RFID Goes Out to Pasture, Matching Ewes With Lambs

By Claire Swedberg

Texas A&M researchers are testing technology from Sapien to capture the identities of ewes and their lambs in Australia, and to match the families together based on their activity.

While all agricultural work is labor-intensive and, in many ways, unpredictable, the challenge is especially felt by sheep farmers. These wooly livestock require a wide range of pastoral graze to thrive, and for that reason are difficult to contain. That makes control over breeding, especially when it comes to understanding genetic lines, an imperfect science for farmers. Ewes give birth in the fields, and both ewes and lambs move freely through pastures. Farmers must find ways in which to leverage the most productive ewes, and ultimately to ensure the most profitable lamb production.

That's where RFID can help. Texas A&M AgriLife Extension is testing a low-frequency (LF) 125 kHz RFID-based system provided by Australian company Sapien to match mother sheep with their lambs, and thereby understand which ewes produce which characteristics.

Reid Redden

The system employs RFID reader panels at paddocks through which sheep travel single file to access water, and its software captures and analyzes the behavior of sheep as they travel, and thereby identifies those that are related. This information enables breeders to predict the ewes that will produce the healthiest, most productive offspring, says Reid Redden, a Texas A&M professor and AgriLife Extension state sheep and goat specialist at San Angelo.

RFID technology is already being used to identify livestock, and to obtain information about animals' health, as the path of custody from one party to another. In Victoria, Australia, that kind of RFID tracking for cattle became mandatory in 2003—which is when Sapien was launched, says Rob Wyld, the company's managing director. The firm has focused on providing farmers with solutions to help them gain their own benefits from the RFID tags that they are mandated to apply to cattle, goats, sheep and other livestock.

In the case of sheep, farmers typically have little visibility into which ewe gave birth to which lambs. That matters on a genetic level, Wyld says, because a ewe that weans twins weighing 30 kilograms (66 pounds) may be a better breeder than one that weans a single 20-kilo (44-pound) lamb. By knowing which of their sheep are the healthiest, most productive breeding ewes, farmers can better manage which animals they intend to breed again.

Traditionally, farmers had to catch lambs, place them in a contained area separate from their mothers and then try to link up particular ewes and lambs. They would try to tag the lambs with printed serial number tags, in order to link them to a specific ewe near which they are physically observed to be located. Sapien calls its technology-based solution to this labor-intensive process the PedigreeScan. This software- and reader-based solution enables farmers to understand, with relative certainty, what types of lambs the ewes deliver, Wyld says. Sapien's PedigreeUnit reader is an alternative to other off-the-shelf readers, he explains, most of which can be cumbersome for farmers.

What's more, the devices often require large batteries and possibly a solar unit to help power them. They could be difficult to install, as well, for users who lack technical backgrounds. "So we built our pedigree unit using less power (about 10 to 15 watts)," Wyld says, in contrast with most traditional readers that require 120 watts to operate. The device comes with a built-in reader from Texas Instruments.

Sapien accomplished the low-power consumption by reducing the read rate and programming the reader to go dormant when no tags are within its vicinity. The panel reader can be attached to a battery via crocodile clips, and can be mounted on the side of a single file lane erected near a water source or other attractor in the paddock.

When ewes and lambs are vaccinated or undergo health inspections, RFID tags are attached to each animal's ears. Every tag is encoded with a unique ID number linked, in the Sapien software, to data about the animal to which it is attached, such as its approximate birthdate and its category as lamb or grown ewe. The software, Wyld says, was developed with support from the Cooperative Research Centre for Sheep, which includes an algorithm that scans through read data to match ewes and lambs. Multiple farms throughout Australia are now using the PedigreeScan system to track the ewes and the lambs to which they give birth.

At Texas A&M, the system is being tested on sheep and goats. "We're testing the applications using the technology developed in Australia," Redden says. The Texas A&M researchers install readers at the entrance to a fenced-in water trough. The paddock through which the animals enter is approximately 2 to 3 feet wide, and they pass through in single-file. Lambs typically follow their mother, at which time their tag ID numbers are also captured. The tags are provided by Shearwell Data, a U.K.-based animal identification technology company.

Rob Wyld

The reader comes with a raceway antenna that is around 2 feet wide, Redden reports. As the animals pass through, the reader captures tag data and stores the date and time each tag is scanned. The operator can visit the PedigreeScan unit on a regular basis and download the content of its memory via a Bluetooth link into a PC laptop. The software can interpret whether a particular animal is passing into or out of the water trough area. However, lambs do not always stay near the mother—and in some cases, lambs that don't belong to a ewe may also follow her. Therefore, the software collects data and analyzes it over a period of time, using many data points collected from multiple events.

In West Texas, a typical sheep pasture measures about 600 acres, according to Redden, and tracking the genetics of sheep without an RFID system can be difficult. With the solution in place, he predicts, the software could help farmers to identify multiple features. For instance, ewes that most often give birth to twins would be more productive breeders. Their own offspring could then be expected to carry on that trait, so they, too, could be good breeders. In addition, some sheep feature higher-quality or parasite-resistant wool, or may simply display overall superior health.

"If the animals were always in confinement," Redden explains, "you don't need [the RFID solution], but sheep are not a confined animal." Additionally, some parts of Texas have larger pastures than others for the same number of sheep, based on the amount of rain and, therefore, the quantity of green grass per animal.

Texas A&M has been testing the technology for three breeding seasons, Redden reports. The researchers have had to work around the variabilities of farming, including the fact that there may be less data collected during heavy rain, since the sheep would less frequently enter the water trough area under such conditions. Read rates have thus far proven to be 95 percent successful, he says, while the software algorithm is matching ewes with lambs at an accuracy rate of about 85 to 95 percent.

The technology has been tested at four different Texas ranches to date, Redden says, on both sheep and some goats. Texas A&M is now evaluating its next plans.