From CedarWiki

Wuppertal OH Spectrometer (wup) Ingest

• Receive data from Peter Knieling in January each year as an ascii file labelled variously wohkz03.dat, wroh07.txt, etc
• Copy data into ~emery/cedar/wrdata/wup
• There are 2 wup .cbf files in /project/cedar/data: wup800723b.cbf and wup010521a.asc. They have approximately the same header, with a difference in contact person and a note that the data from 800723-980926 were revised with a better definition of SLT in May 2001. They are also located in ~emery/cedar/wrdata/wup in .cbf form and in .asc form (is2chr wupname.cbf > wupname.asc). The ascii versions have the 2-line catalog record created in the update ingest for additions removed. These will be added to (or replaced) as needed. If replaced, the letter at the end is increased by 1. If added to, the name remains the same.
• In early Jan 2008, received 2007 data (missing number of lines in first line, and also including for the first time standard deviation of OH temperature over the night), and later received revised values for 1995-2005 (with comma delimiters that could be deleted with Vi editor, but not needed with calc.f). Decision to delete wup800723b and wup010521a and replace both with wup800723c.
• Contacted Peter Knieling to find out about new 2007 'error bar' for OH temperature and to see if this could be provided for 1980-2006 data. If it is only available for 2007, then we will start a new wup070102a file with an additional parameter of -812 (error bar for 812 temperature) and a new kindat.
  • Only std error for 2007, but got 3 .rar compressed files (on Windows PC) of high resolution data (1.2-5 min or ~90 sec spectra in FK files) from 1987-2007.
    • Cannot read the .rar files, so asked on 7 March for ascii - got .zip which were fine (unzip -a file.zip)
  • FK files are the high-resolution files from which the mean nightly files are made, partly by hand. First years July 1980 to July 1983 are missing. Decision to replace wup800723b with c for these years of data only, and to delete wup010521a and replace with wup870509a and b where a is the FK 90-sec data and b is the calculated mean nightly data (100-300K) with a calculated standard deviation and more parameters.
    • ls y1987/* > list.1987 (etc for all years to get list of FK files)
• Wrote calc07.f and calc.f to read list of FK files to compare the FK calc mean and std, my calc mean and std, and that read in the nightly mean files.
  • Discrepancies revealed that the mean nightly files were typed in by hand from the FK file values. Found also some FK problems with %- before Tn not included in means.
  • Got list of daylight savings time in Germany to convert LT in FK files to UT (LT=UT+2h summer, LT=UT+1h winter)
  • FK files missing in 2007 after March - sent by Peter 17 March
  • 7 FK files missing in 2006 - sent by Peter (problem was 516 should be 515)
• Wrote wup_mean_header with new parameters and text and sent to Peter. Wrote wup_spectra_header for new 90-sec high-res data.
  • Revised calc.f to include 2007 (calc07.f) and to make DB format ascii mean nightly files and high-res files. Ran for 2004-2007.
    • Checked daylight savings time switch. Asked Peter if LT begin or end time.
      •  ?Problem 2007 0326 file (start 0325 - start summer) seems 1 UT early (same for 2004 0329 file (start summer 0328) - so revised calc.f