EC Raw Binary Format (CH-CHA)

See here for more info about the EC raw binary format.

Legend

An overview of used abbreviations can be found here: Abbreviations

Setup since 2005

Notes

• For QCL the version and variant according to Table 2 in WE’s sonicread.pdf is given.

(1) –
(2) QCL-A format: Modified data acquisition to also record QCL temperature and pressure. Now there are 5 QCL variables that are detected (there were 3 in previous versions).
(3) For this short time the QCL was run with the CH-DAV QCL setup. For more info see here: ch-cha-2016-data-issues.pdf
(4) QCL-A3 format: see GIN entry 26 Nov 2018. QCL was re-installed at site on 16 Nov 2018, data logging started on 26 Nov 2018. This period is a special case, since the QCL logged the variables in a different order. In addition, the gains on the signal are different. Also, H2O was not recorded at full resolution and looks rather static, but of realistic magnitude.
(5) The separation given for the QCL seems more or less correct, see this image.
(6) QCL is defective and was removed from the site on 11 Apr 2018 (factory repair). The QCL is still part of the raw data, but all its data are empty.
(7) The multiplier of 1000 got set back to 1 so that the QCL H2O concentrations could be saved correctly. Correct values were recorded since 13:32:27 CET 15.05.2020. File with the transition: 2020051513.C00. See GIN entries (internal) from 13 May 2020 and 15 May 2020. Direct link to info text file.
(8) Calibration gas issue: Due to the IRGA75 replacement of in 15 Mar 2017, only a short time period of approx. 2 months is affected by the wrong calibration gas. Affected time period: all IRGA75 CO2 concentrations between 2017011114.C57 and 2017031507.C00. For more info see here.
(9) After checking the raw data plots, the QCL seems defective after 29 Mar 2018 approx. since 15:34h, the  (most QCL sginals stop with record 184 706 instead of 431 990 in the 6-hour raw data file). Nearly half of the file 2018032913.C00 still contains usable data.
(10) Short period where the gain on the H2O signal from QCL-A4 was set to 1000 instead of 1 (by accident).
(11) See GIN entry from 13 May 2020 and 15 May 2020: Instead of the H2O concentration the ambient CH4 concentration got recorded in ppb. This happened due to a change in the incoming data from the QCL. The order of magnitude for the ambient CH4 is e.g. 13000 ppb. Note that this is very different from the dry mixing ratio that we use for flux calculations that is on the order of 2000 ppb.
(12) QCL removed from site 5 Feb 2020 10:50h.
(13) IRGA75 was calibrated on 30 Apr 2019 with correct calibration gas.
(14) Inlet location of QCL-A3: measured at field site field as Tube length (Inlet to QCL in box):  8.57 m, vertical distance of inlet to sonic volume: 35 cm, -5 cm eastward separation, +3 cm northward separation, -35 cm vertical separation, see this photo. I assume that the GA1 separation is the same as was later measured in Feb 2019. Here during this time period in 2018, the QCL (GA2) inlet was moved which means there was a setup change at the site. I assume that the setup change was also the case for GA1. It was not possible to find out more details from the fieldbook.
(15) Note that the file 2020051307.C00 is the transition file from QCL-A2 to QCL-A4.
(16) Note that the file 2020051513.C00 is the transition file from QCL-A4 to QCL-A.
(17) QCL removed from site on 20 Jul 2021.
(18) The LGR from the CH-INO project is now running instead of the QCL.
(19) LGR was removed from the site, but empty data are still part of the datastream.
(20) LGR data is now no longer part of the datastream.
(21) New sonic format: data from the sonic anemometer now includes StaA (status address) and StaD (status data). The format R350-B is described in the bico settings here and here.
(22) Raw binary files now have the file extension *.C*.
(23) Sonic orientation and separation: Info from GIN entry 28 Oct 2022. This info is (slightly) differrent than before, using the new info from this point in time onwards (for future time periods). It was not possible to determine exactly when the setup changed, but since the differrence to the previously reported setup is small, it should not really make a difference in flux calculations.
(24) QCL-D: At the time of writing (Apr 2024) it was not possible to find out the exact setup (sensor separations…) for the QCL. For flux calculations during this time period we use settings similar to the later QCL setups, e.g. during 2015_2 and 2019.
(25) Inlet location of QCL-B: The exact location of the QCL inlet was not documented, but it seems to be very similar to the setup in 2018. Compare the inlet location in this photo from 2012 and this photo from 2018. The location in 2018 was measured (see note (14) above). The main difference between the two years is that in 2012 the inlet was closer to the sonic center. Also in the following years the inlet was close to the main mast, but the vertical separation to the sonic center changed. Info about instrument path length and time response were taken from settings used in 2019.
(26) Sonic orientation/height: Should be 7° / 2.41m more or less and consistent across all years. In a comparison of histograms of wind directions between 2005 and 2023 showed that a sonic orientation of 7° offset to north yields very similar results across years all years. Info from one of the locations.table files that were used for documenting setup info in earlier years listed the sonic orientation at 0°. In 2016, 7° north offset was measured for the north spar, the sonic setup should be approx. the same across all years.
(27) QCL time lag is quite unclear during this time period and most likely around +7s for N2O and CH4.
(28) Wind offset in sonic anemometer data for vertical wind component W: W jumps to higher values and then remains stable around this higher value. This issue can be more or less corrected in EddyPro by defining an offset for the vertical wind measurements. The affected time periods are (inlcuding offsets, using middle timestamp for half-hours): OFFSET between 2023-04-17 10:15 and 2023-04-20 06:15: 1.163599708029197, OFFSET between 2023-07-21 15:45 and 2023-07-21 18:45: 1.1862112857142857, OFFSET between 2023-08-23 14:15 and 2023-08-24 18:45: 1.2251075862068965, OFFSET between 2023-08-26 16:45 and 2023-08-30 08:45: 1.0117830873920455 (during this time period, there seems to be another issue on 28 Aug 2023, but only for approx. one half day, otherwise this offset would also be approx. 1.2), OFFSET between 2023-09-14 00:45 and 2023-09-18 15:45: 1.187201864864865, OFFSET between 2023-09-22 00:45 and 2023-09-26 14:45: 1.1451817454545454, OFFSET between 2023-09-28 08:45 and 2023-09-29 14:15: 1.132388540677966. All these time periods should be calculated together in one run, using a mean offset of 1.173283.

 

Downloads

–

Last Updated on 2 Mar 2025 01:00