Dr Trish Nowak

I am an EPSRC Postdoctoral Fellow in the Department of Computer Science, specialising in artificial intelligence for atmospheric composition and environmental forecasting. My work sits at the interface of machine learning, atmospheric science and environmental intelligence, with a particular focus on improving the skill, transparency and reproducibility of aerosol predictions. My current fellowship, Sub-seasonal Dust Intelligence: Reproducible Monthly AOD Predictions, develops new AI architectures tailored for sub-seasonal forecasting of mineral dust. The project advances methodological foundations in convolutional models and uncertainty quantification, while addressing pressing environmental and societal needs related to dust hazards, climate interactions and early-warning systems.

In addition to fundamental research, I am committed to open science, reproducible workflows and collaborative development with operational and scientific partners. My work contributes to ongoing efforts to bridge data-driven methods with process-based understanding, supporting more accessible, robust and environmentally responsible prediction tools.