What tied together COVID-19 outbreaks in the Chinese city of Wuhan, in northern Italy, in Seattle and elsewhere—and could this knowledge help prepare for, or even head off, future spread of the disease?

University of Maryland researchers are zeroing in on temperature, humidity and other factors that apparently contributed, as part of a fast-tracked broader study designed to give scientists and public health officials a prediction tool for the novel coronavirus pandemic. As of today, the virus has killed more than 10,000 people, infected hundreds of thousands more and is radically disrupting the lives of billions worldwide.

“What we’ve understood so far is that there is a kind of sweet spot for temperatures and a sweet spot for humidity where the virus is likelier to be contagious,” said Fernando Miralles-Wilhelm, professor of atmospheric and oceanic science and interim director of the Earth System Science Interdisciplinary Center (ESSIC).

In a pre-publication paper shared on SSRN, Miralles-Wilhelm and co-authors wrote that the virus spread most rapidly in temperatures between 5 and 11 degrees Celsius (41 to 52 Fahrenheit) and in humidity levels of 50 to 80%. The paper, which has not yet been peer-reviewed for journal publication, also showed that COVID-19 established community spread along an east-west corridor in northern hemisphere latitudes between 30 and 50 degrees, which includes much of the United States, Western Europe and northern China.

Thus, while previous coronaviruses that caused outbreaks, including SARS and MERS, are not believed to be seasonal, the “distribution of significant community outbreaks along restricted latitude, temperature, and humidity are consistent with the behavior of a seasonal respiratory virus,” according to the paper.

Miralles-Wilhelm’s coauthors include Dr. Mohammad Sajadi and Dr. Anthony Amoroso, both of the Institute of Human Virology at the University of Maryland (UMB) School of Medicine in Baltimore, ESSIC Assistant Research Scientist Augustin Vintzileos and researchers working in Iran.

The next step, Miralles-Wilhelm said, is to connect environmental variables to data about human health and how populations live and interact socially, and submit the full paper for publication in coming weeks.

“We are now working fast and furiously to develop a more accurate predictive model that incorporates factors that are more on the human side, like proximity and population density”—a tool that once available, he said, could give national and local health authorities a vital heads-up on heading off the further spread of the virus.