In mid-March, Nancy Reid was settling into a sabbatical in Australia when Canadian Prime Minister Justin Trudeau called on all Canadians to come home. Within days, Reid was on a flight back to Toronto, cutting short a three-month stint meant to deepen collaborations with mathematicians in Australia. “We left with heavy hearts,” says Reid, a statistician at the University of Toronto.

Reid and nine other researchers, all visiting Australia through a programme funded by the Sydney Mathematical Research Institute at the University of Sydney, flew home as borders began closing. Of the 50 planned visits by researchers to the institute, 30 have been postponed or cancelled. Reid says that opportunities for her to take sabbaticals come around only once every half a dozen years. “It could be a while before any of us take short hops to faraway places.”

The pandemic has forced researchers to grapple with how restrictions on their movement will affect the way science is conducted. Some fear that these limitations could lead to a world that favours elite scientists and established teams with less need to network, while others lose collaborators and become marginalized, and fieldwork becomes a luxury. Or, the opposite could turn out to be true: travel restrictions could help to democratize science, if virtual working allows access to more connections for people who cannot easily travel. Many researchers will have to reimagine their projects in a world where travel is limited.

Elite club

The pandemic might already be closing off some collaborations. Caroline Wagner, a science and policy researcher at the Ohio State University in Columbus, has compared coronavirus-related publications in the two years before December 2019 with peer-reviewed papers and preprints on the same topic posted between January and April 2020.

Wagner found that, since the outbreak, research teams — measured by the number of authors on a paper — have become slightly smaller and involve fewer nations. Co-authorship links between China and other scientifically dominant countries have strengthened, whereas participation from developing countries has dropped off, which Wagner says suggests that researchers are working with those with whom they already have strong, established connections. She posted her team’s results as a preprint on the social-sciences repository SSRN on 8 May (C. V. Fry et al. preprint at http://doi.org/10.2139/ssrn.3595455; 2020).

Universities will never be the same after the coronavirus crisis

The shift in favour of smaller networks is probably a by-product of the need for speed. International collaborations can enrich research and enhance creativity — but they take time to establish, says Wagner. “If you’re working on a crisis timescale, then the transaction costs of adding new, additional people is not going to be worth the price in time.”

If existing collaborations with developing-country scientists are faltering, then new ones could slow to a trickle, even in areas outside coronavirus science, says Wagner. The majority of relationships start with an in-person meeting, she says. “If people can’t work side by side for short periods of time, it’s going to dramatically cut international collaboration.” Early-career researchers who haven’t had the opportunity to make those interpersonal connections will be hit hard, says Richard Woolley, a sociologist at Ingenio (CSIC-UPV), Polytechnic University of Valencia, Spain.

Geographically remote countries such as Australia are also exposed because of their highly international scientific workforce, says Keith Jones, a reproductive biologist at the University of Adelaide. Data from the Australian government show that around 35% of Australia’s more than 65,000 PhD and master’s students are from overseas; a report by the Australian Academy of Science published on 8 May (see go.nature.com/2u2dznj) estimates that some 9,000 international students will not resume their research in the country this year owing to financial constraints or travel restrictions resulting from the pandemic.

Such shifts can have long-term implications, says Cassidy Sugimoto, an information scientist at Indiana University Bloomington. “As soon as you start down a research path and get into a pattern of collaborations, it takes a long time to turn that ship around.”

Subject switch

At least in the short term, some researchers will have to change the questions they ask and the projects they work on in a movement-restricted world. As a programme director for the US National Science Foundation, Sugimoto says she has to consider how the research can be done without travelling. “What happens if anthropologists can no longer go to their site? Do we still fund the research?”

Sebastien Kenmoe, a virologist at the Pasteur Centre in Cameroon in Yaoundé, leads a study on febrile illness in sub-Saharan Africa. He had to cancel a trip to the Central African Republic, and is instead developing ways to monitor outbreaks, including those of coronavirus, remotely.

Sputnik moment or budget breaker: How will the pandemic alter research funding?