Mooring data post-processing - Fram Strait guidelines

Note

The content below was parsed from the document mooring_data_NPI_June2024.docx, last updated by L. de Steur 13.06.2024 (formatting changes applied when transferring to Markdown).

On /script_files and data structure

The .m script files require editing to include the right metadata of the mooring, the deployment, and the instrument and creates a plot to take a first look at the quality of the data per instrument.

For ADCPs and CMs one applies the correction for magnetic declination here (call to igrf.m).

For SBE37s or other CTD types there is a running_mean low-pass filter that removes spikes that are e.g larger than 3 or 4 x std within the one month window (can be edited per case).

After an initial look of the figure one can decide to do extra despiking, post-corrections for e.g S drift, or for removing bad pieces of data (all requires manual coding then per script file per instrument and for each unique SN (I just apply an if SN == xx then do so and so... end).

The processed files are stored in dir /processed/mooringID. The metadata and data are stored in the .mat file as a matlab structure data (structure with fields) as follows:

data =

            cruiseID: 'FS2023'

            mooringID: 'F10'

            latitude: 78.8273

            longitude: -1.9513

            instrument: 'Aquadopp'

            sn: 15888

        instrument_depth: '225 m'

            bottom_depth: '2650 m'

            time_start: [2022 9 17 23 0 0]

            time_end: [2023 9 8 6 0 0]

        sample_interval: '3600 s'

            U: [8528×1 double]

            V: [8528×1 double]

            P: [8528×1 double]

T: [8528×1 double]

            time: [8528×1 double]

            time2: [8528×1 datetime]

comment: [ ]

Here, one can add text into the comment on what one has been done (extra) and specific to that data set of that instrument, e.g despiked, corrected for mag.decl. and/or corrected for drift.

---

Attention

Finally: these are not final publishable files but are considered working files. As one works with it, it can still be discovered these data need extra processing, corrections or Q control. Once everything looks good for real and once used in paper publication, these .mat structure files are converted to .cnv using Yannick’s NetCDF-Creator app and published under an entry on data.npolar.no.