1. FieldViewer
FieldViewer - Digital Field Review and Decision-Support Platform
Open Intro
FieldViewer is a web-based field visualization concept designed to support integrated field review and decision support. It combines maps, wells, production data, timelines, completions, well testing, ESP review, and technical outputs into a reusable digital environment.
The goal is to provide a more interactive and reusable alternative to static PowerPoint-based field reviews and disconnected technical datasets.
- Problem addressed
- Disconnected field review material spread across maps, spreadsheets, presentations, and technical files.
- Workflow
- Reusable browser modules for maps, well status, timelines, production, completions, well testing, ESP review, and AI-assisted selected-well context.
- Technologies
- Python, Bokeh, Flask, GIS/CRS workflows, web maps, structured field data, SQLite concepts, AI-assisted interaction concepts.
- Why it matters
- Improves technical communication and field review continuity by keeping related subsurface and production context in one interactive environment.
2. Applied ML
Seismic Attribute ML - Reservoir Property Prediction and Interpretation QC
Python-based workflows for seismic-attribute-driven reservoir property prediction using well-calibrated data, feature testing, blind-well validation, model QC, and geological consistency checks.
- Problem addressed
- Reservoir property prediction requires both statistical testing and geological consistency, especially when well control is limited.
- Workflow
- Feature screening, model training, blind-well validation, prediction review, and interpretation QC using seismic attributes and well data.
- Technologies
- Python, pandas, NumPy, scikit-learn, XGBoost concepts, Jupyter, seismic attributes, well data, reservoir characterization.
- Why it matters
- Supports more disciplined use of machine learning in reservoir characterization by tying prediction outputs back to geoscience checks.
3. Interpretation automation
OpendTect + Python - ML-Assisted Seismic Interpretation Automation
Developing workflows using open-source seismic interpretation packages such as OpendTect and Python-based geoscience libraries to enhance automation of structural and stratigraphic seismic interpretation through machine learning, geometric analysis, and geophysical attribute workflows.
- Problem addressed
- Seismic interpretation workflows can include repetitive manual screening, attribute review, and QC steps that are candidates for automation support.
- Workflow
- Prototype fault detection, horizon analysis, seismic facies, geometric attributes, geophysical attributes, and interpretation QC workflows.
- Technologies
- OpendTect, Python, seismic attributes, geometric analysis, geophysical attribute workflows, machine learning concepts.
- Why it matters
- Connects interpretation experience with open-source tools and Python workflows that can reduce repetitive work and improve QC discipline.
4. Reservoir uncertainty
STOIIP Monte Carlo App - Reservoir Uncertainty and ZMAP Grid Integration
Streamlit-based reservoir volumetrics workflow for STOIIP estimation using Monte Carlo simulation, uncertainty handling, ZMAP reservoir grid input, spatial mapping, and 3D visualization concepts.
- Problem addressed
- Volumetric uncertainty needs transparent assumptions, repeatable calculations, and spatial context.
- Workflow
- Input distribution handling, Monte Carlo simulation, ZMAP grid integration, uncertainty summaries, maps, and visualization concepts.
- Technologies
- Python, Streamlit, NumPy, pandas, ZMAP grids, reservoir volumetrics, 3D visualization concepts.
- Why it matters
- Makes uncertainty review more repeatable and easier to communicate across technical and management audiences.
5. Structural QC
Fault Damage and Amplitude Profile Analysis
Python/Jupyter workflow for extracting fault-normal seismic or grid profiles, analyzing amplitude variation around interpreted fault segments, and supporting structural interpretation QC.
- Problem addressed
- Fault-related amplitude and grid behavior can be difficult to compare consistently along interpreted structural trends.
- Workflow
- Extract fault-normal profiles, compare amplitude behavior, review trends, and support interpretation QC.
- Technologies
- Python, Jupyter, NumPy, pandas, seismic/grid profiles, visualization workflows.
- Why it matters
- Turns structural interpretation checks into repeatable analysis rather than one-off visual inspection.
6. Geoscience utilities
ZMAP / GeoTIFF / Geoscience Data Automation Utilities
Utilities and workflows for ZMAP grid handling, GeoTIFF processing, Excel-to-LAS conversion, report building, visualization, and repetitive geoscience data preparation tasks.
- Problem addressed
- Geoscience projects often lose time to repeated data conversion, formatting, and reporting tasks.
- Workflow
- Read, write, convert, visualize, and package technical data for review and communication.
- Technologies
- Python, pandas, NumPy, geospatial data handling, GeoTIFF, ZMAP, Excel, LAS, reporting utilities.
- Why it matters
- Reduces manual preparation work and improves consistency across technical deliverables.