Kristen,
OK, there two issues at play here. The first is a known feature of the system but one that isn’t well advertised. When data from a co-located instrument is required to calculate a parameter in the dataset of interest (in this case temperature, salinity and pressure to correct the DO measurement and convert to umol/kg), the system will grab the data and those instrument records will be added to the THREDDS catalog. However, they will be incomplete. You can see an earlier post on this issue here: Missing variables in data sets. As a rule of thumb, I only use the data from the sensor of interest in each catalog, ignoring any other instrument files as they are often incomplete. Instead of getting the CTDBP data from the DOSTA directory, you need to get that data from the CTDBP directory.
For the DO data, explicitly use Catalog http://thredds.dataexplorer.oceanobservatories.org/thredds/catalog/ooigoldcopy/public/CE04OSSM-RID27-04-DOSTAD000-recovered_host-dosta_abcdjm_dcl_instrument_recovered/catalog.html and the *DOSTA*.nc files in there.
For the CTD data, explicitly use Catalog http://thredds.dataexplorer.oceanobservatories.org/thredds/catalog/ooigoldcopy/public/CE04OSSM-RID27-03-CTDBPC000-recovered_host-ctdbp_cdef_dcl_instrument_recovered/catalog.html and the *CTDBP*.nc files in there.
The second issue, with the missing QARTOD variables, looks to be an unintended consequence of the changes of migrating to QARTOD for QC processing and the parameter name changes. We are going to need to dig into that further to better understand where the disconnect is and how to correct it. I’m going to open a ticket in our internal system so we can start working on this.
Cheers,
Chris