Preparing for next round of GMAG metadata updates

[DCarSunDr]

There are a couple things we're waiting on before launching a new round of GMAG mastercdf metadata updates:

• There is a slight discrepancy between the recorded station locations between what was listed in the UCLA version of the station metadata spreadsheet and what we have in our mastercdfs, for the following stations (TXT values are from the CDFs):
  BENN
  PDF: Lat 41.360, Lon 262.840
  TXT: Lat 41.36, Lon 263.84 (Longitude difference of exactly 1 degree)
  CCNV
  PDF: Lat 39.190, Lon 240.220
  TXT: Lat 39.19, Lon 240.20 (Longitude difference of 0.02)
  EKAT
  PDF: Lat 64.717, Lon 250.667
  TXT: Lat 64.73, Lon 249.33 (Latitude differs by ~0.013, Longitude differs significantly by 1.337)
  HOTS
  PDF: Lat 47.590, Lon 245.340
  TXT: Lat 47.61, Lon 245.33 (Latitude differs by 0.02, Longitude differs by 0.01)
  RICH
  PDF: Lat 32.980, Lon 262.250
  TXT: Lat 32.98, Lon 263.25 (Longitude difference of exactly 1 degree)

  However, the CDF station location values appear to have remained unchanged at least since 2013. So we need to:

  • Confirm which set of coordinates is the most accurate
  • Add the confirmed station locations to the centralized GMAG station metadata dictionary (see pyspedas issue #1352
  • If the CDF versions of the station locations are incorrect, we need to generate an update list and include them in the update

• For the variometers, we need to:

  • We need to fully implement variometer data retrieval improvements AND
  • Backfill 10 Hz data to present so SPDF can readily locate the data files for ingestion (see Optimizing USGS Variometer data retrieval #392)
  • Request confirmation that the variometer metadata looks good.

• Finally, when all of this is done, we need to:

  • Let SPDF know about any changes/corrections, and ask if they've finished reviewing the variometer metadata.
  • Confirm SPDF is ready for ingestion of updated files
  • Once we get approval, run the update script using the update list file. The update script already should perform verification and update of the masters before finding and handling the data CDFs.

The URL update (see #310) file should be ready to go, so it just needs any additional updates/sites added from the other tasks.

[DCarSunDr]

BENN: the txt value from the CDF appears correct, according to this paper:

LAT:  41.36
LON: 263.84

CCNV: CDF appears correct, according to this site

LAT:  39.19
LON: 240.20

EKAT: CDF appears correct, according to this site. This one is through SPDF, so it might be pulling metadata from a related file. Checking satellite images, this location looks correct, since it's supposed to be located at the Ekati diamond mine, and these are only coordinates which show the mine nearby.

LAT:  64.73
LON: 249.33 

HOTS: the PDF is correct and the CDF metadata is incorrect, according to this site

LAT:  47.59
LON: 245.34

RICH: the txt value from the CDF appears correct, according to this paper:

LAT:  32.98
LON: 263.25

The HOTS metadata needs to be corrected and scheduled for update