Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/terrafloww/rasteret/llms.txt

Use this file to discover all available pages before exploring further.

​
Method Signature

Collection.get_xarray(
    geometries,
    bands,
    *,
    max_concurrent=50,
    cloud_config=None,
    data_source=None,
    backend=None,
    target_crs=None,
    **filters
)
Load selected bands into an xarray Dataset with coordinate information and CF-compliant spatial metadata.

​
Parameters

​
geometries
bbox tuple, pa.Array, Shapely, WKB bytes, or GeoJSON dict
required
Area(s) of interest to load. Accepts (minx, miny, maxx, maxy) bbox tuples, Arrow arrays (e.g. from GeoParquet), Shapely objects, raw WKB bytes, or GeoJSON dicts.
​
bands
list of str
required
Band codes to load (e.g. ["B04", "B03", "B02"]).
​
max_concurrent
int
default:"50"
Maximum concurrent HTTP requests for fetching raster tiles.
​
cloud_config
CloudConfig
Cloud configuration for URL rewriting and authenticated access.
​
data_source
str
Override the inferred data source identifier.
​
backend
StorageBackend
Pluggable I/O backend for authenticated range reads. See create_backend().
​
target_crs
int
EPSG code to reproject all records to before merging.
​
**filters
kwargs
Additional keyword arguments passed to subset() for filtering scenes.

​
Returns

​
dataset
xarray.Dataset
Band arrays in native COG dtype (e.g. uint16 for Sentinel-2). CRS is encoded via CF conventions with a spatial_ref coordinate containing WKT2, PROJJSON, and GeoTransform. Multi-CRS queries are auto-reprojected to the most common CRS.

​
Examples

​
Basic Load

import rasteret

collection = rasteret.load("my_collection")

# Load bands into xarray Dataset
bbox = (-122.5, 37.7, -122.3, 37.9)
ds = collection.get_xarray(
    geometries=bbox,
    bands=["B04", "B03", "B02"]
)

print(ds)
# <xarray.Dataset>
# Dimensions:  (x: 2048, y: 2048, time: 10)
# Coordinates:
#   * x        (x) float64 ...
#   * y        (y) float64 ...
#   * time     (time) datetime64[ns] ...
#     spatial_ref int64 0
# Data variables:
#     B04      (time, y, x) uint16 ...
#     B03      (time, y, x) uint16 ...
#     B02      (time, y, x) uint16 ...

​
With Cloud Cover Filter

# Load only clear scenes
ds = collection.get_xarray(
    geometries=bbox,
    bands=["B08"],
    cloud_cover_lt=10
)

​
With Date Range

# Load summer imagery
ds = collection.get_xarray(
    geometries=bbox,
    bands=["B04", "B03", "B02"],
    date_range=("2023-06-01", "2023-08-31")
)

# Access temporal dimension
print(ds.time.values)

​
Multiple Geometries

import geopandas as gpd

# Load data for multiple fields
fields = gpd.read_file("fields.geojson")
ds = collection.get_xarray(
    geometries=fields.geometry,
    bands=["B08", "B04"]
)

​
With CRS Reprojection

# Force all data to UTM Zone 10N
ds = collection.get_xarray(
    geometries=bbox,
    bands=["B04"],
    target_crs=32610
)

​
Using with rioxarray

import rioxarray

ds = collection.get_xarray(
    geometries=bbox,
    bands=["B04", "B03", "B02"]
)

# Export to GeoTIFF
ds["B04"].rio.to_raster("red_band.tif")

​
Dataset Structure

The returned xarray Dataset includes:
  • Coordinates:
    • x: Easting/longitude coordinates
    • y: Northing/latitude coordinates
    • time: Acquisition timestamps
    • spatial_ref: CRS metadata (WKT2, PROJJSON, GeoTransform)
  • Data Variables: One per requested band, named by band code
  • Attributes: CF-compliant metadata

​
Notes

  • Data is in native COG dtype (typically uint16 for Sentinel-2, Landsat)
  • Multi-CRS collections are automatically reprojected to the most common CRS
  • Use target_crs to force a specific output projection
  • Compatible with rioxarray for geospatial operations

Build docs developers (and LLMs) love

Get started for freeTalk to us