We’ve introduced a new feature in level-2 subsetting functionality (l2-subsetter, SAMBAH) that allows users to subset TEMPO data by specifying a coordinate name and numeric range.
This capability supports arbitrary variable names, making it especially useful for vertical subsetting, such as selecting data within specific altitude, pressure, or other vertical coordinate ranges, allowing you to easily extract slices of data along vertical dimensions.
New Feature: Vertical Range Subsetting for TEMPO Level-2 Data
-
ASDC - jennifer.tindell
- Subject Matter Expert

- Posts: 51
- Joined: Wed Mar 13, 2019 12:51 pm America/New_York
-
ASDC - jennifer.tindell
- Subject Matter Expert

- Posts: 51
- Joined: Wed Mar 13, 2019 12:51 pm America/New_York
Re: New Feature: Vertical Range Subsetting for TEMPO Level-2 Data
How it works:
Include the variable name and numeric bounds in your request, and the service will return only the corresponding subset of the original data.
Known limitations:
Include the variable name and numeric bounds in your request, and the service will return only the corresponding subset of the original data.
Code: Select all
from harmony import BBox, Client, Collection, Request, Dimension
harmony_client = Client()
collection_id = "C3685896287-LARC_CLOUD"
request = Request(
collection = Collection(id=collection_id),
spatial = BBox(-135, 36, -52, 43),
max_results = 3,
concatenate = True,
dimensions=[Dimension('/support_data/ozone_profile_altitude', 30, 40)]
)
assert request.is_valid()
harmony_client.submit(request) - You must provide a valid variable name and an appropriate numeric range. If the variable does not exist in the dataset or the range is invalid, the request will fail or the output file will be empty.
- This functionality is currently available only through programmatic access, such as via harmony-py or the Harmony API.
- Subsetting in SAMBAH applies masking instead of trimming: values outside the range are set to NaN, preserving the original array shape and potentially containing slices composed entirely of NaNs.