# Query open satellite data

Tilebox indexes searchable metadata for public Earth observation catalogs. Use metadata queries to find relevant observations before reading or downloading their image assets.

## Prerequisites

* You have a [Tilebox API key](/docs/authentication).
* You have installed the [Python SDK](/docs/sdks/python/install).

```bash
uv add tilebox shapely
```

## Select the Sentinel-2 catalog

Open the Sentinel-2 dataset and its Level-2A collection:

```python title="Python"
from tilebox.datasets import Client, field

client = Client()
sentinel2 = client.dataset("open_data.aws_earth.sentinel2")
collection = sentinel2.collection("L2A")
```

You can browse other open datasets and inspect their schemas in the [Tilebox Console](https://console.tilebox.com/datasets/open-data).

## Query observation metadata

Query by time and area of interest. The result contains metadata and asset references, but does not download image bytes.

```python title="Python"
from shapely import box

area = box(-109.05, 37.0, -102.05, 41.0)

scenes = collection.query(
    temporal_extent=("2025-10-01", "2025-11-01"),
    spatial_extent=area,
    filter=field("cloud_cover") < 10,
    show_progress=True,
)

print(scenes[["stac_id", "cloud_cover", "platform"]])
```

The result is an `xarray.Dataset`. Use regular xarray operations to sort or select observations:

```python title="Python"
latest = scenes.sortby("time").isel(time=-1)

print(latest.stac_id.item())
print(latest.cloud_cover.item())
```

## Next steps

[Create a Sentinel-2 RGB image](/docs/guides/datasets/access-sentinel2-data)

Read three COG windows and combine them into an RGB image.

[Querying data](/docs/datasets/query/querying-data)

Learn more dataset query patterns.