### References & Citations

# Mathematics > Numerical Analysis

# Title: A Bayesian Approach to Estimating Background Flows from a Passive Scalar

(Submitted on 3 Aug 2018 (v1), revised 18 Sep 2018 (this version, v2),

*latest version 10 Jun 2019*(v3))Abstract: We consider the statistical inverse problem of estimating a background flow field (e.g., of air or water) from the partial and noisy observation of a passive scalar (e.g., the concentration of a pollutant). Here the unknown is a vector field that is specified by a large or infinite number of degrees of freedom. Since the inverse problem is ill-posed, i.e., there may be many or no background flows that match a given set of observations, we adopt a Bayesian approach, incorporating prior knowledge of background flows and models of the observation error to develop probabilistic estimates of the fluid flow. In doing so, we leverage frameworks developed in recent years for infinite-dimensional Bayesian inference.

The contributions in this work are threefold. First, we lay out a functional analytic and Bayesian framework for approaching this problem - a common experimental approach to visualizing complex fluid flows. Second, we define an adjoint method for efficient computation of the gradient of the log likelihood, a key ingredient in many numerical methods. Finally, we identify two interesting example problems with very different posterior structures, which we use to conduct a large-scale benchmark of the convergence of several Markov Chain Monte Carlo methods that have been developed in recent years for infinite-dimensional settings. In particular, we demonstrate that two of these algorithms are capable of resolving a posterior with complex, multi-modal structure.

## Submission history

From: Justin Krometis [view email]**[v1]**Fri, 3 Aug 2018 04:41:36 GMT (5334kb,D)

**[v2]**Tue, 18 Sep 2018 15:30:14 GMT (4341kb,D)

**[v3]**Mon, 10 Jun 2019 19:08:49 GMT (4855kb,D)

Link back to: arXiv, form interface, contact.