86 |
|
|
87 |
To obtain the sources for release1 type: |
To obtain the sources for release1 type: |
88 |
\begin{verbatim} |
\begin{verbatim} |
89 |
% cvs co -d directory -P -r release1 MITgcmUV |
% cvs co -d directory -P -r release1_beta1 MITgcm |
90 |
\end{verbatim} |
\end{verbatim} |
91 |
|
|
92 |
This creates a directory called \textit{directory}. If \textit{directory} |
This creates a directory called \textit{directory}. If \textit{directory} |
116 |
delete; even if you do not use CVS yourself the information can help |
delete; even if you do not use CVS yourself the information can help |
117 |
us if you should need to send us your copy of the code. |
us if you should need to send us your copy of the code. |
118 |
|
|
119 |
|
\paragraph*{Upgrading from an earlier version} |
120 |
|
|
121 |
|
If you already have an earlier version of the code you can ``upgrade'' |
122 |
|
your copy instead of downloading the entire repository again. First, |
123 |
|
``cd'' (change directory) to the top of your working copy: |
124 |
|
\begin{verbatim} |
125 |
|
% cd MITgcm |
126 |
|
\end{verbatim} |
127 |
|
and then issue the cvs update command: |
128 |
|
\begin{verbatim} |
129 |
|
% cvs -q update -r release1_beta1 -d -P |
130 |
|
\end{verbatim} |
131 |
|
This will update the ``tag'' to ``release1\_beta1'', add any new |
132 |
|
directories (-d) and remove any empty directories (-P). The -q option |
133 |
|
means be quiet which will reduce the number of messages you'll see in |
134 |
|
the terminal. If you have modified the code prior to upgrading, CVS |
135 |
|
will try to merge your changes with the upgrades. If there is a |
136 |
|
conflict between your modifications and the upgrade, it will report |
137 |
|
that file with a ``C'' in front, e.g.: |
138 |
|
\begin{verbatim} |
139 |
|
C model/src/ini_parms.F |
140 |
|
\end{verbatim} |
141 |
|
If the list of conflicts scrolled off the screen, you can re-issue the |
142 |
|
cvs update command and it will report the conflicts. Conflicts are |
143 |
|
indicated in the code by the delimites ``<<<<<<<'', ``======='' and |
144 |
|
``>>>>>>>''. For example, |
145 |
|
\begin{verbatim} |
146 |
|
<<<<<<< ini_parms.F |
147 |
|
& bottomDragLinear,myOwnBottomDragCoefficient, |
148 |
|
======= |
149 |
|
& bottomDragLinear,bottomDragQuadratic, |
150 |
|
>>>>>>> 1.18 |
151 |
|
\end{verbatim} |
152 |
|
means that you added ``myOwnBottomDragCoefficient'' to a namelist at |
153 |
|
the same time and place that we added ``bottomDragQuadratic''. You |
154 |
|
need to resolve this conflict and in this case the line should be |
155 |
|
changed to: |
156 |
|
\begin{verbatim} |
157 |
|
& bottomDragLinear,bottomDragQuadratic,myOwnBottomDragCoefficient, |
158 |
|
\end{verbatim} |
159 |
|
and the lines with the delimiters (<<<<<<,======,>>>>>>) be deleted. |
160 |
|
Unless you are making modifications which exactly parallel |
161 |
|
developments we make, these types of conflicts should be rare. |
162 |
|
|
163 |
|
\paragraph*{Upgrading to the current pre-release version} |
164 |
|
|
165 |
|
We don't make a ``release'' for every little patch and bug fix in |
166 |
|
order to keep the frequency of upgrades to a minimum. However, if you |
167 |
|
have run into a problem for which ``we have already fixed in the |
168 |
|
latest code'' and we haven't made a ``tag'' or ``release'' since that |
169 |
|
patch then you'll need to get the latest code: |
170 |
|
\begin{verbatim} |
171 |
|
% cvs -q update -A -d -P |
172 |
|
\end{verbatim} |
173 |
|
Unlike, the ``check-out'' and ``update'' procedures above, there is no |
174 |
|
``tag'' or release name. The -A tells CVS to upgrade to the |
175 |
|
very latest version. As a rule, we don't recommend this since you |
176 |
|
might upgrade while we are in the processes of checking in the code so |
177 |
|
that you may only have part of a patch. Using this method of updating |
178 |
|
also means we can't tell what version of the code you are working |
179 |
|
with. So please be sure you understand what you're doing. |
180 |
|
|
181 |
\section{Model and directory structure} |
\section{Model and directory structure} |
182 |
|
|
183 |
The ``numerical'' model is contained within a execution environment support |
The ``numerical'' model is contained within a execution environment |
184 |
wrapper. This wrapper is designed to provide a general framework for |
support wrapper. This wrapper is designed to provide a general |
185 |
grid-point models. MITgcmUV is a specific numerical model that uses the |
framework for grid-point models. MITgcmUV is a specific numerical |
186 |
framework. Under this structure the model is split into execution |
model that uses the framework. Under this structure the model is split |
187 |
environment support code and conventional numerical model code. The |
into execution environment support code and conventional numerical |
188 |
execution environment support code is held under the \textit{eesupp} |
model code. The execution environment support code is held under the |
189 |
directory. The grid point model code is held under the \textit{model} |
\textit{eesupp} directory. The grid point model code is held under the |
190 |
directory. Code execution actually starts in the \textit{eesupp} routines |
\textit{model} directory. Code execution actually starts in the |
191 |
and not in the \textit{model} routines. For this reason the top-level |
\textit{eesupp} routines and not in the \textit{model} routines. For |
192 |
|
this reason the top-level |
193 |
\textit{MAIN.F} is in the \textit{eesupp/src} directory. In general, |
\textit{MAIN.F} is in the \textit{eesupp/src} directory. In general, |
194 |
end-users should not need to worry about this level. The top-level routine |
end-users should not need to worry about this level. The top-level routine |
195 |
for the numerical part of the code is in \textit{model/src/THE\_MODEL\_MAIN.F% |
for the numerical part of the code is in \textit{model/src/THE\_MODEL\_MAIN.F% |
330 |
\item \textit{global\_ocean.90x40x15} Global circulation with |
\item \textit{global\_ocean.90x40x15} Global circulation with |
331 |
GM, flux boundary conditions and poles. |
GM, flux boundary conditions and poles. |
332 |
|
|
333 |
|
\item \textit{global\_ocean\_pressure} Global circulation in pressure |
334 |
|
coordinate (non-Boussinesq ocean model). Described in detail in |
335 |
|
section \ref{sect:eg-globalpressure}. |
336 |
|
|
337 |
\item \textit{solid-body.cs-32x32x1} Solid body rotation test for cube sphere |
\item \textit{solid-body.cs-32x32x1} Solid body rotation test for cube sphere |
338 |
grid. |
grid. |
339 |
|
|
871 |
\item time-discretization |
\item time-discretization |
872 |
\end{itemize} |
\end{itemize} |
873 |
|
|
874 |
The time steps are set through the real variables \textbf{deltaTMom }and |
The time steps are set through the real variables \textbf{deltaTMom} |
875 |
\textbf{deltaTtracer }(in s) which represent the time step for the momentum |
and \textbf{deltaTtracer} (in s) which represent the time step for the |
876 |
and tracer equations, respectively. For synchronous integrations, simply set |
momentum and tracer equations, respectively. For synchronous |
877 |
the two variables to the same value (or you can prescribe one time step only |
integrations, simply set the two variables to the same value (or you |
878 |
through the variable \textbf{deltaT}). The Adams-Bashforth stabilizing |
can prescribe one time step only through the variable |
879 |
parameter is set through the variable \textbf{abEps }(dimensionless). The |
\textbf{deltaT}). The Adams-Bashforth stabilizing parameter is set |
880 |
stagger baroclinic time stepping can be activated by setting the logical |
through the variable \textbf{abEps} (dimensionless). The stagger |
881 |
variable \textbf{staggerTimeStep }to '.\texttt{TRUE}.'. |
baroclinic time stepping can be activated by setting the logical |
882 |
|
variable \textbf{staggerTimeStep} to '.\texttt{TRUE}.'. |
883 |
|
|
884 |
\subsection{Equation of state} |
\subsection{Equation of state} |
885 |
|
|
886 |
First, because the model equations are written in terms of perturbations, a |
First, because the model equations are written in terms of |
887 |
reference thermodynamic state needs to be specified. This is done through |
perturbations, a reference thermodynamic state needs to be specified. |
888 |
the 1D arrays \textbf{tRef}\textit{\ }and \textbf{sRef}. \textbf{tRef }% |
This is done through the 1D arrays \textbf{tRef} and \textbf{sRef}. |
889 |
specifies the reference potential temperature profile (in $^{o}$C for |
\textbf{tRef} specifies the reference potential temperature profile |
890 |
the ocean and $^{o}$K for the atmosphere) starting from the level |
(in $^{o}$C for the ocean and $^{o}$K for the atmosphere) starting |
891 |
k=1. Similarly, \textbf{sRef}\textit{\ }specifies the reference salinity |
from the level k=1. Similarly, \textbf{sRef} specifies the reference |
892 |
profile (in ppt) for the ocean or the reference specific humidity profile |
salinity profile (in ppt) for the ocean or the reference specific |
893 |
(in g/kg) for the atmosphere. |
humidity profile (in g/kg) for the atmosphere. |
894 |
|
|
895 |
The form of the equation of state is controlled by the character variables |
The form of the equation of state is controlled by the character |
896 |
\textbf{buoyancyRelation}\textit{\ }and \textbf{eosType}\textit{. }\textbf{% |
variables \textbf{buoyancyRelation} and \textbf{eosType}. |
897 |
buoyancyRelation}\textit{\ }is set to '\texttt{OCEANIC}' by default and |
\textbf{buoyancyRelation} is set to '\texttt{OCEANIC}' by default and |
898 |
needs to be set to '\texttt{ATMOSPHERIC}' for atmosphere simulations. In |
needs to be set to '\texttt{ATMOSPHERIC}' for atmosphere simulations. |
899 |
this case, \textbf{eosType}\textit{\ }must be set to '\texttt{IDEALGAS}'. |
In this case, \textbf{eosType} must be set to '\texttt{IDEALGAS}'. |
900 |
For the ocean, two forms of the equation of state are available: linear (set |
For the ocean, two forms of the equation of state are available: |
901 |
\textbf{eosType}\textit{\ }to '\texttt{LINEAR}') and a polynomial |
linear (set \textbf{eosType} to '\texttt{LINEAR}') and a polynomial |
902 |
approximation to the full nonlinear equation ( set \textbf{eosType}\textit{\ |
approximation to the full nonlinear equation ( set |
903 |
}to '\texttt{POLYNOMIAL}'). In the linear case, you need to specify the |
\textbf{eosType}\textit{\ }to '\texttt{POLYNOMIAL}'). In the linear |
904 |
thermal and haline expansion coefficients represented by the variables |
case, you need to specify the thermal and haline expansion |
905 |
\textbf{tAlpha}\textit{\ }(in K$^{-1}$) and \textbf{sBeta}\textit{\ }(in ppt$% |
coefficients represented by the variables \textbf{tAlpha}\textit{\ |
906 |
^{-1}$). For the nonlinear case, you need to generate a file of polynomial |
}(in K$^{-1}$) and \textbf{sBeta} (in ppt$^{-1}$). For the nonlinear |
907 |
coefficients called \textit{POLY3.COEFFS. }To do this, use the program |
case, you need to generate a file of polynomial coefficients called |
908 |
\textit{utils/knudsen2/knudsen2.f }under the model tree (a Makefile is |
\textit{POLY3.COEFFS}. To do this, use the program |
909 |
available in the same directory and you will need to edit the number and the |
\textit{utils/knudsen2/knudsen2.f} under the model tree (a Makefile is |
910 |
values of the vertical levels in \textit{knudsen2.f }so that they match |
available in the same directory and you will need to edit the number |
911 |
those of your configuration). \textit{\ } |
and the values of the vertical levels in \textit{knudsen2.f} so that |
912 |
|
they match those of your configuration). |
913 |
|
|
914 |
|
There there are also higher polynomials for the equation of state: |
915 |
|
\begin{description} |
916 |
|
\item['\texttt{UNESCO}':] The UNESCO equation of state formula of |
917 |
|
Fofonoff and Millard \cite{fofonoff83}. This equation of state |
918 |
|
assumes in-situ temperature, which is not a model variable; \emph{its use |
919 |
|
is therefore discouraged, and it is only listed for completeness}. |
920 |
|
\item['\texttt{JMD95Z}':] A modified UNESCO formula by Jackett and |
921 |
|
McDougall \cite{jackett95}, which uses the model variable potential |
922 |
|
temperature as input. The '\texttt{Z}' indicates that this equation |
923 |
|
of state uses a horizontally and temporally constant pressure |
924 |
|
$p_{0}=-g\rho_{0}z$. |
925 |
|
\item['\texttt{JMD95P}':] A modified UNESCO formula by Jackett and |
926 |
|
McDougall \cite{jackett95}, which uses the model variable potential |
927 |
|
temperature as input. The '\texttt{P}' indicates that this equation |
928 |
|
of state uses the actual hydrostatic pressure of the last time |
929 |
|
step. Lagging the pressure in this way requires an additional pickup |
930 |
|
file for restarts. |
931 |
|
\item['\texttt{MDJWF}':] The new, more accurate and less expensive |
932 |
|
equation of state by McDougall et~al. \cite{mcdougall03}. It also |
933 |
|
requires lagging the pressure and therefore an additional pickup |
934 |
|
file for restarts. |
935 |
|
\end{description} |
936 |
|
For none of these options an reference profile of temperature or |
937 |
|
salinity is required. |
938 |
|
|
939 |
\subsection{Momentum equations} |
\subsection{Momentum equations} |
940 |
|
|
1167 |
The precision with which to write the binary data is controlled by the |
The precision with which to write the binary data is controlled by the |
1168 |
integer variable w\textbf{riteBinaryPrec }(set it to \texttt{32} or \texttt{% |
integer variable w\textbf{riteBinaryPrec }(set it to \texttt{32} or \texttt{% |
1169 |
64}). |
64}). |
1170 |
|
|
1171 |
|
%%% Local Variables: |
1172 |
|
%%% mode: latex |
1173 |
|
%%% TeX-master: t |
1174 |
|
%%% End: |