Unlocking the mystery of lizard movements in a changing climate

October 20, 2021 by Eran Vijayakumar

There’s much we don’t know about lizards. That’s why a team of researchers at U of T and Ohio Wesleyan University set out to help us understand this notoriously complex species better, especially how their behaviour might adapt to our world’s changing climate.

To unlock this mystery, U of T undergraduate students Sophie Berkowitz and Simone Collier are using computational tools to analyze lizard movement and body temperature under the supervision of Vianey Leos Barajas, assistant professor in the U of T Department of Statistical Sciences and the School of the Environment. For this project, the team of students worked closely with ecologists from Ohio Wesleyan University, namely Assistant Professor Eric Gangloff and Wesleyan undergraduate students Ciara Pettit and Sierra Spears, who cared for the lizards and provided the primary data of lizard movement and temperature.  

The team’s research focuses on analyzing movements of lizards when placed in an arena with a “thermal gradient”, meaning an observation area where one end is warmer than the other. The sand-filled arena is one metre long and has a lamp as a heat source. Using open-source object-recognition software, statistical and environmental science student Sophie Berkowitz extracted movement data from hours of lizard footage to prepare a dataset for statistical analysis.  

Most people who don’t spend hours observing lizards don’t know the unique challenges of this type of research, such as when a lizard tries to play hide and seek. Berkowitz explains that a key challenge was tracking the movement of the lizard when it hid from the camera by crawling into the sand or when trying to crawl up the side of the observation arena. “They try to bury themselves in the sand and get caught. Then the ecologists have to fish them out. I think the moral of the story is that lizards have minds of themselves, and our research definitely has to accommodate for that,” she says.

 

To gain insights into how lizards might adapt to climate change, undergraduate students Berkowitz and Collier analyzed video footage of movements of lizards when placed in a sand-filled arena with different temperatures.

 

With the dataset of lizard movement, body temperature, and environmental temperature, mathematics and environmental science student Simone Collier used hidden Markov models to gain insight into the lizard’s movement and changes in behaviour. “What we're trying to do with these models is to determine their thermoregulation behaviours. So, we want to find patterns in the lizards’ movement based on, how they move between these different temperature zones,” says Collier.

Lizards are ectothermic. Unlike humans, they rely on the temperature of their environment to regulate their body temperature. That’s why, when a lizard is put into a world with a rapidly changing climate, its behaviour is destined to change. These changes in behaviour hold important clues to how lizard species might be impacted by temperature changes due to climate change in their natural environment.  

"It's really important for us as scientists to get a better understanding of how their thermoregulation works, and how it is affected by their environment,” says Berkowitz. “The hidden Markov models work on the premise that the observations are the product of an underlying behavioural process. So, if we can understand things that we can't readily observe, we can help understand the future of our lizard species in terms of climate change."

Statistics has many multidisciplinary research applications. This research partnership highlights the intersection of statistics and ecology: Gangloff’s team of ecologists provided the primary data to the statistical and environmental sciences team at U of T for further analysis.

“We get to do a lot of fun projects and work with a lot of diverse students that are quite capable. They are using their statistical skillset to expand the notion of what statistical sciences mean in practice,” says Leos Barajas.

The team is continuing their modelling and plans to publish their results, along with their modelling tools.

In addition to contributing to the field, the two U of T student researchers enjoyed gaining hands-on research experience under the supervision of a professor while building important skills. Both Berkowitz and Collier encourage students to pursue research projects they are interested in. “If you meet a professor that's researching something you're interested in, just reach out and see what comes of it. You never know,” says Collier.