WorldPop Open Population Data

Use in your agent

Load WorldPop file metadata into pandas and Parquet. Checked against the provider's documentation on Nov 1, 2026.

Access

• No account or key. The REST catalogue returns JSON metadata and direct GeoTIFF URLs.
• This example requests the 2000-2020 unconstrained population-count series for Australia without downloading the large rasters.

Request

curl 'https://www.worldpop.org/rest/data/pop/wpgp?iso3=AUS'

Format

JSON data array with title, description, DOI, production date, population year, citation, format, country, ISO3, licence, summary page and one or more GeoTIFF URLs.

Load into pandas and Parquet

Needs Python 3 with requests, pandas and pyarrow.

"""WorldPop Australia population-file metadata -> pandas -> Parquet. No key."""
import requests
import pandas as pd

URL = "https://www.worldpop.org/rest/data/pop/wpgp"
r = requests.get(URL, params={"iso3": "AUS"}, timeout=60)
r.raise_for_status()
df = pd.DataFrame(r.json()["data"])
df["files"] = df["files"].map(lambda values: " | ".join(values))
df.to_parquet("worldpop_aus_wpgp_metadata.parquet", index=False)
print(len(df), "files", int(df["popyear"].min()), "to",
      int(df["popyear"].max()), df["data_format"].unique().tolist())

Run by us on 2026-11-01 (python 3.13.5 | pandas 3.1.0 | pyarrow 26.0.0 | requests 2.34.2). Result: 21 files 2000 to 2020 ['Geotiff']

Limits

• Rate limit: Not stated.
• Raster downloads can be large; the checked worldwide 2020 1 km file was listed as 829.43 MB.
• The metadata example returns 21 annual Australia records and does not download raster cell values.

Quirks

• The files field is a JSON list; flatten or explode it before writing tabular output.
• Product families differ by resolution, constrained versus unconstrained modelling and UN adjustment.
• Read the record-specific licence because some products derived from named building or road sources use ODbL rather than CC BY 4.0.

Terms and attribution

• The checked WorldPop population dataset is CC BY 4.0 and requires appropriate credit plus a licence link.
• Named OpenStreetMap and Microsoft-derived datasets use ODbL and can impose share-alike on altered data.

Sources

• https://hub.worldpop.org/: “Open access spatial demographic datasets built using transparent approaches.”
• https://www.worldpop.org/sdi/introapi/: “Query individual countries using the “?” symbol and iso3 code:”
• https://www.worldpop.org/geodata/summary?id=24777: “The dataset is available to download in Geotiff format at a resolution of 30 arc (approximately 1km at the equator).”
• https://www.worldpop.org/geodata/summary?id=24777: “WorldPop datasets are available under the Creative Commons Attribution 4.0 International License.”