Unlocking the Secrets of Caribou Migration—From 2,000 Miles Away
Lehigh researchers use remote sensing data to reveal how melting snow signals caribou to begin their spring journey
When Joan Ramage, associate professor of earth and environmental sciences, began working with Mariah Matias ’20, G’23 on her master’s degree thesis, the pair never guessed they would discover a key to caribou migration behavior more than 2,000 miles away in Canada’s northwest territories.
Every spring female caribou very suddenly begin to move from forests in northern Canada to their calving grounds in the Arctic tundra hundreds of miles away to have their young. Scientists are interested in understanding more about how the animals know when to start migrating. Utilizing data from remote sensing equipment that included GPS and microwave technology, Ramage, Matias, and their colleagues identified a correlation between the start of the migration and snowmelt. The research was presented at the American Geophysical Union annual meeting in December 2024. The findings also were published in the journal Remote Sensing.
Ramage and Matias’s research indicates that changes in snow texture and consistency when it begins to melt appear to be a cue for the move to calving grounds. Their discovery is perhaps even more fascinating because they were able to identify it without ever setting foot in the northwest territories, using remote sensing data tracked by the Canadian and United States governments.
Interest in studying caribou has grown along with concerns about global warming since Arctic temperatures have warmed nearly four times faster than the rest of the planet. Ramage and Matias studied migration data from 2007 to 2016 for 117 female caribou in the Bathurst Herd, plotting it against data about snowmelt from the same period. The data came from two sources. Herd movement is tracked via GPS collars on caribou used by the Government of Northwest Territories’ Department of Environment and Climate Change. The department collects the data as part of efforts to stem the declining numbers of caribou, which have dropped from 400,000 to less than 8,000 over the last 30 years.
Information about snow was gathered using satellite observations of microwave radiation on the earth’s surface. The data was available from the United States Defense Meteorological Satellite Program (DMSP) which has been collecting information since the 1980s. Microwaves are sensitive to liquid water, so the satellites pick up on water when it begins to appear in the snow, especially at the surface.
Combining Research Interests
Ramage’s research interests include remote sensing technology and snow, but she says that Matias provided the impetus for studying the caribou. A native of Philadelphia who had worked in Ramage’s lab since her first year as an undergraduate earth and environmental science major, Matias gravitated to research linking the environment and animal behavior. A first-generation college student in Lehigh’s Rapidly Accelerated Research Experience (RARE) and an Eckardt Scholar, her undergraduate capstone project centered on studying white-tailed deer in New York state using drone technology. In developing her master’s degree thesis, she once again sought a project focusing on animals in their native habitat. She examined available data to pinpoint a research focus.
“I did a ton of digging online,” Matias explains. “I found an open data set on ArcGIS online, which is an online public service platform for remotely sensed data where people can publish… which is where I found the GPS collar data for the caribou.” Her next step was to find who owned the data and obtain permission to use it. Ramage had done previous research a bit further south in the Yukon, so she knew sources to consult for snow-related data.
"They're kind of busying around for a while, and then out of nowhere, you see this drastic, rapid movement north to the calving grounds. I said, ‘That’s what I need to study. What are they responding to?’”
Pinpointing exactly what to study began with an analysis of the GPS data. Matias studied time series – a series of data points that tracks how a variable changes over time. It allows an analysis of trends, patterns and fluctuations. Sitting at her computer for hours, she watched data points showing movements of the caribou herd. Each tiny point reflected the movement of an individual caribou.
“I was watching them over and over again. You rewind and play it faster, you play it slow and then you start to pick up some of the patterns that seemed unique,” Matias explains. “One of those patterns was during their spring migration. They're kind of busying around for a while, and then out of nowhere, you see this drastic, rapid movement north to the calving grounds. I said, ‘That’s what I need to study. What are they responding to?’”
Painstaking Analysis
Using Passive Microwave Remote Sensing data, they extracted useful information about the freeze-thaw cycles and compared them with the female caribou migration data to see if there were any association or patterns with snow melting. The snow quality is represented by something called brightness temperatures, which combines surface temperature and emissivity. Ramage's team determines whether snow is frozen or melting using the brightness temperature and how it changes over a daily cycle, she explains. The temperature increases and other properties change when the snow starts melting “The brightness temperature also goes up, and it is noticeable” Ramage says.
Data was modeled in collaboration with Eliezer Gurarie, an assistant professor at SUNY College of Environmental Science and Forestry and an investigator in its "Fate of the Caribou Project." By identifying periods of caribou migration and comparing it with the snow quality, they learned that caribou responded to changes in the snow and started their migration when it began to melt. They also concluded that earlier snowmelt can lead to earlier migrations and the melt/refreeze cycle can cause longer migrations and delay the caribou from reaching their calving grounds.
Arriving at these conclusions required hours of painstaking analysis. “I did 2,000 to 3,000 interpolations for each onset of (herd) movement,” Matias says.
Ramage notes that their study does not identify what other factors caused by the melt/refreeze cycle might be influencing migration. “We don't know if it's access to food or changes in how difficult it is for them to walk,” Ramage notes. It may be more difficult for caribou to reach the lichen, which is their primary food source, under the snow. Thawing and refreezing may make it tricky for them to keep their footing. Such questions would need to be answered in a different study.
Ongoing Research
Since graduating from Lehigh, Matias has moved to Charlotte, N.C., where she works for Esri, as a support analyst for its ArcGIS its geographic information system technology. It is the technology she used extensively at Lehigh. The research she began about the caribou is continuing in Ramage’s lab. The Canadian government shared data about four herds, making it possible to test findings found in the first study with other research populations.
Since January 2024, Becca O’Brien, a student in the Integrated Degree in Engineering and Science (IDEAS) program with concentrations in earth and environmental science and mechanical engineering, has been working with Ramage. They are analyzing data about migration and snowmelt for the Bluenose East Herd, which neighbors the Bathurst Herd in the northwest territories. Ramage and O'Brien are interested in whether the results from the Bathurst Herd are typical of multiple herds. Among the ideas they are considering as possible reasons for differences in caribou movement patterns are whether the animals must expend more energy to move through wet and slushy snow, if herds take advantage of snow crusts for greater mobility, and the ease of finding food.
“It's really incredible for students to have this opportunity to work with a diverse set of scientific expertise and data expertise.”
O’Brien, a native of Overland Park, Kansas, says the project offers a research opportunity that combines her two academic concentrations. “I was looking into intersections between engineering and environmental science, and remote sensing is a really good application of that,” she says.
In addition to contributing to knowledge about caribou, Ramage says the research also presents great options for students. “It's really incredible for students to have this opportunity to work with a diverse set of scientific expertise and data expertise,” she states.