PoNHy-map

This repo is a fork of RodolfoChristiansen/PoNHy (Potential for Natural Hydrogen) with changes focused on running PoNHy reliably and mapping results.

General changes in this fork:

• Run script & deps: run_ponhy.sh (venv + optional CPU limits) and requirements.txt for pip users; packaging and other deps pinned so the workflow runs without conda.
• SimPEG v0.24+: Inversion uses active_cells in gravity/magnetic simulations; plotting still uses ind_active where required.
• Stability: ponhy_config_light.yaml for lighter runs (fewer iterations, 1 core, optional analyses off); README notes on avoiding overload.
• Paths & config: Absolute paths in config are left unchanged; clearer errors when the chosen Data folder doesn’t match the config (e.g. Pyrenees config → Data_Pyrenees).
• Non-interactive: When multiple Data folders exist and stdin isn’t a TTY, the first Data folder is used automatically.
• GeoJSON & map: export_to_geojson.py converts result CSVs (or topo/gravity inputs) to WGS84 GeoJSON (points, --polygons, --value-ranges); index.html loads density/gravity/topography polygon layers (e.g. on GitHub Pages from the main branch).

For full documentation (quick start, environment, configuration, troubleshooting, contact), see README_original.md.

---

Adding a new region (different input data)

To run PoNHy on another area (e.g. Southern Africa, Australia) instead of Pyrenees or California, you change the input data and a config file. The code in RodolfoChristiansen/PoNHy does not need to be edited; only data files and YAML config.

Reference: Pyrenees raw data to replicate

Everything in Data_Pyrenees/ is input data. PoNHy never writes into any Data_* folder; all outputs go to timestamped folders (Results_Inversion_*, Results_Generation*). The files below are the reference formats to replicate for other areas. All paths are under the repo root.

Purpose	Pyrenees file	Format
Topography	Data_Pyrenees/Ext_Topo.txt	Tab-separated, no header: X, Y, Z
Gravity	Data_Pyrenees/Ext_Grav.txt	Tab-separated, no header: X, Y, Z, gravity
Magnetic	Data_Pyrenees/Ext_Mag_down.txt	Tab-separated, no header: X, Y, Z, magnetic
Initial geology model	Data_Pyrenees/Updated_Geology_Pyrenees.csv	CSV header: X,Y,Z,Density,Susceptibility
Temperature (H₂)	Data_Pyrenees/temperature_70.csv	CSV header: X,Y,Z,Temperature
Serpentinite density	Data_Pyrenees/Density_Serpentinite.csv	CSV (from inversion or prior)
Serpentinite susceptibility	Data_Pyrenees/Magsus_Serpentinite.csv	CSV (from inversion or prior)
H₂ thermodynamic DB	Data_Pyrenees/DB0/	Zarr datasets (HZ1, LH1); can copy from repo

California uses the same layout with different filenames: Data_California/Topo_data.txt, Grav_data.txt, Mag_data_down.txt, Initial_geology_small_20.csv, Temperature_20.csv, plus Density_Serpentinite.csv, Magsus_Serpentinite.csv, and DB0/. Match column order and units to the Pyrenees files when building data for a new region.

1. Get or prepare data for your region

You need the following in the same grid (same X,Y points for every file; no gaps). Coordinates in meters (UTM) for your region's UTM zone.

File type	Format	Columns (order matters)
Topography	Tab-separated, no header	X Y Z (elevation)
Gravity	Tab-separated, no header	X Y Z (or topo) gravity_value (last column is used)
Magnetic	Tab-separated, no header	X Y Z (or topo) magnetic_value (last column is used)
Initial model (optional)	CSV, with header	X,Y,Z,Density,Susceptibility (one row per cell; density/susc can be contrasts)
Temperature (for H₂)	CSV, with header	X,Y,Z,Temperature
H₂ DB (for H₂)	Same as in Data_Pyrenees/DB0	Copy from repo or provide equivalent

Where to source input data (public/free):

• Gravity: ICGEM (GFZ Potsdam) — global gravity field models (e.g. EIGEN-6C4); calculation service for grids. BGI and national survey archives often provide regional gravity.
• Magnetic: EMAG2 / EMAG2v3 (NOAA NCEI) — global 2-arc-minute magnetic anomaly grid; downloads (GeoTIFF, CSV, etc.). Resample to your grid and convert to the same units as the Pyrenees magnetic file.
• Topography / DEM: SRTM (NASA Earthdata, 1″ or 3″ global), USGS EarthExplorer, Copernicus DEM. Export elevation on a regular X,Y grid in UTM.
• Temperature at depth: Often from geothermal or lithospheric models (region-specific); can use simple conductive gradients or public geothermal datasets where available.
• Initial geology model: From geological maps, prior inversions, or literature; build a CSV with X,Y,Z and density/susceptibility per unit.

Resample all datasets to one common X,Y grid (and Z if needed) in UTM, then export in the column order above.

Public data with similar (output-style) results:

• USGS Geologic Hydrogen Prospectivity — raster and tables for hydrogen accumulation potential in the conterminous US (CC0); not PoNHy output but comparable "H₂ potential" mapping.
• EGI Natural Hydrogen Knowledge Platform — GIS of natural H₂ sites and geological context.
• PoNHy itself does not publish a central repository of result datasets; runs are local. Sharing outputs would mean publishing your own result CSVs/GeoJSON (e.g. via Zenodo, your repo, or a data portal).

2. Create a new data folder

Inside the PoNHy repo (same level as Data_Pyrenees), create e.g. Data_MyRegion. Put your files there. You can use any filenames (e.g. Topo_SA.txt, Grav_SA.txt).

3. Create a new config YAML

Copy ponhy_config_pyrenees.yaml (or ponhy_config_california.yaml) to e.g. ponhy_config_myregion.yaml. Edit at least:

• Paths: TOPO_FILE, GRAV_FILE, MAG_FILE, INITIAL_MODEL (if used), TEMPERATURE_FILE, DENSITY_FILE, MAGSUS_FILE — set to your filenames inside Data_MyRegion.
• Mesh: DX, DY, DZ, DEPTH_CORE, EXPANSION_TYPE, ORIGINAL_X, ORIGINAL_Y, DOMINANT_SIDE to match your grid extent and expansion.
• Geophysics: INCLINATION, DECLINATION, STRENGTH — use the Earth's field for your region (e.g. IGRF or a calculator for your lat/lon).
• Petrophysics / units: Adjust UNIT_LABELS, UNIT_DENS_*, UNIT_MAGSUS_*, SERPENTINITE_LABEL, etc. if your geology differs.

4. Run PoNHy

From the PoNHy folder run ./run_ponhy.sh (or python ponhy.py). When prompted, select your new config (e.g. ponhy_config_myregion.yaml) and your new data folder (e.g. Data_MyRegion). The workflow then uses your input data and writes results to a new timestamped folder.

5. Export to GeoJSON and map

Use export_to_geojson.py with the UTM zone for your region (e.g. Southern Africa often zone 34–36; use --utm 35 for zone 35N; southern hemisphere uses different EPSG codes — if you need that, the export script may need a small change to use the southern-hemisphere UTM EPSG). Then point mapbox_map.html at the new GeoJSON files or replace the existing ones.