Why it matters:
When data collections use different internal time standards (such as Coordinated Universal Time (UTC), Global Positioning System (GPS), or International Atomic Time (TAI)), a user request to subset data between 2025-01-01 00:10:00 and 2025-01-01 00:13:00 will yield subsets with varying systematic time offsets, depending on each collection's time standard.
For example:
TEMPO collections use GPS time standard, expressed as 'seconds since 1980-01-06T00:00:00Z', leading to systematic offset of 18* seconds compared to UTC due to the absence of leap second adjustments in GPS time. As a result, a request above will return the data subsetted between 2025-01-01 00:09:42 and 2025-01-01 00:12:42 in UTC time.
*The current offset as of July 2025 is 18 seconds, but the time offset between GPS and UTC varies over time.
PREFIRE collections use a TAI-like time standard with a custom epoch of "seconds since 2000-01-01 00:00:00Z," leading to systematic offset of 5 seconds compared to UTC due to the number of leap second adjustments since 2000. As a result, a request above will return the data subsetted between 2025-01-01 00:09:55 and 2025-01-01 00:12:55 in UTC time.
Graphical representation:
Subsetting TEMPO data
- Example_L2ss_Subsetting_Of_TEMPO_Data.png (409.53 KiB) Not viewed yet
Subsetting PREFIRE data
- Example_L2ss_Subsetting_Of_PREFIRE_Data.png (415.79 KiB) Not viewed yet
L2ss is used in the following situations:
Programmatically, through the Harmony API when using L2ss-py
In Earthdata Search, when using the Harmony services: PODAAC Level 2 subsetter or Subsetter And Multi-dimensional Batched Aggregation in Harmony (SAMBAH)
Additional information on L2ss-py may be found in its GitHub repository: https://github.com/podaac/l2ss-py