant, VHF would have cost US$4018 per data point.
• For studies needing fewer than 15 locations, VHF is usually cheapest, followed by Argos. For larger samples, GPS is the most cost-effective.

3. What they concluded

• Choosing a tracking method comes down to three steps: (1) be clear about the data the project needs, (2) understand the limits set by the animal and its environment, and (3) work out the net cost per data point of each option.
• GPS is generally the best choice because it gives more frequent and precise locations and has more download options. Its limits are size and weight, and poorer performance under forest canopy, under water and underground. Where these rule out GPS, Argos or VHF usually work adequately, though with lower accuracy and fewer transmissions.
• GPS can be very cost-effective over medium and long studies. VHF is still the cheapest for very short studies needing few locations, or where field work is practical and needed anyway.
• When the animal or environment rules out GPS, researchers may have to compromise on their aims (for example, accepting lower accuracy for a long-distance migrant, using GPS with manual download to save weight, or tracking larger adults), or decide that tracking is not a good use of conservation funds.
• Home-built devices can be much cheaper, but design, building and testing costs should be counted, and unexpected failures can cost time and data. More expensive, robust systems may be more cost-effective overall.
• The paper is very time sensitive: transmitters keep getting smaller, cheaper and more capable, and new options (Fastloc GPS, Iridium and Globalstar download, spread-spectrum and wireless networks) were appearing. The authors stress the decision process rather than the specific figures.

6. Other technologies, skills or innovations

• Argos platform transmitter terminals (PTTs), whose positions are calculated from the Doppler effect by polar-orbiting satellites and graded into location classes (from about 150 m to over 1000 m error).
• GPS units (accuracy under 30 m) with many ways to send data home: Argos, Iridium, Globalstar, geostationary satellites, mobile phone (GSM) networks, or stored on board and downloaded by VHF or UHF receiver, from a drop-off collar, or on recapture.
• VHF radio transmitters located by triangulation or homing from a plane, vehicle, on foot or with automatic towers (reported precision 200 to 600 m).
• Light-level geolocation (archival) tags, which estimate position from sunrise and sunset times and must be retrieved.
• Solar-charged transmitters, duty cycles to save battery, and two-way units whose settings can be changed remotely. The elephant collar weighed 12.5 kg, half of it a lead weight to keep the GPS receiver upright.
• Low-cost home-built GPS devices from other studies: harnesses on pampas deer in Brazil gave 31,596 locations at US$0.07 each, and a unit on a giant anteater cost US$490.

7. Implications for conservation (from the paper)

• Satellite tracking is expensive and budgets often limit the tools conservation researchers can use, so understanding all the costs matters.
• Matching the technology to the research aims and the animal reduces transmitter failures and wasted money. Sometimes tracking may not be an appropriate use of conservation funding.
• Some studies suggest about 30 animals as an optimal sample, and that the number of animals, not the number of locations per animal, matters most for conclusions about a population. The number of animals tracked is the main driver of a project’s total cost.
• As tracking becomes cheaper and more accessible, researchers must make sure the data are analysed properly so that ecological and conservation aims are met.