Forecasting the Impacts of Weather and Social Distancing on COVID-19 Transmission Across the U.S.

Statement of Problem

As the nation continues to respond to the once-in-a-century challenge of the COVID-19 pandemic, there is an urgent need to project how the virus will spread in the coming months. This information is critical to help local, state, and federal leadership make informed decisions about public policy solutions to manage the crisis.

In an initial response to the pandemic, society largely shut down through a variety of social distancing interventions that varied in magnitude and timing, and were designed to help slow the rate of infection, or flatten the curve. Flattening the curve was essential to prevent overwhelming health care systems, particularly with increased demands for ICU beds and ventilators, as well as to improve access to rapid and serologic testing for public health surveillance. Now, as communities have reopened in various stages across the country, we are facing new challenges with schools and colleges attempting in-person learning, a converging flu season, and pandemic fatigue among individuals and institutions for following the proven safety protocols such as masking and distancing that can reduce risk for widespread community transmission.

As the pandemic spreads in varied ways among our cities, it will be critical to provide more robust and reliable data models to guide decision-making regarding the opening and closing of schools, colleagues and workplaces to ensure the safety of all Americans. This will require more precise interventions to protect communities while attempting to maintain economic activity as we navigate the months ahead.


Forecasting the Impacts of Weather and Social Distancing on COVID-19 Transmission Across the U.S.

Our interdisciplinary team across Children’s Hospital of Philadelphia and the University of Pennsylvania is using local data to provide short-term forecasts based on current social distancing practices.

Our interdisciplinary team from Children’s Hospital of Philadelphia (CHOP) and the University of Pennsylvania is using actual case data to longitudinally track COVID-19 transmission and test positivity rates across all U.S. counties, and project case counts for 820 counties with active outbreaks, representing 82% of the U.S. population and 83% of all identified coronavirus cases. The 820 counties include those with state capitals, populations of 40,000 individuals or more (with a minimum population density of 250 people per square mile), and sustained outbreaks, which we define further in this abstract.

Through this model, we are accounting for the impact of weather, health and demographics of the population and city characteristics in order to: define the influence of temperature and humidity on SARS-CoV-2 transmission; determine the impact of social distancing in modifying the trajectory of SARS-CoV-2 transmission during current and future outbreaks; investigate how city characteristics modify the transmission; and forecast the effects of policy change and social distancing practices on the risk for virus resurgence.

The data and methodology behind this modeling project was informed by a peer-reviewed study, published in JAMA Network Open, confirming social distancing as the most effective intervention against COVID-19. When observing how time-varying factors, such as weather and social distancing, impacted COVID-19 transmission risk at the county level between Feb. 25 and April 23, 2020, we found that spring-like temperatures contributed to some reduction in the spread of the virus, but without strong social distancing practices, the beneficial impact of temperature was not fully realized.

The latest updates to the model, released on Jan. 13, project New York (Manhattan), Bronx, Kings (Brooklyn), Queens, and Richmond (Staten Island) counties to have another doubling of daily cases over the next four weeks, having already doubled since the holidays. Additionally, the researchers forecast a continued rise in case numbers for population-dense regions throughout a number of southeastern states such as Alabama, North Carolina and Tennessee. The model’s dashboard with county-level case incidence and test positivity rates also accompanies revised guidance for in-person schooling developed with CHOP’s Division of Infectious Diseases to help school districts and communities around the country negotiate the dynamic challenges of this pandemic and plan for the future. To view a data visualization of our findings, click here, to learn how we validate this model to ensure accurate projections, click here, and to learn more about the methods of this model, see this abstract.

Next Steps

Using this real-time data, we are engaging with local, state, and federal policymakers who are managing the COVID-19 response to support their efforts. We aim to inform a safe approach to reopening society that also accounts for the economic and social impacts shelter-in-place strategies are having on low-income families and communities.

For more on PolicyLab's COVID-19 response, click here.

This project page was last updated in January 2021. 

Suggested Citation

Children's Hospital of Philadelphia, PolicyLab. Forecasting the Impacts of Weather and Social Distancing on COVID-19 Transmission Across the U.S. [Online]. Available at: [Accessed: plug in date accessed here].