A topic overview exploring how machine learning can help screen seismic attributes for porosity and support early-stage carbon-storage assessment.
Carbon storage assessment depends on understanding the geometry, properties and uncertainty of the subsurface. Machine-learning models can help teams interrogate large seismic datasets, identify relationships between attributes and prioritise areas for deeper geoscientific review.
GeonatIQ applies these methods as decision support. Models are developed around the available seismic, well and contextual data, then validated against the physical setting and the decision the work must inform.
The objective is faster, more consistent screening while preserving expert judgement at every material stage.