9 September, 2020
Femke Vossepoel, Instructor

Data assimilation, a geoscience data technique that originates from meteorology and oceanography, is widely used in the oil and gas industry for assisted history matching. The method aims to find a model solution to a given dynamic problem that agrees with the observations given the uncertainties in both observations and models. In this webinar, I will give an introduction to the concept of data assimilation and introduce a number of well-known methods. I will illustrate the use of so-called iterative ensemble smoothers to effectively estimate parameters of a model that predicts the evolution of the COVID-19 pandemic in eight countries, including Norway, England, France, Brazil, and a number of states in the United States. The model used, a SEIR model with age-classes and compartments of sick, hospitalized, and dead, is conditioned on daily numbers of accumulated deaths, the number of hospitalized and, where possible, on the number of cases obtained from testing. We start from a wide prior distribution for the model parameters; then, the ensemble conditioning leads to a posterior ensemble of estimated parameters leading to model predictions in close agreement with the observations. The updated ensemble of model simulations have predictive capabilities and include uncertainty estimates. In this webinar, I will discuss how we can estimate the effective reproductive number as a function of time, and how the evolution of the pandemic depends on local differences in response to intervention measures.

More: [ Ссылка ]

свернуть