The International Society for Bayesian Analysis (ISBA) recently held the 2024 ISBA World Meeting in Venice. STOR-i PhD student, Thomas Newman, received the award for best poster at the conference.

In discussing his poster, titled ‘Probabilistic Inversion Modelling of Gas Emissions: A Gradient-Based MCMC Estimation of PDE Parameters’, Thomas stated: "My poster focuses on accurately estimating the location and rate of gas emissions by inverting ground sensor concentration measurements. In practice, Gaussian plume models are often used to describe gas dispersion. However, these commonly rely on atmospheric stability class-based dispersion parameters which can lead to significant bias in our parameter estimation.

"In this poster, I present a robust approach that jointly estimates dispersion parameters, source characteristics, and sensor errors, while also quantifying uncertainty using gradient-based MCMC methods. I demonstrate this through both a simulation study and an analysis of the Chilbolton dataset, showcasing the benefits of improved parameter estimation in practice."

Remarking on the prize, Thomas added, "I’m honoured to have received the Best Poster Award at the 2024 ISBA World Meeting. It’s incredibly rewarding to have my work recognised by peers in the field. This acknowledgment motivates me to continue pushing forward with my research, and I’m grateful for the support and feedback from the community. Thank you to the organisers, judges, and fellow researchers for the inspiring environment and support!"

Details of the conference can be found here: https://www.unive.it/web/en/2208/home.