Collaborative Research: Understanding Stochastic Spatiotemporal Dynamics of Epidemic Spread to Improve Control Interventions - From COVID-19 to Future Pandemics

This grant will support research that will contribute new scientific knowledge related to how uncertainties in both human behavior and transmission characteristics of a causative pathogen (such as the novel coronavirus in the case of COVID-19) influence the spread of an epidemic, and how the new knowledge thus obtained about epidemic spread can contribute to interventional public health policy measures to effectively mitigate and control an epidemic. The research will advance both the science of predicting epidemic spread as well as national prosperity by enhancing national preparedness for early and effective mitigation of potential future epidemic outbreaks. Mathematical and computational models that can accurately predict an epidemic spread across geographical regions over specified periods of time are critical precursors to developing effective interventions for mitigation such as social-distancing measures and vaccination campaigns (when vaccines become available). However, the limitations of existing predictive models, as evident during the COVID-19 outbreak in the US, underscore the need for new knowledge in this area. This award supports fundamental research to develop novel predictive models of epidemic spread and also to validate model predictions against the extensive COVID-19 spread data only now available. This research involves multiple disciplines including the mathematical theory of partial differential equations, stochastic analysis, control theory, and epidemiology and the results will likely have broader significance in the study of rare-event dynamics in areas such as ecology, climate science and wildfire propagation. Moreover, this cross-disciplinary project, a collaborative effort involving multiple institutions, will broaden the participation of underrepresented groups in research and training, and also advance science and engineering education.

The research will advance the fundamental knowledge of how uncertainties, both in human behavior and pathogen characteristics, influence spatiotemporal, stochastic epidemic dynamics and also yield a control-theoretic framework to analyze interventions for mitigation. Specifically, the project will: (1) develop novel predictive dynamic models based on partial differential equations, (2) uncover effects of the interaction between nonlinearity and uncertainty such as noise-induced bifurcations, (3) study infection spikes using a stochastic approach, (4) validate the models using COVID-19 data, (5) establish a control-theoretic framework to analyze mitigative interventions, using a combination of averaging methods from stochastic analysis and feedback control theory, (6) obtain improved characterization of epidemiologic parameters such as basic and effective reproduction numbers, and (7) identify principles and strategies that can inform interventional public health policy.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.