ASF — quickstart (anonymous search)¶

Run a small Sentinel-1 SLC catalog search over a bounded bbox / time window. No credentials required — asf_search.geo_search (and everything backend._search() invokes for the search mode) is anonymous. EDL credentials are only needed for the download step.

Needs the [asf] extra (pip install earthlens[asf]).

Setup¶

Build the facade in search mode. The bbox is a small Iceland-rift-zone window where Sentinel-1 ascends every few days, so a short time window normally yields a handful of results.

In [ ]:

from earthlens.earthlens import EarthLens

el = EarthLens(
    data_source='asf',
    variables=['sentinel-1-slc'],
    start='2024-06-01',
    end='2024-06-15',
    lat_lim=[63.8, 64.4],
    lon_lim=[-21.7, -21.0],
    path='asf_out',
    flight_direction='ASCENDING',
    max_results=10,
)
el.datasource._mode

from earthlens.earthlens import EarthLens el = EarthLens( data_source='asf', variables=['sentinel-1-slc'], start='2024-06-01', end='2024-06-15', lat_lim=[63.8, 64.4], lon_lim=[-21.7, -21.0], path='asf_out', flight_direction='ASCENDING', max_results=10, ) el.datasource._mode

Dry-run inspection¶

backend._search() returns one RemoteProduct per matching ASFProduct. We can inspect it without downloading anything — useful for sizing a request before paying for the EDL login + bulk fetch.

In [ ]:

products = el.datasource._search()
print(f'{len(products)} product(s) match the request')
for p in products[:5]:
    print(f'  {p.id:62s} {p.metadata["fileName"]}')

products = el.datasource._search() print(f'{len(products)} product(s) match the request') for p in products[:5]: print(f' {p.id:62s} {p.metadata["fileName"]}')

How big is the request?¶

Each ASFProduct carries a bytes property — sum it to get the on-disk weight.

In [ ]:

total = sum(p.metadata['product'].properties.get('bytes', 0) for p in products)
print(f'Total download size: {total / 1e9:.2f} GB across {len(products)} product(s)')

total = sum(p.metadata['product'].properties.get('bytes', 0) for p in products) print(f'Total download size: {total / 1e9:.2f} GB across {len(products)} product(s)')

What's in metadata?¶

Inspecting the first product shows the metadata shape the backend records. In search mode the baseline keys (perpendicularBaseline, temporalBaseline) are intentionally omitted — they only exist on stacked products. See the InSAR stack notebook.

In [ ]:

first = products[0]
for key, value in first.metadata.items():
    print(f'  {key:24s}: {value if key != "product" else "<ASFProduct>"}')

first = products[0] for key, value in first.metadata.items(): print(f' {key:24s}: {value if key != "product" else "<ASFProduct>"}')

Downloading (gated on EDL credentials)¶

When you have EARTHDATA_TOKEN (or EARTHDATA_USERNAME + EARTHDATA_PASSWORD) set, the same backend can fetch:

paths = el.download()   # gated by EDL creds

The call returns a list[Path] of the downloaded SLC archives. Idempotent re-runs skip files already on disk. Credentials and the error path are documented under Authentication.