SANTA FE – A team of national defense scientists at Los Alamos National Laboratory that studies contagions with award-winning accuracy has developed its own U.S. forecast for the spread of the coronavirus.
It’s one model that states are using as they consider the duration of directives on social distancing and restrictions on business and President Donald Trump steps up pressure to follow his road map to gradually reopening a crippled economy.
With support from the U.S. Energy Department, the Los Alamos model builds upon a decade of past experience in forecasting contagions, including the seasonal flu, the Ebola virus and mosquito-borne Chikungunya.
Last year, Los Alamos statisticians beat out more than 20 teams in a CDC competition aimed at improving flu forecasting using supercomputing power. The lab’s “Dante” model was most successful in predicting the peak and short-term intensity of the unfolding flu season – and became the basis for the new COVID-19 model.
That model shows the likelihood that a state has hit its daily infections peak and may be on a downward slope. It also offers hurricane-style probability forecasts for infections and deaths in each state in the coming week, as well as a longer six-week period.
For the state of New York, the most recent forecast late last week showed a 60% chance that infections already have peaked and may be waning.
Behind the results is a model that runs on a supercomputer at the national laboratory but avoids complex, “what if” assumptions, said Sara Del Valle, a deputy group leader of the Information Systems and Modeling Group that creates the forecast.
Instead, it builds primarily upon infection and fatalities data from John Hopkins University. Without delving into state-by-state social distancing measures, the model assumes in general that the effects of implemented interventions will continue into the future.
As leaders worldwide try to get a handle on the coronavirus outbreak, they are turning to numerous mathematical models to help them figure out what might happen next. Some models extrapolate the effects of existing social distancing for a certain time period, or play out a variety of assumptions about the effectiveness of interventions against the virus.
A top New Mexico state health officials says the forecast from Los Alamos is being combined with details about local populations and public health into a unique state model for New Mexico, where an aggressive outbreak in the northwestern corner of the state has disproportionately affected the Navajo Nation and at least two Native American pueblo communities.
Amid aggressive social-distancing and stay-at-home orders, the virus is at a near-standstill in counties such as Santa Fe, a hub for state government, and Los Alamos itself, where the nation’s first atomic bombs were designed.
For every infection confirmed by current testing methods, New Mexico state and private health officials assume about five people are infected, said David Scrase, secretary of the state Human Services Department.
“Los Alamos National Laboratory and our model are really converging closely when it comes to timing of the peak,” said Scrase, also a practicing gerontologist.
It is unclear how many states and local jurisdictions are watching the Los Alamos model, though they include Nevada’s second largest county as it safeguards the Lake Tahoe shoreline, the city of Reno and an extensive rural area.
The model offers straightforward probabilities – the worst, best and most likely forecasts for infections and deaths.
The team behind the Los Alamos model coalesced in 2011 with funding for infectious disease modeling from the National Institutes of Health, growing as doctorate students worked on projects and then joined the staff.
Current projects on Dengue fever in Brazil combines broad sources information from medical clinics to social media posts and satellite imagery.
Del Valle said the team monitored the accuracy of the stability and accuracy of the COVID-19 model for weeks before the public release April 5. Updates take place Mondays and Thursdays.