Hi Jérôme,
Thanks for your comments and sorry for my late answer.
In the end, I’m using a code from the NOC developped for UK with the r4.0.4 version (reference for the previous v3.6 version : O’Neill, C., Saulter, A., Williams, J., and Horsburgh, K.: Storm surge forecasting and other Met Office ocean modelling, 2016. https://digital.nmla.metoffice.gov.uk/download/file/IO_53fa4f69-432c-40bb-9481-8c7dfbd6492d). If it could help someone, I think the problem with my previous case was coming from the removal of the calls to vertical physics. In the code I have now, the zdfphy.F90 routine is also modified with the ln_2d option so that in barotropic mode:
- the function zdf_phy_init is used only for a call to zdf_drg_init
- the function zdf_phy is used only for a call to timing_start and zdf_drg
In addition, the 2-layers case is also working (jpk=2). In the domain_cfg namelist I used these parameters :
!-----------------------------------------------------------------------
&namcfg ! parameters of the configuration
!-----------------------------------------------------------------------
!
ln_e3_dep = .true. ! =T : e3=dk[depth] in discret sens.
! ! ===>>> will become the only possibility in v4.0
! ! =F : e3 analytical derivative of depth function
! ! only there for backward compatibility test with v3.6
! !
cp_cfg = “Caribbean” ! name of the configuration
jp_cfg = 0083 ! resolution of the configuration
jpidta = 544 ! 1st lateral dimension ( >= jpi )
jpjdta = 342 ! 2nd " " ( >= jpj )
jpkdta = 2 ! number of levels ( >= jpk )
jpiglo = 544 ! 1st dimension of global domain → i =jpidta
jpjglo = 342 ! 2nd - - → j =jpjdta
jpizoom = 1 ! left bottom (i,j) indices of the zoom
jpjzoom = 1 ! in data domain indices
jperio = 0 ! lateral cond. type (between 0 and 6)
/
!-----------------------------------------------------------------------
&namzgr ! vertical coordinate (default: NO selection)
!-----------------------------------------------------------------------
ln_zco = .false. ! z-coordinate - full steps
ln_zps = .false. ! z-coordinate - partial steps
ln_sco = .true. ! s- or hybrid z-s-coordinate
ln_isfcav = .false. ! ice shelf cavity
ln_linssh = .false. ! linear free surface
/
!-----------------------------------------------------------------------
&namzgr_sco ! s-coordinate or hybrid z-s-coordinate (default F)
!-----------------------------------------------------------------------ln_s_sh94 = .true. ! Song & Haidvogel 1994 hybrid S-sigma (T)|
ln_s_sf12 = .false. ! Siddorn & Furner 2012 hybrid S-z-sigma (T)| if both are false the NEMO tanh stretching is applied
ln_sigcrit = .false. ! use sigma coordinates below critical depth (T) or Z coordinates (F) for Siddorn & Furner stretch
! stretching coefficients for all functions
rn_sbot_min = 2.0 ! minimum depth of s-bottom surface (>0) (m)
rn_sbot_max = 10000.0 ! maximum depth of s-bottom surface (= ocean depth) (>0) (m)
rn_hc = 0.0 ! critical depth for transition to stretched coordinates
!!! Envelop bathymetry
rn_rmax = 1 ! maximum cut-off r-value allowed (0<r_max<1)
!!! SH94 stretching coefficients (ln_s_sh94 = .true.)
rn_theta = 20 ! surface control parameter (0<=theta<=20)
rn_bb = 0 ! stretching with SH94 s-sigma
!!! SF12 stretching coefficient (ln_s_sf12 = .true.)
rn_alpha = 4.4 ! stretching with SF12 s-sigma
rn_efold = 0.0 ! efold length scale for transition to stretched coord
rn_zs = 1.0 ! depth of surface grid box
! bottom cell depth (Zb) is a linear function of water depth Zb = H*a + b
rn_zb_a = 0.024 ! bathymetry scaling factor for calculating Zb
rn_zb_b = -0.2 ! offset for calculating Zb
!!! Other stretching (not SH94 or SF12) [also uses rn_theta above]
rn_thetb = 1 ! bottom control parameter (0<=thetb<= 1)
Thank you again for your time,
Alisée