[819] | 1 | \documentclass[a4paper,10pt]{article} |
---|
| 2 | %\usepackage{graphicx} |
---|
| 3 | \usepackage{natbib} % si appel à bibtex |
---|
| 4 | %\usepackage[francais]{babel} |
---|
| 5 | %\usepackage[latin1]{inputenc} % accents directs (é...), avec babel |
---|
| 6 | %\usepackage{rotating} |
---|
| 7 | |
---|
| 8 | \setlength{\hoffset}{-1.in} |
---|
| 9 | \setlength{\oddsidemargin}{3.cm} |
---|
| 10 | \setlength{\textwidth}{15.cm} |
---|
| 11 | \setlength{\marginparsep}{0.mm} |
---|
| 12 | \setlength{\marginparwidth}{0.mm} |
---|
| 13 | |
---|
| 14 | \setlength{\voffset}{-1.in} |
---|
| 15 | \setlength{\topmargin}{0.mm} |
---|
| 16 | \setlength{\headheight}{0.mm} |
---|
| 17 | \setlength{\headsep}{30.mm} |
---|
| 18 | \setlength{\textheight}{24.cm} |
---|
| 19 | \setlength{\footskip}{1.cm} |
---|
| 20 | |
---|
| 21 | \setlength{\parindent}{0.mm} |
---|
| 22 | \setlength{\parskip}{1 em} |
---|
| 23 | \newcommand{\ten}[1]{$\times 10^{#1}$~} |
---|
| 24 | \renewcommand{\baselinestretch}{1.} |
---|
| 25 | |
---|
| 26 | \begin{document} |
---|
| 27 | \pagestyle{plain} |
---|
| 28 | |
---|
| 29 | \begin{center} |
---|
| 30 | {\bf \LARGE |
---|
| 31 | Documentation for LMDZ, Planets version |
---|
| 32 | |
---|
| 33 | \vspace{1cm} |
---|
| 34 | \Large |
---|
| 35 | How to modify the initial state: the newstart tool |
---|
| 36 | } |
---|
| 37 | |
---|
| 38 | \vspace{1cm} |
---|
| 39 | S\'ebastien Lebonnois |
---|
| 40 | |
---|
| 41 | \vspace{1cm} |
---|
| 42 | Latest version: \today |
---|
| 43 | \end{center} |
---|
| 44 | |
---|
| 45 | |
---|
| 46 | \section{First step: \textsf{start\_archive.nc}} |
---|
| 47 | |
---|
| 48 | The initial state of a simulation is read by the GCM in the files \textsf{start.nc} and \textsf{startphy.nc}. |
---|
| 49 | A first step to change these files is first to change them into a \textsf{start\_archive.nc} file. This is coming from the Mars GCM, where this file can include all the \textsf{start*.nc} files for the 12 months of one complete Martian Year. |
---|
| 50 | |
---|
| 51 | To do this conversion, you need to use the \textsf{start2archive} tool. |
---|
| 52 | Currently, this tool is well adapted for Venus and Titan, so the corresponding routine is called \textsf{start2archive-VT.F} |
---|
| 53 | It is located in the \textsf{dyn3d} directory, and compiled the same way as the GCM. As inputs, it will need the same \textsf{*.def} files as those used during the simulation that created the \textsf{start*.nc} files. It then creates the \textsf{start\_archive.nc} file from \textsf{start.nc} and \textsf{startphy.nc}. |
---|
| 54 | |
---|
| 55 | \section{Second step: the \textsf{newstart} tool} |
---|
| 56 | |
---|
| 57 | This tool allows many changes in the initial state: |
---|
| 58 | \begin{itemize} |
---|
| 59 | \item change in resolution |
---|
| 60 | \item change in zoom characteristics |
---|
| 61 | \item change in surface characteristics (topography, albedo) |
---|
| 62 | \item change in planetary constants |
---|
| 63 | \end{itemize} |
---|
| 64 | |
---|
[927] | 65 | Currently, this tool is well adapted for Venus and Titan, so the corresponding routine is called \textsf{newstart-VT.F}, located in the \textsf{dyn3d} directory. As for the \textsf{start2archive} tool, it is compiled the same way as the GCM (in sequential only). |
---|
[819] | 66 | |
---|
| 67 | As inputs, it takes the \textsf{start\_archive.nc} file with the initial state to be modified, and the \textsf{traceur.def} and \textsf{z2sig.def} files. |
---|
| 68 | It also needs a specific \textsf{run.def} file, taylored to include only the needed changes. |
---|
| 69 | |
---|
| 70 | To change the resolution, compile \textsf{newstart-VT} in the new resolution. |
---|
| 71 | For the topography, the default behaviour will be to use the previous one (from \textsf{start\_archive.nc}) and to interpolate it to the new resolution. |
---|
| 72 | You may want to use the highest resolution topography file you have to get a finer topography when increasing the resolution. In that case, you can add the line |
---|
| 73 | |
---|
| 74 | \textsf{topoflag=y} |
---|
| 75 | |
---|
| 76 | in the taylored \textsf{run.def} file. Your topography file should be called \textsf{Relief.nc}. This file must include the variable \textsf{RELIEF}, which is the topography in meters. It will be read by \textsf{newstart} and the new surface geopotential will be computed. |
---|
| 77 | This may also be used to modify the topography the way you want. |
---|
| 78 | |
---|
| 79 | One point concerning Venus topography: it should be reversed in the \textsf{Relief.nc} file, so that the GCM runs as if it was looking at Venus with the South pole upward. |
---|
| 80 | |
---|
| 81 | \section{Specific \textsf{run.def} file} |
---|
| 82 | |
---|
| 83 | {\bf Zoom} |
---|
| 84 | |
---|
| 85 | To make a simulation with a zoom, the grid has to be altered. This needs to be done through \textsf{newstart}, using zoom parameters that will be read from the |
---|
| 86 | \textsf{run.def} file. |
---|
| 87 | These parameters include (see specific zoom documentation ?): |
---|
| 88 | \begin{itemize} |
---|
| 89 | \item {\it clon,clat}: coordinates of the zoom center (in degrees) |
---|
| 90 | \item {\it grossismx,grossismy}: resolution increase factor within the zoom area |
---|
| 91 | \item {\it fxyhypb}: logical. If "y" (True) then hyperbolic function for the transition. |
---|
| 92 | \item {\it dzoomx,dzoomy}: fraction of the total domain used in the zoom area (used only if {\it fxyhypb}=y) |
---|
| 93 | \item {\it taux,tauy}: zoom stiffness (in the transition area ?) (used only if {\it fxyhypb}=y) |
---|
| 94 | \item ysinus: logical (used only if {\it fxyhypb}=n). Use sinus of latitude instead of latitude for the transition. |
---|
| 95 | \end{itemize} |
---|
| 96 | |
---|
| 97 | For Venus, the topography being reversed, {\it clat} and {\it clon} must take this into account when targeting a specific feature ! |
---|
| 98 | |
---|
| 99 | {\bf Albedo} |
---|
| 100 | |
---|
| 101 | If you want to change the value of the albedo, you can add the lines |
---|
| 102 | |
---|
| 103 | \textsf{albedoflag=y} |
---|
| 104 | |
---|
| 105 | \textsf{albedo=$<$value$>$} |
---|
| 106 | |
---|
| 107 | in \textsf{run.def} to input the new value. |
---|
| 108 | |
---|
| 109 | For the albedo, we may want to use a map. This is not yet implemented. When it will be, the file containing this map can be used to change the albedo (as for the topography). |
---|
| 110 | |
---|
| 111 | {\bf Other parameters} |
---|
| 112 | |
---|
| 113 | It could be possible to include other parameters in the change (, just by adding optional lines in \textsf{run.def}. The reading has to be hardcoded in \textsf{newstart-VT.F} first, but it's easy. |
---|
| 114 | |
---|
| 115 | The specific heat is already implemented ({\it cpp}). |
---|
| 116 | |
---|
| 117 | \section{Technical aspects} |
---|
| 118 | |
---|
| 119 | The subroutines used for the \textsf{start\_archive.nc} and \textsf{newstart} tools are located in the \textsf{phy$<$planet$>$} directory, since they may slightly vary from one planet to the other. |
---|
| 120 | |
---|
| 121 | For \textsf{start\_archive.nc}, these routines are: |
---|
| 122 | \begin{itemize} |
---|
| 123 | \item \textsf{ini\_archive.F} |
---|
| 124 | \item \textsf{readstart.F} |
---|
| 125 | \item \textsf{readstartphy.F} |
---|
| 126 | \item \textsf{write\_archive.F} |
---|
| 127 | \end{itemize} |
---|
| 128 | |
---|
| 129 | For \textsf{newstart}, they are: |
---|
| 130 | \begin{itemize} |
---|
| 131 | \item \textsf{interp\_vert.F} |
---|
| 132 | \item \textsf{scal\_wind.F} |
---|
| 133 | \item \textsf{wind\_scal.F} |
---|
| 134 | \item \textsf{writerestart.F} |
---|
| 135 | \item \textsf{writerestartphy.F} |
---|
| 136 | \item \textsf{startvar.F90} and \textsf{grid\_noro.F}, which are used to read the \textsf{Relief.nc} file and compute the surface geopotential and the parameters needed for the orographic gravity wave drag. |
---|
| 137 | \end{itemize} |
---|
| 138 | |
---|
| 139 | \end{document} |
---|