How would social distancing change the COVID-19 curve of community spread?

The rapid development of the coronavirus (COVID-19) has prompted unprecedented global health adversities. Curbing the virus diffusion and infection by social distancing, such as minimizing non-essential travel activities, has become a consensus. When social distancing measures are in effect, new cases of infection would be primarily induced by community spread, such as human movements within and between neighboring cities, towns, and communities. How these travel behaviors affected the virus spread on the community scale has been less explored.

Xiang Chen, an Assistant Professor in Geography at the University of Connecticut, and his team members develop an epidemic model for simulating COVID-19 epi curves on the community scale based on different social distancing scenarios. Supported by the NSF-supported COVID-19 Geospatial Fellowship program in the Geospatial Software Institute (GSI) at the UIUC and a grant from the Institute for Collaboration of Health, Intervention, and Policy (InCHIP), this research is published in the International Journal of Geographical Information Science, the flagship journal of GIScience.

In this research, Chen considers social distancing in two dimensions, containment, which refers to how far people may travel, and compliance, which refers to the percentage of people complying with social distancing rules. Based on these two metrics, the team develops the MSEIR model, an epidemic model that can study and predict COVID-19 epi curves at the community scale, which has local policy implications. Using COVID-19 data published by Connecticut Department of Public Health, they apply the model to simulate cases of infections in the early outbreak as well as the recent case resurgence in every Connecticut town.

A framework of compliance and containment in social distancing

The MSEIR model has considered nine social distancing scenarios, or Models 1 through 9. Model 1 represents the most stringent social distancing restrictions and Model 9, the least. The model successfully predicted the early outbreak in Connecticut, which was within the range of Model 5 and Model 9.

Simulating infection cases of the early outbreak in Connecticut

With the model being calibrated by travel survey, they apply it to forecasting infections in the second-wave outbreak and identify high-risk Connecticut towns, such as Ledyard and Lisbon, where the increase in cases would be the steepest.

Predicted infections and case increase in Connecticut towns

The scenario-based analysis can help policy stakeholders to understand how the compliance with and the containment by social distancing rules regulate people’s travel activities and can help predict how different degrees of policy enforcement would shape the epic curves over different phases of policy implementation. These modeling results have the potential to assist stakeholders with strategic decisions about the timing and expected outcomes of relieving social distancing rules in the upcoming vaccination phases.

Chen’s collaborators included Aiyin Zhang from Clark University, Hui Wang from the University of Idaho, and Xiaolin Zhu from the Hong Kong Polytechnic University.



Xiang Chen

I am an Assistant Professor in the Department of Geography at the University of Connecticut, USA. I received a Ph.D. in Geography at The Ohio State University. My research interests are focused on GIScience, community health, and urban studies. I employ GIS instruments (e.g., big data analytics, geovisualization, deep learning) and accessibility theory to unveil the socioeconomic and health inequalities across urban communities. I also have research interests in transportation and remote sensing. During the COVID-19 pandemic, I am heavily involved in various public health research, including mapping, modeling, and projecting cases of infection, discussing socioeconomic impacts on tourism, and exploring the health inequity in food procurement.


Name and email are required. Your email will not be published.

Please provide a username.
Please provide a valid email
Please input your message.