The most ambitious scientific experiment ever to test the effects of global warming trains the heat on ants. Akshai Jain explains why, and how
How would the Pink Panther predict what would happen as the earth heated up? Well, he’d do it the most manageable way. He’d take a small creature, perhaps an ant, and light a fire under it. Then he’d sit back, watch and extrapolate the experiment to all other creatures.
That, however, was then. Much before global warming and climate change had entered the popular imagination. What belonged to the realm of Pink Panther’s comic-mystery universe is now the stuff of the most advanced scientific experiment ever conducted on the effects of global warming. Playing police detective Clouseau are scientists from Harvard University, the University of Vermont and a few other institutions. The sets moved from Paris to two patches of forest in America and, instead of singeing single ants, they trained the equivalent of giant hair dryers onto many colonies of ants. Then they sat back, watched—for six years—and perhaps whistled that famous tune.
Barring Donald Trump and his ilk, few reasonable people still dispute that climate change and global warming are real, and here to stay. The weather does seem to be changing—the extremes of heat, cold, flood and drought are getting only more extreme. Beyond this, however, looking especially at the effect on animals, the narrative swings wildly between science and fiction. In the extreme version, the weather turns perpetually violent and there’s a dramatic collapse of species.
But even the more sober, ‘scientific’ assessments use anecdotal evidence to arrive at vast generalisations.
Leopards in India, we’re told, are on the move, venturing into higher forests as the plains heat up. As are apple trees—whose cultivation has moved from the lower mountains to high elevations. Mackerel and sardines, once found in the oceans bordering southern states like Kerala and Goa, have moved north with the warm currents, to the seas near states like Gujarat. The spread of invasive species like the fall armyworm, which is devastating corn and sugarcane crops, has been aided by rising temperatures in most parts of India.
These observations, while probably true in small pockets, are too meagre to offer firm conclusions. Even in cases where the numbers are adequate, how do you attribute an effect to warming, rather than to, say, a natural change in vegetation, a long-term fluctuation in ocean currents, a change in soil nutrient composition? Or an ecological disruption such as a new railway line, or any one of the dozens of other factors that affect ecosystems?
One solution is long-term monitoring—observing an ecosystem for decades at a stretch. This is as difficult and expensive as it sounds, and impossible in the case of animals that are mobile. The other alternative is to translocate a species to a laboratory setting where changes in climate can be simulated and speeded up—but this too is rarely feasible because recreating all the other facets of their ecosystem isn’t possible.
What might work though is experimenting with insects—observing ants, perhaps, in their natural setting. Many colonies of ants would fit into a small observational forest area a few metres large. Since ants don’t venture very far from their colonies, they would effectively spend their entire lives inside this observational area. If this area could be heated, you’d be able to simulate global warming and, given a little time, be able to gauge its effect on ant species and communities. What you’d be doing, in effect, is the equivalent of heating up an ecosystem at a controlled pace, almost as if you’d cranked up the sun’s temperature.
This sounds like an elegant and scientifically rigorous solution to the problem. But why ants? Because ants are the most successful creatures on land. There are more than 22,000 species of ants spread across every single landmass on earth. Their combined weight is equal to that of all humans on earth. They perform a dizzying array of ecosystem functions—from seed dispersal to replenishing micronutrients in the soil. Thousands of other species of insects, as well as birds, depend on them for survival. Butterflies like the Silver Studded Blue cohabit with the Black Garden Ant. A species of hoverfly lives off a type of wood ant and a clever beetle specialises in getting ants to feed and protect it. In all, the number of insect species that relies on ants is twice the number of all mammal species.
If life is a web, then ants live on every single strand of it. This makes ants a perfect bellwether for climate change. Studying the impact of warming on them is likely to lead to a better understanding of its impact on almost everything else. The impact on ants is in many ways a proxy for the overall impact on biodiversity. Creating a model for studying ants—a complex process that requires iterations, course correction, inclusion of newly discovered interactions and so on—would also lay the groundwork for models to study other creatures.
The case for the experiment was very convincing.
In 2010, the large team of scientists from Harvard University, the University of Vermont, North Carolina State University and the University of Tennessee chose two sites—one in the warm, southern Duke Forest in North Carolina, and a cool, northern site, Harvard Forest in New England—for the experiment. The former was near the centre, and the latter at the northern limits of the deciduous forest landscape of eastern North America. It was a difference of 6.5° of latitude and 5.8°C of temperature.
These are forests of broad-leafed trees, dense yet spaced so as to let air and light down to the forest floor. In winter the trees are bare—in Duke the forest floor is carpeted in leaves, in Harvard a layer of snow remains on the ground for months. Spring brings new growth, but summer is the time when the forest really comes to life.
At the sites, 15 open-topped chambers were created. They were five metres across, and open at the bottom to allow the free movement of ants and other creatures. For ants, with a typical body length of 0.5cm, the chamber was the equivalent of 3.1 square kilometres for the human body. The area was large enough for most ants and ant colonies to spend the bulk of their lives within it.
At the start of the experiment four square wooden nest boxes were placed in each chamber. Their transparent covers made them ideal to observe the composition of ant colonies and the interactions between species. Midway through the experiment four more nest boxes were added.
A complex heating infrastructure was established in the forests to heat each of these chambers through the year, for five years running. The system generated hot air at controlled temperatures, which was then pumped gently into the chambers, warming the soil and resident ant colonies. “We literally put heaters around the forest floor and warmed the ant communities up to see what would happen,” says scientist Shannon Pelini.
Nine chambers at each site were warmed in increments of 0.5°C, from 1.5°C to 5.5°C above the ambient temperature of the surrounding forests. This range spanned the most likely projections of global warming over this century. For purposes of comparison, three chambers had air but no heat; the three remaining chambers had neither air nor heat.
An array of instruments kept real-time records of parameters like temperature and humidity at the sites. A thorough monthly census was conducted of all the chambers—in which details like the numbers of each species of ant and the composition of nest boxes were painstakingly collated. Apart from this, individual experiments exploring facets such as seed dispersal by ants were happening simultaneously, with their own cycle and frequency of observation.
During the six-year study more than 60 species of ground-foraging ants were collected in the chambers. Of these, 30 occurred at both sites. Crematogaster lineolata, a brownish-black ant, was the most abundant at Duke. This species, which is aggressive and has an annoying bite, usually lives in tree stumps. At Harvard Forest, Aphaenogaster rudis, a relatively timid, scavenging species that lives on dead bits of flies and other insects, was the most common. The southern site had, as expected, a far greater abundance of ants—both in numbers and species.
One of the first things the scientists noticed as the ants warmed up was a change in the way different species interacted. It has typically been assumed that how well or badly a species does when faced with global warming would be determined by the species’ thermal tolerance, or ability to cope with heat. However, Sarah Diamond of Case Western Reserve University and her colleagues noticed that, as temperatures increased, species that were less heat-tolerant were also getting more affected (than they were before temperatures rose) by competition from other species.
The physiological effect of warming on individual species was amplified by their greater susceptibility to hostile interactions with other ant species. And contrary to conventional wisdom, which predicts that the effects of climate change are going to be more pronounced in lower latitudes, the effects of negative species interaction seems more pronounced at Harvard Forest.
This altered species interaction, the scientists realised, was an important aspect (and determinant) of how species respond to climate change. ‘Winner’ and ‘loser’ species might be winning and losing not because of their individual ability to adjust, but because their mortal enemies were adjusting better. Perhaps understanding how the ant community as a whole reacted to warming might be more instructive than looking primarily at individual species reactions.
Individually, at Duke, Crematogaster lineolata seemed to do better with warming. The probability of finding it in nest boxes increased, and the chances of it disappearing from the chambers decreased. The opposite happened with Aphaenogaster rudis, an important seed disperser that tends to get bullied by the former. A closer look at interactions indicated that subspecies of Crematogaster were coping better than others, riding roughshod over species like Biondia chinensis.
At Harvard, Temnothorax longispinosus and species of the Camponotus group seemed to do better to the detriment of species of the Myrmica group. Here again, there seemed to be an increase in hostility between species, harming species that were already suffering more due to the increase in temperature.
But things became fascinating when Diamond looked at the effect on the community as a whole. On the face of it, ant populations at both Harvard and Duke seem to have actually increased. In sheer numbers there were more, and nest box occupancy had also increased. But this apparent growth had been driven by just one or two ant species. Many other ant species had actually declined. There had been a consistent shift in the composition of the community towards heat-loving species.
In the normal course, there is a fairly rapid turnover of ant colonies within any particular area. The game of survival entails a constant genetic churn through winning, losing, displacement and occupation. No single species would occupy a nest box for long. This actually keeps the ant community as a whole robust, prepared to handle exigencies and disasters. However, with increasing temperatures, as a few species tend to start dominating the landscape and stay longer in the nest boxes, the community as a whole becomes more susceptible to shocks. This reduced dynamism led to an overall reduction in stability, and resilience—the ability of a community to return to equilibrium after it’s been disturbed.
It seemed that the ant community was witnessing a gradual destabilisation. This was not the ‘regime change’, or the move of a community to an alternative stable state, that some scientists had predicted would occur with global warming. It was also distant from apocalyptic ‘species collapse’ theories which predict that entire species could collapse almost overnight.
In this case, the consequences were somewhere in between—it was almost as if the social and economic fabric of the ant community was coming apart; and in that it was uncannily similar to what happens to human communities when faced with strife.
Contrary to expectations, the destabilisation seemed more pronounced at Duke, the lower latitude site, than at Harvard. This has implications for lower latitude countries like India.
These were very significant findings that indicated that scientists needed a new, more holistic way of assessing the impact of global warming and climate change.
“An important next step,” Diamond and her colleagues concluded in a paper, “is to link the reduction in stability with the demographic and fitness consequences for individual species and biodiversity at the community level.”
They started looking more closely at a single species, Temnothorax curvispinosus, a tiny ant that lives primarily in acorns on the ground. This is an extremely well-studied, and fairly heat-tolerant ant, but its reaction to the rising temperatures in the chambers puzzled the scientists. Even though the rise in temperatures was well within its tolerance limits, its numbers seemed to be declining.
The ant could tolerate temperatures up to 43.5°C, and its survival (numbers) only seemed to be impacted dramatically when temperatures rose more than 4°C above ambient temperatures. In most chambers it should not have been impacted. Yet its numbers had decreased.
On closer look, an interesting fact cropped up: the ants seemed to reproduce well only when the increase in temperature rise was low, about 2°C above ambient. There was an incongruity between the ant’s ability to survive and its ability to reproduce with rising temperatures. Though it could survive higher temperatures well, its ability to reproduce was impacted.
That explained the falling numbers. The reduction in stability of this particular species (and communities where it has a significant presence) was not linked not to its demographic decline (in the short term) but to the reduction in its ability to reproduce, or its fitness. Put differently, the vulnerability of this species to warming was dictated not by its ability to survive higher temperatures but by its diminishing ability to reproduce in them. A population could stop growing, the scientists realised, much before it even came close to the limits of its survival.
If scientists were to understand the effects of warming on community stability, and be able to predict it, they’d have to expand their definition of fitness (or success) to include factors like reproduction. Else, they risked missing or mis-estimating the potential impact of
These are big discoveries. If their impact doesn’t seem that obvious at first, imagine substituting large mammals like tigers and deer for ant species. This work on ants could well be the blueprint for understanding how those mammals will respond to warming.
The warming experiments were finally shut down in 2016, six years after they’d started. This wasn’t because the scientists were done with their work—many questions remain to be answered. But Aaron M Ellison, Deputy Director, Harvard Forests, Harvard University, says that the chambers “had been repetitively sampled enough times that the effects of sampling were beginning to obscure the effects of the experiment. So effectively they (the warming experiments) had run their course.”
The vast volumes of data generated over those years are, however, still being analysed, with many more findings in the pipeline. According to Katharine L Stuble of the University of Oklahoma, this data spans “everything from microbial activity to insect richness to herbivory (foraging)”. By looking at it more holistically she hopes to get a better picture of the “cascading impact of warming across ecosystems” and also explore whether it is possible to predict where warming will have the largest impact. “Will the impacts of warming on populations of individual species scale up and be magnified in their impacts on ecosystem processes, or will these changes mitigate one another, causing less changes in ecosystem processes than may be expected?”
At the University of Vermont, Nicholas J Gotelli and Andrew Nguyen are looking at analysing the evolution of heat shock proteins, proteins produced by cells in response to heat stress, in different species of ants in the chambers. This will reveal whether the ants were able to adapt to warming, and if they did, to what extent. It will also help delineate how much of this adaptation was genetic and how much behavioural. The presence of any significant adaptation to warming is likely to open up many new lines of questioning, and the faint hope that communities like those of ants might be more resilient to warming than we’d imagined.
Scientists who worked on this original warming experiment have also initiated spin-off projects. One group is trying to manipulate both temperature and dominant plant presence along a mountainous elevation gradient to study the impact of climate change. Another scientist is looking at differences in ant physiology between rural and urban areas. Stuble is trying to understand how climatic conditions at the time communities are forming shape their long-term trajectories.
This outlandish ant-warming experiment has given scientists the first real answers about the impact of climate change and global warming. It helped that when confronted with these biggest, most alarming questions, they started with the smallest of creatures.
This essay was published in the April-June 2019 issue. The theme of the issue was ‘Heat’.