search

aodn / content

Tracks AODN Portal content and configuration issues
0 stars 0 forks source link

Biogeochem timeseries datasets not available from portal #398

Open ggalibert opened 5 years ago

ggalibert commented 5 years ago

For example at NRSYON, the bottom WQMs have been replaced by SBE16s which do not collect data in bursts. The chl-a an turbidity data from SBE16 is not found in the usual "ANMN WQM and CTD burst average data" collection!

mhidas commented 5 years ago

This was previously reported as https://github.com/aodn/harvesters/issues/473, but I think this is a better place for it.

Having a collection defined by the sampling regime of the instruments is not very helpful. A more generic collection including all mooring timeseries would be the best (IMHO).

Also, there's been some discussion of doing the burst averaging in the Toolbox, so that the FV01 files would already be averaged and more similar to the non-burst-sampled data.

ggalibert commented 5 years ago

Also, there's been some discussion of doing the burst averaging in the Toolbox, so that the FV01 files would already be averaged and more similar to the non-burst-sampled data.

Not clear agreement from everyone on this yet.

A more generic collection including all mooring timeseries would be the best (IMHO)

Agreed, implementing the moorings products identified at the last long timeseries product working group might encourage us to go down that path.

mhidas commented 5 years ago

Here's another example: http://thredds.aodn.org.au/thredds/catalog/IMOS/ANMN/NRS/NRSROT/Biogeochem_timeseries/catalog.html?dataset=IMOS/ANMN/NRS/NRSROT/Biogeochem_timeseries/IMOS_ANMN-NRS_EKTUZ_20180816T080000Z_NRSROT_FV01_NRSROT-1808-ACLW-USB-24_END-20181214T023003Z_C-20181217T041212Z.nc

Slightly different in that it does seem to be a burst-sampling instrument (4x1s every 15min), but the metadata describing the sampling regime are missing (except for instrument_sample_interval, which is 0!).

ggalibert commented 4 years ago

When the hourly product is on the portal and data harvested in database for CSVs, we could decomission this collection yet keep the burst avg files on THREDDS.

lbesnard commented 3 years ago

@ggalibert is this still current?