# DEA Water Observations (Sentinel-2 Near Real-Time, Provisional, Australia)

## Facts

- Provider: Geoscience Australia (Digital Earth Australia)
- License: CC-BY-4.0 (https://knowledge.dea.ga.gov.au/data/product/dea-water-observations-sentinel-2-nrt/)
- Format: JSON, JPEG (in sampled files)
- Stored at: Amazon S3, s3://dea-public-data/derivative/ga_s2_wo_provisional_3 (ap-southeast-2)
- Access: public, readable without an account
- DOI: https://doi.org/10.26186/146257
- Category: Earth Observation & Geospatial
- Source: https://doi.org/10.26186/146257
- Page: https://scigantic.com/archive/49bc02b1-6aec-57b8-af5f-79f6efc4e872
- Keywords: Earth Observation & Geospatial, satellite imagery, JSON, JPEG, AWS Open Data

## About

Per-scene surface water maps for Australia from Sentinel-2A and 2B, produced by Digital Earth Australia within 24 hours of each overpass by applying the Water Observations (WOfS) classifier to Sentinel-2 near-real-time surface reflectance. Each 10 m pixel is flagged wet or dry, together with bit flags for cloud, cloud shadow, terrain shadow and sensor problems that made it unobservable. The product is provisional and not validated; it is intended for flood extent mapping and emergency management soon after an event, complementing the validated Landsat Water Observations series. The product page describes a rolling three-month window, while the DEA STAC catalogue still indexes 541,954 scenes from 2021 onward. Checked 2026-11-16: version 0-0-1 is laid out by MGRS UTM zone, tile, acquisition date and processing timestamp (e.g. 0-0-1/54/LWJ/2021/07/01/20230223T125452/), each scene holding a single-band *_nrt_water.tif bit-flag Cloud Optimised GeoTIFF (10980 x 10980 pixels), thumbnail and STAC/ODC metadata. Load with odc-stac or rioxarray and decode the bit flags as in the DEA Water Observations notebooks. Bucket region ap-southeast-2. Licensed CC BY 4.0 by Geoscience Australia, containing modified Copernicus Sentinel data; cite DEA Water Observations, doi:10.26186/146257, and Mueller et al. (2016).

## Load the data

```python
import boto3
from botocore import UNSIGNED
from botocore.client import Config
s3 = boto3.client('s3', region_name='ap-southeast-2', config=Config(signature_version=UNSIGNED))
s3.list_objects_v2(Bucket='dea-public-data', Prefix='derivative/ga_s2_wo_provisional_3/', Delimiter='/')
```

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

## Data structure

File types in sampled files: JSON, JPEG

Example file: `derivative/ga_s2_wo_provisional_3/0-0-1/49/HFA/2026/02/07/20260207T053419/ga_s2_wo_provisional_3_49HFA_2026-02-07_nrt.odc-metadata.yaml`

Path pattern: `{prefix}/{seg1}/{seg2}/{seg3}/{seg4}/{YYYY}/{seg6}/{seg7}/{seg8}/{filename}`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/49bc02b1-6aec-57b8-af5f-79f6efc4e872
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/49bc02b1-6aec-57b8-af5f-79f6efc4e872
- Over MCP: `get_schema_card` with id `49bc02b1-6aec-57b8-af5f-79f6efc4e872` at https://api.scigantic.com/mcp

## Related topics

- [Earth Observation & Geospatial datasets](https://scigantic.com/topics/earth-observation-and-geospatial)
- [JSON datasets](https://scigantic.com/topics/json)
- [Satellite imagery datasets](https://scigantic.com/topics/satellite-imagery)
- [JPEG datasets](https://scigantic.com/topics/jpeg)
