Small animals form the majority of animal diversity on Earth, but there are fundamental gaps in our knowledge of how animals too small to wear GPS or satellite tracking tags move across the landscape. Learning how and where these animals prepare for migration, where their rest and refueling stations are, and how they move between different habitats may be critical to reversing the declines of many migratory species.
The Motus Wildlife Tracking System, a collaborative, continent-wide network of connected radio receivers that detect tagged animals from a range of scientific studies, aims to address these information gaps. A project of Bird Studies Canada, Motus deploys open-source radio tracking technology to study movements of animals as small as dragonflies. It is the largest collaborative automated telemetry project in the world. Since its inception in 2012, participating scientists have tagged and tracked more than 9,000 individual organisms and generated more than 250 million detections.
A summer tanager sports its new nano-tag, which will enable any Motus receiving station it passes to detect its presence and share that information with the researchers and project network. Image by David Bell.
Researchers fit 0.2-2.6 gram radio transmitters called nano-tags onto the bodies of birds, bats, and insects using glue, small sutures, and harnesses. The tags transmit coded radio signals that are picked up by nearby receivers. A detection consists of the transmitter’s unique code, its signal strength, and the GPS coordinates of the receiver.
What’s new is the combination of very small transmitters, the coordinated array of automated radio receivers operating across a large geographical area, and the centralization and coordination of the data. A given receiver detects and stores data from any participating researcher’s tagged individuals passing within its vicinity, regardless of who is operating it. Depending on environmental factors, which antennae are used, and the altitude of the tagged animal, receivers detect tags up to 15-20 kilometers (9-12 miles) away. The current distribution of over 300 Motus receivers allows scientists to track wildlife across vast distances and can be seen on this map.
A Motus receiving station, called a sensor gnome, consists of a small, inexpensive computer connected to a radio signal receiver, one or more antennae, open-source software, and a power source. Instructions to build a sensor gnome can be found online.
All Motus tags transmit the same radio frequency (166.380 MHz). Researchers register each transmitter to the Motus database when they activate it, and the combination of the tag’s code and the fixed rate at which it transmits a signal (its burst interval) make each tag uniquely identifiable. There are over 500 possible code options and about 25 different burst intervals, resulting in more than ten thousand unique tag combinations. This means that thousands of individuals can be tagged every year.
A Motus receiving station tower to monitor movements of American woodcock at the Moosehorn National Wildlife Refuge in Maine. Receivers can also be attached to an existing structure, though they have better range in open areas with clear line-of-sight. Image by US Fish and Wildlife Service.
The receiving stations continuously scan for active Motus tags that pass within the range of their antennae. Researchers visit a receiver multiple times per year to download its data and perform maintenance. They can also connect sensor gnome receivers to the internet through a cellular network or ethernet to upload data automatically. Depending on the frequency at which data are uploaded from receivers, researchers can potentially track their study organisms in near real time.
Once tag detection files are uploaded, they are immediately processed on the sensor gnome website. Within minutes, researchers receive a rough summary of the detections made by that receiving station, as well as the data, which they can further process and analyze. Motus provides tools and information to assist with processing of the data.
Motus data are centralized at the Bird Studies Canada national database, and Motus provides a coarse-scale version of the data on its website. The project lists the data, project details, and associated researchers together to encourage users, even those outside the network, to ask new questions, form collaborations, and combine datasets in new ways.
Pros and cons
Automating radio telemetry means receiving stations listen for passing tagged individuals 24/7. They preclude the need to track tagged animals manually, as with standard VHF tags, or to relocate and recapture an individual and remove its transmitter to access the data, as with GPS-based tags and light-level data loggers currently used on small birds. Sensor gnome receivers are as accurate as the hypothetical range of their antennae, making them less accurate than GPS devices but much more accurate than lightweight light-level data loggers.
A recently tagged yellow-billed cuckoo ready to be released. Image by David Bell.
The utility of the Motus network grows as more receivers become operational. If used in areas with many receivers, researchers can potentially generate more data points per tag than is currently possible with other tracking devices used on small birds.
Battery weight limits the number of times small GPS and light-level data-logging devices can turn on to record the location. The primary limitation to Motus transmitters is also battery weight. Smaller batteries limit how long the tags transmit a signal. A higher burst interval increases the likelihood that a tag will be picked up by a receiver but significantly reduces battery life.
Industrial noise can interfere with Motus reception, falsely registering it as detections of tagged animals. David Bell, a coordinator and data analyst for Motus, told Mongabay that researchers could identify false detections when, for example, a receiver gets just one or two detections from a tag, which would imply an animal moving impossibly fast. The project has developed an algorithm to help filter out false detections.
Citizen science and tracking
The potential for citizen scientists to operate Motus receivers has been an important aspect of the project from its inception. “When we were first setting up towers,” Bell said, “our preference was to ask landowners because when you go through parks, it can take a year to get all of the paperwork and permissions in order, whereas if you go directly to a landowner, they will say yes or no, and that’s it.”
Bell says that setting up a receiving station in North America with three antennas costs around $5,000. This cost includes a solar panel, battery, and charge controller capable of powering the receiver day and, importantly, at night, when many species migrate. It also includes the cost of a technician to install the system. Building homemade antennae, mounting antennae on pre-existing structures, and using an existing power-source can substantially lower this cost.
The project team is developing a miniaturized receiving station which uses a $25 Raspberry Pi 0 computer that integrates wifi capabilities, a USB hub, and a smaller antenna. Partnering it with an existing power source would bring the cost of a receiving station down to less than $500.
The smaller, less expensive omnidirectional antenna proposed for these receivers will detect tags from just 3 kilometers (2 miles) away, but, Bell says, having numerous, small antennae within a given area would enable the network to better detect fine-scale movements of tagged individuals.
Using Motus data
In a 2017 study, J. Morgan Brown and Philip Taylor, founder and committee member of Motus, tagged and tracked 107 blackpoll warblers in Nova Scotia, Canada. From detections from over 100 Motus receivers installed across northeastern North America, they found that most of the tagged individuals of this declining Boreal species wandered, up to 200 kilometers (124 miles), away from their ultimate fall migratory pathway over the Atlantic.
The breeding range (in blue) and fall migratory route (dashed lines) of blackpoll warblers. The black rectangle surrounds the inset region of the Gulf of Maine, where birds were captured and tagged (inside the open blue circle). Red dots represent the automated radio telemetry towers in the Motus array during fall 2014 & fall 2015. The dark gray contours represent changes in stable isotopes analyzed in the study. The blackpoll warbler breeding range was provided by BirdLife International (2014). Image is Figure 1 in Morgan Brown and Taylor (2017), Movement Ecology.
Their findings support a more complex migration strategy than was previously thought. Taylor says such findings can broaden what counts as a species’ critical habitat and redefine conservation strategies, which often focus only on a species’ breeding grounds.
In northeastern Colombia, Camila Gómez and her colleagues identified one such critical habitat. After measuring and tagging 133 migratory gray-cheeked thrushes, they found through the Motus network that birds that stayed longer and put on more fat (fuel) at a single stopover site on the northeastern slope of the Sierra Nevada de Santa Marta arrived at their northern breeding grounds up to 30 days faster than leaner birds that left earlier. One tagged individual flew 3,500 kilometers (2,200 miles) non-stop from this stopover site on the way to its breeding grounds.
Motus receivers detected 32 percent of the 133 tagged thrushes. Gómez suspected that some individuals were not detected because they flew to the west coast of North America where there are almost no Motus receivers. Bell described a chicken-and-egg scenario there, in which researchers cannot effectively use tags without an array of receiving stations, and people interested in operating stations may not be compelled to do so until there are research projects with tagged animals.
The study showed the critical importance of this stopover site for this species’ migration and subsequent breeding success, and it points to the likely effect of habitat quality on migration duration and success in general.
“The Gray-cheeked thrush stopover site is not inside a protected area,” Gómez said. “It’s actually a big privately owned coffee farm which has large relics of forest remaining.”
Gómez says the most important elevations for migrating songbirds in the Santa Marta range are between 700 and 1500 meters (2,300 and 4,900 feet) elevation in pre-montane forest, most of which is unprotected.
“Thankfully, the most common crop is shade coffee, and is not as bad for birds, but most of the forest at mid-elevations was cleared and used at some point,” Gómez said, though gray-cheeked thrushes themselves do not use shade-grown coffee as habitat.
A gray-cheeked thrush pauses in Magee Marsh, Ohio during its northward spring migration. Image by Don Faulkner, CC 2.0.
Gómez says that the findings have led to conservation work with coffee growers and landowners in the region, including a project on bird-friendly coffee in conjunction with the Smithsonian Conservation Biology Institute and educational outreach in five rural schools.
Gómez is also participating in the Neotropical flyways project, a joint effort of SELVA, the Cornell Lab of Ornithology, and Bird Studies Canada that aims to discover additional stopover sites and migrant hotspots within Central and South America, which she says will be good places for additional Motus receivers.
Vast, remote areas such as the Amazon basin will likely remain difficult to effectively cover with these types of receiving stations. Taylor says Motus can instead be thought of as one tool for answering specific questions which can be combined with other tracking technologies to form a clearer understanding of migrations and other animal movements.