|
|
|
[[_TOC_]]
|
|
|
|
|
|
|
|
# Chlorophyll and light in MEDUSA
|
|
|
|
|
|
|
|
## The different light settings of MEDUSA
|
|
|
|
|
|
|
|
NEMO-MEDUSA enables you to use different light-Chl configurations. \
|
|
|
|
NEMO can use the BGC model's Chl or not. The different options can be set in NEMO's namelist_cfg : namtra_qsr
|
|
|
|
|
|
|
|
```
|
|
|
|
!-----------------------------------------------------------------------
|
|
|
|
&namtra_qsr ! penetrative solar radiation (ln_traqsr =T)
|
|
|
|
!-----------------------------------------------------------------------
|
|
|
|
! ! type of penetration (default: NO selection)
|
|
|
|
ln_qsr_rgb = .true. ! RGB light penetration (Red-Green-Blue)
|
|
|
|
ln_qsr_2bd = .false. ! 2BD light penetration (two bands)
|
|
|
|
ln_qsr_bio = .false. ! bio-model light penetration
|
|
|
|
! ! RGB & 2BD choices:
|
|
|
|
nn_chldta = 3 ! RGB : Chl data (=1) or cst value (=0)
|
|
|
|
cn_dir = '/work/n01/shared/julpal/ANCIL4.2/' ! root directory for the chlorophyl data location
|
|
|
|
!___________!_________________________!___________________!___________!_____________!________!___________!__________________!__________!_______________!
|
|
|
|
! ! file name ! frequency (hours) ! variable ! time interp.! clim ! 'yearly'/ ! weights filename ! rotation ! land/sea mask !
|
|
|
|
! ! ! (if <0 months) ! name ! (logical) ! (T/F) ! 'monthly' ! ! pairing ! filename !
|
|
|
|
sn_chl ='merged_ESACCI_BIOMER4V1R1_CHL_REG05' , -1. , 'CHLA' , .true. , .true. , 'yearly' , 'weights_reg05_2_eorca1_bilinear.nc' , '' , ''
|
|
|
|
/
|
|
|
|
```
|
|
|
|
The different possibilities are :
|
|
|
|
1. NEMO sees a flat constant chl :
|
|
|
|
- set `ln_qsr_rgb` to `.true.`
|
|
|
|
- and `nn_chldta` to 0
|
|
|
|
1. If you want NEMO to see a climatology :
|
|
|
|
- set `ln_qsr_rgb` to `.true.`
|
|
|
|
- and `nn_chldta` to 1 or 2 (1 for surface chl climatology; 2 for 3D chl climatology)
|
|
|
|
- Fill `sn_chl` and `cn_dir` to point to the right chl climatology file.
|
|
|
|
3. For NEMO to see MEDUSA's chl :
|
|
|
|
- set `ln_qsr_rgb` to `.true.`
|
|
|
|
- and `nn_chldta` to 3
|
|
|
|
4. For NEMO light and MEDUSA's Chl to be fully and properly coupled :
|
|
|
|
- set `ln_qsr_bio` to `.true.` instead of `ln_qsr_rgb`.
|
|
|
|
That way, NEMO will use its own `trcopt` subroutine, and calculate the PAR carefully in 3 bands. \
|
|
|
|
{- ** WARNING ** -} : MEDUSA is not yet tuned for this option.
|
|
|
|
5. If you want MEDUSA to use NEMO's attenuated light :
|
|
|
|
this option is to be found in MEDUSA's namelist : nammsa_ctl
|
|
|
|
```
|
|
|
|
&nammsa_ctl
|
|
|
|
ln_msa_carbchem = .true. !! activates Carbone cycle
|
|
|
|
ln_msa_avgqsr = .true. !! biology sees daily avg light
|
|
|
|
ln_msa_ccd = .false. !! reads ccd file to init ccd field.
|
|
|
|
ln_msa_preico2 = .false. !! atm co2 forced at pre-ind level
|
|
|
|
ln_msa_ukesm_cpl = .false. !! activate coupling in MEDUSA with UKESM.
|
|
|
|
ln_msa_omipdic = .false. !! 2 dic pre-ind and date-related
|
|
|
|
ln_msa_zmp = .false. !! switch to Tom's Oithona + 2 very fast detritus
|
|
|
|
ln_msa_noco2flux = .false. !! debug - cut sea-air co2 fluxes
|
|
|
|
ln_msa_debug = .false. !! add lots of debug print statements
|
|
|
|
ln_msa_nancheck = .false. !! checks if any nans in bio 3D fields
|
|
|
|
ln_msa_check_n_fix = .false. !! check and exceptionally fix crazy
|
|
|
|
!! values for DIC, ALK, T, and S.
|
|
|
|
!! avoids carb chem crashes
|
|
|
|
ln_msa_kill_medusa = .false. !! debug - no bio sources and sinks
|
|
|
|
ln_msa_no_detsink = .false. !! debug - poc don't sink
|
|
|
|
ln_msa_ice_frac = .true. !! mask light with ice fraction
|
|
|
|
ln_msa_nemo_qsr = .true. !! use 3D light field attenuated
|
|
|
|
!! from NEMO if TRUE
|
|
|
|
/
|
|
|
|
```
|
|
|
|
Set `ln_mas_qsr` to `.true.`
|
|
|
|
6. To couple light and chl without using NEMO's *all* *included* subroutine,
|
|
|
|
Simply do both 3. and 5. :
|
|
|
|
- set `ln_qsr_rgb` to `.true.`
|
|
|
|
- and `nn_chldta` to 3
|
|
|
|
- and `ln_msa_qsr` to `.true.`
|
|
|
|
|
|
|
|
## Correcting light - chl and PP with the ice-fraction
|
|
|
|
|
|
|
|
Just to explain what this is about :\
|
|
|
|
In MEDUSA II, the light (`qsr`) that NEMO and MEDUSA get at the surface already take into account the light attenuation for the sea-ice. It sounds reasonable to use this light without correction for the primary production calculations.
|
|
|
|
|
|
|
|
But :
|
|
|
|
1. Some light manages to cross the fully ice-covered regions of the ocean, especially the Arctic, resulting in a significant chlorophyll concentration in these regions.
|
|
|
|
1. The light the model phytoplankton sees in the partially ice-covered regions is an average of ice-free and ice-covered light; whereas, in reality, the phytoplankton should grow in the ice-free parts, at full light. The ice-covered regions being assumed to be dark.
|
|
|
|
|
|
|
|
Here we managed to get the right chl and production at full light in partially ice-covered regions, that we then attenuate with the ice-fraction.
|
|
|
|
The Arctic light-chl problem is solved in the process.
|
|
|
|
|
|
|
|
To include this fix,
|
|
|
|
switch `ln_msa_ice_frac` in the namelist_medusa_cfg to `.true.`
|
|
|
|
|
|
|
|
## Daily cycle or Daily averaged light
|
|
|
|
|
|
|
|
* **By Default, MEDUSA should run with Daily averaged light**.
|
|
|
|
All chl and Production paramaters are tuned for an averaged light.
|
|
|
|
You can double check it's done this way by making sure that `ln_msa_avgqsr` in nammsa_ctl is set to `.true.`
|
|
|
|
|
|
|
|
If you use the fully coupled chl-light scheme, make also sure that `ln_trcdc2dm` is set to `.true.` (dc2dm for "Daily Cycle to Daily Mean" \
|
|
|
|
`ln_trcdc2dm` can be found in `namtrc_dcy` in the namelist_top
|
|
|
|
|
|
|
|
```
|
|
|
|
!-----------------------------------------------------------------------
|
|
|
|
&namtrc_dcy ! Diurnal cycle
|
|
|
|
!-----------------------------------------------------------------------
|
|
|
|
ln_trcdc2dm = .true. ! Diurnal cycle for TOP
|
|
|
|
/
|
|
|
|
```
|
|
|
|
|
|
|
|
* **If, for some reason you need the light daily cycle**,
|
|
|
|
Set **`ln_msa_avgqsr` and `ln_trcdc2dm` to `.false.`** (ln_trcdc2dm, should not be necessary, but, i mention it just for good measure - can't be too prudent) |
|
|
\ No newline at end of file |
