Barangay Boundaries Repository — Philippine Barangay Boundaries GeoJSON & PSGC RDF

Hierarchical per-class Philippine administrative boundaries (GeoJSON) — barangays, municipalities, and cities split by legal class (highly urbanized, independent component, component), plus provinces, regions, and special geographic areas — each polygon annotated with its canonical PSGC code. Generated from PSA Philippine Standard Geographic Code snapshots and NAMRIA administrative-boundary shapefiles, with the full history of quarterly PSGC changes published as a W3C ORG-ontology knowledge graph. The enriched adm0–adm4 files and the pre-enrichment raw NAMRIA conversion are preserved as intermediate pipeline stages.

What this is

This repository is both a dataset and the toolchain that produces it. The dataset's curated output is the hierarchical per-class GeoJSON collection — one file per administrative class (barangays, provinces, municipalities, highly_urbanized_cities, independent_component_cities, component_cities, special_geographic_areas, …) — with each polygon annotated with its canonical PSGC code, name, and status. These are produced from NAMRIA shapefiles enriched with PSA PSGC data; the enriched (adm0–adm4) and raw NAMRIA-converted files are kept as intermediate stages of the same pipeline. The toolchain (brgybnd) also ingests PSA PSGC Excel/PDF snapshots and emits RDF linked data using the W3C ORG ontology plus machine-readable deltas across 17 quarterly snapshots (2021-08-19 → 2026-04-13).

It is built for GIS analysts, researchers, journalists, and civic-tech projects who need authoritative, code-stable Philippine boundary data that can be joined to any PSGC-keyed dataset.

Key features

• Hierarchical per-class extracts — the curated end product: one GeoJSON per administrative class (barangays, provinces, municipalities, highly_urbanized_cities, independent_component_cities, component_cities, special_geographic_areas, …) with cities split by legal class and each polygon tagged with its canonical PSGC code.
• PSGC pcode enrichment — pipeline stage (enriched_t0p005/) that annotates every feature with psgc_id, psgc_code, psgc_name, psgc_status, and a match_confidence score (exact / HUC-map / fuzzy); this feeds the classifier that produces the hierarchical output.
• GeoJSON per administrative level — country (ADM0), region (ADM1), province (ADM2), municipality/city (ADM3), barangay (ADM4), simplified with Douglas–Peucker (tolerance 0.005°).
• 17 quarterly snapshots of PSGC change history as RDF (delta.ttl / .nt / .jsonld) with creation, deletion, transfer, merger, split, rename, and code-change events.
• HUC mapping — resolves the Highly-Urbanized-City / NAMRIA-province mismatch (see The HUC problem).
• 99.945% pcode coverage between PSGC and NAMRIA for the 2023-10-24 snapshot.

Download the data

The curated hierarchical per-class GeoJSON is published as GitHub Releases — as a zipped bundle, individual per-class files, and a checksum manifest. The enriched adm* files and raw NAMRIA conversion are intermediate pipeline stages kept in-repo for traceability. See Releases.

The inaugural release v2026.4.13.0 ships the fully processed 2023-10-24 snapshot:

Asset	Contents	Tier
barangay-boundaries-2023-10-24-hierarchical.zip	per-class GeoJSON (barangays, provinces, regions, municipalities, HUC/ICC/component cities, …)	curated (recommended)
barangays.geojson, provinces.geojson, regions.geojson, …	individual per-class extracts	curated (recommended)
manifest.json	per-file MD5/SHA-256, size, feature count	—
delta.{ttl,nt,jsonld}	PSGC change-history RDF	—
enriched_t0p005/adm{0..4}.geojson	PSGC-enriched, pre-classification (in-repo only)	intermediate stage
raw_t0p005/adm{0..4}.geojson	NAMRIA-converted, pre-enrichment (in-repo only)	intermediate stage

Or grab an individual per-class file directly, e.g.:

# All ~42,000 barangay polygons (curated per-class extract)
curl -LO https://github.com/bendlikeabamboo/barangay-boundaries-repository/releases/download/v2026.4.13.0/barangays.geojson

Quick start

The CLI (brgybnd) converts PSA PSGC snapshots and NAMRIA shapefiles into enriched GeoJSON and RDF. It is GitHub-only (not on PyPI); install directly from the repository with uv:

uv tool install git+https://github.com/bendlikeabamboo/barangay-boundaries-repository.git

Three example workflows:

# 1. Ingest a PSGC snapshot (datafile + changes + press release) for a given date
brgybnd ingest --date 2023-10-24

# 2. Convert NAMRIA shapefiles to simplified GeoJSON (ADM0–ADM4)
brgybnd convert-geo --levels 0,1,2,3,4 --tolerance 0.005

# 3. Build the curated per-class hierarchical GeoJSON (convert → enrich → classify → split)
brgybnd build-hierarchical --date 2023-10-24

Other commands: brgybnd list, brgybnd process, brgybnd process-all, brgybnd delta, brgybnd validate, brgybnd coverage, brgybnd enrich. Run brgybnd --help for the full reference.

Data coverage (snapshot 2023-10-24)

PSGC ↔ NAMRIA pcode matching, from enriched_t0p005/summary.md:

Level	Meaning	PSGC entities	GeoJSON features	Matched	Coverage
ADM0	Country	1	1	1	100.000%
ADM1	Region	17	17	17	100.000%
ADM2	Province	82	88	88	100.000%
ADM3	Municipality / City	1,644	1,626	1,624	99.877%
ADM4	Barangay	42,010	41,825	41,803	99.947%
Total		43,754	43,557	43,533	99.945%

Per-class feature counts (hierarchical extract, NAMRIA v2023-11-06):

Class	Features
Region	17
Province	82
Municipality	1,485
Component City	110
Independent Component City	6
Highly Urbanized City	34
Barangay	42,026
Special Geographic Area	9

Known limitations (represented honestly)

• Submunicipalities: NAMRIA ADM4 has no separate polygons for the 14 Manila submunicipalities, so submunicipalities.geojson is empty. Aggregation from barangays is out of scope.
• 6 unresolved features (unresolved.geojson) and 20 non-administrative features could not be classified to a standard PSGC level.
• 230 PSGC entities (mostly barangays in BARMM and Metro Manila) have no NAMRIA polygon.

PSGC code structure

The Philippine Standard Geographic Code is a 10-digit hierarchical identifier of the form PPCCCMMLLL0:

Digits	Level	Example
PP	Region	1300000000 — National Capital Region
PPCCC	Province / HUC	1339000000 — City of Makati
PPCC CMM	Municipality / City	1339060000 — City of Makati (muni-level)
PPCCCMMLLL	Barangay	1339060014 — a barangay of Makati

A 9-digit correspondence code tracks entity identity across code changes (used to compute deltas). See PSGC reference for the full breakdown.

The HUC problem

Highly Urbanized Cities (HUCs) are administratively independent of any province, so the PSGC records their parent as the region. NAMRIA, however, draws HUC polygons inside their geographic province. A naive parent-lookup therefore misassigns HUCs.

This repository resolves the mismatch with namria/huc_adm2_mapping.json (generated by generate_huc_mapping.py), consumed by both coverage.py and enrich.py. Metro Manila is handled separately: it has no PSGC province-level units, so NAMRIA's four NCR legislative districts are mapped as virtual provinces of type mm_district.

Read the full explainer → HUC mapping.

Outputs / file map

Each snapshot directory (YYYY-MM-DD/) contains:

File	Description
delta.ttl / .nt / .jsonld	PSGC change-history RDF (W3C ORG) for that snapshot
Press-Release-*.md	Source PSA press-release text (extracted from PDF)
hierarchical_t0p005/	curated per-class GeoJSON extracts + classification_report.json + summary.md
enriched_t0p005/	intermediate GeoJSON annotated with PSGC codes (adm0–adm4)
raw_t0p005/	intermediate NAMRIA shapefile → GeoJSON, before enrichment

The full pipeline and architecture are documented in AGENTS.md.

Snapshot history

17 quarterly snapshots are tracked, each producing an RDF delta against its predecessor:

2021-08-19 · 2022-04-29 · 2022-11-08 · 2023-04-18 · 2023-08-15 · 2023-10-24 (first fully enriched release) · 2024-01-23 · 2024-04-23 · 2024-05-08 · 2024-07-12 · 2024-10-18 · 2025-01-30 · 2025-04-23 · 2025-08-29 · 2025-10-13 · 2026-01-13 · 2026-04-13.

Browse the full snapshot timeline → PSGC snapshot history, or the version changelog.

Citation

If this dataset supports your work, please cite it. See CITATION.cff (or use GitHub's "Cite this repository" button) and the citation page for ready-to-paste BibTeX and APA entries.

License

Code is released under the MIT License — see LICENSE.

Data attribution

• PSGC snapshots — © Philippine Statistics Authority (PSA). Source: https://psa.gov.ph/classification/psgc/.
• Administrative boundaries — © NAMRIA, shapefile version 2023-11-06. Source: https://namria.gov.ph/.

These datasets are redistributed for research and interoperability. Please credit PSA and NAMRIA in any downstream product.

Contributing

Issues and pull requests are welcome — especially PSGC/NAMRIA coverage gaps, new snapshot processing, and HUC-mapping corrections. See AGENTS.md for the architecture and conventions before contributing.