Loading GeoViable Environmental Layer Data into PostGIS

GeoViable’s PostGIS database contains no environmental data on a fresh installation. The seven official layers — Red Natura 2000, flood zones, Hydraulic Public Domain, livestock paths, Protected Natural Spaces, and surface and groundwater bodies — must be loaded explicitly before spatial analyses return meaningful results. This page describes both loading methods.

Running an analysis on an empty database is not an error. The API will respond successfully, but every layer will show risk: ninguno with zero spatial intersections — because there is simply nothing to intersect with. If you receive a valid PDF with no afecciones on any layer, check that the data has been loaded before filing a bug report.

Loading Methods

Option A — Manual load from ZIPs (recommended)
Option B — Automated download from MITECO/CNIG

The recommended approach for initial setup is to load from preprocessed ZIP files containing the official MITECO/CNIG shapefiles filtered and reprojected for Galicia.

Step 1 — Request the ZIP files

The preprocessed ZIPs are available on request:

Request data files from: daniel.zas.dacosta@gmail.com

Step 2 — Place ZIPs in `backend/data/`

Copy the received files into the backend/data/ directory. The load_initial_data.py script looks for files with these exact names:

File	Layer
`red_natura_2000.zip`	Red Natura 2000 (ZEPA + LIC/ZEC)
`zonas_inundables_t100.zip`	Flood zones T100
`zonas_inundables_t500.zip`	Flood zones T500
`dph.zip`	Hydraulic Public Domain
`vias_pecuarias.zip`	Livestock paths
`enp.zip`	Protected Natural Spaces
`masas_agua_superficial.zip`	Surface water bodies
`masas_agua_subterranea.zip`	Groundwater bodies

Step 3 — (Optional) Inspect before loading

Before committing to a full load, you can inspect each ZIP’s column names, CRS, geometry types, and column mapping without writing anything to the database:

docker compose exec geoviable-api python -m scripts.load_initial_data --inspect

This is useful for diagnosing NULL columns caused by shapefile field name changes between MITECO dataset versions.

Step 4 — Load all layers

docker compose exec geoviable-api python -m scripts.load_initial_data

The script runs a TRUNCATE + INSERT pipeline for each layer inside a single transaction, then runs VACUUM ANALYZE to update query planner statistics. At the end it prints a summary table with record counts for every layer.

Reload a single layer

If one layer fails or you need to refresh it independently:

docker compose exec geoviable-api python -m scripts.load_initial_data --layer red_natura_2000

Available values for --layer:

`--layer` value	Table
`red_natura_2000`	`red_natura_2000`
`zonas_inundables`	`zonas_inundables` (loads both T100 and T500)
`dominio_publico_hidraulico`	`dominio_publico_hidraulico`
`vias_pecuarias`	`vias_pecuarias`
`espacios_naturales_protegidos`	`espacios_naturales_protegidos`
`masas_agua_superficial`	`masas_agua_superficial`
`masas_agua_subterranea`	`masas_agua_subterranea`

GeoViable includes an update_layers.py script that downloads shapefiles directly from MITECO and CNIG official sources and loads them into the database in one step. This script also runs automatically on the first day of every month via a cron job inside the geoviable-api container.To trigger it manually:

docker compose exec geoviable-api python -m scripts.update_layers

The automated download may fail if MITECO or CNIG change their download URLs — which happens occasionally after dataset updates. Option A (manual load from ZIPs) is more reliable for initial setup. Use Option B for keeping the database current after the first load.

Cron execution logs are written to /var/log/geoviable_update.log inside the container:

docker compose exec geoviable-api tail -f /var/log/geoviable_update.log

Verifying the Load

After loading, connect to the database and run this query to confirm all seven tables have data:

SELECT 'red_natura_2000'            AS tbl, count(*) FROM red_natura_2000
UNION ALL SELECT 'zonas_inundables',          count(*) FROM zonas_inundables
UNION ALL SELECT 'dominio_publico_hidraulico', count(*) FROM dominio_publico_hidraulico
UNION ALL SELECT 'vias_pecuarias',            count(*) FROM vias_pecuarias
UNION ALL SELECT 'espacios_naturales_protegidos', count(*) FROM espacios_naturales_protegidos
UNION ALL SELECT 'masas_agua_superficial',    count(*) FROM masas_agua_superficial
UNION ALL SELECT 'masas_agua_subterranea',    count(*) FROM masas_agua_subterranea;

Run it with:

docker compose exec geoviable-db psql -U geoviable -d geoviable -c "
  SELECT 'red_natura_2000' AS tbl, count(*) FROM red_natura_2000
  UNION ALL SELECT 'zonas_inundables', count(*) FROM zonas_inundables
  UNION ALL SELECT 'dominio_publico_hidraulico', count(*) FROM dominio_publico_hidraulico
  UNION ALL SELECT 'vias_pecuarias', count(*) FROM vias_pecuarias
  UNION ALL SELECT 'espacios_naturales_protegidos', count(*) FROM espacios_naturales_protegidos
  UNION ALL SELECT 'masas_agua_superficial', count(*) FROM masas_agua_superficial
  UNION ALL SELECT 'masas_agua_subterranea', count(*) FROM masas_agua_subterranea;
"

Every row should show a count greater than zero. A count of 0 means that layer either was not loaded, was filtered out entirely by the Galicia bounding box, or failed silently — check the script output and logs. You can also verify layer freshness and load timestamps through the API:

GET /api/v1/layers/status

All layers are stored in EPSG:25830 (ETRS89 / UTM zone 30N). The loading script reprojects every shapefile to this CRS automatically, regardless of the source projection. If you supply your own sample data for testing, ensure it is in EPSG:25830 — coordinates in UTM zone 29N will not spatially intersect with Galicia features stored in zone 30N.

Get Started

Using GeoViable

Deployment

Data & Database

Troubleshooting

Loading GeoViable Environmental Layer Data into PostGIS

Loading Methods

Step 1 — Request the ZIP files

Step 2 — Place ZIPs in `backend/data/`

Step 3 — (Optional) Inspect before loading

Step 4 — Load all layers

Reload a single layer

Verifying the Load

Build docs developers (and LLMs) love

Get Started

Using GeoViable

Deployment

Data & Database

Troubleshooting

Documentation Index

​Loading Methods

​Step 1 — Request the ZIP files

​Step 2 — Place ZIPs in backend/data/

​Step 3 — (Optional) Inspect before loading

​Step 4 — Load all layers

​Reload a single layer

​Verifying the Load

Build docs developers (and LLMs) love

Loading Methods

Step 1 — Request the ZIP files

Step 2 — Place ZIPs in `backend/data/`

Step 3 — (Optional) Inspect before loading

Step 4 — Load all layers

Reload a single layer

Verifying the Load