Dataset Versions

Dataset Versions#

Current dataset version#

CH-CHA FP2025.3 (2005-2024) [current version]#

  • This release adds additional variables to the dataset and applies a different gap-filling for N2O and CH4 fluxes during 2+ weeks after grassland restoration in 2012.

  • release date: 16 May 2025

  • is currently available on demand from the Grassland Sciences group server

  • Differrences to previous version FP2025.2:

    • Correction gap-filling Feb/Mar 2012: Random forest models were not able to predict N2O and CH4 fluxes satisfactorily between 27 Feb 2012 8:00 and 16 Mar 2012 18:00. Fluxes during this time period were gap-filled using the MDS method (with SW_IN, TS and SWC as predictors) instead of random forest.

    • Added new flux variables from NEE partitioning (2005-2024):

      • Partitioned fluxes were calculated from the gap-filled (random forest) NEE versions (3 USTAR scenarios: 16, 50, 84)

      • Using nighttime method (Reichstein et al., 2005):

        • GPP_NT_CUT_16_gfRF, GPP_NT_CUT_50_gfRF, GPP_NT_CUT_84_gfRF

        • RECO_NT_CUT_16_gfRF, RECO_NT_CUT_50_gfRF, RECO_NT_CUT_84_gfRF

      • Using daytime method (Lasslop et al., 2010):

        • GPP_DT_CUT_16_gfRF, GPP_DT_CUT_50_gfRF, GPP_DT_CUT_84_gfRF

        • RECO_DT_CUT_16_gfRF, RECO_DT_CUT_50_gfRF, RECO_DT_CUT_84_gfRF

    • Added new meteo variables (2005-2024):

      • Soil heat flux: G_GF1_0.03_1, G_GF1_0.03_2, G_GF1_0.05_1, G_GF1_0.05_2, G_GF4_0.02_1, G_GF5_0.02_1 (different depths not complete over the years, sometimes only sporadically)

      • Radiation: SW_OUT_T1_2_1, LW_OUT_T1_2_1, NETRAD_T1_2_1, PPFD_OUT_T1_2_2

  • For a list of previous versions see here.

Previous dataset versions#

CH-CHA FP2025.2 (2005-2024)#

  • release date: 7 Mar 2025

  • Differrences to previous version FP2025.1:

    • Flux calculations Level-1, updated fluxes for 2023: The vertical wind component W showed a constant offset during some time periods. Fluxes for these time periods were re-calculated separately, taking the offset into account in the EddyPro settings. See this table, Note (28), for the exact time periods. Other time periods during the same year were also re-calculated, but without the offset time periods.

    • Post-processing Level-2, SSITC test: stricter setting for all fluxes between 2022-05-01 and 2023-09-30. For this test flag, data of medium quality were set to bad quality. This allowed to filter out erratic flux values due to a drift towards negative numbers observed in the vertical wind component W.

    • No new data were added.

CH-CHA FP2025.1 (2005-2024)#

  • initial release

  • release date: 8 Feb 2025

Upcoming updates#

  • RF = gap-filled using random forest

  • MDS = gap-filled using marginal distrubution sampling from Reichstein et al. (2005)

  • ET calculated from LE (RF)

  • maybe GPP (MDS) and RECO (MDS) from NEE (MDS) partitioning

  • maybe ET calculated from LE (MDS)

  • maybe Modified daytime paritioning (NEE) method based on Keenan et al. (2019)

  • more meteo variables

Contents