STRATICA Documentation

Welcome to STRATICA — Stratigraphic Pattern Recognition & Paleoclimatic Temporal Reconstruction.

A Physics-Informed AI Framework for Deep-Time Earth System Reconstruction, Stratigraphic Layer Intelligence, and Paleoclimatic Cycle Decoding via the Temporal Climate Integrity Index (TCI).

Quick Links

• GitHub Repository: https://github.com/gitdeeper8/STRATICA

• GitLab Mirror: https://gitlab.com/gitdeeper8/STRATICA

• Live Dashboard: https://stratica.netlify.app

• PyPI Package: https://pypi.org/project/stratica/

• Research Paper DOI: 10.5281/zenodo.18851076

What is STRATICA?

STRATICA presents the first unified, multi-parameter Physics-Informed AI framework for the systematic reconstruction, computational modeling, and temporal interpretation of Earth’s stratigraphic record across 4.5 billion years.

The framework integrates nine analytically independent stratigraphic and geochemical parameters into a single Temporal Climate Integrity Index (TCI) , achieving paleoclimate classification accuracy of 96.2% across 47 sedimentary basins on 6 continents.

Key Features

✓ Temporal Back-Casting — Transformer-LSTM hybrid architectures reconstruct missing geological records ✓ Nine-Parameter TCI — Integrated composite metric balancing all paleoclimate proxy types ✓ Physics-Informed Neural Network — Hard constraints enforce stratigraphic, thermodynamic, and orbital coherence ✓ Extensive Validation — 47 sedimentary basins, 12 IODP drill cores, 8 ice core records (800,000 years) ✓ Real-Time Dashboard — Interactive paleoclimate exploration and analysis platform

Performance at a Glance

Metric	STRATICA	Previous	Improvement
TCI Classification Accuracy	96.2%	81.4%	+14.8 pp
δ¹⁸O Back-cast RMSD	0.0018 ‰	0.0063 ‰	71% reduction
Milankovitch Cycle Detection	±1,200 yr	±8,500 yr	7x improvement
Magnetostratigraphy Age Accuracy	±3.4%	±11.2%	3.3x improvement
Microfossil Classification (CNN)	93.4%	71.8%	+21.6 pp
Drill Core Processing Speed	4 hrs/200m	6-12 months	500-2000x faster

Contents

Getting Started

• Installation Guide
  • Requirements
  • Install from PyPI
  • Install from Source
  • Install with GPU Support
  • Install with Development Dependencies
  • Verify Installation
  • Docker Installation
• Quick Start Guide
  • Basic Workflow
  • Command Line Interface
  • Jupyter Notebooks
  • Sample Data

API Reference

• Core API Reference
  • StratigraphicAnalyzer
  • TCIIndex
  • TCIResults
  • DataValidator
  • ParameterValidator
• Parameters API Reference
  • Base Classes
  • Lithological Parameters
  • Isotope Parameters
  • Microfossil Parameters
  • Magnetic Parameters
  • Geochemical Parameters
  • Palynology Parameters
  • Varve Parameters
  • Thermal Parameters
  • Cyclostratigraphy Parameters

Case Studies

• PETM Case Study
  • Paleocene-Eocene Thermal Maximum
  • ODP Site 1209B (Shatsky Rise)
  • Key Results
  • Scientific Significance

Reference

• Glossary
• Contact & Support

The Nine TCI Parameters

Parameter	Symbol	Weight	Description
Lithological Deposition Rate	LDR	20%	Rate of sediment accumulation as function of basin subsidence and compaction
Stable Isotope Fractionation	ISO	15%	δ¹⁸O / δ¹³C ratios encoding palaeotemperature and carbon cycle state
Micro-Fossil Assemblage	MFA	12%	AI-classified foraminifera, nannofossils, and palynomorphs
Magnetic Susceptibility	MAG	11%	Ferrimagnetic mineral content recording geomagnetic reversals
Geochemical Anomaly Index	GCH	10%	Trace element signatures detecting bolide impacts and anoxic events
Palynological Yield Score	PYS	9%	Pollen and spore assemblage diversity encoding vegetation history
Varve Sedimentary Integrity	VSI	8%	Annual lamination preservation in lacustrine sediments
Thermal Diffusion Model	TDM	8%	Subsurface heat flow modeling quantifying burial depth and maturity
Cyclostratigraphic Energy Cycle	CEC	7%	Spectral power at Milankovitch orbital periods for astronomical calibration

Citation

If you use STRATICA in your research, please cite:

BibTeX:

@software{baladi2026stratica,
  author = {Baladi, Samir},
  title = {STRATICA: Stratigraphic Pattern Recognition & Paleoclimatic Temporal Reconstruction},
  year = {2026},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.18851076},
  url = {https://github.com/gitdeeper8/STRATICA}
}

APA:

Baladi, S. (2026). STRATICA: Stratigraphic Pattern Recognition & Paleoclimatic Temporal Reconstruction [Software]. Zenodo. https://doi.org/10.5281/zenodo.18851076

License

This project is licensed under the Creative Commons Attribution 4.0 International License (CC-BY-4.0).

See LICENSE for details.

Contact & Support

Principal Investigator: Samir Baladi

• Email: gitdeeper@gmail.com

• ORCID: 0009-0003-8903-0029

• Phone: +16142642074

Affiliation: Ronin Institute for Independent Scholarship Division: Geological Deep-Time & Geospatial Intelligence Division

Indices and Tables

• Index

• Module Index

• Search Page

In the layers of the Earth, time is not lost — it is stored.
Every stratum is a sentence; every basin is a book.
STRATICA is the language in which that book was always waiting to be read.