# US Census Demographics Cloud-Native Mirror: ACS 2020-2024 5-year Indicators and 2020 Census Block Counts (GeoParquet, PMTiles, STAC)

## Facts

- Provider: US Census Bureau; Portolan mirror by Chris Holmes
- License: CC0-1.0 (https://www.census.gov/data.json)
- Format: Parquet, JSON (in sampled files)
- Stored at: Amazon S3, s3://us-west-2.opendata.source.coop/portolan-mirrors/us-census-catalog (us-west-2)
- Access: public, readable without an account
- Category: Demographics & Economics
- Source: https://data.census.gov/
- Page: https://scigantic.com/archive/647b2caf-a96c-5d01-a445-0ac9d2a495a2
- Keywords: Demographics & Economics, Parquet, JSON, AWS Open Data

## About

United States census statistics as cloud-native GeoParquet and PMTiles with STAC metadata, at every census geography from states down to the 242,298 block groups and 8,174,955 blocks of the 2020 Census, plus 33,791 ZIP Code Tabulation Areas (ZCTAs), 32,612 places and 441 congressional districts of the 119th Congress. Thirty-nine collections carry curated American Community Survey (ACS) 2020-2024 5-year indicators: income, poverty, age, race and ethnicity, education, housing cost and tenure, vehicles, commuting, internet access, health insurance, disability and nativity, each estimate with its 90% margin of error and headline statistics with a coefficient of variation. One collection carries 2020 Decennial Census P.L. 94-171 redistricting counts at block level, hive-partitioned by state, and six carry 2020 Demographic and Housing Characteristics (DHC) counts protected by differential privacy. Built from the Bureau's own Summary File, 2024 cartographic boundary files and TIGER/Line 2024 blocks. Start at catalog.json; read with DuckDB, geopandas or pystac. Census data are public domain (CC0 in the Bureau's data catalog); attribute the US Census Bureau. Not an official Census product.

## Load the data

```python
import boto3
from botocore import UNSIGNED
from botocore.client import Config
s3 = boto3.client('s3', region_name='us-west-2', config=Config(signature_version=UNSIGNED))
s3.list_objects_v2(Bucket='us-west-2.opendata.source.coop', Prefix='portolan-mirrors/us-census-catalog/', Delimiter='/')
```

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

## Data structure

File types in sampled files: Parquet, JSON

Example file: `portolan-mirrors/us-census-catalog/block-groups/bg-age-dec2020/AGENTS.md`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/647b2caf-a96c-5d01-a445-0ac9d2a495a2
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/647b2caf-a96c-5d01-a445-0ac9d2a495a2
- Over MCP: `get_schema_card` with id `647b2caf-a96c-5d01-a445-0ac9d2a495a2` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [JSON datasets](https://scigantic.com/topics/json)
- [Demographics & Economics datasets](https://scigantic.com/topics/demographics-and-economics)
