# Flateau V2: PLATEAU Building Footprints and Attributes for 303 Japanese Cities (GeoParquet)

## Facts

- Provider: Pacific Spatial Solutions (from MLIT Project PLATEAU)
- License: CC-BY-4.0 (https://source.coop/pacificspatial/flateau-2-0)
- Format: JSON, Parquet (in sampled files)
- Stored at: Amazon S3, s3://ap-northeast-1.opendata.source.coop/pacificspatial/flateau-2-0 (ap-northeast-1)
- Access: public, readable without an account
- Category: Urban & Civic
- Source: https://source.coop/pacificspatial/flateau-2-0
- Page: https://scigantic.com/archive/d7758747-590d-53e0-a5b7-44293bf16b8c
- Keywords: Urban & Civic, JSON, Parquet, AWS Open Data

## About

Version 2 of Flateau, Pacific Spatial Solutions' conversion of Japan's Project PLATEAU national 3D city models (CityGML, Ministry of Land, Infrastructure, Transport and Tourism) into analysis-ready GeoParquet. LOD0 building footprints and their attributes are extracted for every city where PLATEAU CityGML exists: 303 datasets from fiscal years 2020 to 2025, 29,290,201 buildings in 45 prefectures, all on one fixed schema (NULL where a city lacks a field), with added heights and areas computed from the 3D geometry. Two collections: flateau-v2-by-city (all available layers per city, joinable by gml_id) and flateau-v2-kdtree (KD-tree-partitioned building_lod0 footprints for fast area reads); each has an items.parquet STAC GeoParquet index and catalog.json is the STAC root. Read with DuckDB, geopandas or pystac. Use for urban exposure, building stock, hazard and energy studies. Licensed CC BY 4.0 by Pacific Spatial Solutions; PLATEAU terms of use also apply (CC BY 4.0 compatible).

## Load the data

```python
import boto3
from botocore import UNSIGNED
from botocore.client import Config
s3 = boto3.client('s3', region_name='ap-northeast-1', config=Config(signature_version=UNSIGNED))
s3.list_objects_v2(Bucket='ap-northeast-1.opendata.source.coop', Prefix='pacificspatial/flateau-2-0/', Delimiter='/')
```

Lists the files with anonymous access. Runs on any machine, with no cloud account.

## Data structure

File types in sampled files: JSON, Parquet

Example file: `pacificspatial/flateau-2-0/flateau-v2-by-city/adachi-ku-13121/adachi-ku-13121.json`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/d7758747-590d-53e0-a5b7-44293bf16b8c
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/d7758747-590d-53e0-a5b7-44293bf16b8c
- Over MCP: `get_schema_card` with id `d7758747-590d-53e0-a5b7-44293bf16b8c` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [JSON datasets](https://scigantic.com/topics/json)
- [Urban & Civic datasets](https://scigantic.com/topics/urban-and-civic)
