Ants don’t change their behavior to avoid sublethal warming

This blog post is provided by Elsa Youngsteadt and tells the #StoryBehindthePaper for the paper “Can behavior and physiology mitigate effects of warming on ectotherms? A test in urban ants”, which was recently published in Journal of Animal Ecology. In their paper they explore how ants might react to climate change, and whether they can adapt their behaviour to new conditions.

Spring flowers are bursting earlier, growing zones are marching ever poleward, and bumble bees are shrinking away from their southern range limits. Climate change is well underway and some of its biological effects have been established for decades. But other effects are more subtle. For example, physiologists predict that, as temperatures increase, some animals’ metabolisms will also ramp up. They’ll need more food to get by, they’ll live faster, and die younger; some will go extinct. But are animals really getting hotter as the climate warms up?

Ectotherms are animals that don’t maintain a constant body temperature, instead letting it fluctuate along with the environment. But they do exert some control, for example by moving between sun and shade or retreating underground at certain times of the day or year. Given this ability to thermoregulate, ectotherms might also buffer their own exposure to climate change by subtly shifting their activity to cooler microclimates within their habitat. Although some modeling studies have suggested that this ability would save many insects from heat-induced extinction, it’s not clear that they really have the behavioral flexibility to pull it off. This is one of the main questions my team and I had heading into the spring of 2020.

Of course, the spring of 2020 didn’t go as planned. We had meant to ask this question in the warm tropics, where ectotherms are closer to their maximum heat tolerance and may take more urgent measures to stave off heat stress. But instead of spending the spring in the tropics, we spent it in lockdown, trying to stay sane and figure out some justifiable use of our time. Once we began to creep back into public, we decided to pilot our tropical methods at home in Raleigh, NC. That pilot project never made it out of the temperate zone, but it did get bigger. Ultimately, we found that our own local ants—although not obviously heat stressed—are surely warming up.

Checking for ants during a nighttime transect survey. Photo credit: Sara Prado

To detect behavioral thermoregulation, it’s not enough to just measure an ectotherm’s body temperature. You need to know its preferred body temperature, and what range of temperatures are available in its habitat—in the thermal mosaic created by shade, sun, darker surfaces, lighter surfaces, and little breezes. With those pieces of information, you can see how challenging the thermal environment actually is. (If the animal moved carelessly through the habitat, would it ever encounter intolerable conditions, or only pleasant ones?) And even if no part of the environment is truly pleasant, you can see whether the animal is making the best of it, spending time in the microclimates that bring it closest to its preferred temperatures, even if it’s still way off.

We chose ants as our focal ectotherms because they’re common models for thermal biology research, and because they’re generally important participants in their ecosystems, acting as predators, herbivores, seed dispersers, and so on. To look at a climatic challenge to ant thermoregulation, we worked on an urban warming gradient. All our sites were forested, but the ones located along urban greenways averaged about one degree Celsius warmer in the summer than the non-urban forest sites.

Map and examples of study sites. All sites were forested, but some were hotter than others due to their location within the urban heat island of Raleigh, NC.

We worked with several ant species that were common across most of the study sites, and collected four kinds of data. First, we measured thermal preference—the range of temperatures where each species chose to settle on a gradient from a heat block to an ice bath. (Because the university was still closed, I did this assay in my home office using an apparatus my husband helped build during lockdown. The frozen bowl of our ice cream maker helped keep the cold end cold.) We also measured each species’ maximum heat tolerance. (Sarah Ferriter, an undergrad researcher, took ants to her apartment for this assay.) And finally, research associate Sara Prado measured available and occupied temperatures along transects at each site in the field. For these last two, Sara used operative temperature models of three focal ant species; these were dead, posed ants mounted on thermocouple probes. You can think of them a little bit like a species-specific heat index thermometer: they incorporate effects of air temperature, surface temperature, solar radiation and wind, as well as the size, shape, and color of the ant species. It doesn’t necessarily tell you the body temperature of live ants, but it tells you how comfortable the thermal environment is. If the operative temperature model is hotter than the live ant prefers, then the live ant shouldn’t just hang out in that spot until her own body reaches that unpleasant temperature; she should move along and find some shade.

Operative temperature thermometer for the chestnut carpenter ant (Camponotus castaneus), made from a posed ant specimen (not alive) mounted on a thermocouple temperature sensor. Photo credit: Elsa Youngsteadt
Thermal preference arena. The lanes of the arena span a hot plate and an ice bath; ants indicated their thermal preferences by choosing where to settle within that range. Photo credit: Elsa Youngsteadt

If ants are actually good thermoregulators, then we should find them disproportionately often at places and times where the operative temperatures most closely matched their preferred temperatures. This was the case for four of our five focal species. (The fifth had such a broad range of preferred temperatures that we could hardly ask the question.) But if ants were capable of saving themselves from climate warming by shifting their activity patterns, they should also occupy cooler parts of the transect at sites that were too hot, and hotter parts of the transect at sites that were too cool. This did not happen. Cool-loving ants, for example seemed to use fixed behaviors (like being more active at night) to avoid the hottest conditions across the board—even at sites where the temperature was more comfortable during the day.

Black field ants (Formica subsericea) at a bait station. Ants were only slightly more likely to use a bait station if it was placed in a preferred microclimate than if it was placed in an uncomfortable microclimate that was too hot or too cold. Photo credit: Sara Prado

None of the ants got dangerously hot or cold during our study. But they did get warmer than they preferred to be, and they didn’t shift their behavior to compensate. That means urban ants are hotter than non-urban ants, and as climate change progresses, future ants will probably be hotter than past ants. If this pattern holds, their metabolisms will increase, and they’ll live faster-paced lives. Because ants are social animals, the bottom line for the superorganism—as opposed to just its individual workers—is still unknown. That’s something we’d like to know next. Only this time, I hope it doesn’t take a pandemic to send us back out to our local field sites to find out. 

Author bio

Elsa Youngsteadt is an assistant professor in the Department of Applied Ecology at North Carolina State University.

Lab website:

Read the paper

Read the full paper here: Youngsteadt, E., Prado, S. G., Keleher, K. J., & Kirchner, M. (2022). Can behaviour and physiology mitigate effects of warming on ectotherms? A test in urban ants. Journal of Animal Ecology, 00, 1– 12.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s