There may be missing observations in the prior lake data ingested in Hydrocron

[nikki-t]

It looks like there are lake identifiers with missing observations and in some cases the entire time series is missing.

It looks like there may be 710,684 lake identifiers that are missing 90% or more of SWOT observations.

Initial investigation:

• In some cases this is due to the Load Granule Lambda function timing out which has been fixed with the most recent release.
• An interesting case was for lake id 4520356022 and shapefile SWOT_L2_HR_LakeSP_Prior_010_202_AS_20240201T045215_20240201T050127_PIC0_01.zip, the shapefile was ingested into Hydrocron but the reach does not show up in the DynamoDB table. This is confirmed in both CloudWatch logs for the Load Granule Lambda and Track Ingest Lambda.
• There was also a case where I could not locate the shapefile in the Load Granule Lambda logs, so maybe it wasn't ever submitted?

Next steps:

• Copy a few shapefiles over to SWOT-UAT to do some testing in the SIT environment.
• Scope out issue and work during PI planning.