EDINBURG — Sitting on top of a building at the University of Texas Rio Grande Valley’s Edinburg campus is a short white box, home to the university’s resident bee colony.
The hive is part of a three-year project aimed at pairing computer science with apiarism to develop better ways to monitor bee populations.
The research began in September and is being spearheaded by Dongchul Kim, assistant professor of computer science, and Joanne Rampersad-Ammons, professor in the School of Earth, Environmental and Marine Sciences. Their research is funded by a federal grant from the USDA.
For all her positivity, Rampersad-Ammons, a cheery woman with a Trinidadian accent and a passion for agriculture, has a surprisingly apocalyptic view of the world’s agricultural situation.
“Supposedly in 30 years we’ll need to feed half of this world’s population over. That’s a scary number, you’re talking like three billion more people,” she said. “We need to start now to think about how we’re going to be dealing with that, and it’s probably going to be a scary, scary situation, because food prices are inching up.”
Rampersad-Ammons said that concern is what makes bee research so important.
“When we were writing that grant, it was with me thinking about the crisis that’s about to hit,” she said.
According to Rampersad-Ammons, bees play a particularly important role environmentally and agriculturally.
“One in every three bites of food that you eat is coming from a pollinator of some kind, of which the bees are a huge piece,” she said.
Recently, Rampersad-Ammons said, bee populations have been in an especially perilous position
“In years past we’ve had what’s called colony collapse disorder. They didn’t know what was going on so they just put this label on it,” she said. “Part of it seems to be Varroa mite, which has become a pest throughout the entire United States. Part of it is bees that have African genetics are more predisposed to just get up and leave when they’re in a situation they don’t like, it’s called absconding.”
To better understand those threats to bee populations, Rampersad-Ammons rigged up her colonies with a variety of sensors and cameras, watching individual bees, how they interacted with each other and with their environment. She’s watched a bee toss a pesky housefly out of the hive and even spotted one of those pesky Varroa mites.
“They had already started doing some of this stuff, people putting temperature sensors, humidity sensors, sound sensors, stuff like that. The issue was, all of those will help you look at a beehive on a hive basis, but what if you wanted to look at individual bees? The only way to do that kind of stuff is if you start putting cameras on it,” she said.
Rampersad-Ammons and Kim hope to use their findings to develop AI to better understand bee populations, hopefully opening the door for the creation of technology that can help commercial and hobby beekeepers better care for their hives.
“There’ve been several big groups that have developed computerized hive technology; the thing its, it’s so expensive,” she said. “Most people in this area, when you say agriculture, they think traditional agriculture. But the engineers can change a lot, and there’s high-tech stuff going on.”
Rampersad-Ammons said she hopes the research being conducted by her and her colleague Kim will help apiarists understand threats facing their hives.
“In a lot of cases you actually have to physically open up a hive to figure out what’s going on inside of it. If you put sensors inside, they might help you a little bit. But what if you could actually count every bee? What if you put a scale, a scale would actually register the hive getting heavier when they all get back in the evening,” she said. “You can start tracking what exactly is going on with that population without going into the hive. That’s the kind of stuff that we want to be able to do. It would allow you to go into that hive and pick up on things, and perhaps do something about it before it’s too late for the group.”