Skip to content

Develope obsForge utilities #151

@ilianagenkova

Description

@ilianagenkova

Develop a number of obsForge utilities for the IODA (netcdf and odb/odc formatted) observation files:

A) Develop tools that mimic the functionality of the NCEPLIB-bufr tools:
https://github.com/NOAA-EMC/NCEPLIBS-bufr/tree/develop/utils
A1) sinv - print an inventory of unique satellite IDs (along with their names in characters) found in the IODA files, and the number of observation from each satellite
A2) binv - print inventory of tanks which populated the bufr dump and now the IODA files, and the number of observation from each tank
A3) split_by_subset - split the IODA files to subsets, which represent observation from a single tank
A4) (no analogos tool for bufr) - write a tool that can exract an observation based on it's index or range of indices
A5) debufr - write out the IODA file as ascii table; provide configuration option to write out only certain mnemonics/variables

B) Develop a tool that mimmic the current go_chkdat and minv tools, i.e. extract the earliest and latest observation times from the IODA file

C) Develop reporting of observation counts - daily and monthy, similar to the "status" files from the legacy obsproc, for use by NCO and plotting on the Real Time Data Monitoring (RTDM) website. Coordinate with NCO on the file formatting (the current reports are parsed by grep/sed/awk)

Notes:

  1. obsForge in GFS v17 will not preserve tank info in the IODA files. We will lean on the bufr dump at times for diagnostics
    obsForge in GFS v18 (JEDI) will incorporate tank info in the IODA files - A2) and A3) need to wait until that time.
  2. A and B are smaller stand alone tools. C, on the other hand, is a larger effort that needs to be coordinated with NCO/SDM at some point.

Update: 1)Dan requested that the tools are capable of handling IODA netcdf and ODB/ODC data formats https://github.com/ecmwf/odc
2)Use the APIs Ron has developed to authomaticlaly recognize and read the obs files (wether it's bufr, netcdf, odb) and generate counts for the report

Metadata

Metadata

Labels

No labels
No labels

Type

No type

Projects

Status

No status

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions