SURF: Subsurface Understanding for Robust Emissions Forecasting

Understanding the subsurface is essential for reducing energy consumption, lowering CO₂ emissions, and improving recovery of resources during the planned transition to net-zero society. Developing better subsurface understanding will enhance efficiency and result in reduced emissions from oil and gas production. Computer models of a reservoir allow one to perform numerical experiments to investigate the effects of different choices for the development of the field before making decisions. This project aims to develop smarter methods for adjusting large reservoir models so they better match measured data—a process known as history matching. These models help us predict how underground reservoirs will behave in the future under different operating conditions, but the process of modeling is complex and there are many sources of uncertainty, especially when multiple geological scenarios are involved. To address this, the project introduces new techniques that will make the adjustment process more efficient and better suited to the types of uncertainty found in real field data. Not all models are equally good at predicting future behavior, so the project also focuses on identifying which models provide the most reliable forecasts for decision-making. The methods will be tested on full field-scale models to evaluate their performance.