Research Expertise
I hold a PhD in Statistics from Lancaster University, with expertise in Monte Carlo–based Bayesian inference, including Markov chain Monte Carlo and Sequential Monte Carlo methods. My research also focuses on surrogate modelling for computationally fluid dynamics, using machine and deep learning approaches such as multilayer perceptrons and convolutional neural networks. My work is driven by environmental and atmospheric science applications.
Real-World Implementation
I apply these methods to atmospheric inversion modelling of methane emissions using in-situ observations. My previous work includes real-time estimation of time-varying emission rates and source locations, as well as joint estimation of emission characteristics while correcting for forward-model misspecification. This bridges rigorous statistical theory with operational environmental monitoring.
Current Responsibilities
I currently develop machine learning and deep learning systems to detect and quantify global methane emissions from TROPOMI satellite data. My role involves building, maintaining and deploying automated, near–real-time monitoring pipelines and collaborating with international partners to support rapid identification of emissions and effective mitigation efforts.
