TraceSeis integrates well-log and seismic data to deliver quantitative estimates of reservoir and geomechanical properties — reducing risk in exploration and development decisions through rock physics, seismic inversion, and proprietary software.
Every engagement verifies two things before any inversion is run: that the rocks are sensitive to the property you care about, and that the seismic data faithfully represents the earth's reflectivity. Only then do we quantify.
A rock physics study with SeisRP® evaluates whether changes in porosity, lithology, or fluids would actually be observable in your seismic data — before you spend on inversion.
Pre-stack data is conditioned and verified against synthetics computed in the rock physics study, ensuring the seismic response represents true earth reflectivity.
Optimum rock properties are computed through seismic inversion, verified at well locations, and transformed into reservoir and geomechanical properties you can act on.
Determine — before committing budget — whether rock sensitivity and seismic data quality support quantitative estimation of the reservoir property of interest. Sensitivity analysis, data-quality evaluation, and synthetic-to-real comparison.
A fast, efficient machine-learning methodology for predicting lithofacies from seismic and well data — published in SEG's The Leading Edge and proven in the Niobrara Formation. Data-driven, without reliance on low-resolution physical model fitting.
Inversion to reservoir and geomechanical properties — including TraceSeis' Relative Rock Properties approach, which avoids dependence on Low Frequency Models (LFMs) to deliver more reliable property estimates.
A comprehensive workflow for analyzing well-log data and computing reservoir property relationships: from LAS loading through fluid substitution, AVO modeling, well-log inversion, and full 3D property cubes.
Each component addresses one stage of the risk-reduction workflow — together they take you from raw logs to defensible reservoir property estimates.
POLISH™ is TraceSeis' wireline data preprocessing platform — transforming raw LAS files into pristine, analysis-ready datasets before they ever touch your rock physics workflow. Garbage in, garbage out; POLISH exists so it's never garbage in.
POLISH™ is currently in active development. Capabilities described reflect the current build.
Overall accuracy hides domain failure. A model can score 84% and still fail an entire customer segment. CONFIRM grades model consistency across domains using pure statistics — chi-square and Cramér's V — methods that predate machine learning by decades and can be reproduced by any auditor with a pencil and a contingency table.
Built on the same discipline TraceSeis has applied to seismic data for 20+ years: never trust an estimate you haven't independently verified.
40+ years in oil & gas exploration and field development, for both operators and service companies — borehole seismic, 2D/3D processing, AVO, AVAZ, seismic inversion, and rock physics. Designs workflows to meet specific geophysical objectives, and writes the software when it doesn't exist commercially.
Graduate studies in Geology & Geophysics, University of Houston · B.S., Montana Tech.
Started as field QC for vibroseis crews; rose to Area Geophysicist for GSI's Mexico office.
VSP, check-shot, salt proximity, and synthetic seismogram expertise applied at scale.
Taught data-processing courses across Mexico and South America; became technically responsible for Houston and Mexico processing operations.
Built and managed production workflows reducing risk in petrophysical property estimation through rock physics modeling, AVO, and seismic inversion.
Geoscience solutions through services, consulting, and proprietary software — including the SeisTool® platform.
Created methods for characterizing fractures, brittleness, and qualitative TOC — pre-stack inversion, AVAZ, high-resolution velocities, converted-wave (PS) data.
Including the Relative Rock Properties methodology and machine-learning lithofacies prediction (Niobrara Formation case study, The Leading Edge).
Chaveste, A., Roden, R., and Smith, T. — A data-driven methodology for predicting lithofacies from seismic and well data without reliance on low-dimensionality physical model fitting. DOI: 10.1190/tle44090728.1
An approach to reservoir property estimation that removes dependence on Low Frequency Models, enhancing reliability where LFM construction is poorly constrained.
Shale analysis, producibility prediction, relative flow-rate estimates for water/oil/gas, petrophysical data management, consistent net-pay tabulation, and synthetic curve generation (RHOB, DT).
Prospect generation, independent evaluation of prospects and assets, business development, commercialization and investor identification, multiclient projects, and training.
Integrated enterprise seismic imaging software and processing services — Kirchhoff PSTM/PSDM, tomography, beam migration, RTM, 5-D interpolation, and diffraction imaging.
Whether it's a feasibility study, a full inversion workflow, a SeisTool® license, or an independent CONFIRM™ validation — start with a conversation.
Get in Touch info@traceseis.comHouston, Texas