Fast-track reservoir characterisation of a subtle Paleocene deep-marine turbidite field using a rock physics and seismic modelling-led workflow. 

First Break volume 25, May 2007

Seismic attributes relate to physical phenomena, and can be modelled by theoretical rock physics methods. By calculating and quantifying lithological and fluid-fill variability, it is possible to take the well-based model forward into the seismic world of AVA (Amplitude Variation with Angle) and inversion. By using rock physics to bridge the gap between petrophysics and seismic, we can use a supervised neural network technique to quantitatively predict reservoir properties (such as porosity, shale volume, and saturation) away from the well. Without the use of rock physics, seismic is incapable of measuring any of these parameters directly. Improvements being made in processing to enhance the signal-to-noise ratio will improve our confidence that subtle changes observed in the seismic signal reflect changes in the subsurface environment.