CH-DAV FP2021 (1997-2020)
This version is deprecated. An updated version of this dataset is available here: CH-DAV FP2022 (1997-2022.09)
Contents
Flux Products
For an explanation of variables in output files, please see Variable Abbreviations and ReddyProc Data Output Format
FP2021.2 (recommended)
- Date: 24 Mar 2022
- Identifier:
ID20220324003457
- File:
CH-DAV_FP2021.2_1997-2020_ID20220324003457_30MIN.csv.zip
(timestamp shows END)CH-DAV_FP2021.2_1997-2020_ID20220324003457_30MIN.diive.csv.zip
(same data, but timestamp shows MIDDLE)
- Description:
- PI flux dataset using a constant USTAR threshold for all years (
CUT
=0.287644
), MDS gap-filling and partitioning done inReddyProc
- PI flux dataset using a constant USTAR threshold for all years (
- Difference to FP2021.1:
- Updated VPD values at the start of July 2020
- Due to updated VPD values, gap-filling and partitioning for 2020 was redone, all other years remain unchanged
- Included variables: NEE, GPP, Reco, TA, SW_IN, VPD and auxiliary output
- Recommended flux variables:
- gap-filled:
NEE_CUT_f
,GPP_DT_CUT
,Reco_DT_CUT
,LE_f
,ET_f
- It is recommended to use GPP and Reco from the daytime partitioning method (
_DT
), because daytime fluxes are generally more reliable across years. - measured only (not gap-filled, highest quality):
NEE_CUT_QCF0
,LE_QC0
- gap-filled:
FP2021.1 (deprecated)
- Date: 8 Oct 2021
- Identifier:
ID20211008121006
- File:
CH-DAV_FP2021.1_1997-2020_ID20211008121006_30MIN.csv.zip
- Description: Initial release
- Release candidates:
- rc1: CH-DAV_FP2021.1.rc1_1997-2020_ID20211005003402_30MIN.csv.zip (5 Oct 2021; initial rc)
- rc2: CH-DAV_FP2021.1.rc2_1997-2020_ID20211008121006_30MIN.csv.zip (8 Oct 2021; +separate cols for directly measured fluxes of highest quality, +QCF flags, +USTAR) –> approved
Plots
Description
- Ecosystem fluxes over 24 years from three different IRGAs and sonic anemometers
- This is the second version of the CH-DAV PI dataset. The first version can be found here, but note the differences in post-processing.
- Differences to the previous version of the CH-DAV PI dataset:
- Usage of a constant USTAR threshold, same across years (before: separate, variable USTAR thresholds for each year)
- Gap-filled NEE data was calculated from daytime NEE data with quality flags 0 and 1, and from nighttime NEE data with quality flag 0 (highest quality) (before: also nighttime NEE of quality 1 were included)
- Due to the use of highest quality nighttime fluxes there are more gaps to be filled during nighttime, and nighttime respiration in gap-filled NEE is significantly higher
- The IRGA75 middle years (2005-2016) were corrected for self-heating using a refined approach similar to Kittler et al. (2017). This increased nighttime respiration, while daytime fluxes remained mostly unchanged (minor differences to before).
- Calculated budgets are now closer to the FLUXNET estimates
Key Stats
- Best estimate cumulative carbon uptake (1997-2020): 3.1 kg C m-2 (
NEE_CUT_f
), avg. 128 g C m-2 yr-1 - Available directly measured, highest-quality fluxes:
- directly measured fluxes with quality flag
QCF
= 0 (no gap-filling) NEE_CUT_QCF0
: 123792 values (22.0% of potential values) [1], [2], [3]LE_QCF0
: 97010 values (23.1% of potential values) [1], [3]
- directly measured fluxes with quality flag
- Data basis for budget calculations:
NEE_CUT_f
: 30.1% measured / 69.9% gap-filled [1], [2], [4]LE_f
andET_f
: 59.8% measured / 40.2% gap-filled [1], [5]
[1] … after all quality checks, including outlier removal
[2] … after USTAR
threshold application
[3] … QCF
= 0 (daytime and nighttime)
[4] … measured data include quality flag QCF
= 0 and QCF
= 1 (OK quality) for daytime data, and QCF
= 0 for nighttime data
[5] … measured data include quality flag QCF
= 0 and QCF
= 1 for daytime and nighttime data
Dataset Production
For info about the flux processing chain and flux levels see:: Flux Processing Chain
Software
EddyPro
v6 and v7 for flux calculations (links to details of Level-1 fluxes are given below for each year)- bico for the conversion of binary raw data files to ASCII (2013-2016, 2020)
- fluxrun for the flux calculation using EddyPro (2013-2016, 2020)
- Various versions of FCT (flux calculation using EddyPro) were used for years 1997-2004 and 2017-2019, details are given in the section Used Flux Versions below
- scop v0.1 (self-heating correction for open-path IRGAs) for the self-heating correction of IRGA75 fluxes
- diive v0.21.0 (legacy version) for file merging, quality control, storage correction, outlier removal
ReddyProc
v1.2.2 for detection of the constant ustar threshold, MDS gap-filliing and partitioning, in R Studio v1.3.959
Processing Steps
For an overview of the general processing chain see here: Flux Processing Chain
Steps to create a flux dataset comprising CO2 flux (NEE), LE and H. Note that H2O flux and ET are not prepared, only LE is used from the water fluxes. ET is later calculated (converted from LE) in ReddyProc.
- Level-1 / Data collection of most recent fluxes
- Separate for each IRGA
- Collection of all available Level-1 flux calculations for IRGA62, IRGA75 and IRGA72 fluxes
- Level-1 / Merging
- Separate for each IRGA
- Fluxes for each IRGA were merged, one separate file for each IRGA, yielding three datasets: IRGA62, IRGA75 and IRGA72
- Level-1 / Exclusion of time periods
- Separate for each IRGA
- Some time periods were exlcuded from the dataset due to data issues
- IRGA62 (2001): CO2 flux data between 27 Feb 2001 and incl. 23 Mar 2001 was excluded from the dataset. See here for more information: CH-DAV: 2001
- Level-1.1 / Self-heating correction for IRGA75
- Only for IRGA75
- using the script SCOP
- Level-2 / Quality Flag Expansion
- Separate for each IRGA
- Quality flags were expanded for CO2 flux, LE and H
- No flux outliers were flagged in this step, no absolute flux limits applied. This is done later after the storage correction.
- Step 1:
- Expanded quality flags (
QCF
) were created for each IRGA, based on the merged dataset of the respective IRGA. QCF for fluxes: CO2, LE and H. For example, the quality flags for IRGA72 were created for all years from the merged IRGA72 dataset. - Used quality flags: SSITC, spectral correction factor, spikes (raw data), drop-out (raw data), absolute limits (raw data; except for IRGA72 LE), AGC (only for IRGA75 CO2 and LE); the SIGNAL_STRENGTH normally used for IRGA72 was not applied, because the signal strength was satisfactory, except for one day (28 Oct 2018) where the signal dropped off, but the respective values were already removed by other quality flags.
- QCF: 0=best quality data, 1=OK quality for long-term budgets, 2=bad data, do not use
- Expanded quality flags (
- Step 2:
- Flux values where
QCF = 2
(bad data) were removed from the datasets. More precisely, all flux values whereQCF = 0
orQCF = 1
were retained, meaning that a quality flag with these values must be available for the respective flux values. This step is important for the subsequent merging of the different, separate IRGA datasets, because for successful merging datasets with real gaps where (bad) values are missing are needed. The suffix_QC01
marks that only fluxes of quality 0 and 1 are available in the respective data columns. - The same was done with the QCF flags for CO2 flux, LE and H: flags were only kept in the dataset if the corresponding half-hourly flux was also available.
- Flux values where
- Level-3.1 / Storage Correction
- Separate for each IRGA
- The storage term was added to CO2 flux, LE and H
- Level-3.1 / Merging of IRGA datasets
- sCombine the 3 IRGA datasets (IRGA62, IRGA75 and IRGA72) into one dataset
- Data from the IRGA62 dataset have the suffix
_IRGA72
, data from the IRGA75 dataset have the suffix_IRGA75
, and data from the IRGA72 have the suffix_IRGA72
. - Fluxes (NEE, LE, H) from the 3 IRGAs were then collected in one single column, marked by the suffix
_IRGA7572
. - For the merging, the IRGA72 and IRGA62 fluxes were used as the starting dataset. Then, the IRGA75 fluxes were added, filling data gaps in the time series with IRGA75 values.
- In the time period where IRGA75 and IRGA72 fluxes overlap (2013-2016), priority was given to IRGA72 fluxes, but IRGA75 fluxes were used to fill gaps in the IRGA72 fluxes whenever possible. This was done not only for the fluxes, but also for the QCF data. This way all data points should have their respective correct data flag, even though there is this mixing of IRGAs with sometimes different
QCF
criteria. - After checking the full time series, the IRGA72 water fluxes in 2013 and 2014 were clearly lower than from the IRGA75 during the same time period. It is possible that during these very first IRGA72 years H2O fluxes were not perfect. Therefore, water data from IRGA72 (LE and its quality flag) were removed and not used in subsequent steps. Instead, the IRGA75 data during the same time period was used.
- Level-3.2 / Outlier Removal: Absolute Flux Limits and Despiking
- For all IRGAs combined
- Proceeding with NEE and LE (no H)
- NEE:
- Step 1: Application of absolute flux limits
- Note that this step was done before the application of the Hampel filter in the next step, so that the Hampel filter has a similar data basis (data range) for all years. The IRGA75 fluxes show generally considerably more noise than IRGA62 and IRGA72 fluxes.
- daytime: values outside a physically plausible range of ±50 µmol CO2 m-2 s-1 were considered outliers
- nighttime: values outside a range between +30 and -5 µmol CO2 m-2 s-1 were considered outliers. The upper limit was chosen based on the range of highest quality nighttime NEE, which is mostly below +20 µmol m-2 s-1 during nighttime. The lower limit was chosen to remove strong nighttime carbon uptake spikes which are deemed implausible due to the absence of radiation.
- Step 2: Despiking filter
- daytime and nighttime (in combination): using a Hampel filter using the median absolute deviation (MAD) in a running time window of 432 records (9 days). The limit above which a NEE data point was defined as an outlier was set at 4 sigmas (NEE). The filter were applied multiple times until all outliers were removed from the dataset.
- Step 1: Application of absolute flux limits
- LE:
- Step 1: Application of absolute flux limits
- Note that this step was done before the application of the Hampel filter in the next step, so that the Hampel filter has a similar data basis (data range) for all years. The IRGA75 fluxes show generally considerably more noise than IRGA62 and IRGA72 fluxes.
- daytime: values outside a range of +800 and -40 W m-2 were considered outliers. The limit was chosen because the vast majority of highest quality LE fluxes falls into this range: 95th percentile = +269 W m-2, 5th percentile = +8 W m-2.
- nighttime: LE was trimmed, removing the upper and lower 3% of data, translating to a non-outlier range of between +107 and -52 W m-2. The upper limit was chosen based on the range of highest quality nighttime LE, which is mostly found in this range during nighttime: 95th percentile = +65 W m-2, 5th percentile = -15 W m-2.
- Step 2: Despiking filter
- daytime and nighttime (in combination): using a trimmed mean approach, removing the upper and lower 0.15% of data. In addition, a Hampel filter was applied with 432 days and 10 sigmas
- Step 1: Application of absolute flux limits
- Level-3.3 / USTAR threshold
- Application of a constant USTAR threshold of
CUT = 0.287644
to NEE data, values below this threshold were rejected. - This is the same constant threshold as in the FLUXNET Warm Winter 2020 dataset.
- Threshold detection using
ReddyProc
yielded similar results.
- Application of a constant USTAR threshold of
- Level-4.1 / Gap-filling
- Important:
- Data basis for gap-filling:
- Data from Level-3.3
- For NEE daytime, data of highest quality (
QCF = 0
) and data of OK quality (QCF = 1
) were kept in the dataset to maximize availability of directly measured data (following CarboEurope recommendations). This is a notable difference to the FLUXNET approach where only highest quality data are retained. - For NEE nighttime, only data of highest quality (
QCF = 0
) were retained. - For LE daytime and nighttime, data of highest quality and OK quality were used.
- Lowest quality data (
QCF = 2
) were rejected in all cases.
- Data basis for gap-filling:
- Gap-filling for NEE and LE was done using the MDS method described in Reichstein et al. (2005).
- ET was calculated from gap-filled LE.
- Important:
- Level-4.2 / Partitioning
- Recommended: NEE partitioning was done using the daytime approach described in Lasslop et al. (2010).
- In addition, NEE partitioning using the nighttime method described in Reichstein et al. (2005) was also done. However, we currently recommend to use the daytime partitioning.
Source Data
- Using data from the Non-ICOS Setup 1997-2016 and ICOS Setup since 2014 (ETH binary files), for an overview of raw data see here: EC Raw Binary Format (CH-DAV)
Flux Data (Level-1)
1997-2005: R2-IRGA62
- 1997 R2-IRGA62_FF-201606 / Level-1_ID2016-07-27T220640 [unchanged from previous dataset FP2020]
- 1998 R2-IRGA62_FF-201606 /Level-1_ID2016-07-27T220629 [unchanged]
- 1999 R2-IRGA62_FF-201606 /Level-1_ID2016-07-27T220619 [unchanged]
- 2000 R2-IRGA62_FF-201606 /Level-1_ID2016-07-27T220606 [unchanged]
- 2001 R2-IRGA62_FF-201606 /Level-1_ID2016-07-27T221217 [unchanged]
- 2002 R2-IRGA62_FF-201606 /Level-1_ID2016-07-27T220552 [unchanged]
- 2003 R2-IRGA62_FF-201606 /Level-1_ID2016-07-27T220527 [unchanged]
- 2004 R2-IRGA62_FF-201606 /Level-1_ID2016-07-27T220516 [unchanged]
- 2005 R2-IRGA62-IRGA75_FF-201606 /Level-1_ID2016-07-28T143836 (IRGA62 fluxes) [unchanged]
2005-2016: R350-IRGA75
Note that the Level-1 fluxes shown here were not corrected for IRGA self-heating, the correction is done in a separate step (Level-1.1). The flux version shown in this list are Level-1 calculations without (i.e. before) the self-heating correction.
- 2005 R2-IRGA62-IRGA75_FF-201606 / Level-1_ID2016-07-28T143836 (IRGA75 fluxes) [unchanged]
- 2006 R2-R350-IRGA75_FF-201606 / Level-1_ID2016-06-03T151158 [unchanged]
- 2007 R350-IRGA75_FF-201606 / Level-1_ID2016-06-01T145942 [unchanged]
- 2008 R350-IRGA75_FF-201606 / Level-1_ID2016-05-30T161917 [unchanged]
- 2009 R350-IRGA75_FF-201606 / Level-1_ID2016-05-30T161726 [unchanged]
- 2010 R350-IRGA75_FF-201606 / Level-1_ID2016-05-29T160744 [unchanged]
- 2011 R350-IRGA75_FF-201606 / Level-1_ID2016-05-29T160603 [unchanged]
- 2012 R350-IRGA75_FF-201606 / Level-1_ID2016-06-06T173342 (IRGA75 fluxes) [unchanged]
- 2013 R350-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-01T120000 (IRGA75 fluxes) NEW CALCULATIONS
- 2014 R350-HS50-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-02T120000 (IRGA75 fluxes) NEW CALCULATIONS
- 2015 R350-HS50-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-03T230000 (IRGA75 fluxes) NEW CALCULATIONS
- 2016 R350-HS50-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-02T230000 (IRGA75 fluxes) NEW CALCULATIONS
2013-2020: HS50-IRGA72
- 2012 R350-IRGA75_FF-201606 / Level-1_ID2016-06-06T173342 (IRGA72 fluxes) [unchanged ]
- 2013 R350-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-01T120000 (IRGA72 fluxes) NEW CALCULATIONS
- 2014 R350-HS50-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-02T120000 (IRGA72 fluxes) NEW CALCULATIONS
- 2015 R350-HS50-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-03T230000 (IRGA72 fluxes) NEW CALCULATIONS
- 2016 R350-HS50-IRGA72-IRGA75_FF-202101 / Level-1_ID2021-05-02T230000 (IRGA72 fluxes) NEW CALCULATIONS
- 2017 HS50-IRGA72_FF-201902 / Level-1_ID2019-02-25T202725 [unchanged ]
- 2018 HS50-IRGA72_FF-201902 / Level-1_ID2019-03-02T164252 [unchanged ]
- 2019 HS50-IRGA72_FF-202005 / Level-1_ID2020-05-22T122418 [unchanged ]
- 2020 HS50-HS100-IRGA72_FF202101 / Level-1_FR-20210323-113925 NEW CALCULATIONS, ADDITIONAL YEAR
Meteo Data
NABEL (empa) data (1997-2020) was merged with meteo data from the FLUXNET Drought Study dataset (1997-2020.06)
TA
(NABEL, 35m, prioritized) merged withTA_F
(FLUXNET)SW_IN
(NABEL, 35m, prioritized) was merged withSW_IN_F
(FLUXNET)RH
(NABEL, 35m, prioritized) was merged withRH
(FLUXNET)VPD_F
(FLUXNET)
Upload for FLUXNET / EFDC Data Sharing
CH-DAV / EFDC-ID20210630
- Updated fluxes for FLUXNET data sharing.
- All years where IRGA75 data were available were re-uploaded for FLUXNET data sharing. No new Level-1 fluxes were uploaded for 1997-2004 and 2017-2020.
- Fluxes for the FLUXNET / EFDC upload were prepared before FP2021, for the full (incl. full 2020 data) Warm Winter 2020 dataset.
- IRGA75 fluxes were corrected for self-heating following a refined approach based on Kittler et al. (2017). Details can be found on the server.
- IRGA75 Level-1.1 fluxes (corrected for self-heating, 2005-2016) were combined with IRGA72 Level-1 fluxes (2013-2016). To upload a complete yearly file also for 2005, the IRGA75 fluxes were combined with the IRGA62 Level-1 (2005) fluxes.
- Level-1 (IRGA62, IRGA75, IRGA72) and Level-1.1 (IRGA75 only) fluxes are available as part of the EFDC upload on the grasslandserver here:
/mnt/gl-processing/CH-DAV_Davos/95_database_submissions/2021_submitted_to_EFDC/2021-06-30_updatedFluxes_2005-2016
- Level-2, Level-3 and Level-4 fluxes are produced by FLUXNET.
IRGA75 Level-1.1 fluxes (2005-2016, corrected for self-heating)
- Quality check: Rejected CO2 and LE flux values where AGC > 90 (bad quality)
- Preparation for merging with IRGA72 and IRGA62 data:
- This is necessary so that records from the different IRGAs match after merging data from IRGA75 with IRGA72 during the years of parallel measurements (2013-2016)
- For CO2, keep storage, mole fraction and SSITC flag where we have CO2 flux available
- For LE, keep storage, H2O mole fraction and SSITC flag where we have LE flux available
- For H, keep storage and SSITC flag where we have H flux available
- Rename variables to FLUXNET Requirements
IRGA62 Level-1 fluxes (2005)
- No additional quality check
IRGA72 Level-1 fluxes (2012-2016)
- Quality check:
- No additional check, signal strength is good. In 2012 and 2013, low values for signal strength translates to good signal.
- Preparation for merging with IRGA72 data:
- This is necessary so that records from the different IRGAs match after merging data from IRGA75 with IRGA72 during the years of parallel measurements (2012-2016)
- For CO2, keep storage, mole fraction and SSITC flag where we have CO2 flux available
- For LE, keep storage, H2O mole fraction and SSITC flag where we have LE flux available
- For H, keep storage and SSITC flag where we have H flux available
- Rename variables to FLUXNET Requirements
Merging IRGA72+IRGA75+IRGA62
- Datasets were combined into one merged dataset (2005-2016)
- IRGA72 fluxes were prioritized in overlapping years (2013-2016), gaps were filled with IRGA75 fluxes (if available for the IRGA72 gap), IRGA62 fluxes for 2005 were added
- After checking the full time series, the IRGA72 water fluxes in 2013 and 2014 were clearly lower than from the IRGA72 during the same time period. It is possible that during these very first IRGA72 years H2O fluxes were not perfect. Therefore, for 2013-2014, IRGA75 data were used for LE, SLE, LE_SSITC_TEST and H2O.
- Exported each year to separate file with yearly range (2005-2016)
- Uploaded yearly files to EFDC / FLUXNET (9 Jun 2021)
Downloads
- ReddyProc R-script for gap-filling and partitioning:
Page Updates
- 2 Feb 2022: Added download for gap-filling and partitioning script.
- 1 Feb 2022: Mark dataset rc2 as final. Added missing details about post-processing. Added references. Updated link to
diive
script. - 8 Oct 2021: Added rc2, Added link to variable list
- 5 Oct 2021: Initial publication
References
- Lasslop et al. (2010). Separation of net ecosystem exchange into assimilation and respiration using a light response curve approach: Critical issues and global evaluation: SEPARATION OF NEE INTO GPP AND RECO. Global Change Biology, 16(1), 187–208.
- Reichstein et al. (2005). On the separation of net ecosystem exchange into assimilation and ecosystem respiration: Review and improved algorithm. Global Change Biology, 11(9), 1424–1439.
Last Updated on 5 Oct 2024 13:20