xgapfilling

Gap filling using direct sampling and extremes. Explore fracture density reconstruction with Jupyter notebook examples and synthetic well log data. This repository is associated with paper:

Fracture density reconstruction using direct sampling multiple-point statistics and extreme value theory.

Ana Paula Burgoa Tanaka ^1,2*, Philippe Renard¹, Julien Straubhaar¹

¹ Centre for Hydrogeology and Geothermics, University of Neuchâtel, Neuchâtel, Switzerland
² Santos Basin Unit, Petrobras, Santos, São Paulo, Brazil

* corresponding author: ana.burgoa@unine.ch

Journal reference

A. P. B. Tanaka, P. Renard, J. Straubhaar (2024). Fracture density reconstruction using direct sampling multiple-point statistics and extreme value theory. Applied Computing and Geosciences. Volume 22.

https://doi.org/10.1016/j.acags.2024.100161

Data description

Due to restrictions on sharing the original data used in the reference, we generate synthetic data for the examples provided in this repository. The synthetic log analysis does not add content to the manuscript. It is available for the dissemination of the application of the proposed methodology and for the ease of the execution of the code. People are also encouraged to try it with their own dataset.

Files

In the "data" folder, you will find:

• Synthetic well log fracture density data in gslib format

  • Filename: syn_well_p10.gslib

• Synthetic well log fracture density data with gap in gslib format

  • Filename: syn_well_p10_gap.gslib

• Synthetic well log enriched fracture density data with gap in gslib format

  • Filename: syn_naive.gslib

• Synthetic well log enriched fracture density data with gap in gslib format

  • Filename: syn_lift.gslib

Dependencies

Use the package manager pip to install the following packages.

pip install geone matplotlib numpy scipy jupyter-notebook