and \citet{lupkes.gryanik_JGR15} (\np{ln_Cd_L15}{ln\_Cd\_L15}) parameterizations
\texttt{\#out\_of\_place.}
Surface turbulent fluxes between sea-ice and the atmosphere can be computed in three different ways:
To use it, an atmospheric vertical grid and specific atmospheric forcing files must be provided to ABL1D.
This is because the model expects atmospheric data on its vertical grid and not only near the surface as usually done.
Another specificity of ABL1D is that it can be dynamically driven by geostrophic wind or horizontal air pressure gradient,
instead of being classicaly relaxed toward the large-scale wind forcing.
To generate the ABL1D vertical grid and atmospheric forcings, specific tools and an associated namelist are provided in the ABL\_TOOLS directory.
They have been developed specifically to deal with ECMWF atmospheric products (such as ERA-Interim, ERA5 and IFS) on their native vertical eta-coordinates.
The namelist is used to setup the ABL1D vertical grid (\forcode{&nml_dom}), atmospheric forcing options (\forcode{&nml_opt}),
input atmospheric filenames (\forcode{&nml_fld}) and outputs filenames (\forcode{&nml_out}).
\begin{listing}
\nlst{namelist_abl_tools}
\label{lst:namelist_abl_tools}
\end{listing}
Each of the three steps needed to generate the atmospheric forcings corresponds to a tool:
\begin{itemize}
\begin{itemize}
\item Constant value (\forcode{Cd_ice=1.4e-3}):
\item main\_uvg\_hpg (optional):\\
default constant value used for momentum and heat neutral transfer coefficients
geostrophic wind or horizontal pressure gradient computation on ECMWF eta-levels
Alternative turbulent transfer coefficients formulation between sea-ice
and atmosphere with distinct momentum and heat coefficients depending
on sea-ice concentration and atmospheric stability (no melt-ponds effect for now).
The parameterization is adapted from ECHAM6 atmospheric model.
Compared to Lupkes2012 scheme, it considers specific skin and form drags
to compute neutral transfer coefficients for both heat and momentum fluxes.
Atmospheric stability effect on transfer coefficient is also taken into account.
\end{itemize}
\end{itemize}
\subsection{ABL1D namelist}
\begin{listing}
\nlst{namsbc_abl}
\caption{\forcode{&namsbc_abl}}
\label{lst:namsbc_abl}
\end{listing}
ABL1D model is activated by adding ABL sources directory to the sources list file (\*\_cfgs.txt) and
by setting \np[=.true.]{ln_abl}{ln\_abl} (and \np[=.false.]{ln_blk}{ln\_bkl}) in \nam{sbc}{sbc}. \\
It is fully compatible with Nemo Standalone Surface module and can be consequently forced by sea surface temperature and currents external data.\\
Atmospheric forcing files needed by ABL1D must be specified directly using the \np{sn_wndi}{sn\_wndi}, \np{sn_wndj}{sn\_wndj},
\np{sn_tair}{sn\_tair} and \np{sn_humi}{sn\_humi} parameters from the \nam{sbc_blk}{sbc\_blk}.\\
When using geostrophic wind (\np[=.true.]{ln_geos_winds}{ln\_geos\_winds})
or horizontal air pressure gradient (\np[=.true.]{ln_hpgls_frc}{ln\_hpgls\_frc}) as dynamical guide, additional \np{sn_hpgi}{sn\_hpgi}
and \np{sn_hpgj}{sn\_hpgj} parameters must be provided using geostrophic wind/pressure gradient i/j-components files generated during the pre-processing steps.\\
Note that due to fldread limitations, the interpolation weight filenames must be different between 2D and
3D atmospheric forcings (even if it is the same weight file).
\subsubsection{Tracers and Dynamics relaxation time}
ABL1D tracers needs to be relaxed toward atmospheric temperature (\np{sn_tair}{sn\_tair})
and humidity (\np{sn_humi}{sn\_humi}) forcings to provide a top boundary condition to the model and to avoid the formation of biases due to the lack
of representation of some important atmopheric processes such as advection and convection.
This relaxation time can be setup independently inside the ABL and above the ABL and it is expressed in hours.\\
The recommanded values for the tracers relaxation time is typically 3 times the ocean model timestep inside the ABL (\np{rn_ltra_min}{rn\_ltra\_min})
and 1 ocean model timestep above the ABL (\np{rn_ltra_max}{rn\_ltra\_max}).\\
\\
The dynamical relaxation time inside (\np{rn_ldyn_min}{rn\_ldyn\_min}) and above (\np{rn_ldyn_max}{rn\_ldyn\_max}) the ABL is only needed in two cases:
\begin{itemize}
\item when geostrophic wind / horizontal pressure gradient options are not used.
\item when geostrophic wind / horizontal pressure gradient options are used and the geographical domain includes the equatorial band
where the geostrophic equilibrium is too weak to contrain efficiently ABL1D dynamics.
\end{itemize}
The recommanded minimum and maximum dynamical relaxation values are identical to the tracers relaxation times.\\
\subsubsection{Turbulent vertical mixing lenght and constants}
The ABL1D turbulence scheme used to compute eddy diffusivities for momentum and scalars relies on a TKE pronostic equation (following \citet{cuxart.bougeault_QJRMS00})
which depends on mixing lenght scales and turbulent constants.
To address the ABL1D sensitivity to these parameters, various mixing lenght formulations and turbulent constants sets are provided in namelist:
\begin{itemize}
\item Three different mixing length scales can be selected using \np{nn_amxl}{nn\_amxl}:\\
(0) \citet{deardorff.ea_BLM80}\\
(1) PBL height distance function (as in Nemo TKE scheme)\\
(2) \citet{bougeault.lacarrere_MWR89}\\
\item Three different sets of turbulent constants are proposed:\\
\citet{cuxart.bougeault_QJRMS00}, \citet{cheng.canuto.ea_JAS02} and \citet{lac.chaboureau.ea_GMD18}\\
\begin{table}[htbp]
\centering
\begin{tabular}{|l|c|c|c|}
\hline
& CBR00 & CCH02 & MNH54 \\
\hline
rn\_Cm & 0.0667 & 0.1260 & 0.1260 \\
\hline
rn\_Ct & 0.1667 & 0.1430 & 0.1430 \\
\hline
rn\_Ce & 0.40 & 0.34 & 0.40 \\
\hline
rn\_Ceps & 0.700 & 0.845 & 0.850 \\
\hline
rn\_Ric & 0.139 & 0.143 & ? \\
\hline
rn\_Rod & 0.15 & 0.15 & 0.15 \\
\hline
\end{tabular}
\end{table}
\end{itemize}
More details about the turbulence scheme parameters and their effect on ABL properties can be found in \citet{lemarie.samson.ea_GMD21}.
