ORCHIDEE-NEWPHENO

This work has been performed by Nicolas Viovy and evaluated for LAI by Fabienne Maignan. The main purpose of this branch and of this wiki page is to merge these changes on phenology into the main branch of ORCHIDEE and to evaluate the overall performances of this new merged version. This work is being performed by Nicolas Vuichard. A branch ​ORCHIDEE_NEW_PHENO has been dedicated to these developments.

Brief description of the changes on phenology

A full description of the changes can be found in the paper of F. Maignan (​see her paper here) The main changes are :

• the new phenology scheme uses the winter solstice calculated for each orbital condition, as the reference date for GDD calculations instead of the end of the former growing cycle (time_lowgpp).
• the new phenology version now includes a specific climate driven senescence model for crops, based on the work done in Bondeau et al. (2007) for the DGVM LPJ managed Land (LPJmL). This crops senescence model is simply based on a GDD threshold.
• An age-based senescence is added for grasses into the new phenology version (stomate_turnover.f90)

Updates compared to Fabienne's version

• Displacement of the routine 'rad' from stomate_season to global/solar.f90. This part of code was already used in weather.f90 and has been consequently defined as a function named 'downward solar_flux' into solar.f90
• Since the reference date for GDD calculations is now based on winter solstice, supression of the time_lowgpp variable.
• Restructuration of stomate_turnover (+correction of a bug when 'mixed' phenology was used for crop (former model))
  • line 262 : problem with parenthesis ?

               WHERE ( ( biomass(:,j,ileaf) .GT. 0.0 ) .AND. &
                    ( leaf_meanage(:,j) .GT. pheno_crit%min_leaf_age_for_senescence(j) ) .AND. &
                    ( moiavail_week(:,j) .LT. moiavail_crit(:) ) .OR. &
                    ( ( t2m_month(:) .LT. t_crit(:) ) .AND. ( t2m_week(:) .LT. t2m_month(:))))
    

    It should be

               WHERE ( ( biomass(:,j,ileaf) .GT. 0.0 ) .AND. &
                    ( leaf_meanage(:,j) .GT. pheno_crit%min_leaf_age_for_senescence(j) ) .AND. &
                    (( moiavail_week(:,j) .LT. moiavail_crit(:) ) .OR. &
                    ( ( t2m_month(:) .LT. t_crit(:) ) .AND. ( t2m_week(:) .LT. t2m_month(:)))))
    

  • line 316 : suppression of

               WHERE ( leaf_meanage(:,j) .LT. pheno_crit%min_leaf_age_for_senescence(j)) 
                  turnover_time(:,j)= pheno_crit%max_turnover_time(j)
               ENDWHERE
    

  • line 360 : test "where (senescence(:,j))" should be a test on type of pheno == 'crop'

and test "if (natural(j))" should be a "else" statement. =>because it may create a bug in case of crop pft base on a 'mixed' phenology => No action done == no flag 'senescence' and 'not(natural)'

Evaluation of this new version

Different versions have been used in order to evaluate this version:

• ref_pheno is the version we used as reference. It corresponds to the version 195
• init_pheno is the version taken from Fabienne and adpated as described above (except for time_lowgpp supression) without the new crop senescence
• new_pheno is the version taken from Fabienne and adpated as described above (with the supression of time_lowgpp) without the new crop senescence
• init_pheno_crop is the version taken from Fabienne and adpated as described above (except for time_lowgpp supression) without the new crop senescence
• new_pheno_crop is the version taken from Fabienne and adpated as described above with the new crop senescence

On Fluxnet sites

Several comparisons have been done between these versions, first based on performances on fluxnet sites for carbon (GPP, TER, NEE) and energy fluxes (LH, SH). The sites have been splited according to the IGBP classification: ENF : Evergreen Needleleaf Forest EBF : Evergreen Broadleaf Forest DNF : Deciduous Needleleaf Forest DBF : Deciduous Broadleaf Forest GRA : Grasslands WSA : Woody savannas CRO : Croplands

Comparisons between :

• init_pheno and ref_pheno
  • for ​ALL sites
  • for ​ENF sites
  • for ​EBF sites
  • for ​DBF sites
  • for ​GRA sites
  • for ​WSA sites
  • for ​CRO sites
• new_pheno and init_pheno
  • for ​ALL sites
  • for ​ENF sites
  • for ​EBF sites
  • for ​DBF sites
  • for ​GRA sites
  • for ​WSA sites
  • for ​CRO sites
• init_pheno_crop and init_pheno
  • for ​CRO sites
• new_pheno_crop and init_pheno_crop
  • for ​CRO sites

At global scale

We also perform simulations at global scale over the period 1980-2010 and compare the LAI against NDVI product from MODIS. Here below are the results in terms of correlation (monthly time scale) for the different versions

• tag195 VS init_pheno_crop : ​pdf
• tag195 VS new_pheno_crop : ​pdf
• tag195 VS rev998 : ​pdf
• new_pheno_crop VS rev998: ​pdf