GeoBootstrap

A lightweight, transparent Python workflow demonstrating:

• Experimental variograms
• Variogram model fitting (spherical, exponential)
• Simple kriging
• Spatial block bootstrap uncertainty
• Resource-style calculation using thickness × density (RD)

This project is intended for educational and prototyping use in geoscience and resource-style spatial estimation workflows.

---

Why this exists

Many geostatistics examples are either too abstract or locked inside black-box tooling. GeoBootstrap aims to be a readable "glass box" reference you can adapt to your own datasets.

---

Features

• Omnidirectional experimental variogram calculation
• Spherical and exponential model fitting via non-linear least squares
• Simple kriging using covariance form
• K-means-like spatial block assignment
• Block bootstrap resampling for uncertainty on total in situ estimates
• Polygon masking with consistent contour styling
• Spyder-friendly file dialogs with console fallback

---

“Start Here” checklist (for your zip)

1. Unzip the folder anywhere (e.g., Desktop).

2. Double-click run_windows.bat.

3. When asked for the borehole file, select sample_boreholes.txt.

4. When asked for the polygon file, select sample_polygon.txt.

5. Confirm the plots appear and the summary prints in the console.

6. Replace the sample files with your own data:

   • Borehole file with headers: Easting, Northing, Thickness, RD
   • Polygon file with x,y points per line

7. Run run_windows.bat again and select your real files.