Tags :: STAC

• Landsat Collection 2 Level 2
  • Bands
  • Item Assets
  • Get Landsat acquistion time
• References

Landsat Collection 2 Level 2

Collection URL :: https://planetarycomputer.microsoft.com/api/stac/v1/collections/landsat-c2-l2

Providers ::

• NASA (producer, licensor)
• USGS (producer, processor, licensor)
• Microsoft (host)

License :: Public Domain

DOI ::

• Landsat 4-5 TM Collection 2 Level-2 : (10.5066/P9IAXOVV)
• Landsat 7 ETM+ Collection 2 Level-2 : (10.5066/P9C7I13B)
• Landsat 8-9 OLI/TIRS Collection 2 Level-2 : (10.5066/P9OGBGM6)

Bands

Item Assets

Get Landsat acquistion time

This shows how to get the acquistion time of Landsat observations for the AmeriFLUX tower, US-Akn + a 5000m bounding box.

Per the STAC Item specification the datetime property is the representative searchable date+time of the asset. In the case of Landsat (and most other EO sensors) this will be the acquisition time.

# stdlib imports
import datetime
import warnings

# third party imports
import geopandas as gpd
import pandas as pd
import planetary_computer
from pystac_client import Client

warnings.simplefilter(action="ignore", category=UserWarning)


# Query bbox
bbox = (
    -81.61905431253872,
    33.337402921165584,
    -81.51154503758427,
    33.42759674839017,
)


# Construct time query
start_date = datetime.datetime(2011, 1, 1)
end_date = datetime.datetime(2022, 12, 31)
date_query = [start_date, end_date]

# Open connection to planetary computer
client = Client.open(
    "https://planetarycomputer.microsoft.com/api/stac/v1",
    modifier=planetary_computer.sign_inplace,
)


# Search Landsat Collection 2 level 2
items = client.search(
    collections=["landsat-c2-l2"],
    bbox=bbox,
    datetime=date_query,
).item_collection()


# Marshal to geopandas and make sure datetime col is interpreted as such
df = gpd.GeoDataFrame.from_features(items.to_dict(), crs="epsg:4326")
df.datetime = pd.to_datetime(df.datetime, format="mixed")

print(df.datetime)  # or save to csv, whatever you want

The output will look something like the following;

• Localize

  It is important to note that the time will be in UTC+0.0. If you need to localize the time, this UTC GeoJSON contains the geometry of every UTC time zone and their offsets. A “quick” spatial merge will give you the offset. E.g.

  # Read in timezone data and join
  utz_timezone = gpd.read_file("utc_tz.geojson")
  df = gpd.sjoin(df, utz_timezone, how="inner")
  
  df["hr"] = df.datetime.dt.hour + df["name"].astype(int)
  print(df["hr"])
  

References

• Planetary Computer Landsat C2L2