The Lithofluids module provides access to Predict, seismic inversion, and Stochastic Rock Physics Model.

Predict is an interface designed for prediction, calibration and QC of wells and models. Tools include spectral filters, integrated synthetics and displaying/calibrating inverted P impedance, S Impedance and Density volumes. At well locations, multiple low-frequency volumes can be loaded, compared and evaluated for bias prior to use in inversion. The module includes a Bayesian supervised classification engine for making quantitative predictions using inverted seismic data and depth-dependent, statistical rock physics models. After calibrating at well locations, generate individual probability volumes for all candidate lithology and fluid combinations. Using Predict, you can interactively calibrate and test classification parameters using wireline and upscaled logs and see predictions based on your inversion products update on the fly. For more information, refer to Using the Predict Window. 

Constrained Simultaneous Inversion (Absolute) and Constrained Simultaneous Inversion (Relative) processes are also included. Absolute inversion solves for absolute P and S impedance, as well as density and uses a low-frequency model as input, while running a relative inversion creates relative P and S impedance outputs, as well as density, gradient and curvature impedances, and impedance for each angle stack.

The module also includes the Stochastic Rock Physics Model. Using input end-member depth-dependent trends, different lithology and fluid combinations can be created and stochastically forward-modelled. The ability to discriminate between different combinations and the variation in response as a function of depth and other variables, including reservoir quality and fluid saturation, can be quickly assessed using this software. The Stochastic Rock Physics Model uses Bayesian updating to make quantitative predictions based on stochastic rock physics models, hence providing reliable information on fluid discrimination.