| Data Model | A thematic analysis category (e.g., Least Cost Electrification). Container and configuration for Scenarios, Vector Datasets, and display settings. Can be public or private (is_public). |
| Scenario | A specific model run within a Data Model (e.g., Baseline 2025). Linked to a Vector Dataset and one or more Scenario Files. Ordered by presentation_order. |
| Scenario File | A CSV uploaded to a Scenario containing modeled output attributes. Must include an id column. Supports low_zoom_as_points rendering option. |
| ScenarioData | Database table (JSONB) storing all imported Scenario File attribute data per feature. Powers the /feature/ API endpoint. |
| ScenarioDataMetrics | Denormalized key-value table for fast aggregation queries. Auto-populated on Scenario File import. Powers the /summaries/ API endpoint. |
| Vector Dataset | Named container for spatial geometry files (.geojson, .gpkg, .zip, .kml). Includes rich metadata fields. Supports EN/PT translations. |
| Raster Dataset | Container for Cloud-Optimized GeoTIFF (COG) imagery files. Displayed as raster overlay layers on the map. |
| Reference Dataset | Container for supporting documents and tables (.pdf, .csv, .xlsx, .docx, etc.). Listed in the Downloads section. |
| PMTiles | Single-file vector tile archive served via HTTP range requests. Generated from vector files using Tippecanoe. |
| COG | Cloud-Optimized GeoTIFF. A raster format optimised for streaming from object storage. Must be in EPSG:3857 with 256×256 block size for the platform's raster rendering. |
| presentation_order | Integer field on DataModel and Scenario controlling sort order in the frontend sidebar and dropdown. Lower numbers appear first. |
| is_public (DataModel) | Boolean on DataModel. If False, the model is hidden from the API for non-superusers and does not appear on the frontend. |
| metric_field_types | JSON dict on DataModel mapping CSV column names to "numeric" or "string". Auto-populated on Scenario File import. Powers the summaries API. |
| visualization_column | The CSV column whose unique values determine map feature colours, as configured by color_coding. |
| low_zoom_as_points | ScenarioFile option that renders polygon centroids at low zoom (5–10) and full geometry at high zoom (11–14). Improves performance for large polygon datasets. |
| Celery | Distributed task queue for async processing (PMTiles generation, CSV import). |
| RabbitMQ | Message broker for Celery. In production, uses a dedicated vhost: proenergia_vhost. |
| Tippecanoe | Command-line tool that generates PMTiles from FlatGeobuf vector data. |
| tile-join | Tippecanoe companion tool that merges multiple PMTiles files (used for low_zoom_as_points). |
| GDAL / ogr2ogr | Geospatial data library for format conversion and reprojection. |
| i18n | Internationalisation. The platform supports Portuguese (default) and English via i18next. Backend models support name_pt and description_pt translation fields. |
| proenergia_db | The production PostgreSQL database name. |
| DATABASE_URL | Primary environment variable for the database connection (postgis://...). |