Brazil Data Cube — CBERS-4 / AMAZONIA-1 / MODIS VI¶

The bdc endpoint (INPE's Brazil Data Cube STAC v1) is the only global source of CBERS-4 / CBERS-4A and AMAZONIA-1 imagery, plus 16-day MODIS Terra/Aqua vegetation indices over South America. Every currently-exposed BDC collection reads anonymously (verified by the A1 token probe in planning/stac-extensions/captures/).

Forward-compat note: earthlens ships a BdcTokenSigner + Collection.requires_token infra so a future BDC tier requiring $BDC_ACCESS_TOKEN can land catalog-only. Today no curated row uses either.

The endpoint + curated collections (no network)¶

In [ ]:

from earthlens.stac import Catalog

cat = Catalog()
e = cat.endpoints["bdc"]
print(f"url   : {e.url}")
print(f"signer: {e.signer}")
bdc_keys = [k for k in cat.datasets if k.startswith("bdc/")]
print(f"curated: {bdc_keys}")

from earthlens.stac import Catalog cat = Catalog() e = cat.endpoints["bdc"] print(f"url : {e.url}") print(f"signer: {e.signer}") bdc_keys = [k for k in cat.datasets if k.startswith("bdc/")] print(f"curated: {bdc_keys}")

CBERS-4 WFI 16-day NDVI composite over São Paulo (live)¶

CBERS4-WFI-16D-2 is BDC's flagship cube — every 16 days, every pixel south of Mexico, fused from the China-Brazil CBERS-4 WFI sensor. The NDVI asset is a single-band float COG.

In [ ]:

import tempfile

from earthlens.earthlens import EarthLens

el = EarthLens(
    data_source="bdc",                   # alias 'brazil-data-cube' works too
    start="2024-01-01", end="2024-12-31",
    dataset="bdc/CBERS4-WFI-16D-2",
    variables=["NDVI"],
    aoi=[-46.8, -23.7, -46.3, -23.2],     # São Paulo
    path=tempfile.mkdtemp(), max_items=1,
)
paths = el.download()
paths

import tempfile from earthlens.earthlens import EarthLens el = EarthLens( data_source="bdc", # alias 'brazil-data-cube' works too start="2024-01-01", end="2024-12-31", dataset="bdc/CBERS4-WFI-16D-2", variables=["NDVI"], aoi=[-46.8, -23.7, -46.3, -23.2], # São Paulo path=tempfile.mkdtemp(), max_items=1, ) paths = el.download() paths

AMAZONIA-1 WFI surface reflectance (recipe)¶

Same request shape — only dataset and variables change. AMAZONIA-1 is Brazil's first domestically-built Earth-observation satellite; BDC is the only public source globally.

In [ ]:

amz = EarthLens(
    data_source="bdc",
    start="2024-01-01", end="2024-12-31",
    dataset="bdc/AMZ1-WFI-L4-SR-1",
    variables=["BAND2", "BAND1", "BAND3"],  # red, green, NIR
    aoi=[-60.0, -3.5, -59.5, -3.0],          # over the Amazon basin
    path=tempfile.mkdtemp(), max_items=1,
)
amz_paths = amz.download()
amz_paths

amz = EarthLens( data_source="bdc", start="2024-01-01", end="2024-12-31", dataset="bdc/AMZ1-WFI-L4-SR-1", variables=["BAND2", "BAND1", "BAND3"], # red, green, NIR aoi=[-60.0, -3.5, -59.5, -3.0], # over the Amazon basin path=tempfile.mkdtemp(), max_items=1, ) amz_paths = amz.download() amz_paths

Inspect¶

In [ ]:

from pyramids.dataset import Dataset

for label, ps in (("CBERS-4 NDVI", paths), ("AMAZONIA-1 SR", amz_paths)):
    ds = Dataset.read_file(str(ps[0]))
    print(f"{label:18} bands={ds.band_count} epsg={ds.epsg} shape={ds.read_array().shape}")

from pyramids.dataset import Dataset for label, ps in (("CBERS-4 NDVI", paths), ("AMAZONIA-1 SR", amz_paths)): ds = Dataset.read_file(str(ps[0])) print(f"{label:18} bands={ds.band_count} epsg={ds.epsg} shape={ds.read_array().shape}")

See also¶

• Migration mapping for MODIS/061/MOD13Q1 / MYD13Q1 → bdc/mod13q1-6.1 / myd13q1-6.1 in docs/reference/stac/migration-from-gee.md.
• The BdcTokenSigner recipe lives in docs/reference/stac/authentication.md — wire it in via a per-collection signer: bdc-token override when INPE next exposes a token-gated tier.