elephants2

UC San Diego researchers studied more than 400 elephants at the Udawalawe National Park in Sri Lanka. Credit: Credit: Udawalawe Elephant Research Project, CC BY-ND

Years of Losses Left Elephant Groups Smaller and More Isolated

In A Nutshell

  • About 69% of the original female elephants tracked in a Sri Lankan park were gone within 12 years, presumed dead or departed for good.
  • Elephant groups grew smaller and more isolated, and the weakest social ties were the most likely to change.
  • Researchers propose that the population crossed a tipping point around 2012, when deaths and departures began to outnumber newcomers.
  • Even long-term residents lost touch with old companions, and the authors warn that delayed effects may hide a decline.

By the end of a 12-year study in Sri Lanka, roughly 69% of the female Asian elephants tracked from the start were gone, either presumed dead or departed for good. Those who remained had not rebuilt the close-knit circles they once relied on, and the groups left behind were smaller and more cut off from one another.

Published in the journal Science, the work followed 443 individually identified females in the 119-square-mile Udawalawe National Park from 2007 to 2018, a stretch long enough to watch relationships shift.

Social animals lean on their communities for food, care of young, and relief from stress, the authors note. Shermin de Silva, an associate professor whose lab produced the study, said in a university release that “for many social species, which includes ourselves as much as elephants, we have to remember that social connectivity is very much a matter of sharing space and building relationships throughout our lives.” Elephants can live 60 years or more, and the authors say that long life in slow-breeding species can mask a collapse in progress.

Asian Elephant Social Bonds Splintered After Losses Outpaced Arrivals

Udawalawe’s elephants, a mix of seasonal and long-term occupants, numbered an estimated 804 to 1,160 in 2007 and 2008. Across 1,665 observation days, researchers logged 7,744 sightings of groups that included at least one identified female. Whenever those elephants were seen traveling together and acting affectionately, a social connection was counted. Each year’s connections became a map of who spent time with whom, similar to a friendship web among people.

A statistical method then sorted each map into communities, clusters of elephants that spent more time with one another than with outsiders. In elephants, these clusters usually, though not always, consist of female relatives.

Two versions of the map were built: one covering all 443 adult and subadult females, and a smaller one covering 59 “residents,” the long-term regulars seen in every year of the study. Comparing them let researchers separate the effect of elephants disappearing from changes in how the remaining elephants related to one another.

Permanent losses rose over the years while arrivals dwindled, and around 2012 deaths and departures began to outnumber newcomers. Afterward, the average group shrank, the number of groups grew, and groups became more isolated from one another. The 59 residents, whose groups stayed stable beforehand, showed the same pattern one to three years later. Time spent observing elephants did not change enough to explain the shift, according to the paper.

elephants
Asian elephants thrive in groups that work together to access food and care for their young. Credit: Udawalawe Elephant Research Project, CC BY-ND

Weaker Social Bonds Were Most Likely to Dissolve

Losing elephants was only part of the story. Even among the residents, who stayed in the population throughout, relationships shifted. Weak ties, the connections between elephants that spent less time together, were far more likely to change than strong ones, though strong bonds became more vulnerable after 2012. Over time, remaining elephants were less likely to reconnect with former companions.

In the authors’ view, weak ties act as built-in backups that cushion a group against disruption, much like a spare route when a main road closes. Losing them could leave a group less able to withstand the next blow.

Elephant society also has layers, with close family units linking into larger, looser circles. Those larger circles broke down after the shift, but the layered structure did not vanish entirely.

elephants
The new findings serve future studies seeking to understand shifts in animal society structures. Credit: Udawalawe Elephant Research Project, CC BY-ND

Asian Elephants Had Not Rebuilt Their Social Bonds by 2018

Other species often bounce back. Earlier studies of savanna elephants in Africa and orcas found survivors working to regain lost connections, sometimes by bonding with unrelated animals. This population showed no such recovery.

Researchers propose that the population crossed a tipping point, a threshold past which a system settles into a new pattern. Among all 443 females, group sizes and connections bounced around before 2012, then appeared to settle into the new arrangement. Lead author Anastasia Madsen of the University of California San Diego said in the same release: “Once we move past the tipping point, it’s much harder to go back to previous states.”

What drove the turnover remains unclear. Possible factors include illegal killings and conflict with people, both known threats to Asian elephants. More than half of Asian elephant populations live on poor-quality habitat after losing nearly two thirds of their probable historic range. At Udawalawe, grasses have been displaced by unpalatable shrubs and invasive species, and the authors say scarce food may limit group sizes and discourage newcomers.

Among the authors’ biggest worries is the delay seen in the residents, since delayed effects may keep conservationists from spotting trouble in time. Deaths from poaching or conflict can be directly observed, but deaths from indirect effects, such as food scarcity, may go unnoticed, especially when a death looks the same as an elephant leaving for good.

Head counts alone can miss a society coming apart, and the authors call for a precautionary approach to protecting social species before the damage becomes obvious.


Paper Notes

Limitations

All data came from one elephant population passing through Udawalawe National Park, and only adult and subadult females were tracked, so the patterns may not hold for males, other elephant populations, or other species. Researchers could not always tell deaths from permanent departures, and the authors say formal population estimates are still urgently needed to confirm that the trends reflect an overall decline, though earlier published estimates already indicated this population was likely shrinking. The tipping-point explanation is a proposal based on observed patterns rather than a directly tested cause. The authors also note that monitoring such shifts in real time is difficult, and that baselines are hard to establish when human pressures on wildlife are now everywhere.

Funding and Disclosures

Funding came from three US Fish and Wildlife Asian Elephant Conservation grants, the International Elephant Foundation, the National Science Foundation, the University of Pennsylvania, the University of California, San Diego, and Trunks & Leaves Inc. The authors declare no competing interests.

Publication Details

Anastasia E. Madsen, T. V. Pushpakumara, U. Sameera Weerathunga, Devaka K. Weerakoon, and Shermin de Silva wrote “Increased demographic turnover and isolation precede social fragmentation in Asian elephants.” Madsen is affiliated with the University of California, San Diego, and the University of Cincinnati. De Silva is affiliated with the University of California, San Diego, EFECT in Colombo, Sri Lanka, and Trunks & Leaves Inc. Madsen and de Silva are the corresponding authors. The paper appeared in the journal Science, volume 394, issue 6825, on November 12, 2026, after acceptance on September 16, 2026. DOI: 10.1126/science.aef6714.

About StudyFinds Analysis

Called "brilliant," "fantastic," and "spot on" by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

John Anderer

Associate Editor