Changes between Version 10 and Version 11 of DevelopmentActivities/CMIP6/DevelopmentsCMIP6/zo_evaporation


Ignore:
Timestamp:
2016-02-23T10:39:40+01:00 (8 years ago)
Author:
nvuilsce
Comment:

--

Legend:

Unmodified
Added
Removed
Modified

  • DevelopmentActivities/CMIP6/DevelopmentsCMIP6/zo_evaporation

    v10 v11  
    160160Since the update of the 11-layer hydrological scheme into the trunk of ORCHIDEE in 2012, a strong positive bias on latent heat flux has been diagnosed with the 11-layer scheme at winter time for deciduous species compared to observations and compared to the 2-layer scheme. The figure below shows how the seasonal variations of the latent heat flux simulated with the 2-layer scheme and the 11-layer scheme compare to observations for different PFTs: 
    161161 
    162 ||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_BoENF/lh_month_9874/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_CROc3/lh_month_22172/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_CROc4/lh_month_21877/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_GRAc3/lh_month_7894/lh_month.gif,100%)]]|| 
    163 ||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeDBF/lh_month_3593/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeEBF/lh_month_15717/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeENF/lh_month_20645/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TrEBF/lh_month_21204/lh_month.gif,100%)]]|| 
     162||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_BoENF/lh_month_9874/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_CROc3/lh_month_22172/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_CROc4/lh_month_21877/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_GRAc3/lh_month_7894/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeDBF/lh_month_3593/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeEBF/lh_month_15717/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeENF/lh_month_20645/lh_month.gif,100%)]]||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TrEBF/lh_month_21204/lh_month.gif,100%)]]|| 
    164163 
    165 ||The lower correlation between the monthly latent heat flux with the 11-layer scheme and the observations compared to the correlation between the 2-layer scheme and the obs for Temperate Deciduous Broadleaf forests (TeDBF) is due to the overestimation of the latent heat flux at winter time when there is no leaves. As an example, on the right side, is the latent heat flux at Hainich site:||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeDBF/DE-HaiFL1.CWRR.2425_FL1.Choi.2425/fluxlat_Day_15790/fluxlat_Day.gif,100%)]] || 
     164||The lower correlation between the monthly latent heat flux with the 11-layer scheme and the observations compared to the correlation between the 2-layer scheme and the obs for Temperate Deciduous Broadleaf forests (TeDBF) is due to the overestimation of the latent heat flux at winter time when there is no leaves. As an example, on the right side, is the latent heat flux at Hainich site:||[[Image(http://dods.extra.cea.fr/work/p86ghatt/OL2/PROD/ref2425/FL1.CWRR.2425/fluxnet_taylor_diff_FL1.CWRR.2425_FL1.Choi.2425_TeDBF/DE-HaiFL1.CWRR.2425_FL1.Choi.2425/fluxlat_Day_15790/fluxlat_Day.gif,50%)]] || 
    166165 
     166This overestimation of the latent flux when there is no contribution of transpiration but only bare soil evaporation is not observed over the cropland sites when the soil is bare. 
    167167 
     168Consequently, one searched for a process/formula embedded into ORCHIDEE parametrized differently for trees and herbaceous PFTs and that may impact on the latent heat flux Analysis of the code. This is the case of the formulation of the roughness length (z0 parameter). In ORCHIDEE, one assumes that the roughness length of the herbaceous PFTs varies with the LAI, while the roughness length of Trees PFT does not. See the block of code here in condveg: 
     169https://forge.ipsl.jussieu.fr/orchidee/browser/trunk/ORCHIDEE/src_sechiba/condveg.f90?rev=2425#L1526 
     170 
     171Some authors have proposed formulations of z0 which vary with the LAI as it is roughly done for herbaceous PFTs into ORCHIDEE. For instance, [attachment:su_2001.pdf Su et al. (2001)] developed such formulation which has been used and tested recently at fluxnet sites by [attachment:ershadi_2015.pdf Ershadi et al. (2015)]. The formulation of Su et al., 2001 has be implemented into ORCHIDEE. Inclusion the Su formula for z0 leads to reduce/cancel the bias on latent heat flux at winter for deciduous forest (TeDBF pfts). Results obtained with this formulation are shown below: