Geological Modeling and Prospectivity Mapping for Natural Hydrogen from Organic-Rich Rocks: Case Study of the Irati Formation, Paraná Basin

• Grant number: 25/26838-8
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: January 01, 2026
• End date: December 31, 2027
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Geology
• Principal Investigator: Tercio Ambrizzi
• Grantee: Stephanie San Martin Canas Janowsky
• Host Institution: Instituto de Energia e Ambiente (IEE). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 24/00949-5 - Climate Crisis and Disasters Resilience Research Center (CLIMARES), AP.CEPID

Abstract

This project aims to develop predictive models to evaluate the potential (prospectivity) of organic-rich rocks for the generation and accumulation of geologic hydrogen, as well as for the production of blue hydrogen through methane reforming with associated CO2 geological storage. The stored CO2 will also be used for enhanced gas recovery (EGR) in shale gas production, with the Irati Formation of the Paraná Sedimentary Basin as the main case study. To achieve this, the project will follow a methodological framework that includes the compilation and integration of geochemical, petrophysical, and structural data from wells in the REATE platform and from the literature, forming a robust georeferenced database; the construction of 3D geological models; and the development of predictive models using machine learning algorithms and deep learning architectures, combined with data mining techniques, spatial interpolation, and multicriteria analysis to generate favorability and uncertainty maps. The analysis will incorporate elements analogous to petroleum systems in order to propose an adapted system for the context of geologic hydrogen. The results will be integrated with geothermal data and studies on CO2 geological storage, allowing the evaluation of regional synergies among geologic hydrogen, blue hydrogen, geothermal energy, shale gas, and CO2 storage. The proposed approach combines geological modeling, advanced artificial intelligence techniques (including machine learning algorithms such as Random Forest, Support Vector Machine, k-Nearest Neighbors, and XGBoost, and deep learning architectures such as deep neural networks and hybrid models) along with geospatial analysis. The results will directly contribute to understanding the processes of generation and trapping of geologic hydrogen in the Irati Formation, support the identification of favorable areas, and assist in developing sustainable exploration strategies for new energy resources in the Paraná Basin. (AU)