38CVC Applying weather and climate modeling techniques to human health outcomes: data-informed simulation, inference and forecasting

Climate and weather research has led to many advances in computing and modeling: from the first numerical weather predictions, to data assimilation techniques used to support reanalysis and forecasting, to recent developments in physics-informed machine learning and artificial intelligence. In the last two decades, similar and analogous methods have been leveraged to support better simulation, analytics and forecasting for contagious systems affecting human health. Unlike climate and weather, health systems are dynamically driven by biological and social processes that change in time and space. Such dynamics require a different sort of inference, one that estimates changing system parameters, not just state variables. Here I will give an overview of some of this work, including the development of infectious disease forecasting, data-informed counterfactual simulation, inference of the critical epidemiological properties of pathogens, such as the SARS-CoV-2 virus, and estimation of other types of social contagion.